Radioanalytical Chemistry

Bernd Kahn, Editor

Radioanalytical Chemistry
Library of Congress Control Number: 2006925171

ISBN 10: 0-387-34122-6 Printed on acid-free paper.

ISBN 13: 978-0387-34122-4


C 2007 Springer Science+Business Media, LLC
All rights reserved. This work may not be translated or copied in whole or in part without the written
permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York,
NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use
in connection with any form of information storage and retrieval, electronic adaptation, computer
software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they
are not identified as such, is not to be taken as an expression of opinion as to whether or not they are
subject to proprietary rights.

9 8 7 6 5 4 3 2 1

springer.com
Contents

Acknowledgments..................................................................... vii

1 Introduction ........................................................................ 1
Bernd Kahn

2 Radiation Detection Principles ................................................. 7

Moses Attrep, Jr.

3 Analytical Chemistry Principles ............................................... 39

Jeffrey Lahr and Bernd Kahn

4 Radioanalytical Chemistry Principles and Practices ....................... 64

Jeffrey Lahr, Bernd Kahn, and Stan Morton

5 Sample Collection and Preparation............................................ 77

Robert Rosson

6 Applied Radioanalytical Chemistry ........................................... 93

Bernd Kahn, Robert Rosson, and Liz Thompson

7 Preparation for Sample Measurement......................................... 121

Bernd Kahn

8 Applied Radiation Measurements ............................................. 134

John M. Keller

9 Radionuclide Identification ..................................................... 163

Bernd Kahn

10 Data Calculation, Analysis, and Reporting ................................. 189

Keith D. McCroan and John M. Keller

v
vi Contents

11 Quality Assurance............................................................... 220

Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

12 Methods Diagnostics ........................................................... 244

Bernd Kahn

13 Laboratory Design and Management Principles........................... 261

Charles Porter and Glenn Murphy

14 Laboratory Safety ............................................................... 292

Arthur Wickman, Paul Schlumper, Glenn Murphy, and Liz Thompson

15 Automated Laboratory and Field Radionuclide Analysis Systems .... 318

Harry S. Miley and Craig Edward Aalseth

16 Chemistry Beyond the Actinides ............................................. 338

Darleane C. Hoffman

17 Mass Spectrometric Radionuclide Analyses ............................... 362

John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

Appendix A1: ASTM Radiological Standard Test Methods, Practices,

and Guides (ASTM 2005) ..................................................... 410

Appendix A2: Standard Methods, part 7000—Radioactivity

(Standard Methods 2005) ...................................................... 416

Appendix A3: Applicable ANSI Standard Procedures

(ANSI 2005)...................................................................... 417

Appendix A4: Applicable ISO Standard Procedures (ISO 2005)....... 418

Appendix B: Regulation- and Standard-Setting Agencies .............. 418

Appendix C....................................................................... 420

Glossary................................................................................. 421

References .............................................................................. 437

Index ..................................................................................... 467

Acknowledgments

The following members of the Editorial Advisory Board (EAB) contributed enor-
mously to this textbook by suggesting format, reviewing successive drafts, and
some even writing chapters (as indicated in chapter headings). It was a great plea-
sure to work with all contributors, and they have my heartfelt thanks for their
participation.
r Moses Attrep, Jr., LANL-retired
r Darleane Hoffman, LBL
r Kenneth Inn, NIST
r John Keller, ORNL
r Harry Miley, PNNL
r Stan Morton, RESL-IDO-retired
r Glenn Murphy, UGA
r Dick Perkins, PNNL
r Charles Porter, EPA-retired
r Jake Sedlet, ANL
r John Wacker, PNNL

Isabel Fisenne and A. L. Boni were members of the EAB for several years and
then resigned.
My deepest sympathy goes to the family of Jake Sedlet, who died while this
work was in progress.
I appreciate the efforts of others who contributed by writing a chapter:
r Craig Aalseth, PNNL
r Gregory Eiden, PNNL
r Pam Greenlaw, EML
r Jeffrey Lahr, GT
r Scott Lehn, PNNL
r Keith McCroan, EPA
r Robert Rosson, GT
r Paul Schlumper, GT
r Linda Selvig, Boise ID

vii
viii Acknowledgments

r Liz Thompson, GT
r Arthur Wickman, GT

Each chapter is credited to the main author or authors, but many portions of
their original manuscripts were moved to other chapters for better text integration.
This effort was supported by NNSA–DOE under grant DE-FG07-01ID14224.
Dan Griggs and Stephen Chase were successive project officers. I am grateful to
them for their active involvement in the preparation of this text, and to Dale Perry
and Terry Creque of NNSA for participating in the EAB meetings.
I thank the Georgia Institute of Technology, my professional home when the
book was written, and my coworkers there. Liz Thompson was the editorial writer
who untiringly worked on every aspect of the text to prepare it for publication with
great insight and competence. Robert Rosson and Jeff Lahr are radiochemists who
were most helpful in working with me and writing specific chapters. Jean Gunter
was the skilled coordinator of all EAB meetings, who made attendance a pleasure.
I wish to express my deepest respect and thanks to the two radiochemists who
taught me, William S. Lyon at ORNL and Charles D. Coryell at MIT.
This book is dedicated with love to my wife, Gail, and my parents, Alice and
Eric Kahn.
1
Introduction
BERND KAHN

1.1. Background
Radioanalytical chemistry is devoted to analyzing samples for their radionu-
clide content. For this purpose, the strategies of identifying and purifying the
radioelements of interest by chemical methods, and of identifying and measur-
ing the disintegration rate (“activity”) of radionuclides by nuclear methods, are
combined. Radioanalytical chemistry can be considered to be a specialty in the
subdiscipline of nuclear and radiochemistry.
This textbook was written to teach radioanalytical chemistry in the classroom
and support its application in the laboratory. Its emphasis is on the practical as-
pects of the specialty, notably setting up the laboratory, training its staff, and
operating it reliably. The information presented herein, outlined in Section 1.4,
is the accumulated product of a century of nuclear chemistry and radiochemistry
practice.
Radioanalytical chemistry was first developed by Mme. M. Curie, with contribu-
tions by many other distinguished researchers, notably E. Rutherford and F. Soddy.
These pioneers performed chemical separations and radiation measurements on
terrestrial radioactive substances during the 20 years following 1897 and in the
process created the very concept of radionuclides. Their investigations defined
the three major radiation types, confirmed the emission of these radiations by the
nucleus and the associated atomic transformations, established the periodic table
between bismuth and uranium, and demonstrated the distinction between stable
and radioactive isotopes.
Thereafter, cosmic rays were observed and explained, and many cosmic-ray-
produced radionuclides were identified. The number of known and character-
ized radionuclides increased dramatically with the development and application
of nuclear-particle accelerators in the 1930s and nuclear-fission reactors in the
1940s.
Since Mme. Curie’s time, applications of radioanalytical chemistry have pro-
liferated. Modern practitioners of nuclear and radiochemistry have applied chem-
ical and nuclear procedures to elucidate nuclear properties and reactions, used

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

1
2 Bernd Kahn

radioactive substances as tracers for solving problems, and measured radionu-

clides in many types of samples. The work plays an integral part in research
related to chemistry, physics, medicine, pharmacology, biology, ecology, hydrol-
ogy, geology, forensics, atmospheric sciences, health protection, archeology, and
engineering. Applications that generate great interest include forming and charac-
terizing new elements, determining the age of materials, and creating radioactive
reagents for specific tracer use, notably targeting selected tissues and organs with
radionuclides for diagnosis and treatment. New and clever analytical methods
are developed to monitor ever-more radionuclides at ever-lower concentrations in
persons and in the environment.
The increasingly common use of radioactive materials has engendered an entire
industry dedicated to monitoring the locations and amounts of those radionu-
clides. Radionuclides are monitored in the environment, in effluent and process
streams, and in workers by nuclear research laboratories, nuclear fuel cycle facil-
ities, radiological measurement contractors, and government agencies. Ongoing
monitoring efforts examine the products of routine operation, research projects,
waste processing, and storage; additional monitoring may be ordered in response
to contamination incidents, nuclear site closures, or terrorist actions. Throughout
the world, radioanalytical chemists collaborate in investigations of contamination
from nuclear weapon tests and nuclear accidents.
Because many of the required skills are the same, a competent nuclear chemist
or radiochemist can move from one of the above-mentioned fields to another.
Radioanalytical chemists combine separation techniques of classical chemical
qualitative and quantitative analysis with the radiation detection techniques of
nuclear physics. Practitioners polish techniques for processing distinctive sample
matrices, separating various sets of radionuclides, and improving detection sen-
sitivity. In some instances, informed use of radiation detection techniques elimi-
nates any need for chemical separations, or mass spectrometry replaces radiation
detection.
Despite the continuing interest and expanding applications, the number of prac-
titioners in nuclear chemistry and radiochemistry and its radioanalytical chemistry
specialty has decreased to an alarmingly low level from its peak in the 1950s and
1960s. Two causes have combined to produce the situation. For a variety of reasons,
many academic institutions no longer teach the discipline, and so the number of
pertinent degree programs and the students who graduate from them have declined
markedly. At the same time, the senior members of the radiochemistry community
are leaving the profession by retirement, and so many of the practitioners who
contributed to the flourishing of radioanalytical chemistry no longer are active.
International and U.S. agencies recognize the seriousness of this problem and
have taken initial steps to reverse the trend. This textbook was written to contribute
to these efforts. First, it provides a newly prepared text for expanded programs
in teaching students the fundamentals of radioanalytical chemistry. Secondly, it
attempts to preserve and transmit the practical aspects of this specialty for training
professionals that are newly assigned to radioanalytical chemistry tasks.
 1. Introduction 3

1.2. Information Sources for Radioanalytical Chemistry

Nuclear chemistry and radiochemistry are described in many excellent texts. Two
of the more widely used ones are by Friedlander et al. (1981) and Choppin et al.
(1995). A five-volume Handbook of Nuclear Chemistry, edited by Vertes et al.
(2003), has been published recently. It includes pertinent applications such as
activation analysis and tracer use, and an excellent brief history by Friedlander
and Herrman (2003). Radioanalytical chemists can obtain information from these
texts and others on such vital aspects of the work as the sources of radionuclides,
radiation detection, radiation interactions, and applications to varied fields. Other
useful books on these topics were written within the past 15 years by Ehmann and
Vance (1991), Navratil et al. (1992), and Adloff and Guillaumont (1993).
Several books that address certain aspects of radioanalytical chemistry are valu-
able sources of information for the radiochemist. The CRC Handbook of Radioan-
alytical Chemistry (Tolgyessy and Bujdoso 1991) is a two-volume set that devotes
the majority of its more than 1700 pages to physical constants, mathematical
tables, radionuclide data, and activation data. It includes charts and tables with rel-
evant data for solvent extraction, ion exchange, masking reactions, isotope dilution
analysis, radiometric titrations, and organic and inorganic radioreagent methods. A
table of radiochemical analysis methods is reproduced from NCRP (1976b). This
handbook is an updated and condensed version of Nuclear Analytical Chemistry,
Tolgyessy’s five-volume work of 20 years earlier (Tolgyessy 1971).
A text with a scope similar to this book is Radiochemical Methods in Analysis
(Coomber 1975). The text contains such relevant chapters as “Separation methods
for inorganic species” and “The use of tracers in inorganic analysis.” A chapter
titled “Determination of radioactivity present in the environment” contains infor-
mation geared toward sample collection.
Elementary Practical Radiochemistry (Ladd and Lee 1964) contains 20 brief
experiments that illustrate detection techniques such as measurement of ingrowth
and decay, as well as ion exchange, extraction, and coprecipitation. The text Ra-
dioisotope Laboratory Techniques (Faires and Boswell 1981) primarily addresses
nuclear physics, radionuclide production, and counting techniques. It briefly men-
tions laboratory apparatus but omits discussion of separation techniques.
The section “Radioactive Methods” in volume 9 of the Treatise on Analytical
Chemistry (Kolthoff and Elving 1971) discusses radioactive decay, radiation detec-
tion, tracer techniques, and activation analysis. It has a brief but informative chapter
on radiochemical separations. A more recent text, Nuclear and Radiochemistry:
Fundamentals and Applications (Lieser 2001), discusses radioelements, decay,
counting instruments, nuclear reactions, radioisotope production, and activation
analysis in detail. It includes a brief chapter on the chemistry of radionuclides and
a few pages on the properties of the actinides and transactinides.
Radioanalytical chemistry methods have been published for a wide variety
of samples. Compilations of methods developed in the United States during
World War II were published in the National Nuclear Energy Series (Coryell and
4 Bernd Kahn

Sugarman 1951). A set of monographs published by the National Research Council

over a period of several years, entitled The Radiochemistry of [Element] (NAS-
NRC 1960a), traverses the entire periodic table. Another set (NAS-NRC 1960b)
of monographs is on radiochemical techniques. Laue and Nash (2003) edited
symposium presentations of recently developed chemical and radiation detection
methods, together with overviews of historical developments and current needs.
New methods were published in the United States in the journal Analytical
Chemistry and the Journal of the American Chemical Society, and corresponding
publications in other countries. From 1949 to 1986, Analytical Chemistry issued
biennial reviews of nucleonics articles; a similar set of reviews in the field of water
chemistry also held articles of interest to the radioanalytical chemist.
Journals such as Radiochemistry, Radioactivity and Radiochemistry, and the
Journal of Radioanalytical and Nuclear Chemistry currently are vehicles for pub-
lishing radioanalytical chemistry methods. Groups such as the American Society
for Testing and Materials (ASTM) and the American Public Health Association
(APHA) publish standard methods for radiochemistry (see Section 6.5), among
other topics.
A manual devoted to radioanalytical chemistry methods was published by
the U.S. Environmental Protection Agency (EPA 1984) and another one was
updated periodically by the Environmental Measurements Laboratory of the
U.S. Department of Energy (Chieco 1997). An intergovernmental task group re-
cently prepared the Multi-Agency Radiological Laboratory Analytical Protocols
(MARLAP) manual (EPA 2004) to guide radiation survey and site investigations,
this detailed work addresses many aspects of radioanalytical chemistry and is
on the Internet at www.eml.doe.gov/marlap. Additional sources of methods are
the manuals published at major DOE facilities—Oak Ridge National Laboratory
(ORNL), Savannah River Site (SRS), Los Alamos National Laboratory (LANL),
and Idaho National Engineering Laboratory (INEL), among others. Each has a set
of manuals to describe every method employed at the site, including radioanalytical
chemistry methods.
No modern textbook is complete without the use of electronic information
sources to augment hard-copy sources. Online resources have several advantages:
they are always accessible (never closed and rarely restricted), unfettered by the
bounds of printing (and so usually fairly extensive in content and illustration),
and generally up to date (if maintained by professionals). This format, however,
is impermanent and subject to change with time. Online references are current as
of the viewing date noted with each Web address cited, but the Web site content
may change, the address may change, or the site may not be maintained.

1.3. Radioanalytical Chemistry Program Elements

This textbook was written to build the knowledge base that a competent radioana-
lytical chemist should possess. From these chapters, the student should be able to
attain the knowledge necessary to perform skillfully the following duties:
 1. Introduction 5

r Operate the radioanalytical chemistry laboratory safely and in accord with reg-
ulations, with assured quality and cost-effectiveness;
r Plan and perform identification and measurement of radionuclides through a
combination of chemical separation and radiation detection methods;
r Select sample size, sample processing, radiation detection instruments with pe-
ripherals, and measurement period to match detection sensitivity specifications;
r Identify radionuclides by their chemical behavior and radiation, and resolve
ambiguities by expanding or revising separations and measurements;
r Calculate radionuclide concentration values and their uncertainty with consid-
eration of the radionuclide decay scheme, sample processing losses, radiation
detection efficiency, and interfering radiation.

These integrated activities require both professional knowledge and hands-on

experience. Hence, this textbook is designed for a radioanalytical chemistry lecture
course and an associated laboratory course. A laboratory manual and an instruc-
tor’s guide are in preparation to support the text. The prerequisite study program
should include (1) analytical chemistry lectures and associated wet-chemistry
laboratory and (2) nuclear physics lectures and associated radiation detection
laboratory.

1.4. Use of This Text

The book is intended as a guide to the functioning radioanalytical chemistry lab-
oratory. It presents a one-semester course at the senior or graduate level to train
students to work in, manage, or interact with a laboratory devoted to radionuclide
measurement. Such a laboratory has two components: the wet laboratory to pre-
pare samples for measuring radionuclides, and the counting room to perform the
measurements. The samples being considered here are primarily for environmental
monitoring in media such as air, water, vegetation, foodstuff, wildlife, soil, and
sediment. Samples may also be for bioassay, process control, and various research
and application uses.
Nuclear physics as applied to radionuclide identification and measurement is
introduced in Chapter 2 to refresh the reader’s memory. Detailed information perti-
nent to the main radiation detectors routinely used in the radioanalytical chemistry
counting room is given in Chapter 8. Chapter 9 discusses radionuclide identifica-
tion by decay scheme and gives examples.
Chapter 3 presents general information on the purification processes in the
wet laboratory that underlie radioanalytical chemistry, as well as the background
information in analytical chemistry necessary to apply those processes. Informa-
tion specifically associated with the behavior of radionuclides in aspects such as
their low concentration and the effect of radiation emission is given in Chapter 4.
Chapter 5 describes the form of the usual samples submitted to the radioanalytical
chemistry laboratory and the treatment of various sample matrices to indicate the
context in which analytical results are considered.
6 Bernd Kahn

The chemistry work in the radioanalytical laboratory is described in Chapters 6

and 7. Chapter 6 discusses the radioanalytical chemistry process from initial sample
preparation to complete purification and gives examples of separation methods for
the commonly encountered radionuclides. Chapter 7 describes preparation of the
counting source that is submitted for the measurements by radiation detection
instruments presented in Chapter 8.
The next four chapters are devoted to various aspects of data interpretation, data
presentation, and quality assurance. Chapter 9 considers interpretation of data for
radionuclide identification by decay scheme. Chapter 10 reviews the important
topics of data calculation, measurement uncertainty, data evaluation, and report-
ing the results. Chapter 11 describes the quality assurance plan that must govern
all laboratory operations. Chapter 12 discusses methods diagnostics to correct
the analytical and measurement problems that can be expected to plague every
laboratory.
The following two chapters address laboratory operation. Chapter 13 describes
laboratory practice, design, and management in terms of a model radioanalytical
chemistry laboratory and its staffing, maintenance requirements, and costs of oper-
ation. Chapter 14 discusses the practice of laboratory safety and the management
elements that reinforce the safety culture. Although the placement of these chap-
ters in the text is intended to maintain learning continuity, their contents should be
considered at the very beginning of laboratory practice.
The last three chapters consider special aspects of radioanalytical chemistry
that have become increasingly important and visible. Chapter 15 describes the
automated systems that are used to measure radionuclides in the counting room
and in the environment. Chapter 16 is devoted to identification and measurement of
the radionuclides beyond the actinides. These are research projects at the cutting
edge of radiochemistry that apply novel rapid separations in order to measure
a few radioactive atoms before they decay. Much must be inferred from limited
observations. In Chapter 17, several versions of mass spectrometers combined with
sample preparation devices are described. The mass spectrometer, applied in the
past as a research tool to detect a small number of radioactive atoms per sample,
is now so improved that it serves as a reliable alternative to radiation detection for
radionuclides with half-lives as short as a few thousand years.
In total, it is anticipated that these chapters will allow the student to begin to
understand the practice of radioanalytical chemistry. The authors wish you a good
journey and look forward to your success.
2
Radiation Detection Principles
MOSES ATTREP, JR.1,2

2.1. Introduction
As noted in Chapter 1, a background in analytical chemistry is essential to the
practice of radioanalytical chemistry. The parallels between the two fields will
become evident as analytical and radioanalytical chemistry principles are covered
in Chapters 3 and 4, and culminate in the discussion of applied radioanalytical
chemistry in Chapter 6. However, the theoretical underpinnings of the two disci-
plines are markedly different.
The primary distinction between analytical chemistry and radioanalytical chem-
istry is the nature of the transformations being examined. The analytical chemist
is concerned with chemical transformations, brought on by the interaction of
an atom’s valence electrons with its physical environment. The radioanalytical
chemist, on the other hand, is primarily interested in the nuclear transformation
of a given atom. For practical purposes, the physical environment of the atom has
no effect on the nuclear event. Consequently, many of the instrumental methods of
detection most widely utilized in the normal course of analytical characterization
have little use in the radioanalytical laboratory.
Instead, the radioanalytical chemist focuses on the detection of radiation, the
by-product of a nuclear transformation. The analyst must understand the types
of radiation that may be encountered and the way that each interacts with matter.
With this knowledge, the analyst can adapt the method of detection to the particular
radionuclide of interest. The goal of this chapter is to provide a brief review of
nuclear chemistry as it relates to the principles of radiation detection. Next, an
overview of the operating principles of commonly used detectors is provided as a
basis for understanding the material presented in Chapter 8.

2.2. Radioactive Decay and Types of Radiation

When the unstable nucleus of a radioactive isotope undergoes a nuclear transfor-
mation, i.e., decays, a new atomic species is formed with the concomitant emission

1
344 Kimberly Lane, Los Alamos NM 87544.
2
Los Alamos National Laboratory (retired), Los Alamos, NM 87545.

7
8 Moses Attrep, Jr.

TABLE 2.1. The three common types of radiation emitted from unstable nuclei
Types of radiation Symbol Description
Alpha α A helium-4 nucleus composed of two protons and two neutrons;
mass is approximately 4 Da; charge +2; no spin
Beta β An electron; mass ∼1/1822 Da; charge −1 or +1; spin 1/2
Gamma γ Electromagnetic radiation; no mass; no charge

of one or more forms of radiation. The three basic types of radiation are alpha (α),
beta (β) and gamma (γ ) radiation (see Table 2.1). Three primary modes of nuclear
decay are alpha-particle emission, beta-particle emission, and gamma-ray emis-
sion. Above atomic number 90, spontaneous fission (SF), a natural decay mode in
which the nucleus spontaneously divides into two large fragments, becomes an in-
creasingly important mode of decay. During all of these processes, mass, charge,
and energy must be conserved. Thus, the sum of the mass numbers A and the
atomic numbers Z of the products (or daughters) must equal those of the initial ra-
dionuclide (the parent). Additionally, the mass of the parent must equal the masses
of the daughter and emitted particles plus mass equivalents of the kinetic energy
of the products. The known radioactive nuclides and their radiation are given in
tables and charts of nuclides together with the stable isotopes (Firestone 1996).
The specific mode of decay listed in Table 2.1 depends upon the nature of
the parent’s instability relative to the lower energy states to which it may decay,
and functions to transform the radioactive nucleus into a more stable nucleus.
Alpha emission (α) occurs in the neutron-deficient lanthanides and is the major
mode of decay in many heavy radionuclides of the actinides and transactinides.
Beta radiation, which can take the form of either negatron (β − ) or positron (β + )
emission, results from an imbalance in the neutron-to-proton (N /Z ) ratio of the
parent and occurs in the isotopes of elements throughout the periodic table. Gamma
radiation (γ ) is the result of an excited nucleus de-exciting to lower energy levels.
A fourth mode of decay is SF, where the parent nucleus fissions into two large,
neutron-rich fission fragments (FF). The FF may be stable or may de-excite by
neutron emission to lower mass nuclei and by β − emission to higher atomic number
nuclides until further decay is no longer energetically possible.
Each of these decay process is explained in more detail in the following sections.
All processes described below are exoergic; i.e., they give off energy.

2.2.1. Alpha-Particle Decay

Alpha decay is characterized by the emission of an alpha garticle, which is equiv-
alent to a 4 He nucleus. This mode of decay decreases the original nucleus by two
protons and two neutrons for a total mass loss of about 4 Da (or atomic mass units,
amu), and a reduction in nuclear charge by 2. That mass and charge is carried away
by the alpha particle, as shown in Eq. (2.1):

A
ZX → A−4
Z −2 Y + 42 α (2.1)
 2. Radiation Detection Principles 9

Because the momentum and energy evolved in the decay must be conserved,
they are distributed between the product nucleus (sometimes called the recoil
nucleus) and the emittedg alpha particle. The unit of energy commonly used in
describing nuclear decay and radiation is the electron volt and its multiples. One
electronvolt equals 1.6 × 10−19 J, which is numerically equal to the electron charge
e in coulombs.
Given the energy difference Q between isotopes X and Y , the kinetic energy of
the alpha particle is QY/(Y + α) (where Y and α are masses) and the kinetic energy
of the recoiling isotope Y is Qα/(Y + α). The lighter alpha particle is quite ener-
getic between 2.5 and 9 MeV, although more commonly between 4 and 6 MeV. In
contrast, the recoil energy of the heavier product nucleus typically is about 0.1 MeV.
A radionuclide may emit several alpha-particle groups of different energies and
intensities, but every alpha particle in a given group is of the same energy. As shown
in the example of 241 Am in Section 9.3.7, the most energetic alpha-particle group
decays to the ground state. Less energetic groups decay to states slightly more
energetic than the ground state and then decay promptly by conversion electron
(CE) and gamma-ray emission (Section 2.2.3) to the ground state.

2.2.2. Beta-Particle and Electron-Capture Decay

Negatron (β − ) emission occurs when a nucleus has an excess of neutrons with
respect to protons, as compared to the stable nucleus isobar (same A value, where
A = N + Z ). This type of transition in effect converts a neutron to a proton.
Conversely, positron emission (β + ) in effect converts a proton to a neutron to
attain greater stability. It occurs when the radionuclide neutron-to-proton ratio is
deficient as compared to the stable nucleus of the same A value. These processes
are represented in Eqs. (2.2) and (2.3):
A
ZX → ZA+1 Y + ν + β − (2.2)
+
A
ZX → ZA−1 Y +ν+β (2.3)

The ν represents the antineutrino; ν is the neutrino. Neutrino and antineutrino

emissions serve to balance the energy and rotation before and after decay. Neutrinos
have no charge and little mass; as a result, they interact to a vanishingly small degree
with matter and are difficult to detect without elaborate apparatus. The neutrino (or
antineutrino) must be included in the decay equation to conserve energy, angular
momentum, and spin. The neutron, proton, beta particle, and neutrino all have a
nuclear spin of 1/2. A fuller discussion of this topic is in nuclear chemistry texts
such as Choppin et al. (1995).
The neutrino and antineutrino groups carry somewhat more than half of the
decay energy, while the beta-particle group carries somewhat less than half. The
kinetic energy of the product nuclide is very small because of the several-thousand-
fold smaller beta-particle mass. This recoil energy, in the range of a few electron
volts, nevertheless may cause chemical change such as displacement of an atom
from a crystal lattice.
10 Moses Attrep, Jr.

A radionuclide can decay by emitting one or several energy groups of beta

particles. The beta-particle energy in a given group ranges from nearly zero to the
maximum energy E max . The value of E max is characteristic of each radionuclide
and is generally between a few kiloelectron volts and 4 MeV.
Positron emission occurs only when the energy difference between the parent
radionuclide and the products exceeds 1.02 MeV (the energy equivalent of the
sum of the masses of an electron and a positron). The atom’s recoil, as for beta-
particle emission, is a few electron volts. At lesser energy differences, a proton
in the nucleus can be converted to a neutron by electron capture, i.e., the capture
by the nucleus of an atomic electron from, most probably, an inner electron shell
(see discussion below of CEs). The process of electron capture parallels positron
emission and may occur in the same isotope. It is accompanied by emission of a
neutrino and characteristic X rays due to the rearrangement of atomic electrons.
Electron capture may also be signaled by the subsequent emission of gamma rays.
Examples of these decays are given in Sections 9.3.4 and 9.3.6.

2.2.3. Gamma-Ray and Conversion Electron Emission

Gamma-ray emission follows the above-mentioned modes of decay when the decay
leaves the product nucleus in an excited state. The nucleus is capable of further
de-excitation to a lower energy state by the release of electromagnetic energy. No
change in N or Z accompanies this release, only a change in the energy and spin
of the nuclide. The excited nucleus is said to have a significant lifetime when it
remains excited for longer than a nanosecond (10−9 s) without releasing its energy.
In this case, it is called a metastable nucleus, and marked with an “m” as in 238m U.
The state-to-state transition energy for gamma-ray emission ranges from a few
kiloelectron volts to generally less than 3 MeV. The energy of the gamma ray is dic-
tated by the nuclear structure of the emitting radionuclide, and thus is characteristic
of that radionuclide.
An excited nucleus may also de-excite by emitting a conversion electron (CE).
This is not a two-step mechanism, where a gamma-ray leaves the nucleus and then
transfers energy by “kicking” the electron out of an atomic orbital. Instead, electron
location probability calculations indicate that orbital electrons are on occasion
located inside the nucleus. When an orbital electron encroaches on the nuclear
space, the excitation energy of the nucleus may be transferred not to a gamma ray
but to that electron, in a process called internal conversion. The resultant CE is
ejected with a kinetic energy equal to the excitation energy of the nucleus minus
the binding energy of the electron.
Conversion electron fractions ε K , ε L , etc., are tabulated as the fractions from
the respective K, L, etc., atomic shells relative to the number of gamma rays of
the pertinent energy (Firestone 1996). The CE and gamma-ray emissions are often
observed together, especially in the decay of high-Z radionuclides and those that
emit weaker gamma rays. Little CE emission occurs in low-Z radionuclides or
with energetic gamma rays because the CE/gamma-ray fraction is proportional to
Z 3 and inversely proportional to Q, the energy of the transition process.
 2. Radiation Detection Principles 11

Ejection of a CE (as well as the process of electron capture) leaves behind an

inner shell electron vacancy. To fill this vacancy, an electron from an outer shell
may fill the inner shell (moving from, say, an L shell to the K shell). X-ray emis-
sion commonly accompanies this rearrangement. The X-ray energy reflects the
difference in binding energies of the two shells involved in the transition. Several
transitions are usually possible (L3 to K, L2 to K, etc.), so that X rays of several
different energies appear, with designations such as Kα1 , Kα2 , and Kβ . Essentially
instantaneously, the electron vacancy is filled by a cascade accompanied by X
rays with ever-lower energies. The energy depends on the electron structure of the
individual element, so that each X-ray energy is characteristic of the element from
which it is emitted. The X-ray energies observed for each element are tabulated
(Firestone 1996) together with their fractional emissions.
Auger electron emission is an alternative to X-ray emission. The process consists
of one electron from the outer shell filling the inner shell, but instead of X-ray
emission, a second electron, the Auger electron, is emitted from the outer shell.
It has a kinetic energy that is equal to the difference in binding energies for the
two shells minus the additional binding energy of the Auger electron. This leaves
behind two electron vacancies in the higher energy shell that are, in turn, filled
instantaneously.
X-ray emission and Auger electron ejection are competitive processes, espe-
cially in low-Z elements. As the Z value of the material becomes higher, X-ray
emission becomes the more likely process. This is reflected in tabulations of fluo-
rescence yield, which is the ratio of X-ray transition to Auger electron plus X-ray
transitions; this value increases with Z from 0.01 at fluorine to 0.97 at and above
polonium (Firestone 1996).

2.2.4. Spontaneous Fission Decay

Decay by SF has been observed in the elements heavier than thorium (Z = 90). At
californium (Z = 98), SF begins to compete favorably with alpha-particle emission
as a mode of decay, and becomes the primary decay mode for many of the higher
atomic number actinides and the transactinides. As with all fission, SF releases
neutrons. Unlike the induced fission process described in Section 2.3.2, SF takes
place without addition of energy. Radioisotopes for which SF is an important decay
mode may be used as neutron sources.
The SF process that results in two nearly equal mass fragments (a process called
“symmetric fission”) has been observed in 259 Fm (1.5 s). More commonly, SF
occurs as “asymmetric fission,” a split of the parent radionuclide into two unequal
large FF. As in neutron-induced fission, many different asymmetric mass (and
charge) divisions with varying yields can result, with mass numbers from about 70
to 170, each with many isotopes. Hundreds of different nuclides can be produced.
Figure 2.1 displays the predominantly asymmetric mass yields as a function of
mass number (dubbed “mass–yield curves”) that have been measured for several
SF and neutron-induced fission nuclides.
12 Moses Attrep, Jr.

FIGURE 2.1. Mass–yield curves for some fission products, including 235 U and 239 Pu. The
curve labels refer to the compound nuclei, as in 235 U + 1 n = 236 U. (From Vertes et al. 2003,
p. 222.)

2.2.5. Radioactive Decay

Radioactive decay is a random first-order process described by
dN
= −N λ (2.4)
dt
where dN /dt is the disintegration rate of a given radionuclide in which dN is the
change in the number of its atoms (N ) that undergo nuclear transformation in the
time (t) interval, dt. The proportionality constant λ is the decay constant; it is in
units of inverse time, such as s−1 . The solution to this differential equation is
Nf = N0 e−λt (2.5)
The subscripts f and 0 respectively represent the final and initial number of atoms.
As shown in Eq. (2.4), the disintegration rate is proportional to the number of
atoms; hence the equation can be written as
   
dN dN
= e−λt (2.6)
dt f dt 0
If the rate is expressed as activity A in units of net disintegrations per unit time,
then successive measurements performed in the same detector without any change
in detection characteristics or sample location can be related by
Af = A0 e−λt (2.6a)
The equation defines a straight line in semilogarithmic coordinates with slope −λ,
 2. Radiation Detection Principles 13

FIGURE 2.2. Radioactive decay. The ordinate is in count rate and the abscissa is the time in
hours. (From Vertes et al. 2003, p. 263.)

as displayed graphically in Fig. 2.2. Note that the two straight lines pertain to two
radionuclides that are measured in the same sample. The curves may be plotted
in terms of the net count rate, RG – RB , shown in Eq. (8.1) as the measured count
rate from which the background count rate has been subtracted, if the counting
efficiency—the number of observed counts per disintegration—remains the same
for all measurements. The curved line in Fig. 2.2 is the total net count rate, the
sum of the two individual net count rates.
Measurements of the slope of the line in Fig. 2.2 or the disintegration rate at
two separate times can be used to calculate the half-life t1/2 , i.e., the time period
during which the radionuclide decays to one-half of a previous value. The half-life
is ln 2/λ, i.e., t1/2 = 0.693/λ.
The disintegration rate can be determined from the measured values of radionu-
clide mass m in grams and the half-life t1/2 in seconds. Equation (2.7) relates the
decay rate to the mass in terms of Avogadro’s number Av of 6.02 × 1023 atoms
per mol and the isotope mass number A in g/mol:
dN 0.693mAv
= (2.7)
dt At1/2
The mass at radionuclide disintegration rates commonly encountered in the lab-
oratory is extremely small: at a typical disintegration rate of 1 disintegration s−1
or Bq, the mass may be about 1 × 10−15 g. Weighable amounts of radionuclides
occur for radionuclides that have half-lives of billions of years. For such long-
lived radionuclides, it has been commonplace to calculate the disintegration rate
from the known half-life and measured mass, or the half-life from the measured
14 Moses Attrep, Jr.

mass and disintegration rate. Recently, masses have been measured by mass spec-
trometer for radionuclides with half-lives of as short as few thousand years (see
Chapter 17).
In some instances, a radionuclide decays not to a stable nuclide but to a second
radionuclide (or even an entire chain of radionuclides) before a stable nuclide is
finally reached. In that case, when the second radionuclide is separated from the
first, it immediately begins to grow into the first radionuclide while decaying in the
separated portion. For two successive radionuclides, the disintegration rate of the
daughter (subscript 2) is related to the disintegration rate at the time of separation,
t = 0, of the parent (subscript 1) by
       
dN dN λ2 −λ1 t −λ2 t dN
= (e −e )+ (e−λ2 t )
dt 2 dt 1(t=0) (λ2 − λ1 ) dt 2 (t=0)
(2.8)

The equation indicates that for a daughter that has a much larger value of λ than
its parent, i.e., a short-lived daughter of a long-lived parent, the second part of
the first term on the right approaches 1 and the third part approaches (1 − e−λ2 t ).
This is called secular equilibrium; i.e., the daughter reaches the same disintegra-
tion rate as the parent (see Fig. 9.10). The second term on the right describes
the decay of the separated daughter. Two other distinctive cases can occur (see
Section 9.3.1). Transient equilibrium refers to the value of λ2 that is only some-
what larger than that of λ1 , so that the daughter disintegration rate after some
time exceeds at an almost constant ratio the disintegration rate of the parent (see
Fig. 9.11). In the case of no equilibrium, λ2 is less than λ1 ; in this scenario, the
daughter decays at a slower rate than the parent after initial ingrowth, as depicted
in Fig. 9.12.
Calculations for longer decay chains under some conditions can be simplified
by assuming that short-lived daughters and long-lived parents have the same dis-
integration rate. In some complex chains, the Bateman equation (in the same form
as Eq. (2.8), but with terms added to describe further decays) can be used to de-
termine the ingrowth and decay pattern of three or four successive radionuclides
(Evans 1955).

2.3. Formation of Radionuclides by Nuclear Reactions

Some primordial radionuclides are still with us because their half-lives are near or
exceed 1 billion years or because they are shorter-lived progeny of these long-lived
radionuclides. All other radionuclides are formed by reaction of a nucleus with
atomic or subatomic particles or radiation. Reactions with high-energy cosmic rays
form some radionuclides in nature. Others are man-made, mostly in accelerators,
nuclear reactors, and nuclear explosions. The hundreds of such radionuclides are
shown in the Chart of Nuclides (Parrington et al. 1996) on both sides of the stable
elements and at masses heavier than the stable elements.
 2. Radiation Detection Principles 15

2.3.1. Energy Requirements for Nuclear Reactions

Common nuclear reactions are viewed as an interaction of a subatomic particle
such as a neutron or proton with the nucleus of an atom to form a compound
nucleus in an excited, or higher energy, state. One or more reactions then occur
that result in a different nucleus and emission of a nuclear particle or a gamma
ray, or even nuclear spallation to form two smaller atoms. The path of the reaction
depends on the energy available during the reaction as well as the energy states in
the nucleus. The reactions must conserve mass, energy, momentum, spin, nucleon
number, and charge. An example is the following reaction:
27
Al + 1 H → 27
Si + 1 n + Q (2.9)
Aluminum-27 reacts with a proton (1 H or p) to form silicon-27 and a neutron.
The symbol Q represents the energy of the reaction, in million electron volts. The
reaction may be written in a shorthand format as 27 Al(p,n)27 Si. The 13 protons of
aluminum plus 1 proton for 1 H yield a total of 14 protons on the left side of the
equation, and silicon has 14 protons on the right side of the equation. A total of
28 protons plus neutrons are on both the left and right sides of the equation, thus
balancing the reaction.
The energy Q related to the nuclear reaction is determined from the differences
in the masses M of the reactants and the products converted to million electron
volts so that, for the example reaction, Q = [M27 Al + M1 H − (M27 Si + M1 n )] ×
931.5. The masses are expressed in atomic mass units as neutral atoms and the
conversion factor is 931.5, in units of million electron volts per atomic mass unit.
A more convenient calculation is to use, instead of M, the commonly tabulated
mass excess or defect . The quantity  is the atomic mass minus the mass
number (A) for the nuclide, expressed in million electron volts. These quantities
for the individual reactants and products can be substituted in the calculation
of Q. For this example, Q = (27 Al) + (1 H) − (27 Si) − (1 n) = −17.194 +
7.289 + 12.385 − 8.071 MeV = −5.591 MeV. The negative value of Q shows
that the kinetic energy of the proton is required for the reaction.
Before the proton energy can be specified, the energy required to overcome
the coulombic barrier at the nucleus of the 27 Al target must be estimated. The
maximum repulsion that the proton will experience is when the distance between
the center of the proton and the center of the 27 Al is at the minimum defined by the
radii. The coulombic barrier is estimated to be about 3.1 MeV. Hence, the energy
of 5.591 MeV calculated above is sufficient to overcome the coulombic barrier. A
momentum correction is needed to adjust for dissipation of the projectile kinetic
energy by both products. The required energy of the proton must be at least 5.591
(28/27) = 5.8 MeV for the reaction to occur with reasonable yield.

2.3.2. Production of Radionuclides

The rate of production of atoms by induced reactions is proportional to the number
of target atoms, the rate of incoming projectiles, and the probability that the reaction
16 Moses Attrep, Jr.

will occur. For a thin target irradiated by a well-defined particle beam, the number
of target atoms is expressed by the term nx, where n is the number of atoms of
target nuclei per cubic centimeters and x is the thickness of the target material in
centimeters. The beam intensity is I , the number of incoming particles per second.
The probability of the reaction is given by the cross section with symbol σ and
units of centimeter square per atom. The cross section is tabulated in units of barns
(b), where 1 b is 10−24 cm2 .
The rate of production, P, in s−1 of atoms by reactions such as that shown in
Eq. (2.9) is
P = nxIσ (2.10)
In the case of neutron activation in a nuclear reactor, the rate of formation of
atoms is given by
P = N Φσ (2.10a)
where Φ is the neutron flux expressed in neutrons s−1 cm−2 and N is the number
of atoms in the sample, i.e., the product mAv /A in Eq. (2.7).
At constant irradiation intensity, the accumulated number of product atoms is
the integral of P over the time of production. In a simple case, the result is Pt.
It is less than this value when the projectile flux applicable to this reaction is
depleted significantly by parallel reactions; the number of target atoms decreases
significantly as a result of the reactions; and the number of product atoms decreases
significantly because of radioactive decay or further nuclear interactions.
If the produced atom is radioactive, the rate of radionuclide production in terms
of the disintegration rate [shown in Eq. (2.4)] is Rλ. The disintegration rate of the
accumulated atoms, balancing the production and decay rates, is then
dN
= N σ (1 − e−λt ) (2.11)
dt
in units of disintegrations per second.

2.3.3. Sources of Neutrons

The most common example of neutron activation is the capture of a neutron by
a nucleus, followed by the emission of a gamma ray, designated as (n,γ ). This
reaction produces an isotope of the target nuclei that has a mass number increased
by 1. Competing reactions can occur, such as (n,2n), (n,p), and (n,α), that depend
on the energy of the neutron and the relative stability of the products. Neutrons exist
in variety of energy ranges, from slow thermal neutrons (0.025 eV) to the high-
energy neutrons produced in cosmic showers and accelerators (about 14 MeV).
Listed below are some typical neutron sources and their associated energy range.
r A low-Z element (e.g., Be) mixed with a high-specific activity alpha-particle
emitter (such as 226 Ra or 239 Pu) produces neutrons by 9 Be(α,n)12 C with an
energy range of 3–5 MeV. The neutron generation rate depends on the amounts
of low-Z element and alpha-particle emitter and is relatively low.
 2. Radiation Detection Principles 17

r Some spontaneously fissioning radionuclides, such as 252 Cf, produce 3–4 neu-
trons per fission with an energy spectrum of average energy of about 2–3 MeV.
The neutron generation rate depends on the amount of 252 Cf, which emits ∼(2–
3) × 1012 neutrons s−1 g−1 .
r Accelerators provide a variety of nuclear reactions for production of neutrons.
Cockcroft–Walton accelerators can generate 14.8 MeV neutrons by accelerating
deuterons (2 H) onto a tritium target to produce 108 –1011 neutrons s−1 . Cyclotrons
and linear accelerators can produce high-energy neutrons with a broad spectrum
of energies in spallation reactions that result from the bombardment of heavy
elements by charged particles.
r Nuclear reactors are the most common source of neutrons. Inside the reactor, a
sustained nuclear reaction of fissile material produces fast neutrons. When 235 U
is used as the reactor fuel, 2–3 MeV neutrons are produced, along with other
neutrons at other energy ranges. A neutron moderator slows the fast neutrons to
reduce their energies to the thermal level. This is required to continue the chain
reaction by further absorption of neutrons by surrounding atoms of 235 U. Other
neutron energy ranges are the epithermal between 0.1 and 1 eV and resonances
between 1 eV and 1 keV.

2.3.4. Nuclear Fission

As discussed in Section 2.2, SF is a “natural” nuclear decay mode. Nuclear fission
can also be induced in numerous nuclides by irradiation with projectiles such as
neutrons, protons, and deuterons.
The fission process can be described in terms of the liquid drop model. The drop
is capable of undergoing various deformations considered as vibrational modes. As
the drop becomes more distorted, it eventually breaks into two primary fragments.
The two fragments are usually unequal in mass, with a mass distribution that
depends on the manner in which the drop splits. In nuclear fission, absorption of a
neutron induces similar oscillations in the target radionuclide that distort its shape
until it splits into the two primary FF. A wide array of possible mass combinations
exists for the FF. Shown below are three such possibilities for 235 U.
1
0n + 235
92 U → 142
54 Xe + 90
38 Sr + 4 10 n
1
0n + 235
92 U → 139
56 Ba + 94
36 Kr + 3 10 n (2.12)
1
0n + 235
92 U → 144
55 Cs + 90
37 Rb + 2 10 n

The sum of the atomic masses of the heavy fragment, the light fragment, some
additional free neutrons, and the released energy equals the mass of the fissioning
atom. The Q value for such neutron-induced fission reactions is in excess of
200 MeV per fission.
For a given radionuclide, some fission-fragment pairs are more common than
others. This is easily seen in the mass–yield curves of Fig. 2.1, which show the spec-
trum of atomic masses given off by the fission of several radionuclides. Atom fission
18 Moses Attrep, Jr.

yields range up to 6–7% for an individual fission product at the heavy and light
mass peaks. These peaks represent the mass fragments that are most common when
the parent nucleus splits. The particular shape of the spectrum depends on both the
identity of the fissioning nucleus and the energy of the projectile. Fission product
yields, in the form of tables that give the yield of each fragment nuclide at each
mass number (from approximately 66 to 171), have been compiled over the course
of several decades, and are available online at http://ie.lbl.gov/fission/endf349.pdf
(January 2006); these are also published in a series of reports by England and
Rider (1994).
Fission products are neutron-rich and thus are negatron emitters. At each mass,
a decay series may consist of as many as five radionuclides that decay, one into the
other, until the chain stops at a stable element. In a decay series, the fission yield
may begin at a smaller value for the initial short-lived radionuclides and increase for
subsequent radionuclides to the maximum shown in the figure. Among these fission
products are ones with half-lives from days to years that are readily measured for
monitoring nuclear reactors and nuclear weapon tests.
The fission process releases the large amounts of energy used in nuclear reactors
and weapons and also several neutrons. The number of these neutrons depends on
both the process and the fissioning atom, as indicated in the 235 U decay equations
above. When the number of neutrons exceeds one per fission and the energy of the
neutrons is suitably moderated to induce further fission, a chain reaction is induced
that can be used for energy production.

2.4. The Basis for Detecting Radiation

Detection of radionuclides based on their radiations—alpha, beta, or gamma—
commonly occurs by one of two processes. Electrons bound to the atoms or
molecules of the material through which charged particles such as alpha and
beta particles pass are released in a process called ionization. The separated elec-
trons and ions can be collected at two electrodes by imposing a potential dif-
ference across the detector space; their presence is then measured as pulses or a
current.
Alternatively, the transfer of the radionuclides’ energy to these bound electrons
raises them to an excited state in the atom or molecule. When the excited species
returns to its ground state energy level, the excited atom or molecule may emit
electromagnetic energy in the ultraviolet to visible (UV/Vis) region. This light can
be detected by a photomultiplier tube (PMT) in a scintillation counter system.
Less common detection of electrons is by photons emitted during their decelera-
tion in a medium. The two mechanisms are creation of bremsstrahlung (continuous
X-ray spectrum) and Cherenkov radiation (visible light). Cherenkov radiation is
detected with a PMT; bremsstrahlung X rays are detected as discussed below for
gamma rays.
Detection of gamma or X rays by ionization or excitation is a secondary process
where the first step is the formation of a free electron by interaction with matter.
 2. Radiation Detection Principles 19

A less common process is the creation of a positron–electron pair, as discussed

below.

2.4.1. Alpha Particles and Ion Formation

The energy per ion formation, w, is a measured value that has been tabulated for
many gases (ICRU 1979). Average values are 35 eV in air and 26 eV in argon; w
is in the same energy range for many other gases. Hence, a 5-MeV alpha particle
theoretically has enough energy to create 192,000 ion pairs in argon. That the
average pair formation energy greatly exceeds the first ionization potential for the
gas—e.g., 15.7 eV is needed to remove the first electron from argon—suggests
that some of the required pair formation energy is devoted to effects other than
ioniziation, such as the excitation discussed above.
Ionization per unit path length is not constant when a charged particle loses
energy. As the alpha-particle energy loses energy along its path, the stopping
power of the medium, defined as the average energy loss per path length, gradually
increases to a maximum as the probability of interaction of the alpha particle with
the atoms or molecules along its path increases. The peak value at about 0.7 MeV
in air is 2.0 MeV mg−1 cm−2 , compared to 0.8 MeV mg−1 cm−2 at 5.0 MeV. Below
0.7 MeV, the stopping power decreases to 0 MeV at zero energy (ICRU 1993).
The alpha particle loses most energy in interactions with electrons. Below
0.1 MeV, collision with the nucleus of the material along its track contributes
significantly to energy loss. The unit of mass per area (or distance times density) is
used instead of distance for stopping power to obtain similar values for the entire
range of elements. In fact, stopping power values increase by about an order of
magnitude as a function of the atomic mass of the element with which the alpha
particle interacts.
Because alpha particles are monoenergetic, all in a given energy group can be
expected to stop at the same distance. A small deviation occurs because the energy
of the alpha particles after multiple interactions has a normal distribution.
The range () of an alpha particle does not vary linearly with energy because
of the above-cited pattern of variation in stopping power. An empirical relation of
range to its initial energy E α is given by Eq. (2.13) (Friedlander et al. 1981):
 = k (E α )a (2.13)
The constants k and a are determined experimentally. For example, the range of
an alpha particle in air at STP (standard temperature and pressure conditions) is
given by Eq. (2.13a) (Eichholz and Poston 1979):
 = 0.31 (E α )3/2 (2.14)
where the alpha-particle energy is in million electron volts and the range is in
centimeters.
Figure 2.3 is a graph of the range calculated by the continuous-slowing-down
approximation (CSDA) of an alpha particle, in both air and aluminum. The data
that generated the graph (ICRU 1993) indicate that a 5-MeV alpha particle has a
20 Moses Attrep, Jr.

10

7
Energy (MeV)

6
Aluminum
5
Air
4
Aluminum:
3 d = 2.70 g/cm3
Air:
2 d = 1.29 g/cm3

0
0 2 4 6 8 10 12
Range (mg/cm2)

FIGURE 2.3. Alpha-particle range in air and aluminum. (Data from ICRU 1993, pp. 192
and 213.)

range of 3.4 cm (CSDA range 4.37 mg cm−2 divided by the density of air, 1.29 mg
cm−3 ) in air. Equation (2.13) gives a value of 3.47. The Bethe equation for the
energy loss of charged particles is complex but indicates that the loss is related
to the electron density, i.e., the atom density times the atomic number Z . The
empirically derived Bragg–Kleeman rule suggests that the range in one material
relative to another is proportional to ρA0.5 (Evans 1955), where ρ is the density of
the attenuating material.

2.4.2. Beta Particles and Ion Formation

At relatively low energies, the ion formation process by beta particles and CE
is similar to that of alpha particles but their range is much greater because of
the 7000-fold smaller mass and 2-fold smaller charge. For example, the stopping
power for electrons in aluminum is 1.5 keV mg−1 cm−2 at 1000 keV, reaches a
maximum of 300 keV mg−1 cm−2 at 0.1 keV, and then decrease to 0 at 0 keV.
Energy transfer is basically by electron–electron interaction, but bremsstrahlung
contributes significantly at beta-particle energies above about 500 keV. The rela-
tivistic velocity of higher energy electrons results in little change in the stopping
power between about 700 and 4000 keV, and a gradual increase in stopping power
above that energy.
As for alpha particles, the relation of range to energy of electrons (or beta
particles) is not linear. All monoenergetic electrons in a group have the same
 2. Radiation Detection Principles 21

10.00

1.00
Energy (MeV)

Polystyrene

Aluminum
0.10

Polystyrene: d = 1.06 g/cm3

Aluminum: d = 2.70 g/cm3

0.01
0.1 1 10 100 1000 10000
Range (mg/cm2)

FIGURE 2.4. Beta-particle range–energy curve (log/log) in aluminum and polystyrene. (Data
from ICRU 1984.)

range. A set of range–energy curves is shown in Fig. 2.4, with range in units of
mg cm−2 in aluminum and polystyrene. Range values in terms of this unit can
be applied to various substances within a restricted range of Z . An equation for
beta-particle range similar to the equation for alpha particles above is
 n
 = 412 E β (2.14)

where n = 1.265 − 0.0954 ln E, and E refers to the E max , maximum energy

of the beta-particle group. Simpler equations have been proposed for portions
of the range–energy curve for which linear approximations are possible (Evans
1955).
The energy spectrum of a beta-particle group takes the form of a continuum
that has different shapes for different radionuclides; three are shown in Fig. 2.5.
Because of the energy distribution of beta particles, the relationships observed for
the interaction of beta particles with matter are not as simple as those of alpha
particles.
The typical curve of beta-particle attenuation in aluminum absorbers, shown
in Fig. 2.6, at lower energies resembles the exponential attenuation observed for
gamma rays (see Section 2.4.4). The final part of the line curves downward to reach
the distinct range associated with E max . Attenuation curves for the various beta-
decay radionuclides differ because of the different beta-particle energy spectra, but
both the characteristic range and the approximately exponential attenuation have
been used to estimate maximum beta-particle energies (Evans 1955).
22 Moses Attrep, Jr.

2.5

2
Beta-distribution function

1.5
Y-90
Sr-89
S-35
1

0.5

0
0 0.2 0.4 0.6 0.8 1
(MeV)

FIGURE 2.5. Three typical beta-particle spectra. (Data from ICRU 1997, pp. 107–108.)

0.1
R/ R0

0.01

0.001
0 100 200 300 400 500
Absorber thickness (mg/cm2)

210
FIGURE 2.6. Beta-particle attenuation curve: Bi in aluminum absorbers. (Based on
Zumwalt 1950, p. 40.)
 2. Radiation Detection Principles 23

2.4.3. Bremsstrahlung and Cherenkov Radiation

These two forms of electromagnetic radiation are produced by charged particles—
electrons in the present context—that slow down in matter (“bremsstrahlung” in
German means “braking radiation”). An electron moving with decreasing velocity
in the electric field of atoms emits part of the energy associated with this decrease
in velocity as continuous X rays. The maximum-energy X ray equals the energy of
the electron (or beta particle or positron). The X-ray intensity is very small at this
maximum energy but increases exponentially with decreasing energy. The fraction
of beta-particle energy converted to bremsstrahlung is approximately (3.5 × 10−4 )
ZEmax . Thus, in the passage of a 1-MeV beta-particle group through lead (Z =
82), approximately 3% of the energy is converted to bremsstrahlung, while through
beryllium (Z = 4) it is 0.14%. Most of the X rays are at energies below 0.1 MeV.
Cherenkov radiation (named after its discoverer, P.A. Cherenkov) is bluish light
emitted forward when a very energetic charged particle, traveling in a transparent
medium at a velocity faster than light could in the same medium, slows to the
velocity of light in that medium. This situation occurs when the refractive index
of the medium is well above 1, such as 1.33 in water. A low-energy cut-off for
Cherenkov radiation decreases with the refractive index. In water, beta particles
below an energy of 0.265 MeV do not stimulate Cherenkov radiation. The number
of photons per electron is about 200 at 1 MeV and about 600 at 2 MeV (Knoll
1989). The Cherenkov light at an energy of about 2.5 eV is detected by a liquid
scintillation counter. Less than 0.1% of the electron kinetic energy is transferred
to Cherenkov radiation.
Bremsstrahlung is of interest in radioanalytical chemistry because some of the
energy of electrons stopped in detector-shielding material is converted to X rays
that can penetrate the shield. Cherenkov radiation permits scintillation counting
of radionuclides in plain water samples if the electron energy is sufficiently high
and the detection system is sufficiently sensitive.

2.4.4. Gamma Radiation Interactions

Gamma rays are electromagnetic radiation without mass and charge. They are far
more penetrating than electrons of the same energy, and much more penetrating
than alpha particles. Three major gamma-ray interactions produce free electrons
that can be detected: Compton scattering, photoelectric effect, and pair formation.
The interactions have been given respective interaction coefficients σ, τ , and κ
that define the exponential attenuation of gamma rays in matter and are used to
calculate the extent of shielding (i.e., gamma ray removal from a beam), interaction
(i.e., energy deposition), and radiation detector efficiency. The total attenuation
coefficient μ for gamma rays is the sum of the individual coefficients. The ratio of
the gamma-ray flux after attenuation, If , to that before attenuation, I0 , is a function
of the distance traveled x:
If
= e−μx (2.15)
I0
24 Moses Attrep, Jr.

Compton
electron

Incident photon Atomic

electron

Compton
scattered
photon

FIGURE 2.7. Compton scattering. (Figure from Evans 1955, p. 675.)

The units of μ can be cm2 g−1 , cm2 atom−1 , or cm−1 ; respective units of x are g
cm−2 , atom cm−2 , or cm. The interaction fraction with the medium is I0 − If ,

(I0 − If )/I0 = 1 − e−μx (2.15a)

Compton scattering occurs when a gamma ray interacts with an electron bound
weakly (i.e., an outer orbital electron) to an atom or molecule and transfers a
fraction of its energy to the electron. The resulting weaker gamma ray departs at
an angle, as illustrated in Fig. 2.7. The energy distribution between the electron and
the produced gamma ray depends on the angles at which the two move relative to
the initial gamma ray. Much or all of the energy transferred from the initial gamma
ray to the electron is deposited in the absorbing material in which the electron
slows down.
The electron energy is distributed between zero and a maximum energy that is
somewhat less than the energy of the incoming gamma ray, E γ . The maximum
electron energy is

2E γ /0.51
E max = E γ   (2.16)
1 + 2E γ /0.51

For example, if the incoming gamma-ray energy is 1.0 MeV, the maximum
electron energy is 0.8 MeV. The energy of the scattered gamma ray is the en-
ergy of the initial gamma ray minus that of the ejected electron. That gamma ray
 2. Radiation Detection Principles 25

TABLE 2.2. Photon interaction and energy absorption coefficients for common materialsa
Mass attenuation (cm2 g−1 ) Mass energy absorption (cm2 g−1 )
Energy
(keV) Water SiO2 CaCO3 Water Air Muscle Bone
50 0.208 0.282 0.485 0.0394 0.0384 0.0409 0.158
100 0.165 0.158 0.181 0.0252 0.0231 0.0252 0.0386
200 0.136 0.135 0.125 0.0300 0.0268 0.0297 0.0302
300 0.118 0.106 0.107 0.0320 0.0288 0.0317 0.0311
500 0.096 0.087 0.087 0.0330 0.0297 0.0327 0.0316
1000 0.0707 0.0636 0.0636 0.0311 0.0280 0.0308 0.0297
2000 0.0494 0.0447 0.0448 0.0260 0.0234 0.0257 0.0248
3000 0.0397 0.0363 0.0366 0.0227 0.0205 0.0225 0.0219
a Data from Shleien (1992).

usually has sufficient energy to interact further, by more Compton scattering or

otherwise.
The two solid curves in Figs. 2.8 and 2.9 show the distinction between gamma-
ray attenuation, represented by μ0 , due in part to Compton attenuation σ0 , and
gamma-ray energy absorption, represented by μa , due in part to Compton energy
absorption σa . Fractional attenuation refers to the number of gamma rays that
interact in the medium relative to the number of gamma rays that enter the medium.
Fractional energy absorption refers to the average amount of energy absorbed
per interaction relative to the energy involved in each interaction. Because some
gamma rays are scattered and carry some energy away from the point of interaction,
the energy absorption fraction is always less than the attenuation fraction, as shown
in Table 2.2.
Values from the upper solid curves (labeled “Total attenuation”) in Figs. 2.8
and 2.9 are applied with Eq. (2.15) to calculate the fractional transmission at a
thin absorber of gamma rays in a collimated beam. Equation (2.15a) is applied to
values from the lower solid curves (marked “Total absorption”) to calculate the
fractional deposition of energy from gamma rays in the absorbing material; more
generally, these calculations give the radiation dose, in energy per unit mass, for
radiation protection and material response purposes. Table 2.2 gives the values
of these coefficients for some typical materials of interest; many others are also
available.
In practical applications, when the beam is broad and the attenuating material is
thick, fractional transmission of gamma rays is greater than the values calculated
from the upper curve in Figs. 2.8 and 2.9. While the gamma rays scattered in a very
thin attenuator are removed from the beam and do not reach a point in the beam
behind the scattering material, in a thick attenuator, some of the scattered gamma
rays again are scattered and will reach the point in the beam behind the scattering
material. The fraction of increased radiation to which this point is exposed had
earlier been determined empirically for various materials shapes, and thicknesses,
26 Moses Attrep, Jr.

Sodium iodide
Sodium iodide cm2/g

sr m0 r
ss r

ma r

kr

s
tr

sa r
r
sa

s
r

a
r

s
s
r

Mev

FIGURE 2.8. Gamma-ray interaction coefficients as function of energy in NaI. (From Evans
1955, p. 716.)

and is now calculated by Monte Carlo simulation. This “buildup factor” B (Shleien
1992) exceeds 1.0 and is used to multiply If in Eq. (2.15). The buildup factor
depends on gamma-ray energy, source-shield configuration, and shield thickness.
It usually is reported in terms of the dimensionless relaxation length μ0 x.
The second mechanism by which gamma radiation loses energy is the photo-
electric effect. In this interaction, all of the energy is transferred from the gamma
 2. Radiation Detection Principles 27

Lead
Lead cm2/g

ss r m0 r

ma r

kr
tr
sr

ss

sa r
r

r
ss
r

Mev

FIGURE 2.9. Gamma-ray interaction coefficients as function of energy in Pb. (From Evans
1955, p. 717.)

ray to a strongly bound interior orbital electron. When the energy transmitted by
the gamma ray exceeds the electron binding energy E BE , the electron is ejected
from the atom or molecule with energy E e according to Eq. (2.17):
E e = E γ − E BE (2.17)
The photoelectric effect becomes more likely with lower E γ and as electrons
are more tightly bound in their shells (i.e., K > L > M). The missing electron
28 Moses Attrep, Jr.

is replaced instantaneously by another electron with the emission of an X ray or

Auger electron as described in Section 2.2.3. The electrons stopped in the absorbent
material plus the stopped associated X rays and Auger electrons deposit energy
E γ . In relatively small absorbers, when the X ray escapes without being stopped,
the energy deposited is E γ minus the energy for the escaped X ray.
The third mechanism of gamma-ray interaction is pair formation. The gamma ray
transforms itself near a nucleus into a positron and an electron. For pair formation
to occur, the energy of the gamma ray must be at least twice that of the electron
rest energy, mc2 , of 0.511 MeV, so that the energy must be equal to or greater than
1.022 MeV.
After pair formation, the electron and positron move in opposite directions so as
to conserve energy and momentum. Each has a kinetic energy of 0.5(E γ −1.022)
MeV. The moving electron causes ionization and excitation. The positron behaves
similarly until it slows down near an electron and both are annihilated, creating
two gamma rays that are directed away from each other, each with an energy
of 0.511 MeV. These two 0.511-MeV gamma rays then proceed to interact by
Compton scattering or the photoelectric effect. This annihilation process occurs
within about 0.1 cm in solid materials, but can be at a greater distance in air. Because
of the potential for interaction at a distance from the source and for Compton
scattering, the fraction of energy deposition is less than that for interaction, as
discussed for Compton scattering.
The interaction coefficients for the three modes of gamma interaction with
sodium iodide and lead are shown in Figs. 2.8 and 2.9. The photoelectric effect
declines as the energy increases, as does Compton scattering. The jagged edge in
the photoelectric effect curve occurs at the binding energy of K orbital electrons. A
fine structure that is not shown at this edge arises from the various X-ray energies
discussed in Section 2.2.3. Pair formation starts at 1.022 MeV and increases as the
energy increases. The total interaction at any given energy is the sum of the three
mechanisms. The electrons generated by these mechanisms will act as described
above for beta particles and CEs to ionize or excite molecules in systems used for
radiation detection.
Gamma rays also interact with matter by other reactions, two of which occasion-
ally may be considered. One interaction is coherent (Rayleigh) scattering at low
energies. The gamma ray is absorbed by an atom and reemitted with unchanged
energy. The direction of the gamma ray is changed due to scattering from bound
electrons. The scattered gamma ray commonly proceeds to interact by the pho-
toelectric effect. Because coherent scattering produces interference patterns, the
process is used in crystallographic analysis to make structural determinations. Its
attenuation coefficient decreases with energy and accounts for only a few percent
of attenuation below 0.2 MeV in lower Z elements, though it is still significant in
lead to 1 MeV.
Another interaction is photodisintegration, generally a (γ ,n) reaction. This re-
action can occur if photons have energies of 8 MeV or higher. Exceptions occur at
energy thresholds of 1.67 and 2.23 MeV for 9 Be and 2 H, respectively, to produce
8
Be and 1 H plus neutrons.
 2. Radiation Detection Principles 29

2.5. Detectors
2.5.1. Gas-Filled Detectors
Gas-filled detector systems collect and record the electrons freed from gaseous
atoms and molecules by the interaction of radiation with these atoms and
molecules. The systems have been classified into the primary categories of ioniza-
tion chambers, proportional counters, and Geiger–Mueller (G-M) counters. The
detectors have a variety of designs, but essentially consist of a chamber, which
serves as the cathode, and a center wire, which serves as the anode. The electri-
cal field is characterized by chamber shape and radius, the wire radius, and the
applied voltage (Knoll 1989). The chamber is filled with a gas and a potential is
applied between the two electrodes. The configuration of the system is shown in
Fig. 2.10.
Ionized electrons in the gas are collected onto the anode. Figure 2.11 shows the
response of the system in terms of the number of electrons or ions produced in
the gas when the applied voltage is increased. The curves show six regions, each
of which has different properties. The higher and lower curves pertain to particles
that deposit more and less energy, respectively, in the detector.
In Region 1, many electrons and ions produced in the gas recombine because
the voltage applied between cathode and anode is not large enough to collect all
electrons. This region is not useful for counting radiation.
Region 2 is the saturation region. The potential difference is sufficient to collect
all freed electrons. The resulting very weak current or pulse over brief periods is
measured with extremely sensitive electrometer devices. A detector working in the
saturation region is called an ionization chamber. Its output is proportional to the
deposited radiation energy. Internal or thin-window ionization chambers are used
as alpha-particle and fission-fragment detectors. External samples are measured
for beta particles and gamma rays with ionization detectors; the latter are larger
and contain counting gas at elevated pressure.

Cathode
Source Anode

Output signal

Window

FIGURE 2.10. Configuration of a gas-flow proportional counter.

30 Moses Attrep, Jr.

FIGURE 2.11. Variation of pulse height with applied voltage in a counter. (From Knoll 1989,
p. 162.)

In Region 3, the proportional region, the applied voltage is strong enough that the
electrons freed by the initial radiation are accelerated, so that they, in turn, ionize
additional atoms or molecules (secondary ionization) to free more electrons. This
electron multiplication generates an avalanche toward the anode for each primary
electron that was freed. The applied voltage domain is called the proportional
region because each avalanche is characterized by the same electron multiplication
at a given applied voltage. The output signal is directly proportional to the deposited
energy, although each pulse is many times larger than in the ionization region. As
the applied voltage is increased, the amplification increases uniformly for the entire
range of deposited energy.
A problem in proportional counters operating at these higher voltages is produc-
tion of a secondary avalanche due to molecular excitation, which interferes with
the detection of subsequent pulses. To prevent excitation that causes this secondary
avalanche, a quenching agent is added to the fill gas (see Section 8.5.1). The multi-
plication factor in this region is about 104 , which requires the measurement device
to be far less sensitive than that for the ionization chambers. The proportional
counter system has a preamplifier and a linear amplifier.
In Region 4, the proportionality of the output signal to the deposited energy at
a given applied voltage no longer applies. Amplification of the greater deposited
energy reaches its limit while that for the lesser deposited energy continues to
increase. This region is called the region of limited proportionality and is usually
avoided as a detection region.
 2. Radiation Detection Principles 31

FIGURE 2.12. Plateau regions for both alpha and beta particles in a proportional counter.

The third useful region for detection of radiation is Region 5, the G-M region.
This applied voltage is high enough that any deposited energy produces sufficient
secondary electrons to discharge the entire counting gas. The linear amplifier is
no longer needed. One no longer can distinguish between a small and a large
deposition of energy. This discharge must be quenched so that the next pulse can
be detected. Either the applied voltage must be removed briefly or quenching gases
(see Section 8.5.1) must be added.
In Region 6, an electrical discharge occurs between the electrodes. This voltage
region has been used for some purposes but generally is avoided because the
discharge can disable the detector.
Although gas-filled detectors are operated in the second, third, and fifth voltage
regions, it would be a mistake to assume that a particular detector can be used in all
of them simply by changing the applied voltage. Detector components are designed
to be used in a single voltage domain. The operating voltage then is selected at a
plateau region on the basis of the curve of count rate vs. applied voltage, shown in
Fig. 2.12.
In proportional and G-M detectors, this curve begins with zero count rate at a
relatively low voltage, reaches a plateau at intermediate voltage, and then increases
when a continuous electrical discharge occurs. The initial increasing count rate
represents the increasing number of pulses that are sufficiently amplified to pass
the lower discriminator. At the plateau, all pulses are detected. In a proportional
counter, a first plateau is reached for the much larger alpha-particle pulses, and
a second plateau at higher applied voltages is then reached for beta particles as
well as alpha particles. Detectors are operated near the middle of the plateau to
avoid erroneous data due to small fluctuations in applied voltage. In proportional
counters, the applied voltage is reduced to measure alpha particles but not beta
particles.

2.5.2. Solid-State Detectors

The operating principle for solid-state detectors is analogous to that of the gas-filled
detector except that collection at the electrodes of electrons and electron holes
32 Moses Attrep, Jr.

Radioactive source

n-type semiconductor

Depleted region

p-type semiconductor

Conducting metal

FIGURE 2.13. Schematic of a reverse bias p-n junction detector.

replaces electrons and positive ions. One advantage is the thousandfold greater
electron density in solids for the much greater stopping power that is desirable for
detecting gamma radiation. Another advantage is the 10-fold smaller value of w,
the energy absorption per ionization, which yields better energy resolution. The
solid medium currently is hyperpure germanium or silicon that has been prepared
to function as semiconductor material. A schematic of a reverse bias p-n junction
detector is shown in Fig. 2.13.
A p-type material is one with positive holes and an n-type material is one with
electrons in excess. To create these types of matrices, small amounts of impu-
rities are incorporated into some host materials. For example, if the host mate-
rial is silicon, which has four valence electrons, it may be “doped” with boron,
which has three valence electrons. This creates regions in the matrix where there
are “holes” for electrons to occupy; it becomes a p-type material. The energy
for creating an electron–hole pair is 3.7 eV in silicon and 3.0 eV in germa-
nium (Knoll 1989). Thin wafers of silicon diode surface barrier detectors that
have a very thin layer of gold on the surface are used for alpha-particle spec-
troscopy. Hyperpure germanium detectors, typically a closed-end coaxial, 6 cm in
 2. Radiation Detection Principles 33

diameter and 6 cm or more in length, are currently used for gamma-ray spectro-
scopy.
Conversely, if phosphorus or arsenic, each with five outer shell electrons, is
inserted into the silicon crystalline material, then the resulting material will have
an “excess” of electrons in the lattice, producing the n-type material. The placement
together of an n-type and p-type creates a p-n junction.
When a positive charge is applied to the n-type semiconductor and a negative
charge is applied to the p-type material, the positive holes are attracted to the neg-
ative electrode and the electrons are attracted toward the p-n junction. This creates
the depletion layer. Radiation enters the n-type side where the depleted region
serves as the radiation-sensitive volume. There, electron–hole pairs are created
that will be rapidly collected to create the pulse for amplification. Silicon semi-
conductors of this type are generally used for beta-particle and CE spectroscopy.
Surface-barrier detectors for measuring alpha particles are formed from n-type
silicon with an oxidized p-type surface. A very thin gold layer is evaporated onto
this surface to function as one of the electrodes. The sample that emits alpha
particles faces this side.

2.5.3. Scintillation Detection Systems

Certain materials that become excited by the absorption of radiation are able to
de-excite through emission of light. These scintillators—or phosphors—enable
radiation detection through the observation of flashes of luminescence. The amount
of light observed is proportional to the amount of radiation energy absorbed by
the medium. A scintillation detector utilizes a phosphor (to absorb radiation and
produce light), optically coupled to a photomultiplier tube (PMT), which detects
the light and converts it into an electrical signal. A computer then collects the
signal produced by the PMT.
The phosphor can be an inorganic crystal or an organic solid, liquid, or gas.
Materials function well as a scintillator for radiation if they efficiently absorb and
convert radiated energy to light, exhibit good transparency for the light to escape the
material, and emit a frequency of light that matches the photon-detection system.
The photons emitted by the phosphors and detected by PMTs typically are in the
2.5–3.6-eV UV/V is energy range. Some common scintillation detectors are given
in Table 2.3.
One effective medium in common use for gamma-ray measurements is thallium-
activated sodium iodide [NaI(Tl)]. The relatively high-Z iodine atom provides
a high attenuation coefficient for interacting with energetic gamma radiation. It

TABLE 2.3. Some common scintillators

Scintillation material Physical state Radiation primarily detected Conversion efficiency (%)
Anthracene Solid Beta 4.5
NaI(Tl) Solid Gamma 10
Liquid scintillators Liquid Alpha, beta 2–4
Plastic scintillators Solid Alpha, beta, gamma 3.0
34 Moses Attrep, Jr.

results in an appreciable fraction of counts in the full-energy peak, which is useful

for spectral analysis (see following section). The crystal is sealed in an aluminum
can because it is hygroscopic, with a light pipe at one end for coupling to the PMT.
The inclusion of thallium in the NaI(Tl) detector provides energy levels that
are somewhat lower than the conduction band. According to the band theory of
crystalline solids, the valence electrons commonly occupy the highest filled atomic
electron band. Above this energy is a band gap, and higher still is the conduction
band. Incoming radiation (if sufficiently energetic to surmount the band gap) can
excite the valence electrons to the conduction band; from here the electrons drop
back, de-exciting by the emission of light. The thallium impurity facilitates the
de-excitation of electrons to the valence band. The benefit of a smaller energy
gap is that the excited electron will shed its excess energy more quickly for more
efficient detector response. Second, the energy emitted will not be the same as that
which caused the initial excitation; consequently, photon emission will not cause
re-excitation (i.e., the detector is transparent to its radiation).
Some other scintillation materials, such as cesium iodide and bismuth ger-
manate, have characteristics that are less favorable than NaI(Tl) for general use, but
recommend them for some special measurements. For example, CsI and NaI(Tl)
can be combined for coincidence or anticoincidence counting by distinguishing
between output from the two detectors by their pulse shapes.
Silver-activated zinc sulfide [ZnS(Ag)] has been used since radioactivity was
first measured to detect alpha particles. It is relatively insensitive to electrons and
gamma rays because it is not transparent to its own radiation. Radiation interactions
within the detector are not recorded; only its surface, where alpha particles interact,
emits scintillations. The ZnS is doped with silver to shift its scintillations to a longer
wavelength for better PMT response.
Organic scintillators function similarly to inorganic crystals. One major differ-
ence is that impurities are not necessary to produce energy levels for de-excitation
because the selected organic substances quickly shed energy by passing through
multiple vibrational states. Organic scintillators that have resonance structures,
such as anthracene, are excellent scintillators. The effect is similar to doping; the
de-excitation photon is rarely of the same energy as the absorbed radiation, so that
the organic scintillator is transparent to its own emissions.
Organic scintillators have relatively low counting efficiency for gamma rays
because of their low-Z carbon, oxygen, and hydrogen atoms. Any spectral analysis
has to be based on Compton-edge energy measurements because of the low full-
energy peak intensity. Solid organic scintillators can be useful for beta-particle
detection because they can be exposed without a container. They can also be
machined to large volume in many shapes.
In a liquid scintillation (LS) system, the sample is mixed with a cocktail that
consists of an organic scintillator dissolved in an organic solvent. The cocktail
and the usual aqueous sample form an emulsion. The radiation emitted by the
intimately mixed radionuclide deposits its energy in the solvent, which transfers it
to the scintillator. The scintillations are then detected by the PMT. The LS counter is
useful for detecting alpha particles and low-energy beta particles from samples that
 2. Radiation Detection Principles 35

are dispersed throughout the LS detector, whereas in other detectors the radiation
would be absorbed in the sample.

2.5.4. Silicon and Germanium Spectrometers

Solid-state detectors—notably silicon for alpha particles and germanium for
gamma rays—are particularly effective for spectral analysis. Spectral analysis
is used both to identify the radiation by its characteristic energy peak and to de-
termine its activity based on the count rate within this peak. Further advantages
of restricting analysis to the narrow band of the characteristic energy peak are re-
ducing interference from other radionuclides in the sample and from background
radiation. Other materials, such as cadmium telluride and cadmium zinc telluride,
have been developed for specific advantages such as inclusion of a high-Z material,
but are not in routine use.
The detection systems for alpha-particle and gamma-ray spectrometers are dis-
tinctly different. For alpha particles, a thin silicon detector and a thin sample are
placed in a small vacuum chamber to eliminate energy loss in air. No further
shielding is needed because the chamber is a shield for ambient background alpha
particles while other ambient radiation is not detected in the thin solid to any sig-
nificant extent. For gamma rays, a large germanium detector and samples as large
as necessary are placed in a radiation shield of lead or steel.
The spectrometer systems require highly stable power supplies and amplifiers to
support analysis at high-energy resolution. The germanium detector has a cryostat
for cooling during application. A Dewar flask that contains liquid nitrogen coolant
is commonly used, but electromechanical cooling can be substituted. Various geo-
metric arrangements enable the germanium detector to view the sample vertically
or horizontally.
The detected radiation is displayed to show near-Gaussian peaks (see Fig. 2.14)
in a 1024-, 2048-, or 4096-channel spectrometer. The spectrometer is calibrated for
energy by matching the channel at the midpoint of the peak to the known energy of
gamma radiation emitted by a set of radioactivity sources. Typically, the channel
width is set to approximately one-fourth of the detector peak resolution, but some
compromise is needed because the resolution changes with energy.
The energy resolution of a recorded set of many identical pulses is the extent
to which individual values deviate from the expected single value. If a normal,
Gaussian energy distribution can describe the deviation from the expectation value
that results in the observed peaks, a simple view (see Section 10.3.4) suggests that
the standard deviation σ for this peak equals the square root of the number of
events for each peak. The number of events is E/w, where E is the deposited
energy and w is the energy required to form an electron–hole pair. The standard
deviation in units of eV is σw ; relative to the energy it is σw /E; hence

σw w E w
= = (2.18)
E E w E
36 Moses Attrep, Jr.

FIGURE 2.14. Peak resolution in gamma-ray spectra from Ge and NaI(Tl) detectors. (From
Friedlander et al. 1981, p. 259.)

The resolution of a peak is commonly described by the full width at half max-
imum, FWHM. Since FWHM equals 2.36σ , to describe the resolution in these
terms,

w √
FWHM = 2.36E = 2.36 wE (2.19)
E
At the w value of 3.0 eV for germanium and 3.7 eV for silicon, the equation yields
the FWHM of 4.7 keV for germanium at 1.33 MeV and of 9 keV for silicon at 4
MeV. The resolution measured at these energies is about 3-fold better in germanium
and 1.5-fold worse in silicon.
Two sets of factors that work in opposite directions √ are not considered in
Eq. (2.19). The equation overestimates the FWHM by F, where F is the em-
pirically observed Fano factor. The existence of this factor is attributed to the
circumstance that the generated electrons do not necessarily act independently in
producing the ionization pulse, and so the peak is narrower than when attributed
to random events. On the other hand, detector drift, noise, and incomplete carrier
collection each contributes to widening the FWHM (Knoll 1989).
The counting efficiency of these detectors is calibrated with radionuclide stan-
dards or Monte Carlo simulation (Briesmeister 1990). Typically, the alpha-particle
detector has the same efficiency for thin samples at all commonly encountered
 2. Radiation Detection Principles 37

energies, while the gamma-ray detector efficiency has a maximum near 100 keV
for samples of various dimensions.
Commercially available spectrometer systems include a computer for operating
the system, storing data, analyzing data, and providing information output. These
functions are used to set counting times and periods, identify peaks by channel and
energy, store and subtract background, calculate gross and net counts in multiple
peak regions, calculate count rates, and convert count rates to disintegration rates
on the basis of stored decay scheme data. Statistics software programs are available
to estimate output data uncertainty.

2.5.5. Other Spectrometers

Alpha-particle spectral analysis before development of solid-state detectors was
accomplished with the Frisch grid chamber. In this large ionization detector, the
effect on pulse height due to sample location and ion drift is minimized. A colli-
mator restricts incoming alpha particles to a limited volume, and a grid that is as
transparent to electrons as possible shields the sensitive area from the positive ions
that are generated. Although the silicon diode is far more convenient because of
its stability, large-area Frisch grid chambers continue to be used for alpha-particle
spectral analysis of surfaces larger than about 20 cm2 .
When designed for optimum detection efficiency, gas-filled proportional coun-
ters are effective as X-ray spectrometers for photons of energy below about 80 keV.
The important criterion is good counting efficiency, achieved by absorbing a large
fraction of the X rays entering the detector. These detectors are larger than the de-
tectors used for alpha- and beta-particle counting, filled with a gas that has higher
electron density such as xenon, and at a pressure of several atmospheres. The
window is a beryllium disk to maintain the pressure while minimizing attenuation.
Gamma-ray spectral analysis before development of solid-state detectors was
by the NaI(Tl) detector. Some of these detectors are much more efficient than
germanium detectors, both because iodine has a higher Z value than germanium
and also because larger sodium iodide detectors can be formed. No cooling is
necessary, although fairly uniform temperature is needed to eliminate calibration
drift. Because the resolution is controlled by the counting statistics of energy
conversion at the PMT, the resolution is much worse for NaI(Tl) detectors than
for germanium detectors, typically by a factor of 30 near 0.66 MeV, as shown in
Fig. 2.14. This spectrometer system is still used in selected situations where good
resolution is not crucial, i.e., if only a few radionuclides are to be measured at one
time, and if ease of application and portability are important factors.
Older LS counter systems (see Section 8.3.2) have three channels for distin-
guishing energies; new ones have full energy spectrometers. These are directly
applicable for distinguishing alpha particles by energy. Because beta particles are
emitted as a spectrum from zero to maximum energy, and the spectra have various
shapes, identification of beta-particle emitters by energy is less feasible.
The energy spectrum presented by the LS system does not exactly reflect
the beta-particle spectrum because some energy is not deposited or is lost in
38 Moses Attrep, Jr.

transmission. When multiple beta-particle energy groups are detected, only the
more energetic beta particles in the group with the highest maximum energy can
be seen by themselves because the other groups overlap.
The advantage of an LS counter as alpha-particle spectrometer is the minimal
energy loss because the sample is integral to the detector medium. The disad-
vantages are the relatively poor resolution associated with the PMT, some edge
effects where the alpha-particle emitter is near the vial wall, and a higher radiation
background count rate. Typical FWHM is around 500 keV at 5 MeV. About 2-fold
better resolution is achieved by special pulse timing discrimination (McDowell
1992).
3
Analytical Chemistry Principles
JEFFREY LAHR and BERND KAHN

3.1. Introduction
To separate and purify the radionuclide of interest in the sample, the analyst can de-
pend on the similar behavior of the stable element and its radioisotopes. Chemical
reactions involving the radionuclide will proceed with essentially the equilibrium
and rate constants known for the stable element in the same chemical form. Slight
differences result from small differences between the isotopic mass of the radionu-
clide and the atomic mass (i.e., the weighted average of the stable isotopic masses)
of the stable element. Because of this similarity in chemical behavior, many ra-
dioanalytical chemistry procedures were adapted from classical quantitative and
qualitative analysis. For the same reason, new methods published for separating
chemical substances by processes such as precipitation, ion-exchange, solvent ex-
traction, or distillation are adapted for and applied to radionuclides. One exception
occurs when the radionuclides to be separated are two or more isotopes of the same
element. Here, effective separation can be accomplished by mass spectrometer (see
Chapter 17).
This chapter reviews classical separation techniques and their roles in the ra-
dioanalytical chemistry laboratory. Practical application to specific radionuclides
and sample types is discussed in Chapter 6.
Chemical separation of a radionuclide from other radionuclides is intended
to recover most of the radionuclide while removing most of the accompanying
contaminants. Purification specifications usually can be relaxed by selecting a
radiation detector that measures the radionuclide of interest without detecting
some of the contaminant radionuclides, either by discrimination against certain
types of radiations or by spectral energy analysis.
The measure of purification is the decontamination factor, DF, defined as the
concentration of the interfering radionuclide before separation divided by its con-
centration after separation. The required DF depends on the initial concentrations
of the interfering radionuclide and the radionuclide of interest, and the extent of
acceptable contamination when counting the emitted radiation. The DF must be
large if the radionuclide of interest is a small fraction of the total initial radionu-
clide content. The interfering radionuclide that remains in the source that is counted

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

39
40 Jeffrey Lahr and Bernd Kahn

should contribute not more than a few percent to the count rate of the radionuclide
of interest. Evaluation of the fractional contribution of the radiation from inter-
fering radionuclides must consider the radioactive decay of the radionuclide of
interest and the interfering radionuclides.
The separation method generally is selected from available methods that are de-
scribed as purifying the radionuclide of interest from the identified contaminant ra-
dionuclides. Scavenging steps can be inserted that separate the major contaminant
radionuclides from the radionuclide of interest. The selected method must be tested
to determine all DF values for the contaminant radionuclides or at least to demon-
strate for the sample under consideration that none interfere with the radiation
measurement. A separation step will have to be repeated or additional separation
steps will have to be added when a single step does not achieve the required DF.
Radionuclide analysis methods are published in analytical chemistry and ra-
diochemistry journals, and in methods manuals issued by nuclear facilities such
as government laboratories. For example, the Environmental Measurements Lab-
oratory Procedures manual, HASL-300 (Chieco 1997), is an excellent source.
Standard methods for radionuclide analysis (see Section 6.7) are available, and
should be used whenever appropriate. If conditions differ from those to which
published methods have been applied, radionuclide recovery and decontamination
must be tested and additional process steps may have to be inserted.
If no applicable analytical method is found, then individual separation steps
have to be selected and combined sequentially, each step to remove one or more
contaminants until all are removed to the extent necessary. The first step must match
the sample form, each subsequent purification step must match the preceding step,
and the final step must produce the counting source in its specified form. Each step
must give high recovery of the radionuclide of interest and the required removal
of interfering radionuclides. Suitably designed tracer tests can provide otherwise
unavailable information.
Each separation method also must be evaluated for suitability with the stable
(nonradioactive) dissolved substances in the sample. These substances may inter-
fere in separations by competing or blocking reactions. Separation or scavenging
steps must be introduced to remove these interfering stable substances or reduce
them to acceptable amounts.
The techniques for separating and purifying radionuclides as part of the radio-
analytical chemistry process are discussed in the following sections. Although
separate sections present the different techniques, the analyst is expected to com-
bine separation techniques that produce optimum analytical efficacy.

3.2. Precipitation Separation

3.2.1. Principle
Whether and to what extent a salt precipitates is characterized by the solubility
product K sp . The solubility product should be recalled from general chemistry
texts as the equilibrium constant describing the formation of a slightly soluble (or
nearly insoluble) ionic compound from its component ions in solution. Consider
 3. Analytical Chemistry Principles 41

the chemical reaction

Aa Bb (s) = a Ab+ (aq) + bB a− (aq) (3.1)

In Eq. (3.1), Ab+ is the cation with charge b+ and B a− is the anion with charge
a − . The solubility product is the product of the thermodynamic activities of the
component ions in solution at equilibrium:

K sp = [Ab+ ]a [B a− ]b (3.2)

Here, the term “activity” is used differently than in other chapters of this text-
book; it stands for the thermodynamic value that replaces the molar concentration
of a substance. The units of K sp are mol/l to the power (a + b). Values are listed in
handbooks, such as the CRC (Weast 1985) and Lange’s Handbook of Chemistry
(Dean 1999), as well as in general chemistry textbooks and online.
The solubility product may be calculated from the free energy of formation G
of the solids and aqueous forms (see Weast 1985) in Eq. (3.3), with the absolute
temperature T and the gas constant R in consistent units:

−G
ln K sp = (3.3)
RT
If the product of the activities of the two components initially in solution is equal
to or less than the K sp of the product compound, then no precipitate is formed.
When the product of the known initial activities of the two components exceeds the
K sp value, a precipitate is formed. One can then estimate the activity of the residual
ion of interest (say B a− ) left in solution from the known value of K sp .Consider for
simplicity that concentrations are so low that they can be substituted for activity,
and designatethe initial concentrations of the two components by a zero subscript,

K sp K sp
(B a− )b = = (3.4)
b+
(A ) a [(A )0 − [(a/b) (B )0 ] + [(a/b) (B a− )]]a
b+ a−

The fraction of B a− remaining in solution after precipitation, (B a− )/(B a− )0 ,

can be estimated by dividing both sides of Eq. (3.4) by (B a− )0 , so that
 a− b 
(B ) K sp
= a− b     (3.5)
(B a− )b0 (B )0 [(Ab+ )0 − (a/b) (B a− )b0 + (a/b) (B a− )b ]a

For example, if the iodide reagent added as carrier for radioactive iodine is
precipitated as AgI by adding silver nitrate to solution, where a = b = 1, the
fraction of iodine remaining in solution after the precipitation, (I 1− )/(I 1− )0 , is
 1− 
(I ) K sp
= 1− (3.6)
1−
(I )0 (I )0 [(Ag )0 − (I 1− )0 + (I 1− )]
1+

The last term in the denominator of Eq. (3.6) can be ignored when it is very small
compared to the difference between the other two terms within the brackets. The
42 Jeffrey Lahr and Bernd Kahn

K sp of silver iodide is 8.49 × 10−17 (mol/l)2 (Weast 1985). Typically in analyzing

radioiodine, the amount of iodide carrier added to 0.035-L solution is 10 mg or
0.078 mmol, and the amount of silver added is 1 ml of 1 M-solution, i.e., 1 mmol,
hence the initial concentrations are 0.0021 M iodine and 0.027 M silver. As a
result, the fraction of iodide that remains in solution after precipitation is 1.6 ×
10−12 .
The result may be modified by the influence of relatively high concentrations of
silver nitrate and possibly other substances in solution on the activity coefficients
for silver and iodide, but precipitation of iodide essentially is complete. Simple
calculations of solubility from K sp values are only approximations of solubility
behavior due to factors such as incomplete dissociation, complex ion formation,
and ion pair formation.

3.2.2. Practice
A first separation step for a radionuclide may be coprecipitation or scavenging
removal with reagents suggested by group separations applied in the past for
qualitative and quantitative analysis. One incentive is to scavenge, as soon as
possible, any relatively intense contaminant radionuclides to retain a sample that
can be handled with less radiation exposure and contamination potential. Group
precipitation can reduce the sample volume by carrying radionuclides from a
large initial water sample on a solid that is then dissolved in a relatively small
volume. This strategy can also serve to place the radionuclides into a solution
that no longer contains interfering substances and is more amenable to subsequent
chemical processing. For example, individual rare earth elements then can be
separated from each other on an ion-exchange column with a complexing agent
under closely controlled conditions such as precise pH values.
Carriers that have proved effective for coprecipitation are the hydrous oxides
of the metals, particularly of iron, other transition metals, and aluminum. Their
efficacy is due to their large surface area, gelatinous character, and ability to coag-
ulate. Some distinction among carried ions can be achieved because, as indicated
by Table 3.1, ions are precipitated as hydroxides at various pH thresholds.

TABLE 3.1. Precipitation of ions as hydrous oxides at various

pH values
pH threshold Species precipitated
pH 3 Sn+2 , Zr+4 , Fe+3
pH 4 U+6 , Th+4
pH 5 Al+3
pH 6 Zn+2 , Be+2 , Cu+2 , Cr+3
pH 7 Sm+3 , Fe+2 , Pb+2
pH 8 Ce+3 , Co+2 , Ni+2 , Cd+2 , Pr+3 , Nd+3 , Y+3
pH 9 Ag+1 , Mn+2 , La+3 , Hg+2
pH 11 Mg+2
 3. Analytical Chemistry Principles 43

TABLE 3.2. Group precipitation

Type of precipitate Carrier example Representative elements precipitated
Hydroxide Ferric hydroxide—Fe(OH)3 and Coprecipitates rare earth and actinide
other hydrated oxides like elements without difficulty.
Mn(OH)2 and Al(OH)3 Used in many procedures as first step to
concentrate from water, urine, and
dissolved solid samples.
Phosphate Alkaline earths, especially Ra, U, Th, Ac, Pd, Ba, Al, Bi, In, Zr, plus
calcium rare earth elements
phosphate—Ca3 (PO4 )2
Oxalate Calcium oxalate—CaC2 O4 Precipitates actinides and rare earth
elements.
Can separate Ra from Pb, Bi, Po, and
Ca at pH 2.
Th at pH 3.5.
Sr, Ba, and Y carriers precipitate as
oxalates.
Precipitates actinides from urine and
leaves behind organics
Sulfate Barium sulfate—BaSO4 Coprecipitate actinides for counting.
Lead sulfate—PbSO4 Carries radium for counting.
Precipitates essentially all ter- and
quadrivalent cations; i.e., all elements
from Pb to Cf in addition to Ba, La,
and the light lanthanides (Sill and
Williams 1969).
Fluoride Neodymium fluoride—NdF3 Precipitates actinides for counting by
alpha spectrometry.
Sulfide Lead sulfide—PbS Pb sulfide at pH 3.5 to 4 carries Po; can
separate Pb, Bi, Po from Ca.
Carbonate Strontium carbonate—SrCO3 Sr, Ba carriers
Nitrate Barium nitrate—Ba(NO3 )2 Fuming HNO3 at 60% conc. will
Strontium nitrate—Sr(NO3 )2 precipitate Sr and Ba but not most Ca.
Chromate Barium chromate—BaCrO4 Carries Ra, Pb, and Ba—separates from
Sr; Final precipitate for Pb.
Iodate Cerium iodate—Ce(IO3 )4 Carries Th, Po, Pt, Pa, and Ce—not
Am+6 ,U+6 , and Pu+6 (Hindman
1986)

Under favorable circumstances, an initial precipitation reaction can be selected

that will separate the radionuclide of interest from most contaminants. An example
from the group separations in Table 3.2 is the precipitation of strontium nitrate in
concentrated nitric acid. The element of interest may be accompanied by several
other elements, usually in the same periodic group, as indicated in the table. Barium
and radium nitrate, for example, are also insoluble. If such specific precipitation
reactions can be applied directly, the only additional processes will be separation
from these similar elements and preparation of the counting source.
Otherwise, procedures can be selected to separate specific radionuclides in a se-
ries of steps performed in different chemical environments to free the radionuclide
44 Jeffrey Lahr and Bernd Kahn

of interest from all accompanying radionuclides and nonradioactive substances

that would interfere with the radionuclide measurement. Each separation is effec-
tive if it recovers most of the radionuclide of interest while carrying little of the
contaminants. Special techniques, such as homogeneous precipitation, have been
suggested to enhance separation (Gordon et al. 1959).

3.3. Ion-Exchange Separation

3.3.1. Principle
Radionuclides in an aqueous solution can be retained on a solid ion-exchange
medium by stirring the solid in the solution or passing the solution through a col-
umn filled with the solid. In the ideal case, the radionuclide of interest is collected
almost entirely on the ion-exchange medium after sufficient time to reach equi-
librium while the contaminants remain almost entirely in solution, or vice versa.
If the purified radionuclide retained on the ion-exchange medium emits gamma
rays, it is then counted on the ion-exchange medium. If not, or if the yield (see
Section 4.7) must be determined, the radionuclide is eluted from the ion-exchange
medium for further processing of the elutriant solution to prepare a counting source.
Column techniques can achieve more complex results, such as collecting sev-
eral radionuclides on the ion-exchange medium and then selectively eluting each
radionuclide.
The mass-action concept of ion-exchange between an ion Ab+ initially in solu-
tion and an ion B a+ on the ion-exchange (ix) medium, at equilibrium, is represented
by Eq. (3.7):
a Ab+ (aq) + bB a+ (ix) = a Ab+ (ix) + bB a+ (aq) (3.7)
The equilibrium or selectivity constant E A/B for aqueous ions Ab+ that replace
some ions fixed on the ion-exchange medium, B a+ is
[B a+ ]b [Ab+ ]aix
E A/B = (3.8)
[Ab+ ]a [B a+ ]bix
The brackets represent the thermodynamic activity. At low concentrations in water,
the concentration value can be used for activity; at higher concentrations or with
other ions in solution, the concentration value must be multiplied by the activity
coefficient. The activity of the ions on the ion-exchange medium is not readily
available; the mole fraction, defined as the moles of an individual component
divided by the total moles of all components in the phase, has been used in its
place (Rieman and Walton 1970).
One way of bypassing calculation of E A/B to estimate the selectivity of a specific
ion-exchange resin for various ions is to measure the distribution coefficient DV
for individual ions, including radionuclides. (The mass distribution coefficient DM
is also used for this purpose.) The volumetric distribution coefficient is the ratio of
 3. Analytical Chemistry Principles 45

the ion concentration on the ion-exchange medium relative to the ion concentration
in solution at equilibrium (both values in units of mol/l or Bq/L:

[Ab+ ]ix
DV = (3.9)
[Ab+ ]aq

The larger the value of DV , the greater is the selectivity of the resin for that
ion. On ion-exchange resins (see below), trivalent ions are bound more strongly
than divalent ions, which are in turn bound more strongly than monovalent ions.
Among monovalent ions, the order of selectivity generally is Cs > Rb > K >
Na > Li (Walton and Rocklin 1990), which is in the order of the ionic radius. The
value of DV can vary with the saturation of the ion-exchange resin by the backing
ion; if the concentration of Ab+ is low, the concentration of B a+ will be relatively
high in both the solution and the resin phase.
The value of DV is determined by shaking or stirring a specified amount of
solution to which the ion of interest, such as the radionuclide, had been added with
a specified amount of ion-exchange medium. When the radionuclide distribution
between solid and liquid phase reaches equilibrium, the concentration of the ra-
dionuclide in each phase is measured. The batch of ion-exchange medium must
be saturated initially with the specified backing ion and the initial solution must
contain the same ion at a specified concentration. For example, if the radionuclide
is a cationic radionuclide such as 42 K+ and the system for comparison is a sodium
salt, then the ion-exchange medium must be in the sodium form and the solution
that contains 42 K+ must be at a specified sodium backing-ion concentration. The
concentration of nonradioactive potassium ion must also be specified. Any other
radionuclides from which the radionuclide of interest is to be separated must be
equilibrated under identical conditions of known volume ratios, ionic concentra-
tion, temperature, and equilibration period.
A column experiment can also yield the value of DV . The radionuclide-bearing
solution is added at one end of an ion-exchange medium column and is then eluted
and collected in incremental volumes that are measured for radionuclide content.
Ideally, the concentration of the eluted radionuclide is distributed in the incremental
volumes as a Gaussian curve. For a distance of movement along the column (in this
case, the length of the resin column), l, cross-sectional area a, interstitial column
volume occupied by solution, i, and elutriant volume Ve measured to the peak of
the radionuclide concentration curve, Eq. (3.10) applies
Ve
DV = (3.10)
la−i
Because DV is dimensionless, the numerator and denominator on the right-hand
side of Eq. (3.10) must be in the same units of volume, say cm3 . If the radionuclide
can be measured while it is moving in the column (for example, with a gamma-
ray detector behind a shield that has a slit so that only a narrow width of column
is observed), then Eq. (3.10) can be applied for movement through the column
46 Jeffrey Lahr and Bernd Kahn

without monitoring the effluent. As indicated in the preceding paragraph, the same
backing ion and identical conditions must be used in all comparisons.
Column separations are more efficient than batch separations because a column
can be viewed as a series of sequential batch separations. For column separation of
two radionuclides, the difference between corresponding values of DV represents
the number of column volumes separating the two elution peaks; if the Gaussian
curves are sufficiently narrow, little contamination may occur. For batch separa-
tions, in contrast, the value of DV indicates that some contaminant always remains,
and sequential batch separations may be needed to reduce such contamination to an
acceptable amount. Batch separations are useful if the value of DV differs by, say,
2 orders of magnitude and the solid:liquid volume ratio is selected for acceptable
discrimination. The easiest separation is a cation from anions or vice versa, so that
one form can be essentially completely retained while the other form is only in
interstitial water.

3.3.2. Practice
The process of ion-exchange was first discovered and studied in natural inorganic
compounds, of which the most abundant are the clay minerals, especially zeolites.
The latter are microporous, crystalline aluminosilicate minerals. Numerous natu-
rally occurring and synthetic zeolites exist, each with a unique three-dimensional
structure that can host cations, water, or other molecules in its void space (cavities
or channels).
Several inorganic compounds were found to be useful for specific separations
involved in radioanalytical work (Amphlett 1964). These include hydrous oxides
of chromium(III), zirconium(IV), tin(IV), and thorium(IV); aluminum molyb-
dophosphate [(NH4 )3 PO.4 12MoO.3 3H2 O]; zirconium phosphate [Zr(HPO4 ).2 H2 O
or ZrO.2 P2 O5 ]; molecular sieves (activated synthetic crystalline zeolites); and sil-
ica gel. These inorganic substances are amorphous and eventually degrade to
fine powders. They are useful for batch operations but not for high performance
chromatography.
Paper can also function as an ion-exchange medium. It has very low capacity
but is suitable for separating radionuclides at their low concentrations. Typically,
paper chromatography is performed on strips through which a selected solvent
flows, and distinguishes radionuclides by their path lengths along the strip.
Once ion-exchange resins were synthesized in 1935, these organic exchangers
replaced zeolites in applications (industrial and analytical) and in scientific investi-
gation. The advent of nuclear power in mid-twentieth century stimulated renewed
interest in inorganic exchangers that were stable at high temperatures and could
withstand the effects of high doses of radiation.
The ion-exchange resins commonly used for separations in a radioanalytical
chemistry laboratory are solid organic structures with ion-exchange sites, or with
attached substances that function as exchange sites. Some common organic ion-
exchange media are listed in Table 3.3. The nature of the functional group defines
the properties of the resin. Resins with fixed positive charge groups can exchange
 3. Analytical Chemistry Principles 47

TABLE 3.3. Types of ion-exchange resins

Cation-exchange resins Functional groups Chemical formula
Strongly acidic Sulfonic acid groups RSO−
3H
+

Moderately strong acidic Substituted phosphoric acids PO(OH)2

Weakly acidic Carboxylic acid group RCOOH

Anion-exchange resins Functional Groups Chemical Formula

Strongly basic Tetraalkylammonium groups [RN(CH3 )3 ]+ Cl−

Weakly basic Tertiary amine groups [RN(CH3 )3 ]+ Cl−
Secondary amine groups [RNH(CH3 )2 ]+ Cl−

anions and are thus called anion-exchange resins; resins with fixed negatively
charged groups exchange cations and are called cation-exchange resins.
A few commercially available ion-exchange resins are used for most published
separations. They are defined by functional group, ion-exchange capacity (in
meq g−1 ), mesh size, cross-linkage, density, and ionic form. Each descriptor af-
fects the value of DV in qualitatively understood ways (Korkisch 1989), so that the
analyst may select a resin with optimum characteristics for the intended separation.
The ionic form of the purchased resin can be replaced by washing the resin with
a solution that contains the new ion until complete conversion is demonstrated by
tests of the effluent solution.
Selection of ion-exchange column dimensions is based on the amount of the
resin needed to hold an amount of ion related to the sample (its capacity) and
to achieve separation of the radionuclide. A longer and narrower column permits
better separation, but the cross-sectional area must be sufficient to minimize wall
effects that lead to the liquid flowing along the wall in preference to passing through
the resin. Although the analyst may desire the most rapid flow permitted by column
flow resistance, the flow must be sufficiently slow to permit local equilibrium and
to avoid disrupting column structure by channeling or introducing air bubbles.
Flow may be either upward or downward; the former tends to decrease disruption
of column structure by air bubbles.
Columns must be filled carefully for uniformity of the resin bed without voids,
and maintained to prevent drying and the resulting separation within the column.
In many applications, columns can be reused for a limited number of times after
being washed with water and then regenerated with the ionic solution that re-
turns them to the original form. Resin reuse is not possible for applications that
partially destroy resin, clog the column with solids, coat the resin with emul-
sions, or load the ion-exchange sites with various ill-defined ions. Commercial
ion-exchange resins usually are relatively stable but decompose slowly with time,
or more rapidly when attacked by strong oxidizing agents (Rieman and Walton
1970).
The separation typically is performed by adding from a few milliliters to a few
liters of the radionuclide-bearing sample to a resin column of about 5–50 ml,
and then washing the radionuclides sorbed on the column with water or a dilute
reagent. In the sorption phase, radionuclides of interest either are retained at the
48 Jeffrey Lahr and Bernd Kahn

inflow end of the resin or move through the column but do not break through. The
radionuclide is then eluted with a selected reagent of experimentally determined
number of column volumes. Several radionuclides may be eluted in succession
with different types or strengths of elutriant. Used in the scavenging mode, the
radionuclide of interest passes through the column while contaminants remain on
the resin.
Researchers systematically examined values of the distribution coefficient for
various resins and solutions across the entire periodic table. These distribution
coefficients are available in tables or graphs of the distribution coefficient vs. acid
concentration. One well-known example, in Fig. 3.1, shows ln DV vs. concentra-
tion of HCl (Kraus and Nelson 1956). Similar figures are available for HBr, HNO3 ,
HClO4 , and H2 SO4 , as well as selected acid/alcohol combinations (Korkisch
1989).
The distinction among oxidation states in Fig. 3.1 suggests oxidation or reduc-
tion as a convenient technique for eluting a radionuclide that is strongly retained
on the ion-exchange resin at its original oxidation state. Taken to its extreme, a
cation held on the ion-exchange resin is removed completely by being converted to
an anion, and vice versa, or to a nonionic state. The same principle applies when a
metal ion, by addition of a complex-forming agent, is retained on or removed from
the ion-exchange resin. Such complexes may be uncharged, of the same charge as
the original ion, or of the opposite charge. The curves in Fig. 3.1 show the effect
of complex formation on retention by ion-exchange in the metal–chloride system.
Such complexes carry a negative charge when fully coordinated, and are adsorbed
by the anion exchanger. The analyte may be eluted by changing the concentration
to cause dissociation of the anionic complexes, or by changing the oxidation state
of the metal ion.
Use of complex-forming agents in solution can increase the selectivity of an
ion-exchange procedure. An example is given in Fig. 3.2 for separating rare earth
ions complexed with ethylene dinitrilotetraacetic acid (EDTA). The rare earth ions
are held on the cation-exchange resin when not complexed and are released by
changing the pH value so that they become complexed. The curves in Fig. 3.2
indicate that the radionuclide represented by the curve on the extreme right (La3+ )
can be completely retained on the column under the study conditions at pH from
0.4 to 2.8, and eluted completely above pH 3.9. The La+3 is separated completely
at pH 2.8 from the radionuclide represented by the fifth curve (Zn+2 ) from the
right, and from all ions to the left of the Zn+2 curve. It is partially separated from
Sm+3 and UO+2 2 ions represented by the other two curves on the right. By operating
at pH 3.4, approximately one-half of La+3 is retained, and is separated from all
radionuclides to the left except UO+2 2 .
Ion-exchange resins have also been loaded with counter ions for in situ pre-
cipitation; for example, an anion-exchange resin in the sulfate form can collect
226
Ra and 90 Sr, or one in the tetraphenyl borate form can collect 137 Cs (Cesarano
et al. 1965). The radionuclide of interest is then eluted with a solution in which it
dissolves.
 No adsorption M 12
Slight adsorption M

Strong adsorption
3. Analytical Chemistry Principles

FIGURE 3.1. DV on anion-exchange resin as a function of HCl concentration. (Figure from Kraus and Nelson 1956.)
49
 50 Jeffrey Lahr and Bernd Kahn

100

90

80
Metal ion on resin (%)

70 Th+4 Yb+3

Hg+2 Zn+2 Y+3 Sm+3 UO2+2 La+3

60

50
Sc+3

40

30

20
Cu+4

10 Zr+4 Th+4
Bi+3 Yb+3
Fe+3
0
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0
pH

FIGURE 3.2. Elution of rare earths from a Dowex-50 cation-exchange resin, as a function
of EDTA solution pH. (Figure from Fritz and Umbreit 1958, p. 513.)

3.4. Liquid–Liquid Extraction

3.4.1. Principle
In liquid–liquid extraction, an immiscible liquid—usually an organic solution—is
combined with the sample in aqueous solution in an extraction flask and shaken to
achieve good contact between the liquids. A reagent that functions as an extractant
may have been added to one phase or the other. Information on the distribution ratio
D and the extraction yield E that indicate the extent of purification from specified
contaminants is available from many studies (Sekine and Hasegaw 1977). The
information should describe the extractant, the organic solvent and the conditions
of purity, reagent concentrations, volumes, required time, and temperature. The
value of D reflects the ratio of the radioelement solubility in the organic phase to
that in the aqueous phase, hence the type of solvent and the chemical form of the
radionuclide to be extracted may be inferred from radioelement solubility data.
If the initial conditions of the extraction procedure are not identical to those for
reported extractions, the extent of extraction must be tested.
The two liquids must be as mutually immiscible and nonreactive as possible. The
interface between the two liquids should be reasonably clean to avoid carryover
of one phase in the other as part of an emulsion. Some time may be required for
phase separation after mixing.
The distribution ratio is defined in analogy to Eq. (3.9) as

Corg
D= (3.11)
Caq
 3. Analytical Chemistry Principles 51

The subscripts org and aq refer to the organic and aqueous phase, respectively, and
C is the concentration of the radioelement at equilibrium.
The extraction yield E is defined as the ratio of the amount of solute extracted
to its initial total amount. Relative to the volume of solution V ,
D
E=  (3.12)
Vaq
D+
Vorg
For extractions repeated n times with fresh portions of the organic phase at
constant Vaq and Vorg , the total recovered fraction E r is defined as
  
Vorg −n
Er = 1 − 1 + D (3.13)
Vaq
From Eq. (3.13), the fraction of solute remaining in the aqueous phase is [1 +
D(Vorg /Vaq )]−n .

3.4.2. Practice
Extractant reagents and solvents are selected for optimum separation of the ra-
dionuclide of interest from contaminants, and also with regard to chemical stabil-
ity, purity, and minimal hazard potential. The reagents and solvents may have to
be stored under conditions that avoid degradation, and purified to remove minor
contaminants that can interfere with separations. Use of chemicals listed as haz-
ardous material may be feasible with appropriate care, but later disposal may be
difficult.
Equilibration usually is reached in a few minutes, but should be checked for
each method. Typically, the radionuclide in the aqueous phase is extracted into
the organic phase under one set of conditions, and is then back-extracted under a
second set of conditions. Washing steps commonly are inserted after extraction to
improve the specificity of the radionuclide transfer. A single extraction and back-
extraction cycle may suffice for purification or several cycles may be necessary. In
analogy to ion-exchange systems, column separations have been developed with
countercurrent flow of the two liquids.
Solvent extraction systems may be characterized by type of reagent used as
extractant or by chemical species extracted. Data for the extraction of elements is
in tabular form listed by element, plots of D vs. pH curves in periodic table format,
and distribution coefficient values for multiple elements plotted at specific phase
compositions for a given extraction system. The many solvent extraction systems
that have been reported provide many options to select a suitable system to solve
a given problem. A brief description of extractant categories is given here with
examples, noting that many reagents will fit more than one category according
to the environment in which they are used (Marcus and Kertes 1969, Sekine and
Hasegaw 1977).
Nonsolvating solvents may be used for nonelectrolyte molecular species such as
rare gases, halogens, interhalogen compounds, and some metal halide complexes.
52 Jeffrey Lahr and Bernd Kahn

Examples include benzene, hexane, chloroform, carbon tetrachloride, toluene,

and carbon disulfide. A well-known early example is the extraction of halogen
molecules such as I2 and Br2 , which are only sparingly soluble in water, into car-
bon tetrachloride or chloroform. The reduced form (the halides, I− and Br− ) is
then back extracted into water.
Ion-pair-forming extractants may be used for anions, anionic metal complexes,
and weak acids and bases. Examples include quaternary ammonium salts such as
tricaprylmonomethyl ammonium chloride (Aliquat 336); polyphenyloniums such
as tetraphenylarsonium chloride; cationic dyes such as rhodamine B and alizarin
blue; large anionic extractants such as tetraphenylborate (TPB); and heterocyclic
polyamines such as 2,2 -dipyridyl (DIP).
Chelating extractants may be used for ionic salts. Examples include beta
diketones such as thenoyltrifluoroacetone (TTA), acetylacetone, benzoylacetone,
and dibenzoylmethane; 8-hydroxyquinoline (oxine); oximes such as dimethyl-
glyoxime; nitrosophenyl compounds and nitrosohydroxylamines such as N -
nitrosophenyl hydroxylamine (cupferron); and diphenylthiocarbazone (dithizone).
Alkylphosphoric acids are acidic extractants sometimes called liquid cation
exchangers due to exchange of H+ ions for the extracted cations. They have been
used for many metal ions. Examples include mono-(2-ethylhexyl)phosphoric acid
(MEHP), dibutylphosphoric acid (DBP), and di-(2-ethylhexyl)phosphoric acid
(DEHP).
High molecular weight amines are basic extractants sometimes called liquid
anion exchangers. They have been used for anionic metal complexes. Examples
include tetraphenylborate (TPB), trioctylamine (TOA), triisooctylamine (TIOA),
and trilaurylamine (TLA).
The high-molecular-weight amines R3 N can be considered liquid ion-exchange
media because of the reaction

R3 NH+ A− (org) + B − (aq) = R3 NH+ B − (org) + A− (aq) (3.14)

This equation is analogous to Eq. (3.7). The amine and its inert solvent—for
example, benzene, toluene, xylene, or kerosene—strongly extract the mineral acids
HA and then extract anions B − that are retained by ion-exchange.
Figure 3.3 shows ln D with triisooctyl amine of many polyvalent metal ions
that form anionic chloro complexes in hydrochloric acid. Additional distribution
data are available for metal nitrates, sulfates, and oxy anions that can be extracted.
These ions can then be back-extracted into less acid systems in which they are no
longer anionic complexes.
Neutral extractants may be used for uncharged metal complexes, ionic salts, and
strong acids. Examples include ethers such as diethylether and diisopropylether;
ketones such as methylisobutyl ketone (MIBK or hexone); and neutral organophos-
phorous compounds such as tri-n-butylphosphate (TBP), tri-n-octylphosphine
oxide (TOPO), triphenylphosphate (TPP), and triphenyphosphine oxide (TPPO).
Such neutral extractants have long been used for extracting actinides and lan-
thanides from nitric acid solutions.
 3. Analytical Chemistry Principles 53

FIGURE 3.3. ln D as function of HCl concentration (in 0.11 M triisooctylamine in xylene)

for extraction by amines. [Figure from Marcus and Kertes (1969), p. 960.]

One example is diethyl ether as an extractant for iron as the chloro complex
HFeCl4 or the cyano complex NH4 Fe(SCN)4 . Many of the radionuclides that form
these molecules can be separated from each other under different conditions related
to redox potential or pH.
54 Jeffrey Lahr and Bernd Kahn

Another example is the use of TOPO in cyclohexane solution. It combines with

various metals in the form Ma+ Cla (TOPO)2 (De et al. 1970). Plots of D for several
elements are shown in Figs. 3.4 and 3.5 for extraction from aqueous solutions as a
function of HNO3 and HCl strength. Trends can be predicted but the value of D is
difficult to predict because of the many factors that influence it. Beyond the obvious
ones of oxidation state and relative liquid volumes are the type and concentration
of organic solvent, concentration of TOPO, temperature, and concentrations of
aqueous phase contents. Acids are also extracted and compete with radionuclide
extraction, although one can achieve equilibrium by first washing the organic phase
with successive fractions of the aqueous phase without the radionuclide until the
organic phase is saturated with the acid.

3.5. Solid Phase Extraction

A separation technique that shares some benefits of ion-exchange and solvent
extraction processes is solid phase extraction (SPE), also called reversed phase
partition chromatography or extraction chromatography. This technique immo-
bilizes extractants on an inert polymeric support to retain specific radionuclides
from a contacting solution (Cerrai and Ghersini 1970). Experimental applications
of SPE began 50 years ago, and development for sample purification began in the
1970s. As in the case of ion-exchange separation, a liquid sample passes through
a column, a cartridge, a tube, or a disc that contains a sorbent to retain the analyte.
After the sample has passed through the sorbent, the retained analyte is washed
and then eluted.
Extractants from liquid–liquid systems such as HDEHP [di(2-ethylhexyl) or-
thophosphoric acid] and CMPO/TBP (carbamoylmethylphosphine oxide deriva-
tive and tri-n-butyl phosphate) are supported on the solid material, as are newer
ion-selective crown ethers (such as 4,4 (5 )-di-t-butylcyclohexano 18-crown-6 for
Sr). Various SPE columns available commercially from Eichrom Industries have
proven useful to separate radionuclides such as Sr, Tc, Ra, Ni, Pb, Am, Pu, Th, U,
Np, Cm, and lanthanides. These columns usually are small (approximately 2 ml
resin bed). Their effectiveness depends on their specificity for the ion that includes
the radionuclide of interest, but the small volume limits the amount (i.e., less
than 10 mg) of carrier that can be retained. The specificity of each product shows
promise for development of procedures for sequential radionuclide analyses from
a single sample aliquot. (Burnett et al. 1997, Horowitz et al. 1998)
Analogous systems are commercially available as 3 M EmporeTM Rad Disks.
These come in the form of impregnated PTFE (polytetrafluoroethylene) mem-
branes, which are used as filters for aqueous samples (Schmitt et al. 1990). Filter
dimensions constrain this type of system to carrier-free separations. These filters
with the retained radionuclide may then be washed, dried, and counted directly or
the radionuclide may be eluted for further processing. Current products include
filters for Ra, Sr, Tc, and Cs.
 3. Analytical Chemistry Principles 55

1000

Plutonium(IV)

100

10 Uranium(VI)

Plutonium(VI)

D 1.0

0.1
Chromium(VI)
Chromium(VI)

0.01

Americium(III)

0.001
0 2 4 6 8 10 12 14
Nitric acid molarity

FIGURE 3.4. D as a function of nitric acid pH for extraction of metal ion into 0.1 M TOPO
in cyclohexane. (Figure from Martin et al. 1961, p. 99.)
56 Jeffrey Lahr and Bernd Kahn

1000
Uranium(VI)

Plutonium(IV)

100
Plutonium(VI)

D 10

1.0
Chromium(VI)

Plutonium(III)

0.1
0 2 4 6 8 10 12

Hydrochloric acid molarity

FIGURE 3.5. D as a function of hydrochloric acid pH for extraction of metal ion into 0.1 M
TOPO in cyclohexane. (Figure from Martin et al. 1961, p. 102.)

3.6. Gas Separations

3.6.1. Distillation and Vaporization
Purification by distillation is one of the oldest separation techniques. Raoult ex-
plained the partition of components between liquid and vapor in terms of their vapor
pressure. The total pressure of a vapor is equal to the sum of the partial pressures
exerted by each component. Each component partial pressure is proportional to
 3. Analytical Chemistry Principles 57

the fractional concentration of the substance in the liquid:

P1 = X 1 P10 (3.15)
where P1 is the vapor pressure of substance 1 over the solution, X 1 is the mole
fraction of substance 1 in solution and P10 is the vapor pressure of the pure substance
at the same temperature. This is a colligative property related to the number of
atoms and not to any characteristics of the substance.
A component at high concentration, such as the solvent, ideally behaves as pos-
tulated by Raoult’s law. Gaseous and liquid solutes at relatively low concentrations
often do not. This deviation from ideal behavior is attributed to the circumstance
that the nearest neighbors of atoms at low concentration usually are not identical
atoms, but rather solvent atoms.
For such solutes, Henry’s law is applicable:
Ca = k Pa (3.16)
where Pa is the partial pressure of substance a, Ca is the concentration of the
substance in solution and k is a constant characteristic of the particular gas–liquid
system. The gaseous atom concentration is proportional to the solute atom con-
centration, but the constant is not necessarily associated with the mole fraction.
Each component of the system is independent of all others. As temperature rises,
so do the partial pressures in the system (Wilson et al. 1968).
Many elements have chemical forms with sufficiently high vapor pressures at
low temperatures—i.e., low boiling points—to permit separation as a gas from
contaminants with higher boiling points. Figure 3.6 summarizes the elements that
can be separated by various aspects of this technique.
The noble gases dissolved in water, as well as the halogens and hydrogen, carbon,
nitrogen, and oxygen (as elements or in a number of compounds) can be flushed
from the solution with an inert gas for collection. Radon is carried from previously
sealed radium-containing solutions to measure the activity of its 226 Ra parent
(Curtiss and Davis 1943), as discussed in Section 6.4.1. Fission-produced xenon
and krypton radioisotopes and neutron-activated 41 Ar are flushed from samples of
reactor coolant water, collected as the gas, and counted. Similarly, 14 C in the form
of dissolved CO2 or CO is flushed with an inert gas through the distillation system
and collected in NaOH or Na2 CO3 solution. Mercury salts can be reduced to the
metal, processed by steam distillation, and collected as liquid mercury on a cool
surface.
Volatile oxidation states of ionic radionuclides in solution that have been sep-
arated by distillation include group IVA, VA, and VIA halides (GeCl4 , AsCl3 ,
SeBr4 , SnCl4 , SbCl3 ), and group VIII oxides (RuO4 , OsO4 , Re2 O7 , Tc2 O7 ). Other
volatile solutes include SiF4 , SO2 , HMnO4 , CrO2 Cl2 , VCl4 , and TiCl4 (DeVoe
1962), and hydrides of some of the cited elements (Bachmann 1982).
A simple radioanalytical chemistry distillation apparatus is shown in Fig. 3.7.
The boiling flask contains the sample solution with any needed reagents to maintain
the radionuclide of interest in its volatile form and any contaminants in nonvolatile
forms. An inflow tube with mouth submerged beneath the solution is available for
 58 Jeffrey Lahr and Bernd Kahn

H He
abcd a
Li Be B C N O F Ne
a bdf bcd abcd abcd abcd a
Na Mg Al Si P S Cl Ar
a d bd abcd abcd abcd a
K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
a d d g e d bd bd abcd bcd abd ad
Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
a d d d d cd cd a a a a bd bd bcd abd ad
Cs Ba La Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
a a d d d cd cd d a ad a ab ad ab ad
Fr Ra Ac
a
Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
d d d d

Form in which elements are volatile

(a) Element
(b) Hydride
(c) Oxide
(d) Halides
(e) As permanganic acid
(f) As boric acid
(g) As chromyl chloride

FIGURE 3.6. Distillation separations by species. (Figure from Coomber 1975, p. 307.)

carrier gas flow or reagent addition. The condenser tube is cooled with water and
leads into the collector, which may either be dry or contain a solution to col-
lect the condensate. For optimum decontamination, the distillation process should
minimize liquid-droplet carry-over.
Tritium as tritiated water undoubtedly is the most common radionuclide purified
by distillation. For separating tritiated water from samples such as biological ma-
terial or vegetation, azeotropic distillation can be more effective (Moghissi et al.
1973). In this method, an organic solvent not miscible with water is mixed inti-
mately with the tritiated aqueous sample. An excellent website on azeotropic distil-
lation can be found at http://www.chemstations.net/documents/DISTILLATION.
PDF. Cyclohexane, which forms a constant-boiling mixture at 94◦ C is used, as
are other organic liquids, such as para-xylene and toluene. The distillation ap-
paratus is assembled with the azeotropic distillation receiver shown in Fig. 3.8,
rather than the usual collection flask. During distillation, the azeotrope separates
into an aqueous and an organic phase in the receiver. The aqueous phase is used
for analysis while the organic phase returns to the distilling flask. Distillation can
continue until essentially all of the water has been collected, achieving complete
distillation. This is particularly important in tritium purification, where incomplete
distillation will enrich the residue in 3 H (see Section 4.8) and yield results that are
underestimates by about 10%.
Many of the elements footnoted in Fig. 3.6 by the letters c or d have been
separated from compounds that have higher boiling points by heating the salts
in a volatilization apparatus at a relatively elevated temperature. The vapor is
 3. Analytical Chemistry Principles 59

FIGURE 3.7. Simple distillation apparatus. (Figure modified from Chemweb 2005.)

collected on a cold plate suitable for counting or subsequent separations. Several

radionuclides can be separated from the same source by successively increasing the
temperature at the sample to each boiling point and replacing the cooled target for

To condenser

FIGURE 3.8. Azeotropic distillation receiver. (Figure

modified from Chemweb 2005.) From distilling flask
60 Jeffrey Lahr and Bernd Kahn

each radionuclide. The technique is occasionally used for rapidly separating a more
volatile radionuclide from an irradiated target, but is limited in the radioanalytical
chemistry laboratory to preparations of purified compounds in cases where the
apparatus and skill are available. An example of this separation is the volatilization
of plutonium chloride from soil in a gas chromatograph (Dienstbach and Bachmann
1980). Another example is the collection of 90 Sr as the chloride salt on a cooled
target when the source is heated between 900 and 1400◦ C; when heated above
1400◦ C, its daughter 90 Y is volatilized (Sherwin 1951).
Radionuclides that are volatilized during ashing or freeze-drying, can be col-
lected in a commercially available apparatus. For freeze-drying, the sample is
placed into a vacuum distillation apparatus and the vaporized radionuclide is col-
lected at a cooled surface or on a sorbent material.

3.6.2. Separation of Gases

Noble gases in air can be collected and concentrated by sorption at very low
temperature on material such as charcoal, silica gel, or molecular sieve. Sorbed
radioactive gases can be counted directly for gamma rays or flushed from the
collectors for separation and measurement by application at elevated temperature
of a vacuum or a stream of inert gas such as helium. The conditions for separation by
sorption typically are selected from reported applications and adjusted empirically
for the samples at hand. The components of the sample matrix can have major
influence on the degree of uptake and the saturation capacity of the sorption medium
available for the gas of interest.
Noble gases collected from air or water (see Section 3.6.1) may be separated
from each other in the carrier gas by selective condensation on sorption media,
followed by selective volatilization at temperatures near their boiling points of the
separated gases (see Section 15.4.2). Radon can be sorbed from air or water and
then measured directly by counting gamma rays emitted by 222 Rn or 220 Rn progeny.
Water vapor in air is collected on these sorption media at measured flow rates
and then removed from the sorbent by distillation. The tritium activity is reported
relative to the amount of water. Collection of various gaseous forms of tritium,
14
C, and radioiodine is discussed in Section 5.3.2.
Binding of molecules in nonionic states to these materials is attributed to van
der Waal’s forces at specific sorption sites. Calculations have been performed to
match theoretical predictions to measured values with respect to rate of sorption
(Blue and Jarzemba 1992), approach to equilibrium, and relative binding for differ-
ent molecules (Underhill 1996). Rates of sorption appear to depend on molecular
diffusion in at least three regions: the air or water medium, the immediate envi-
ronment at the sorbent surface, and within the sorbent material (Kovach 1968).
Commercially available sorption media may be rated by capacity for a specified
substance.
In molecular sieves, the dimensions of molecules considered for sorption can be
matched to the sorbent network, rated by active site diameter (Karge and Weitkamp
1998). Most molecular sieves are stable to 500◦ C. Molecular sieves can be selected
 3. Analytical Chemistry Principles 61

to separate molecules based on size, shape, polarity, and degree of unsaturation.

The range of pore sizes commercially available is approximately 0.3–0.8 nm. A
mixture of gases or vapors of different molecular sizes can be separated by passage
through a series of molecular sieves arranged so that only the smallest molecules
are sorbed by the first zeolite and only the smallest of the remaining compounds
by a second zeolite.

3.7. Electrodeposition Separation

In electrodeposition, an ion is transferred from solution to the surface of an elec-
trode and retained as a solid. Typically, a metallic ion is reduced and deposited as
the metal at the cathode. Deposition of oxy-ions at the cathode and oxidation at the
anode also are employed. The rate of deposition depends on the applied voltage
and factors such as the rate of stirring, volume and nature of the solution, and the
material and area of the electrode. Little deposition occurs until a critical voltage is
reached, after which the rate of deposition increases rapidly with increasing volt-
age. Consider the half reaction in which n is the number of transferred electrons
per atom of A:
−
(aq) + ne ⇔ A(s)
An+ (3.17)
The energy of this reaction E, in volts, is defined as
RT 1
E = Eo + ln n+ (3.18)
nF [A(aq) ]
E o is the voltage, relative to a standard hydrogen electrode (SHE), that is required
for a reduction half reaction at extreme dilution; this value is tabulated for redox
couples in handbooks (Weast 1985). The term [An+ ] is formally defined as the
thermodynamic activity of species A at 25◦ C (298 K) and 1 atmosphere, but often
is applied as concentration when radionuclides are at very low concentrations. The
term R is the gas constant, 8.314 volt-coulomb/Kelvin-mol. The Faraday constant
F is 96,485 coulomb/mol, and T is the temperature in degrees Kelvin. To write
the equation with logarithms to the base 10, these terms are multiplied by 2.303;
for n = 1 and T = 298 K, the coefficient is 0.05916 V.
As an example, for applying Eq. (3.17) to the reduction of silver ion to metallic
silver,
+
Ag(aq) + e− ⇔ Ag(s) (3.19)
The voltage required for electrodeposition is
0.05916 [Ag(s) ]
E = Eo + log  − E over
1 +
Ag(aq)

1
= E o + 0.05916 log  − E over (3.20)
Ag+
(aq)
 62 Jeffrey Lahr and Bernd Kahn

H He

Li Be B C N O F Ne

Na Mg Al Si P S Cl Ar

K Ca Sc Ti V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
X X X X X X
Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te I Xe
X X X X X X X X
Cs Ba La Hf Ta W Re Os Ir Pt Au Hg Tl Pb Bi Po At Rn
X X X X X X X X X
Fr Ra Ac
X X
Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
X X X X X X X

FIGURE 3.9. Electrodeposition separations (shown with X). (Data from Blanchard et al.
1960.)

The overvoltage or overpotential E over is inserted in Eq. (3.20) to adjust for other
processes that compete in the system and make electrodeposition less than ideally
efficient. These processes are irreversible and include the effects of the decompo-
sition of water, other solutes, and imperfections in the electrode surface. Because
of these processes, a greater potential difference than calculated from the reference
potential and the ionic concentration must be applied in order to achieve deposi-
tion. For the same reason, spontaneous deposition, inferred from a positive value
of E o , may not occur if the overvoltage exceeds it. Overvoltage effects occur at
both the cathode and the anode.
The positive value of E o for many metals (for Ag+ , E ◦ = 0.799 V) and the ad-
ditional effect of low concentration from the second term on the right in Eq. (3.20)
suggest that these metals will deposit readily at the cathode to which a low negative
voltage, i.e., a few tenths of a volt, is applied, or even spontaneously. The equi-
librium voltage E o applies to an electrode—in this case, the cathode–uniformly
constructed of the same metal as the ion; the initial voltage required for deposition
is different for a cathode of a different material. Only when a coating of the same
metal has been deposited on the cathode surface, the value of E o will apply.
The current, which is proportional to the amount deposited, can be calculated
to be small at low concentrations. Practical requirements for applied voltage and

TABLE 3.4. Electrodeposition or reduction to metal: grouping in classes

(Lindsey 1964)
Group Name Range
A Precious metal group Gold to mercury
B Copper group Copper, bismuth, antimony, arsenic
C Lead group Lead to tin
D Zinc group Nickel to zinc
E Mercury cathode group Alkali metals, lanthanons, etc.
F Anodic group Metals anodically deposited as oxides
 3. Analytical Chemistry Principles 63

current will differ because of the above-cited competing reactions by the water
and other substances in solution.
Most metals except the alkalis and alkaline earths can be electroplated at the
cathode with suitable applied voltage from acid solutions. Relatively early expe-
rience with electrodeposition of various metals is summarized in Fig. 3.9. The
process typically is applied for carrier-free or low-concentration samples to pre-
pare sources for alpha-particle spectral analysis. It is also useful for depositing thin
sources for counting radionuclides that emit beta-particles with low maximum en-
ergy.
Because electrodeposition usually is the last procedural step, the composition of
the solution can be controlled for ease of electrodeposition. The reagents that are
present may consist of an acid to avoid hydrolysis and a redox agent to retain the
radionuclide in a state suitable for final reduction. As shown in Table 3.4, metals
can be grouped for electrodeposition analysis based on their standard electrode
potentials.
Some elements are not suitable for electrodeposition from aqueous solution as
the metal. Among these are the radionuclides plutonium, uranium, and thorium,
which are prepared for alpha-particle spectral analysis by deposition of oxides.
Other metals, such as lead, can also be deposited as oxides under empirically
derived conditions (Laitinen and Watkins 1975).
4
Radioanalytical Chemistry Principles
and Practices
JEFFREY LAHR,1 BERND KAHN,1 and STAN MORTON2

4.1. Introduction
The preceding chapter describes methods that can be applied to the analysis of both
radionuclides and their stable element counterparts. One important difference to
consider is the extremely low amounts of radionuclides generally analyzed in the
radioanalytical chemistry laboratory. As noted in Section 13.1, the environmental
samples in a radioanalytical chemistry laboratory typically are in the picocurie to
nanocurie (0.037 to 37 Bq) range; this corresponds to a radionuclide sample mass
of around 10−15 g, depending on half-life and molecular weight of the radionuclide.
Precise measurement of such a low amount can be achieved because radionuclides
emit energetic radiation.
The low concentration of the typical radionuclide sample affects some aspects
of its behavior as a solute. One cause is believed to be interactions with surfaces
and other solutes that are not taken into consideration for species at the more
usual concentrations. The chemical status of a radionuclide sample may also be
affected by the energy liberated when a radionuclide is created and the energy de-
posited when the emitted radiation passes through the solvent medium (Vertes et al.
2003).
The peculiarities of radionuclides at extremely low concentration were of great
interest to radiochemists during the first half of the twentieth century and are
described in excellent texts such as Wahl and Bonner (1951) and Haissinsky (1964).
Some of these effects still are not clearly understood. Considered in this chapter
are issues pertinent to radioanalytical chemistry, including:
r Production of radionuclides
r Conventional equilibria values extended to very low concentrations
r Low-concentration “radiocolloidal” behavior
r Sample preservation

1
Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of
Technology, Atlanta, GA 30332
2
Radiobioassay Programs, General Engineering Laboratories, 1111 North Mission Park
Blvd. Suite #1070, Chandler, AZ 85224

64
 4. Radioanalytical Chemistry Principles and Practices 65

r Sample dissolution
r Carrier or tracer addition
r Separation for purification
r Source preparation for radiation measurement

The term “radiocolloidal” was applied in the early days of radiochemistry to

describe certain behavior patterns of radionuclides in aqueous solution that did
not conform to expectations based on the normally higher concentrations of the
chemical forms assigned to them. Because many of these deviations were related
to failure to pass through filters or to remain uniformly distributed in solution,
colloidal behavior was inferred. This behavior was observed to affect in different
ways many of the steps in the radioanalytical chemistry process itemized above,
with the potential of invalidating analytical results because the radioisotope did
not consistently follow the known path of its stable isotopes.
On the other hand, the low concentration of a radionuclide provides opportunities
for use as tracer in chemical and physical studies. In radioanalytical chemistry, one
benefit is that the addition of a stable-element “carrier” permits analysis without
the requirement of quantitative analyte recovery. Another benefit is the opportunity
to deposit very thin sources that minimize self-absorption in a source of alpha-
and beta-particle radiation.
Radionuclides with extremely long half-lives or in very large amounts will not be
at such extremely low concentrations, as indicated by Eq. (2.7). Radionuclides with
half-lives in excess of 109 years, such as 238 U and 232 Th, can be measured by some
of the conventional techniques of analytical chemistry as alternatives to radiation
measurement. Radionuclides at concentrations much above 104 Bq/l usually are
not submitted to a low-level radioanalytical chemistry laboratory because of the
threat of contaminating other samples.
Another distinction pertains to radionuclides that have no stable isotopes. Chem-
ical analysis for these radionuclides has no basis in conventional analytical chem-
istry except as studies performed with the usual small amounts, based on similari-
ties in chemical behavior to homologous stable elements according to their location
in the periodic table. When sufficiently large amounts of these radionuclides are
produced and purified to permit observation by microchemical manipulations, any
conclusions must consider the impact of the intense radiation on the observed
chemical reactions.
The atomic mass difference between the radionuclide and the mix of its stable
isotopes in nature, although minor in terms of its effect on chemical equilibrium
and reaction rates, provides opportunities for separation, identification, and quan-
tification at low concentration by mass spectrometer, as discussed in Chapter 17.
The mass difference ratio is at its extreme for tritium (T or 3 H) relative to the sta-
ble isotopes 1 H and 2 H. This distinction causes minor separation between ordinary
water with molecular mass 18 and tritiated water (HTO) with molecular mass 20
during distillation, and can be applied to enriching tritiated water in the laboratory
by electrolysis.
66 Jeffrey Lahr, Bernd Kahn, and Stan Morton

4.2. Effects in Producing Radionuclides

When a radionuclide is produced by fission or activation processes or in radioac-
tive decay of a radionuclide to its progeny, the product has a different form than
its antecedent and moves from its original site. That is, the nuclear reaction may
rupture the chemical bond between the radioactive atom and the molecule of
which the atom was a part, and the newly created radioactive atom may have sev-
eral new electron configurations. This result is described as the Szilard–Chalmers
effect.
The Szilard–Chalmers effect permits separation of radionuclides at high specific
activity and purity from the matrix in which they are produced. Preparation of
the pure product is an empirical process because of the complex interaction
of the three sequential steps: producing the free radioactive ion or atom, maintain-
ing the product in its new form in the sample matrix, and separating the product
from the matrix. Success of the process is evaluated empirically in terms of the
specific activity of the product relative to the matrix or, alternately, the fractional
yield of the product.
As discussed in Sections 2.2.1, the recoil energy of a product nuclide following
alpha-particle emission is about 0.1 MeV; in Section 2.2.2, it was noted that the
recoil energy observed following beta-particle emission is of the order of a few
kiloelectron volts. Recoil is also observed during gamma-ray emission and electron
capture and when an atom is produced in reactions such as (n,γ ) or (n,α) (Vertes
et al. 2003). The recoil energy of the produced radioactive atom generally is
sufficient to perturb its original position in a molecule or crystal.
The recoil energy is not the only factor that controls the degree to which the new
radionuclide separates itself from the original matrix, and some of the recoil energy
it is dissipated by other mechanisms. Moreover, the recoiling atom may return to
its original molecule or crystal site, especially when the product remains the same
element as its antecedent, as in an (n,γ ) reaction or after an internal transition. The
product nuclide and its antecedent nuclides may rapidly interchange to reduce the
specific activity of the product.
The last step in the collection process is to separate the product. Coprecipitation
and solvent extraction are common techniques for producing pure, high specific
activity, radionuclides such as chlorine or bromine. To keep the produced radionu-
clide separated at high specific activity, a chemical form is selected for the product
that has a slow rate of interchange with the product, such as certain oxyions or
organic compounds. Separation techniques that have been found especially con-
venient are sweeping a gaseous radionuclide from the medium and collecting a
recoiling radionuclide on a surface that faces a deposited solid.

4.3. Anticipated Low-Concentration Behavior

Analytical processes at the conventional concentrations discussed in Chapter 3
involve such concepts as the mass action law and the solubility product; these
tools can be applied to the analysis of radionuclides at extremely low, “trace-level,”
 4. Radioanalytical Chemistry Principles and Practices 67

concentrations with some precautions. For example, competing reactions with a

second substance, at too low a concentration to be identified but at a much higher
concentration than the radionuclide, can totally overshadow the expected reaction
of the radionuclide.
Some of the early studies of radionuclide behavior in solution and as solids
resulted in findings that could be explained by extrapolating the behavior of the
ions or molecules from the conventional to “trace-level” amounts or concentrations.
Other findings could be attributed to the unusual environment faced by the relatively
few radioactive atoms that were detected only because they emitted radiation.
Various anomalies in low-concentration behavior have been reported. In precip-
itation, a trace-level radionuclide that does not exceed its solubility product may
coprecipitate with a second substance. In electrodeposition, despite adjustment for
concentration (or more accurately, thermodynamic activity) by the Nernst equa-
tion, the predicted voltage may not be suitable because of the factors discussed in
Section 3.7 pertaining to overvoltage and an electrode surface not fully covered by
the element of the radioisotope (Haissinsky 1964). In volatilization or sublimation
of a radionuclide, distinctive effects have been attributed to the situation that atoms
at trace-level are not bonded to each other but to the other atoms in the medium
or on a surface. A radioactive cation sorbed on a negatively charged solid particle
that it does not fully neutralize will behave like a negative particle.
A radionuclide at trace level can be incorporated in a precipitate, either by
intention or inadvertently. In the absence of a visible precipitate, the radionuclide
may appear to act as a colloid (i.e., act insoluble and stratify or settle), as suggested
by the term “radiocolloid.”
Such precipitation or colloidal behavior occurs under conditions that would be
expected if the radionuclide hydrolyzed to form a hydrous oxide at higher con-
centration than inferred from radiation measurements. At a very small solubility
product K sp (see Section 3.1), any unknown small amount of stable ion in solution
may be sufficient to cause such an effect. Coprecipitation of trace-level radionu-
clides with another insoluble ion, such as Ra+2 with BaSO4 and Pu+3 with LaF3 ,
incorporates the radioactive atoms within the crystal structure in various ways or
sorbs it on particle surfaces (Kolthoff 1932). A precipitate such as Fe(OH)3 in
neutral or slightly basic solution can scavenge from solution many tracer-level
radionuclides that hydrolyze under the conditions of the procedure (see Table 3.1
for the effect of pH).
In practice, tests are needed to determine the fraction of a radionuclide that is
carried under the conditions of the process. For example, radioactive strontium
tracer is carried by BaCO3 in slightly basic solution and by Y(OH)3 in more
strongly basic solution. Such carrying generally is intended for the former but not
for the latter. Carrying by Y(OH)3 can be prevented by adding even a small amount,
such as 1 mg, of stable strontium ion. When no stable element is available, as for
plutonium, behavior due to hydrolysis must be avoided by strongly acidifying the
sample, possibly to as low pH as 2.
Much trace-level behavior has been understood only after detailed studies. In the
absence of visible evidence, behavior must be inferred from radiation measure-
ments during the reaction. In some cases, an observed effect can be attributed
68 Jeffrey Lahr, Bernd Kahn, and Stan Morton

to various mechanisms, as in the above-cited processes of coprecipitation as

mixed crystals or by surface sorption, colloid formation, or sorption on suspended
material, and retention on filters by conventional filtration or ion exchange on cel-
lulose. In each instance, the second effect has been termed “radiocolloidal” at one
time or another.

4.4. Sample Loss by “Radiocolloidal” Behavior

Samples submitted for analysis that contain radionuclides at extremely low con-
centration present a challenge to the analyst because these radioactive atoms un-
expectedly can deposit on surfaces. Such behavior has been widely observed and
was termed “radiocolloidal” (Hahn 1936). Radionuclides at trace levels are lost
from solution by sorption on container walls (Korenman 1968), suspended solids
(Wahl and Bonner 1951), immersed glass slides (Eichholz et al. 1965), metal foils
(Belloni et al. 1959), and filters (Granstrom and Kahn 1955). Filters of cellulose,
cotton wool, and glass wool retain various radionuclides at trace levels.
This loss is of greatest concern at the very beginning of the sample path, as when
a water sample is poured into a sampling bottle. Observations of losses stimulated
studies of radionuclide sorption on container walls of different types of glass and
plastics, pretreated or coated with reagents, and from aqueous solutions either
unmodified or with added carriers, acids, or complexing agents (Kepak 1971).
Washing containers with various solutions was tested for recovering some of the
sorbed radionuclides (Hensley et al. 1949).
A simple explanation for this sorption, although qualitative and incomplete, is
that a significant fraction of the radionuclide in solution is retained at a few con-
tainer and filter surface sites by ionic or electrostatic sorption (Kepak 1971). Frac-
tional retention varies among radionuclides, presumably reflecting their affinities
for the sites. Even a few available sites can be detected by radiation measurement
if they are of the same magnitude as the number of radioactive atoms in solution.
According to this explanation, the fractional sorption at these few sites becomes
immeasurably small when carrier is added to the sample. Various reagents can also
compete for the sites. Conversely, the fewer the competing ions, the greater the
fractional sorption (Thiers 1957); this is the case with deionized water. Pretreatment
of glass or plastic container surfaces can reduce radionuclide sorption by saturating
the sites; treatment by heating or washing glass with a basic solution, on the other
hand, can increase sorption.
Tests have shown that some glasses and plastics are better than others in min-
imizing surface sorption losses for specific radionuclides (Eichholz et al. 1965),
but the results are so variable that actual samples should be tested to assure that
loss is minimal for the samples and containers under consideration. Unfortunately,
test results for actual samples can be ambiguous at the low count rates of typical
samples, while tests with radioactive tracers at higher concentrations may not ap-
ply unless the tracers are in the same chemical form as the radionuclides in the
sample. Alternate methods that counteract the effect of sorptive behavior are listed
in Section 12.3.2.
 4. Radioanalytical Chemistry Principles and Practices 69

4.5. Sample Preservation

A liquid sample must be preserved so that the radionuclide in solution remains
at the same concentration during transportation and storage by preventing loss
or fractionation of the radionuclide. The only acceptable loss is by radioactive
decay for which Eq. (2.6) can compensate. Problematic sources of radionuclide
loss include volatile forms, the above-cited deposition on container walls, and
incorporation in suspended solids.
Filtration at the time of sampling eliminates suspended solids as a sorption
medium, but the filter may retain some radiocolloidal radionuclides. Addition
of acids, carrier, or other reagents can eliminate observable losses of specific
radionuclides by deposition or sorption. Prevention of transfer of radionuclide
state from soluble to insoluble or vice versa is important if measurements are
intended to distinguish oxidation states or degree of complexation in the original
sample, or predict radiation exposure to persons drinking the water. Any reagent
considered for addition to preserve one radionuclide must be evaluated for adverse
effect on preserving other radionuclides.
Sample preservation is also required to prevent bacterial action and changes
of form, such as souring of milk and decomposition of vegetation and biological
samples, as discussed in Sections 5.5–5.9. These changes can cause sample con-
tainers to leak or even explode, make chemical separation of radionuclides more
difficult, and change the form of the radionuclide. Sample storage on ice or in a
freezer is the common treatment. A preservative such as formaldehyde is added
when freezing is not an option.

4.6. Dissolution of Solids

Solids are dissolved so that they can be treated by conventional wet-chemistry
processes. The dissolution treatment must be tested to assure that no portion of
the radionuclide of interest is lost. The initial steps for organic solids are drying
and ashing. Radionuclides in a chemical form with high vapor pressure may be
partially lost by volatilization well below their boiling points. Radionuclides such
as gaseous tritium and 14 C forms can be collected during the drying and ashing
steps. Solids then are processed by either wet dissolution with acid or fusion with
solid reagents.
A sample prepared for radionuclide analysis generally should be completely
dissolved, without residual solids that could retain radionuclides. In contrast,
stable-element analysis permits residual solids if the dissolution process has been
demonstrated to dissolve completely the substance of interest.
Processing solids by leaching—i.e., incomplete dissolution of the matrix—
is acceptable for radionuclide analysis only if knowledge of the retention pro-
cess supported by leaching tests assures that the radionuclides of interest are
leached to a reproducible extent. Without positive knowledge that leaching re-
covers large and consistent fractions of the radionuclides of interest, results are
uncertain.
70 Jeffrey Lahr, Bernd Kahn, and Stan Morton

4.6.1. Acid Dissolution

As indicated in Table 4.1, most samples dissolve at least partially in strong mineral
acids such as HF, HClO4 , HNO3 , and HCl, used individually and in combinations.
Conventional aqua regia digestion consists of treating a geological sample with a
3:1 mixture of hydrochloric and nitric acids. Nitric acid destroys organic matter and
oxidizes sulfide material. It reacts with concentrated hydrochloric acid to generate
aqua regia:
3 HCl + HNO3 = 2 H2 O + NOCl + Cl2 (4.1)
Aqua regia is an effective solvent for most base metal sulfates, sulfides, ox-
ides, and carbonates. Some elements, however, form very stable diatomic oxides,
referred to as refractory species. Aqua regia provides only a partial digestion
for most rock-forming and refractory elements. Hydrofluoric acid can destroy
silicate matrices completely to liberate trapped trace constituents. Basic solu-
tions can dissolve tissue and many anionic forms of inorganic ions. Complex-
ing solutions such as EDTA are used under conditions that dissolve specific ions
(Perrin 1964).
Pressurized vessels in microwave ovens improve dissolution at elevated temper-
atures and pressures (Kingston and Haswell 1997). The advantages compared to
boiling solutions in beakers on hot plates are a shorter dissolution period under con-
trolled conditions of temperature and pressure with lesser quantities of reagents,
and—possibly more importantly—avoiding corrosion in hoods and ducts. One
disadvantage is a limitation on sample size due to vessel capacity for sample plus
reagents. Limited amounts, 1–3 g, of soil, biota, or vegetation are completely

TABLE 4.1. Acid dissolution

Dissolution Dissolution
Dissolution behavior behavior
Acid Character behavior (metals) (salts) (organics)
HCl Strong acid, Metals above H in the Oxides, sulfides, Organic salts
nonoxidizing electrochemical carbonates,
series phosphates
HNO3 Strong acid, strong Metals, nonferrous Metal sulfides Most organics
oxidizing agent alloys
H2 SO4 Strong acid, Metals Oxides, carbonates, Most organics
moderate phosphates
oxidizing agent
HF Weak acid, strong Metals Silicates Most organics
oxidizing agent
HNO3 /HCl Strong acid, very Metals including gold Most salts Most organics
(1/3) strong oxidizing and platinum
(aqua- agent
regia)
HNO3 /HF Strong acid, strong Metals Carbides, nitrides and Most organics
oxidizing agent borides of Ti, Zr,
Ta, and W.
 4. Radioanalytical Chemistry Principles and Practices 71

(>99%) dissolved by HF and HNO3 in 90-ml vessels in a few hours under pre-
scribed conditions (Garcia and Kahn 2001).

4.6.2. Fusion
Fusion is employed to decompose solids that are difficult to dissolve in acids but re-
act at high temperatures to form soluble compounds. To promote the fusion process,
a substance called “flux” is employed. Mixing the flux with the insoluble matrix
and applying heat can produce complete decomposition. A favorable reagent mix
and amount can be selected from published reports (Sulcek and Povondra 1989,
Bock 1979), but the sample matrix must be tested to confirm the effectiveness of
reagent choice and amount. Fusion can decompose environmental and biological
matrices (Sill et al. 1974, Williams and Grothaus 1984), even with high silica and
alumina content.
Fusion processes can be grouped into acid–base reactions (carbonates, borates,
hydroxides, disulfates, fluorides, and boron oxide) and redox reactions (alkaline
fusion agent plus oxidant or reductant). Common fluxes are listed in Table 4.2.
Fluoride-pyrosulfate and carbonate-bisulfate fusions are used to decompose soil
and fecal samples.
Disadvantages of the process are loss of volatile radionuclides at the elevated
temperature and the large amount of the fusion reagent added to the sample. Han-
dling the crucibles at the specified high temperatures is a safety concern. Platinum
crucibles are expensive, while other crucibles (e.g., nickel and iron) are attacked
by some reagents and contribute a contaminant to the sample.

TABLE 4.2. Common fluxes (Dean 1992)

Melting point Types of crucible Type of substances
Flux (◦ C) used for fusion decomposed
Na2 CO3 851 Pt For silicates, and silica-containing
samples; alumina-containing
samples; insoluble phosphates and
sulfates
Na2 CO3 plus an — Pt (do not use with For samples needing an oxidizing
oxidizing agent Na2 O2 ) or Ni agent
such as KNO3 ,
KClO3 , or Na2 O2
NaOH or KOH 320—380 Au, Ag, Ni For silicates, silicon carbide, certain
minerals
Na2 O2 Decomposes Fe, Ni For sulfides, acid-insoluble alloys of
Fe, Ni, Cr, Mo, W, and Li; Pt
alloys; Cr, Sn, Zn minerals
K2 S2 O7 300 Pt, porcelain Acid flux for insoluble oxides and
oxide-containing samples
B2 O3 577 Pt For silicates and oxides when alkalis
are to be determined
CaCO3 plus NH4 Cl — Ni For decomposing silicates in the
determination of alkali element
72 Jeffrey Lahr, Bernd Kahn, and Stan Morton

4.7. Carrier or Tracer Addition

The low concentration of radionuclides has stimulated near-universal application
of an isotope dilution technique (“reverse isotope dilution”) that permits mea-
surement of chemical recovery, termed “yield.” As commonly applied outside the
radioanalytical chemistry laboratory, isotope dilution consists of the addition of a
known amount of radionuclide tracer to a sample that contains its stable element,
i.e., the natural mixture of stable isotopes of the element. The tracer is added at the
beginning of the procedure and then measured in the separated and purified sample.
If the added tracer and the stable isotopes of interest are assured at the beginning
to be in the same chemical and physical form, then the fraction of recovered tracer
represents the fraction of recovered stable isotope.
Application of this technique avoids the need for quantitative recovery, which
enormously simplifies the analytical process. Description of the isotope dilution
technique in Section 4.7.1 is used to frame the discussion of reverse isotope dilution
with carriers and tracers in Sections 4.7.2 and 4.7.3, respectively.

4.7.1. Isotope Dilution Technique

Isotope dilution begins with the addition of a known amount of calibrated radioac-
tive tracer solution Ri to a sample of mass m i in solution. By measuring the mass
recovered m f and the radionuclide amount at the end of the procedure, Rf , one
calculates the initial mass. Designating Si and Sf as the specific activity (i.e., the
ratio of radionuclide amount to the mass of the same element, at the beginning
and end, respectively) and m Ri as the mass associated with the radionuclide tracer
initially gives the following relationships:
Ri
Si = (4.2)
m Ri
and
Ri Rf
Sf = = (4.3)
m Ri + m i mf
The last term in Eq. (4.3) holds because the specific activity is a ratio that re-
mains constant throughout the procedure, if radionuclide and solid have perfectly
interchanged when combined. Then
 
Si
m i = m Ri (4.4)
Sf − 1
To calculate m i if m Ri  m i
Ri Rf
= (4.5)
mi mf
In typical isotope dilution, one calculates m i = m f Ri /Rf .
 4. Radioanalytical Chemistry Principles and Practices 73

4.7.2. Carrier Addition

For the purpose of radiochemical analysis, one uses reverse isotope dilution by
adding a known mass of stable carrier solution to the radionuclide sample solution,
instead of adding a radionuclide tracer to the stable element. The mathematical
expression is identical to Eq. (4.5):
m i Rf
Ri = (4.5a)
mf
Here, m f /m i is the yield.
To assure accuracy, the carrier reagent is calibrated by mass measurement iden-
tical to that of the sources prepared for counting. For confirmation, the weight of
the precipitate at 100% yield can be predicted from the amount of carrier reagent
dissolved in the carrier solution, the amount pipetted into the radionuclide solution,
and the ratio of the molecular weight of the precipitate to that of the carrier reagent.
The amount of carrier added to the radionuclide solution conventionally is se-
lected for ease of weighing with an accuracy of 1% or better for yields of 50–100%.
Low carrier amounts for processing thin sources for alpha-particle spectrometry
(see Section 7.2.1) may require specially selected purification and yield determi-
nation techniques.
Under the following circumstances, special efforts are required for yield deter-
mination:
r A significant amount of the chemical carrier is in the original sample, that is, m Ri
is not far less than m i . This circumstance becomes obvious when yields exceed
100%. The amount in the sample must be measured before radionuclide analysis.
If stable isotopes of the radionuclide of interest are found in the sample, each
yield factor must be corrected for the stable isotopes initially in the sample by
adding their weight, m Ri , to the carrier weight, m i . If the amount m Ri is marginal,
it may be possible to increase the carrier weight so that the weight originally in
the sample becomes negligible.
r The radionuclide and its carrier have more than one chemical state. When mul-
tiple oxidation states are present, sufficient oxidation and reduction reactions
must to be performed after the carrier is added to the radionuclide solution to
ensure that the carrier and radionuclide are in a common state. If a fraction of the
radionuclide is solid or gaseous, that fraction must be dissolved. If the radionu-
clide has formed a complex molecule or ion, sufficient reagent must be added to
complex the carrier as well.
r No stable isotopic carrier is available (e.g., for promethium, technetium, radium
and its progeny, and the actinides and transactinides). Either nonisotopic carrier
or isotopic radioactive tracer can be substituted (see next section). The special
problems that arise are discussed in Section 7.2.2.
r Some processing steps do not include the carrier. If the sample initially is a
solid, the carrier may be added after dissolution. This will require prevention of
radionuclide loss during dissolution. If the carrier is added to the solid, then the
problem becomes that the carrier in solution and the radionuclide in the solid
74 Jeffrey Lahr, Bernd Kahn, and Stan Morton

may not respond identically to dissolution. Another gap may occur at the end of
the procedure if carrier yield is determined by a method other than precipitation,
such as spectrometric measurement of an aliquot, or if the precipitate is dissolved
for liquid scintillation counting. In such cases, the last step must be quantitatively
controlled.

4.7.3. Tracer Addition

A known activity of a second radionuclide may be added to the solution as a
tracer for the radionuclide of interest when no stable carrier is available, or if
measuring the tracer is more convenient than measuring the carrier. Preferably,
both radionuclides are isotopes of the same element, e.g., 85 Sr tracer for 90 Sr
measurement. The initial count rate of the radionuclide of interest is calculated
from its final count rate by Eq. (4.5a), taking m i and m f to be the initial and final
net count rates, respectively, of the tracer.
As with the carrier, the tracer must be mixed with the radionuclide of interest so
that they behave identically in every step of the procedure. If the possibility exists
that the chemical form of the added tracer differs from that of the radionuclide
of interest, both must undergo sufficient chemical reactions to achieve complete
interchange of the two radionuclides.
The tracer and its radiation detector must be selected to avoid cross-talk between
their radiations in measuring the two radionuclides. A tracer must be selected that
emits different radiations than the radionuclide of interest, or emits the same type
of radiation at an energy sufficiently different for resolution by spectral analysis.
Use of radioactive tracer is common for actinides that are measured by alpha-
particle spectral analysis. A correction factor may be applied if cross-talk cannot
be entirely avoided but is small enough to maintain the reliability of the activity
calculated for the radionuclide of interest.

4.8. Sample Purification

Carrier addition is widely used, not only for yield determination, but because the
added mass avoids problems of radiocolloidal behavior. Other advantages of added
carrier are the ability to use precipitation for radionuclide separation and purifica-
tion, and avoidance of unintended coprecipitation during scavenging. When carrier
is not added because no stable isotope is available or the counting source must be
very thin, radionuclide deposition on container walls or on suspended solids must
be avoided by applying the techniques discussed in Section 4.4, notably use of
nonisotopic carrier or low pH.
The need for carrier addition to achieve radionuclide precipitation or prevent ra-
diocolloidal effects is shown by applying the discussion in Section 3.2 and Eq. (3.6)
to the precipitation of 129 I. The amount of 1 Bq carrier-free 129 I is 2.1 × 10−15
mmol, and its concentration in 0.035 l of solution is 5.9 × 10−14 mmol l−1 . No
precipitate is expected in a solution with silver nitrate added to a concentration of
 4. Radioanalytical Chemistry Principles and Practices 75

28 mmol l−1 because the product of the concentrations of the two reagents of 1.7 ×
10−12 (mmol l−1 )2 is less than the solubility product for silver iodide of 8.49 ×
10−11 (mmol l−1 )2 . The precipitate may form, however, if even small amounts of
additional iodide were in the water from natural or man-made sources, or if the
thermodynamic activities exceeded component concentrations because of other
salts in solution.
Although chemical reactions of radionuclides are assumed, for practical pur-
poses, to be identical to those of their natural stable isotope mixtures, minor dif-
ferences in equilibrium distribution exists, notably in the case of tritium relative
to stable hydrogen, but also to a small extent for others, such as 14 C relative to
stable carbon. During water distillation to purify it for tritium analysis, the tri-
tium concentration of the residue is enriched by about 10% relative to the vapor.
Only when all of the water is distilled will the specific activity of tritium (and
the concentration in water, in units of Bq l−1 ) be the same in the distillate and
the sample (Baumgartner and Kim 1990). Moreover, because of such differences
in equilibrium constants, the tritium concentration can be enriched approximately
10-fold by electrolysis of a water sample from which the produced hydrogen gas
escapes into the air (NCRP 1976a).
Chemical procedures for purifying radionuclides that have no stable isotopes
are based on studies of radionuclide behavior performed over the years. A crucial
aspect of analytical procedures for these elements is addition of reagents that
maintain the radionuclide in the redox state appropriate for the separation step.
The essential tools of radioanalytical chemistry practice are the 50-ml glass
centrifuge tube and the stirring rod that are used for precipitation separations.
Carriers are added with pipettes, and reagents, from dropper bottles. Heat is applied
by a Bunsen or Meeker burner, and solutions are cooled in an ice bath. Plastic is
substituted for glass for convenience if no direct heat needs be applied, and by
necessity when hydrofluoric acid is used. The precipitate is separated from its
supernatant solution during the purification stage by centrifuging, and in the final
stage, on a 2.5-cm-diameter filter in a filtration apparatus.

4.9. Counting Source Preparation

Conventional source preparation is by carrier precipitation. In the absence of iso-
topic carrier, a nonisotopic carrier can carry the radionuclide when both are in-
soluble under identical conditions. Correction for the yield of the radionuclide
may be necessary to adjust for the different solubility product of the nonisotopic
carrier. The nonisotopic carrier for source preparation may be a different carrier
than for purification, i.e., different nonisotopic carriers may be used in a sequence
of precipitation separations.
As discussed in Chapter 7, a source may also be deposited by volatilization
or electrodeposition for alpha or beta particle counting by Si diode or propor-
tional counter, or added as a solution or suspension to scintillation cocktail for
liquid scintillation counting. For electrodeposition, as indicated by Eq. (3.20), the
76 Jeffrey Lahr, Bernd Kahn, and Stan Morton

half-electrode equilibrium potential depends on the tabulated standard reduction

potential, the equilibrium concentration of the radionuclide of interest, other fac-
tors associated with the reactions of water and other ions in solution, and on the
characteristics of the electrode surface.
For example, the standard potential of Ag+ is 0.799 V. If the concentration of
carrier-free 110m Ag in solution is about 2 Bq l−1 , the concentration of silver is about
10−16 M initially and possibly 10−17 M at equilibrium. For the value of log (Ag+ )−1
of 17, the added potential 0.059 × 17 is 1.00 V. In fact, the concentration of Ag+
in the solution may be orders of magnitude higher because silver in nature, in
the laboratory, and accompanying the radionuclide will reduce the concentration-
related voltage proportionately. The effect on the voltage by these factors must be
measured for the solution processed and the electrode in use.
Radionuclides at very low concentrations that have a large value of E 0 deposit
spontaneously on metals that are readily oxidized in the solution. For example,
polonium is routinely purified and prepared for counting by spontaneous deposition
on a nickel planchet. Spontaneous deposition has been demonstrated with other
metals (Blanchard et al. 1960). Polarographic separation has been achieved by
reducing a radionuclide to an amalgam in a drop of mercury (Kolthoff and Lingane
1939).
An alternative electrochemical source separation and preparation technique is
electrophoresis. A drop of the radionuclide solution is pipetted at one end of a
moistened paper strip that has an electric potential difference along its length. Each
ionic radionuclide moves at its own rate down the strip. Locations of individual
radionuclides are identified by their radiation or by dyes affected by the chemical
form (Deyl 1979).
Loss of a fraction of a radioactive daughter from a radioactive source can be
expected due to the recoil discussed in Section 4.2. For example, the radioactive
daughter 224 Ra from a very thin source of 228 Th can transfer to a facing surface
by recoil. This process can be applied beneficially to prepare a pure source of
the radioactive daughter or can be a problem in contaminating the detector during
alpha-particle spectrometry.
A similar situation is presented by a sample of soil or rock from which the
radioactive progeny 222 Rn of 226 Ra leaves as a gas. The gas may be collected for
counting as a purified radionuclide or its loss can cause an erroneous result in
counting the sample by gamma-ray spectral analysis.
A different mechanism is attributed to the observed loss of 210 Po deposited on
a metal backing. The 210 Po spreads over an extended area (Sill and Olson 1970),
presumably because it has sufficient vapor pressure in the deposited metallic or
hydride form.
5
Sample Collection and Preparation
ROBERT ROSSON

5.1. Introduction
Sample collection and preparation are inextricably linked to the practice of radio-
analytical chemistry, despite the fact that two separate groups will perform these
activities. Ideally, the radioanalytical chemist is part of the team that plans the
sampling effort and prepares the quality assurance project plan (see Section 11.1).
Such participation in the planning phase benefits the entire analytical process. The
radioanalytical chemist can tailor the analytical approach to both the sample media
and the radionuclides in the media. Information on the purposes for which the lab-
oratory produces data can enable the radioanalytical chemist to devise laboratory
practices to match required detection sensitivity, sample submission rate, and re-
porting style. On the other hand, the analyst can describe the capabilities and limits
of the laboratory to the planning group. Such participation in the planning process
can prepare a cohesive sampling and laboratory effort that produces defensible
results for the client.
To serve as an effective member of the team, the radioanalytical chemist must
have some knowledge of sampling methods pertaining to various matrices. The
planning dialogue can influence specifications of sample collection by location,
frequency, size, and techniques, particularly when the suite of radionuclide samples
is from a complex environment. Sections 5.2 to 5.9 also briefly describe protocols
for sample preservation between collection and analysis.

5.2. Sample Information

Sampling and sample treatment should be considered in terms of the data quality
objectives (DQO) approach (Section 11.3.1), which integrates the monitoring pro-
gram, sample collection program, radioanalytical chemistry program, and report-
ing of results. The magnitude of such initial planning and iterative improvements
must be matched to the size of the monitoring program, with more extensive efforts
for a larger program.

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

77
78 Robert Rosson

Those planning the sampling effort should initially consider the following:
r The radionuclides produced or used in the facility or project
r Radionuclide activity levels
r Variability in radionuclide activity with time and location
r Chemical forms of the radionuclides
r Sampling circumstances, e.g., routine operation or specific incidents
r Radionuclide transport by various media.
These initial items of information guide sampling location, volume, and initial
preservation steps. Implicitly, they also guide laboratory selection; incident re-
sponse will likely require a quick turnaround time and the capability of handling
higher radioactivity levels (see Section 13.1), which narrows the choice of appro-
priate laboratories.
Once the sample has been taken, and a laboratory chosen, this information
goes with the sample as part of the chain of custody documentation. When the
laboratory accepts the sample, the sample information is reviewed to extract those
facts directly pertinent to the analytical process:
r Sample form and matrix
r Analytical requirements
r Format for results
r Sample load

The sample form and matrix type control the initial sample treatment in the labora-
tory. The analytical requirements dictate the chemical and instrumental procedures.
The format specifies units for reporting values, uncertainty, and detection limits,
and the context in which this information will be reported. The sample load controls
laboratory staffing, scheduling, and turnaround time.
The DQO checklist for the laboratory is a comprehensive compilation of all
these pieces of information. It will include descriptions of the following:
r Sample type
r Description of site parameters (e.g., flow rate of the sampled medium)
r Sample collection, storage, and preservation methods
r Sample matrix, including the radionuclides expected and the radionuclides of
interest
r Predicted radionuclide concentrations
r Required detection limits
r Available radiation detection instruments
r Available radioanalytical chemistry methods
r Estimated detection limits for selected method and instruments
r Requirements of sample size and counting times to meet specifications
r Personnel skill, time, and cost estimate
The usual types of samples are described in Table 5.1 and the following sections;
their analysis is discussed in Section 6.2. Some of these sample types are obtained
with sampling equipment such as air and water collectors placed at selected
 5. Sample Collection and Preparation 79

TABLE 5.1. Sample storage and preservation

Sample type Collection Storage container Preservation
Air filter 5-cm-diameter Filter Petri dish None
Air cartridge Charcoal cartridge Ziplock bag None
Air tritium Desiccant column Air tight container None
Water Bottle Gallon cubitainers Add acid, base, or
complexing agent
Milk Grab Gallon cubitainers Refrigerate or add
formaldehyde
Vegetation Grab Ziplock bag Refrigerate or freeze dry
Biota Grab Ziplock bag Refrigerate or freeze dry
Soil, sediment Grab Wide-mouth jar Refrigerate or freeze dry
Urine Grab Gallon cubitainers Refrigerate, or add acid
or complexing agent

locations (Budnitz et al. 1983), while other samples are collected by hand, as
in the case of soil and vegetation. Particularly important for hand-collected sam-
ples is the reported description of sampling site and process in sufficient detail
to permit comparisons among sampling periods and locations. All information
regarding the sample should be reported on the chain of custody form (see Section
11.2.5) that accompanies the sample to the laboratory.
Identifying the type of sample matrix and radionuclide of interest helps to de-
termine a pretreatment protocol. Some form of pretreatment is usually required
for environmental samples. The half-lives of both the radionuclides of interest and
interfering radionuclides must be considered to decide how quickly to perform
the analysis. The radiation type and energy must be known to select the radiation
detector, and possibly the radioanalytical chemistry method to prepare the source
for counting.
The concentration of the radionuclides of interest may be inferred from the sam-
pling location. Natural and man-made background radiation values are summa-
rized by the National Council on Radiation Protection and Measurements (NCRP
1987a) and the United Nations Scientific Committee on the effects of atomic ra-
diation (UNSCEAR 2000a,b).
Estimation of the detection limit of the measurement instrument in the time
period available for counting, linked with the required detectable concentration and
the expected concentration of the radionuclide to be determined, guides selection
of sample size for the analysis. Calculation of the minimum detectable activity
is discussed in Section 10.4.2. Additional documents of interest are Altshuler
and Pasternack (1963), Pasternack and Harley (1971), and Currie (1968). The
terms minimum detectable concentration and lower limit of detection also have
widespread use. A document that addresses these and other topics pertinent to
radiation monitoring and measurement was developed by a committee of the Health
Physics Society (EPA 1980a).
All sample information must be taken into account to develop an estimate of
the time and resources necessary to process the sample to the satisfaction of the
80 Robert Rosson

client, as discussed in Section 13.9. For example, an estimate of needed sample

volume to obtain valid result with the detector at hand may show that the proposed
sample is too large to accommodate the separation procedure unless pretreated to
reduce the volume and measured for a longer period in a low-background system.

5.3. Atmospheric Samples

Air sampling is required around nuclear facilities and in populated areas that may
be exposed to elevated radionuclide levels to evaluate radiation exposure from
external or inhaled radionuclides. Airborne radionuclides may be in the form of
a gas, vapor, or particles. Different sampling techniques are employed depending
on the radionuclide of interest, its form, and the sample volume required to reach
the detection limit. Samples may be collected at fixed stations or from vehicles
moving on the ground or in the air.
Air sampling equipment consists of a framework or housing, a sample collector,
a collector holder, an air pump, and control and recording equipment, including
a flow control device, flow-rate meter, timer, and data recording or transmitting
system, as shown in Fig. 5.1 (see Sections 15.2. and 15.4 for examples). The
system should provide structural support, shelter, and a tight connection for the
collector. The air pump must provide the required airflow. The system must be
maintained for continuous operation and protected from vandalism. The number
and location of stations must be selected to provide the information for which they
are operated, e.g., to detect and quantify airborne radionuclides that are released
at a site or distributed across an area.

5.3.1. Particulate Radionuclides

Airborne particles commonly are collected from air on filter media such as organic
membranes, paper (cellulose fiber), or glass fibers. Membrane filters are fragile and

FIGURE 5.1. Air sample collector.

 5. Sample Collection and Preparation 81

can be used only with low-flow-rate systems, while paper and glass fiber filters are
sturdier and more easily handled. These filters have high collection efficiency for
particles with sizes in the respirable range, i.e., those with an aerodynamic median
diameter of around 0.3 μm (Lockhart et al. 1964). All except membrane filters
have low-pressure drops that permit high flow rates for collecting large volumes
of air to achieve increased measurement sensitivity. The filters can meet standard
practice requirements of 99% removal of respirable particles at the operating air
velocity and pressure drop (NCRP 1976b, Corley et al. 1977, EPA 2004).
In view of the influence of particle size on inhalation radiation exposure, some
collectors of discrete particle sizes are deployed, typically for research rather than
routine monitoring. Impactors with metal plates or Mylar foils that are sprayed
with silicone to reduce particle bounce (ACGIH 2001) have proven useful, but can
challenge radioanalytical chemistry preparations. An aerosol spectrometer with a
series of metal screens (EPA 2004) may present less of a chemical preparation
problem.
Collector location and collection period must be selected to avoid excess mass
loading. Excessive deposition of matter on the filter will reduce the air flow-rate,
change retention efficiency, and result in some collected solids falling off when
the filter is handled.
The duration of sampling (start and end times) and the flow rate through the filter
must be recorded. The type of pump, the kind of filter substrate (cellulose, glass
fiber, polystyrene, etc.), and even the housing for the sampling station should be
recorded because these factors can affect data application. Interruption in sample
collection or the air-flow record should be avoided because the former causes a
gap in the monitoring program and the latter makes the radionuclide measurement
pointless because the air volume is unknown.
Radionuclides in particulate samples collected on filter substrates usually are
measured first by gamma-ray spectral analysis of the filter enclosed in a thin-walled
envelope. Gross alpha- and beta-particle analysis by low-background proportional
counter follows (see Section 7.2.4). The particles collected from the air stream
include radioactive progeny of natural 222 Rn and 220 Rn. These radionuclides always
present a radiation background to any targeted measurement of radionuclides. To
eliminate interference from these progeny in counting, the filters are held for a time
between collection and counting to allow for their decay; the short-lived daughters
of 222 Rn decay mostly in 2 h and those of 220 Rn decay in 4 days. The long-lived
progeny of 222 Rn—210 Pb, 210 Bi, and 210 Po—and cosmic-ray-produced 7 Be remain
on the filters after 4 days.
A specified fraction of the filter is taken for radiochemical analysis. The first
step is filter dissolution, as discussed in Section 6.2.1. The remainder of the filter is
archived for possible future analyses. The archive should also contain unexposed
(blank) filters.
Airborne particulate radionuclides are also collected in deposition trays. An
upward-facing adhesive “gummed film” in the tray has been used to retain the de-
posited solids during the collection period (NCRP 1976b). The tray collects solids
during dry periods and precipitation events unless a sliding cover is activated to
close the tray during precipitation. The cover can be moved by a sensor of increased
82 Robert Rosson

conductivity that causes the cover to close the tray at the beginning of precipita-
tion and opens it after the end (Chieco 1997). The measured deposition during dry
periods (for a collector with a cover) or during the entire period (for a collector
without cover) can be compared to the results measured for a rainwater collector
filter discussed in Section 5.4.3. The particles collected during precipitation may
include some dry deposition that the rainwater washed out of the collector and
onto the filter.

5.3.2. Gaseous Radionuclides

The usual gaseous radionuclides are tritium, 14 C, radioiodine, and noble gases.
Tritium (written below as T) can exist as water vapor, hydrogen gas, and organically
bound compounds such as CH3 T. Carbon-14 can be found in CO2 , CO, and organic
gases. Gaseous radioiodine is believed to exist in three forms: as I2 , HIO, and
organic-bound such as CH3 I. The noble gases include neutron-activated 41 Ar,
several fission products of krypton and xenon, and naturally occurring 222 Rn and
220
Rn.
Tritium collection. Tritium in air is usually in the form of water vapor and less
commonly in the elemental or organic-bound forms. It is generated in nature by
cosmic-ray interactions, and at nuclear reactors and tritium-production facilities
by ternary fission and neutron activation. Tritium as HT tends to oxidize to
water vapor in air. Conversion to and from organic-bound tritium occurs in biota
(NCRP 1979).
Tritiated water vapor is collected by condensing or freezing it, by bubbling air
through water to exchange tritiated water with nontritiated water, and by sorption
on silica gel or molecular sieve (Corley et al. 1977). A standard method calls for
drawing filtered air through a silica gel column at a rate of 0.10–0.15 l min−1 ,
as shown in Fig. 5.2 (APHA 1972). Sufficient silica gel must be in the column

Glass wool

Silica gel

Glass wool FIGURE 5.2. Silica gel column for collecting tritiated
water vapor.
 5. Sample Collection and Preparation 83

to sorb the maximum humidity expected for the volume of sampled air. Column
dimensions must accommodate this silica gel volume and the selected air flow
rate. The collected water or used silica gel is taken to the laboratory for tritium
analysis (see Section 6.4.1).
Special systems are used to collect separately the tritiated hydrogen gas (HT)
and organic-bound tritium from air. A two-component collection train consists of
a molecular sieve column, followed by addition of a stream of H2 carrier gas to the
sampled air before it passes through a palladium-coated molecular sieve column.
The first column collects tritiated water and the second column collects the two
other forms (Ostlund and Mason 1985). The two columns are treated separately at
the laboratory, as discussed in Section 6.2.2, to provide the three forms for separate
analyses.
Carbon-14 collection. Carbon-14 is formed in nature by cosmic-ray interactions.
It is in all carbon-containing compounds that are in equilibrium with 14 C in air at
a specific activity of 0.23 Bq g−1 carbon. Concentration measurements in carbon-
containing compounds that are no longer at equilibrium with air, such as dead trees,
are used to determine their “age”—the time period since the end of equilibrium
with airborne carbon—in terms of the fractional radioactive decay. Fluctuations of
cosmic-ray production of 14 C in air over the centuries must be considered in this
determination (NCRP 1985a). Carbon-14 is also produced at low rates in nuclear
reactors, mostly by the (n,p) reaction with 14 N and the (n,α) reaction with 17 O.
Collection of 14 C in air generally is a research effort. A large volume of air is
pumped through a collection train that consists of a bubbler with barium hydroxide
solution and a tube that contains aluminum–platinum (0.5%) catalyst, where it is
heated to 550◦ C, and then pumped through a second bubbler with barium hydroxide
solution. Air passes through the first solution, in which 14 C in the form of CO2
precipitates as barium carbonate and then passes with added CO carrier gas through
the oxidation tube in which 14 C as CO or CH4 is converted to CO2 . The gas finally
passes through the second solution in which the newly formed CO2 precipitates as
barium carbonate. The two 14 C samples are taken to the laboratory for measuring
the emitted beta particles (Kahn et al. 1971).
Radioiodine collection. Monitoring for radioiodine is associated with fallout,
nuclear facilities, and medical use. The most common radioiodine is 131 I, but
shorter-lived 132 I, 133 I, and 135 I are also fission products, and others, such as 125 I,
are also used in nuclear medicine. Some long-lived 129 I is produced in nature and
some is formed in fission and released when processing spent reactor fuel. Iodine
in air can be both gaseous and particulate. The existence of radioiodine in a number
of chemical states complicates sample collection and analysis.
The typical environmental airborne radionuclide collection train consists of a
filter followed by an activated-charcoal cartridge; particulate iodine is retained on
the filter and gaseous iodine on the cartridge (Corley et al. 1977). A more elaborate
collector train distinguishes the various forms of radioiodine. The collectors, in
order, are an air filter for particles, cadmium iodide on Chromosorb-P for I2 , 4-
iodophenol on alumina for HIO, and activated charcoal for organically bound
iodine, such as CH3 I (Keller et al. 1973).
84 Robert Rosson

Collection efficiency for radioiodine from the air stream by activated charcoal
is subject to type of charcoal, size of the granules, depth of the sampling bed, and
flow rate. The grain size should be in the range 12–30 mesh and the sampling rate,
0.03–0.09 m3 m−1 , for optimum collection (APHA 1972.). Air streams can also
be monitored for radioiodine with silver zeolite and molecular sieve, but activated
charcoal is the commonly used sorbent largely due to lower cost.
Noble gas collection. The radioactive argon, krypton, and xenon radioisotopes
are monitored near nuclear power plants and related facilities to determine the
magnitude of releases of gases generated by fission or neutron activation. Radon
radioisotopes and their particulate progeny are measured in homes and mines to
determine whether their airborne concentrations are below radiation protection
limits.
Grab sampling techniques for noble gas radionuclides include collection in
a previously folded large plastic bag or a previously evacuated sampling flask.
Low-volume samples can be taken with a hand pump; larger volume samples are
collected in metal containers under pressure of 10–30 atm (Corley et al.1977).
Flowing air is sampled for noble gases by sorption on cooled activated charcoal
or by cryogenic condensation.
Radon in air can be collected and measured with an evacuated Lucas cell (see
radium analysis in Section 6.4.1). The cell is taken to the sampling site and filled
by opening the stopcock. After closing the stopcock, the cell is returned to the
laboratory and counted between 2 h and a few days after collection. Radon is also
collected on charcoal cartridges that are measured directly by gamma-ray spectral
analysis. Radon progeny are collected on filters. They can be measured to esti-
mate radon parent concentrations. Numerous in situ detectors of radon and radon
progeny, including ionization detectors, proportional counters, and scintillation
detectors, are in use (EPA 2004).

5.4. Water
Types of samples include rain, surface water, ground water, drinking water, seawa-
ter, and wastewater. The various types have different chemical constituents, solids
content, and pH that may require distinctive collection and analysis techniques.
Information should be obtained on the expected radionuclide content of the water
to guide initial processing, sample preservation, and any requirement for rapid
shipping and prompt analysis.
Practical experience suggests that water is the easiest matrix to process. Prob-
lems with water samples include the distinction between dissolved and suspended
radionuclides and the preservation of dissolved radionuclides in the sample in
their original form. The sampling location and process can modify the sample; for
example, radionuclides in water may deposit on pipe and valve surfaces and later
may dissolve again.
Filtration of water samples at collection removes from the water the suspended
and colloidal fractions (for separate analysis, if desired) but some dissolved
 5. Sample Collection and Preparation 85

radionuclides may be retained on the filter and associated particulate matter. If

filtration is postponed until the sample reaches the laboratory, some radionuclides
may in the interval transfer between the water and suspended materials. This prob-
lem does not apply to samples from public water supply systems for water that
already has been treated to remove suspended solids.
Sorption of radionuclides onto the container walls is a common problem. The
problem does not arise for all radionuclides and types of water, but its occurrence
must be checked and may be prevented as discussed in Section 4.5. The usual
sample preservation technique is mild acidification, but stronger acid may be nec-
essary to preserve transuranium radionuclides. A basic solution must be used for
radioiodine.

5.4.1. Grab Sample Collection

Grab samples provide initial information and periodic checks for surface water
or water supply systems. The simplest collection from surface water is bottle or
dipper submersion. These methods are effective for collecting water near the top
of the water column. To collect grab samples far below the surface, one needs an
extendable tube sampler with a check valve to fill the tube with the sample.
A bailer can be used for surface water collection but is primarily intended for
collecting ground water. When the bailer is lowered into the water, liquid flows
into the bailer through the bottom check valve; when the bailer is retrieved, the
check valve closes to retain the water. Other types of collectors are also available
(Brown et al. 1970).

5.4.2. Continuous Surface Water Sampler

Unattended continuous samplers are used for ongoing radionuclide measurements.
A small fraction of the surface flow is pumped into the sampler. One type of
automated sampler (see Fig. 5.3) collects samples at regular intervals of minutes
to days in a set of sample containers. A single container can be placed at the
collector to receive a composite sample.

5.4.3. Rainwater Collector

A rain pan collector accumulates a rainwater sample deposited on a known area.
The collector funnels rainwater into a sample bottle. A filter followed by cation-
and anion-exchange resin columns can be inserted in the line between the sam-
pling pan and collection bottle, as shown in Fig. 5.4. The filter collects particulate
radionuclides, the resins collect cationic and anionic radionuclides, and nonionic
radionuclides remain in the water. The pan collects both dry and wet depositions.
Collectors with movable shields have been developed that open during rainfall,
as discussed in Section 5.3.1. The pans are washed after each collection period
to remove the radionuclides associated with dry deposition and clean the pans for
86 Robert Rosson

FIGURE 5.3. Automated surface water sample collector.

the next collection period. Acid or another preservative is added to newly placed
collection bottles.
Depending on the program, rain samples are collected weekly, monthly, or quar-
terly. An adjacent rain gage is desirable to record precipitation data, or the collector
may be placed near a meteorological station so that radionuclide measurements can
be related to weather data. Radionuclide data are reported relative to the amount
of water, area of deposition, and rainfall for the period.

5.4.4. Groundwater Collection

Radionuclides in groundwater are identified and quantified to determine sources of
discharges or leaks into the ground. They can indicate the presence of radionuclides
in subsurface soil, and predict radionuclide concentrations in surface water reached
by groundwater. Groundwater usually is collected from wells that have been drilled
 5. Sample Collection and Preparation 87

FIGURE 5.4. Rainfall collector. Rain pan

Glass wool packing

Ion-exchange resin

at locations and depths selected on the basis of groundwater flow models. Well
drilling and water sampling must follow established guidelines to yield acceptable
samples (Wood 1976, EPA 1986, Shuter and Teasdale 1989). Several wells may
be bundled at a given site to reach various depths. The number of wells drilled
is limited by their cost. A different sampling system consists of collecting water
from existing supply wells. The frequency of sampling is related to the estimated
subsurface flow rate; for example, if the estimated flow rate is a few meters per
year, only an annual sample is needed.
Samples that represent groundwater are collected after several well volumes have
been discarded unless little water is available. The sample is acidified promptly
for preservation. If one of the radionuclides is not stable in acid, a second sample
is collected and preserved as appropriate.

5.5. Milk
Milk is frequently analyzed for radionuclides in a monitoring program because
it is one of the few foods that reaches the market soon after collection. Milk
commonly is consumed within 7–11 days after milking and occasionally reaches
the consumer within 2 days. It may contain relatively short-lived radionuclides and
be a dietary source of fission-produced 89 Sr, 90 Sr, 131 I, 137 Cs, and 140 Ba, as well as
naturally occurring 40 K. A 4-l sample usually is collected for analysis. Cow’s milk
samples can be collected to represent a specific herd in the form of raw milk, or a
regional pool of pasteurized milk. Goat’s milk is collected when this medium may
88 Robert Rosson

be a significant source of radionuclide intake by humans. Powdered milk may be

analyzed for the longer-lived radionuclides 90 Sr and 137 Cs.
Milk samples should be refrigerated and delivered to the laboratory within 1–
2 days. For longer storage, the milk sample should be frozen or a preservative
added such as methiolate, formaldehyde, or thimerosal (DOE 1987, EPA 1984),
as indicated in Table 5.1. Preservatives may interfere with radioiodine separation
(Murthy and Campbell 1960), but can be added if radioiodine is measured directly
by gamma-ray spectral analysis.

5.6. Vegetation and Crops

Decisions concerning sampling vegetation and crops include selection of sites,
type, and collection technique (e.g., height above ground for cutting grass). Food
may be obtained from a private garden, a store, or a distributor. Collection protocols
are mainly site specific; crop selection depends on local growth and consumption,
guided by availability, location, season, and farming practice. Radionuclide path-
ways to vegetation and crops are surface deposition and root uptake.
Radionuclides in grasses and leaves are analyzed as indicators of radionuclides
in the environment and links in the food chain to animals that feed on them.
Food (crop) samples are analyzed to relate radionuclide concentrations to human
radiation doses due to consumption. Studies of radionuclide pathways from source
to consumer can identify the category of crops—i.e., leafy vegetables, grains,
tubers, or fruits—to be sampled.
Sample masses of several kilograms may be needed to achieve specified ana-
lytical sensitivity. Sample size and preparation typically should provide 20–40 g
of ash from the edible portion. As a rule of thumb, ash weight is approximately
1% of moist weight. The sample should be weighed at the time of collection and
analysis. For many sample types, refrigeration and analysis without undue delay
are necessary to avoid decomposition.

5.7. Meat and Aquatic Species

Edible tissues from terrestrial animals, aquatic animals, and fowl are measured
for radionuclide content to estimate the radiation dose to humans from food con-
sumption. Organs and tissue that concentrate radionuclides, such as radioiodine
in thyroids or radiostrontium in bones and teeth, are analyzed as indicators of
such radionuclides in the environment. Domestic animals are sampled at the nor-
mal time of slaughter. Wildlife samples can be obtained from hunters or by special
collections. Because radionuclide concentrations in animal tissues vary widely, de-
pending on species, feed or inhalation intake, location, and individual metabolism,
multiple animals must be sampled to establish a normal range of radionuclide con-
centrations. Radionuclides that emit gamma rays can be measured in live animals
by a technique similar to whole-body counting in humans.
 5. Sample Collection and Preparation 89

Samples from aquatic species, i.e., fish and shellfish, may be the edible portion
to focus on human consumption, or the whole fish as indicator of radionuclide
levels in the system. Fish can be considered in several categories such as bottom
feeders (e.g., catfish), pan fish (blue gill), and predators (bass). Often, more than
one fish per species must be collected to accumulate the sample mass needed for
analysis. Fish are not good indicators of local radionuclide distribution because of
their mobility: i.e., fish caught at one location may have ingested the radionuclide
elsewhere, within the limits suggested by their normal movement or imposed by
dams.
Fish are sampled by electroshocking or with hoop nets, trout lines, or rod and
reel. Certain fish will float to the water surface after shocking (pan fish and preda-
tors), but bottom feeders such as catfish are more easily caught with hoop nets or
trout lines. Collection with rod and reel simulates sports angler catches.
Samples of meat and fish must be refrigerated for prompt analysis or frozen
for extended storage. Fish and other small animals may be dissected promptly
after collection or just before analysis. Samples should be weighed when fresh,
dried, and ashed as needed for the specified purpose in reporting radionuclide
concentrations.

5.8. Soil and Sediment

5.8.1. Soil
Soil usually is an easy matrix to collect in bulk quantity, but one that is difficult
to process. This means that the collection regimen must be highly selective in
allowing all pertinent, but no extraneous, soil to be sampled (see also Section
11.3.3). The site must be selected with regard to the radionuclide information
required of the program, which may include the consideration of all or some of
the following factors:
r amount of area deposition and accumulation
r deposition pattern
r transfer to vegetation or crops
r transfer to groundwater or surface runoff
r resuspension into air
r external radiation exposure

Soil can accumulate radionuclides over long time periods but may not be useful for
observing short-term trends. Measurements can provide retrospective information
on radionuclide levels. Site description, location, collection date and time, sample
area and depth, and sampling method (core, template, trench) should be recorded
along with the sample identification number and name of sample collector.
Surface soil can be collected with a scoop or cookie cutter. The scoop sampler is
convenient for collecting a grab sample and transferring it directly into a collection
container. A cookie cutter is a rigidly framed apparatus that can be sunk into the
90 Robert Rosson

soil to a selected depth to select the material within the frame for removal and
analysis (FRMAC 2002). Vegetation, debris, roots, and stones should be removed
at the time of collection to obtain a sample pertinent only to soil and assure that
the soil mass is sufficient for analysis. Sandy soil usually retains only a small
fraction of radionuclides to which it is exposed, while clay can be highly retentive.
In agricultural land, surface soil with elevated radionuclide levels may have been
turned over and had chemicals added that affect soil retention capability.
If the interest in radionuclide distribution extends beneath the soil surface, cor-
ing is the preferred sampling technique. Cores are taken to a known depth at the
sampling site. Depth profiles are separated at selected vertical intervals to provide
information on area deposition and downward movement of a radionuclide. Nu-
merous cores may have to be taken because radionuclide retention can vary with
changes in soil constituents in a relatively small area. Samples at greater depths
can be obtained from well-drilling cores.

5.8.2. Sediment
Radionuclides in sediment are indicators of facility releases to bodies of water
and runoff from surfaces in the drainage region. Samples represent the fraction of
radionuclides precipitated or sorbed on suspended matter that has settled to the
bottom. If such radionuclides are in sediment, elevated levels usually can be found
where suspended material tends to settle as the water flow rate decreases, e.g., at
widening channels, on the inner side of bends, and behind dams.
Variables that affect radionuclide concentrations in sediment are water move-
ment, water quality, the presence of aquatic vegetation, and sediment characteris-
tics such as clay and organic fractions. Sediment samples collected at time intervals
should not be expected to have a consistent pattern of radionuclide concentrations
because the location sampled previously is difficult to find precisely, sediment
may have moved downstream, and older sediment may be covered by more recent
deposits.
A striking effect of the change in water quality is the increased deposition found
for 137 Cs in estuaries where fresh water contacts saltwater (Corley et al. 1977).
Impounded water that stratifies, with anoxic (oxygen deficient) conditions near
the bottom during winter and mixing in spring, can affect radionuclide solubility
(Carlton 1992). In “blackwater” streams, humic acids can complex radionuclides to
keep them soluble. Changes in water level may periodically cover soil, vegetation,
and debris that take up radionuclides, notably in swamps.
The dredge sampler for sediment acts like a scoop in collecting sediment from
the stream bed surface. The dredge has drawbacks because the sampling location
on the riverbed is uncertain, it collects only the top 15 cm of sediment, and fine
material can escape along with the water on return to the surface. A core sample
collected by the slide hammer method can be used to collect a 0.7-m-long sediment
sample in water as deep as 5 m (Byrnes 2001).
Collected samples are frozen if preservation is necessary. Both moist weight at
the time of collection and dried weight before analysis should be measured.
 5. Sample Collection and Preparation 91

5.9. Bioassay Samples

Radiobioassay is the determination of the kind, quantity, and location of radionu-
clides in the body by direct measurement (in vivo) or by in vitro analysis of material
excreted from the body. Spectral analysis permits rapid analysis of radionuclides
that emit gamma rays. Computer systems with data analysis in terms of metabolic
models support routine use of bioassay procedures for assessing internally de-
posited radionuclides. The whole-body counter is an example of an in vivo proce-
dure.
Need for bioassay in humans depends on the exposure potential and the ra-
diotoxicity of a radionuclide and the degree of radionuclide control and radiation
monitoring at the location. Exposure assessment takes into account the physical
properties, the quantity, and the concentration of the radionuclide in the process,
and monitoring data on surface and airborne contamination. Monitoring levels that
exceed 10% of guidelines such as the derived air concentration for each radionu-
clide are a signal for instituting a bioassay program (NCRP 1987).
Radiotoxicity depends on energy deposition in tissue or organs by the radionu-
clide, the specific tissue exposed to the radionuclide, and the tissue radiation sen-
sitivity. Energy deposition by a radionuclide is a function of its emitted radiations
and half-life. Biokinetic studies have identified for most radionuclides of interest
the pattern of movement through the body and the effective turnover rate (the
sum of the biological and radioactive turnover rates). Biokinetic information also
identifies the appropriate type of sample to be collected among blood, urine, feces,
saliva, breath, hair, teeth, nasal swipes, and tissue obtained incidental to unrelated
operations, and collection frequency. The measured radionuclide concentrations
are combined with biokinetic information to calculate the committed dose equiv-
alent, the indicator of radiation impact on the subject (NCRP 1987b).
Urine is the most common bioassay sample type because it is nonintrusive and
readily available, and provides some insight into radionuclide intake, retention,
and excretion, but is not ideal for all radionuclides. A preservative can be added
to urine samples immediately after collection. Bioassay samples are refrigerated
for prompt analysis and kept frozen for extended storage.

5.10. Quality Control in Sample Collection

Quality control (QC) is as important when collecting samples in the field as it is in
the laboratory. A valid and defensible laboratory result is meaningless if the sample
has been collected improperly or the extent of its reliability is not known (see
Section 11.3.3). Many of the pitfalls associated with collection and preservation of
environmental samples are enumerated in MARSSIM (EPA 2000c) and HASL-300
(Chieco 1997). The steps taken for field QC vary with the matrix being sampled.
The following QC samples that pertain to the sample collection process should be
integrated with the QC samples appropriate to laboratory practices that are listed
in Section 11.2.9:
92 Robert Rosson

r Collocated samples to indicate the variability in the results of sampling at a given

location.
r Field trip blanks to identify radionuclide contamination in the collection material
(e.g., filter or sorbent) and container.
r Background samples that are from surveillance sites at a direction or distance
where no radionuclides from the monitored facility or incident are expected.
The program DQO specifies the types, locations, and collection frequency of these
QC samples.
Many types of environmental samples, notably vegetation and soil, are almost
inevitably inhomogeneous in ways that may prevent uniform distribution of a ra-
dionuclide throughout the matrix. An inhomogeneous radionuclide sample may
be due to the nature of the collected material, inadequate preparation, or “hot parti-
cles,” i.e., radionuclides that are not distributed uniformly in a solid, but associated
only with a subset (see Section 10.3.8). Some providers of reference materials state
the minimum sample aliquot to be analyzed to obtain a valid result. A larger sam-
ple than needed can be taken to eliminate extraordinary efforts in taking reliable
aliquots for radionuclide analysis. Analysis of multiple replicate samples indicates
the extent of reproducibility, and the blank samples assure that contamination has
been avoided before or during transport. Background collection prevents attribu-
tion of radionuclides found in these samples to the facility or incident that is being
monitored.
6
Applied Radioanalytical Chemistry∗
BERND KAHN1 , ROBERT ROSSON1 , and LIZ THOMPSON1

6.1. Introduction
A radioanalytical chemistry procedure is a series of steps that leads to the measure-
ment of radiation (or sometimes mass), with the goal of unambiguously identifying
the radionuclide of interest and determining its amount in a sample. The chem-
istry component of the analysis begins with the sample collection and handling
described in Chapter 5 and ends with the source preparation discussed in Chapter
7. The radiation emitted by the radionuclide is then measured (i.e., the sample is
“counted”), as discussed in Chapter 8. This chapter addresses the practical aspects
of the chemical and radiochemical separation processes surveyed in Chapters 3
and 4, which are central to separating interfering radionuclides and solids from a
radionuclide in preparation for counting.
Numerous separation methods of the types cited in Chapter 3 were developed and
applied in radioanalytical chemistry during the past century. The first 30 years were
devoted mostly to nuclear chemistry applications for identifying and characterizing
the naturally occurring radionuclides. In the following years, attention shifted to
the man-made ones; these activities continue, as exemplified by the work described
in Chapter 16. Currently, many methods are devoted to monitoring radionuclides
in the environment, facility effluent, process streams, and workers.
The radioanalytical chemist who is responsible for such monitoring selects or
designs procedures that meet the client’s specifications of sample type, list of
radionuclides, measurement reliability and sensitivity, and response time. The an-
alyst also considers the limits imposed by prescribed sample size, solution volume
appropriate for chemical separation, and radiation detection instruments at hand.
Potentially applicable procedures are selected for these criteria by literature review
and then evaluated in a methods development and testing process. A chemical sep-
aration procedure can be devised either by selecting the most applicable published
method and introducing any needed modifications or by combining pertinent sep-
aration steps.
One of the main efforts in measuring a radionuclide of interest is removing
interference due to other radionuclides by chemical separation, detector selection,

∗
This text owes the genesis of its content to Dr. Isabel Fisenne, Department of Homeland
Security.
1
Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of
Technology, Atlanta, GA 30332

93
94 Bernd Kahn, Robert Rosson, and Liz Thompson

or spectral analysis. Some sample characterization is always applied to limit the

number of radionuclides and their intensity to be expected in a sample. One ap-
proach is the following breakdown by origin, pathway, and half-lives (note that
the indicated half-lives and number of radionuclides are approximate):
r Samples that contain a few long-lived (t1/2 > 1 year) radionuclides. This mixture
is encountered in monitoring fallout from long-ago tests and in effluent, radioac-
tive waste, and decommissioning actions at nuclear facilities. Some subcate-
gories can simplify separation method selection: (1) Very long-lived (104 –1011
years) radionuclides may be considered when planning protection for thousands
of years. (2) A single radionuclide may have to be distinguished from the nat-
ural radiation background in contamination incidents, including possible ter-
rorist action. (3) Long-lived naturally occurring radionuclides may have to be
measured.
r Samples that contain approximately 10 radionuclides with intermediate to long
(1 day–1 year) half-lives. This mix is found when monitoring distant locations for
recent fallout from atmospheric nuclear tests and effluent from waste tanks and
stacks at operating nuclear power plants. A subcategory is samples that contain
natural radionuclides with half-lives in this range as progeny in terrestrial decay
chains and cosmic-ray-produced radionuclides.
r Samples that contain numerous radionuclides from certain portions of the pe-
riodic table with short (1 h–1day) and intermediate half-lives. This mixture is
found in monitoring process streams at operating nuclear facilities, fallout from
nuclear tests, or accidents within hours to days after occurrence. Subcategories
include samples from activation analysis, nuclear chemistry studies, and nature.
r Samples that contain a few very short-lived (<1 h) or extremely long-lived (>1011
years) radionuclides. These are encountered in nuclear chemistry studies con-
ducted in an academic setting, and include newly prepared and naturally occur-
ring radionuclides.
Circumstances will modify the contents of these categories with regard to ra-
dionuclides and half-lives; hence each sample set should be defined in its own
terms for planning the analysis. Occasional unusual radionuclide mixtures must
be anticipated during a project. The first three categories are discussed in Section
6.4. Examples of very short-lived radionuclides in the fourth category are given in
Chapter 16; very long-lived radionuclides are best measured by mass spectrometry
(see Chapter 17).
Once the radionuclide content of a sample is specified, available detectors must
be selected to measure the emitted radiation. The choice of detection technique
influences the need for and extent of the radiochemical separation, as discussed in
Chapter 7.
Chemical separation and radiation detection can work together in discriminating
against interfering radiation. Chemical separation may not be necessary if the
radiation from contaminant radionuclides is not detected by the selected detection
instrument or does not interfere with the measurement, e.g., by spectral analysis.
 6. Applied Radioanalytical Chemistry 95

In some instances, purification may not be necessary but can ease detection and
improve measurement.
Two or more radioisotopes of the same element that cannot be measured by
spectral analysis require integration of effective separation of impurities and radi-
ation detection of the selected distinguishing decay characteristics. To determine
the amounts of 89 Sr and 90 Sr in a sample, for example, interfering radionuclides
such as 226 Ra and 140 Ba must be removed; only then can the two strontium ra-
dioisotopes be distinguished in terms of the radioactive decay of 89 Sr, ingrowth of
the 90 Y daughter, and detector response to beta-particle energies, as discussed in
Section 6.4.1.
A testing program is needed to confirm the reliability of the selected proce-
dures, especially to achieve high chemical yield and a sufficient decontamination
factor for every interfering radionuclide. The analyst must estimate the allowable
concentrations of radionuclide impurities and matrix components; below this con-
centration, interference in measuring the radionuclide of interest is not of concern.
An ongoing quality assurance program provides feedback for maintaining or
improving each procedure during routine application. Changes are needed when
quality control results fall outside the limits of acceptability discussed in Section
10.5. Ambiguous results concerning the presence or absence of radionuclides of
interest require methods diagnostics, the topic of Chapter 12. Reported methods
that appear to be better should stimulate efforts to replace existing methods.

6.2. Initial Sample Treatment

The samples that are received for analysis are described in Chapter 5, and the
receiving actions in Section 13.3. Incoming samples can be divided into three
process streams: one for direct measurement, another for direct chemical analysis
of aqueous solutions, and the third for conversion to forms that are soluble in
aqueous solution. All solid samples are weighed; some will be dried and possibly
ashed, and weighed again after each process. Samples to be counted directly are
placed in tared counting container and weighed. Volumetric aliquots for gross
activity measurements and for chemical analysis are taken from liquid samples.
Samples scheduled for dissolution may be processed with suitable acids or bases
or by fusion. Special treatment is given to samples that must be removed from a
collector, require concentration, or are in gaseous form.

6.2.1. Aerosol Collectors

Air filters and gummed deposition collection films first are measured by gross
alpha- and beta-particle counting and gamma-ray spectral analysis. Usually, 1/2
of the sample then is dissolved to perform radiochemical analysis of the deposited
radionuclides. The filter can be dry ashed, and then totally dissolved with an
HNO3 –HF treatment.
96 Bernd Kahn, Robert Rosson, and Liz Thompson

Whether some insoluble residue remains depends on the type of mass loading.
For example, soil on the filter may constitute a partially insoluble matrix that retains
some radionuclides, including naturally occurring uranium and thorium. Insoluble
residues of ashing and subsequent acid dissolution require fusion, as discussed in
Section 6.2.7. The two fractions are combined after the melt is dissolved in dilute
HNO3 ,and the solution is evaporated almost to dryness and is then taken up in
dilute HNO3 .
Some readily soluble filter materials are available (see Section 5.3.1), but the
most favorable material may not meet the specifications for large filter area, rapid
airflow, and long collection period. Paper filters generally are sufficiently rugged
to meet these requirements.
Removing the accumulated particles by washing is unlikely to be quantitative
or even consistent. Acid leaching can leave highly insoluble forms of several ra-
dionuclides, notably the actinides, on the filter. As an exception, some fallout
radionuclides (such as 90 Sr) are completely dissolved by acid treatment enhanced
by refluxing. The undissolved material (principally silica) is removed by filtration.
This process can be applied only to separations that have been verified experimen-
tally to show no radionuclide loss.

6.2.2. Gaseous Radionuclide Collectors

Elements that have intermediate or long-lived radionuclides of interest in the
gaseous state are discussed in Section 5.3.2 with the collection techniques for
each. The gases are either brought into the laboratory in an airtight and possibly
pressurized container or collected on a sorbent such as charcoal, molecular sieve,
or silica gel. The samples are measured by gamma-ray spectral analysis either
directly in the collector or transferred to another container that has been calibrated
for measurement. If separation and purification are required, the gases are flushed
into a closed system for processing and subsequent measurement.

Tritium Samples
Airborne 3 H may be collected in any of the forms discussed in Section 5.3.2.
Samples may arrive with several forms in the same collector for separation at the
laboratory, or the forms may already be separated at collection. Most commonly,
tritiated water vapor is brought to the laboratory as a water sample. When collected
on silica gel or molecular sieve, tritiated water vapor is removed by heating the
collector material in a distillation flask above the boiling point of water. The
vapor is transferred to a separate flask and condensed. A measured amount of the
condensed water is analyzed for tritium in a liquid scintillation (LS) counter in
terms of becquerel per gram water. If the collector is silica gel, the result must
be corrected for dilution by a small amount of water originally in the silica gel
(Rosson et al. 2000). In addition, a small isotope effect, mentioned in Section 4.8,
results in distilling a greater fraction of H2 O than HTO unless the entire sample is
distilled.
 6. Applied Radioanalytical Chemistry 97

The other two forms of tritium gas usually are submitted on a palladium-coated
molecular sieve collector. Any HT is converted to HTO by palladium-catalyzed
oxidation in the collector; the water is removed by heating it above its boiling
point and condensing the vapor. The sorbent with the remaining organical-bound
tritium is mixed with a Hopcalite catalyst and heated to 550◦ C to oxidize organic
gases to water, which is distilled and condensed (Ostlund and Mason 1985).

C-14 Samples
The sample for 14 C analysis, usually collected as part of a research project, has
been processed at the time of collection. The 14 C is in the form of barium carbonate,
precipitated from a barium hydroxide solution. Two samples may be submitted if
the CO2 form has been separated from the CO and organic-bound 14 C forms. The
barium carbonate precipitate can be measured directly in an LS or proportional
counter or further purified.

Radioiodine Samples
Gaseous forms of iodine generally are collected undifferentiated on an activated
charcoal cartridge after passing through a filter on which particulate iodine is col-
lected. In special studies, the different gaseous forms may be retained on separate
collectors that are brought to the laboratory for analysis of 131 I. Direct gamma-ray
spectral analysis of filter and cartridge for the characteristic 0.364-MeV (85%)
gamma ray of 131 I is the preferred method of measurement.
A fresh fission-product mixture also contains 132 I (2.3 h), 133 I (20.8 h), and 135 I
(6.6 h) that can be measured by gamma-ray spectral analysis. Measurement of 129 I
is discussed in Section 6.4.1.

Noble Gas Samples

The noble gases can all be collected at liquid nitrogen temperatures on charcoal,
or can be sorbed selectively by controlling the collector temperature near the indi-
vidual condensation temperatures. They can be further purified in the laboratory
to separate the noble gases from each other and from other gases collected with
them.
The common radioisotopes of krypton and xenon are relatively short-lived,
fission-produced radionuclides; these are measured directly by gamma-ray spectral
analysis and require no sample treatment. One exception is 85 Kr (10.7 years), a
radionuclide that emits mostly beta particles (0.687 MeV maximum energy), with
a gamma ray of 0.514 MeV (0.43%) that is too low in abundance for measurement
except at high 85 Kr levels. A simple technique for 85 Kr analysis is to concentrate
it by collection at a low temperature, dissolving it in an LS cocktail, and then
measuring it in an LS counter (Shuping et al. 1970). The method is used to measure
the 85 Kr background in ambient air. A 7-week decay interval generally reduces
all the shorter lived krypton and xenon fission products to concentrations that no
longer interfere with this measurement. A gas-separation system was constructed
98 Bernd Kahn, Robert Rosson, and Liz Thompson

to separate krypton from xenon by sequential condensation and heating on sorption

columns (Momyer 1960).
Airborne 222 Rn collected in a scintillator-coated Lucas cell is placed on a pho-
tomultiplier tube in the laboratory and counted (see discussion for 226 Ra analysis
by radon measurement in Section 6.4.1). Radon collected on a charcoal cartridge
or other sorbent is measured by gamma-ray spectral analysis. In both cases, a 4-h
period of ingrowth is needed to accumulate the radon progeny that contribute to
the count rate.

6.2.3. Water
The sample may arrive unfiltered or separated as filtered water and the filter that
contains the solids. The water sample is preserved with dilute acid or a preservative
suitable for a radionuclide such as 131 I that may be lost from an acid solution.
Water without suspended solids is ready for evaporation to measure the gross
alpha- and beta-particle activity, measure gamma rays by spectral analysis, and
perform radiochemical analysis. The solids usually are counted similarly and then
processed for dissolution as described in Section 6.2.1 for subsequent radionuclide
analysis.
When a liquid volume of several liters is needed to achieve a specified sensitiv-
ity for radionuclide detection, the volume usually is reduced for ease of analysis.
Evaporation is a simple approach. A faster process—and one that is necessary for
samples of seawater or other high-salt-content solution—is precipitation of the
radionuclides of interest from the large sample volume with a carrier in bulky in-
soluble forms such as phosphate, hydroxide, or carbonate. The precipitate bearing
the radionuclides of interest is separated by decanting most of the solution and
filtering the rest, and is then dissolved for further radionuclide purification.
Certain radionuclides may be collected from even larger volumes (i.e., hundreds
of liters) by passing a stream of water through a large filter system to retain
insoluble forms and then through large ion-exchange columns to retain cations and
anions. The filter and columns can be analyzed with a gamma-ray spectrometer.
Radionuclides that do not emit gamma rays can be eluted from the ion-exchange
columns with a relatively small volume of a high-salt solution, a complexing agent,
or a redox reagent. The filter is processed as discussed in Section 6.2.1.

6.2.4. Milk
Many persons consume milk soon after collection, and the pathway for a few
radionuclides from origin to milk includes accumulation steps due to large-area
grazing by cows and their lactation process. Before the advent of gamma-ray
spectral analysis, the radionuclides 89 Sr, 90 Sr, 131 I, 137 Cs, and 140 Ba were measured
after chemical separation. Today, only the 89/90 Sr pair requires chemical separation.
The brute-force approach to sample preparation was evaporating and ashing a
liter of milk. The process requires care to avoid splattering and loss of ash from
 6. Applied Radioanalytical Chemistry 99

the large mass of organic material. A major improvement was precipitation of the
organic phase of the milk with trichloroacetic acid, followed by radiochemical
analysis of radiostrontium in the aqueous phase (Murthy et al. 1960). Currently,
whole milk is passed through ion-exchange resin to sorb the radionuclide of inter-
est, notably radiostrontium, on cation-exchange resin and radioyttrium on anion-
exchange resin, at controlled pH. The radionuclide is then eluted from the resin
for measurement (Porter and Kahn 1964).

6.2.5. Vegetation and Crops

Vegetation and crop samples are treated for subsequent purification by removing
the bulk of the matrix—typically 99%—in oven drying and ashing. After dissolving
the ash with mineral acid, a minor residue of insoluble silica can be volatilized by
boiling with HNO3 –HF solution.
One common distinction between processing food and vegetation samples is
that the edible portion in food is analyzed separately to provide information on ra-
dionuclide consumption. Additional treatment of food before analysis may include
washing and other preparation steps.

6.2.6. Meat and Fish

Samples of animal products may be processed whole to determine their radionu-
clide content, dissected to distinguish radionuclide concentrations in various or-
gans, or separated to analyze only edible portions. Questions may arise about what
constitutes the edible sample portion, which differs among persons and population
groups. These distinctions affect radiological health protection monitoring because
metabolic parameters result in accumulation of radionuclides in specific organs
or tissue, such as radiostrontium in bone and radioiodine in the thyroid (Cember
1996).
Unless silica is present (e.g., due to soil or sediment in intestines), the ashed
material usually is soluble in dilute HNO3 ; minor residual silica is removed by
HNO3 –HF treatment (see Section 6.2.5). Soft tissue may also be dissolved in basic
solution.

6.2.7. Soil and Sediment

Soil and sediment often are considered the most difficult matrix to dissolve. They
may be leached under restricted conditions, dissolved completely in small amounts,
or dissolved after fusion. The method is selected on the basis of which radionuclide
is to be analyzed and in what form, sample origin, and the difficulty and cost of
analysis.
Any leaching process must be tested to determine the leached fraction of a
radionuclide as a function of type and strength of acid, type of soil or sediment,
100 Bernd Kahn, Robert Rosson, and Liz Thompson

leachate/sample ratio, leach period, and temperature. The leached fraction varies
for different soils and radionuclides.
Total dissolution is needed for less soluble radionuclide forms. This goal may
be achieved by repeated treatment with solutions of the same acid or sequential
treatment with different acids. Total dissolution of gram amounts is achieved rela-
tively rapidly in a microwave oven with programmed temperatures and pressures
(Garcia and Kahn 2001). The factors discussed above also control total dissolution
and must be determined experimentally. Sample mass is limited to a few grams,
compared to order-of-magnitude larger amounts that can be treated by leaching.
Refractory species (such as zirconium oxide) are incompletely atomized at the
temperatures of a flame or a muffle furnace. If the radionuclide of interest may be
in a refractory form, the material must be mixed with fusion reagents and melted
at a high temperature. After cooling, the solidified melt is dissolved in a HNO3
solution. Fusion mixtures that have been tested for many types of soil are listed in
Table 4.2. Sample masses are limited by practical consideration of the final sample
size, taking into consideration the large amounts of fusion reagents added to the
sample.

6.2.8. Bioassay Samples

Urine samples are measured directly by spectral analysis for radionuclides that
emit gamma rays. To measure radionuclides that emit only alpha or beta particles,
a general approach for urine pretreatment is boiling to dryness and then ashing.
The ash is dissolved in mineral acid to prepare the aqueous solution for analysis
(Chieco 1997). Other bioassay samples are pretreated similarly.
More convenient procedures for urine analysis include separation of the radionu-
clide of interest from urine by precipitation or by sorption on a cation-exchange
resin column. Ferric hydroxide is a precipitate that carries many radionuclides from
a neutral or basic solution. The precipitate is washed with water and then dissolved
in dilute mineral acid for further purification of the radionuclide of interest. For ion-
exchange sorption, a complexing agent such as ethylene dinitrilotetraacetic acid
(EDTA) is added to the urine at a selected pH value to complex interfering ions,
such as calcium for 90 Sr analysis. After passage of the urine through the column
and washing the column with the complexing agent and water, the radionuclide of
interest is eluted from the column with a suitable reagent (Sunderman and Townley
1960).

6.2.9. Smears
Paper or cloth smears are used, often in response to regulations, to wipe surfaces of
specified area (e.g., 100 cm2 ) to check for removable radionuclides. The smears are
counted directly by gamma-ray spectral analysis. For gross alpha- or beta-particle
measurements, thin smears are counted in a proportional counter or immersed in a
cocktail for LS counting. Further analysis for a radionuclide of interest that emits
 6. Applied Radioanalytical Chemistry 101

only alpha or beta particles requires ashing the smear and subsequent dissolution
with dilute mineral acid.

6.3. Carriers and Tracers

Carriers are added to radionuclide solutions, as discussed in Section 4.7, to perform
precipitation separations, avoid radiocolloidal behavior, and to eliminate the need
for quantitative recovery by allowing the analyst to calculate the fractional chemical
recovery (“yield”). A radioactive tracer (see Section 4.7.3) can be added to the
sample in addition to or instead of the carrier to measure yield. Tracer addition
may make yield determination easier than carrier addition, or may be necessary if
no carrier is available.

6.3.1. Carriers
Cationic carriers usually are obtained as chloride or nitrate forms, and anions, in
their sodium form. The reagents must be sufficiently pure so that any impurity
does not significantly affect its mass determination or radionuclide measurement.
Reagent blanks must be counted after carrier preparation to check purity, and any
significant source of radiation must be removed from the carrier.
The carrier solution should be maintained in a stable form, which may require
addition of a reagent such as a dilute acid to a carrier that may otherwise hydrolyze
over the period of use. The carrier should be filtered after preparation, checked
periodically, and discarded if solids have formed or other changes are apparent.
Care must be taken not to introduce contamination into a carrier; typically, a master
solution is stored separately from frequently used fractions.
The carrier concentration in the solution should be determined in replicate with
sufficient precision so that the uncertainty of yield measurement does not exceed
1%. The precipitate for carrier mass determination should in all ways be identical
to that obtained at the end of the procedure to avoid chemical form differences
such as water content.
The carrier is pipetted into the sample solution at the beginning of the analytical
process and mixed thoroughly with the solution. The chemical state of the carrier
must be identical to that of the radionuclide. If any difference is possible, steps
must be inserted into the procedure at this point to achieve identical form for carrier
and tracer. A common process is to oxidize and reduce the carrier and radionuclide
through all possible oxidation states. For example, if the radionuclide 131 I is not
known to be in the form of iodide and the carrier is added as iodide, the mixture is
oxidized through molecular iodine, iodate, and periodiate, and then reduced back
to iodide.
Nonisotopic carriers can be used when no isotopic carrier exists. Rhenium is a
carrier for technetium, lanthanum for promethium and plutonium, and barium for
radium. Such chemical homologs do not have identical solubility products, hence
their yields may differ and must be compared by testing. Group separations can be
102 Bernd Kahn, Robert Rosson, and Liz Thompson

performed (see Table 3.2), such as iron hydroxide precipitation to carry tetravalent
uranium and rare earth radionuclides.
So-called “holdback” carriers are added to improve separation of one radionu-
clide from another at very low concentration. A holdback carrier differs slightly
from a regular carrier in its function: a regular carrier functions to “carry” the ra-
dionuclide of interest out of solution, while a holdback carrier’s function is to keep
an interfering radionuclide in solution while the radionuclide of interest is pre-
cipitated with its regular carrier. Holdback carrier addition prevents radiocolloidal
behavior exhibited in sorption of a contaminant radionuclide on the precipitate of
the radionuclide of interest. The holdback carrier effectively maintains a consistent
decontamination factor.

6.3.2. Tracers
Tracers are used as radiometric means of determining the yield of the radionuclide
of interest. Tracers are selected for convenience in availability and counting the
emitted radiation. The two radiations may be distinguished from each other by
spectral analysis, e.g., 242 Pu tracer for 239 Pu by counting alpha particles, or by use
of different detectors, as in counting the gamma rays of the 85 Sr tracer for 90 Sr that
is determined by counting beta particles emitted by its 90 Y daughter.
Although isotopic tracers are preferred, application of nonisotopic tracers pro-
vides more choices for selecting a radionuclide with suitable radiation, e.g., 133 Ba
tracer by gamma-ray spectral analysis for 226 Ra by counting alpha particles. As
stated in the preceding section, tests must be performed to determine whether
yields for the two nonisotopic radionuclides are identical or have a constant ratio.
The radioactive tracer must not be contaminated with other radionuclides that
interfere with the yield measurement. Some actinide-series tracers have radioactive
progeny with slow ingrowth that should be identified from their decay scheme chain
and require periodic purification to remove these progeny.
The amount of tracer to be added needs to be balanced; too much may result
in some radiation cross-talk between the radionuclide of interest and the tracer,
while too little reduces the precision of yield determination. Typically, the tracer
concentration is selected to be a few times the expected concentration of the
radionuclide of interest.

6.4. Radionuclide Separation and Purification

Radiochemical analyses are presented in this section according to the categories
in Section 6.1 for selected radionuclides that have been measured throughout the
world for many years. These methods are discussed briefly to suggest approaches
for selecting and improving these and other methods.
Chemical separations were important contributors to early efforts for radionu-
clide identification and measurement when only gas ionization and solid scintilla-
tion detectors were available. Chemical separation identified the radionuclide by
 6. Applied Radioanalytical Chemistry 103

element. Subsequent measurements with absorber foils (see Section 2.4.2) iden-
tified the type of emitted radiation and estimated its energy. The half-life was
determined by repeated measurements under the same conditions. This approach
was difficult for a radionuclide that emits radiation of several types and energies, for
more than one radioisotope per element, or for incompletely purified radionuclides.
These measurement techniques are reflected in the early compilations of radio-
analytical chemistry methods (Coryell and Sugarman 1951, NAS-NRC 1960). As
spectral analysis achieved better resolution and higher counting efficiency, chem-
ical separations were avoided when possible in favor of spectral analysis.
Currently, only a few routinely measured radionuclides such as 3 H, 14 C, 55 Fe,
90
Sr, 99 Tc, 129 I, 147 Pm, and many of the elements heavier than bismuth remain
dependent on chemical separations. Chemical separation still is needed for a few
reasons; a radionuclide emits no gamma rays, emits too low a fraction of gamma
rays to meet required detection levels, requires removal of solids for counting (e.g.,
alpha particles), or must be detected with greater sensitivity.
In the future, mass spectrometry (see Chapter 17) may supersede radiochemical
analysis for long-lived radionuclides and require a different set of chemical sepa-
rations. This trend is opposed to a certain extent by chemical separation processes
introduced to achieve ever lower minimum detectable activity requirements and
by the continued interest in identifying newly created radioelements.

6.4.1. Long-Lived Radionuclides

The long-lived radionuclides associated with nuclear fission and concomitant neu-
tron activation are listed in Table 6.1, together with the long-lived terrestrial ra-
dionuclides and the major cosmic-ray-induced radionuclides in nature. Of the
fission and activation products in Table 6.1, all but the minor ones are commonly
measured; of the naturally occurring radionuclides, 40 K, uranium, thorium, and
radium are frequently encountered in samples, together with radioactive progeny
of the latter three.

H-3 Analysis
Tritium is produced in nuclear reactors and devices by ternary fission and neutron
activation of deuterium, boron, and lithium, and is produced naturally by cosmic
rays. It is analyzed in air, water, and biota samples and for bioassay. Water vapor in
air is collected by condensation on a cold surface, by bubbling through water, or on
a sorbent material such as silica gel from which it is then flushed above the boiling
point of water. Water from biological material is collected as vapor by heating
samples just above the boiling point of water and then condensing the vapor. The
NCRP has published a survey of tritium measurements, which provides detailed
information (NCRP 1976a).
The sample—typically about 20 ml water—is distilled for purification. Reagents
and holdback carriers may be added to the distilling flask to prevent volatile
forms of contaminant radionuclides from being distilled. Initial distillation may be
104 Bernd Kahn, Robert Rosson, and Liz Thompson

TABLE 6.1. Commonly analyzed long-lived to extremely long-lived radionuclides

Category Radionuclide Half-life Radionuclide Half-life
Major fission products 3H 12.3 years 125 Sb 2.7 years
85 Kr 10.7 years 129 I 16,000,000 years
90 Sr 28.8 years 137 Cs 30.2 years
99 Tc 214,000 years 147 Pm 2.62 years
106 Ru 367 days
Minor fission products 79 Se 65,000 years 135 Cs 2,300,000 years
93 Zr 1,500,000 years 151 Sm 90 years
107 Pd 6,500,000 years 155 Eu 4.9 years
Neutron activation in 134 Cs 2.06 years 239 Pu 24,100 years
Fission 233 U 159,000 years 240 Pu 65,700 years
236 U 23,400,000 years 241 Pu 14.4 years
237 Np 214,000 years 241 Am 433 years
Common neutron 55 Fe 2.7 years 109 Cd 453 years
Activation products 60 Co 5.27 years 152 Eu 13 years
59 Ni 75,000 years 154 Eu 8.5 years
63 Ni 100 years
Naturally occurring 3H 12.3 years 227 Ac 21.8 years
radionuclides 10 Be 2,700,000 years 228 Ra 5.76 years
14 C 5730 years 228 Th 1.91 years
26 Al 740,000 years 230 Th 7300 years
32 Si 280 years 231 Pa 32,800 years
39 Ar 269 years 232 Th 1.41 × 1010 years
40 K 1.28 × 109 years 234 U 245,000 years
87 Rb 4.8 × 1010 years 235 U 7.04 × 108 years
210 Pb 22.3 years 238 U 4.47 × 109 years
226 Ra 1,600 years

performed to separate the volatile forms, and the remaining water is then distilled
for tritium analysis. For extracting water from solid samples or tritium sorbents,
azeotropic distillation in a reflux system conveniently maintains a constant boiling
point and liquid cover. The organic solvent is refluxed while the heavier distilled
phase is 99.9% water, as described in Section 3.6.1.
The distilled water is counted in an LS system at a selected water-to-cocktail ratio
(usually 1:1). Samples may be counted without purification if other radionuclides
are known to be absent or can be differentiated clearly from tritium by pulse-
height discrimination in the detection system, and if chemicals that cause excessive
quenching or fluorescence are known to be absent. Section 15.4.3 illustrates an
extension of this measurement technique, where 3 H is measured in flowing water
with a scintillation counter.
Collection and analysis complexity introduced by forms other than tritiated
water is discussed in Section 6.2.2. Tritium is released in these forms from nuclear
facilities or converted to them in the environment. These forms generally occur
in lesser magnitude than tritiated water; they should be differentiated because of
their different pathways and radiation impacts after they enter the body. Biological
material can be dried to collect tritium as water vapor and then ashed to collect
organic tritium as HTO vapor.
 6. Applied Radioanalytical Chemistry 105

Tritium has also been counted as a gas or vapor mixed with the counting gas
in a flow-through ionization or proportional counter. Flow-through detectors are
suitable as effluent monitors for tritium when tritium concentrations in air are high
relative to concentrations of other radionuclides. Discrimination between tritium
and radionuclides with more energetic beta-particle groups is achieved by pulse-
height control for the proportional counter.

Fe-55 Analysis
55
Fe is formed by the (n,γ ) reaction with 54 Fe. At the same time, 59 Fe is formed by
the (n,γ ) reaction with 58 Fe. Both radioisotopes are produced in iron and steel cas-
ings, vessels, or supports for nuclear weapons and reactors. 55 Fe (t1/2 = 2.73 years)
decays by electron capture with a K α X-ray energy of 5.89 keV (24.5%) and a K
Auger electron energy of 5.2 keV (61%). In contrast, 59 Fe (44.5 d) emits beta par-
ticles with maximum energies of 0.466 (53%) and 0.274 MeV (45%) and gamma
rays of 1.29 (43%) and 1.10 MeV (57%). The following procedure (Chieco 1997)
determines both radionuclides.
Iron chloride or nitrate carrier is added and precipitated to carry the radionuclides
as hydroxide, dissolved in strong HCl and collected as the chloro-complex on an
anion-exchange column. After washing to remove contaminants, iron is eluted
from the column with dilute HCl. Cobalt, manganese, and zinc holdback carriers
are added to the solution and iron is precipitated as the cupferate at 10◦ C. The
cupferate complex is destroyed by wet ashing and iron oxide is converted to the
chloride by boiling in HCl. The iron ion is complexed in an ammonium phosphate–
ammonium carbonate electrolyte and electroplated on a tared copper disc. The disk
is weighed to determine the iron yield and then is sprayed with a thin acrylic coating
to prevent oxidation of the iron.
59
Fe beta particles are counted with a proportional detector or its gamma rays
are analyzed with a Ge detector and spectrometer. The sample is then measured
for 55 Fe content with a thin Ge detector and spectrometer or xenon-filled X-ray
proportional detector with a thin (e.g., 140 mg cm−2 ) beryllium absorber. The 55 Fe
count rate is adjusted for background, the 59 Fe contribution, self-absorption in the
plated sample, and the chemical yield, and converted to the disintegration rate. The
activity of both radioisotopes is corrected for radioactive decay from the sampling
date.
If no 59 Fe is in the sample, as confirmed by sample analyses, it can be added
as yield tracer for 55 Fe analysis. After ion-exchange purification, the elutriant is
measured by LS counting 55 Fe and 59 Fe in different energy regions. Cross-talk
from the 59 Fe beta particles in the 55 Fe Auger electron region must be corrected,
as well as color quenching from chemicals.

Sr-90 Analysis
Analysis of 90 Sr is a classical example of a precipitation separation. Precipitation in
concentrated (around 60%) to fuming (>86%) nitric acid separates strontium and
106 Bernd Kahn, Robert Rosson, and Liz Thompson

barium from most other elements (Sunderman and Townley 1960), and is partic-
ularly useful because so many other fission-produced radionuclides are insoluble
in basic solution but soluble in acid.
After strontium carrier is added to a small volume (<10 ml) of 90 Sr solution,
sufficient fuming nitric acid is added to attain a nitric acid concentration of 14–
16 N. The solution with strontium nitrate precipitate is cooled in an ice bath
and then centrifuged. The supernatant solution is thoroughly decanted and the
strontium nitrate precipitate is dissolved in water. Barium and yttrium carriers
are added. Precipitation of barium chromate at pH 5.5 removes 140 Ba and natural
radium from the supernatant strontium solution (for counting, if needed, of these
two separated radioelements). Precipitation of yttrium hydroxide in basic solution
then removes the 90 Y daughter that has grown into the 90 Sr parent. Ammonium
oxalate is immediately added to the supernatant solution to precipitate strontium
oxalate. The precipitate is washed and dried in the filter holder with alcohol and
ether, promptly weighed for yield determination, and counted with a beta-particle
detector (Chieco 1997) such as a proportional detector.
The measured count rate must be corrected for contribution by 90 Y daughter
ingrowth that begins immediately after the hydroxide scavenging precipitation. If
time is not of the essence, the measurement can be delayed for 20 days until 90 Y
ingrowth is essentially complete, to reach equal disintegration rates for 90 Sr and
90
Y. Individual counting efficiencies must be applied for 90 Sr and 90 Y because the
two are not identical in the proportional counter.
An alternative is to store the supernatant solution after the barium chromate
scavenging precipitation for 20 days until 90 Y approaches equilibrium with 90 Sr,
and then precipitate yttrium hydroxide with the 90 Y. The hydroxide is dissolved in
dilute acid, and oxalic acid is added to precipitate yttrium oxalate. The precipitate
is washed and dried, promptly weighed for yield determination, and counted with a
proportional detector (Sunderman and Townley 1960). To the supernatant solution
from the yttrium hydroxide precipitation, ammonium oxalate is added to precipitate
strontium oxalate for weighing to determine the strontium yield. The 90 Sr activity
is calculated from the 90 Y count rate adjusted for 90 Y ingrowth into the 90 Sr parent,
90
Y decay since its separation from 90 Sr, yttrium and strontium yields, and 90 Y
counting efficiency.
One variant is that the radionuclides are dissolved and then measured with an LS
counter. Another variant is that 85 Sr tracer is used for strontium yield measurement
when 90 Y is measured to determine the 90 Sr activity.
Samples with high calcium content can be problematic because some calcium
nitrate may also precipitate in nitric acid at the indicated strength. The additional
calcium will cause overestimation in gravimetric strontium yield. The calcium
nitrate precipitate (after thorough draining to remove the explosion potential of
the nitric acid–alcohol mixture) can be dissolved in a few milliliters of absolute
alcohol with little loss of strontium nitrate precipitate. A possibly safer removal of
calcium contamination is by reducing the nitric acid strength to 60% or dissolving
the strontium and calcium nitrate precipitates in a small amount of water and then
precipitating strontium nitrate with concentrated nitric acid for a second or even
 6. Applied Radioanalytical Chemistry 107

third time. One interference is a silica precipitate in the strong nitric acid solution
from samples such as dissolved soil; this precipitate can be removed by filtration
after dissolving the strontium nitrate in water.
As indicated in Section 6.4.2 below, the presence of 89 Sr requires calculation
by simultaneous equations to determine the disintegration rates of 89 Sr and 90 Sr in
the same sample. Both radionuclides can be expected in a mixture of shorter-lived
fission products within a year after formation.
Changes in 90 Sr purification were developed mainly to eliminate use of large
volumes of fuming or concentrated nitric acid that corrode hoods, ducts, and lab-
oratory equipment. Early modifications applied selective ion-exchange sorption
to separate strontium from calcium. For example, calcium in the sample is com-
plexed by EDTA at pH 4.6 so that strontium, barium, and radium can be retained
on a strong-base cation-exchange resin while calcium passes through the column
(Porter et al. 1967). The strontium is then selectively eluted with the same so-
lution at pH 5.1 and precipitated as the carbonate for yield determination and
counting.
A more recent technique uses the Eichrom column discussed in Section 3.5. After
initial precipitation and dissolution to reduce the solution volume, the solution is
adjusted to the specified acidity and passed through the column, which retains
strontium. Strontium is eluted and then precipitated as the carbonate.

Tc-99 Analysis
Technetium does not have a stable isotope. Of its long-lived isotopes, 95 Tc, 97 Tc,
98
Tc, and 99 Tc, only the latter is a fission product. Short-lived 99m Tc (6.0 h) pre-
cedes 99 Tc. It is the daughter of 99 Mo (66 h), a radionuclide widely used in nu-
clear medicine. The half-lives of all other technetium fission products are less
than 1 h.
The most stable oxidation states of technetium are +4 and +7. Hydrazine and
hydroxylamine reduce technetium to +4. Atmospheric oxygen, hydrogen perox-
ide, and strong nitric acid oxidize technetium to +7. Some reactions can be slow.
Other, less stable, oxidation states exist. In the lower oxidation state, technetium
as TcO2 is carried on ferric hydroxide precipitate. In the upper oxidation state,
technetium is the anion TcO− 4 . It is sorbed on anion-exchange resins, extracted
into oxygen-containing solvents such as hexone (methyl isobutyl ketone) and trib-
utylphosphate, and coprecipitated with rhenium as the phenylarsonium pertech-
netate, (C6 H5 )4 AsTcO4 . It can be distilled as Tc2 O7 from sulfuric and perchloric
acids (Anders 1960).
The above-cited reactions are used to separate 99 Tc for measurement. Rhenium is
a common nonisotopic carrier for precipitation or yield measurement. The shorter-
lived 95 Tc (60 days) has been used as isotopic tracer for yield measurement (Anders
1960). Technetium can be separated from rhenium by coprecipitating the +4 state
with ferric hydroxide while rhenium remains in solution. Technetium is separated
from the fission-produced radionuclides 103 Ru and 106 Ru by distillation with Re2 O7
from sulfuric acid while ruthenium remains behind because it is volatile only in its
108 Bernd Kahn, Robert Rosson, and Liz Thompson

highest oxidation state (RuO4 ). An automated separation procedure is described

in Section 15.3.4.
The beta-particle group emitted by 99 Tc with a maximum energy of 0.294 MeV
can be measured by proportional or LS counter. The latter is more suitable for the
low energy, but the former may provide more sensitive detection. 99 Tc does not
emit gamma rays.

I-129 Analysis
If a thermal neutron source is available, 129 I can be determined at low concentra-
tion by measuring activation-produced 130 I (12.4 h); at the same time, the specific
activity can be measured by activation of stable 127 I to 128 I (25 min). The cross-
sections for neutron activation are 6.1 barn for 127 I and 9 barn for 129 I. Although
129
I is a fission product, it is also produced in the environment by cosmic ray neu-
trons. Attribution to an anthropomorphic source must be checked by background
measurements of comparable matrices in terms of the specific activity. Data from
such measurements are available (NCRP 1983).
At a magnitude of 0.1 Bq per sample, 129 I can be measured directly with an LS
counter or a Ge gamma-ray spectrometer in its low-energy region. The radionuclide
emits 0.153 MeV (100%) maximum energy beta particles, 0.040 MeV (7.5%)
gamma rays, K X-rays in the 0.029–0.034 MeV (70%) region, and conversion and
Auger electrons in the 0.025–0.039 MeV (22%) region.
The radionuclide with added carrier can be separated from most other long-
lived radionuclides by precipitating silver or palladium iodide from dilute nitric
acid solution. As indicated in Section 6.3.1, the oxidation state of radioiodine,
of which there are several, must be well defined before a separation step can be
trusted. Other purification techniques are solvent extraction of iodine oxidized to
I2 into carbon tetrachloride followed by back extraction of the reduced iodide form
into water, or sorption of I− on anion-exchange resin followed by elution with a
strong chloride solution (Kleinberg and Cowan 1960). These processes also lend
themselves to concentrating the radionuclide from larger solution volumes to attain
a lower detection limit.

Cs-137 Analysis
Initially, 137 Cs was analyzed by adding cesium carrier and precipitating a com-
pound that is specific for cesium in dilute acid solution in the presence of sodium,
such as the cobaltinitrite, phosphomolybdate, silicotungstate, or tetraphenylbo-
rate. Known contaminants, such as 32 P or 99 Mo in phosphomolybdate, are then
removed after dissolving this precipitate by a second precipitation with a different
reagent or by a scavenging precipitation for the contaminant. The purified sample
is weighed for yield and then counted with a proportional or G-M counter (Finston
and Kinsley 1961).
137
Cs is collected from larger volumes of water on large cation-exchange resin
columns. Anion-exchange resin columns loaded with one of the above-cited pre-
cipitation reagents make these columns specific for cesium. A column loaded with
 6. Applied Radioanalytical Chemistry 109

ammonium hexacyanocobalt ferrate effectively retains radiocesium (Boni 1966)

from liquids with high salt content.
Once gamma-ray spectral analysis became available, the 0.660-MeV gamma ray
of the 2.55-min 137m Ba daughter was counted for identification, distinction from
other radionuclides, and measurement. Little interference from other gamma rays
is encountered with a Ge detector. 134 Cs may accompany 137 Cs in reactor-produced
mixtures. The 2.55-years 134 Cs emits several gamma rays, notably at 0.605 (98%)
and 0.795 MeV (88%).

Pm-147 Analysis
147
Pm is one of three relatively long-lived rare earth fission products; the other
two are 144 Ce (284 days), which is discussed in the following section, and 155 Eu,
with a low fission yield. 147 Pm emits a 0.284-MeV maximum beta-particle energy
group and no gamma rays. In contrast, 144 Ce emits gamma rays of 0.134 MeV
(10.8%) and 155 Eu emits gamma rays of 0.087 MeV (31%) and 0.105 MeV (20%).
Contamination by 152 Eu and 154 Eu is possible if a sample that contains trace
amounts of europium, such as soil, was irradiated with neutrons. Both of these
radionuclides emit gamma rays.
The typical purification method for rare earths is coprecipitation with ferric
hydroxide, dissolution in dilute acid, precipitation as fluoride in strong mineral
acid solution, dissolution in strong nitric acid with boric acid to complex fluo-
ride, and precipitation for counting as the oxalate in dilute acid solution (Steven-
son and Nervik 1961). Because 147 Pm has no stable isotope, another rare earth
(such as lanthanum) is added as carrier. The 147 Pm precipitate can be counted
with a proportional counter, or can be dissolved and measured with an LS counter
because of the low beta-particle energy. If small amounts of the other rare earth
radionuclides are detected by gamma-ray spectrometric analysis, the beta-particle
count rate of 147 Pm can be calculated by difference.
147
Pm can be separated from cerium and europium because it remains in the +3
oxidation state when cerium is oxidized to +4 or europium is reduced to +2. 144 Ce
can be scavenged in dilute acid solution by adding cerium carrier, oxidizing to the
+4 form with sodium bromate, and then precipitating it with HIO3 as Ce(IO3 )4 .
Eu-155 can be separated by adding holdback europium carrier, reducing europium
to the +2 state with zinc powder, and then carrying 147 Pm on ferric hydroxide
precipitate (Stevenson and Nervik 1961).
Rare earths can be separated from each other on ion-exchange columns, as
illustrated in Fig. 3.2. Once separation conditions have been defined, the method
can be simplified for purifying one radionuclide such as 147 Pm from interfering
rare earth radionuclides.

Ra-226 and Ra-228 Analysis

Radium analysis in water supplies is required under the Safe Drinking Water Act.
The US EPA specifies that 226 Ra and 228 Ra must be detected at a concentration of 1
110 Bernd Kahn, Robert Rosson, and Liz Thompson

pCi/l (0.037 Bq/l). Radium analysis may also be needed in soils, ores, manufactured
materials such as concrete, and various contaminated media.
The conventional analysis of 226 Ra in water is coprecipitation with barium sulfate
carrier from a 1-l volume. After decanting the supernatant solution, barium sulfate
is dissolved in EDTA solution and stored in a sealed container for 1–4 weeks to
allow for ingrowth of its 222 Rn (3.82 days) daughter. The 222 Rn gas is flushed with
air (sufficiently aged in a tank to assure the absence of radon) into an evacuated 100-
to 200-cc ZnS(Ag)-coated “Lucas” cell. A photomultiplier tube views scintillations
at the ZnS(Ag) surface through a light pipe at one end of the cell (Lucas 1957). The
measured alpha particles are emitted by 222 Rn and its progeny 218 Po and 214 Po. The
ingrowth of 214 Po, controlled by its precursors 214 Pb and 214 Bi, reaches equilibrium
with that of 222 Rn within 4 h.
The activity of 226 Ra is calculated from the count rate of the three alpha particles
of its progeny. Adjustments are made for the ingrowth and decay of 222 Rn, the
detector counting efficiency, and any losses during the precipitation of barium
sulfate and the transfer of the gas into the cell.
The analysis of 228 Ra conventionally followed that of 226 Ra in precipitating with
barium sulfate and dissolving with EDTA solution. After ingrowth of 228 Ac (6.13 h)
for 2 days, the 228 Ac is precipitated with yttrium carrier as yttrium oxalate. The
precipitate is washed and dried with alcohol and ether, weighed, and then counted
with a proportional detector. The barium that remained in the EDTA solution again
is precipitated as sulfate to determine its yield. The 228 Ra activity is calculated
from the 228 Ac count rate and counting efficiency, the yields of barium and yttrium
carriers, and the ingrowth and decay of 228 Ac (Percival and Martin 1974).
Various other methods of radium purification by coprecipitation, ion exchange,
and radon emanation are available (Kirby and Salutsky 1964). In a recent method,
both 226 Ra and 228 Ra are collected by a barium sulfate precipitate which is weighed
to determine the barium yield, and then counted by Ge gamma-ray spectral analysis.
The counted gamma rays are emitted by the 214 Pb and 214 Bi progeny of 226 Ra
and the 228 Ac progeny of 228 Ra. The measurement for 228 Ra can be performed
immediately because 228 Ac coprecipitates with barium sulfate, but delay by 1–4
weeks is needed for ingrowth of the 222 Rn daughter of 226 Ra in the barium sulfate
(Kahn et al. 1990). Corrections are required for yield and incomplete ingrowth to
calculate the activity.

Uranium Analysis
The uranium isotopes often encountered in the radioanalytical chemistry laboratory
are listed in Table 6.2. In natural uranium, the relative decay rates at equilibrium
are 1.0 Bq 238 U, 1.0 Bq 234 U, and 0.045 Bq 235 U. Enriched (containing relatively
more 235 U and 234 U) and depleted (relatively more 238 U) combinations are also
encountered, as are 236 U in neutron-irradiated mixtures and 233 U from some pro-
cesses. These uranium isotopes emit alpha particles, characteristic 13-keV L X
rays, and generally several weak gamma rays. Several isotopes have numerous
minor alpha-particle or gamma-ray transitions that are not listed.
 6. Applied Radioanalytical Chemistry 111

TABLE 6.2. Common uranium radionuclides

Alpha particles Gamma rays
Isotope (MeV) Percent (MeV) Percent Comments
232 U 5.139 0.28 0.058 0.202
5.264 31.2 0.129 0.067
5.320 68.6
233 U 4.729 1.6 0.054 0.019
4.783 13.2 0.317 0.0080
4.796 0.28
4.824 84.4
234 U 4.605 0.24 0.053 0.12
4.724 27.4 0.121 0.041
4.776 72.4
235 U 4.209 5.7 0.144 10.5
4.322 4.7 0.163 4.8
4.358 17 0.186 54
4.392 54 0.205 4.7
4.411 2.1
4.501 1.7
4.555 4.5
4.597 5.4
236 U 4.445 26 0.045 0.079
4.494 76 0.113 0.019
238 U 4.042 0.26 0.063 4.9 (from 234 Th)
4.150 23 0.092 5.5 (from 234 Th)
4.200 73 0.766 0.29 (from 234 Pa)
1.001 0.89 (from 234 Pa)

One approach to measuring natural uranium is chemical analysis. For calculating

the decay rate, the specific activity of 238 U is 12.4 Bq/mg (0.33 pCi/μg). A common
method for detection of microgram amounts is fluorimetry of a fused aliquot. Laser
fluoroscopy has been developed as a more sensitive method. If a nuclear reactor is
available, neutron activation or the fission reaction can detect nanogram amounts
(Gindler 1962). Measurement by mass spectrometer has been applied, as discussed
in Section 17.8.1.
The stable oxidation states of uranium are 0 (as metal), +4, and +6. Uranium
has also +3 and +5 states. In nature, uranium exists as relatively insoluble oxides
such as UO2 ,U3 O8 , and UO3 and as the relatively soluble cation UO2+ 2 . Both +4
and +6 states form soluble complexes with numerous inorganic and organic ions.
Uranium has been extracted from aqueous solutions with many organic solvents
and sorbed on either cation- or anion-exchange resins under various conditions. A
commonly used procedure that also applies to the actinides is sorption on an anion
exchange resin of the chloride complex from strong hydrochloric acid, followed
by selective elution at lesser acidity. If precipitation is desired, ammonium uranate,
(NH4 )2 U2 O7 , is formed with NH4 OH, but many other radionuclides will accom-
pany the uranium salt under the basic conditions. For radiation measurement, the
uranium source typically is prepared by electrodeposition (Talvitie 1972), notably
by the American Society for Testing and Materials (ASTM) standard method
112 Bernd Kahn, Robert Rosson, and Liz Thompson

C1284-94 listed in Appendix A-1. Other actinides that emit alpha particles are
also prepared with this method. After purification, uranium isotopes usually are
identified and measured by alpha-particle spectrometry with silicon diodes. Tracer
232
U can be added to the initial solution and counted with the uranium isotopes of
interest.
Table 6.2 indicates some uranium isotopes that emit gamma rays in a significant
fraction of decays; these can be measured directly by gamma-ray spectral analysis.
In measuring 235 U by its most intense gamma ray of 0.186 MeV, interference from
226
Ra is possible. For 238 U, the listed gamma rays are emitted by the daughter
234
Th (24.1 days) and its daughter 234 Pa (1.2 min) if they are at equilibrium.

Plutonium Analysis
Plutonium isotopes of interest are listed in Table 6.3. All except 241 Pu emit alpha
particles, 14-keV L X rays, and multiple weak gamma rays; 241 Pu emits beta par-
ticles that decay to 241 Am, which emits alpha particles and gamma rays. Single or
multiple neutron reactions with 238 U and 235 U form all of the isotopes. The isotopes
that emit alpha particles have numerous unlisted minor gamma-ray transitions.
Among plutonium oxidation states from +3 to +6, the most stable forms are
+3 and +4. Conversion between oxidation states is used for purification from
other radionuclides. Plutonium is oxidized to the +4 state by hydrogen peroxide,
permanganate, and nitrite, and reduced to the +3 state by bisulfite and ascorbic
acid. A strongly acidic or complexing solution is needed to maintain the selected
state and avoid hydrolysis with polymerization (Coleman 1965).
An early conventional method for plutonium analysis was coprecipitation in
acid solution with a rare earth fluoride, dissolution in aluminum nitrate solution
with sodium nitrite to maintain the +4 oxidation state, and extraction into thenoyl-
trifluoroacetone (TTA) in benzene. Plutonium was back-extracted into dilute HCl,
the acid was evaporated, and plutonium was taken up in HCl–NH4 Cl solution

TABLE 6.3. Common plutonium radionuclides

Alpha particles Gamma rays Beta particles
Isotope (MeV) Percent (MeV) Percent (E max in MeV)
238 Pu 5.359 0.013 0.043 0.039 NA
5.454 28.7
5.498 71.1
239 Pu 5.105 11.5 0.052 0.027 NA
5.129 15.1 0.129 0.0062
5.161 73.3
240 Pu 5.123 26.5 0.045 0.045 NA
5.168 73.4 0.104 0.0070
241 Pu No alpha NA No gamma NA 0.0208
(Measure 241 Am
daughter)
242 Pu 4.856 21 0.045 0.042 NA
4.900 78.9 0.103 0.0088
 6. Applied Radioanalytical Chemistry 113

for transfer to an electrodeposition cell and deposition on a stainless steel disk

(Coleman 1965).
Currently, anion-exchange resins are used to purify and separate transuranium el-
ements. Plutonium and other radionuclides are coprecipitated with iron hydroxide.
The precipitate is dissolved in strong nitric acid and adjusted to 8 N, with sodium
nitrite added to control the plutonium oxidation state at +4. The solution is passed
through an anion-exchange resin column to retain plutonium. After washing the
resin with more 8 N HNO3 , plutonium is eluted with 10 N HCl and dilute NH4 I,
repeatedly evaporated with nitric acid to convert it to the nitrate, and then taken
up in H2 SO4 –(NH4 )2 SO4 solution at pH 2.0–2.3 (IAEA 1989). The plutonium is
electrodeposited as described above for uranium. An alternative is coprecipitation
with 0.1 mg of a rare earth such as lanthanum fluoride and collection on a smooth
filter (EPA 1984).
Plutonium, uranium, and the other actinide radionuclides can be separated on
crown-ether extractant columns (Eichrom columns are discussed in Section 3.5)
by selective sorption and elution. A sequential process collects uranium on the
first column and the other elements on the second column. Plutonium and other
actinides are then eluted with solutions specific for each element.
For tracing plutonium yield, 242 Pu is available. The plutonium isotope peaks are
measured with a Si diode and alpha-particle spectral analysis. The 239 Pu activity
is calculated from the product of the 242 Pu activity and the 239 Pu /242 Pu peak area
ratio. The 239 Pu activity includes any contribution from 240 Pu because the energies
of their peaks are almost identical. Other plutonium isotopes (except 241 Pu) are
measured at the same time in terms of characteristic peak areas at the energies
listed in Table 6.3. A mass spectrometer is used to separate 239 Pu from 240 Pu if
they must be reported separately. As indicated in Table 6.3, the emitted gamma
rays can be used for intense sources but are too weak for sensitive measurements.

Am-241 Analysis
The long-lived isotopes of americium are 241 Am (458 years), 242 Am (152 years),
and 243 Am (7400 years). All are formed by multiple neutron interactions with
uranium and plutonium. Of particular interest is 241 Am because it identifies the
erstwhile presence of its 241 Pu parent. 241 Am emits alpha particles but can also
be measured by its 0.059-MeV (36%) gamma ray. 242 Am has several isomers that
mostly emit beta particles, and 243 Am emits alpha particles.
Americium in aqueous solution is in the +3, +5, and +6 states, respectively,
as Am+3 , AmO+1 +2
2 , and AmO2 . The trivalent state is most common, but higher
oxidation states are achieved by strong oxidation. The highest oxidation states can
be used for separating americium from curium and rare earth radionuclides with
ammonium persulfate, (NH4 )2 S2 O8 (Penneman and Keenan 1960).
The trivalent actinides such as 241 Am follow the same precipitation reactions
as the trivalent rare earth radionuclides, notably with insoluble hydroxides, flu-
orides, and oxalates. Numerous solvent extraction and ion-exchange separations
from other trivalent radionuclides are reported. Americium radionuclides can be
114 Bernd Kahn, Robert Rosson, and Liz Thompson

separated from other rare earth and actinide radionuclides by meticulous control
of ion-exchange column elution.
A convenient tracer for 241 Am is 243 Am (if the latter is not in the sample).
Both conventionally are measured with a silicon diode by alpha-particle spectral
analysis. The disintegration rate is calculated as discussed above for plutonium.

6.4.2. Radionuclides with Intermediate Half-Lives

A mixture in this category may include the radionuclides listed in Table 6.4 as
well as some longer-lived radionuclides in Table 6.1. With few exceptions (89 Sr
among the major fission products in this list), the radionuclides are detected by
gamma-ray spectral analysis. Because in many cases the shorter-lived radionu-
clides have a much higher decay rate, sequential gamma-ray spectral analysis,
obtained promptly and then after intervals such as 1 day, 1 week, and 1 month,
allows more precise analysis in separate measurements for the shorter-lived and
longer-lived radionuclides.
The radionuclide content in a sample can be predicted for a known source, i.e., a
nuclear reactor or device, or a hospital. Conversely, the source may be identified by
the radionuclide content of the sample. This is particularly true when, for example,

TABLE 6.4. Common radionuclides with intermediate half-lives

Category Radionuclide Half-life Radionuclide Half-life
Major fission products 89 Sr 50.5 days 133m Xe 2.19 days
90 Y 64.1 h 140 Ba 12.8 days
91 Y 58.5 days 140 La 40.3 h
95 Zr 64.0 days 141 Ce 32.5 days
95 Nb 35.0 days 143 Ce 33.0 h
99 Mo 66.0 h 143 Pr 13.6 days
103 Ru 39.4 days 144 Ce 284 days
105 Rh 35.4 h 147 Nd 11.0 days
131 I 8.04 days 149 Pm 53.1 h
132 Te 78.2 h 151 Pm 28.4 h
133 Xe 5.25 days 153 Sm 46.8 h
Minor fission products 121 Sn 27.1 h 127m Te 109 days
123 Sn 129 days 129m Te 33.5 h
125 Sn 9.62 days 131m Te 30 h
126 Sb 12.4 days 156 Eu 15.2 days
Neutron activation 82 Br 35.3 h 136 Cs 13.1 days
in fission 124 Sb 60.2 days 239 Np 2.35 days
Common neutron 51 Cr 27.7 days 59 Fe 44.6 days
Activation products 54 Mn 312 days 65 Zn 244 days
58 Co 70.8 days 110m Ag 252 days
Naturally occurring 7 Be 53.3 days 224 Ra 3.64 days
radionuclides 210 Po 138 days 227 Th 16.7 days
210 Bi 5.01 days 231 Th 25.5 h
222 Rn 3.82 days 234m Th 24.1 days
223 Ra 11.4 days
 6. Applied Radioanalytical Chemistry 115

the mix is purely natural, or consists of fission products; certain activation products
may also be identified. In principle, when the relative amounts of various fission
products resemble the distribution for one of the fissionable isotopes (see Fig. 2.2),
instantaneous fission can be inferred. The absence of shorter-lived fission product
suggests that a considerable interval between production and measurement has
occurred.
In practice, the origin of a radionuclide mixture usually is more difficult to
identify because of ongoing radioactive decay between origin and sample, differing
degrees of retention at the origin, and partial retention along the pathway. At a
nuclear reactor, for example, fuel element cladding retains most, but not all, of the
fission products, and reactor coolant controls and waste processing retain more.
Many of activation products that may originate from nuclear reactor or nuclear
medicine facilities can be predicted from experience, but not necessarily associated
with a specific facility.
Radioisotope multiples, e.g., 89/90 Sr, 103/106 Ru, 141/143/144 Ce, 134/136/137 Cs,
133/135
Xe, 235/238 U, or 238/239 Pu, are particularly useful for attributing origin and
inferring the time from formation to sampling. With the 89/90 Sr pair, for example, if
the generation rate of 89 Sr atoms relative to 90 Sr atoms were the same, the 89 Sr:90 Sr
activity ratio is inversely proportional to the half-lives, i.e., 208. The ratio is about
170 because of the higher fission yield (the relative atom generation rate) of 90 Sr
compared to 89 Sr. After accumulation for 500 days, the ratio is reduced to about
25 because 89 Sr no longer accumulates while 90 Sr still is far from equilibrium.
Even here, conditions usually are complex because various fuel elements may be
exposed for different periods and the two radioisotopes may have been released at
different times.
More confident attribution usually is possible when radionuclides in a sample
can have only limited origin from, say, atmospheric fallout and one or two nuclear
facilities. For tritium, 14 C, 129 I, or uranium in the environment, for example, the
specific activity (radioisotope/stable isotope ratio) can indicate the origin. Cer-
tain activation products can be attributed to specific nuclear medicine or reactor
facilities.
Screening measurements for gross alpha and beta particles may not be partic-
ularly useful for fission-product mixtures because generally alpha particles are at
extremely low concentration and beta particles are contributed by numerous ra-
dionuclides. A gross screening measurement is useful for suggesting additional
analyses if it clearly exceeds the sum of the radionuclides measured individually.

89
Sr-90 Sr Analysis
Determination of 89 Sr together with the long-lived 90 Sr (see Section 6.4.1) is a
widespread analytical endeavor for radiation protection because of the similarity
of strontium to calcium in their absorption into bone mass. Strontium separation
is performed as described in Section 6.4.1 whether or not 89 Sr is in the sample.
The distinction between treating a sample that is known to have no 89 Sr and one
116 Bernd Kahn, Robert Rosson, and Liz Thompson

TABLE 6.5. Calculation of 89 Sr and 90 Sr /90 Y disintegration rate

Description Formula
Initial measurement R1 = A90 Sr ε90 Sr + A90 Sr ε90 Y D190 Y + A89 Sr ε89 Sr
Second measurement R2 = A90 Sr ε90 Sr + A90 Sr ε90 Y D290 Y + A89 Sr ε89 Sr D289 Sr
(R1 D289 Sr −R2 )
Solution for A90 Sr at the initial count A90 Sr = [ε90 Sr (D289 Sr −1)+ε90 Y (D289 Sr D190 Y −D290 Y )]

[R1 −A90 Sr (ε90 Sr +ε90 Y D190 Y )]

Solution for A89 Sr at the initial count A89 Sr = ε89 Sr

in which it may be present relates to the radionuclide measurements intended to

measure 89 Sr and 90 Sr separately.
The decay characteristics of 89 Sr, 90 Sr, and 90 Y (see Figs. 9.3 and 9.4) suggest
three approaches to distinguishing between the two radiostrontium isotopes: by
half-life, the parent–daughter relation of 90 Sr and 90 Y, and beta-particle energy. The
simple approach, described below as Alternative 1, is to perform measurements
in a proportional counter twice with an interval of 2–4 weeks to obtain the net
count rates (background subtraced), R1 and R2 , shown in Table 6.5. The different
counting efficiencies ε for 89 Sr, 90 Sr, and 90 Y are determined for the detector in
use. The decay constants λ are 0.0137 days−1 for 89 Sr and 0.260 days−1 for 90 Y.
The decay factor fraction D89 Sr for 89 Sr and the ingrowth fractionD90 Y for 90 Y
are calculated. For practical purposes, the decay fraction for 90 Sr (28.5 years) of
0.998 in a 4-week period can be considered to be 1.0. With these factors, the
disintegration rates A of 89 Sr and 90 Sr may be determined.
Alternative 1 consists of measuring the strontium precipitate as soon as possible
(within a few hours) after yttrium separation, and repeating the measurement after
2–4 weeks. As shown by the equations in Table 6.5, the first value consists of beta
particles from 89 Sr and 90 Sr, with a small contribution from the recently separated
90
Y. The latter is the calculable fraction D190 Y relative to the 90 Sr beta-particle
activity. The second value consists of the same contributors, except that the 89 Sr
beta-particle emission rate has decreased according to its known half-life, while
the 90 Y emission rate has increased to become almost the same as that for 90 Sr,
i.e., D2 for 90 Y is almost 1.0.
Alternative 2 consists of counting the 90 Sr precipitate as before, then dissolving
the precipitate in dilute acid and storing it for 2–4 weeks so that the 90 Y daughter
grows to near equilibrium. The 90 Y then is precipitated initially as hydroxide and
secondly as oxalate (see 90 Sr analysis in Section 6.4.1). The yttrium precipitate is
measured as soon as possible (within a few hours) after separation. This measure-
ment provides the 90 Y activity. The 90 Sr activity is equal to the 90 Y activity when
corrected for radioactive ingrowth and decay of 90 Y. The 89 Sr activity is calculated
from the initial measurement shown in Table 6.5 by subtracting the contribution
by the activity of 90 Sr with a small activity of ingrown 90 Y.
Alternative 1 is more precise when 90 Sr is the major component and Alternative
2, when 89 Sr is the major component. An interval in Alternative 1 of a few days
 6. Applied Radioanalytical Chemistry 117

instead of 2–4 weeks is feasible because 90 Y can be calculated for any time period,
but reduces the count rate, and hence the precision. In distinction, if results need
not be reported promptly, then the first measurements can be performed 3 weeks
after yttrium separation, when the 90 Sr/90 Y pair is in equilibrium, and the second
measurement, after 7 weeks when 89 Sr has decayed by about one half-life. The
difference in the count rate then is almost entirely due to 89 Sr decay.
The two radioisotopes of strontium and 90 Y can also be distinguished by energy
discrimination with absorber foils in a proportional counter or by spectrometer in
an LS counter. These methods usually are less precise than the two above-cited
alternatives.
The other cited radioisotope pairs among fission products are measured simul-
taneously by gamma-ray spectral analysis. Before such analysis became available
in the 1950s, these pairs were distinguished after chemical separation by dual mea-
surements such as those described above for 89/90 Sr before and after an interval
during which the shorter-lived radionuclides had decayed sufficiently to measure
the difference with the needed precision.

6.4.3. Short-Lived Radionuclides

Prompt submission to the laboratory and rapid handling for chemical separation
and counting are crucial for measuring a fission-product or activation mixture
with half-lives between 1 h and 1 day. The numerous short-lived fission products
have been tabulated with their fission yields, precursors, and progeny. Sequential
counting, promptly at first and then at increasing intervals as the shorter-lived
radionuclides disappear, is the common approach. Rapid separation may be needed
to remove radionuclides that are so intense that they obscure gamma rays from other
radionuclides. 24 Na (15.0 h) and the short-lived noble gases are examples of such
obscuring radionuclides. As examples of a rapid separation, 24 Na can be carried
on a precipitate of sodium zinc uranyl acetate, NaZn(UO2 )3 (C2 H3 O2 )9 .6H2 O, by
combining solutions of a sodium and a zinc salt with uranyl acetate in acetic acid.
The noble gases can be removed from solution by agitation or solvent extraction.
The radionuclides in this category that emit beta particles also emit gamma
rays that can be detected by spectral analysis. Short-lived radionuclides that emit
alpha particles occur in the natural decay chains and usually are identified by other
members of the decay chain that emit gamma rays. One caution to consider is
that air filters and other surfaces in the environment collect particulate progeny of
220
Rn and 222 Rn that emit alpha particles, beta particles, and gamma rays with half-
lives of minutes to hours. Observation of such emissions and decays has misled
unprepared observers into attributing these radiations to man-made radionuclides.

6.4.4. Very Short- and Extremely Long-Lived Radionuclides

Measurement of radionuclides that have half-lives of minutes or even seconds is
a research activity that requires careful preparation and testing (see Chapter 16).
118 Bernd Kahn, Robert Rosson, and Liz Thompson

Numerous techniques have been reported (Kusaka and Meinke 1961). Fundamen-
tally, all processes for sample preparation, purification, and measurement must
be brief and highly effective. Sample purification before production is helpful.
Miniature ion exchange, solvent extraction, and precipitation systems have been
developed for processing small volumes. Mechanical transfer from purification to
the counting system enhances prompt measurement. Measurement results must
take into account the radioactive decay of the radionuclide during measurement
shown in Eq. (10.2).
Radionuclides with intermediate, short, and very short half-lives include progeny
of long-lived terrestrial radionuclides and products formed by cosmic-ray irradia-
tion of gases in the upper stratosphere. All three terrestrial decay chains—starting
from 238 U, 235 U, and 232 Th—include radionuclides with half-lives of seconds and
minutes that are readily detected because they are supported by their long-lived
precursors. Prompt analysis after collection is needed to analyze very short-lived
cosmic-ray-produced radionuclides such as the isotopes of chlorine, 34m Cl, 38 Cl,
and 39 Cl, with half-lives of minutes.
For radionuclides in the very long-lived to extremely long-lived category, the
primary consideration is that a longer half-life is associated with the lower prob-
ability of decay per atom and lesser decay energy. As a consequence, analysis
for such radionuclides requires large samples, thorough purification, and mini-
mal radiation background. Moreover, energetic gamma rays are not emitted. Mass
spectrometric separation and analysis (see Chapter 17) generally is the preferred
alternative.

6.5. Standard Methods

As indicated in Section 11.2.8, several organizations publish standard methods for
radionuclide testing. Prominent among these are the ASTM, Standard Methods,
American National Standards Institute (ANSI), and International Organization
for Standardization (ISO). Some of these methods are radioanalytical chemistry
procedures while others are guides for actions such as processing samples, using
standards and controlling detectors.
ASTM is a consensus standards organization with committees that develop
and approve standard test methods. Committee members represent industry, uni-
versities, and government. Several committees publish methods for radioactivity
measurement; the main ones are D19, water (of which subcommittee D19.04 deals
specifically with radiochemical analysis of water), C26, Nuclear Fuel Cycle and
E10, Nuclear Technology and Applications.
Method development and publication requirements differ slightly among com-
mittees. All D19.04 methods must undergo round robin testing prior to approval;
C26.05.01 methods must undergo single operator testing. Appendix A-1 lists some
current ASTM standards and methods.
Standard Methods for the Examination of Water and Wastewater is a text, first
published in 1905, that has remained current through the efforts of its advisory
 6. Applied Radioanalytical Chemistry 119

committee. This committee has over 500 professionals from the ranks of the Amer-
ican Water Works Association, the American Public Health Association (APHA),
and the Water Environment Federation. Their combined effort produces the Stan-
dard Methods text; part 7000 of the text deals with the testing of water samples
for radionuclides. Appendix A-2 lists the Standard Methods, part 7000 proce-
dures. The APHA intersociety committee published Methods of Air Sampling and
Analysis (2nd ed.) in 1977.
ANSI is the U.S. representative of ISO. Appendices A-3 and A-4 list ANSI and
ISO procedures, respectively.
The benefit of a standard method is the scrutiny and evaluation that it has received
by fellow professionals. For that reason, a regulatory agency or customer may
require application of specified standard methods in the radioanalytical chemistry
laboratory. If another method is preferred because of its greater flexibility, rapidity,
or ease in application, it may be compared initially or periodically with a standard
method to confirm its reliability. Development and approval of these standards
takes much time, effort, and money. For that reason, the standards are available
from these agencies for a fee, generally ranging from $50 to $75.

6.6. Methods Development or Modification

A method development and testing function is a prerequisite to initiating laboratory
operation. Each selected method must be tested for reliability in recovering the
radionuclide of interest and removing impurities that interfere with measurement.
Method modifications are inevitable if the sample matrices in the laboratory differ
from those for which the selected method was designed.
In a functioning laboratory, development and testing is required for trouble
shooting (see Chapter 12) and introducing improved methods. Problematic pro-
cesses must be examined to determine whether the problem is due to method
formulation, analyst error, or sample matrix complications. The method as written
may be insufficiently stable to handle the wide range of radioactive or stable im-
purities in the samples, and may require more detailed specifications, added steps,
or replacement.
In contrast, the effort in performing a selected method may be excessive if the
set of samples has fewer interfering materials, either stable or radioactive, than
the matrix for which the method was designed. Simplification may subtract one
or more steps that were designed to remove interfering materials that are not in
the sample to the extent foreseen. Such changes must be tested to confirm that
the deleted steps indeed were unnecessary and that their deletion does not induce
other problems.
Even if a method is successful, a better method may become available with
greater ease in handling, better reproducibility of results, and lower cost. Improve-
ments may be feasible for the chemical procedure, radiation detection instruments,
and computer hardware and software.
120 Bernd Kahn, Robert Rosson, and Liz Thompson

One type of improvement is sequential analysis to use a single sample for

analyzing several radionuclides. This approach is beneficial if the sample amount
is limited by collection, shipping, or handling restrictions, or the largest possible
amount is required for adequate detection sensitivity, or the initial analytical steps
are identical for several radionuclides. The method must be tested to assure that
cross-contamination and other possible interference among the analyses for the
several radionuclides are minimal and/or acceptable.
The development and testing function may be part of the analyst’s responsibility
or may reside with a separate group. The former can be beneficial because it adds
to the challenge of the analyst’s work and draws on the analyst’s knowledge of the
specific process, but requires that management allot sufficient time to perform both
routine and research functions. The latter approach may benefit from the broad
knowledge of specialists devoted entirely to development and testing.
7
Preparation for Sample Measurement
BERND KAHN

7.1. Introduction
The transition from radiochemical separation described in Chapter 6 to instrumen-
tal radiation detection in Chapter 8 is source preparation for counting. The analyst
wants to prepare a source that represents the radionuclide in the collected sample,
can be measured reliably by its radiation, and is stable. The analyst selects a de-
tector that is sensitive to the radiation that characterizes the radionuclide, stable as
defined by its QA program, and calibrated for efficiency and—if needed—energy.
Source preparation concerns are addressed here for the four types of detectors that
are described in Chapters 2 and 8, but these considerations can apply to sources
prepared for measurement by other detectors.

7.2. Counting Alpha and Beta Particles with

a Gas Proportional Detector
7.2.1. Source Preparation
Sources conventionally are prepared by precipitation to permit gravimetric mea-
surement of stable isotopic carrier yield, as discussed in Section 6.3. The isotopic
carrier must be in the same chemical state as the radionuclide of interest, or the
sample must be processed to achieve this requirement when carrier is added. The
precipitating agent is selected to obtain a pure precipitate of the radioelement with
a large, but not necessarily quantitative, yield and reproducible weight. If this pre-
cipitate does not completely purify the sample, as is often the case, then previous
separation steps should have done so. The various purification steps must eliminate
extraneous solids that will add to the carrier yield and contaminant radionuclides
that will add to the count rate. The purification steps must reduce such contaminants
to a small fraction of the amount to be weighed and counted. Occurrence in nature
of significant amounts of the isotopic carrier in the sample must be determined in
control samples to correct the yield value.

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

121
122 Bernd Kahn

The carrier weight generally is in the range of 10–30 mg and the yield is specified
to be sufficiently high—typically above 50%—to reduce the uncertainty associated
with yield determination to about 1%. To maintain constant weight and counting
efficiency, the precipitate should not be volatile, hygroscopic, flaky, or powdery.
Any drying or heating during source preparation must not affect comparison with
initial carrier weight due to a change in chemical form or differences in degree of
hydration.
Sources are also prepared by pouring a liquid or slurry onto a planchet and then
evaporating it to dryness. Care must be taken to avoid nonuniform distribution
because solids tend to dry as rings and to accumulate near the lip of the planchet.
Adding a drop of a solvent such as alcohol that reduces surface tension in water
before the onset of dryness can reduce such uneven deposition (Chieco 1997).
Planchets have been designed with concentric ridges to distribute evaporated solids
more evenly. Evaporation on a level surface prevents lopsided accumulation of
solids. Evaporation must be sufficiently slow by controlling heat lamp distance to
avoid sample loss by spattering.
Electrodeposition on metal disks is used to prepare very thin sources of the
radionuclides discussed in Section 3.7. At very low concentrations in water,
certain radionuclides may be deposited on metal disks by sorption or sponta-
neous electrodeposition (Blanchard et al. 1960). The former yields a relatively
low percent deposition on more noble metals, as in the case of 144 Ce deposited
on gold. In contrast, spontaneous electrodepostion results in near 100% yield
on a less noble metal, for example, in depositing 110m Ag or divalent 59 Fe on
magnesium. Yield must be determined separately for this source preparation
step and then combined with the yield observed for the chemical purification
steps.

7.2.2. Other Yield Determinations

Alternatives for yield determination in the absence of stable isotopic carrier are
nonisotopic stable carrier and radioactive tracer, as discussed in Sections 4.5
and 6.3. Technetium, promethium, and the elements heavier than bismuth have
radioactive but no stable isotopes.
The yield for a low-mass sample, e.g., 1 mg or less for alpha-particle mea-
surement, can be determined with nonisotopic carrier in an aliquot taken before
preparing the counting source. The analytical technique can be instrumental, such
as colorimetry or atomic absorption spectrometry. Subsequent source preparation,
by precipitation, evaporation, or electrodeposition, must be quantitative or highly
reproducible so that a reliable yield value for this final step can be included in the
total yield.
If no carrier or distinctive tracer is available, recovery must be quantitative or
estimated by separate yield determinations with a solution to which has been added
a known amount of the radionuclide of interest. This yield is measured with each
batch of samples. Separate determinations require consistent and high yield.
 7. Preparation for Sample Measurement 123

TABLE 7.1. Source thickness relations

Dimension Ring and disk filter Planchet
Diameter (cm) 1.6 5.0
Area (cm2 ) 2.0 20
Area mass (mg/cm2 ) (20 mg sample) 10 1.0
Thickness at density 3.0 g/cm3 (cm) 0.0033 0.00033
Energy of β particle stopped from sample bottom (keV) 80 15
Energy of α particle stopped from sample bottom (MeV) 7 1

7.2.3. Source Form

Thin sources are prepared for counting alpha and low-energy beta particles that
are readily absorbed in solids because counting efficiency depends on sample
thickness. Counting efficiency is determined as a function of source thickness by
counting several radionuclide sources in the range of source thickness expected
for the usual fluctuations in carrier yield. Sources are prepared by mixing a ra-
dionuclide standard with carrier and preparing filter sources or evaporated slurries
on planchets for various carrier masses. A smooth curve is drawn through the
measured points to plot counting efficiency (i.e., the net count rate/disintegration
rate) vs. sample mass.
The counting efficiency calibration must match the dimensions, uniformity, and
electron density of the routinely analyzed sources. Filtering a precipitate with
suction usually prepares a sample of uniform thickness unless the sample is too
small to cover the filter or so thick that it piles up at the edge.
A source is sufficiently thin if it has an area mass (density × thickness) of a
few milligrams per square centimeter. The relation of area mass, mass, thickness,
and area is shown in Table 7.1 for sources on a ring and disk or on a planchet. An
example of counting configuration is shown in Fig. 8.4. Area masses of 10 and
1 mg/cm2 will stop beta particles with energies below 80 and 15 keV, respectively,
in an aluminum-like material (see Fig. 2.4). A source with an electron density
similar to aluminum and area mass of 10 mg/cm2 will stop alpha particles below
about 7 MeV energy, as shown in Fig. 2.3.
Although the electron density of the actual source affects attenuation, the listed
values suggest the magnitude of the energies of the alpha particles and weak beta
particles from the bottom of the 10-mg/cm2 source will be stopped. Alpha particles
of that energy and a progressively higher fraction of beta particles that originate
nearer the surface that faces the detector will reach this surface. Other barriers
will combine to further reduce the energy of the emerging particles; these barriers
include any source covering, the air between source and detector, and the detector
window. If the sum of attenuating media exceeds the range of the alpha and beta
particles, none of the particles are detected.
Thinner sources are measured with higher counting efficiency per unit mass, but
the highest count rate per sample is attained with an “infinitely thick” source, the
thickness of which equals the maximum particle range. This configuration can be
useful for detecting the presence of very low levels of radioactivity, although it is
124 Bernd Kahn

of questionable value for quantification. The mass of an “infinitely thick” source

is irrelevant for efficiency calculations (except that the surface of a thicker sample
is nearer the detector window above it) because radiation emitted only near the
surface of such sources is detected.
In a very thin source (<0.1 mg/cm2 ), the alpha particle flux at the surface that
faces the detector, Rs in particles/s, is one-half of the disintegration rate, as shown
in Eq. (7.1a) from Finney and Evans (1935):
 
Aa 2x  x 2
Rs = − for x <  (7.1a)
4m  

In the equation, A is the source activity in Bq,  is the alpha particle range
in mg/cm2 , a is the source area in cm2 , m is the source mass in mg, and x is
the source thickness in mg/cm2 . The equation indicates that, for samples with a
constant activity per unit mass, the surface flux decreases linearly with thickness
to one-fourth the disintegration rate when the source thickness equals the range,
which is about 7 mg/cm2 for the 5.5-MeV alpha particle emitted by 241 Am. Beyond
this thickness, the surface flux is predicted to remain constant:

Aa
Rs = for x ≥ (7.1b)
4m
Equation (7.1b) shows that the counting efficiency varies inversely with the sam-
ple mass when the thickness exceeds the alpha-particle range, i.e., at “infinite
thickness.”
The counting efficiency, discussed in Section 8.2.1, is related to the flux at
the surface in terms of the source-detector geometry and attenuation of alpha
particles between source surface and the sensitive-detector volume. Table 8.3 pre-
dicts a 31% counting efficiency for the detector-source configuration shown in
Fig. 8.4 (but with the ring-and-disk replaced by the 20-cm2 planchet source).
In practice, the counting efficiency is considerably lower than 31%, as indi-
cated by the measured alpha-particle counting efficiency curve shown in Fig.
7.1. The reduction in efficiency even at 0 mg/cm2 is due to alpha particles that
pass through the source, air, and window at angles not perpendicular to the de-
tector for distances that exceed their range. The curve in Fig. 7.1 is approxi-
mately a straight line from 0 to 60 mg (i.e., 3 mg/cm2 ) and then curves, as the
count rate becomes constant at about 130 mg, the alpha-particle range in the
source.
The decrease in counting efficiency with thickness in Fig. 7.1 emphasizes the
benefit of a thin source for counting alpha particles at low levels. Thin sources can
be prepared by using only a small amount of carrier and a light precipitating reagent,
with purification processes other than precipitation such as electrodeposition, ion
exchange, and solvent extraction, and by flaming the source to remove organic
compounds.
For beta particles, reasonable estimates of the curve ε/ε0 for counting efficiency
as function of mass have been based on their approximately exponential attenuation
 7. Preparation for Sample Measurement 125

35%

Geometric efficiency
30%
Alpha (Am-241)

25% Beta ( Cs-137 )

20%
Efficiency

15%

10%

5%

0%
0 20 40 60 80 100 120 140 160
Weight (mg)

FIGURE 7.1. Alpha- and beta-particle efficiency self-absorption curves for 241Am and 137 Cs,
respectively, on 5-cm-diameter planchet measured with proportional counter.

over part of the attenuation curve (itself an empirical finding), so that

ε0 (1 − e−μ x )
ε= (7.2)
μx

In Eq. (7.2), ε is the counting efficiency, subscript 0 refers to the efficiency at

zero source mass, x is the area mass of the sample in mg/cm2 , and μ is an empir-
ically derived attenuation coefficient in cm2 /mg. This attenuation coefficient has
been related to maximum or average beta particle energy for specific source-to-
detector configurations and source backing material (Evans 1955). Typical curves
of counting efficiency vs. sample thickness are shown in Fig. 7.1 for 137 Cs on a
planchet and in Fig. 7.2 for 89 Sr,90 Sr, 90 Y, and 14 C on a ring and disk. The value
of μ is purely empirical because Eq. (7.2) does not consider beta particles that are
emitted at an angle, scattered back, or not in conformance with exponential atten-
uation (notably near the maximum energy). Back scattering significantly increases
the beta-particle counting efficiency above its geometrical efficiency.
More recently, reliable Monte Carlo simulation of electron interactions in the
source, its environment, and the detector has been used to calculate the detector
efficiency curve. The efficiency curves in Fig. 7.2, calculated with the Monte
Carlo n-particle code, version 4, from available beta spectral data, agree with
measured efficiency values within the uncertainty of the measurements (Nichols
2006).
126 Bernd Kahn

0.60

0.50
Counting efficiency
0.40 Yttrium-90
Strontium-89
0.30 Strontium-90
Carbon-14
0.20

0.10

0.00
0 10 20 30 40
−2
Sample thickness (mg SrCO3 cm )

FIGURE 7.2. Beta-particle efficiency self-absorption curves for 89 Sr, 90 Sr, 90 Y, and 14 C as
1.6-cm-diameter ring-and-disk source, calculated by Monte Carlo simulation. (from Nichols
2006).

7.2.4. Gross Counting

Determination of gross alpha- or beta-particle activity is a measurement for which
the source is prepared for counting without chemical separation. The measurements
are performed to:
r assure that the maximum permissible concentration of even the most hazardous
radionuclide in the sample is not exceeded;
r eliminate need for further analysis if only a single radionuclide is known to be
present;
r compare to the sum of individually analyzed radionuclides for checking whether
major radionuclides have been missed;
r indicate the magnitude of total alpha- or beta-particle activity in an unfamiliar
sample.

For the first three applications, a radionuclide- and mass-specific counting ef-
ficiency must be selected. For the fourth application, a thin sample—below
2.5 mg/cm2 for alpha-particle counting—should be prepared so that efficiency
values are similar at commonly encountered energies. For counting beta particles,
the sample should not exceed 10 mg/cm2 . An intermediate-energy (e.g., 0.6–0.8
MeV β max ) radionuclide standard provides reasonable efficiency estimates except
that the activity of a radionuclide that emits only low-energy beta particles will be
underestimated.
An aliquot of the sample as liquid or slurry usually is dried on a planchet and
then counted for gross activity measurement. In some instances, the radionuclides
are coprecipitated by a carrier such as ferric hydroxide or manganese dioxide that
is expected to collect many of them (ICRU 1972), and this solid is filtered or
poured onto the planchet as slurry and then dried. Under favorable circumstance,
 7. Preparation for Sample Measurement 127

the radionuclide concentration is high enough that a thin sample can be prepared
and attenuation by the sample itself is not a major concern. A thicker sample
may yield detectable beta-particle counts in the available time period from a less
radioactive sample, but the self-absorption fraction is in doubt if the radionuclides
are not identified.
Gross alpha- and beta-activity counting can be a useful screening process. It
should not be used to delineate radioactivity levels without associated specific
radionuclide measurements because of uncertainty about what and how much is
being measured. Moreover, some radionuclides may be lost in processing due to
volatility during evaporation or incomplete carrying during precipitation.

7.3. Counting Alpha and Beta Particles with a Liquid

Scintillation Detector
7.3.1. Source Preparation
Radionuclides typically are measured in a liquid scintillation (LS) counter as an
aqueous solution that is pipetted into a vial, mixed with a scintillation cocktail, and
counted within a radiation shield by a dual photomultiplier tube (PMT) system. As
described in Section 8.3.2, the cocktail consists of a scintillating substance in an
organic solvent such as benzene. The mixture may be a cloudy emulsion, but this
lack of visible-light transparency does not interfere with the effective transmission
of the generated 3-eV photons to the PMT.
Commercial LS systems are designed to use 20-ml screw-top vials available in
glass or plastic. The sample volume that results in the most sensitive detection per
unit volume is usually 10 ml aqueous solution per 10 ml cocktail, but the optimum
volume ratio should be determined empirically. Smaller or larger sample vials
have been used in systems designed to meet specific needs, such as limited sample
volume or greater sensitivity, respectively. Very large solid scintillator systems
have been designed for whole-body counting (Horrocks 1974).
An aqueous sample may be added to the cocktail directly, after minor prior
processing, or at the end of a radiochemical separation procedure. Direct addition
is the equivalent of gross activity counting discussed in Section 7.2.4 except that
some spectral analysis may be possible. Alpha particles can be differentiated from
beta particles by deposited energy, pulse shape, and decay time. Self-absorption
is of no concern. Quenching and luminescence, discussed in Section 8.3.2, often
occur. Identification by maximum beta-particle energy is approximate, and requires
comparison to radionuclide standards.
A clearer understanding of what is being measured results from separating and
purifying the sample radiochemically before counting. Sample volume should
be sufficient for mixing with the cocktail and for taking an aliquot to measure
yield. Quenching and luminescent impurities should be removed to the extent
necessary for reliable measurement from samples destined for LS counting. Some
quenching of the final sample can be anticipated—water is a quenching agent—but
128 Bernd Kahn

a consistent level of quenching leads to more reliable results than wildly variable
levels.
The main advantage of the LS counter over the gas proportional counter is the
intimate mixing of the radionuclide with the scintillating detection medium to elim-
inate loss of alpha and beta particles by absorption outside the detector. Moreover,
the LS geometrical efficiency is almost 100%. The practical counting efficiency is
above 90% for alpha and more energetic beta particles. Energy loss by quenching
is minor except for low-energy beta particles from radionuclides such as tritium.
One disadvantage of the sample–cocktail mix is impermanence. The stability
of the prepared vial contents should be tested over time. Typically, a sample can
be recounted after a few days or weeks, but should not be considered reliable after
longer periods.

7.3.2. Beta Particles

LS counters are suitable for measuring radionuclides that emit only very low-
energy beta particles or electrons, notably tritium (E max = 18.6 keV). When tritium
is measured as tritiated water and its activity is reported relative to water weight
or volume, no yield measurement is needed. Liquid samples, e.g., water from the
environment, process streams, urine, or dissolved solids, can be counted directly or
purified by distillation. Results for purified samples are more reliable to the extent
that the radioelement can be identified, quenching is stabilized, and luminescent
contaminants are removed. Reagents may have to be added to the distillation flask
to hold back other potentially volatile radionuclides, such as radioactive iodine,
carbon, ruthenium, or technetium.
Examples of other radionuclides counted with an LS system because they emit
very low-energy beta particles or electrons are 63 Ni (E max = 66 keV) and 55 Fe
(E e = 5.2 keV, 61%), respectively. Each radionuclide normally is purified to re-
move interfering substances, including other radionuclides, and then prepared as a
solution with suitable volume for LS counting and yield determination. Radionu-
clides that emit energetic beta particles are measured by LS counting because their
counting efficiency is near 100% and sample processing is convenient.

7.3.3. Alpha Particles

Measuring alpha particles with an LS counter is an attractive option because the
counting efficiency is near 100% and no self-absorption problem exists. After
the usual sequence of separations for radionuclides such as thorium, uranium,
and transuranium isotopes, the radionuclide is prepared in the final solution for
counting and yield determination. A tracer that emits alpha particles at a sufficiently
distinct energy is added initially to measure yield. The factor that controls detection
sensitivity is the background, typically of 1–2 c/m in the alpha-particle energy
region of the LS counter.
Alpha particles emitted by a mixture of several radionuclides can be analyzed
simultaneously by energy discrimination with an LS spectrometer. Such separation
 7. Preparation for Sample Measurement 129

is better than for beta-particle continua because alpha particles are emitted with
discrete energies. Resolution of alpha-particle energy peaks in conventional LS
systems is about 0.50 MeV. It is reduced to about 0.40 MeV by deoxygenating
the sample to reduce quenching. This resolution is far poorer than for solid state
detectors (see Section 7.4), but spectral analysis can be useful for radionuclide
mixtures in which only a few alpha-particle groups occur at widely separated
energies. A photon/electron rejecting alpha liquid scintillation (PERALS) system
has been developed that discriminates against beta particle scintillations by their
longer decay time, improves the resolution to about 0.25 MeV, and reduces the
background to about 0.001 c/m (McDowell 1992).

7.3.4. Special Applications

Ingenious permutations of the conventional combination of aqueous solution with
organic cocktail have helped the analyst. Some samples are extracted into an or-
ganic solvent that is soluble in the scintillation cocktail to eliminate water quench-
ing. Sample purification and counting can be integrated by collecting a dissolved
or airborne radionuclide on a sorption or ion-exchange matrix that contains scin-
tillating material and is transparent to its scintillations. This material is washed
and placed directly into the counting vial for measurement.
Collected particles and prepared precipitates have been counted by dispersing
and stabilizing them in a cocktail with silica gelling agent (ICRU 1972). Air filters
and smears have been counted by placing them directly into a scintillation cocktail.
The counting efficiency for each type of sample must be calibrated for specific
configurations, radionuclides, and interfering substances.
Radon-222 gas in water is counted after pipetting the water sample without
agitation (to avoid losing the gas) into a counting vial that contains an organic
cocktail solvent not miscible with water. Almost no air space should remain in
the vial. The vial is closed tightly, shaken to transfer the radon gas to the or-
ganic phase in which it is far more soluble (EPA 1978), and counted after a
4-h interval to permit 222 Rn progeny ingrowth. The system can be calibrated
by pipetting the same volume of a 226 Ra standard aqueous solution into a vial
that contains the usual volume of immiscible cocktail and sealing the vial. Af-
ter 4 weeks to accumulate the 222 Rn progeny (three alpha particles from 214 Po,
218
Po, and 222 Rn and two beta particles from 214 Pb and 214 Bi) in the standard
to near equilibrium, the mixture is shaken to transfer the 222 Rn to the organic
phase, and the standard is counted. Alpha and beta particles are counted together
at a composite counting efficiency of about 420%, with a background of about
20 c/m.
Flow-through systems have been developed for monitoring radionuclides, es-
pecially tritium, in facility effluent, as described in Section 15.4.3. Typically, the
sample accumulated in each intermittent collection period is mixed automatically
with the scintillation cocktail and transferred to a counting chamber. While the
sample is counted, the next sample is collected. After counting, the sample is
discharged and the counting chamber is flushed for cleaning.
130 Bernd Kahn

Solid scintillators such as anthracene have been tested as alternatives for contin-
uous measurements (ICRU 1972). The scintillation material and the radionuclide
collector are combined, as discussed above for scintillation beads (Winn 1993).
The operating characteristics of these devices, such as the sample volume, count
accumulation period, energy resolution, detection efficiency, and background must
be arranged to satisfy radionuclide concentration limit specifications.
The LS counter also measures Cherenkov radiation (see Section 2.4.3) in water
without addition of scintillation cocktail. The radiation is generated in water by
electrons at energies above 0.265 MeV. Hence, about one-half of beta particles in a
group with E max of 0.8 MeV generate this light, and the fraction is larger for groups
with higher maximum beta-particle energies. The advantages of this process for
detecting beta particles are that the water sample volume can be the full 20 ml and
the sample is stable. The drawbacks are that quenching still can occur, few photons
are produced per beta particle, and the PMT usually is not optimal for detecting
the wavelength of the Cherenkov radiation. A 1-MeV beta particle produces about
200 Cherenkov scintillations in water compared to 10,000 scintillations in an LS
cocktail (Knoll 1989).

7.4. Counting Alpha Particles with a Silicon

Detector and Spectrometer
The main radionuclides that are measured in this system are isotopes of thorium,
uranium, and plutonium. Others are the longer-lived isotopes of heavy elements
such as radium, protactinium, neptunium, americium, and curium. Conventionally,
an isotopic tracer of known activity that emits alpha particles is added at the
beginning of the radiochemical procedure (see Section 6.3.2). The solution from
which the sample for counting will be deposited is thoroughly purified by the
radiochemical procedure to remove interfering radionuclides and solids.
The critical requirement for alpha-particle spectrometry to achieve good char-
acteristic peak resolution, shown in Fig. 9.1 and described in Section 8.3.3, is the
preparation of a very thin and uniform source. Less rigorous criteria for resolution
can be met if the several radioisotopes to be measured and the tracer have widely
different alpha-particle energies, only a single radionuclide is known to be present
either initially or after chemical separation, or the measurement is for screening
purposes. A sample may be prepared for screening with limited prior separation
or none, and without tracer addition, if previous tests have demonstrated that the
process causes no losses for all radionuclides of interest.

7.4.1. Tracer Addition

Isotopes available as tracers include 229 Th, 232 U, 242 Pu, and 243 Am. Only one or
two tracers usually are available per element because of the requirements of dis-
tinctive alpha-particle energy, long half-life, and absence of contamination by the
 7. Preparation for Sample Measurement 131

radioisotopes of interest. Some radioactive tracers must be purified at intervals of

a few percent of the progeny half-life to remove ingrown radioactive progeny that
emit alpha particles. Prior tracer purification is not necessary if the impurity is re-
moved as part of the chemical separation before counting sample preparation, or if
the ingrown alpha-particle peaks do not interfere with the peaks of the radioisotope
of interest and the tracer.

7.4.2. Source Preparation

Electrodeposition is the preferred method for preparing a thin, smooth, uniform,
and stable source for counting, as discussed in Section 3.7. Empirically developed
procedures are applied to the heavy elements in aqueous solution because they
cannot be reduced to the metal; the source is deposited at the cathode in what is
believed to be hydroxy forms after partial reduction. Deposition is in acid solution,
under various conditions that yield near-quantitative recovery from the solution.
Test results specify the volume, reagent content, applied volts and amperes, and
optimum time period (see Section 6.4.1). The cathode can be a platinum disk, but
a polished stainless steel disk is satisfactory.
In an application of the spontaneous deposition discussed in Section 7.2.1,
sources of 210 Po are prepared on a nickel planchet. This process separates this
radionuclide from other radon progeny or heavy elements that emit alpha particles.
Sources can also be prepared by coprecipitating actinides with 1 mg or less of
rare earth—e.g., lanthanum—carrier. Reasonable alpha-particle peak resolution
results when the precipitate is counted on a filter that has been prepared with
additional rare earth substrate (Chieco 1997).
Very thin and uniform sources have been prepared on targets in a mass spec-
trometer or a thermal volatilization system (see Section 3.6.1). Such preparations
require a relatively intense radioactive source.

7.5. Counting Gamma Rays with a Germanium

Detector and Spectrometer
Samples with radionuclides that emit gamma rays are counted conveniently with
an intrinsic, high purity, germanium (Ge) detector with spectrometer. Gamma rays
from the sample can be counted with minimal processing—often the sample is
counted directly in a container that has been calibrated for counting efficiency.
The container is tared for weight and calibrated for volume; when filled, it is
weighed to calculate the activity per this mass and determine source density for
calculating the self-absorption factor. Samples counted to report activity relative
to the flow rate, such as filters and sorption media for air and water samples, must
conform to a calibrated source volume even if volume and mass do not enter into
the calculation of activity.
132 Bernd Kahn

7.5.1. Source Preparation

Many sources can be prepared directly for gamma-ray spectral analysis. Exceptions
are sources that require concentration, stabilization, or radionuclide purification.
Concentration improves detection sensitivity, stabilization should prevent changes
in source configuration or content, and purification can eliminate interference with
counting characteristic peaks.
Liquid samples are concentrated by evaporation or by separating the radionu-
clide of interest with processes such as filtration, precipitation, solvent extraction,
ion exchange, sorption, and distillation. Solid samples are dried, ashed, com-
pressed, or physically sorted. Gases can be sorbed, condensed, or compressed.
Some steps may be performed at collection, others, in the laboratory. Recovery
of the radionuclide of interest must be quantitative or at a previously determined
fraction.
For measurement reproducibility, the source must be stable and of uniform
shape. Sources should be prepared so that solids do not shift or decompose. Ra-
dionuclides in liquids should not settle, float to the surface, or adhere to container
walls. Gaseous radionuclides should not escape from the solid or liquid matrix to
collect in vapor spaces or leak from the container. The applied counting efficiency
must take into consideration any known nonuniformity of radionuclide distribution
in the source.
The good energy resolution a Ge detector system reduces the possibility that a
second gamma-ray peak resulting from the presence of other radionuclides in the
sample will obscure the gamma-ray peak of the radionuclide of interest. If this
does occur, the interfering radionuclide must be removed.

7.5.2. Direct Measurement

Direct measurement of relatively large samples is feasible because energetic
gamma rays—those above about 0.1 MeV—will reach the detector with only minor
attenuation in large, multiliter, samples. The Ge counting efficiency, although less
than for a NaI(Tl) detector of the same dimensions, can be optimized by placing
a large Ge detector in close proximity to a large sample. Sensitivity is improved
with re-entrant beakers that surround the detector, with well-type detectors, and
with multiple detectors connected in parallel to view a large sample. Larger sam-
ples may be of irregular shape for which the geometry can only be approximated.
Extreme examples of the latter are animals and humans monitored in whole body
counters.

7.5.3. Calibration
Source containers must be sufficiently rigid to maintain their shape. They must
resist attack and decomposition from contents such as organic solvents, acids,
bases, and biological material. They must be closed to prevent spillage, permit
storage, and protect the detector and its environment from contaminants. In some
 7. Preparation for Sample Measurement 133

instances, samples are sealed to prevent inleakage of air or escape of gases. One
technique is to place a solid or liquid sample in a metal can and then seal its
lid (Chieco 1997). Gaseous samples are transferred to an evacuated container
connected by tubing, valves, and a pump with the sample container.
Efficiency calibration requires a sample of defined shape and volume placed
at a defined location relative to the detector. A sample frame may be needed for
reproducible placement. Calibration can be performed with a radioactivity standard
or by Monte Carlo simulation (see Sections 8.2 and 10.5).
Although gamma rays are much less subject to attenuation than alpha and beta
particles, a density correction is needed if the density of the sample deviates sig-
nificantly from the density of the calibration standards. The effect of density on
self-absorption for both the standard and the sample is estimated by Eq. (7.2); μ
for this purpose is the photon attenuation coefficient in cm2 /g and x is the sample
area density in g/cm2 . Values for μ in some common materials are listed in Table
2.2 and in its cited reference. If a large set of samples with consistent density is
analyzed, it may be possible to prepare radioactivity standards at the same density
to avoid the need for correction. Interpolating efficiency values as a function of
density is feasible at energies above 0.1 MeV because the effect of minor density
difference on counting efficiency is small.
When initial nonuniformity cannot be eliminated by mixing the sample, e.g., in
whole-body counting, the radionuclide distribution can be estimated by calibrating
with standards in various configurations and conditions. The results can establish
a range of possible counting efficiencies. Changing sample location relative to the
detector for separate counts may yield further information concerning radionuclide
distribution. Monte Carlo simulations can also provide needed efficiencies.
A simple example of sample nonuniformity is an air cartridge that may have
retained a radionuclide either on the front surface or with decreasing concentration
in depth, or distributed uniformly throughout. Counting the cartridge from both
sides and comparing the results to a simulation model will permit calculating the
activity as a function of depth distribution.
8
Applied Radiation Measurements
JOHN M. KELLER

8.1. Introduction
The radiation detection systems employed in radioanalytical chemistry labora-
tories have changed considerably over the past sixty years, with significant im-
provement realized since the early 1980s. Advancements in the areas of material
science, electronics, and computer technology have contributed to the develop-
ment of more sensitive, reliable, and user-friendly laboratory instruments. The
four primary radiation measurement systems considered to be necessary for the
modern radionuclide measurement laboratory are gas-flow proportional counters,
liquid scintillation (LS) counters, Si alpha-particle spectrometer systems, and Ge
gamma-ray spectrometer systems. These four systems are the tools used to identify
and measure most forms of nuclear radiation.
Some operating parameters can be considered in common for all these detec-
tors, notably detection efficiency and the radiation background. Additional param-
eters pertain only to detectors with associated spectrometers. These parameters
concern the radiation energy peaks that identify and quantify radionuclides, and
include energy calibration, energy resolution, peak-to-Compton ratio, and peak
shape.
Each of the detection instruments has a specialized field of application. The gas-
flow proportional counter is the laboratory workhorse for measuring radionuclides
that emit alpha and beta particles in samples before radionuclide separation and
then again in the purified fractions. The LS counter also measures radionuclides that
emit alpha and beta particles and is particularly useful for measuring low-energy
beta particles. The alpha-particle spectrometer is applied for purified samples to
measure radioisotopes, notably of the heavy elements, that emit alpha particles. The
gamma-ray spectrometer is the laborsaving device for measuring photon-emitting
radionuclides, in most cases without chemical separation.
Many ingenious applications of these detectors go beyond such specialization
so that their use can be a matter of matching sample characteristics, radiation
types, the radionuclide of interest and associated impurities, required sensitivity,
and convenience in sample preparation, or even analyst preference. Various other
detection systems and instrumental techniques that are mentioned at the end of

Oak Ridge National Laboratory, Oak Ridge, TN 37831

134
 8. Applied Radiation Measurements 135

this chapter should be considered when available. Some of these other detectors
may be more appropriate than the four conventional ones for analyzing samples
with very high or low activity and for radionuclides that otherwise would require
complex radiochemical separations prior to analysis.

8.2. Common Operating Parameters

8.2.1. Counting Efficiency
Dependable radiation measurement requires a professional understanding of ra-
diation detector calibration, as it applies to each detector and sample. Several
calibration techniques are applied:
r Prepare and count a radionuclide standard for each radionuclide and source–
detector configuration.
r Prepare and count selected radionuclide standards to determine curves of count-
ing efficiency vs. energy for each source–detector configuration.
r Determine separately and then combine all factors that constitute the counting
efficiency.
r Apply Monte Carlo simulation to determine curves of counting efficiency vs.
energy for each source–detector configuration.

A careful analyst will apply more than one of these techniques to assure reliable
calibration. The first three will be discussed here to review the basic elements of
counting. The fourth approach is a numerical analysis technique that should be
given proper grounding in an appropriate text (see Briesmeister 1990) and proper
introduction to the student after the basics of counting are understood.
The crucial characteristic that applies to all detector calibration is the practical
counting efficiency—the ratio of the observed measurement to the activity of the
sample in Bq (or pCi) per sample that may be converted to activity per unit mass
(or volume or flow rate). In the simplest case, a measure of the practical counting
efficiency ε is given by the relationship

RG = Aε + RB (8.1)

where A is the disintegration rate (or activity), RG is the observed count rate of the
sample, and ε is the counting efficiency. A measured background count rate RB
must be subtracted from the observed count rate to calculate the actual count rate
of a sample. The observed count rate is also referred to as the gross count rate and
the value after subtracting the background, the net count rate. Most measurements
of the count rate of a sample are performed to determine its disintegration rate, but
some measurements are performed only to compare count rates, e.g., in half-life
measurements or for samples taken to evaluate a purification step.
In the first two calibration approaches cited above, a certified radionuclide
standard is measured to obtain the practical counting efficiency of the detector,
136 John M. Keller

according to Eq. (8.2):

RG-std − RB
ε= (8.2)
Astd
Here, RG-std is the count rate observed for the standard and Astd is its reported
activity. The calibration standard must be traceable to a national standardization
agency—in the United States, the National Institute of Standards and Technology.
Calibration may be by direct counting of a source in the form prepared by a
standardization group, or it may involve processing a standard solution to prepare
a counting source. In an indirect approach to calibration, a source may be counted
that was previously counted in the laboratory with a calibrated detector.
The radionuclide standard must be accompanied by a certificate (see
Section 11.2.6) with detailed descriptions of its chemical, physical, and radio-
logical characteristics and the uncertainty of the reported disintegration rate. The
uncertainty of calibration depends on the reported uncertainty of the standard,
compounded by the uncertainty due to source preparation and measurement.
The calibration value can be accepted if it agrees, within the calculated uncer-
tainty, with the value obtained from another standard or by other means, such as
a calibration curve, calculations of individual parameters described below, and by
Monte Carlo simulation. The reliability of these approaches depends on meticu-
lous description of the source, the detector, and their environment, and must be
defined by the uncertainty associated with these factors.
The third listed calibration approach involves the separate determination and
subsequent combination of all factors that constitute the counting efficiency. These
factors depend on detector volume, source-to-detector geometry, and the extent of
radiation attenuation and scattering. They have differing degrees of impact in
different detectors. Some approximate counting efficiency and background values
for the detectors discussed here are listed in Table 8.1. The listed PERALS (Photon

TABLE 8.1. Typical values of radiation detector counting efficiency and background
Sensitive
Detector volume Radiation Energy Efficiency Background
type (cm3 ) type (keV) (%) (c/m) Conditions
Proportional 28 alpha 4000–9000 20 0.05 Anticoincidence
beta 400–3500 45 1
LS 20 alpha 4000–9000 >95
20 beta 18 25 2 10 ml H2 O
20 beta 250–3500 >90 5 20
LS 1 alpha 4000–9000 99.7 0.001 PERALS
Si spec 0.06 alpha 4000–9000 25 0.0001 Electrodeposited
source
Ge spec 85 gamma 100–103 0.15 0.5 Flat source
500–505 0.063 0.3 Flat source
1000–1006 0.035 0.15 Flat source
2-000–2007 0.011 0.011 Flat source

Note: maximum beta-particle energy is given

 8. Applied Radiation Measurements 137

electron rejecting alpha LS) is an LS counter specially designed for improving

alpha-particle spectral analysis by increasing counting efficiency and decreasing
background.
For calculating the practical counting efficiency from all contributing factors,
parameters must be considered for the source–detector system, the detector itself,
and the radioactive sample. The source–detector system contributes the following:
r Intrinsic efficiency of the detector, εi
r The geometric relation of sample and detector, or solid angle 
r Factor for attenuation loss in the sample, detector window, and intervening
material l
r Factor for scattering into the detector by sample support and shield s

The detector electronic system controls the following:

r Factor for pulse losses and discrimination in the pulse processing system p

The radionuclide and the sample affect the following:

r The decay fraction of the measured radiation, f
r The fractional extent of radioactive decay or ingrowth, D
r The chemical recovery fraction or yield Y
r The sample mass or volume, m or v
The radionuclide concentration, a, the “massic activity,” is calculated from the
count rate and the combined factors, which constitute the overall counting effi-
ciency by:
RG − RB A
a= = (8.3)
εi lsp f DY m m
If the count rate is in c/s, then the units of A are Bq and of a are Bq per unit mass.
Discussion of these individual parameters follows.
The intrinsic efficiency of a detector is a function of its dimensions and the
electron density of its detection volume. For alpha- and beta-particle detection
in a typical 1-cm deep, gas-filled proportional counter, creation of only a few
electron–ion pairs is needed to achieve high probability of detection. The Poisson
distribution (see Section 10.3.5), with probability Pr(x) for the number of events
x with a mean number of events μ (see Eq. 10.14), suggests that the probability is
only 0.01 when the number of events is zero at a mean number of 4.6:
Pr(0) = e−μ (8.4)
Therefore, the probability of detecting an event is 1–0.01, or 99%. The specific
ionization in air is more than 50 ion pair per cm for beta particles and about 400
times as great a value for alpha particles. The specific ionization in the filling gas is
similar. Hence, a travel distance of 0.1 cm will exceed 99% intrinsic efficiency for
counting both alpha and beta particles in the typical proportional counter, although
most particles deposit only a fraction of their energy in the gas.
138 John M. Keller

TABLE 8.2. Relation of intrinsic efficiency to

gamma-ray energy in 6-cm-thick Ge detector
Energy (MeV) Efficiency (%)
0.1 >99
0.5 94
1.0 87
3.0 83

In an LS counter, the cocktail is about a thousand times as dense as counter gas,

hence most beta particles and all alpha particles deposit their full energy in it. This
energy deposition produces enough scintillations for all but near-zero-energy beta
particles to be recorded.
The sensitive region of the Si diode typically is about 200 μm thick, compared to
the range of 20–60 μm for alpha particles with energies of 4–9 MeV, respectively.
Hence, detectors are constructed to absorb the entire energy and produce sufficient
electrons for good resolution at essentially 100% intrinsic efficiency.
The intrinsic efficiency for detecting gamma rays in the Ge detector, estimated
in terms of the attenuation coefficient, usually is less than 100% except for low
energies. Table 8.2 shows the percent efficiency in a typical 6-cm-thick detector
as a function of gamma-ray energy. The figures are based on Eq. (2.12).
The efficiency is somewhat lower for the noncollimated gamma rays emitted by a
sample because some would not traverse the entire thickness of the detector. Larger
detectors can be expected to have larger intrinsic efficiency values for energetic
gamma rays.
To optimize the geometry factor, a source is placed as close to the detector
as feasible without causing damage or contamination. A second requirement is
preparing the sample so that its diameter is appreciably less than the diameter of
the detector. For the simple case of a point source and a detector with a circular
aperture, the solid angle () can be calculated as follows:
⎛ ⎞
1⎝ d ⎠
= 1−  (8.5)
2 d +ρ
2 2
d

Here d is the distance from the point source to the detector face and ρ d is the
radius of the detector window. The solid angle approaches 2π when d approaches
0, i.e.,  approaches 0.5 for a point source near the detector surface. Such opti-
mized geometry is desirable in low-level measurements, but not necessary under
other circumstances. Some geometry factors for a circular source with radius b,
calculated as a function of d/ρ d and b/ρ d (Bland 1984), are given in Table 8.3.
Reducing the sample mass and the amount of materials between the sample and
the detector reduces radiation attenuation l. This effort is most important for alpha
particles and least important for energetic gamma rays, as discussed in Chapter 7.
The fractional self-absorption within the sample can be estimated for beta particles
and gamma rays by using Eq. (7.2). The fractional attenuation of gamma rays in
 8. Applied Radiation Measurements 139

TABLE 8.3. Geometry factors for extended source

b/ρ d d/ρ d
0.10 0.15 0.20 0.25 0.30 0.35 0.40
0.05 0.4502 0.4257 0.4018 0.3786 0.3562 0.3346 0.3141
0.10 0.4500 0.4255 0.4015 0.3783 0.3558 0.3342 0.3137
0.15 0.4498 0.4252 0.4011 0.3777 0.3552 0.3335 0.3129
0.20 0.4495 0.4247 0.4004 0.3769 0.3543 0.3325 0.3118
0.25 0.4490 0.4240 0.3996 0.3759 0.3531 0.3312 0.3104
0.30 0.4484 0.4232 0.3986 0.3747 0.3517 0.3297 0.3087
0.35 0.4478 0.4222 0.3973 0.3731 0.3499 0.3277 0.3066
0.40 0.4470 0.4210 0.3958 0.3713 0.3478 0.3254 0.3042
0.45 0.4460 0.4196 0.3940 0.3691 0.3453 0.3227 0.3013
0.50 0.4449 0.4179 0.3917 0.3665 0.3425 0.3196 0.2981
0.55 0.4434 0.4159 0.3892 0.3636 0.3392 0.3161 0.2943
0.60 0.4418 0.4135 0.3862 0.3601 0.3354 0.3120 0.2901
0.65 0.4398 0.4108 0.3826 0.3560 0.3309 0.3074 0.2854
0.70 0.4373 0.4072 0.3784 0.3513 0.3258 0.3021 0.2800
0.75 0.4344 0.4031 0.3735 0.3457 0.3199 0.2960 0.2740
0.80 0.4304 0.3977 0.3672 0.3390 0.3131 0.2892 0.2673
0.85 0.4253 0.3910 0.3596 0.3312 0.3052 0.2814 0.2598
0.90 0.4184 0.3823 0.3504 0.3218 0.2961 0.2729 0.2514
0.95 0.406 0.3706 0.3387 0.3106 0.2854 0.2648 0.2420
1.0 0.394 0.355 0.3239 0.2970 0.2729 0.255 0.2316

the sample covering, detector window, and intervening space is calculated by

Eq. (2.12). The fractional attenuation of beta particles can be measured for these
materials, or estimated from the energy of beta particles required to pass through
the material, as suggested by Fig. 2.4. Alpha particles of a given energy are entirely
absorbed by intervening materials unless the sum of the material thickness is less
than the range for the energy of the particle.
Scattering into the detector, s, depends on the environment of the source and
the detector. One subcategory is back scattering from the sample support, which
significantly increases the count rate for beta particles when the sample is placed
near the detector. It is about 20–30% of the direct radiation count rate for a ring-
and-disk source for the efficiency values shown in Fig. 7.2 with an end-window
proportional counter. Back scattering increases with beta-particle energy and the
Z value of the backing material.
Backscattering is minimal for alpha particles because they are mostly stopped
in the backing material (NCRP 1985b). A typical factor is 1.03. Backscattered
gamma rays add to the Compton continuum a broad peak at an energy of about
0.2 MeV.
Material around the source and detector, notably the detector housing, cause scat-
tering into the detector. The opportunity for scattering into the detector increases
when the source is more distant. This scattering adds a few percent to the count
rate for end-window Geiger–Mueller (G-M) detectors when the sample is 2 cm or
more distant (Zumwalt 1950), but little for gas-flow proportional counters with the
sample only about 0.3 cm from a relatively large window. Scattering, attenuation,
140 John M. Keller

and self-absorption effect in G-M counters also differ from those in end-window
proportional counters because of the typically larger sample-to-detector distances
and smaller detector window (Steinberg 1962). Scattered energetic gamma rays do
not affect the efficiency in gamma ray spectrometers because they only increase
the Compton-scattering region. Material between source and detector also scatters
direct radiation away from the detector, but this effect is small when, as is common,
the amount of intervening material is small.
Source radiation that produces relatively weak pulses is lost (to the extent of frac-
tion 1 − p) when the pulse processing system of a detector discriminates against
the low-energy pulses that constitute noise; discriminators are, however, set to min-
imize count loss. Such discrimination prevents large Ge detectors from measuring
low-energy (<30 keV or <3 keV, depending on construction) X rays, reduces the
counting efficiency of LS counters in measuring low-energy radionuclides such as
tritium, and causes a loss of <1% in proportional counters. Measurements with a
pulse generator or recording low-energy electron or X-ray pulses can identify the
energy cutoff to evaluate whether the discriminator setting is acceptable.
The factor p also refers to loss of counts at all energies if radiation arrives at
the detector more rapidly than individual pulses can be processed. The dead time
(τ ) is the detector recovery period before it is capable of recording a second event.
The dead time depends on the period and shape of the pulse and is an inherent
property of the detecting systems (e.g., gas, liquid, or solid state detectors) and
their electronic support. If the count rate is sufficiently low, the average interval
between pulses is sufficiently long that the dead time is insignificant. For higher
count rates, the following “nonparalyzable” model for correction can be applied:
Robs
Rcor = (8.6)
(1 − Robs τ )
Rcor is the corrected count rate and Robs is the observed rate. Count rates should
be limited so that the correction does not add more than about 30%.
The dead time is long in a G-M counter because of its large and wide pulses,
but short in a proportional counter. For a G-M detector with a dead time of 0.5 μs,
a measured count rate of 200 c/s implies a loss of 11% according to Eq. (8.6). In
contrast, a proportional counter with a dead time of 5 μs has a loss of only 0.1%
at the same measured count rate.
The loss in count rate can be determined by counting a set of samples with
increasing known activity and plotting the relation of count rate to activity. The
extent of deviation from a straight line at higher count rates indicates the loss.
Another technique for determining the dead time compares the count from a split
source when combined with the sum of the separately counted components (Knoll
1989).
The distinction between a second event that only adds to the pulse height of
the first event or is counted separately is a function of pulse height discrimination
and random pulse arrival time, as shown in Fig. 8.1. Further distinctive behavior,
also shown in the figure, is a second event that is counted although the first pulse
has not completely died away, i.e., the second pulse has some contribution from
 8. Applied Radiation Measurements 141

FIGURE 8.1. Resolving time (from Cember 1996, p. 350).

both events. The time required for the first pulse to die away completely is termed
the resolving time. The terms “dead time” and “resolving time” are often used
interchangeably.
The impact of such loss depends on pulse shape and pulse application. In integral
counting, all distinct pulses with height above an acceptance value are counted.
In spectral analysis, the pulse height matters. A clock records the live time for
spectrometers as the time interval to use in calculating the count rate. A pulse
that results from two radiations detected within the resolving time is known as a
“coincidence”; the coincidence is recorded at the summed energy and results in two
lost pulses at the energies corresponding to the individual radiations. Coincidences
occur randomly, but can be expected with much greater frequency when counting
radionuclides that emit two radiations simultaneously.
Radiations that are emitted simultaneously and measured for standardization
purposes (see Section 9.3.5) include two gamma rays and beta particles with
gamma rays. Occurrence of coincidences can be inferred from information in
the compilations of radionuclide decay characteristics cited in Chapter 9. The
counting efficiency ε for such coincidences is the product of the efficiency for
each radiation. In simplified terms, the fractional detection of a coincident gamma
ray relative to either of the two gamma rays is ε1 ε 2 /ε 1 = ε 2 , which is significant
only for large values of ε1 and ε 2 .
In recent years, calibration by cascade summing has become a more recognized
measurement technique with the introduction of the larger high-purity germanium
(HPGe) detectors which have a higher intrinsic efficiency. Cascade summing or
true coincidence summing results when two gamma rays are emitted from the
nucleus at nearly the same time and are then detected simultaneously at an energy
that is the sum of the two incident gamma rays. True coincidence summing is
geometry-dependent and should not be confused with the count rate-dependent
random summing mentioned above. The probability of detecting both gamma rays
142 John M. Keller

is proportional to the square of the solid angle subtended by the detector, hence
the sum peak count rate will increase as the source to detector distance decreases.
Coincidence summing can introduce the problem of misattributing summed
gamma rays as single gamma rays when counting sources close to large HPGe de-
tectors. Use of n-type HPGe detectors and Be windows can result in cascade
summing problems at lower energy X rays. Gamma rays can even sum with
Bremsstrahlung radiation from associated beta particles (Gilmore and Hemingway
1995). Cascade summing problems are observed with radionuclides commonly
used as calibration standards such as 60 Co, 134 Cs, and 152 Eu.
The decay fraction f of an emitted radiation and its half-life that are used for
measuring a radionuclide and calculating the extent of decay or ingrowth are tab-
ulated in tables of isotopes and on the Internet at the BNL nuclear data bases
(http://www.nndc.bnl.gov/ index.jsp). The decay scheme should be examined, as
discussed in Section 9.3, before selecting the decay-fraction value. When mea-
suring gamma rays, for example, the gamma-ray fraction must be used instead
of a decay fraction from an excited state that includes conversion electrons. Any
applicable conversion-electron fraction must be included when converting the beta-
particle count rate to activity. Characteristic X rays may be detected when counting
beta particles or gamma rays.
Processing a radionuclide with minimal decay between collection and
measurement—that is, D is almost 1.0—optimizes detection sensitivity and min-
imizes the uncertainty of decay correction. Some radioactive decay is inevitable
and may be useful for reducing interference from shorter-lived radionuclides.
A separation procedure with the yield Y below unity is expected in radioanalyt-
ical chemistry. For quality assurance purposes, the range of acceptable yields may
be set between 0.5 and 1.0. Lower or higher yield fractions suggest occurrence
of analytical process problems. Carrier or tracer addition to determine yield is
described in Section 6.3.
Appearance of the mass (or volume) term, m, in the denominator of Eq. (8.2)
suggests that large samples permit more sensitive measurements. Analyzing ever
larger samples is a tempting solution for attaining lower detection levels, but
larger samples require ever more complex processing and may cause reduction
in factors for the other variables in Eq. (8.3), notably those related to detector-
sample geometry, radiation attenuation, and chemical yield.

8.2.2. Radiation Background

Radiation detectors should have a low and stable radiation background to reduce
interference and uncertainty in sample measurements (see Section 10.3). Contrib-
utors to the radiation background for a detector are the following:
r Radionuclides in the detector and nearby components of the detector system
r Radioactive samples placed near the detector
r Radioactive contamination on the detector, sample holder, and other components
r Radionuclides in the counting room walls, floor, and air
 8. Applied Radiation Measurements 143

r The radiation environment near the counting room

r Other radionuclides in the source that is being counted.

Specifications in purchasing detectors and tests of the received instrument sys-

tems can limit the contamination in detector materials and associated nearby com-
ponents such as the sample holder, preamplifier, and radiation shield to acceptable
levels. The background due to cross-contamination from other samples and placing
highly radioactive sources near the detector can be prevented by careful laboratory
practices. For example, solid and liquid radionuclide sources should be enclosed
as thoroughly as is feasible when brought to the counting room.
A reasonable criterion for controlling radiation background from the cited ori-
gins is to limit the fraction that any component contributes to the total background.
Controls should be more restrictive if great efforts are being made to reduce the ra-
diation background or if the background interferes with measuring a radionuclide
of interest at its specified detection sensitivity.
Understanding, stabilizing, and minimizing the radiation background in a count-
ing system maintain confidence in observed net count rates and the calculated
detection sensitivity. Stability is important because the net count rate of a sam-
ple is calculated as the difference between two measurements—the sample gross
count rate and the background count rate—that are performed with the detec-
tor at different times. The implicit assumption is that the background has not
changed between the measurements. Fluctuations in the background count rate
have an especially important impact on the uncertainty of the measured sample
content when the background count rate is equal to or larger than the net count
rate.
Of primary importance in achieving a low and stable background is locating
the counting room in an environment of relatively low terrestrial radionuclide
background. Direct radiation from terrestrial radionuclides can be from the ground
and from structural materials such as concrete, brick, glass, and wallboard. In
addition, airborne radon and its particulate progeny emanate from the ground and
walls. Radon and its progeny are major causes of radiation background fluctuation
because their indoor concentrations depend on ambient conditions such as soil
moisture, air pressure, and inversion conditions. The progeny are sources of both
external radiation and surface contamination.
Radiation from radon and its progeny has been minimized by site selection
and controlled by sealing floor and walls to reduce emanation and by passing
ventilating air through a sorbent that removes the gas with its particulate progeny.
The air volume within the detector shield can be kept to a minimum or filled with
aged, radon-free, air. Samples are stored within containers to avoid deposition of
radon progeny immediately before counting and to await the decay of progeny
deposited earlier, in view of the short half-lives of 214 Pb (t1/2 = 27 min) and 214 Bi
(t1/2 = 20 min). Air filtration can also reduce the concentration of other particulate
airborne radionuclides, such as fission products from nuclear tests or incidents.
The counting room should be located at a distance from sources of penetrating
radiation, i.e., gamma rays and neutrons. Avoidance of sources that contribute
 144 John M. Keller

to background fluctuation by being turned off and on periodically is especially

desirable.
Shielding certain detectors contributes significantly to reducing the radiation
background. Alpha-particle detectors need no shielding because the detector walls
are sufficiently thick to stop external alpha particles. Beta-particle detector con-
tainers usually are shielded by 5-cm-thick lead bricks that are more than enough
to stop all external beta particles and also stop a large fraction of low-energy
gamma rays. Gamma-ray detectors are shielded with 15-cm-thick lead shields or
the equivalent in steel.
The low-energy components of cosmic rays can be controlled by shielding with
heavy metal (Pb, stainless steel, and in some cases Hg). The energetic (“hard”)
component of cosmic radiation is not substantially reduced by shielding with
15 cm of lead or the equivalent in other metals. Cosmic-ray components are further
reduced with thick walls of low-Z material such as water or borated polyethylene
or by placing the detection rooms far underground.
Typical radiation backgrounds for the detectors discussed here are listed in
Table 8.1 as criteria for background control, but experience suggests that the back-
ground depends strongly on detector and shielding material, as well as radiation
in the immediate environment. In gamma-ray spectral analysis, the background
decreases with increasing energy, as shown in Fig. 8.2, with peaks due to gamma
rays from radon progeny, 40 K, and annihilation radiation.

1000

511
800

600
Counts

400

352
1461
200
609

1764

0
0 500 1,000 1,500 2,000
Energy (keV)

FIGURE 8.2. Gamma ray spectral background for an HpGe detector.

 8. Applied Radiation Measurements 145

The extremely low background count rates given in Table 8.1 for two types of
alpha-particle detectors suggest that few, if any, counts will be accumulated during a
background measurement. In that case, statistical considerations in Sections 10.3.1
and 10.4.1 for calculating the uncertainty due to the background and the minimum
detectable activity do not apply. To decide whether a measured value near zero is
due to the background may depend on the presence of a recognizable peak or on
inferences based on experience concerning background fluctuation in the energy
region of interest.
Gamma rays from the thermal neutron capture of cosmic ray neutrons within the
detector and shielding materials are emitted by cadmium in graded shield liners
for gamma-ray spectrometry systems. One of the naturally occurring isotopes of
cadmium (113 Cd) has an extremely high cross section (∼20,000 barns) for thermal-
neutron capture and emits a 558-keV prompt gamma ray. Background due to such
prompt gamma rays should be significantly lower than the 511-keV annihilation
peak and the 1461-keV 40 K peak in the detector background.
The detector, its neighboring structures, and the counted source itself also con-
tribute to the background. Reducing the detector volume can decrease the back-
ground count rate, but may also reduce the counting efficiency. This interaction
can be evaluated in terms of maximizing the figure of merit ε 2 /RB , where ε is
the counting efficiency and RB is the radiation background count rate. The fig-
ure of merit also permits comparison of sensitivity of two detector systems. For
example, if 89 Sr is counted with an efficiency of 0.96 and a background of 20
c/m in an LS counter and with an efficiency of 0.40 and a background of 1 c/m
in an anti-coincidence-shielded gas proportional counter, the respective figures
of merit are 0.046 and 0.16. In this comparison, the gas proportional counter is
more sensitive when other factors, such as sample volume and counting period, are
equal.
Radioactive contaminants in filters, planchets, detectors, and shields may be
primordial radionuclides and their progeny in aluminum (e.g., thorium), lead (e.g.,
uranium progeny), and filters (40 K). Materials with low radionuclide content should
be selected from carefully screened supplies. Man-made radionuclides have con-
taminated steel and other metals during processing (from airborne fallout radionu-
clides) or reprocessing (from radioactive tracers or medical irradiation sources
such as 60 Co, 137 Cs, or 226 Ra).
A common and effective way to reduce the external background is use of an
anticoincidence system (see Figs. 8.3 and 8.4). A separate detector shields the
sample detector. When the shield counter detects a pulse, the electronics system
momentarily turns off the detector to reduce the background count rate.
Research that requires extreme gamma-ray detection sensitivity, such as neutrino
and double beta decay measurements, has led to enormous reduction in gamma-
ray spectrometer background. Contributing to this improvement are extremely low
radionuclide contents in materials for the germanium detector, container and shield,
and reduction of the cosmic-ray background by underground location, electronic
anticoincidence, muon shielding, and shield material with low photon production
in muon interactions (Heusser 2003).
146 John M. Keller

Amp

Amp Gate

FIGURE 8.3. Schematic drawing of an anticoincidence system for Compton suppression.

The radionuclide that is being counted is another source of background, both by

contaminating the detector and its immediate environment, and by emitting other
radiations that interfere with counting the radionuclide of interest.
A specific cause of potential contamination is counting radionuclides that emit
alpha particles and decay to progeny radionuclides; such sources cannot be covered,
and the progeny may leave the source due to recoil (see Section 2.2.1). One possible
means of control is to keep the recoiling radionuclide on the source mount by an
applied negative charge.
A background of sorts in gamma-ray spectral analysis is the Compton con-
tinuum (see Section 8.3.4), which may underlie characteristic photon peaks, as
shown in Fig. 2.14. The continuum is due to scattering interactions in the detec-
tor and the subsequent exit of scattered gamma rays. This situation is recognized
in the description at purchase of a germanium detector by its peak-to-Compton
ratio. Larger detectors have better, higher, ratios because the increase in mul-
tiple Compton scattering with a final photoelectric interaction results in fewer
counts in the continuum and more in the peak. A second detector surrounding
the primary detector and operated in the anticoincidence mode (see above) can
count photons scattered from the primary detector and subtract them from the
continuum.
Photoelectric interaction in materials surrounding the detector can result in
characteristic X rays in the lower energy region of the gamma-ray spectrum. For
example, the K α (72 keV) and K β (85 keV) X rays are almost always part of the
background in a spectrum of a detector shielded with lead. Commercially available
lead shields for gamma-ray-spectrometer detectors are lined with thin cadmium
and copper layers to attenuate these lead X rays.
Beta particles that pass through the can that holds the gamma-ray detector
produce another sample-related background. If the detector is sufficiently shielded
 8. Applied Radiation Measurements 147

to stop all beta particles, it will still record the lower-intensity Bremsstrahlung
produced in the shield, as described in Section 2.4.3.
The reverse—gamma-ray background in beta-particle detectors—also occurs
due to electron-producing interactions of gamma rays in the walls of the gas-filled
proportional and G-M detectors. The magnitude of the detection efficiency for
energetic gamma rays, i.e., above 0.1 MeV, typically is about 1% of that for beta
particles (Knoll 1989).
An effective but expensive technique for monitoring background radiation is
operation of one of each set of detectors in the background measurement mode
at all times to alert the operator to fluctuations due to line noise, radon in air, or
external radiation sources. Another approach is to monitor the count rates recorded
in anticoincidence detectors for identifying fluctuations in the external background.
Detector contamination between regular background tests may be inferred from a
sudden increase to unusually high sample count rates.

8.3. Detectors
8.3.1. Gas-filled Detectors
Different types of radiation detectors that use a counting gas are based on a similar
design (see Fig. 2.10). Detectors are either sealed or with a continuous flow of gas.
A high voltage applied across the volume of gas in a conducting container forms an
electrical field between two electrodes—the detector wall and a central electrode.
As radiation passes through this electrical field with enough energy (∼25–35 eV
per ion pair) to ionize the counting gas, a flow of electrons to the wire anode creates
a current pulse between the electrodes. Based upon detector design, this electrical
signal is measured as a current, accumulated charge, or pulse which can be related
to the incident radiation. Examples of operating parameters for the three basic
types of gas-filled detectors are listed in Table 8.4. As described in Section 2.4.1,
detector design and operating voltage determine detection performance.
The ionization chamber generally integrates the current over a brief period and
measures it with a sensitive electrometer. An early version was an electroscope

TABLE 8.4. Characteristics of common gaseous detectors

Ionization chamber Proportional counter G-M counter
Gas Air or other gas Argon Helium or argon
Quench gas — Methane (10%) Halogen organic (3%)
Dimensions (cm) — 5.7 dia. × 1.0 3.5 dia. × 10
Window diameter (cm) — 5.7 3.0
Window area density — 0.08 or 0.5 0.50 (β) or shield (γ )
(mg/cm2 )
Applied volts approx. 100 800 (α) or 1600 (α + β) 1000–2000
Amplification Yes Pre-amp + linear amp No
Measurement Electrometer Pulse count Pulse count
148 John M. Keller

charged by friction, in which the current was measured by the rate of separation of
two light metal foils that was a measure of their accumulation of electric charge.
The relatively large pulses from alpha particles are measured with a Frisch grid
chamber (Knoll 1989) with a resolution of about 40 keV. Its advantage is the
capability to measure and perform spectral analysis for samples with relatively
large areas, e.g., 500 cm2 .
A pressurized and sealed ionization chamber is used as calibrator for individual
radionuclides that emit gamma rays. Their activity is measured in terms of the
radiation dose, as discussed in Section 8.3.5.
The G-M counter is a simple and relatively inexpensive gas-filled tube with
a count-rate meter; an amplifier may also be present. The G-M detector counts
alpha particles, beta particles, and gamma rays with a very thin window, counts
beta particles and gamma rays with a thicker window, and counts gamma rays only
with a thick shield. Alpha and beta particles interact in the gas; gamma rays interact
mostly in the walls, from which electrons enter the gas. The intrinsic efficiency for
counting gamma rays relative to beta particles depends on the amount and type of
solids surrounding the detection gas.
The fill gas differs among types of detector. Almost any gas is suitable for an
ionization chamber. The counting gas for a proportional counter must have an
added organic quenching agent. The most commonly used gas is a mixture of
argon fill gas (90%) and methane quench gas (10%) readily available as P-10 gas.
In the G-M counter, if the tubes are sealed, the quenching agent usually is a small
amount of Br2 or Cl2 ; if the detectors are flow type, the quenching agent may be
organic compounds such as ethyl alcohol and ethyl formate. The fill gas is usually
a noble gas; for example, Q-gas is 97% He and 3% an organic compound.
The most common detector is the gas-flow proportional counter, although ion-
ization chambers and G-M counters can be usefully applied in the laboratory.
Several types of gas-flow proportional counters are used in modern counting
laboratories.
The first type is a manual detector, either with a thin (typically 80 μg/cm2 )
window beneath which the sample is placed, or without a window but with a
sample tray that slides under the detector and becomes the detector bottom. Use of
the internal proportional counter eliminates external attenuation of low-energy beta
particles but introducing the source into the detector can cause contamination or
reduce the potential difference to a value below the applied voltage. The detection
volume may be cylindrical (typically about 1 cm high and 3 cm in radius), or may
be a hemisphere in the classical 2πgroportional counter. For either a thin window
or no window, continuous gas flow is maintained. A thicker window (typically 0.5
mg/cm2 ) can seal the counting gas in the detector.
The second commonly available system is a detector coupled to an automatic
sample changer. In this arrangement, planchets or disks from a stack of 30 or
50 are fed individually by tray to the location beneath the detector window. The
controlling computer sets the counting time of typically 10 to 100 min, moves the
tray, and records the time and accumulated counts. The computer also controls
whether alpha or beta particles or both are measured and whether samples are to
 8. Applied Radiation Measurements 149

5.70 cm.
Window and detector diameter

Guard detector
80 m g cm–2 Window
Depth of retaining
detector
window Sample detector ring Window to
carrier
distance
0.97 cm.
0.32 cm.
Slide Assembly

Measurements +– 0.005 cm 1.65 cm.

(2 standard deviations) Source diameter

Horizontal view

Mylar SrCO3 Filter

Plastic Plastic
spacer spacer

Air
Plastic Plastic disk Stainless
ring steel pan
Expanded source mount

FIGURE 8.4. Configuration of the source and detector for the Tennelec LB5100, an end-
window, gas-flow, anticoincidence proportional counter (from Nichols 2006).

be counted repeatedly. A typical pancake-shaped detector is shown in Fig. 8.4.

It includes a guard detector connected in anticoincidence to reduce the external
radiation background. The sample is inserted immediately beneath the detector. A
ring-and-disk sample is shown, but a planchet may be substituted by removing the
spacer.
A third, usually more expensive, system has multiple detectors to count multiple
samples simultaneously. Configurations available commercially are modular with
groups of four detectors arranged over a slide-out drawer. Modules can be added
for 8, 16, or even 64 detectors. This counting system is useful for low-level samples
that require long counting time (i.e., 24–96 h). A high sample load can justify the
cost.
Two internal counters facing each other and electronically combined to produce
single pulses constitute a 4π detector. A thin source deposited on a thin backing is
placed between the two detectors. Measurements at near 100% counting efficiency
approach the absolute disintegration rate of a radionuclide that emits beta particles.
A gas-flow proportional counter is operated at a low bias voltage (∼600–800 V
with P-10 gas) if alpha particles are to be detected in the “alpha only” mode. By
increasing the bias voltage to ∼1500 V with P-10 gas, the counter will respond
to both alpha and beta particles. Operation at this higher voltage is referred to as
“simultaneous” or “α + β” mode. These voltages can be decreased by increasing
150 John M. Keller

the linear amplifier gain and vice versa. The optimal operating voltage for each
mode is selected after finding the plateau by plotting count rate vs. bias voltage with
an alpha-particle and a beta-particle source (see Section 2.4.1). The pulses require a
preamplifier and an amplifier for counting. Pulses are proportional to energy depo-
sition, hence are hundreds of times larger for alpha particles than for beta particles.
Plateau curves are generated with a pure alpha-particle emitter (238 Pu, 244 Cm)
and a pure beta-particle emitter (63 Ni, 90 Sr/90 Y, 99 Tc) unless a combination of
radionuclides is selected to represent an actual sample. The analyst chooses, on
the basis of the two curves, a bias voltage for “alpha only” mode that is below the
point where beta particles just begin to be detected (see Fig. 2.11). At this voltage,
and also by application of pulse-shape discrimination, beta-particle “cross talk” is
minimized. The voltage for α + β counting is selected near the midpoint of the
beta-particle plateau. This plateau is not quite horizontal because an increase in
the applied voltage adds a few counts that previously had just been excluded by
the lower-energy discriminator.
The next step is to determine the counting efficiency for alpha- and beta-particle
measurements. The alpha-particle counting efficiency for thin samples with a pro-
portional counter is almost constant over the energy range of 4–9 MeV. This
energy range includes the naturally occurring isotopes of thorium, uranium and
their progeny, and also the actinides in global fallout from nuclear weapon test-
ing. A few naturally occurring rare earth elements (e.g., 147 Sm, 144 Nd, and 152 Gd)
emit alpha particles in the 2–3 MeV range, but these radionuclides have very low
decay rates associated with their long half-lives and are not of interest at most
radioanalytical chemistry laboratories.
Alpha particles that enter the gas proportional detector deposit energies of the
order of 1 MeV, far more energy than is needed for detection, hence alpha-particle
counting efficiency is a function of the source–detector geometry minus losses from
alpha-particle attenuation in the sample mass, intervening air, and detector window.
The ratio of count rate to disintegration rate (cpm/dpm) for alpha particles from a
very thin source mounted on stainless steel and counted within a windowless 2π
gas-flow proportional counter is ∼51.5%. The additional 1.5% over the expected
50% from the 2π geometry is due to backscattering of the alpha particles from a
stainless steel source mount.
End-window proportional counters with thinwindows that separate sample from
detector have the alpha-particle counting efficiency shown in Table 8.1 for thin
samples on planchets. The lesser counter efficiency relative to the internal detector
is due to less than 2π geometry and attenuation in sample, air space, and window.
Beta particles deposit typically about 2 keV of energy, in the range of 0.5–10
keV, in the gas proportional counter. Pulse height discrimination readily distin-
guishes these beta-particle pulses from the much more energetic ones generated
by alpha particles to provide a “beta only” mode. Cross-talk with alpha particles
can occur because the various angles at which the particles enter the detector
lead to considerably lower energy deposition for a small fraction of the particles.
Beta particles emitted with near-zero energy are not detected when their energy is
deposited in the sample, air, and detector window.
 8. Applied Radiation Measurements 151

Unlike alpha particles, a significant fraction of beta particles that enter the
detector has been backscattered. Due to the energy distribution of beta particles,
the overall beta-particle counting efficiency for a proportional counter depends on
the fraction of low-energy beta particles absorbed in the sample, air, and window.
To quantify the activity for 63 Ni (66.9 keV) or 99 Tc (292 keV), for example, a
separate standard is needed for each radionuclide. The beta counting efficiency is
in the 65–75% range for radionuclides with maximum energies of 0.3–3.5 MeV
on steel planchets in a windowless gas-flow proportional counter. The efficiency is
less in an end-window counter (see Table 8.1), and can reach zero for beta-particle
groups with maximum energy below about 30 keV.
Other versions of the gas proportional counter system permit measurements
of special sources or radionuclides. Radioactive gas samples can be counted in
a detector that has valves for drawing a vacuum or admitting gases. A measured
volume of the gas sample is mixed with counter gas and counted in the detec-
tor (ICRU 1972). Alternatively, a flowing gas stream can be mixed with flowing
counting gas. Radioactive liquid samples can be placed in a thin-walled container
beneath an end-window counter, or a flowing stream can be passed through this
container.
Modified versions of the gas proportional counter are used for spectral analysis
of X rays in the energy region of a few kiloelectron volts. Detectors are constructed
with a greater depth and a window with a low-Z material such as Be, and with a
high-Z counting gas such as Xe at a pressure of several atmospheres. The resolution
is about 0.6 keV at 5 keV (Knoll 1989).

8.3.2. Liquid Scintillation Counters

Counting with LS systems is the most widely used technique for measuring ra-
dionuclides that emit only low-energy beta particles. LS counting can be used for
measuring any radionuclide in any liquid sample, but it is the preferred method
for low-energy beta emitters, especially 3 H, and volatile radionuclides, e.g., 99 Tc
as TcO−4 , that could be lost during evaporation or drying before counting in a
gas-flow proportional detector. Examples of radionuclides routinely measured by
LS are listed in Table 8.5.

TABLE 8.5. Low energy and potentially volatile

b-particle-emitting radionuclides
Radionuclide Emax (keV) t1/2 Volatile Forms
3H 18.6 12.3 years H2 O, H2
14 C 156.5 5730 years CO, CH4 , CO2
32 P 1709.0 14.3 days —
35 S 167.4 87.2 days SO3 , H2 S
36 Cl 709.0 301,000 years Cl2 , HCl
45 Ca 258.0 163 days —
63 Ni 66.9 100 years NiCl2 , Ni(NO3 )2
99 Tc 292.0 213,000 years TcO− 4
152 John M. Keller

Traditional LS applications in radioanalytical chemistry are minor compared

to the wide use of LS techniques in the life sciences (biological and medical) for
measuring labeled proteins, genetic materials (RNA and DNA), and other types of
biological tracer work. Even nonvolatile forms of 32 P and 45 Ca that are frequently
used to label biological and medical samples are counted by LS, as indicated by
Table 8.5. Cherenkov radiation from 32 P beta particles may be counted directly
in water or other biological fluids without adding scintillation cocktail. The LS
technique is also useful in counting radionuclides smear samples collected for
health physics (radiation protection) applications.
The basic design of a commercial LS counter was developed more than fifty
years ago. A charged particle that moves through the mix of sample and scintillation
cocktail transfers energy in ∼5 nanoseconds with a yield of ∼10 photons per
keV to generate many photons at energies of a few electron volts that distribute
isotropically. The set of simultaneously generated photons per charged particle
creates a pulse in each of the two photomultiplier tubes (PMTs) that face the
sample vial. The signal from each PMT is fed into a coincidence circuit which
produces an output pulse only when the two PMT signals occur within the resolving
time of the circuit (∼20 nanoseconds). The coincidence pulse drives a gated circuit
that passes only a valid signal and excludes randomly produced electronic noise
from the PMTs.
A significant improvement to LS instrumentation was summing the pulses from
each PMT for an output proportional to the total intensity of the scintillation
event. This design for an LS system with pulse coincidence detection and summa-
tion (Fig. 8.5) has remained basically unchanged over the last forty years. More
recently, instrument vendors have replaced a traditional three-channel system with
a multichannel analyzer (MCA) and software for spectral analysis.
The spectrometer permits confirmation of the observed radionuclide by its spec-
trum, within the limitations of spectral distortion by energy losses and the radiation
background associated with the measurement. It also permits setting the lower and
upper energy discriminators for optimum ε 2 /RB values for each radionuclide of
interest. The energy ranges beyond the region of interest usually are measured
at the same time to confirm the absence of contaminant radionuclides. Two or
three radionuclides may be detected simultaneously by such discriminator control

PMT 1

Sample Coincidence Sum Amp ADC MCA

PMT 2

FIGURE 8.5. Functional diagram of a modern liquid scintillation counter.

 8. Applied Radiation Measurements 153

if their maximum energies are considerably different so that their beta-particle

groups can be clearly distinguished.
Most commercially available LS systems have automatic sample changers to
measure multiple sources. A computer controls the system to specify counting
schedule, counting period, energy regions, and data recording, calculations, and
storage. Concentrations for specified radionuclides, the counting uncertainty, and
corrections for luminescence and quenching are calculated as discussed below.
For LS counting, a radioactive sample is dispersed in a scintillation counting
solution referred to as a scintillation cocktail (see Section 2.4.2). In the radioan-
alytical chemistry laboratory, typically 10 ml of aqueous sample is shaken with
10 ml of cocktail in a 20-ml glass or plastic counting vial. The solvent absorbs
the energy deposited by the charged particle and transfers it to the scintillator
compound. Additional considerations for the solvent are high transmission of the
photons emitted by scintillator compounds and a low 14 C content (to ensure low
background for counting).
Solvent molecules that can effectively transfer the nuclear radiation energy have
a high density of π electrons; they include aromatic compounds such as toluene,
xylene, and pseudocumene (1,2,4-trimethylbenzene). These solvents were used for
many years, but now have been replaced in most commercial scintillation cock-
tails with more environmentally friendly solvents such as diisopropylnaphthalene
(DIN), dodecylbenzene (LAB), and phenylxylylethane (PXE).
The primary scintillator compound becomes excited (electrons are elevated to a
higher energy state) when the energy deposited by radiation is transferred from the
solvent. The excited scintillator compound promptly returns to the ground state
while releasing energy in the form of a photon that can be detected by a PMT. One
of the more widely used scintillator compounds is PPO (2,5-diphenyloxazole).
Characteristics that define an effective scintillator compound include a fluores-
cence wavelength (the energy of the emitted photon) that is effectively detected
by the PMT, a brief decay time (interval for return to ground state), and a good
quantum yield (number of photons emitted per number of molecules excited).
The peak wavelength emitted by the primary scintillator should be within the
sensitive range (300–425 nm, or 4–3 eV) of the PMT used in the LS system.
Some older scintillation cocktail recipes used a secondary scintillator to shift the
initial photon emitted to a slightly longer wavelength (400–420 nm range). Most
modern LS systems do not require this secondary scintillator because new PMTs
are sensitive to the photons emitted by the primary scintillator.
Because the solvents used for most scintillation cocktails are not good solvents
for aqueous samples or ionic species, detergents or surfactants are included in the
cocktail to help emulsify aqueous samples. The formulation of a scintillation cock-
tail can be complex, but most laboratories purchase prepared scintillation cocktails.
The counting efficiency for charged particles with an LS counter can approach
4π geometry, as shown in Table 8.1. It decreases to about 25% for 3 H because of
low beta-particle energies.
Counting efficiency for LS is reduced or degraded by several factors, notably
chemical and color quenching. Chemical quenching occurs when an impurity in the
154 John M. Keller

(dN/dE )

Unquenched

Emax
Quenched

Beta energy (keV)

FIGURE 8.6. Example of quenching in a beta-particle energy deposition spectrum in an LS

counter.

sample or scintillation cocktail interferes with energy transfer between the solvent
and the scintillator. Color quenching is the attenuation of scintillator photons in the
solution due to light absorption. The difference between the two types of quenching
lies in the type of energy absorbed; chemical quenching absorbs deposited charged-
particle energy while color quenching absorbs emitted-photon energy.
Loss of energy, i.e., a reduction in the pulse height recorded at the PMT, is the
result in both cases as quenching shifts the energy spectrum to lower energies (see
Fig. 8.6). Any pulses shifted to below the low-energy discrimination are lost and
reduce the counting efficiency. Factors that can influence the degree of quenching
include the sample matrix, sample preparation, and choice of scintillation cocktails.
Several techniques have been developed to correct the effect of quenching.
The traditional method for quench correction is to add an internal standard to
each sample to determine the counting efficiency for each sample matrix. This
method continues to be one of the most accurate but is labor intensive and expen-
sive. A set of samples is counted without the internal standard. A duplicate set,
with a small volume (e.g., 0.1 ml or less), of known activity of the internal stan-
dard spiking solution is added to each sample, and is then counted. The counting
efficiency for each sample is as follows:

Efficiency = (cpms+i − cpms )/dpmi (8.7)

The values of cpms and cpms+i are the net count rates for each sample without
and with the added internal standard, respectively, and dpmi is the activity of
the internal standard added. The activity, dpms , for each sample is calculated by
dividing the cpms for each sample by the efficiency for each sample. This method
 8. Applied Radiation Measurements 155

assumes that the chemical form of the radionuclide added as the internal standard
is the same as in the sample and that the added internal standard spiking solution
does not affect sample quenching.
Instrument vendors developed various techniques to measure the degree of
quenching in each sample with quench correction curves. The quench correc-
tion curve is prepared from a set of increasingly quenched standards with the same
level of activity. Then a plot of efficiency vs. the degree of quenching is prepared
for each radionuclide of interest. The success of quench correction curves depends
on skill in identifying the quenching agent and measuring the degree of quenching,
which is typically represented by a quench indicating parameter (QIP). Techniques
developed to define various QIPs are discussed in LS instrument manuals, vendor
literature, and other LS application descriptions (Horrocks 1974). The main types
of QIP used for quench correction curves are either based on the sample spectrum
or an external standard spectrum. The QIPs based on sample spectra are derived
from mathematical transformations of spectral data that provide a measurement
related to either the average or maximum spectral energy.
An external standard spectrum to determine the QIP is popular with users and
instrument vendors. The external gamma-ray source (e.g., 137 Cs, 133 Ba, or 152 Eu)
that is part of the detector system induces a Compton-electron spectrum in the
scintillation cocktail. Each sample is automatically counted with and without the
external standard. The Compton-electron spectrum produced in each sample vial
is applied with mathematical techniques to derive a QIP for a quench correction
curve.
One method offered by instrument vendors to determine the activity of a radionu-
clide is efficiency tracing (ET) (Takiue and Ishikawa 1978) for most beta-particle
radionuclides except tritium. The ET method uses a single unquenched 14 C stan-
dard to calibrate the LS system by collecting the spectrum and determining the
efficiency in several energy regions. Then, an unknown sample is counted in the
same energy regions. The cpm measured for the unknown in each counting region
is plotted vs. the efficiency determined for the 14 C standard in the same regions.
The generated curve is extrapolated to 100% efficiency to obtain the dpm for the
unknown activity.
As an example, the information in Table 8.6 summarizes the data collected for
a 14 C standard (134,900 dpm) in six energy regions of the beta-particle spectrum.
The table also includes the data from a simulated unknown sample with 90 Sr and
90
Y in secular equilibrium with a total beta-particle-activity of 1037 dpm.

TABLE 8.6. Counting data collected to demonstrate ET method.

Energy Region (keV) 14 C (cpm) 14 C Efficiency (%) 90 Sr + 90 Y (cpm)
100–2000 13,550 10.04 777
80–2000 26,788 19.86 825
60–2000 45,897 34.02 877
40–2000 71,228 52.80 918
20–2000 100,951 74.83 971
0–2000 130,157 96.48 1032
156 John M. Keller

1050

1000
y = 2.8057x + 765.31
950
Y (CPM)

900
90
Sr +
90

850

800

750
0 10 20 30 40 50 60 70 80 90 100 110
14
C Efficiency

FIGURE 8.7. Example of efficiency tracing for direct measurement of activity.

The cpm of the “unknown” is plotted vs. the 14 C efficiency from each energy
region as shown in Fig. 8.7. The equation in Fig. 8.7 is the best fit to the data as
determined by linear regression. This linear equation is then used to extrapolate
the count rate of the “unknown” at 100% 14 C efficiency, which is equivalent to the
activity (dpm) for the 90 Sr + 90 Y in the sample. The calculated activity for this
example of 1045 dpm is within −0.8% of the known value.
Chemiluminescence and photoluminescence are other forms of interference
that can reduce the accuracy of LS techniques. “Chemiluminescence” describes
the emission within the scintillation cocktail of photons that result from a chemical
reaction; common initiators are samples with an alkaline pH or the presence of
peroxides. Photoluminescence can occur when the scintillation cocktail is exposed
to ultraviolet light. Some substances in the cocktail, notably the scintillator, are
excited and then emit light when the species return to ground state. The effect of
photoluminescence is reduced in LS systems by decay when the sample train is
held in a dark environment for a few minutes prior to counting. On the other hand,
chemiluminescence may have a slow decay rate that requires a change in sample
preparation to eliminate the chemical that causes it. Some LS systems identify
luminescence by pulse shape and indicate its relative extent.

8.3.3. Alpha-Particle Spectrometers

Most modern alpha-particle spectral analysis is performed with a semiconductor
such as a surface barrier detector with a very thin dead region in front of the active
 8. Applied Radiation Measurements 157

Charged particle
Preamp Amplifier ADC MCA
detector

Bias
Supply

FIGURE 8.8. Block diagram of an alpha-particle spectrometer system.

(depletion) zone of the detector. Only a very thin dead region is acceptable because
of the low penetrating power of alpha particles. The sample is counted in a small
vacuum chamber that provides a nearly air-free path between the sample support
plate and the detector. Without the vacuum, the alpha-particle spectrum is degraded
by interactions with air molecules.
Commercially available systems integrate in a single unit the high-voltage power
supply, amplifier, analog-to-digital converter (ADC) and multichannel analyzer
with the counting chamber, detector, and preamplifier, as shown in Fig. 8.8.
In general, larger detectors have worse resolution. A good compromise for most
laboratory applications is a Si diode detector with an active area of 3–5 cm2 , a
depletion zone of about 0.02 cm, and a resolution of 15–20 keV. This resolution
is sufficient for baseline separation of 239 Pu (5.15 MeV) from 238 Pu (5.50 MeV)
despite complications due to multiple alpha-particle groups per radionuclide. The
counting efficiency for surface barrier detectors depends on the geometrical factor
and can be expected to be constant at fixed geometry for very thin sources that
emit alpha particles in the typical energy range of 4–9 MeV. As described Section
7.2.3, preparation of a uniform and very thin source, usually by electrodeposition,
is essential for obtaining good peak resolution. This requirement may be relaxed
to the extent that low-energy tailing at a peak does not interfere with calculating
the activity, identifying the peak energy for the alpha particle, or measuring lower-
energy alpha particles.
Spectrometer energy is calibrated for these energies with available thin sources
of alpha particles at energies known within 0.1 keV. The energy peaks in an
acceptable system are almost symmetrical. Many radionuclides emit additional
alpha-particle groups at energies slightly lower by 30–50 keV than the main char-
acteristic peak. These are not clearly resolved from the characteristic peak, but do
not interfere seriously with calibration or detection because they are at much lesser
intensity.
Detectors within this range of dimension and resolution are suitable for most en-
vironmental radioanalytical chemistry applications. The alpha-particle emitters of
interest include isotopes of thorium, uranium, neptunium, plutonium, americium,
and sometimes curium (244 Cm at 5.80 MeV). Most of the major alpha particles
on this list can be resolved. Exceptions are 233 U from 234 U and 239 Pu from 240 Pu,
which require mass spectrometric separation to resolve them (see Table 6.3). A
158 John M. Keller

pair that can be distinguished only after chemical separation is 238 Pu and 241 Am
at 5.50 MeV.
Computation of activity is simplified because the same counting efficiency ap-
plies to all alpha particles in the usual energy range. The activity of a radionuclide
is calculated simply from the activity of the added tracer multiplied by the net
accumulated counts for the peak of the radionuclide of interest and divided by the
net counts for the tracer peak. Separate values of the counting efficiency and the
yield are not needed, although they may be of interest to monitor tracer activity
and process yield, respectively.
Alpha particles are also measured by spectral analysis in Frisch grid cham-
bers (see Section 8.3.1) and specially designed LS counting systems (see Section
7.3.3). Although solid-state detectors are most commonly used, the other detection
systems, if available, can simplify the processing of certain samples.

8.3.4. Gamma-Ray Spectrometers

That nuclear radiation stimulates light emission in certain substances was discov-
ered at the beginning of the nuclear age. A flash of light soon was associated with
passage of single radiation. Some materials were found to be more sensitive and
in some forms to be particularly responsive to certain radiations. As technology
improved, these scintillation materials were coupled to a PMT to record the flashes
of light. The PMT detects the photons that constitute the flash of light from the
scintillator and converts each set of photons to an electron at the photocathode.
The number of electrons is then amplified for a factor of 107 –1010 by being ac-
celerated through a series of dynodes. Commercial PMTs are available that have
appropriate dimensions, provide suitable amplification, optimize responses at a
range of photon energy, and minimize thermionic noise.
The most common scintillation crystals used in these systems were sodium
iodide doped with thallium, referred to as NaI(Tl) detectors (see Section 2.4.3).
The advantages of the NaI(Tl) crystal for gamma-ray detection are high detection
efficiency and ability to machine the detectors in various dimensions and shapes. A
common configuration is a cylinder sealed in an aluminum can, where the sample
is placed on one end while the other end is coupled by light pipe to a PMT.
Some cylinders are manufactured with a central well into which a small sample
may be placed for better counting efficiency. Cylindrical crystal dimensions are
reported by diameter and height; the 3 × 3 crystal used as comparison standard
for reporting relative counting efficiency in terms of 60 Co has a diameter of 7.6 cm
and a height of 7.6 cm.
In the late 1940s, the pulses generated in a PMT were amplified and sorted by
pulse height with a single-channel analyzer that was moved incrementally through
the entire gamma-ray energy region. This analysis depicted a spectrum such as the
one in Fig. 2.14 for each radionuclide that emits a single gamma ray, or a multiple
gamma-ray spectrum for two or more gamma rays. In time, ease of analysis was
improved, first by placing 10 or 20 channels in parallel and ultimately with systems
that had 1024 channels or more, by multiples of 2x .
 8. Applied Radiation Measurements 159

Lithium-drifted semiconductor detectors were the next generation of detectors

for gamma-ray spectrometers. Based on lithium-drifted germanium and silicon,
they were designated as Ge(Li) and Si(Li), respectively. The Ge(Li) detector was
applied to the entire gamma-ray energy range while the Si(Li) detectors were
applied for low-energy gamma rays and X rays. The Si(Li) detectors, because
of their lower Z value and smaller size, were more suitable for lower gamma-
ray energies and continue to be good choices for X-ray spectrometers in modern
counting laboratories.
The lithium drifted detectors provided greater resolution but at a higher cost
and lesser counting efficiency relative to the NaI(Tl) detectors. Additionally, the
Ge(Li) detector must be kept at liquid nitrogen temperature at all times to prevent
lithium ions from drifting in the germanium to destroy the detector.
The intrinsic HPGe detector now has replaced the Ge(Li) detector. This detector
is made of sufficiently pure germanium that does not require lithium drifting for its
radiation-sensitive properties. The HPGe is not subject to adverse effects if allowed
to come to room temperature, but must be cooled to liquid-nitrogen temperature
during use to achieve good resolution.
The most common type of detector in current use for gamma-ray spectrometry
is the coaxial HPGe detector in an aluminum can. These detectors are available
with either a p-type or n-type crystal. The efficiency of a p-type detector typically
decreases quickly below ∼100 keV whereas the n-type detectors can be used at
energies as low as 3 keV with a beryllium window.
Selection of detector type and size should be controlled by its application, both
immediate and long-term. In addition to cost, important performance characteris-
tics of a semiconductor detector for gamma-ray spectrometry are the resolution,
efficiency and peak-to-Compton ratio.
Resolution (see Section 2.5.4) is reported as the peak full width at half maximum
height (FWHM) for solid state detectors, and as FWHM divided by the peak
centroid energy for NaI(Tl) detectors. The FWHM increases slowly with peak
energy.
A standard method is described in IEEE standard test procedures (IEEE 1996) to
define the relative detection efficiency for a coaxial detector. The relative efficiency
εR for a coaxial detector is defined as follows:
εHPGe
εR = (8.8)
εNaI(Tl)
In Eq. (8.8),

εHPGe = absolute efficiency at 1332 keV for a coaxial detector measured

25 cm from the source.
εNaI(Tl) = absolute efficiency at 1332 keV for a 3 × 3 NaI(Tl) scintillation
detector measured 25 cm from the source.
The absolute efficiency in this calculation is the ratio of the net number of counts
measured in the 1332-keV full-energy peak divided by the number of gamma rays
emitted by the 60 Co source at this energy during the same time interval.
160 John M. Keller

TABLE 8.7. Resolution vs. efficiency

NaI(Tl) (30 mm × 30 mm) HPGe (∼30% Relative Eff.)

Energy (keV) εabs (%) FWHM (keV) ε abs (%) FWHM (keV)
60 (241Am) 50 8 22.4 1.2
(14%)
662 (137 Cs) 32 50 5.0 1.5
(7.5%)
1332 (60 Co) 25 95 2.5 1.9
(7.1%)

Application of this relation is shown in Table 8.7 for comparing the efficiency
and the resolution of two types of detectors with different dimensions. The com-
parison has been extended to three photon energies to demonstrate that comparison
at 1332 keV may not be sufficient for the purposes of the user because both rel-
ative efficiency and resolution change with energy. The comparison for practical
application also depends on sample dimensions and source-to-detector distance.
Table 8.7 data support the observations of better resolution but worse efficiency in
the HPGe detector than the NaI(Tl) detector.
The peak-to-Compton ratio is an important indication of detector performance
for measuring lower-energy gamma rays in the presence of higher energies. The
Compton continuum that is from 0 to ∼200 keV below the full-energy peak (see
Eq. 2.16) of 60 Co increases the spectral background across the entire low energy
region below the Compton edges. Most detector manufacturers report the peak-to-
Compton ratio as the full-energy peak count rate divided by the net count rate for
the same number of channels in the middle of the Compton continuum. This ratio
typically ranges from 40 to 70 for 60 Co measured with coaxial HPGe detectors
currently produced. The peak-to-Compton ratio is higher for detectors with better
efficiency and resolution.
The spectral response of a detector is more complex than described in Section
2.4.4 because of the bulk of the detector. The observed Compton continuum con-
sists of single plus multiple successive scattering interactions. When such multiple
Compton scattering interactions are terminated by a photoelectric interaction, the
pulse is added to the full-energy peak. Most of the counts in a full-energy peak
for gamma rays above 100 keV are due to such multiple scattering plus a final
photoelectric interaction.
Photoelectric interactions are recorded at the full energy peak despite the binding
energy loss (see Eq. 2.17) because the emitted X ray that follows electron emission
is detected simultaneously with the gamma ray. In small detectors, where the X
ray has a high probability of escape, a small second (“escape”) peak is seen at
the energy predicted by Eq. (2.17), which is below the gamma-ray energy by 28
keV in the NaI(Tl) detector and by 10 keV in the Ge detector. The dual peaks are
readily apparent at low gamma-ray energies.
Interaction by pair production results in a spectrum that includes escape peaks
at the full energy minus 511 keV and minus 1022 keV, when either one or both
of the positron annihilation photons do not interact with the detector. Compton
scattering of these photons adds to the continuum.
 8. Applied Radiation Measurements 161

76 mm
IR window
1.3 mm 5 mm

Ge crystal
22.5

Lithium contact
(1.0 mm Ge equivalent)

8 mm End cap (AI)

Crystal holder (Cu)

70 mm

1.6 mm

Core hole
(length:45.5 mm)

7.5 mm 0.75 mm
15.4 mm

25.4 mm 3 mm

25.4 mm

FIGURE 8.9. Detailed description of a Ge detector (from Wang et al. 2002).

The previously cited sum peaks occur for two or more coincident gamma rays,
for example, at 2505 keV for 60 Co. Interactions outside the detector commonly
are detected as a peak at 511 keV due to annihilation radiation and at about 200
keV due to Compton scattering at 180◦ . Gamma rays produced by cosmic-ray
interactions in or near the detector are observed as discussed in Section 8.2.2.
All of these interactions by gamma rays with the detector can be modeled by a
gamma-ray simulation program such as the previously cited Monte Carlo n-particle
code, version 4. Modeling requires precise information on the location and material
of the source, detector, and surroundings. A description of the detector such as that
shown in Fig. 8.9 must be obtained from the supplier because the detector container
is sealed.
Selection of the type of detector for gamma-ray counting should be informed by
the analyst’s intention for the measurement. For example, NaI(Tl) detectors have
excellent efficiency and are useful for counting low-activity samples with few
radionuclides that emit gamma rays. Germanium detectors have superior energy
resolution, as shown in Fig. 2.14, for identifying numerous radionuclides that
emit gamma rays in a sample. Both systems must be evaluated in terms of their
background characteristics. Sodium iodide crystals of the same size will have
higher backgrounds inherently in spectral analysis because the peaks cover a wider
energy range; larger detectors have higher backgrounds because of their size.
Current efforts to select better detectors for spectral analysis have identi-
fied several materials, notably cerium-doped lanthanum chloride or bromide for
162 John M. Keller

scintillation systems and cadmium zinc telluride for solid-state systems. The en-
ergy resolution of the new crystalline scintillators is better than two-fold that of
thallium-doped sodium iodide and the detectors are less responsive to temperature
change. One drawback is the 138 La radionuclide that constitutes 0.09% of lan-
thanum in nature; moreover, thorough purification from thorium is necessary. The
new solid-state material has much better counting efficiency (due to its higher Z)
than the same volume of germanium and can be used at room temperature, but
resolution comparable to germanium has not yet been achieved.

8.3.5. Other Radiation Detectors

Some other detectors in common use either measure radiation in terms of exposure
or by tracks. Among the former are ion chambers, dosimeters of various kinds,
and photographic film. Tracks can be observed in other types of photographic film,
cloud chambers, and track-etch films.
Radiation exposure is defined as the energy deposited by radiation per unit
mass. For a gamma ray emitted by a point source, the radiation exposure rate i is
related to the gamma-ray decay fraction f , the gamma-ray energy E (in MeV),
the activity of the radionuclide a(in Bq), the linear energy absorption coefficient
μa , the density of air ρ, and the distance x:
(4.7 × 10−15 ) f Eaμa
I = (8.9)
4πρ x 2
If a radionuclide emits several gamma rays, their values on the right side in Eq.
(8.9) are summed. The unit of exposure rate is coulomb/kg/s. Typical exposure
rates are in microcoulomb/kg/s (3881 Roentgen equals 1 coulomb/kg− ). Thus,
the activity of a gamma ray can be calculated from its decay characteristics f
and E and a measurement of the exposure rate. Exposure rates are read directly
with ionization chambers and exposures, as light emission with thermoluminescent
dosimeters and as darkening of developed photographic film. These readings are
calibrated by exposure to a radiation standard. A common laboratory application
is measurement of single radionuclides that emit gamma rays by insertion into a
calibrated ionization chamber (Knoll 1989).
A sensitive measurement of alpha-particle tracks consists of placing a very weak
source on suitable photographic emulsion for a long period and then developing
the film and counting the number of tracks (Knoll 1989). Although track length
depends on the angle at which the alpha particle penetrated the film, tracks of alpha
particles that almost parallel the surface can indicate the alpha-particle energy.
Tracks of alpha particles or heavier ions can also be counted by etching material,
such as certain plastics, with a basic solution because the tracks dissolve faster than
the nonirradiated material. These tracks can be viewed with a microscope or can
be enlarged by an electrical discharge to view by the lesser magnification of a card
reader (Knoll 1989). Cloud chambers provide visual observation of alpha particles
and electrons that pass through a gas saturated with a vapor that condenses on the
ions formed along their tracks (Ehmann and Vance 1991).
9
Radionuclide Identification
BERND KAHN

9.1. Introduction
Systematic identification of potential radionuclides of interest usually is considered
at the beginning of a project, with further needs developing if results are questioned
in the course of the project. In practice, one identifies a radionuclide by finding
a match of measured decay characteristics to listed values. This comparison may
not be a simple matter. The effort entails selecting appropriate radiation detectors,
correctly interpreting the resulting data, and being aware of distinctive forma-
tion and decay characteristics that can distinguish otherwise similar radionuclides.
Correct radionuclide identification can be crucial to planning protective measures,
especially in emergency situations, by defining the type of radiation source and its
radiological hazard. Discussed here are the information used for radionuclide iden-
tification, the sources of this information, and the application of the information
in the radioanalytical chemistry laboratory.
A radionuclide is defined by its half-life and type, fraction, and energy of emit-
ted radiation. Also of interest are its atomic number (element) Z , its mass number
(isotope) A, whether it exists in an excited or ground state, and any parent–progeny
relation, as discussed in Chapter 2. The radioanalytical chemist needs this informa-
tion initially to plan the analytical program for measuring expected radionuclides
appropriately, and later to identify and quantify any unexpected radionuclides. The
professional should be familiar with the decay schemes of commonly encountered
radionuclides and with sources of information for all decay schemes.
Each radionuclide among the more than one thousand that are known has a
unique decay scheme by which it is identified. For this reason, among others,
researchers have studied decay schemes over the years and their reported infor-
mation has been compiled and periodically updated. The compiler surveys the
reported information for each radionuclide and attempts to select the most reliable
information for constructing a self-consistent decay scheme. The fraction of beta
particles that feed an excited state must match the fraction of gamma rays plus
conversion electrons emitted by the excited state. The energy difference between
any two states must be consistent with the energies of the transition radiations plus
the recoil energy of the atom that emitted the radiations.

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

163
164 Bernd Kahn

Compilations of decay scheme information are issued to maintain this informa-

tion reasonably current, notably in the Table of Isotopes (Firestone 1996) and in
supporting nuclear databases, such as that maintained by the Brookhaven National
Laboratory at http://www.nndc.bnl.gov (Jan. 2006). These resources organize in-
formation by Z and A. Compilations prepared in the last two decades can be
considered reliable except for the few radionuclides that are still being created
(see Chapter 16), rare and incompletely described radionuclides, recent improve-
ments in precision, and occasional errors. Such problems may be identified and
then resolved by referring to two or more independent sources or to the original
decay scheme research cited in the compilation.
Because the compilations are used for various nuclear physics purposes, much
of the recently published information is far more extensive than needed by the
radioanalytical chemist. Notably, information on numerous radiations emitted by
radionuclides at less than 1% wastes the time of the analyst who wishes to identify
a radionuclide by one or two major radiations. Detailed information on radiations
of minor intensity becomes useful only when these radiations affect measurement
results for the characteristic radiations or lead to misidentification.
Simplified decay information has been extracted from detailed compilations for
practical use in identifying and measuring radionuclides (Kocher 1977, NCRP
1985b), and also is given in wall charts that usually are arranged by proton vs.
neutron numbers, as shown in Fig. 16.2. These charts provide a useful overview
of stable and radioactive nuclides to indicate patterns of radionuclide formation,
decay, and parent–progeny relations. Tabulations of radionuclides ordered by ra-
diation type and energy, separated for longer and shorter half-lives, are given in
Appendix D of the Table of Isotopes and other compilation (Wakat 1971, Shleien
1992) for quickly matching measured with known values to identify a radionuclide.
The main tools of the radioanalytical chemist for identifying and quantifying
a radionuclide are chemical separation and radiation measurement, especially by
spectral analysis. The former separates radionuclides by element, i.e., Z value,
and removes potentially confusing contaminants. The latter determines the type of
radiation, its energy, and relative intensity at various energies. These measurements
are repeated during the period of study to determine the half-life of radionuclides
that are sufficiently short-lived to observe a decrease in the count rate.

9.2. Decay-Scheme Knowledge for Planning and Analysis

The radioanalytical chemistry laboratory is designed, staffed, and operated to con-
sider radionuclide characteristics for the following purposes:
r Perform measurements within time periods consistent with half-lives
r Determine radionuclide activity by measuring emitted radiations that have suf-
ficiently large and accurately known decay fractions
r Recognize detection limitations and uncertainty imposed by the measured
radiation
 9. Radionuclide Identification 165

r Utilize spectral analysis for radionuclides with appropriate radiations

r Select radiation detectors that optimize detecting the radionuclides of interest
r Chemically separate radionuclides for element identification
r Chemically separate radionuclides for contaminant removal before radiation
measurement
Some of the above-listed purposes may be addressed simultaneously.

9.2.1. Half-Life Measurements

The half-life t1/2 of a radionuclide is calculated by performing two successive
measurements of the purified sample with a time interval sufficient to observe
significant decay:
t1/2 = 0.693(t2 − t1 )/ln C1 /C2 (9.1)
In Eq. (9.1), t is the time of measurements on the first and second occasion, C is the
net count rate at those times, ln represents the natural logarithm, and 0.693 is the
natural logarithm of 2. The equation is another form of Eq. (2.6a). A more precise
determination of the half-life is obtained with more measurements plotted on a
graph of ln count rate vs. time of measurement. If several radionuclides are in the
counted sample, the curve of best fit is not a straight line, but may be resolved into
component straight lines when only a few radionuclides are present (see Fig. 2.1).
The half-life must be known to bring radionuclides with relatively short half-
lives to the laboratory promptly before they have decayed below the level needed for
precise measurement. The measured activity is then compensated for radioactive
decay in terms of the half-life. The accuracy of the reported concentration depends
on the reliability of both the half-life and the decay interval values.
Half-life values can be used to distinguish among radionuclides that emit similar
radiations. For a sample with a mixture of radionuclides, repeated measurements
over intervals similar to the shorter half-life reduce interference by the shorter-lived
radionuclide when measuring the longer-lived ones.
Half-lives of relatively long-lived radionuclides may not be measurable if the
count rate does not decrease significantly in the time period available for analysis.
A count rate that actually increase in this interval indicates the presence of parent–
progeny pairs or chains, as discussed in Section 9.3.1.
.

9.2.2. Type and Energy of Radiation

The radiation type guides selection of the detection system and sample size, which
in turn affect the magnitude of the radiation background and the detection limit,
as discussed in Section 8.2. The gas-filled proportional and liquid scintillation
(LS) counter respond efficiently to alpha and beta particles. Low-energy X rays
can also be detected in them, while energetic gamma rays pass through with little
interaction. Thin silicon diodes respond efficiently to alpha particles and low-
166 Bernd Kahn

energy beta particles and electrons. A large Ge detector responds efficiently to

gamma rays and beta particles.
Application of shielding can make a detector more selective. A relatively thin
window or shield prevents alpha particles and low-energy beta particles from en-
tering an end-window proportional counter. A thicker shield prevents beta particles
from entering the Ge detector.
Pulse selection or rejection can distinguish by type and energy of radiation and
also optimize the ratio of counting efficiency to the background count rate. Alpha
and beta particles can be distinguished from each other by pulse height or pulse
shape selection in the proportional and LS counters. Two or three beta-particle
groups with distinctly different maximum energies can be counted simultaneously
in LS counters. Pulse-height rejection at low energy is designed to reduce back-
ground noise, but also rejects radiation at very low energy. A typical Ge spectral
analysis system designed for measuring energetic gamma rays has a low-energy
cutoff near 30 keV, but systems designed to count lower-energy gamma rays and X
rays reduce the cutoff to 1 keV or less. The low-energy cutoff for the proportional
counter typically rejects only electrons below 0.1 keV.
Detection of energetic gamma rays is optimized by a thick Ge detector, and of
low-energy X rays by high-Z gas in a proportional counter. Increasing sample size
improves gamma-ray measurement efficiency per unit sample mass until a maxi-
mum value is reached, beyond which the increasing distance of sample portions
from the detector begins to reduce the average counting efficiency. For beta-particle
measurement, backscattering by a thick, high-Z material increase counting effi-
ciency, but increasing the bulk of the sample soon decreases efficiency because
of self-absorption. Alpha particles are much more sensitive to count-rate loss by
self-absorption.

9.2.3. Radiation Decay Fraction

The radiation fraction per radionuclide decay is one of the factors for calculating
radionuclide activity from measurement that are described in Section 8.2. The
intensity of a characteristic peak measured by spectral analysis or of a count rate
measured by proportional or LS counter must be divided by the decay fraction
(among other factors) to obtain the activity of the radionuclide. Decay fraction
values can be obtained from the above-cited decay scheme compilations.
Some interpretation may be required to obtain parameters of interest from the
values reported in compilations. Gamma-ray intensities may be given in fractions
that can be used directly, or relative to a selected—usually the most intense—
reference gamma-ray intensity set at a value of 100. In the latter case, all other
gamma-ray intensities must be calculated by multiplying the given intensity of the
reference gamma ray by the appropriate fraction.
The fractional emission of conversion electrons may be reported directly, or
relative to the associated gamma ray as cek /γ and cek /celmn , where γ represents
the decay fraction of the gamma ray and ce represents the decay fraction by
conversion electrons from the k, l, m, and n shells. The decay fraction for the entire
 9. Radionuclide Identification 167

internal transition can be summed from this information on the assumption that
contributions are minimal from electron shells more distant from the nucleus. The
energies and fractions of characteristic X rays that accompany conversion electron
transitions may be reported in compilations or the energies and relative intensities
from the various K and L subshells are tabulated separately in compilations such
as Table 7 in Appendix F of the Table of Isotopes (Firestone 1996). The fractions
of characteristic X rays per conversion electron from that shell can be calculated
by multiplying the relative intensities from the shell by the K or L fluorescence
yield (see Table 3 in the same Appendix F).
The uncertainty information for these values that is needed to estimate the
uncertainty in the calculated radionuclide activity may be given in the compila-
tion or in the cited original studies. Any consistent deviations beyond this un-
certainty in quality assurance test comparisons from the reported values may
suggest, among other causes, an erroneous decay fraction in current use by the
laboratory.

9.2.4. Application of Radiation Detection and Spectral

Analysis
The end-window proportional counter detects beta particles with energies above
approximately 25 keV because lower-energy beta particles are absorbed in the thin
window, a narrow air gap, a thin film covering the sample, and the sample itself.
Lower- energy beta particles can be detected in very thin samples measured in
an internal proportional counter. An end-window proportional or Geiger–Mueler
(G-M) counter with sufficient space between window and source for inserting
aluminum foils and sheets can be used to obtain a curve of count rate vs. absorber
thickness for determining the maximum beta-particle energy (see Section 2.4.2)
from the beta-particle range.
All alpha particles in the conventional energy range are detected by the pro-
portional counter unless the attenuation between source and detector exceeds 4
mg/cm2 . They can be distinguished from beta particles by operating the counter
at an applied voltage too low to detect beta particles, or by pulse-height discrim-
ination to measure only the more energetic alpha-particle pulses if the detector
is operated at the higher applied voltage for detecting both alpha and beta parti-
cles. Very thin foils (about 1–12 mg/cm2 ) can be used to determine alpha-particle
energies by their range (see Section 2.4.1).
LS counters are useful for detecting radionuclides that emit beta particles with
energies of only a few kiloelectron volts such as 3 H (18.6 keV maximum energy).
For counting 3 H, an energy range from 1 to 19 keV can be selected to count most
tritium beta particles with a reasonably low background count rate. Considerably
more energetic beta particles (in the range of several hundred to several thousand
kiloelectron volts) can be counted separately at the same time in the energy range
from 20 keV to their maximum energy. If two beta-particle groups with suffi-
ciently different maximum energies are present, the upper range can be further
subdivided.
168 Bernd Kahn

Use of the LS counter for alpha-particle spectral analysis is discussed in Section

8.3.2. Source preparation is simpler, but energy resolution is worse than with the
solid-state detector. Special source preparation and electronic pulse-shape selec-
tion can improve resolution.
The Ge detector with gamma-ray spectrometer is the measurement instrument
of choice for radionuclides that emit gamma rays. Its high resolution minimizes
interference from other gamma rays in the sample so that many radionuclides can
be identified and quantified unambiguously in a single sample. The gamma ray is
displayed as a full-energy peak, together with a Compton continuum, that usually
is not used for identification or quantification but interferes with lower-energy
peaks. At energies above 1022 keV, the annihilation interaction peaks discussed
below are also produced.
The radionuclide is identified by the energy at the midpoint of the characteristic
full-energy peak. It is quantified in terms of the count rate in the channels that
define the full-energy peak. Subtracted from this count rate is the count rate in
these channels due to other gamma rays discussed in Section 10.3.7; in simple
cases, the background count rate per channel is the average of the count rates in
one or more channels on each side of the peak.
If the radionuclide has multiple gamma rays, a set of several of the more intense
ones can be considered characteristic to provide more convincing identification
and more counts for reduced uncertainty. Weaker gamma rays are excluded from
this set if they have a less reliable decay fraction values and more interference by
gamma rays from other radionuclides.
Gamma-ray spectral analysis of a radionuclide that emits only low-intensity
gamma rays provides results of marginal reliability unless the radionuclide has
been purified so that little interference is expected, or is in large amount. Examples
of radionuclides measured with gamma rays of low decay fraction are plutonium
isotopes and 85 Kr.
Radionuclides are also measured by gamma-ray spectrometer if they emit
positrons that are then annihilated with the creation of 511-keV gamma rays. The
efficiency of counting annihilation radiation is slightly less than that of 511-keV
gamma rays emitted by a source because some annihilation occurs in solids beyond
the source. The 511-keV annihilation radiation is also created by gamma rays more
energetic than 1022 keV that are emitted by the source. These interactions produce
not only a 511-keV peak in the detector, but also escape peaks that correspond to
the full gamma ray energy minus 511 and 1022 keV.
The characteristic X rays of the atomic electron energy levels of an excited
daughter isomer are also measured conveniently by the Ge detector with gamma-
ray spectrometer. These X rays are associated with conversion electrons emitted
by the isomer, and their energies and emission fractions are listed in radionuclide
compilations (see Section 9.2.3). Conventional detectors that measure gamma
rays above about 30 keV are used to identify K X rays from elements heavier than
Xenon.
Thin Ge detectors with spectrometer systems are designed to detect gamma
rays with energies as low as a few kiloelectron volts to measure K X rays of
lighter elements and L X rays of heavier elements. Peaks are resolved for energy
 9. Radionuclide Identification 169

sublevels such as Kα1, Kα2, and Kβ that differ by several kiloelectron volts for
the more energetic K X rays. Low-energy X rays are also spectrally analyzed with
gas proportional counters designed for that purpose, as described at the end of
Section 8.3.1.
Care should be taken to avoid misidentifying a radionuclide by peaks that are
artifacts of the detector. Most important are the escape peaks from annihilation
radiation cited above, and from low-energy radiation at the full energy minus
10 keV for the K X ray and 1.2 keV for the L X ray in Ge. Sum peaks are
displayed for a radionuclide that emits two or more gamma rays simultaneously
(“in coincidence”), as discussed in Section 9.3.5 for 60 Co, i.e., a 2505-keV peak
for two gamma rays with energies of 1173 and 1332 keV. Although not a peak,
the Compton edge, approximately 200 keV less than the full-energy peak (see
Eq. 216) in the usual gamma-ray energy region, may be misinterpreted as such
when seeking barely detectable radionuclides. Similarly, a slight increase in the
Compton continuum due to the backscattered electrons is associated with this
edge at about 200 keV. The radiation background in this energy region consists
of peaks for annihilation radiation, progeny of 222 Ra and 220 Rn gas, the terrestrial
radionuclides 40 K, uranium, and thorium, and cosmic-ray-produced radionuclides,
as discussed in Section 8.2.2.
The Si detector with spectrometer is used with thin sources to identify and
quantify radionuclides that emit alpha particles. All alpha particles are in the ap-
propriate energy range for detection unless attenuated in a thick source. Chemical
separation of the element of interest and meticulous preparation of the source usu-
ally are needed to obtain well-resolved peaks. Figure 9.1 shows the spectrum of a

100 234
238
U U
4200 4776
232
U
5320

75
Counts

50

4723
4150 5264
25

235
U
4392

0
3000 4000 5000 6000 7000
Energy (keV)

FIGURE 9.1. Alpha-partial spectrum of uranium with 232 U tracer, counted for 60,000 s.
170 Bernd Kahn

FIGURE 9.2. Conversion electron s measured whith a Si(Li) detector (from Knoll 1989,
p. 464).

separated uranium source, with the large peaks for 238 U and 234 U, the usual small
peak for 235 U, and the peak for the radioanalytical tracer 232 U. Many radionuclides
emit several alpha-particle groups at energies that—because of the small energy
difference and the lesser quality of the detector and the source cannot be resolved.
Some smaller peaks of lower-energy alpha particles are shown in Fig. 9.1. The
combined count rates of the unresolved peaks must be matched to the same com-
posite group in a known source for calibration, or the combined decay fractions of
all unresolved peaks must be used for calculating activity.
For some radionuclide mixtures, a group separation, e.g., for actinides, is sat-
isfactory for measuring its component radionuclides by alpha-particle spectral
analysis. As discussed in Section 6.4.1, further chemical separation is needed for
radionuclides that emit alpha particles of almost the same energies, or even a
mass spectrometer for radioisotopes of the same element with almost identical
alpha-particle energies such as 239 Pu and 240 Pu.
A thin solid-state detector with spectrometer is also useful for identifying and
quantifying conversion electrons. Figure 9.2 shows the spectrum of conversion
electrons from a thin source of 244 Cm. For radionuclides that also emit beta par-
ticles, the conversion electron spectrum may be underlain by the beta-particle
continuum.

9.2.5. Application of Chemical Separation

Chemical separation to characterize a radionuclide by element and purify it for
quantification by radiation measurement is needed when direct spectral analy-
sis does not achieve identification and quantification. Separation procedures are
 9. Radionuclide Identification 171

selected on the basis of the chemical behavior of the radioelement of interest and of
possible interfering radioelements. When little is known about the presence of var-
ious radionuclides in the sample, an attempt is made to separate the radioelement
of interest from all conceivable radioactive contaminants. At best, the identity and
approximate amounts of contaminant radionuclides are known, and a purification
scheme can be developed as discussed in Chapter 6.
A purification procedure can be considered appropriate if separation from other
radionuclides is so effective that, in the absence of the radioelement of interest, no
radioactivity is detected in the purified sample. That is, the measured radioactivity is
zero within the measurement uncertainty. If the radioelement of interest is present,
then contaminants should not observably increase its measured activity. To plan
the procedure, the amount of each contaminant radionuclide that remains after
applying the separation methods is calculated in terms of its decontamination
factor (DF) (see Section 3.1). Meeting the stated criteria depends on the initial
concentrations of the radioelement of interest and of each contaminant radionuclide
in the sample, and on the achieved DF. Because these values generally are not
known initially, reasonable concentrations must be assumed and subsequently
revised when initial measurement results become available.
The other important purpose of chemical separation is to remove nonradioactive
substances that interfere with the selected separation procedures and measurement
of the purified source. In many instances, dissolved solids must be removed from
the sample solution to measure carrier yield without interference and to obtain thin
sources for counting alpha and beta particles, as indicated in Sections 7.2 and 7.4.

9.2.6. Radionuclide Confirmation

Under usual conditions, the above-described processes of detector selection, chem-
ical separation, and spectral analysis are sufficient for identifying a radionuclide.
Often, only one or two of these processes must be applied. A question of identity
may arise under special circumstances:
r Two radionuclides emit the same type of radiation with almost identical energies
r Count rates are too low for clear peak energy identification
r The emitted radiation cannot be examined by spectral analysis
r Chemical separation is not possible for radioisotopes of the same element
r Chemical separation is imperfect for detecting a minor constituent

Resolving uncertainty is especially important if the radionuclide is relatively haz-

ardous or characterizes a source or circumstance of particular concern.
Radionuclides are confirmed by applying redundant processes. A radionuclide
is identified by gamma-ray spectral analysis and checked by chemical separa-
tion followed by a second spectral analysis measurement. Different analysts are
assigned to analyzing the same sample by different chemical separations. The
expected absence of gamma rays is confirmed by gamma-ray spectral analysis
after the usual measurement for alpha- or beta-particle activity. Measurements are
repeated to determine the half-life or a parent–daughter relation. Tabulations of
172 Bernd Kahn

energy and half-life are scanned to look for other radionuclides with the measured
decay scheme.

9.3. Decay Scheme Information and Application

Decay schemes for seven sets of radionuclides—89 Sr, 90 Sr/90 Y, 137 Cs/137m Ba, 40 K,
60
Co, 22 Na, and 241 Am—that exhibit the various aspects of spontaneous nuclear
transformations are shown in Figs. 9.3–9.9 as a guide to their interpretation and
application. Each decay is discussed separately in Sections 9.3.2–9.3.7. These
figures were selected from earlier publications—the first six from Lederer et al.
(1967), and the seventh from Martin and Blichert-Tolf (1970)—to show the con-
ventional formalism for radioanalytical chemistry purposes, without the details
that are useful to some practitioners, but excessive and confusing for radionu-
clide identification and quantitation. The data cited in these sections are more
recent (2005), from the Brookhaven National Laboratory nuclear data tables, than
the ones shown in the figures. Uncertainty in the reported values generally is as-
sociated with the last numeral on the right. As the minor differences between
Figs. 9.3–9.8 and Fig. 9.9 suggest various compilers prepare decay schemes in
similar but not identical formats.
The following conventions apply to the information of primary interest to the
radioanalytical chemist presented in these decay schemes:
r element symbol below energy level line, with value of A to upper left and value
of Z to lower left
r half-life at energy level line in years, days, min, or s
r the energy at each level relative to the ground state, in MeV
r the decay energy (Q) in MeV
r alpha-particle decay, α, an arrow (sometimes with a double line) to lower left
r beta-particle decay, β − , an arrow to lower right
r positron decay, β + , a vertical line followed by an arrow to lower left

FIGURE 9.3. Decay scheme for 89 Sr (modified from Lederer et al. 1967, p. 392).
 9. Radionuclide Identification 173

FIGURE 9.4. Decay scheme for 90 Sr/90 Y (modified from Lederer et al. 1967, p. 392).

FIGURE 9.5. Decay scheme for 137 Cs (modified from Lederer et al. 1967, p. 399).
174 Bernd Kahn

FIGURE 9.6. Decay scheme for 40 K (modified from Lederer et al. 1967, p. 384).

FIGURE 9.7. Decay scheme for 60 Co (modified from Lederer et al. 1967, p. 389).

FIGURE 9.8. Decay scheme for 22 Na (modified from Lederer et al. 1967, p. 382).
 9. Radionuclide Identification 175

241
FIGURE 9.9. Decay scheme for Am (modified from Martin and Blichert-Tolf, 1970,
p. 144).

r electron capture (EC) an arrow to lower left

r internal transition by gamma ray and conversion electrons with characteristic X
rays and Auger electrons, a vertical arrow downward
r alpha-particle energy, next to appropriate arrow, in MeV
r gamma-ray energy, the energy of the excited state to right on energy level line,
or above appropriate arrow for multiple gamma rays (see Figs. 9.7 and 9.9), in
MeV
r the decay fraction, next to appropriate arrow, in percent

Additional information necessary for radionuclide identification and analysis but

not shown in Figs. 9.3–9.9 is given in Sections 9.3.2–9.3.7; it includes the fraction
of conversion electrons and X rays in an internal transition and the beta-particle
maximum energy.
Also in these figures is information primarily of interest to the physicist that can
provide some guidance to the radioanalytical chemist:

r nuclear spin, e.g., 5/2−, 2+, shown at left on energy level line, that relates to the
intensity of transition
r gamma ray transition polarity, e.g. M4, E0, shown sideways on internal transition
vertical arrow
r log ft value, e.g., 12.1, 8.6, listed after the beta-particle energy, that defines the
spectral distribution of beta particles

Applications of these items are discussed in nuclear physics texts and summarized
in the nuclear and radiochemistry texts cited in Chapter 1.
176 Bernd Kahn

9.3.1. Radioactive Decay and Ingrowth Rate

First consideration usually is given to the half-life shown on or next to the line
in the decay scheme that represents the radionuclide energy level. For example,
the 2.55-min half-life of 137m Ba in Fig. 9.5 suggests that, once 137m Ba is separated
from its 137 Cs parent, rapid methods must be applied to separate and count it. The
90
Y 64-h (Fig. 9.4) half-life requires prompt attention and priority for analysis
after separation from its parent 90 Sr. The longer half-lives of all other examples do
not require special scheduling efforts for analysis.
Short half-lives enable identification by decay measurement. They also require
precise records of the time of sample collection, separation from the longer-lived
parent, and the beginning and end of counting to calculate ingrowth and decay.
Decay correction for reporting the activity at the time of collection is needed for all
radionuclides, but is significant under usual sample analysis scheduling only for
radionuclides with short and intermediate half-lives; 89 Sr is an example of the latter.
Equations (2.6) and (9.1) describe radioactive decay. Figure 2.1 shows two
straight lines on a semilogarithmic scale that represent decay by two radionuclides
in the sample, observed by measurements repeated over a period of time. The pre-
cision of each calculated slope, characterized by the decay constant λ, is optimized
by obtaining large net count rates, numerous repeated measurements, reliable back-
ground count rates for subtraction, and meticulously reproduced source–detector
configurations for the sequential measurements. The value of the decay constant
can be obtained with a least-squares fit for each of the component lines. Because
the important aspect of these measurements is their relative values, the counting
efficiency of the detector is not needed.
The curve toward a horizontal asymptote of the measured line in Fig. 2.1
suggests the presence of at least a second radionuclide with a smaller decay
constant, i.e., a longer half-life. Because the curved line represents the sum of two
or more straight lines, the contributing straight lines with their decay constants
may be determined by fitting first the line for the longest-lived radionuclide, and
then, in turn, lines for the shorter-lived components. Fitting multiple lines reduces
the precision compared to a single straight line. Measurements at times later than
shown in Fig. 2.1 may identify a third component if the line continues to curve
toward the horizontal, or at least suggest an upper limit for a third component if
the difference between the measured count rate and the background count rate is
near zero and hence uncertain.
Equation (2.8) describes the decay of a radioactive parent with ingrowth of its
radioactive daughter. The three types of ingrowth relations—secular, transient, and
no equilibrium—are discussed in Section 2.2.4 and shown in Figs. 9.10–9.12.
The secular equilibrium example of the 226 Ra – 222 Rn pair in Fig. 9.10 is analo-
gous to the 90 Sr – 90 Y and 137 Cs – 137m Ba pairs discussed below. The slope of the
line that represents the disintegration rate of the parent with its half-life appears
horizontal because of negligible decay on the scale of 10 daughter half-lives. It
is approached by the disintegration rate of the daughter after about six half-lives.
The two disintegration rates become equal beyond that time.
 9. Radionuclide Identification 177

FIGURE 9.10. Radioactive daughter ingrowth—secular case (from Vertes et al. 2003,
p. 275).

Transient equilibrium occurs when the half-life of the daughter is only somewhat
shorter than that of the parent (Fig. 9.11). The slopes of the disintegration rate of
the daughter approaches that of the parent after about six half-lives. The important
distinction from secular equilibrium is that the disintegration rate of the daughter

FIGURE 9.11. Radioactive daughter ingrowth—transient case (from Vertes et al. 2003,
p. 273).
178 Bernd Kahn

FIGURE 9.12. Radioactive daughter ingrowth—no equilibrium case (from Vertes et al. 2003,
p. 276).

ultimately exceeds that of the parent by the factor t1 /(t1 – t2 ), where t is the half-life
and the subscripts 1 and 2 refer to the parent and daughter, respectively, as inferred
from Eq. (2.8).
No equilibrium is reached when the daughter half-life exceeds that of the parent.
As shown in Fig. 9.12, the daughter first accumulates as the parent decays, and
will remain for some time after the parent no longer can be detected. The daughter
will decay with its own half-life when, for practical purposes, the parent no longer
remains.
The decay constant of a daughter in equilibrium with its parent may be inferred
from repeated measurements of the count rate during ingrowth by use of Eq. (2.8),
or directly determined by chemically separating the daughter and then performing
repeated measurements of the count rate as the daughter decays.

9.3.2. Beta-Particle Decay: 89 Sr and 90 Sr/90 Y

The 89 Sr decay scheme in Fig. 9.3 is simple. For practical purposes, the radionuclide
emits a beta particle group with a maximum energy of 1495 keV (plus the associated
neutrino group discussed in Section 2.2.2). The half-life is 50.5 days. An obscure
beta particle group of 590 keV maximum energy, followed by gamma rays of
909 keV with an intensity of about 0.01%, is of no practical consequence in
radioanalytical chemistry.
The 90 Sr/90 Y decay scheme in Fig. 9.4 shows a parent and daughter (progeny)
pair. The parent has a long 28.8-years half-life and the daughter has a short 64.0-h
 9. Radionuclide Identification 179

half-life. Each emits a pure beta-particle group in 100% of the decay, which is
directly to the ground state. The maximum beta-particle energies are 546 keV for
90
Sr and 2280 keV for 90 Y. The minor (0.0115%) radiation emitted by 90 Y of
a 519-keV maximum-energy beta-particle group followed by 1760-keV gamma
rays is usually of no practical consequence. If the 90 Y is separated from 90 Sr, the
subsequent daughter ingrowth is described as secular equilibrium discussed in
Section 9.3.1.

9.3.3. Beta-Particle Decay to Excited State: 137 Cs/137m Ba

The 137 Cs/137m Ba decay scheme in Fig. 9.5 shows two beta-particle groups emitted
by 137 Cs. The 1176-keV maximum-energy beta-particle group decays to the ground
state in 5.6% of the decays and the 514-keV maximum-energy beta-particle group
decays in 94.4% of the decays to an excited state in 137 Ba. The 137m Ba is in secular
equilibrium with 137 Cs. This metastable state decays with a 2.55-min half-life.
The decay of 137m Ba is 85.1% by 661-keV gamma rays and 9.4% by conversion
electrons.
The sum of conversion electron and gamma-ray decay of 94.4% equals the
percent decay of the lower energy beta-particle group. The conversion electrons
consist of K electrons at 624 keV (7.66%), L electrons at 656 keV (1.39%), and
M, N, and O electrons at about 660 keV (total of 0.3%).
The X rays that accompany conversion electrons are X(K) at about 32 keV
and X(L) at about 4.5 keV. With a Ge detector, several groups of X(K) energies
can be distinguished, such as Kα1 at 32.2 keV (3.6%), Kα2 at 31.8 keV(2.0%),
and Kβ at 36.4 keV (1.3%). The sum of these X-ray transitions is less than the
fraction for conversion electrons because other transitions at lesser intensities occur
and because Auger electrons are also emitted. The Auger electrons have average
energies of 26 keV (0.8%) for the K shell and 3.7 keV (7.3%) for the L shell.
An important point in measurements is that conversion electrons are counted
with the usual end-window beta-particle detectors and not with gamma-ray spec-
trometers. The decay fraction counted under these conditions with a beta particle
detector is 1.094. The K Auger electrons may also be counted with beta particles
and conversion electrons in an internal proportional counter. In a Ge detector and
spectrometer, in addition to the gamma ray, the various K X rays are counted.

9.3.4. Electron Capture and Beta-Particle Decay: 40 K

The decay scheme of the naturally occurring radionuclide 40 K (see Fig. 9.6) is
89.1% by beta-particle emission, 10.9% by electron capture, and 0.001% by
positron decay. The 1311-keV maximum-energy beta-particle decay is to the
ground state of 40 Ca. Electron-capture decay is followed instantaneously by emis-
sion of a 1461-keV gamma ray from the excited state of 40 Ar to its ground state.
Electron capture decay to the ground state is 0.2% per disintegration.
Electron-capture decay is associated with emission of 3-keV (1%) argon K X
rays and of K Auger electrons at an average energy of 2.7 keV and an intensity
180 Bernd Kahn

of 7.3%. Positron decay produces two 511-keV gamma rays in the course of
annihilation, or 0.002%.
The 1461-keV gamma ray at 0.107 per disintegration and the beta-particle group
at 0.891 per disintegration are commonly measured. The other radiations—notably
positron decay—are at too low intensity for practical purposes.

9.3.5. Beta Particles and Gamma Rays in Coincidence: 60 Co

The decay scheme of 60 Co shown in Fig. 9.7 consists, for measurement purposes,
of a 318-keV maximum-energy beta-particle group (99.88%) followed instanta-
neously by a 1173-keV gamma ray and a 1332-keV gamma ray. Minor beta-particle
groups that constitute 0.12 and 0.009% of the decay go to the indicated lower ex-
cited state of 60 Ni. The major beta-particle group and the two following gamma
rays constitute 99.9% per disintegration. Conversion electrons are about 0.03% of
total decay, and the characteristic K X rays at about 7.5 keV are 0.01% of total
decay.
The major beta-particle group is in coincidence with the two gamma rays that
follow it, and one gamma ray is in coincidence with the other. These relations
can be applied to detect 60 Co in the presence of numerous other radionuclides
by counting only radiations in coincidence with each other. For example, a beta-
particle detector and a gamma-ray detector can be coupled in a system to count
each types of radiation separately and also to count in a third recorder those beta
particles and gamma rays that are detected simultaneously, as defined by the time
constant of the system. In the case of 60 Co, either one or both gamma rays may be
recorded for beta-gamma coincidences.
A special benefit of counting radiations in coincidence is that the process
permits the absolute measurement of the activity of the radionuclide. For the
beta-particle counting efficiency εβ and the gamma-ray counting efficiency εγ (in
count/disintegration), the activity A is related to the three count rates. The count
rate Rβ of beta particles in the beta-particle counter, the count rate Rγ of gamma
rays in the gamma-ray counter, and the coincidence count rate Rβγ (in count/s)
yield A (in Bq) by:
R β = εβ A (9.2)
R γ = εγ A (9.3)
Rβγ = εβ εγ A (9.4)
The three equations can be solved for A while eliminating the counting efficiencies:
Rβ Rγ
A= (9.5)
Rβγ
Eq. (9.5) becomes complicated (NCRP 1985b) when some beta particles are
counted in the gamma-ray detector or vice versa, or the decay scheme is more
complex than for a beta particle followed by a single gamma ray. For example,
conversion electrons may be in coincidence with beta particles and X rays, or
 9. Radionuclide Identification 181

several beta particle groups or gamma rays that have different counting efficien-
cies may be emitted, some in coincidence and others not in coincidence.
For 60 Co, gamma–gamma coincidence counting can be performed with a single
Ge detector and gamma-ray spectrometer by recording separately the peak count
rates for the two coincident gamma rays and their coincidences at the summed
full-energy peak. Corrections are necessary for the nonpeak count rate and any
significant angular correlation between the two gamma rays (NCRP 1985b). Coin-
cidences between other pairs of radiations are also used for absolute measurement.

9.3.6. Positron Emission and Electron Capture: 22 Na

The decay scheme of 22 Na in Fig. 9.8 typifies positron emission accompanied by
some electron capture. The main decay is by 546-keV maximum-energy positron
group (90.5%) and electron capture (9.5%) to the 22 Ne-excited state. The latter de-
cay is followed essentially instantaneously by the emission of a 1275-keV gamma
ray (99.95%). A very small fraction of positron emission has 1820-keV maximum
energy (0.056%) and goes to the ground state. The fraction of 0.85-keV K X ray
is minimal (0.2%). The fraction of Auger K electrons at an average energy of
0.82 keV is 9.2%.
The major positron group and the gamma ray can be measured by coincidence
counting (see above for 60 Co). In addition to the noted gamma ray, 511-keV anni-
hilation radiation is detected by the gamma-ray detector; it is generated at the rate
of two per positron, i.e., 1.81 per disintegration.

9.3.7. Alpha-Particle Decay and Gamma Rays: 241 Am

The 241 Am daughter of 241 Pu shown in Fig. 9.9 emits mainly a 5486-keV alpha
particle (84.5%) to an excited state of 237 Np. Emission of two other alpha particles
at successively slightly lower energy and considerably less intensity is typical
of many alpha-particle emitters. In addition, two alpha particles of less than 1%
intensity each have slightly higher energies, and 18 alpha particles of less than
0.02% intensity each have been reported between energies of 4757 and 5469 keV.
The energy difference between the decay energy Q and the alpha particle to the
ground state is due to the atom recoil energy discussed in Section 2.2.1.
The main gamma ray has an energy of 59.5 keV at an intensity of 35.9%. Two
other gamma rays of 26.3 and 3322 keV have intensities of greater than 0.1%, and
many gamma rays of extremely low intensity are emitted with energies between
922 and 43 keV. The L X ray of 14 keV has an intensity of 36.9%, while the K X
ray of about 100 keV has an intensity of only about 0.003%. Numerous conversion
electrons are emitted; the most intense groups are near 37 keV (33%) and 11 keV
(17%). Numerous L Auger electrons at about 10 keV are also emitted.
The radionuclide activity usually is determined by measuring the 5486-keV
alpha particle or the 59.5-keV gamma ray. Americium-241 is often measured to
determine its beta-particle-emitting 14.35-years 241 Pu parent.
182 Bernd Kahn

9.4. Radionuclide Characterization for Special and

Routine Occurrences
The first effort in a characterization project usually is devoted to screening the
sample for radionuclide content. Screening includes:
r Prediction of the radionuclides and their concentrations from a known or assumed
source
r Attribution to source from other information
r Inference of the radionuclide mixture and overall decay rate in submitted samples
based on type of sample and prescreening radiation scans of sample
r Initial gamma-ray spectral analysis
r Initial gross alpha- and beta-particle counting
r Observations of decay rates in gamma-ray spectra and gross counts
r Rapid chemical separations and radiation measurements for important radionu-
clides with little or no gamma-ray emission
r Integration of information developed within laboratory and by others
r Confirmatory and quality control measurements

Subsequent measurements pursue the implications of the screening results for

monitoring persons and places, and seek radionuclides that are inferred to be
present but would not have been found by such screening.
Sample characterization for radionuclide content under emergency response
conditions requires a combination of rapid processing and careful evaluation that
is best planned in advance. Included in advance planning must be a prescreening
facility for distinguishing between samples with higher and lower radionuclide
levels for sending them to separate laboratories with detectors dedicated to mea-
suring these different levels. Protocols must be written to assure availability of
skilled analysts, sample tracking and quality assurance for data reliability, and
minimal cross-contamination to maintain the viability of detectors and samples.
Exercises should be performed at appropriate intervals to test these plans and
accustom analysts to emergency response actions.
Radionuclide identification to explain puzzling radiation measurement results
or respond to clients concerning a routinely submitted sample requires the same
efforts. The main distinction is that usually the sample radionuclide content and
level is more narrowly defined and the response can be relatively unhurried.

9.4.1. Screening Measurements

Screening measurements of incident response samples begin as soon as possible
after sample receipt and logging. A necessary first step is prescreening to prevent
a sample from being taken to a laboratory if its radionuclide content is unsuitably
high for the usual controls of personnel radiation exposure and contamination of
the analyst, the laboratory, and radiation detectors. External radiation is measured
with conventional low-level radiation monitors that include the G-M detector and
 9. Radionuclide Identification 183

the NaI(Tl) detector with spectrometer. Surface contamination is monitored by

wiping with paper or cloth smears the outside of shipping and sample containers
and measuring radionuclides on the smears with a G-M or LS counter.
Once the accepted sample is moved to the appropriate laboratory, aliquots of the
sample are transferred to tared containers that provide reproducible measurements.
The weight is recorded for solids and the volume for liquids. Samples for gamma-
ray spectral analyses are placed in closed containers. Samples for gross alpha- and
beta-particle measurements with a proportional (or G-M) counter are stabilized by
evaporating liquids or slurrying and drying solids and by covering them with thin
film. Samples for measurement with an LS counter are mixed with scintillation
cocktail in a counting vial.
The results of brief initial counting—typically for a period of 1–5 min—are
reviewed to determine whether they provide sufficient information on radionuclide
content. If not, a longer counting period, say 1–2 h, is selected to provide as much
reliable data as possible on the presence or absence of specified radionuclides
within the necessarily limited period.
Detector quality control records are reviewed to assure that control samples
and the radiation background have been measured recently and that the detectors
in use are within control limits (Section 11.2.10). Brief control source and back-
ground measurements are performed before the screening process begins to assure
that detectors continue to operate appropriately and have not been contaminated
recently. The detection limit in terms of activity per sample is calculated for all
radionuclides of interest to determine whether a null result will meet radionuclide
detection requirements for the submitted samples.
Gamma-ray spectral analysis typically is performed with a Ge detector plus
spectrometer for radionuclides that emit gamma rays in the energy range of
30–4000 keV. All peaks are noted, and those believed to be full-energy peaks
are tentatively identified by comparing them to a table of gamma-ray energies for
all radionuclides. Assignment of multiple peaks to a radionuclide increases confi-
dence in the identification. Wide peaks are tested for resolution to consider whether
more than one gamma ray exists per peak energy region. Energies of background
radiation—both in the sample and at the detector—and escape peaks are listed to
eliminate such peaks from consideration. Concentrations of the tentatively identi-
fied radionuclides are estimated on the basis of listed gamma-ray decay fractions,
sample mass or collector flow rate, and available detector counting efficiencies as
function of energy.
If the decay of peaks in the gamma-ray spectrum is observable, the half-lives
associated with these peaks are estimated and compared to tabulated values to
confirm identification by energy. Gamma-ray peaks that appear constant are con-
sidered to be emitted by radionuclides that have half-lives long compared to the
observation period.
At the same time, other aliquots are counted with gas proportional and LS coun-
ters to determine the presence of alpha and beta particles. The detectors are operated
in a mode that distinguishes between alpha and beta particles. The LS counter can
suggest maximum beta-particle energies and show alpha-particle energies from
184 Bernd Kahn

spectrometric readings. The activity of total alpha and beta particles per sample
is estimated by assuming a typical or intermediate counting efficiency. These ef-
ficiency values must be revised once the samples are better characterized because
factors such as beta-particle energy, sample self-absorption (in the proportional
counter), and quenching (in the LS counter) affect counting efficiency.
If the gross alpha-particle activity is sufficiently high, a thin sample—of the order
of 1 mg/cm2 —is prepared for alpha-particle spectral analysis with a Si detector
plus spectrometer. The energy range of interest usually is 4–10 MeV.
Measurements are repeated at specified intervals to observe radioactive decay
or ingrowth. Counting usually is performed after increasing intervals, e.g., 0.5, 2,
8, and 24 h, as the initially rapid overall decay becomes slower. Decay or ingrowth
measurements for screening are ended when no further change is observed in the
count rate.
Any change in the net count rate associated with alpha- and beta-particle count-
ing provides a composite decay–rate curve that may be analyzed for a few com-
ponent half-lives by the procedure discussed in Section 9.3.1. These half-lives are
used to identify the radionuclides in the sample by scanning a list of radionuclides
tabulated by half-life. If the sample has many radionuclides with short half-lives,
only a general conclusion can be drawn concerning the presence of such radionu-
clides.
Initial measurements may stimulate further analysis with different sample mass
or type of measurement. A larger sample is prepared if the count rate was too low
for reliable measurement, and a more favorable geometry or longer counting period
may be selected. Too intense a sample suggests use of a smaller sample, greater
distance of source from detector, and shorter counting period. If the gamma-ray
spectrum suggests the presence of gamma rays below 30 keV, the sample can
be measured with an X-ray spectrometer to energies as low as 1 keV. A limited
table of gamma-ray energies and radionuclide half-lives used for identification—
either on paper by computer—may be modified to include additional radionuclides
and remove ones inappropriate for the recorded spectra. Other detectors that are
considered more useful for the type of sample or expected radionuclides may be
selected for further measurements.
When no radiation is observed above the detector background, the estimated
detection limit is compared to the regulatory limit or radiation protection guidance
to determine whether sample analysis results are sufficiently sensitive. If not,
larger samples, longer counting periods, and more efficient detection are required
for more sensitive measurements. Any radionuclides that have been tentatively
identified and quantified are compared to initial predictions to assure consistency
or consider reasons for differences.
Additional samples may have to be collected if an observed radionuclide can
be attributed either to the source of interest or to the ambient background due to
another source. These samples are analyzed to examine the pattern and concentra-
tion range of the radionuclide and establish as reliably as possible the distinction
between the two sources.
 9. Radionuclide Identification 185

9.4.2. Information from Others

Prior to or in parallel with the screening measurements, available information is
collected to suggest the set of radionuclides that can be anticipated in the sub-
mitted samples. A known radionuclide origin, such as a nuclear weapon, nuclear
reactor, nuclear fuel cycle facility, accelerator, hospital nuclear medicine group,
or irradiation source, will limit the list of potential radionuclides in a sample and
possibly identify specific radionuclides. Even better are results of measurements
performed at the site of origin to identify the radionuclides to be analyzed.
Various transport mechanisms from source to sample can be expected to modify
the relative amounts of the component radionuclides. The type of sample may
suggest the influence of the transport mechanism on the relative radionuclide
content of the sample. For example, an air filter retains solid particles in preference
to gases, and a filtered water sample contains only soluble radionuclides. Radiation
monitoring of the sample in its container at collection, during transport, and on
arrival at the laboratory may indicate the magnitude or absence of emitted gamma
radiation, the extent of radioactive decay during transport, and the presence or
absence of specific radionuclides.
Results obtained at other laboratories can suggest the identity and quantity
of radionuclides at the origin even if different samples are analyzed. Cooperation
with other groups that participate in the response can be decisive in focusing on the
appropriate set of radionuclides instead of wasting time following false trails. On
the other hand, confirmation of reported radionuclides in the laboratory is necessary
because information from others on occasion has been unreliable and misleading.

9.4.3. Rapid Radiochemical Separations

A thin source for alpha spectral analysis should be prepared if screening or outside
information indicates the presence of radionuclides that emit alpha particles but few
or no gamma rays. The presence of alpha particles is predicted by a positive result
for gross alpha particle activity and by gamma rays attributed to some actinide
radionuclides. Suitable analytical methods are discussed in Chapter 6. A simple
approach for a liquid sample is to purify actinides with a small ion-exchange
resin column and prepare a thin source by electrodeposition. For solid samples,
purification is preceded by ashing and dissolution. Tests performed in anticipation
of an emergency should demonstrate the shortest time period required for these
processes to function reliably. Calculations can suggest the sample amount needed
to meet the detection limit requirements in a reasonable counting period.
If the gross beta-particle measurement is well in excess of the total beta particle
activity inferred from gamma-ray spectral analysis, then radionuclides that emit
only beta particles must be measured. The search usually can be narrowed by
information on other radionuclides or from screening. The relatively few radionu-
clides that emit energetic beta particles but not gamma rays include 32 P, 89 Sr, 90 Sr,
91
Y, 99 Tc, and 147 Pm. Methods presented in Chapter 6 can be tested for obtaining
186 Bernd Kahn

results rapidly by reducing the number of steps and simplifying the remaining
steps, but the reduced effort must provide sufficient DFs for anticipated impuri-
ties. Incomplete decontamination is acceptable only if the impurity is minor and
the measurement technique corrects for the impurity.
Radionuclides that emit only low-energy beta particles may not be detected by
gross beta-particle analysis with an end-window proportional counter, but can be
detected, although at low efficiency, by LS counting. These radionuclides include
3
H, 14 C, 32 Si, 33 P, 35 S, 45 Ca, 59 Ni, 63 Ni, 228 Ra, and 241 Pu. The long-lived radionu-
clides among these have very low decay rates and usually are not of concern as
immediate health hazards.

9.4.4. Quality Assurance

Some aspects of conventional quality assurance discussed in Chapter 11 cannot
be applied to screening processes that require data reporting soon after sample
receipt. Nevertheless, statements based on measuring radionuclides—notably their
concentrations relative to exposure limits or their absence—must be carefully
checked in emergencies because of their impact on efforts for protecting humans
and area control and remediation.
Screening in response to emergencies is particularly subject to problems due
to the unexpected situation and the possibly unprepared persons who collect and
transport samples. In the laboratory, problems can arise due to large differences in
radionuclide contents among samples, unexpected radionuclide mixtures, unusual
modes of detector operation, and the pressure to report analytical results immedi-
ately. Problematic actions that need to be prevented because they adversely affect
results include:
r Unreliable chain of custody descriptions of collection process, sample content
r Sample contamination during collection or delivery
r Laboratory cross-contamination
r Application of unfamiliar rapid methods
r Operation of detectors in unusual modes
r Hurried data calculation, analysis, and reporting

Special preparation is needed to anticipate and prevent these problems.

A sound approach is to prepare written instructions and conduct emergency
exercises. A special reception area for radiation scanning of incoming samples,
described in Section 13.3 for routine samples, should be prepared for receiv-
ing emergency samples. Control-source and background measurements must be
interspersed with sample measurements despite the urgency given to measuring
incoming samples. Collection and analysis of selected samples in duplicate should
be specified. Calculations and data interpretation should be checked by at least a
second person. Any results that may lead to major action decisions in the matter of
concern must be evaluated for consistency with internal and outside information
before they are reported.
 9. Radionuclide Identification 187

The results that guide emergency response decisions must be susceptible to

subsequent evaluation. Chain-of-custody information, procedural instructions and
instruction changes, analyst comments, radiation measurement records, and QC
data must be in writing and preserved despite the confusion that tends to accompany
emergency response. Occurrence of contamination, instrument failure, analytical
problems, and calculating error must be annotated. Samples and measured sources
must be retrievable for subsequent confirmatory or expanded measurements.

9.5. Identification Needs

Radionuclide monitoring requires identification and quantification for the follow-
ing interrelated purposes:
r Associate radionuclides with their origin
r Relate identified to expected radionuclides
r Determine extent and magnitude of radionuclide contamination
r Compare radionuclide concentrations to concentration limits
r Show absence of specified radionuclides

The essence of any monitoring program is to distinguish between radionuclides

contributed by the activity, incident, or process that is being monitored and radionu-
clides that occur in the sample as “background.” The background in the sample
conventionally is considered to consist of the terrestrial and cosmic-ray-produced
radionuclides. For careful measurements at low concentrations, the background
must be considered to include man-made radionuclides in fallout from nuclear
devices tested in the environment, a power source from a burned-up satellite, and
airborne and waterborne discharges from nuclear power fuel cycle facilities. Man-
made radionuclides also reach the environment by discharges from radionuclide
production, research, and application, from stored and buried wastes, and from
various incidents (Eisenbud 1987).
Analysis is planned for the radionuclides that are expected from the object of
monitoring, with emphasis on radionuclides for which relatively large amounts are
estimated to occur, be released, and reach the sample. Under some circumstances,
radionuclides that are measured most readily or most sensitively may be used as
tracers for accompanying radionuclides. Gamma-ray spectral and gross activity
analyses are useful for testing predictions of radionuclide amounts, releases, and
transfers, and also to seek unexpected radionuclides.
For initially undefined sources of radionuclide origin, such as certain accidental
releases or terrorist actions, the wider net that must be cast by screening is described
in Section 9.4.1. Measurements obtained near the origin of the radionuclides pro-
vide the best guidance for analyzing samples collected at greater distances. The
radionuclides selected for analysis and the precision of analysis are guided by the
plans to identify the source at the origin and to understand the circumstance that
led to the dispersion of the radionuclides.
188 Bernd Kahn

Once the radionuclides of interest are identified, a sample collection pattern

should be developed to delineate the region of contamination and the radionuclide
concentrations that affect radiation exposure. The usual concern is to identify loca-
tions of external exposure by direct radiation and internal exposure by inhalation
and food consumption. The lower limits of detection required for analysis are
specified by the customer to assure reliable detection in accord with the regulation
or guidance that limits the concentration of each radionuclide. The lower limit of
detection usually is required to be 2–10 times below the concentration limit to
assure reliable measurement at the limit.
Establishing the absence of a radionuclide becomes important when radioan-
alytical chemistry is used to demonstrate that a radiological incident, such as a
threatened terrorist act, did not occur. Absence of a substance in a sample is a
relative conclusion that could be altered by a more sensitive measurement. A ra-
dionuclide is reported as measured or as “less than” the lower limit of detection. A
more sensitive detection method may replace the “less than” description with an
actual value, or continue to report “less than,” but at a lower value. Such “less than”
values are based on net count rates that may be zero or a positive value sufficiently
near zero that is too uncertain, as discussed in Section 10.4, to be reported. This net
count rate can be the difference between the gross count rate and (1) the detector
background count rate or (2) the count rate in the sample attributed to background
from various sources, as defined above.
10
Data Calculation, Analysis,
and Reporting
KEITH D. MCCROAN1 and JOHN M. KELLER2

10.1. Introduction
The reputation of the radioanalytical chemistry laboratory is based on the extent to
which its reported results are judged to be reliable and reported in a form responsive
to customer needs. The effort to obtain and provide such data requires the competent
execution of every analytical step in the process, as well as adherence to the quality
assurance tenets discussed in Chapter 11. The result of these interlocking activities
should be an accurate and defensible data set. This chapter addresses the processing
and evaluation of those data.
Four types of calculations are typically performed that pertain to measured
radionuclide values:
r Conversion of the instrument signal (counts or count rate) and other measured
quantities to an activity concentration, typically expressed in becquerels (Bq) or
picocuries (pCi) per unit mass (or per unit volume, area, or time);
r Calculation of the uncertainty of the result in terms of an estimated standard
deviation or a multiple of the standard deviation;
r Calculation of the critical value, which is used to decide whether the radionuclide
is “detected”; and
r Calculation of the minimum detectable activity (MDA).

These types of calculations are discussed in Sections 10.2 through 10.4.

For the reasons discussed in Chapter 11, quality control (QC) measurements
must be performed. Replicate measurements are only one aspect of the overall
program of QC. The additional calculations associated with QC measurements
fall into three main categories: instrumental, analytical, and methodological. These
topics are covered in Section 10.5.
Preparing, compiling, and presenting the information developed by a radioana-
lytical chemistry laboratory is not merely a clerical exercise; it is fully as important
as the analysis and measurement effort. The work must assure that the reported val-
ues meet the data quality objectives (see Section 11.3.1) in being reliable, providing

1
National Air and Radiation Environmental Laboratory, USEPA, Montgomery, AL 36115
2
Oak Ridge National Laboratory, Oak Ridge, TN 37831

189
190 Keith D. McCroan and John M. Keller

the information required by the initial plan, and conveying to the reader a clear
understanding of the radiological impact of the subject. Section 10.6 describes the
process of data review, the steps that ensure the viability of the data. Section 10.7
addresses some of the important details of data presentation.

10.2. Calculating the Activity

A typical radioanalytical measurement involves radiation counting, although other
measurement techniques, such as ICP-MS, are sometimes used. The initial pro-
cessing of the raw counting data usually involves two steps. The first step is division
of the gross sample count CG by the length of the counting period tG to calculate
the gross count rate RG . The second step is subtraction of the detector background
count rate RB from the gross count rate to calculate the net count rate RN .
Conversion of the net count rate to activity (or massic activity or volumic activity)
produces the final result that is reported to the client. The mathematical model for
calculating the final result a often has a form similar to that shown in Eq. (8.2),
and in Eq. (10.1):
(CG /tG ) − (CB /tB ) RG − RB
a= = (10.1)
εY m DF εY m DF
where
a is the massic activity of the sample (or volumic activity)
CG is the gross count observed when the sample is counted
tG is the duration (live time) of the counting period for the gross sample
measurement
CB is the count observed when the background (or blank) is counted
tB is the duration of the counting period for the background (or blank)
ε is the practical counting efficiency
Y is the chemical yield
m is the mass of the sample aliquot analyzed (or volume)
D is the decay or ingrowth factor
F is the decay fraction (or radiation emission probability)
RG is the gross count rate (CG /tG )
RB is the background (or blank) count rate (CB /tB )
If the count times are expressed in seconds and the mass in grams, then the result
of the calculation is given in becquerel per gram (Bq/g).
The practical counting efficiency ε represents the probability that any particular
photon or particle of radiation emitted by the sample source will be recorded by
the detector. As explained in Section 8.2, its value may depend on many factors,
including the detector, the type and energy of the radiation, the composition of
the source, and the geometry of the source–detector configuration. It includes the
loss factor in the pulse analysis system and attenuation and scattering fractions
associated with the sample-detector system. All of these factors are discussed
further in Section 8.2.
 10. Data Calculation, Analysis, and Reporting 191

The mass m in Eq. (10.1) may represent mass of material that is moist, dried, or
ashed. The concentration may be reported on the basis of moist weight to indicate
the radionuclide concentration in the environment, or on the basis of dried or
ashed weight for reliable comparison with measurements of other samples or by
others. For such comparisons, aliquots of a sample are weighed moist as collected,
then after drying at 110◦ C, and then again after thorough ashing, both at least
overnight.
The decay or ingrowth factor D is a correction factor for decay or ingrowth of
the radionuclide before and during counting. For the simple situation where the
radionuclide of interest is not supported by a parent in the sample, the decay factor
is given by

1 − e−λtR
D = e−λtD × (10.2)
λtR
Here λ denotes the radionuclide decay constant,tD denotes the time from sample
collection to the start of sample counting, and tR denotes the real time, or clock
time, of the sample counting measurement, which may be longer than the live
time tG . When the time tR is short relative to the half-life of the radionuclide, the
product λtR may be very small, and as λtR approaches zero, the correction factor
for decay during counting, (1 − e−λtR )/λtR , approaches 1. If the value of this factor
is sufficiently close to 1, it may be omitted from the expression for D. When it is
included, it must be calculated carefully to avoid large round off errors when λtR
is small.

Note: For small values of λtR , the factor (1 − e−λtR )/λtR may be approximated
either by e−λtR /2 or by the sum of the first few terms of the series 1 − λtR / 2! +
(λtR )2 / 3! − (λtR )3 / 4! + · · ·.

The model for a gamma-ray spectral analysis is somewhat more complicated

than suggested by Eq. (10.1), and in the worst cases may be much more compli-
cated. In gamma-ray spectrometry, the gross count CG equals the sum of counts in
channels that represent a photopeak. The radionuclide being measured may or may
not be present in the instrument background (i.e., as a peak in the background spec-
trum), but the total background correction always includes a contribution from the
Compton baseline of the spectrum, which is usually estimated from the observed
counts in several channels on each side of the peak. The shape of the baseline
under the peak is often assumed to be a straight line, although stepped functions
are sometimes used. If the radionuclide being measured emits measurable gamma
rays at more than one energy, the activity is typically calculated as a weighted av-
erage of the results obtained for selected peaks. In some cases, gamma rays from
two or more radionuclides may have energies so nearly equal that the associated
peaks in the spectrum overlap and combine to form multiplets. Such individual
peaks often can be resolved by nonlinear regression. For all these reasons, most
laboratories rely on software packages to perform the required data analysis for
gamma-ray spectrometry.
192 Keith D. McCroan and John M. Keller

10.3. Measurement Uncertainty

The uncertainty of each result must be reported to indicate its reliability. This
information is necessary when comparing a measured value with a guidance level
or regulatory limit to state with confidence that the true value is above or below the
level or limit. Measurement uncertainty must also be considered when comparing
results within or between laboratories.
An internationally accepted guide to the subject of measurement uncertainty is
the Guide to the Expression of Uncertainty in Measurement, commonly referred to
as “the GUM” (ISO 1995). The GUM defines standard terminology, notation, and
methodology for evaluating and expressing measurement uncertainty. This chapter
uses the GUMs terms and symbols, some of which are summarized briefly below.
For more information, see the GUM itself (ISO 1995), NIST Technical Note 1297
(NIST 1994), or the Multi-Agency Radiological Laboratory Analytical Protocols
Manual (MARLAP) (EPA 2004).
The uncertainty of a measurement is defined in the GUM and in the International
Vocabulary of Basic and General Terms in Metrology (ISO 1993) as a “parameter,
associated with the result of a measurement, that characterizes the dispersion of the
values that could reasonably be attributed to the measurand.” Thus, the uncertainty
may be used to find bounds for the likely value of the measurand. The traditional
approach to uncertainty in radioanalytical chemistry, which is also the GUMs ap-
proach, is to express the uncertainty of a result as either an estimated standard devia-
tion or a multiple thereof. An uncertainty expressed as a standard deviation is called
a standard uncertainty. The same concept has traditionally been called a “one-
sigma uncertainty.” A multiple of the standard uncertainty, such as the “two-sigma”
or “three-sigma” uncertainty, is called an expanded uncertainty. A laboratory may
report the uncertainty of a result as either the combined standard uncertainty or an
expanded uncertainty, but the type of uncertainty should always be explained.
Radioanalytical measurements can have many causes of uncertainty. One of the
best known is the inherent randomness of nuclear decay, radiation emission, and
radiation detection, which is often referred to as “counting statistics.” Radiation
laboratories generally recognize the existence of counting statistics and account
for it when they report the uncertainty of a measurement. The component of the
total uncertainty due to counting statistics is the counting uncertainty, which has
traditionally been called the “counting error.”
In some situations, other components of the total uncertainty may be as large as
or larger than the counting uncertainty. A laboratory that reports only the count-
ing uncertainty may give data users an unrealistic view of the data’s reliability.
All known sources of uncertainty and all the uncertainty components should be
evaluated if they are believed to be significant.

10.3.1. Uncertainty Evaluation and Propagation

In radioanalytical chemistry measurements, as in many analytical measurements,
the final result is not obtained by direct observation of a measuring instrument.
 10. Data Calculation, Analysis, and Reporting 193

Instead, one uses a mathematical model of the measurement to relate the values of
input quantities, which are observed or previously measured, to the value of the
desired output quantity (the measurand), which must be calculated. The mathe-
matical model is an equation or a set of equations that describe exactly how one
calculates the value of the measurand from the observed values of the input quan-
tities. The model for a radioanalytical measurement often resembles Eq. (10.1),
but for the purpose of explaining uncertainty evaluation, a model that follows the
GUM is written abstractly in the form
Y = f (X 1 , X 2 , . . . , X N ) (10.3)
where X 1 , X 2 , . . . , X N denote the input quantities, and Y denotes the output quan-
tity (not the yield in this case), or measurand.
A particular observed value of an input quantity X i is called an input estimate.
If we denote the input estimates by lowercase variables x1 , x2 , . . . , x N , then the
output estimate, y, is calculated as follows:
y = f (x1 , x2 , . . . , x N ) (10.4)
Each input estimate xi has an uncertainty, u(xi ). In principle the standard uncer-
tainty of an input estimate can be evaluated in many ways. For example, if the value
of the input quantity can be measured repeatedly, then a series of observations of
the quantity, xi,1 , xi,2 , . . . , xi,n , can be obtained to calculate the arithmetic mean
x̄i and experimental standard deviation s(xi,k ):


1 n  1  n
x̄i = xi,k and s(xi,k ) =  (xi,k − x̄i )2 (10.5)
n k=1 n − 1 k=1

The experimental
√ standard deviation of the mean s(x̄i ) is obtained by dividing
s(xi,k ) by n:


s(xi,k )  1 n
s(x̄i ) = √ =  (xi,k − x̄i )2 (10.6)
n n(n − 1) k=1

One then lets xi equal the arithmetic mean of the observations and lets u(xi ) be
the experimental standard deviation of the mean:
xi = x̄i
u(xi ) = s(x̄i ) (10.7)
This type of uncertainty evaluation is an example of what the GUM calls a Type A
evaluation of uncertainty, which is defined as an evaluation of uncertainty based
on the statistical analysis of series of observations (ISO 1995).
Many methods of uncertainty evaluation do not involve the statistical analysis
of series of observations; these are called Type B evaluations. One of the most
common Type B methods of uncertainty evaluation in radioanalytical chemistry is
the common practice√of estimating the standard uncertainty of an observed count
C by its square root C (see Section 10.3.4).
194 Keith D. McCroan and John M. Keller

Another way to perform a Type B evaluation is to estimate the maximum possible

absolute error b in an input estimate xi and to assume that the distribution of the
estimator is either rectangular or triangular, with half-width equal to b. To evaluate
the standard deviation of the estimator, if the distribution is rectangular with half- √
width b, the standard deviation (i.e., the standard uncertainty of√xi ) equals b/ 3.
If the distribution is triangular, the standard deviation equals b/ 6.
A rectangular distribution is a more conservative assumption than a triangular
distribution. A rectangular distribution is appropriate if one believes that the true
value lies in the interval xi ± b, and all possible values in this interval are equally
likely. A triangular distribution is more appropriate if one believes that values near
the center of the interval are more likely than those near the edges. For example,
a rectangular distribution is ideal for evaluating the uncertainty due to rounding
a number. A triangular distribution may be used to evaluate the uncertainty as-
sociated with the capacity of volumetric glassware or pipetting devices, where
the manufacturer states a tolerance for the true value but the nominal capacity is
assumed to be most likely.
In both of the preceding examples, the value of the half-width b is usually given.
In other cases, the determination of b may require experience and professional
judgment. For example, one would use judgment to estimate the maximum possible
deviation of a meniscus from the capacity mark in a volumetric flask.
Sometimes a pair of input estimates xi and x j may be correlated, because they
are not determined independently of each other or because there is some effect in
the measurement process that influences the observed value of each. The estimated
covariance of xi and x j is denoted by u(xi , x j ). A Type A evaluation of covariance
may be performed in some cases by making a series of paired observations of the
two input quantities, (xi,1 , x j,1 ), (xi,2 , x j,2 ), . . . , (xi,n , x j,n ), and performing the
following calculations:

1 n
xi = x̄i = xi,k
n k=1
1 n
x j = x̄ j = x j,k
n k=1
1 n
u(xi , x j ) = (xi,k − x̄i )(x j,k − x̄ j ) (10.8)
n(n − 1) k=1

Other possible methods for estimating the covariance u(xi , x j ) are described by
MARLAP (EPA 2004).
All the uncertainties u(xi ) and covariances u(xi , x j ) of the input estimates com-
bine to produce the total uncertainty of the output estimate, y. The mathematical
operation of combining the standard uncertainties and covariances of the input
estimates xi to obtain the standard uncertainty of the output estimate y is called
propagation of uncertainty. The standard uncertainty of y obtained by uncer-
tainty propagation is called the combined standard uncertainty of y and is de-
noted by u c (y). The following general equation, which the GUM calls the “law of
 10. Data Calculation, Analysis, and Reporting 195

propagation of uncertainty,” describes how the combined standard uncertainty is

calculated:

 N  
 ∂ f 2 
N −1 N
∂f ∂f
u c (y) =  u 2 (xi ) + 2 u(xi , x j ) (10.9)
i=1
∂ x i i=1 j=i+1
∂ xi ∂ x j

In this equation, ∂ f /∂ xi , called a sensitivity coefficient, denotes the partial deriva-

tive of the function f (X 1 , X 2 , . . . , X N ) with respect to X i , evaluated at X 1 = x1 ,
X 2 = x2 , . . . , X N = xn . A sensitivity coefficient ∂ f /∂ xi represents the ratio of
the change in the output estimate y to a small change in one input estimate
xi .
If the input estimates are uncorrelated, then all the “covariance terms,” which are
those terms that involve u(xi , x j ), are zero, and the uncertainty equation reduces
to the simpler form shown in Eq. (10.10) below:

 N 
  ∂ f 2
u c (y) =  u 2 (xi ) (10.10)
i=1
∂ x i

The following applications show how Eq. (10.10) can be applied to some simple
models when the input estimates are uncorrelated. In these equations, the variable
xi denotes input estimates and the variable y denotes the calculated output estimate;
a and b are constants.
Application 1: Addition and Subtraction
y = x1 ± x2

u c (y) = u 2 (x1 ) + u 2 (x2 )
Application 2: Multiplication
y = x1 x2

u c (y) u 2 (x1 ) u 2 (x2 )
u c (y) = x22 u 2 (x1 ) + x12 u 2 (x2 ) = +
y x12 x22
Application 3: Division
x1
y=
x2

u 2 (x1 ) x12 u 2 (x2 ) u c (y) u 2 (x1 ) u 2 (x2 )

u c (y) = + = +
x22 x24 y x12 x22
Application 4: Exponential
y = aebx
u c (y)
u c (y) = abebx u(x) = b u(x)
y
196 Keith D. McCroan and John M. Keller

+ +2 +3

FIGURE 10.1. A normal distribution.

10.3.2. Expanded Uncertainty

Most laboratories multiply the combined standard uncertainty u c (y) by a factor k,
which is typically a value such as 1.96 or 2. The laboratory then reports the product
of k and u c (y) as the uncertainty of the result y. The purpose of expanding the
uncertainty in this manner is to obtain an interval about the measured result y ±
k u c (y) that has a high probability of containing the true value of the measurand.
When k = 1.96 or 2, this product has traditionally been called the “two-sigma”
uncertainty. The GUM calls the product k × u c (y) an expanded uncertainty and
calls the factor k itself a coverage factor.
The common choice of k = 1.96 or 2 is based on the assumption that the
distribution of the measured result is approximately normal, or Gaussian. If the
distribution is normal and centered on the true value of the measurand, there is
approximately a 95% probability that the measured result will fall within 1.960
standard deviations of the true value.
Figure 10.1 shows a normal distribution with intervals about the mean μ repre-
sented in terms of multiples of the standard deviation σ . The area under the curve
between μ – σ and μ + σ is approximately 68.3% of the total area, and between
μ – 2σ and μ + 2σ it is approximately 95.4%. More than 99.7% of the area falls
between μ – 3σ and μ + 3σ .
The combined standard uncertainty is only an estimate of the true standard
deviation, rather like an experimental standard deviation, and for this reason the
probability that the expanded uncertainty interval y ± 1.96 u c (y) will contain the
true value of the measurand may actually be less than 95%. In fact it is misleading
for a laboratory to use as precise a coverage factor as k = 1.96, which implies that
the true standard deviation is known and that the coverage probability really is
95%. The use of k = 2 better indicates the approximate nature of the uncertainty
evaluation and does not imply so strongly that the coverage probability is exactly
95%.

Note: Although it is common to use a fixed coverage factor, such as 2, to calculate

the expanded uncertainty, the GUM also describes a method for calculating a cov-
erage factor k p for each measurement to provide a desired approximate coverage
probability p. The method is derived by treating the standard uncertainty of each
 10. Data Calculation, Analysis, and Reporting 197

input estimate u(xi ) as if it were an experimental standard deviation with a specified

number of degrees of freedom ν i . When u(xi ) is determined by a Type A method of
evaluation, ν i is the actual number of degrees of freedom for the evaluation. When
u(xi ) is determined by a Type B method of evaluation, the value of ν i is based on
the estimated relative standard deviation of u(xi ) (i.e., the uncertainty of the uncer-
tainty). Then the number of “effective” degrees of freedom ν eff for the combined
standard uncertainty u c (y) is evaluated. Finally, the coverage factor k p is chosen
from a table of percentiles of Student’s t-distribution with ν eff degrees of freedom.
For example, if the number of effective degrees of freedom is 12, then the coverage
factor k0.95 for 95% coverage is the 97.5% percentile of the t-distribution with 12
degrees of freedom, which equals 2.179. For more information about this method
of calculating coverage factors, see the GUM (ISO 1995) or MARLAP (EPA 2004).

10.3.3. Application to Radioanalytical Chemistry

A typical mathematical model for a simple radioanalytical measurement might
have the form of Eq. (10.1), which is shown again below:
(CG /tG ) − (CB /tB ) RG − RB
a= = (10.1)
ε×Y ×m× D× F ε×Y ×m× D× F
where all the symbols are defined in Section 10.2. If there are no correlations
between any of the input estimates, the law of propagation of uncertainty (with
some algebraic manipulation) implies that
  
u 2 (CG )/tG2 + RG2 u 2 (tG )/tG2 + u 2 (CB )/tB2 + RB2 u 2 (tB )/tB2
u c (a) =
ε2 Y 2 m 2 D 2 F 2
 2 
u (ε) u 2 (Y ) u 2 (m) u 2 (D) u 2 (F) 1/2
+ a2 × + + + + (10.11)
ε2 Y2 m2 D2 F2

The uncertainty of the input quantities in this model is discussed below in Section
10.3.8. Some quantities, such as the decay factor D and the yield Y , are not directly
observed but are calculated from other observed quantities. Their uncertainties are
calculated by uncertainty propagation before being used in the equation above.
The uncertainties of the count times, u(tG ) and u(tB ), and the uncertainty of the
decay factor, u(D), usually are negligible and can be omitted from the uncertainty
equation. Then the simplified uncertainty equation becomes

 2 
u 2 (CG )/tG2 +u 2 (CB )/tB2 u (ε) u 2 (Y ) u 2 (m) u 2 (F)
u c (a) = +a ×
2 + 2 + +
ε2 Y 2 m 2 D 2 F 2 ε2 Y m2 F2
(10.12)
The uncertainty of a count time may be relatively significant if the count time is
very short (say less than 1 min), if the number of counts is very large, or if the dead
time rate is large. The uncertainty of the decay factor is most significant when the
radionuclide is short-lived and the decay time is either uncertain or long.
198 Keith D. McCroan and John M. Keller

The preceding uncertainty equations presume that all pairs of input estimates are
uncorrelated, which may or may not be true. One of the most common examples of
correlated input estimates in radioanalytical chemistry occurs when the chemical
yield Y is calculated from an equation involving the counting efficiency ε. This
happens in measurements by alpha-particle spectrometry with an isotopic tracer. In
this case, the uncertainty equation can be simplified by treating the product ε × Y
as a single variable. What happens in effect is that the efficiency cancels out of the
activity equation and for this reason its uncertainty can be considered to be zero:

 2 
u 2 (CG )/tG2 + u 2 (CB )/tB2 u (ε × Y ) u 2 (m) u 2 (F)
u c (a) = +a ×
2 + +
ε2 Y 2 m 2 D 2 F 2 (ε × Y )2 m2 F2
(10.13)
Another example of correlated input estimates occurs when both the counting
efficiency ε and the yield Y depend on the mass of a precipitate or residue on the
prepared sample source. In this case, dealing with the correlation is less simple. It
may be necessary to replace the variables ε and Y in the activity equation by the
expressions used to calculate them, or to include the covariance term for ε and Y
in the uncertainty equation.

10.3.4. Counting Statistics

Nuclear decay, the emission of radiation from a decaying atom, and the detection
of emitted radiation by a detector are inherently random phenomena. Their occur-
rences cannot be predicted with certainty, even in principle, although they can be
described probabilistically. The randomness of these processes causes the result
of a radiation-counting measurement to vary when the measurement is repeated
and thus leads to an uncertainty in the result, called the “counting uncertainty.”
Suppose one places a radioactive source in a radiation counter and acquires
counts for a period of time t, and the total number of counts observed in this time
period is C. A common Type B method of evaluating
√ the standard uncertainty of
C is to take its square root. That is, u(C) = C. The rationale for this approach,
and some of its limitations, are described below.
One option for evaluating the uncertainty of the number of counts observed is
to perform a Type A evaluation described in Section 10.3.1, where one repeats
the measurement n times, obtaining the values C1 , C2 , . . . , Cn , and calculates the
arithmetic mean C̄, the experimental standard deviation of the values s(Ci ), and
the experimental standard deviation of the mean s(C̄). The arithmetic mean C̄ is
then used as the measured value and the experimental standard deviation of the
mean s(C̄) is used as its standard uncertainty.
When one performs repeated measurements of this type with a long-lived ra-
dioactive source, what is often observed is that the experimental standard deviation
s(Ci ) is approximately equal to the square root of C̄. One can explain this obser-
vation by probability theory if one makes the following assumptions about the
measurement problem:
 10. Data Calculation, Analysis, and Reporting 199

r The source in the detector contains a large number, N , of atoms of some long-
lived radionuclide.
r Each atom of the radionuclide has the same probability p of decaying during the
counting period, emitting the radiation of interest, and producing a count. When
the half-life of the radionuclide is long, p is a very small number.
r All atoms of the radionuclide decay and produce counts independently of each
other. Decay by one atom that produces a count has no impact on whether another
atom decays and produces a count.
r No atom produces more than one count during the counting period.
r Counts are not produced by other sources.

Given these assumptions, the distribution of the observed total number of counts
according to probability theory should be binomial with parameters N and p.
Because p is so small, this binomial distribution is approximated very well by the
Poisson distribution
√ with parameter Np, which has a mean of Np, and a standard
deviation of N p. The mean and variance of a Poisson distribution are numerically
equal; so, a single counting measurement provides an estimate of the mean of
the distribution Np and its square root is an estimate of the standard deviation
√
N p. When this Poisson approximation is valid, one may estimate the standard
uncertainty of the counting measurement without repeating the measurement (a
Type B evaluation of uncertainty).
Although all the assumptions stated above are needed to ensure that the distri-
bution of the total count is binomial, not all the assumptions are needed to ensure
that the Poisson approximation is valid. In particular, if the source contains several
long-lived radionuclides, or if long-lived radionuclides are present in the back-
ground, but all atoms decay and produce counts independently of each other, and
no atom can produce more than one count, then the Poisson approximation is still
useful, and the standard deviation of the total count is approximately the square
root of the mean.

Note: One does not evaluate the uncertainty of a count rate, R = C/t, by taking the
square root of R. If the total count C has a Poisson distribution √
and t has √
negligible
uncertainty, the standard uncertainty of R is given instead by C/t or R/t.

The Poisson approximation is not always valid. For example, if the half-life of
one or more radionuclides in a source is short relative to the count time t, the Pois-
son distribution may not be a good approximation of the binomial distribution. Also
note that some radiation-counting measurements involve the counting of particles
emitted by a radionuclide and its short-lived progeny, where one atom may produce
several counts as it decays through a series of short-lived states. A well-known
example of this type of measurement is alpha-counting 222 Rn and its progeny in
an alpha scintillation cell, or “Lucas cell.” For such measurements, neither the
binomial nor Poisson model is valid. Another example is the use of the Poisson
model to estimate the total uncertainties of gross and background counts measured
by gamma-ray spectrometry. While the use of the Poisson model is applicable in
this case, there are some complications to consider (see Sections 10.2 and 10.3.7).
200 Keith D. McCroan and John M. Keller

10.3.5. The Poisson Distribution

If X is a random variable with a Poisson distribution, and μ denotes its mean,
then the following equation describes the probability, Pr(X = x), of observing
any particular value x (for non-negative integers x):

μx e−μ
Pr(X = x) = (10.14)
x!
When the mean μ is small, the distribution of values about μ is skewed because
Pr(X = x) cannot be negative. When μ is large—say 20 or greater—the Poisson
distribution approaches a Gaussian distribution like the one shown in Fig. 10.1 (but
√
with the standard deviation σ = μ). The probability function for the Gaussian
approximation of the Poisson distribution is shown in Eq. (10.15):
1 (x−μ)2
Pr(X = x) = √ e− 2μ (10.15)
2πμ
Figure 10.2 shows how the Poisson distribution approaches the Gaussian ap-
proximation as the mean μ increases. In the figure, the height of a bar over a
value x represents the actual Poisson probability p(x) of observing that value,
while the height of the smooth curve over the point x represents the Gaussian
approximation of this probability. When μ is only 3, it is clear that the Pois-
son distribution is skewed and the Gaussian approximation does not describe it
well. When μ = 10, the distribution is more symmetric and superficially Gaus-
sian in shape. By the time μ reaches 20, the Gaussian approximation is good
enough for many purposes, and the approximation only improves as μ increases
further.

p(x) p(x)
=3 = 10

x x
0 1 2 3 4 5 6 7 8 9 10 11 12 0 2 4 6 8 10 12 14 16 18 20 22 24
p(x) p(x)
= 20 = 36

x x
0 4 8 12 16 20 24 28 32 36 40 44 48 0 6 12 18 24 30 36 42 48 54 60 66 72

FIGURE 10.2. Poisson probability functions with Gaussian approximations.

 10. Data Calculation, Analysis, and Reporting 201

10.3.6. Statistical Test for a Poisson Distribution

Many common mathematical formulas used for calculations of uncertainty, detec-
tion capability (see Section 10.4), and quality control in the radioanalytical chem-
istry laboratory are based on the assumption of Poisson counting statistics and are
valid only if that assumption is valid. In particular, the formulas depend on the fact
that the mean and variance of a Poisson distribution are numerically equal. For
this reason it may be a good idea to test the Poisson assumption from time to time.
When a statistical population has a distribution that is approximately normal, a
chi-square test can be performed on a random sample from the population to check
its variance. Since the Poisson distribution is approximately normal whenever its
mean is large enough (e.g., 20 or more), the chi-square test can be adapted to check
whether the mean of a set of counting data equals its variance. For example, if a
long-lived source is counted n times with the same detector to generate the counts
C1 , C2 , . . . , Cn , then one may calculate the chi-square statistic given by Eq. (10.16):
1  n
χ2 = (Ci − C̄)2 (10.16)
C̄ i=1
Under the hypothesis that the data distribution is Poisson, this statistic has
(approximately) the chi-square distribution with n – 1 degrees of freedom; so,
one can test the hypothesis by comparing χ 2 to appropriate percentiles of that
distribution. Chi-square tables are commonly given in books and handbooks about
statistics; many are also found on the Internet, though you must be careful to choose
a reliable source. Generally, one does not expect the variance of the data to be less
than the mean. The question in most instances is whether the variance is larger. So,
one usually performs a one-sided test rather than a two-sided test. For example, if
n = 10, one might compare the calculated value of χ 2 to the 95th percentile of the
chi-square distribution with nine degrees of freedom, which equals 16.9. Then, if
χ 2 > 16.9, one would conclude that the distribution was not Poisson and that the
variance was larger than the mean.
Note: It is appropriate to choose the 95th percentile for a one-sided test at the 5%
level of significance, or the 2.5th and 97.5th percentiles for a two-sided test. Other
significance levels are possible, but 5% is common.
Although a one-sided test is most common, a two-sided version of the test might
be used in some situations. In this case one would compare the value of χ 2 to the
2.5th percentile and the 97.5th percentile of the chi-square distribution with (n – 1)
degrees of freedom. For example, if n = 10, these percentiles are 2.7 and 19.0.
Any value of χ 2 less than 2.7 or greater than 19.0 indicates that the distribution is
not Poisson.

10.3.7. Spectrometry
Uncertainty evaluation for gamma-ray spectral analysis is more complicated
than for most other radiation measurements. First, gamma-ray counts for one
202 Keith D. McCroan and John M. Keller

radionuclide are observed in several channels of a multichannel analyzer and form

a more-or-less Gaussian photopeak that must be identified and integrated. Second,
the counts estimated to be associated with the Compton baseline under the peak
must be subtracted from the total peak counts, as discussed in Section 10.2. Third,
any gamma-ray-emitting radionuclides of interest that are present in the detector
background must be subtracted from the peak counts. Fourth, mathematical tech-
niques may have to be applied to resolve peaks from different gamma-ray emitters
that overlap to form multiplets. Each of these issues affects the uncertainty evalua-
tion. At times, other complications include summing, escape peaks, and Compton
edges. On the other hand, observation of a peak yields additional information by
confirming the presence of the specific radiation, compared to the simple use of
counts to calculate the uncertainty discussed in Section 10.3.4.
The problem of uncertainty evaluation for background corrections in alpha-
particle spectrometry is that the background count rate may be extremely low.
Counting for several days may yield only a few, or even zero, counts. At least two
issues may arise. First, the Poisson uncertainty estimate is relatively imprecise
when the mean count is so low. In the worst case, both the background count CB
and the gross count CG are zero, so that the uncertainty equation will calculate
zero uncertainty for the measured activity, which is inappropriate for a laboratory
√ it is possible to observe√zero counts (C = 0), the
to report. In any situation where
Poisson uncertainty estimator C should be replaced by C + 1. Low-level alpha
particle spectrometry is a situation where zero counts may indeed be observed.
Second, it becomes important to consider possible variability in the background
(e.g., from spurious counts), which may be relatively large in comparison to the
Poisson uncertainty.

10.3.8. Other Components of Uncertainty

The many other possible sources of uncertainty in a radioanalytical measurement
include:
r Measurements of masses and volumes
r Variable instrument background
r Dead time
r Tracer activity
r Calibration
r Contaminants in tracers and reagents
r Values of physical constants such as radionuclide half-lives and radiation emis-
sion probabilities
r Laboratory subsampling of heterogeneous material

MARLAP (EPA 2004) describes methods for evaluating the uncertainties of

masses and volumes, but it also points out that these components of uncertainty
tend to be negligible. Generally these uncertainties can be neglected if balances
and volumetric devices are maintained and used properly.
 10. Data Calculation, Analysis, and Reporting 203

The uncertainty due to varying instrument background can be significant. If the

background varies, the assumption of pure Poisson counting statistics to evaluate
the uncertainty of the net count rate for a sample may seriously underestimate
the uncertainty. Options include replicate background measurements to determine
its uncertainty (Type A evaluation), or to evaluate an additional component of
uncertainty to be added to the Poisson counting uncertainty.
Uncertainty due to dead time can be avoided by keeping the dead time rate low
(e.g., less than 10%). If the dead time rate is not negligible, the gross count rate
must be calculated for the live time rather than the clock time (real time); if the
dead time rate is not excessive, no additional uncertainty component for dead time
is usually needed.
The uncertainty due to the concentration of a tracer used to measure the chemical
yield may be significant. It should be evaluated and propagated when determining
the uncertainty of the yield.
Calibration uncertainty should be evaluated and propagated. The causes of cali-
bration uncertainty include uncertainties in the reported values of calibration stan-
dards, measurement uncertainty in the laboratory such as counting statistics, and
in processing standards.
Significant contamination in tracers and reagents should be measured and cor-
rected for if it cannot be eliminated. Corrections for contaminants may introduce
uncertainty that should be propagated when calculating the combined standard un-
certainty of the sample activity. Options for assessing such contamination include
direct measurement of the contaminant’s concentration in particular solutions, or
replicate measurements of reagent blank samples.
Uncertainties in physical constants may or may not be significant, depending on
the measurement. Half-lives in particular tend to be well known and usually con-
tribute only negligible uncertainty to the result of a radioanalytical measurement.
Errors in radiation emission probabilities may contribute additional uncertainty
to some measurements. Uncertainties for both categories are available from ta-
bles of isotopes and the public Web site of the National Nuclear Data Center at
Brookhaven National Laboratory.
Laboratory subsampling of heterogeneous material may contribute significant
uncertainty to the result of the measurement, especially if the sample contains tiny
high-activity particles, i.e., “hot particles.” Currently, the most fully developed the-
ory of sampling errors is that of the French geologist Dr. Pierre Gy. The theory ex-
plains that the relative uncertainty due to subsampling decreases if one either takes
a larger subsample or grinds the sample to reduce particle sizes before subsampling.
MARLAP summarizes the simpler aspects of Gy’s sampling theory. It suggests a
simple equation that can be used by default to evaluate the relative standard uncer-
tainty due to subsampling particulate solid material, such as soil, when “hot parti-
cles” are not suspected and a reasonable effort has been made to homogenize the
sample before subsampling (EPA 2004). The equation suggested by MARLAP is

 
1 1
ur = − × kd 3 (10.17)
mS mL
204 Keith D. McCroan and John M. Keller

where
ur denotes the relative standard uncertainty due to subsampling
mS is the mass of the subsample (aliquot)
mL is the mass of the entire sample
k is, by default, 0.4 g cm−3
d is the maximum particle diameter (size of a square mesh), in cm
For example, if a very large sample (m L = ∞) is ground until it passes through
a sieve with mesh size d = 0.1 cm, and a subsample of mass m S = 0.25 g is
removed for analysis, this equation estimates the relative subsampling uncertainty
to be u r = 0.04, i.e., 4%.

10.4. Radionuclide Detection

One of the most controversial topics in analytical and radioanalytical chemistry
involves measures of detection capability and the methods used to decide whether
an analyte is present in a laboratory sample. Despite the controversy and confu-
sion, the basic theory is simple. The only difficulties that ought to exist are those
associated with the application of the theory.
The term detection capability refers to the ability of a measurement process to
detect the analyte, where to “detect” the analyte means to decide, based on the
result of an analysis, that the analyte is present in the sample. Detection capability
is often called “sensitivity,” although the term sensitivity is also commonly used
to mean the ratio of the change in an output quantity to the change in an input
quantity (e.g., the slope of a calibration curve). The latter meaning of sensitivity
is the one implied by the term “sensitivity coefficient” in Section 10.3.1.
The laboratory may often be required to make detection decisions about samples,
but when the analyte activity is low enough, the relative uncertainty in the result
may make it difficult to distinguish between a small positive activity and zero. The
performance characteristic of the measurement process that describes its detection
capability is called the minimum detectable value, minimum detectable activity,
minimum detectable concentration (MDC), or lower limit of detection (LLD).
These terms have been used to denote the theoretical concept of the smallest
true value of the analyte in a sample that gives a specified high probability of
detection.

10.4.1. The Critical Value and the MDA

The common approach to detection decisions in radioanalytical chemistry is based
on statistical hypothesis testing. In a hypothesis test, one formulates two mutually
exclusive hypotheses, called the null hypothesis and the alternative hypothesis,
and uses the data to choose between them. The null hypothesis is presumed to be
true unless there is strong evidence to the contrary. When such evidence is present,
the null hypothesis is rejected and the alternative hypothesis is accepted.
 10. Data Calculation, Analysis, and Reporting 205

Typically in radioanalytical chemistry, the null hypothesis is the hypothesis

that no analyte is in the sample. Even if no analyte is present, the net result
of the measurement has uncertainty, and, if the measurement were repeated a
number of times, a distribution of results about zero, including both positive
and negative values, should be observed. Although results near zero are most
likely, in principle there is no upper or lower bound for what the result might
be. Observation of a positive result in a single measurement does not necessar-
ily constitute strong evidence that the analyte is present. The result must exceed
some positive threshold value, called the critical value, to lead one to conclude
that the analyte is really present. The question is: how to determine the critical
value?
Before choosing the critical value, one specifies one’s tolerance for Type I errors,
defined as erroneous rejection of the null hypothesis when it is true. Type I errors
are sometimes called false positives or false rejections. Regardless of the choice
of the critical value, there is always the probability of a Type I error. For large
enough critical values, this probability is small and generally can be tolerated. The
tolerable probability that one specifies is called the significance level of the test and
is usually denoted by α. In radioanalytical chemistry, it is common to set α = 0.05.
If α = 0.05, then analyte-free samples should produce false positive results at a
rate of only about one per twenty measurements.
The MDA for a radioanalytical measurement process is defined as an estimate
of the smallest true activity (or massic activity or volumic activity) in a sample
with a specified probability 1 – β that the net count rate will exceed the critical
value, to conclude correctly that the sample contains the analyte. The probability
β of a Type II error is defined as the error of failing to reject the null hypothesis
when it is actually false. A Type II error is also called a false negative or false
acceptance. It is common to choose α = β = 0.05, but other values of α and β
can be selected.

10.4.2. Calculation of the Critical Value and MDA

In radioanalytical chemistry, the critical value can be calculated by either of two
common approaches. One is based on repeated measurements of blank samples,
and the other on the assumption of Poisson counting statistics. The former method
is generally applicable to any measurement process for which the distribution
of measurement results is approximately normal. The latter method is useful
when the Poisson assumption is valid, but it may give misleading results in other
situations.
Suppose the laboratory performs measurements of n blank samples, where the
blanks simulate real analyte-free samples as nearly as possible and the results
B1 ,B2 , . . . ,Bn are expressed as yield-corrected activity. Then the laboratory may
calculate the arithmetic mean B̄ and the experimental standard deviation s(Bi ) of
the values (see Eq. 10.5), and it may blank-correct the absolute activity found in
a subsequent measurement of a real sample by subtracting the mean B̄ from it. In
this case the critical value L C for the net absolute activity is calculated as shown
206 Keith D. McCroan and John M. Keller

in Eq. (10.18):

1
L C = t1−α (n − 1) × s(Bi ) × 1+ (10.18)
n
In this equation t1−α (n − 1) denotes the (1 − α)-quantile of the Student’s
t-distribution with n − 1 degrees of freedom. For example, if α = 0.05 and n = 10,
then t1−α (n − 1) = t0.95 (9) = 1.833.
If one performs the series of blank measurements described above, calculates
L C by Eq. (10.18), and subsequently calculates the blank-corrected activity A for
an analyte-free sample, the probability of observing a value of A greater than L C
is approximately equal to α. If one makes the detection decision by comparing A
to L C , the false positive rate should be approximately α.
If α = β and the number of blank measurements n is not too small (say
at least 5), and if the measurement variance does not increase rapidly with activity,
the minimum detectable absolute activity L D may be approximated as shown in
Eq. (10.19):
LD = 2 × LC (10.19)
Typically, one expresses the MDA as a massic or volumic activity by dividing
L D by the amount of sample analyzed. If α = β or if variance increases too rapidly
with activity, Eq. (10.19) is inappropriate. For these cases see MARLAP (EPA
2004).
When the radiation counting statistics are essentially Poisson for the background
count and there are no interferences, the critical net count rate (critical value of
the net count rate) may be calculated as follows:

 
CB 1 1
SC = z 1−α + (10.20)
tB tG tB
where
SC is the critical net count rate in s−1
CB is the observed background count
tB is the background count time in s
tG is the sample (gross) count time in s
α is the significance level (e.g., α = 0.05 by default)
z 1−α is the (1 − α)-quantile of the standard normal distribution (equal to
1.645 when α = 0.05).
In this case one makes the detection decision for a sample by comparing the
observed net count rate RN to the critical net count rate SC .
When Eq. (10.20) is used for the critical value and α = β, a commonly used
approximation formula for the MDA is

 
z2 1 1
+ 2z RB +
tG tG tB
MDA = (10.21)
K
 10. Data Calculation, Analysis, and Reporting 207

where
z = z 1−α = z 1−β
RB is the background count rate in s
tG is the sample (gross) count time
tB is the background count time
K is an appropriate estimate of the denominator by which the net count
rate is divided to calculate the final result of the analysis.
The denominator represented by K above might, for example, be the product
ε × Y × m × D × F, which appeared in the previous example of a mathematical
model for a radioanalytical measurement (Eq. 10.1). If there is significant vari-
ability in this denominator, a somewhat low estimate of its value should be used
to avoid underestimating the MDA.
In many cases, the assumption of pure Poisson counting statistics is invalid
because other sources of variability affect the distribution of measurement results.
In these cases, the laboratory should consider determining the critical value and
MDA based on a series of blank samples, as in Eqs. (10.18) and (10.19). When
the Poisson model is valid but the background level is so low that the Poisson
distribution is not approximately normal, other formulas for the critical value and
MDA tend to give the best results, as discussed in MARLAP (EPA 2004).
Note: Radioanalytical chemists should avoid the common mistake of using the
MDA as a critical value.

10.5. Calculations for Quality Control

The quality of measurements in the radioanalytical chemistry laboratory is eval-
uated by QC measures for both instrumental measurements and sample analysis,
as outlined in Section 11.2.8. QC ensures that the measurement process remains
in a state of statistical control so that uncertainty estimates are valid, and ideally
should help to keep measurement uncertainties small. QC may also provide assur-
ances that the quality of measurements does not decrease when there are changes
in personnel, instrumentation, or methods.
Initial QC is devoted to assuring that analytical methods and detection devices
are suitable for the purpose of the program and includes estimating the uncertainty
of analytical results. Subsequent methodological QC efforts during data processing
evaluate the acceptability of the results and add information on the degree of
reliability.

10.5.1. Instrument Quality Control Calculations

Radiation detection instruments are checked at frequent intervals to assure that
they are operating and responding correctly. To estimate analytical uncertainty and
sensitivity, information must be obtained about the stability of pertinent instrument
208 Keith D. McCroan and John M. Keller

parameters, especially background count rate and counting efficiency. Additional

parameters for spectrometers are energy response and peak resolution.
At installation, all instrument controls (voltage, gain, bias, discriminator, etc.)
must be set according to supplier instructions, and the settings must be recorded for
periodic inspection and consideration of need for future adjustments. Initial tests
should check whether the count rate for a test source and the radiation background
count rate are reasonable and meet purchase specifications. Spectrometers should
be tested with sources that emit radiation at known energies to assure that the
desired range of energies can be observed and that peak shape and resolution are
acceptable relative to purchase specifications.
Once accepted, the instrument must be calibrated for counting efficiency and, if a
spectrometer, for energy response. For the former, the radionuclide standards must
be prepared by the national calibration facility—NIST in the United States—or
by another facility in a manner traceable to NIST (ANSI/IEEE 1995). Standards
used for calibration may be supplied as a point source, an extended source of
the same geometric configuration as the samples that will be counted, or as a
sealed solution which is converted by the user to the desired form (see Section
8.3). A certificate that contains all appropriate information described in Section
11.2.6 must accompany all standards. The typical relative standard uncertainty of
radionuclide standards is in the range of 1–2%.
Handling the standard, placing the source near the detector, recording data,
and processing results all must be meticulous operations because the reliability
of all future calculations of activity from count rate depend on the efficiency
calibration. Accumulated calibration counts should be sufficiently large to achieve
a relative counting uncertainty of 0.01 or less (at least 10,000 counts). To perform
measurements in a brief period (such as 1–10 min), the source usually is prepared
to yield a relatively intense count rate, but not so intense that the count rate is
affected by dead-time losses (see Section 8.2.2) or system contamination becomes
a threat.
For calibrating energy response in terms of kiloelectron volts per channel, any
set of radionuclides can be used, as long as their radiation energies are better
known than the precision of measurement that will be reported for results from the
calibrated instrument. For example, if radiation energies are to be reported to four
significant figures, the calibration radiation should be known with uncertainty in
the fifth significant figure, e.g., 1.1732 MeV for the lower energy 60 Co gamma ray.
Calibrated instruments at specified control settings are checked by measuring
an instrument-control source and the radiation background as described in Section
11.2.9. Initially, both measurements are repeated to establish a mean value and
calculate the standard deviation and its multiples for defining acceptable limits
for operation. A useful device for keeping track of instrument performance is the
control chart shown in Fig. 10.3. The control chart consists of parallel lines at the
mean value and the upper and lower 2σ and 3σ uncertainty values; its x-axis is
the calendar. The lines are horizontal for both the source and the background if the
source rate of radioactive decay is negligible during the time period of the chart.
For a control source that has observable decay during the period of observation,
 10. Data Calculation, Analysis, and Reporting 209

15,000

+3s limit +2s limit Mean

14,500

14,000

Counts
13,500

13,000
s limit s limit

12,500
01/01/2000 12/31/2000 01/01/2002 01/01/2003 01/01/2004 12/31/2004
Date

FIGURE 10.3. Typical radiation detection instrument control chart.

the mean and bounding values for the source must curve downward with the slope
defined by the radioactive decay constant.
The instrument control source is not a standard. Its main criteria are stability—
i.e., no observable change in the radionuclide content (other than by decay) or
configuration—despite frequent handling over the period of use, and known ra-
dioactive constituents. Its radionuclide content is selected to have a count rate that
is high but not excessive, as discussed above, so that a 1- to 10-min counting pe-
riod is sufficient to accumulate about 10,000 counts. The background count usually
requires a much longer time period—50 to 1000 min, depending on the type of
detector—but still will have a larger relative standard deviation than the source.
Hence, the upper and lower bounds for the source control chart will be much closer
together, i.e., have a smaller relative standard deviation, than for the background
control chart.
During routine operation of the instrument, the control source and the back-
ground are measured at selected intervals and the number of counts is entered on
the control chart. This may be done by hand or by computer, as in Fig. 10.3. A
measurement point that falls within the 3σ control limits is acceptable.
Any value outside the 3σ limits indicates that a problem exists and correction is
required. The instrument should be checked for easily corrected problems such as
control settings, computer instructions, applied voltage, radioactive contamination,
ambient radiation levels, temperature changes, dirt, and noise. Points outside of
the 3σ limit may occur, for example, if small errors in the decay correction cause
an upward or downward trend in the data that results in outlying values over time.
If the problem cannot be corrected at this level of trouble shooting, the instrument
is scheduled for repair. Values that are too low generally result from component
failures or a decrease in the applied voltage. Values that are too high often result
210 Keith D. McCroan and John M. Keller

from high ambient radiation levels—notably high radon concentration in counting-

room air—or an increase in the applied voltage.
Normal instrument operation should result in control chart values that lie within
the 2σ limits about 95% of the time and between the 3σ limits more than 99% of
the time if source and background counts are randomly distributed. Nonrandom
distribution requires evaluation of instrument operation, whether the distribution
is too narrow or too wide. Any obvious trend in the distribution so that the center
line no longer represents the mean value should also be evaluated, leading either to
redetermining the mean and standard deviation or considering the problems cited
above.
Instead of the above-cited control chart to test the hypothesis that the measured
values are randomly distributed, a tolerance chart may be established to compare
periodic measurements with established acceptance limits (see ANSI N42.2, Ap-
pendix A). In this system, an acceptable relative standard deviation—say 2% for
the standard source count and 10% for background count—is selected and hori-
zontal lines that represent these values are drawn as upper and lower acceptance
limits. These limits should be wider than the statistically based limits; if they were
narrower, they would be exceeded frequently. They need not be symmetrical on
the positive and negative side. Their purpose is to keep the counting efficiency
and background count rate within limits that give acceptable results even if the
instrument and its output are subject to nonrandom fluctuations.

10.5.2. Analytical Quality Control Calculations

The usual QC sample program for routinely analyzed sample sets consists of
laboratory blanks, laboratory-fortified (or spiked) blanks, and matrix spikes (in
EPA terminology), as described in Section 11.2.9. The laboratory blank is a sample
prepared with laboratory water. The laboratory-fortified blank is laboratory water
with a known amount of radioactive tracer added. This tracer usually is the same
radionuclide that is being analyzed. The matrix spike is a replicate of an actual
sample to which a known amount of radioactive tracer has been added. At selected
intervals, all are processed in the laboratory and counted exactly as are the routinely
analyzed samples.
A number of different test statistics are used for evaluating the results of radio-
analytical chemistry QC analyses. The test statistic for a QC analysis should take
into account the combined standard uncertainty of measurement.
Laboratory blank measurements represent the radiation background count rate
of the samples: the sum of the count rates due to detector background and any
radioactive contaminants in the prepared counting source. Ideally, the latter con-
tributes little to the background count rate. The cause of any significant increase
over the detector background count rate should be investigated and—if possible—
eliminated. The recommended test statistic for a reagent blank analysis is computed
as follows:
A
Z rb = (10.22)
u c (A)
 10. Data Calculation, Analysis, and Reporting 211

Here A is the measured sample activity and u c (A) is its combined standard un-
certainty. Acceptance limits for Z rb may be analogously to 3σ control limits on
a control chart or at other values based on the desired false rejection rate; they
could, for example, be based on the estimated number of effective degrees of free-
dom for the combined standard uncertainty u c (A), described in the note in Section
10.3.2. Thus, the result of the reagent blank analysis A is found to be acceptable
if its absolute value does not exceed a specified multiple of its combined standard
uncertainty u c (A).
Laboratory-fortified blanks and matrix spikes both test the analyst’s ability to
obtain the expected result. The extent to which the net radionuclide concentration
of the fortified blank (corrected for yield and radioactive decay) deviates from the
expected value for the tracer radionuclide concentration is a measure of analytical
bias. Any consistent deviation from the expected value should be investigated to
eliminate the cause. Typical causes are the wrong counting efficiency, an analytical
problem with interchange between carrier and tracer, unreliable yield determina-
tion, or erroneous tracer radionuclide concentration.
The recommended test statistic for a blank spike is computed as follows:
A−K
Z bs =  (10.23)
u 2c (A) + u 2c (K )
Here A is the measured sample activity, K is the added spike activity, and u c (A)
and u c (K ) are the combined standard uncertainties of A and K , respectively.
Acceptance limits for Zbs may be set at value such as 3.
The percent recovery (%R) for a blank spike is computed as in Eq. (10.24):
A
%R = × 100% (10.24)
K
Here Arepresents the measured spiked sample activity and K is the actual activity
of the spike added.
The difference between the activity of the matrix spike and that of the unspiked
matrix should be approximately equal to the activity of the added radioactive tracer.
The recommended test statistic for a matrix spike analysis is computed as in Eq.
(10.25):
aS − aU − aK
Z ms =  (10.25)
u 2c (aS ) + u 2c (aU ) + u 2c (aK )
Here aS is the measured spiked sample result, aU is the unspiked sample result,
aK is the spike added, and u c (aS ), u c (aU ), and u c (aK ) are the estimated standard
deviations in aS , aU , and aK , respectively. All variables are again in the same units.
If required, the percent recovery for a matrix spike is computed as in Eq. (10.26):
aS − aU
%R = × 100% (10.26)
aK
where aS represents the measured spiked sample result, aU the unspiked sample
result, and aK the actual concentration of spike added.
212 Keith D. McCroan and John M. Keller

Any difference in tracer results between the spiked and unspiked blanks and
the spiked and unspiked matrix usually can be attributed to the effect of different
chemical and physical forms of the radionuclide of interest and its tracer. In that
case, the tracer does not represent the radionuclide of interest, and some treatment
of the tracer plus radionuclide of interest, such as a redox process, will be needed
to place them into identical forms. The cause of such inconsistency needs to be
examined and resolved if the difference exceeds the uncertainty of the two values.
Accumulated results from laboratory-fortified blanks and matrix spikes are a
measure of both precision and bias. To determine precision, numerous replicate
values can be examined and the standard deviation can be calculated. The mean
value is compared to the expected tracer concentration to determine bias. For this
application, the results collected at different periods must be seen to belong to the
same set, i.e., there is no obvious temporal difference among measurements due
to analytical or measurement problems. The calculated standard deviation value
can be compared to the suitably propagated components for counting and for the
rest of the analytical process. Causes of unexpectedly large or small values of the
standard deviation should be examined.
To obtain reliable values of the standard deviation, the accumulated tracer-
related count must be sufficiently large, e.g., at least several hundred counts. The
amount of tracer that is added must be similar to the amount of the radionuclide
of interest in order to represent that level of activity but large enough to achieve a
relatively small standard deviation of counting in a reasonable period. The latter is
especially of concern for the matrix spike, where the result is the difference between
counts that are often relatively small for the spiked and unspiked samples.
Combined analytical and measurement bias is also evaluated by periodic anal-
yses of interlaboratory comparison samples described in Section 11.2.10. The
radionuclide concentrations in these samples should have been measured with
great care. Samples are submitted for blind analysis, i.e., they are not identified as
test samples.

10.5.3. Analytical Method Test Quality Control Calculations

The applicability of radioanalytical chemistry methods is tested initially with ra-
dioactive tracers and realistic mock samples. A tracer that can be measured reliably
and conveniently, such as a radionuclide that emits gamma rays, is preferred. If
initial tracer tests are successful, tests are repeated with the media that will be an-
alyzed. These tests must demonstrate that the radionuclide of interest is recovered
consistently with good yield and that no interfering radionuclides or solids remain.
The extent of reproducibility is determined by analyzing actual samples in repli-
cate for chemical and radionuclide yield. Replicate samples are identical samples
from the same batch, processed and counted separately to assess the variability or
uncertainty in the analysis. The recommended test statistic for a duplicate analysis
is computed using Eq. (10.27):
a1 − a2
Z dup =  2 (10.27)
u c (a1 ) + u 2c a2 ()
 10. Data Calculation, Analysis, and Reporting 213

Here a1 and a2 are the two measured sample results, and u c (a1 ) and u c (a2 ) are their
combined standard uncertainties. Acceptance limits for Z dup may be set at ±3 or
at other values if desired.
If required, the relative percent difference (RPD) for a pair of duplicate analyses
may be computed using Eq. (10.28):
|a1 − a2 |
RPD = × 100% (10.28)
(a1 + a2 )/2
where a1 and a2 represent the two measured results. If the average of the two
results a1 and a2 is less than or equal to zero, the RPD is undefined.
Applicability of radioactive tracer tests to actual radionuclide measurements is
always in question because the radionuclide in the sample may be in a different
chemical or physical form. If the radionuclide has multiple oxidation states in
nature, e.g., iodine or plutonium, a tracer study will be applicable only if the initial
step in the procedure provides for the interchange of all possible oxidation states,
or if tracers have been used in all possible oxidation states. Tests can become even
more elaborate if the radionuclide of interest is an integral part of a solid sample
matrix, e.g., biological material or soil.
In such cases, it may be desirable to reproduce the actual sample with tracers; for
example, radioactive tracer is added to the roots of the growing vegetation that is to
be analyzed. If going to such lengths is not feasible, all portions usually discarded
during the testing of the procedure should be analyzed for the radionuclide to check
for losses not indicated by chemical yield measurements.

10.6. Data Review

Erroneous values have a way of creeping into results despite the control measures
in the Quality Assurance Plan (QAP). The data review process must be designed to
remove such mistakes. Data review should accompany each step of the analytical
process from sample receipt to calculation of activity. Thereafter, weeding out
erroneous values becomes a management responsibility during data compilation
for presentation and retention. Especially for the reality checks described below, the
data compiler must be knowledgeable about radioanalytical chemistry processes
as well as the pattern of analytical results related to sources of radionuclides, or
must be assisted by specialists in these topics.
The data review process is conducted at the following points in the analytical
process:
r Sample collection and shipment data review at sample receipt.
r Sample identification review at initial sample preparation and each subsequent
sample transfer.
r Sample amount, analytical method application, sample response, and yield re-
view during analysis.
r Sample counting and computer data processing review at radiation measurement.
214 Keith D. McCroan and John M. Keller

r Calculational results review.

r QC results review, including instrumental, analytical, and methodological QC,
to maintain statistical control of the measurement process.
r Confirmation of the preceding steps during data compilation.
r Search for missing samples and data during data compilation.
r Reality check of analytical results during data compilation.

The listed items emphasize that each sample must be processed with alertness at
every step. At collection, the sample must be accompanied by all of the informa-
tion required to calculate results and to assure that the correct sample is collected
in the specified manner. The sample must be preserved and stored appropriately.
Familiarity with the sampling plan and previous shipments is desirable to enable
recognition of deviations from the plan. Care in sample identification avoids sam-
ples being switched, lost, or analyzed for the wrong constituents. Observation
of strange behavior during analysis may indicate reasons for unusual yields or
counting results. Application of inappropriate calculations and factors, by hand or
computer, are common causes of error.
Problems are best recognized and corrected at the time of occurrence. If found
later, during data compilation, they may be difficult or even impossible to remedy.
At the very least, however, false results will not be reported.
Findings based on QC results of unacceptable circumstances during a period
of operation (e.g., unduly elevated sample radiation background or deviation of
reported tracer concentration addition from the expected value) invalidate sample
data for the entire period to which the QC data pertain. This period extends back to
the previous QC tests, although an investigation may narrow the problem period by
dating the cause of error, e.g., onset of contamination or error in reagent preparation.
The inverse inference, that correct results in the QC program validate the processes
during the period since the previous QC test, is not necessarily true because errors
may have occurred between QC tests.
One type of reality check is comparing a radionuclide measurement with values
for similar samples collected at the same time or previously, either in the same
program or reported by others. It is usually invoked when the result under consid-
eration appears to be unusually low or high. As indicated in Chapter 12, the best
response is reanalysis. If the questioned value is confirmed, an unexpected situa-
tion has been found. If the second result does not confirm the questioned value,
the source of error should be sought in the analysis, radiation detection, sample
identification, preservation, or possible contamination.
Reports of false positives or false negatives are a common problem, especially
for gamma-ray spectral measurements with data analysis by computer. A false
negative value may be suspected because the radionuclide is known to be at the
monitoring site. A false positive value may be suspected because the radionuclide
is unlikely to be at the location or in the sample—for example, a short-lived fission
product where a nuclear reactor has been inactive for years. Results and calculations
should be reexamined if the reviewer has doubts about the presence or absence of
a reported radionuclide. Repetition of the measurement for a longer period or with
 10. Data Calculation, Analysis, and Reporting 215

a more sensitive detector can be ordered; ultimately, the sample can be reanalyzed
if portions are still available.
In a gamma-ray spectrum, the reexamination may consist of considering details
of the sample plus detector background. For example, peaks in soil samples at
766 and 835 keV that may be falsely attributed to 95 Nb and 54 Mn, respectively,
actually can be minor gamma rays in the natural uranium and thorium decay
chains. For measurements near or below the MDA, the usual peak identification
and quantification software can be replaced with one that is more or less responsive
to channel-by-channel fluctuations.
A reality check may also focus on scanning the raw data and initial calcu-
lations to seek major deviations from the norm. This search typically unearths
typographical errors such as decimal-point shifts and digit reversals, or shifts
of entire lines or columns. Obvious identifiers of error are improbable weights
and volumes, unacceptable yields, and unusual detection efficiencies. More subtle
sources of error are wrong decay schemes or interference by the natural back-
ground.
The chemical and physical characteristics of a radionuclide that affect its inter-
action with the sample matrix may offer suggestions whether its presence can be
expected. For example, thorium and radiozirconium compounds are insoluble in
water except under highly acid conditions and would not be expected in drinking
water. Anions generally are not sorbed on soil and remain in groundwater. Gener-
alities such as these, however, can be invalid under local conditions. Radiocesium
ions are strongly retained in some soil types but not in others. Radioiodine can be
found in airborne particles and also as a gas. Plutonium and the radiolanthanide
compounds are insoluble in water near neutral pH values but can be soluble as
organic complexes at the same pH values.
Uranium and technetium are soluble in water when oxidized but insoluble in
reduced form. Radioisotope pairs, parent–progeny relations, and specific activity
may provide guidance in assigning the origin of radioactive material and identify
questionable results for a specified location. Certain radioactive materials have
“signatures” of uranium or plutonium isotopes at known ratios. Ratios for other
radionuclide pairs, such as 89/90 Sr and 103/106 Ru, can suggest the time interval
since formation. The time interval can also be suggested by ingrowth of shorter-
lived progeny into long-lived parents. The 134/137 Cs ratio provides information on
the neutron environment at origin. The specific activity of 3 H, 14 C, and 129 I can
distinguish between natural and anthropomorphic creation.
The data reviewer may be guided in expecting a certain radionuclide concentra-
tion in a sample matrix by the radionuclide concentration in a related matrix and
the reported transfer factor between these matrices. Relations exist among environ-
mental samples of air, rain, vegetation, and milk, and between industrial samples
of airborne particles and workers’ urine. These relations are complex and often
time- and site-related, but calculations ranging from rule of thumb to an elaborate
computer program may suggest the magnitude of radionuclide concentration ratios
among such matrices. A major inconsistency in the concentration of a radionuclide
among related sample matrices suggests reevaluating the analytical data, although
216 Keith D. McCroan and John M. Keller

the conclusion may be that the data are reliable but the applied transfer coefficients
are inappropriate.
A common problem in radioanalytical chemistry laboratories, especially those
that analyze environmental samples, is the existence of an overwhelming number
of samples without detectable radionuclide activity, which can lead analysts and
data reviewers to expect such results as a matter of course. Reviewers who expect
undetectable results for samples may tend to interpret small positive values as false
positives even when they are not. The tendency to let one’s preconceptions influ-
ence one’s judgment in data interpretation must be resisted. Detection decisions
and other evaluations of the data should always be based on objective criteria. If
the laboratory’s objective criteria indicate the presence of an unexpected radionu-
clide, an investigation may be needed to confirm that the measurement process is
performing properly.

10.7. Data Presentation

The main criteria for presenting radioanalytical chemistry data are correctness,
clarity, and usefulness to the intended audience. In some instances, the customer
or regulator may prescribe the report format. When the format is left to laboratory
management, its design should be directed by the data quality objectives plan (see
Section 11.3.1) to respond to the program purposes. Tables, graphs, footnotes, and
explanatory text should be combined to present the information unambiguously.
Data presentation should be guided by the following items:
r Results that indicate an exceeded limit or guidance value should be highlighted.
r Every result should be accompanied by an explicit measure of uncertainty, such
as the combined standard uncertainty or an expanded uncertainty. The measure
of uncertainty should be clearly explained, and for an expanded uncertainty, the
coverage factor should be stated.
r Results should be rounded appropriately to avoid (1) unnecessary loss of in-
formation and (2) misrepresentation of measurement accuracy. One acceptable
rounding method is to round the reported uncertainty to two figures and to round
the result of the measurement itself to the same number of decimal places as its
uncertainty.
r Detectable and undetectable results should be clearly distinguished.
r The basis for detection decisions should be explained.
r All analytical results must be available in a form that permits review and use in
further calculations.
r Missing data and questionable results must be listed and explained.
r Subsequently changed results due to corrections or new information must be
reported when available, and replace earlier erroneous results.
The data presenter has a wide choice of formats, but should realize that critical
concerns arise at two levels—the value associated with a guidance or limiting level
and the critical value, which distinguishes between a detected and a nondetected
 10. Data Calculation, Analysis, and Reporting 217

FIGURE 10.4. Change in radionuclide concentration with time.

result. The guidance or limiting level is established by a regulatory agency or by

management. It may be direct, as a maximum contaminant level in drinking water,
or indirect in relating the concentration in a sample to a radiation dose limit. The
critical value (see Section 10.4) is arbitrary to the extent that it is related to a
chosen false positive rate; association with 5% false positives is conventional, as
discussed in Section 10.4.
An unresolved point of contention is the extent to which undetectable results
should be reported. One view is that all data accepted after review must be reported.
Values that are statistically indistinguishable from zero, including negative results,
should be included to permit subsequent data review, analysis, and numerical
treatments, including summing or averaging by collection location or time. The
opposing view proposes a reader-friendly table that presents only the detected
values, while all other results are stated to be undetectable. One resolution is to
218 Keith D. McCroan and John M. Keller

FIGURE 10.5. Pattern of radionuclide concentration in water.

prepare both types of table, the former for the professional reader and the latter
for overview by management and the public. Typically, the client of a laboratory
should be knowledgeable enough to interpret a full report that includes all data,
but tables and data summaries prepared for others may be tailored to the needs of
the audience.
Multiple results, such as weekly data in an annual report, can be tabulated
as the mean value with its standard deviation or values of the mean, highest,
and lowest value by location and sample medium. Additional information should
be provided if elevated radionuclide levels occur at specific times or locations.
Patterns of changing radionuclide concentrations can be presented as time-line
graphs. Spatial patterns of radionuclide concentrations can be presented as maps
with concentration isopleths, which indicate the points at which each radionuclide
has a specified constant concentration.
Presentations in figures can clarify and emphasize relations of radionuclide
concentrations to release levels or release points. Figure 10.4 compares the higher
radionuclide concentrations in river water downstream (“US 301 bridge”) from a
 10. Data Calculation, Analysis, and Reporting 219

source to the lower concentrations upstream (“Augusta”). The downstream results

are highly variable from measurement to measurement, but the gradual decrease
with time is apparent in the graph. Figure 10.5 shows that the elevated uranium con-
centrations in groundwater supplies are clustered in only one of the physiographic
regions marked by the heavy lines. Further pictorial information concerning con-
tributions of radionuclides from various sources or the relation of radionuclide
concentrations to their limits can be presented in bar graphs or pie charts.
The extent of detail depends on the purpose of the report. Even a brief progress
report should characterize samples and their data so that they can be traced to more
detailed records. The significance of the results in terms of magnitude, reliability,
and pertinence must be understandable by the reader. Detailed reports for long peri-
ods or at the completion of the project should also provide all necessary information
on sample collection and processing, analytical and radiation detection methods,
quality assurance, and data processing, either as part of the report or by references.
The names of the persons who participated in the program and their individual
contributions are an important part of the record to permit further detailed review.
The units for reporting data are still under discussion at the time of this writing.
The universal trend is toward SI units, which include the becquerel (Bq) for activity
and the meter, kilogram, and second for length, mass, and time, respectively.
However, in the United States at this time, regulatory agencies use the curie (Ci)
for activity, defined as 3.7 × 1010 Bq. Moreover, usage such as river mile markers
and cubic feet per second (cfs) remains common. A typical response in the United
States is to report curies and its submultiples mCi, μ Ci, and pCi (10−3 , 10−6 , and
10−12 Ci, respectively) with SI units for length, volume mass, and time, with the
exception that locations, flows, and such may remain in earlier English units.
11
Quality Assurance
LIZ THOMPSON,1 PAMELA GREENLAW,2 LINDA SELVIG,3 and KEN INN4

11.1. Introduction
Quality assurance (QA) describes the effort by a laboratory organization to pro-
duce trustworthy results. Every laboratory, no matter how small, must maintain
a continuing effort to confirm instrument calibration, measurement reproducibil-
ity, and applicability of analytical methods. These efforts must be documented so
that the results achieved by the laboratory can be used confidently in a decision-
making program. Various decision-making programs are supported by radioana-
lytical chemistry. A partial list of these programs includes
r protection of workers, population, and the environment in accord with regulations
and professional guidance;
r determination that contractual requirements have been met, with regard to facility
operation or decommissioning;
r verification that a decontamination response is necessary for a nuclear incident
with the potential for environmental contamination;
r litigation regarding nuclear attribution or worker compensation;
r demonstration of compliance with guides or regulations for pollution control,
e.g., for drinking water or air quality;
r comparison of results among laboratories;
r confirmation of compliance with international trade agreements, e.g., food
import;
verification of compliance with nuclear nonproliferation treaties such as com-
prehensive test ban monitoring; and
r publication of research.

This list emphasizes the importance of the data generated by a radioanalytical

chemistry laboratory. The results produced by radiochemistry procedures are often
subject to intense scrutiny and even expression of skepticism by sample submitters,

1
Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of
Technology, Atlanta, GA
2
Department of Homeland Security, Washington, DC
3
Centennial High School, Boise, ID
4
National Institute of Standards and Technology, 100 Bureau Drive, Gaithersburg, MD
20899-8462

220
 11. Quality Assurance 221

regulators, or members of the public. Laboratory results are routinely questioned

in court, and must be defensible (see Section 11.5). The subject of contention may
be a few positive or nondetected values or the entire contents of a report. The basis
for such challenges is inconsistency with results from another laboratory, changes
from earlier measurements, differences from predictions, or generalized distrust
of the sampling process or laboratory. In some unfortunate instances, skepticism is
justified. Bad data are produced by inappropriate analyses, and also by unreliable
sampling, failed sample preservation, confusion in sample transfer, errors in data
processing and reporting, and—worst of all—dishonesty. The contamination status
of important sites (see Section 11.3.3) has been questioned as a result of monitoring
failures. To prevent recurrence of these failures, a body of QA measures was
developed and instituted.
As these QA measures became more pervasive and intricate, their formal organi-
zation as a quality assurance plan (QAP) was deemed necessary. Every laboratory
should prepare a QAP that meets all the requirements stated by its operating li-
cense, its nuclear materials license (see Section 13.8.2), and all local, state, and
federal regulations. In addition, the laboratory must prepare a QAP suitable for
each customer who may have different format preferences or requirements.
Although the specific content of each QAP is determined by the individual entity,
certain common elements must, by both consensus and regulation, be addressed.
Consensus standards are QA formats that have been approved for broad use. They
define the elements that must be addressed in a quality system, for that system to
be considered comprehensive and acceptable. Consensus standards often have a
“point of view” because they apply to a particular field or industry. For instance,
ANSI-ASQ E4-1994 (American Society for Quality [ASQ], 1995) was developed
for environmental programs by the ASQ and authorized by the American Na-
tional Standards Institute (ANSI). It serves as the basis for the QA plan of the
EPA (2000a), among many other entities. Some standards are even more specific
in their scope: ASME NQA-1-1989 (ASME, 1989) sets forth requirements for the
establishment and execution of QA programs for siting, designing, constructing,
operating, and decommissioning of nuclear facilities. The elements of a QAP dis-
cussed in this chapter (Sections 11.2.1–11.2.13) are those of the format suggested
by ANSI/ISO/ASQ Q9001-2000 (ISO, 1987), but the goal of this chapter is to
indicate the utility of QA measures to the student rather than to insist on a rigid
and formalized structure.
The laboratory that participates in large projects also must conform to the data
quality objectives (DQO) of the project for which the samples are submitted and
the data are used, and the quality assurance project plan (QAPP) for the project.
The DQO and QAPP are intended to address every aspect of the project, including
justifying the need for a sampling effort, selecting sampling sites and matrices,
collecting samples, reporting the entire project results, and interpreting these re-
sults. These planning tools are discussed in Section 11.3. The DQO process is
mentioned here only as it applies to the overall QA effort.
The fraction of time devoted to QA measures is an important economic
consideration that balances the situation with the funds dedicated to the program.
222 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

Application of these measures to reduce the frequency of reanalysis required

in the radioanalytical chemistry laboratory can be justified by a cost–benefit
analysis. The general purpose of supporting acceptance of analytical results has
a less determinate economic benefit. The time and effort devoted to QA has
increased from about 10% of the total analytical workload to 25–30% at present,
as regulatory agencies have required more quality control (QC) measurements
and more supporting documents. This increasing cost includes QC measurements,
organization and actions instituted to ensure correct and defensible data output.
The costs and benefits of QA are considered in Section 11.4.
Acronyms are used pervasively in this chapter. Although they may annoy the
reader, these terms are in common parlance when discussing QA, both in the
literature and during communication with certain regulatory agencies.

11.2. The Quality Assurance Plan

The Multi-Agency Radiation Survey and Site Investigation Manual (MARSSIM)
describes the quality system as consisting of all efforts devoted to produce quality
results that are “authentic, appropriately documented, and technically defensible”
(EPA, 2000c, Chapter 9). The purpose of the quality manual or QAP is to de-
scribe, in a single document, all elements of the quality system. This includes all
laboratory policies regarding quality and all of the QA measures implemented by
the laboratory. Some records may be included in the QAP itself, while others are
filed as specified in the QAP. The QAP defines every element of laboratory oper-
ation and represents the laboratory’s commitment to quality. The interconnected
elements of QA policy that must be included in the QAP are
r organization,
r personnel training,
r laboratory operating procedures,
r procurement documents,
r chain of custody records,
r standard certificates,
r analytical records,
r standard procedures,
r QC sample analysis program and results,
r instrument testing and maintenance records,
r results of performance demonstration projects,
r results of data assessment,
r audit reports, and
r record retention policies.

These categories are also addressed by other standards and guidance documents,
as mentioned in Section 11.1; however, the order, title, description, degree of
importance, and amount of detail given each topic differs among publications. For
a point-by-point comparison of different standards, see MARSSIM (EPA, 2000c,
Appendix K) or Appendix B of ASQC (1995).
 11. Quality Assurance 223

The categories listed above are described in the following subsections. The
degree of detail is intended as guidance for management to formulate an effective
QA laboratory organization and to enable an auditor to overview the pertinent
components of the organization in a single document.

11.2.1. Organization
The QAP must contain a staffing chart that presents the laboratory organization
and describes the function and responsibility of each individual within the organi-
zational hierarchy (see Fig. 13.7). The resume of each staff member should also
be copied and filed in the plan. This information defines the chains of authority
and communication and records the qualifications for analyzing samples and re-
porting results so that responsibilities can be assigned to the appropriate analytical
and management staff persons. Every staff person is considered responsible for
performing duties and reporting in accord with the QAP. The information must be
kept current.
A QA officer must be designated to be directly responsible for managing the
QAP with respect to preparing the plan, supervising its application, and review-
ing its results. The QA officer must be competent in the fields of radioanalytical
chemistry and QA. He or she assists in reviewing analytical data reports to address
problems with analyses and radiation detection instruments and recommend prob-
lem solutions. The QA officer must be free to discuss results and findings directly
with supervisors, analysts, and operators. He or she must report directly to upper
management, with authority to recommend and initiate corrective action. In a large
laboratory, the QA officer supervises a separate and independent QA staff; in a
small laboratory, the QA officer may also fulfill other functions.
The QA officer and staff procure, store, and dispense QC samples, report QC
results, and evaluate the implication of these results for the analytical program.
The QA officer and staff prepare radionuclide sources for counting and solutions
for “spiking,” i.e., tracing the radionuclide of interest.

11.2.2. Personnel Training

The QAP must document all staff hiring and training. A competent staff is the foun-
dation of reliable analysis. Assurance of competence begins by preparing position
descriptions that specify the schooling, experience, skills, and responsibilities of
each position, and hiring supervisors, analysts, and operators whose qualifications
match those descriptions (see Section 13.7.1). Once hired, the person must be
trained to perform the assigned duties. Such training should address both general
knowledge that can be conveyed in lectures, and specific analytical and instru-
mental operations that are taught by practice. Emphasis should be placed on work
practices that conform to the QAP and safe operation (see Chapter 14).
Periodic evaluations of employee performance should be recorded, with notes
on deficiencies that require greater effort, additional training, or replacement. The
QC program is an important means of checking analysts’ and operators’ skills and
reliability.
224 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

Continuing education should be available to expand analyst and operator skills

and provide opportunity for advancement. Cross-training for other laboratory po-
sitions permits flexibility in work assignments. Attention to new methods and
improved instrumentation assists the laboratory in maintaining a state-of-the-art
program.

11.2.3. Laboratory Operating Procedures

Laboratory procedures must be documented and readily accessible in the QAP
manual. Individual procedures must be available for review when resolving an
analytical problem, and the entire file for auditing laboratory operation. The file
should contain all currently applied radioanalytical chemistry and instrumental
procedures, variants of procedures (e.g., for special matrices or contaminants),
and ancillary procedures (e.g., for reagent preparation, calibration, and QC). Each
procedure and procedural change should be signed and dated. The QA officer is
responsible for assuring that actual laboratory operation is reflected exactly in the
manual. Periodic updating is necessary to formalize as revisions the insertions,
deletions, and additions that can be expected to accumulate in the analyst’s or
operator’s copy of a procedure.
Each radioanalytical chemistry procedure should be written in conventional
style, with its title descriptive of radionuclide and medium, introductory discus-
sion and method summary, list of equipment and reagents, numbered analytical
steps, numbered measurement steps, and equations for calculating sample activ-
ity. Notes and comments should identify the origin of the procedure and address
special precautions and optional variants. Appendices should contain information
such as reagent preparation, instrument operation, and tabulated values for use in
calculations.
Analogously, measurement procedures should include a descriptive title, intro-
duction, numbered procedural steps, and radionuclide activity calculation should
describe operation of radiation detection instruments. Appendices should record
initial setup instructions, dial settings, maintenance actions and schedule, and in-
structions on applying computer software for control, information storage, and
calculation. The location should be given for files of instrument description and
operating instruction manuals and of records of purchase, repair contracts, re-
pair history, and QC data. Modifications in the instrument, computer control, and
operating instructions should be listed, dated, and signed.
All QA processes should be described in sufficient detail to perform them with-
out added spoken instructions. These procedures include instrument calibration,
QC measurements for radioanalytical chemistry, radiation detection, and ancillary
instrument use, data review, and interactions with laboratory staff, both at speci-
fied intervals and in response to incidents. The locations should be given of filed
records such as certificates, QC results, and reports of actions and audits.
Procedures must be included in the QAP for all other aspects of laboratory
operation that can affect data quality, including sample receipt, storage, transfer,
and disposal, equipment application, and data collection, processing, reviewing,
 11. Quality Assurance 225

and reporting. Guidance must be presented for ancillary activities for worker safety,
laboratory security, and emergency response.

11.2.4. Procurement Control Documents

Written instructions and procedures must be prepared to control and document the
receipt of equipment, supplies, and services that affect the quality of measurement.
Examples are reference materials (e.g., radionuclide standards), reagents, supplies
(e.g., graduated cylinders, pipettes, planchets), and computer software and hard-
ware. Controls should ensure that only correct items are accepted. If specifications
are not met or the material is otherwise unsuitable, the QA officer should be notified
and the material returned.
Items with specified storage conditions and shelf life must be controlled by
written instructions to prevent storage and use beyond limiting conditions, e.g.,
temperature, humidity, and expiration date. Equipment is used for its intended pe-
riod of time and under the correct conditions, and undergoes maintenance work at
specified intervals. Similar control measures must be prepared for installing, test-
ing, using, and maintaining computer software and hardware. Test and calibration
results must be documented and maintained.
National consensus standards that pertain specifically to computer software are
available for reference, such as ANSI/IEEE Standard 730-2002 (IEEE, 2002). The
user of equipment driven by software should refer to such documents to understand
how the software will react under given conditions. When a user is unfamiliar with
the intricacies of a software program, the results and their presentation are open
to question.

11.2.5. Chain of Custody and Sample Handling

Each analytical result obviously must apply to the sample for which it is reported.
Any mix-up must be avoided. The Chain of Custody system was developed to
ensure that the reported result applies to the sample collected at the recorded
time and location, and not to another sample inserted by mistake or to deceive.
The recorded parameters are available for comparison with those specified in the
monitoring protocol. Meticulousness in adhering to protocol is crucial for samples
analyzed for legal or regulatory purposes. The format of the Chain of Custody
form must match the QAP for all samples.
A sample Chain of Custody and Analysis Request form is shown in Fig. 11.1.
The form must be signed by all its custodians with a listing of the periods of
custodial control. It reports the sample type, its origin, the times of collection and
transfer, and any treatment.
The practical application of the Chain of Custody form in the laboratory is
twofold: information for sample analysis and data calculation, and formal evi-
dence of the sample’s history. The form must be preserved for both purposes. The
information on the form may also serve as criterion for sample rejection at the
laboratory because of contents with elevated radioactivity or hazardous material.
226 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

FIGURE 11.1. Sample Chain of Custody and Analysis Request form.

 11. Quality Assurance 227

Once the sample is accepted at the laboratory, an internal chain of custody

begins, controlled by a Laboratory Information Management (LIM) system that
may be manual or computerized. A unique sequential identification (ID) number
is assigned to the sample and affixed to its containers as they move through the
laboratory. The ID is recorded by the sample preparer, the analyst, and the radiation
detector operator, and is associated with all records of chemical yield, counting
data, calculations, and reported values.

11.2.6. Standards Certificates

Standard radioactive Material (SRM) solutions are radioactive materials with ac-
curately known radionuclide content and radioactive decay rate or rate of particle
or γ -ray emission. They are used primarily to calibrate radiation detection in-
struments and to prepare QC samples that test analytical accuracy. The supplier
prepares radionuclide standard solutions in flame-sealed glass ampoules. Other
standard radionuclides are in the form of point sources (usually on thin backing)
or as solids in configurations that represent actual samples.
The QA officer is responsible for appropriate use of SRMs and evaluation of
the resulting instrument calibration data, as discussed in Section 8.3.1. The SRM
solutions and their dilutions should be stored in containers that are tightly stoppered
and meticulously labeled to describe the contents and their origins.
Activity calibration of SRMs must be performed by the National Institutes of
Standards and Technology (NIST) or be traceable to NIST. If the radionuclide
standard source is purchased from a supplier other than NIST, its accompanying
certificate must confirm that the standard is traceable to NIST. Traceability requires
reporting an unbroken chain of comparisons to stated references. NIST traceability
is discussed at http://ts.nist.gov/traceability/ (December 2005). A terminal date for
use of a specific standard is appropriate when the radionuclide decays to a small
fraction of its original concentration, an initially minor interference has become
major, or chemical decomposition occurs with time.
The SRM must have a calibration certificate with key information; an example
is shown in Fig. 11.2. Standards certificates must be filed with the sample ID
numbers they accompanied, and the calibrated instrument identifiers.
All certificates must contain information that identifies the physical, chemical,
and radiological properties of the SRM. Physical properties include density and
mass. Chemical properties identify composition such as chemical form, acidity,
and salt content. Radiological properties are the name of the radionuclide, the
decay scheme, half-life, daughter radionuclides, reference time for calculating
decay, and radionuclide impurity amounts. The most important information is the
radionuclide activity or concentration and the uncertainty of this value. Details
must be provided about the technique used to determine activity and the basis for
the uncertainty value. Technical information such as the production method also
may be given on the certificate.
To view or download a NIST certificate in its entirety, go to the following
URL: https://srmors.nist.gov/pricerpt.cfm (December 2005). NIST produces a
228 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

FIGURE 11.2. Summary SRM certificate for 133 Ba (Certificate available at https://srmors.
nist.gov/certificates/view cert2gif.cfm?certificate=4241C) (December 2005).

great many SRMs, all of which are included in this list; the 4000 series consists of
radionuclide SRMs. Click an SRM number—e.g., 4241c for the Barium-133 point
source—and then click on the “C” to view the certificate for specific radionuclide
or matrix. The full NIST certificate contains more than the information shown in
Fig. 11.2 and may be several pages of documentation and instructions on use.
Radionuclides that are used as comparison sources, e.g., for determining instru-
ment count rate stability and as tracers, must be of sufficient radiochemical purity
and activity to eliminate interference in counting and permit correction for decay,
but in most instances they need not have an accurately known disintegration rate.
The absolute count rate also is unimportant for energy calibration because only
the energy must be accurately known.

11.2.7. Analytical Records

The basic recording medium in the chemistry laboratory and the counting room
has always been the notebook. The numbered notebook pages are dated and signed
 11. Quality Assurance 229

by the user. Each analyst enters all pertinent information in a personal notebook,
including the sample ID, what operations were performed (at least by reference to
a method or an instrument), calculations, and any pertinent observations. Ancillary
activities, such as QC measurements, reagent preparations, or instrument modifi-
cations, are also recorded. Unanticipated occurrences and findings also should be
recorded. Changes or corrections should not be erased or whited out, but rather
indicated by restating the information and drawing a line through the superseded
information.
Separate notebooks are associated with sample log-in and each instrument, such
as analytical balance and pH meter in the laboratory and radiation detector in the
counting room. Each use is recorded in terms of operator, date, sample ID, and
pertinent notes. All completed notebooks are filed permanently.
More recently, storage of information in computers has tended to superseded use
of notebooks. The computerized LIM system can include all such information and
relate it to sample ID and QC data presentation. Use of notebooks often continues
on a limited basis. This dual data record may be a lag in technology development
or reasoned record duplication in recognition of computer imperfections. In either
case, storage of all analytical records, including raw data printouts, whether hard
copy, disk, or tape, is mandatory, and the information must be retrievable, as
dictated by the QAP.

11.2.8. Standard Procedures

American national standard procedures for performing tasks such as calibrat-
ing and operating instruments are written by specialists under the aegis of pro-
fessional societies and published for the ANSI. Many standards are available
for nuclear procedures; ANSI standards for some of the radiation detection in-
struments discussed in this text are listed in Appendix A-3. A standard that
defines traceability of radionuclides, ANSI N42.22 (1995), is included in the
list.
ANSI procedures should be considered for use if applicable. Their correct ap-
plication ensures that the procedures under consideration are deemed generally
acceptable in the profession and require no further demonstration of validity. Some
clients require that these standards be used. Standard methods for radiochemical
analysis are discussed in Section 6.5.

11.2.9. Quality Control Samples

A radioanalytical chemistry laboratory must have QC procedures to verify that the
quality of combined chemical analysis and radiation measurement complies with
accuracy requirements. The QC program is the direct-measurement component of
the QA program, and is controlled by the QA officer. A detailed description of QC
activities is part of the laboratory QAP manual.
Internal QC samples are prepared by the QC officer and inserted into the sample
flow at specified intervals, normally to accompany each batch. The conventional
QC sample types are as follows:
230 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

r Reagent or laboratory blank: To verify that method interference caused by con-

taminants in solvents, reagents, glassware, and other sources during sample
processing is known and minimal.
r Replicate: To measure the reproducibility of an analysis.
r Matrix spike (an actual sample to which a known amount—the spike—of the
radionuclide of interest has been added): To check yield by comparing the results
of parallel analyses for the spiked and unspiked matrix. The amount of added
spike is taken 1–20 times that of the radionuclide of interest to permit precise
measurement of both.
r Traceable reference material: To test initially the accuracy and reproducibility of
a procedure without considering the influence of the matrices. Sufficient spike is
added for precisely measuring the counting source and any normally discarded
fractions.
r Blind traceable reference material: To evaluate the accuracy of the analyst.
Sufficient spike is added for precise measurement.

All but the fourth QC sample type are inserted in batches of routine samples. Al-
though inserting these QC samples without the knowledge of the operator (“blind”)
ensures that they are not treated with special care, this may not be possible in small
laboratories, where the analyst may recognize the QC samples. Distinctive features
are their radionuclide concentration (i.e., radionuclide concentrations are zero in
blank samples and elevated in spiked samples), salt content (blank samples may
have none), or the manner of their insertion into the sample flow. The last of the
above types also is used for external QC in the form of interlaboratory comparison
samples (see Section 11.2.11).
The results of QC sample analyses are evaluated with statistical tests (see Sec-
tion 10.5) and presented in tandem with the sample batch results to which they
correspond. Good QC results give confidence that the laboratory procedures and
personnel operated satisfactorily for that batch. Bad QC results show the need for
investigation and remeasurement. The QA officer keeps track of the laboratory’s
QC efforts, and the results are filed as specified by the QAP.
The fraction of QC samples depends on the number of samples per batch. If a
batch consists of about 20 samples, then as much as 20% of the sample load is
devoted to QC samples. Larger batches not only reduce the fractional cost of QC
analysis but also place more sample results at risk of a bad QC result.

11.2.10. Instrument Control Records

The QA effort for radiation detection instruments is directed toward correct instal-
lation, accurate calibration, and stable operation. For correct installation, the in-
strument must be shown to function as designed by the manufacturer and specified
by the purchaser. Accurate calibration depends on reliable radionuclide standards
handled correctly. Stable operation is demonstrated by comparing control param-
eters such as the count rate for a stable radiation source and the detector radiation
background measured at selected intervals.
 11. Quality Assurance 231

The QA manual should have a directory for file locations of instrument manuals
and control records. Control records for each instrument should contain the date of
the test, the name of the tester, the results, and any pertinent observations. Among
these observations may be notes related to deviations and actions taken to correct
these deviations. The QAP should require that all instrument control records are
retained for the life of the instrument, and longer if subsequent reviews may occur.
Once radiation detection instruments are operational, they must be calibrated
with radionuclide standards for response in terms of counting efficiency and, for
some detectors, with radionuclides in terms of energy and peak resolution. The
counting efficiency is affected by radiation type and energy and by sample di-
mensions, density, and placement. Counting efficiency results obtained with one
standard should be confirmed by replicate measurements and with other standards.
Such additional measurements also can be used to estimate the standard deviation
of the efficiency values. In many instances, the efficiency values can be checked by
calculations based on the geometry and associated factors (see Section 8.2.1) and
by computer simulation. Curves or equations that represent trends in the variation
of efficiency with factors such as radiation energy and sample thickness (see Figs.
7.1 and 7.2) can provide efficiency values by interpolation where none were mea-
sured and call attention to efficiency measurements that are questionable because
of their difference from interpolated values.
The results are recorded in permanent form, defined by estimated uncertainty
values (see Section 10.3), and applied in the calculations that are used for re-
porting radioactivity levels. Calibrations are repeated at specified intervals. More
recent values are compared with earlier ones to resolve whether the new values are
within the uncertainty of the old values, should supersede them, or require further
measurements.
Background QC: At specified intervals, in many instances daily or for each
batch, the background count rate for each system must be measured. The count
rate is recorded and plotted on control charts, either by hand or by computer. The
mean value of the background is found by averaging at least 20 measurements.
The ±2σ and 3σ geviations are calculated from the individual and mean values,
and these multiple-standard-deviation lines are plotted on the control chart (see
Section 10.5.1). Once the control chart is established, each newly measured value
is recorded. The measurement should be repeated if it falls outside the 2σ band to
distinguish between a random event and an instrumental problem. Remedial action
with the detector or its environment is necessary if the repeated measurement is
beyond the 3σ band.
Instrumental QC: For a routinely operating radiation detector, a count rate com-
parison source is measured initially about 20 times to establish the mean value and
the standard deviation. A control chart with lines for the mean value and ±2σ and
3σ values is prepared. The source then is counted at frequent intervals—typically
daily or per batch—to demonstrate that the system is functioning properly. The
count rate is recorded and plotted on the control chart for the specific system,
as discussed above. This value is compared with the 2σ warning) limits and the
3σ (out-of-control) limits, and the procedure is repeated if the 2σ boundary is
232 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

exceeded. Appropriate action must be taken if the measurement is beyond the out-
of-control levels. One option is to continue use of the detector but recalculate the
mean value and warning limits for the control chart. This approach is reasonable
only if the values have reached another stable level. The other option is instrument
repair.
Spectrometers have additional parameters for which response should be tracked.
These include the count rate in spectral energy regions of interest, peak channel
numbers at selected energies, difference in channel numbers between two spec-
ified energy peaks, peak energy resolution at specified energies, and the “peak
to Compton” ratio (see Section 8.3.4). A QC chart can be established for each
parameter by replicate measurements. This chart will display each data point as
well as lines at the mean value and at ±2σ and 3σ values.
The number of instrument QC checks typically represents about 5% of the total
number of measurements made for a set of samples. The QC data and records of
problems and responses should be kept accessible for the life of the instrument, and
possibly beyond, to indicate the extent of instrument reliability. Changes in values
may be examined for periodic patterns related to parameters such as temperature
or airborne radon concentrations to consider remedial measures.

11.2.11. Performance Demonstration

Proficiency testing by interlaboratory comparison is the definitive process for
demonstrating reliability of a radioanalytical chemistry laboratory at the time of
measurement. In contrast to the QC samples described in Section 11.2.9, which
are controlled by the QA officer within the laboratory, results of an interlaboratory
comparison are evaluated and reported by an external group. The external group
distributes the test matrix as a “blind” sample to participating laboratories, i.e., the
radionuclide concentration in the distributed matrix is known to the external group
but is unknown to the participating laboratories.
Typically, the external group is a testing laboratory such as the Radiological
and Environmental Sciences Laboratory (RESL), a DOE laboratory that admin-
isters the Mixed-Analyte Performance Evaluation Program and DOE Laboratory
Accreditation Program. The RESL obtains NIST-traceable SRM solutions, spikes
a matrix with a known aliquot of the standard, and divides the matrix into many
identical solution or solid fractions. The testing laboratory measures the radionu-
clide concentration in several representative fractions to ensure that they agree
with the predicted value for the dilution and with each other within the estimated
uncertainty of measurement. The interlaboratory comparison fractions are then
distributed to participating laboratories at prearranged times.
Several matrices and radionuclides may be prepared for distribution to match
the sample matrices and radionuclide concentrations handled by the participating
laboratories. The radionuclide activity in these matrices usually is a compromise
between the relatively low activity in the samples processed by the participating
laboratories and the higher activity that improves precision. As a rule of thumb,
the spiked concentration of the unknown sample should be about 20 times the
 11. Quality Assurance 233

TABLE 11.1. Interlaboratory comparison testing levels

Typical MDCb
Measurement Radionuclidea (per L or per sample)
I BETA particle activity 90 Sr 1 Bq (27 pCi)
Maximum energy > 100
keV
II ALPHA particle activity 228/230 Th or 232 Th 0.02 Bq (0.5 pCi)
isotopic analysis
234/235 Uor 238 U 0.02 Bq (0.5 pCi)
238 Puor 239/240 Pu 0.01 Bq (0.27 pCi)
241 Am 0.01Bq (0.27 pCi)
III GAMMA ray activity 134 Cs, 137 Cs, 60 Co 0.2 Bq (5 pCi)
a A laboratory may elect to be tested for a specific radionuclide or for the category. The testing
laboratory will select the test radionuclide if a category is requested.
b The upper bound of the testing range should not exceed 20 times the stated MDC.

minimum detectable concentration (MDC) for a radionuclide. Performance on

test samples with analyte concentrations much less than 20–200 times the MDC
may not meet the statistical requirements for bias and precision testing. Table 11.1
shows the MDC for a few radionuclides.
Upon receipt of results reported by the participating laboratories, the testing
laboratory publishes a summary report. The report contains the value and uncer-
tainty obtained by the testing laboratory, the values and uncertainties reported
by each participating laboratory (usually with encoded names), and notification
that each reported value is acceptable, acceptable with a warning (i.e., barely ac-
ceptable), or unacceptable. For Table 11.1 and Fig. 11.3, the testing laboratory
is the Environmental Measurements Laboratory (EML). The EML administered
the Quality Assessment Program to test the quality of the environmental mea-
surements reported to the Department of Energy by its contractors. Data and
reports from recent Quality Assessment Program cycles can be found online at
http://www.eml.doe.gov/qap/reports/ (December 2005).
Table 11.2 shows the EML testing results of one laboratory for two different
matrices. The “EML Value” listed in the tables is the mean of replicate EML deter-
minations for each nuclide. The contractors’ results (“Reported Value”) are divided
by the “EML Value” to obtain the “Ratio.” The column titled “Last Evaluation”
gives the laboratory a comparison of its present and preceding performance.
This ratio is expressed graphically in Fig. 11.3 as the distribution of the resulting
ratios for the 40 K analysis performed by the 109 laboratories participating in the
program for that period. The lines on the graph define ranges of acceptability
for the ratio value; these control limits are based on the historical distribution of
data, are different for every matrix, and are specifically defined for each Quality
Assessment Program period.
The upper and lower limits define the total range of acceptable results; a value
above the upper (95th percentile) or below the lower (5th percentile) limit is
marked Not Acceptable (N) by the testing laboratory. The upper middle (85th–
234 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

TABLE 11.2. Individual laboratory results for two matrices.

Reported Reported EML EML Last
Radionuclide value error value error Ratio Evaluation evaluation
Matrix: AI (Air filter) in units of Bq/filter

241 Am 0.286 0.017 0.340 0.040 0.841 W —

60 Co 31.930 0.530 33.500 0.870 0.953 A A
137 Cs 102.500 1.670 99.700 2.300 1.028 A A
54 Mn 45.500 0.800 43.800 1.130 1.039 A W
238 Pu 0.509 0.022 0.520 0.010 0.979 A A
239 Pu 0.323 0.015 0.330 0.010 0.979 A A
90 Sr 2.550 0.120 2.800 0.140 0.911 A A
234 U 0.244 0.013 0.240 0.003 1.017 A A
238 U 0.226 0.012 0.240 0.010 0.942 A A
Matrix: SO (Soil) in units of Bq/kg
241 Am 14.8 1.9 15.6 1.0 0.949 A —
212 Bi 50.9 2.1 60.6 4.0 0.840 A A
214 Bi 56.2 63.4 67.0 2.3 0.839 W A
137 Cs 1,504.0 24.0 1,450.0 73.0 1.037 A A
40 K 659.0 11.0 636.0 33.0 1.036 A A
212 Pb 56.4 1.1 57.9 2.9 0.974 A A
214 Pb 63.4 1.2 71.1 2.3 0.892 A A
239 Pu 29.8 1.6 23.4 1.1 1.274 W A
234Th 113.3 3.5 127.0 7.1 0.892 A —
234 U 126.0 5.2 120.0 0.5 1.050 A A
238 U 128.9 5.3 125.0 0.3 1.031 A A

1.6

1.4 Upper

Upper
1.2 middle

1
Ratio value

Lower
middle
0.8
Lower

0.6

0.4

0.2

0
0 20 40 60 80 100 120
Lab ID number

FIGURE 11.3. EML interlaboratory comparison results for 40 K in vegetation.

 11. Quality Assurance 235

95th percentile) and lower middle (5th–15th percentile) limits define the warning
levels; depending on whether a result lies inside or outside the middle limit lines,
it is labeled Acceptable (A) or Acceptable With Warning (W), respectively. Other
parameters, such as the mean of all submitted values and the mean of acceptable
submitted values (all submitted values except outlying values), may be reported.
Any results in the W or N categories for a participating laboratory should trigger
a thorough QA effort within that laboratory to remedy the situation (see Chapter
12). The simplest causes are typographic, data transfer, and calculating error. A
second line of consideration is an error in analysis or counting that affects only
the intercomparison sample and can be checked by reanalysis. If these causes are
not applicable, the bad result reflects an inherent defect in analysis or counting
efficiency. This conclusion is supported if previous interlaboratory comparison
results were at or near the warning level. The results may hint at what problem to
seek by being biased low or high.
In view of the importance placed on agreement of results by a participating
laboratory with those reported by the testing laboratory, the testing laboratory must
take great care in preparing test samples and calculating its published radionuclide
concentration value and uncertainty. Nevertheless, errors have occurred in standard
certificates and testing laboratories. A significant difference between the mean of
acceptable values by many participating laboratories and the testing laboratory
value suggests that such an error may have occurred. For the values in Fig. 11.3,
for example, the ratio for the mean of all acceptable values of 1.07 indicates
reasonable agreement of results by the testing and tested laboratories.
The existence of a testing laboratory for interlaboratory comparison is so impor-
tant for assuring reliability of radioanalytical chemistry laboratories that it should
be supported by either a Federal agency or a group of laboratories. In the absence
of a testing laboratory, a cooperative venture may be arranged among several lab-
oratories to perform round-robin testing of a single procedure or a collaborative
study to develop and test standard methods. In these cases, results are compared
among several laboratories. Agreement among all participants within specified un-
certainty values usually satisfies the participants but disagreements cannot always
be resolved by attributing bad results to a given laboratory.
Even acceptable interlaboratory comparison results may require further QC ef-
forts at a laboratory if the test sample matrix is not identical to routine sample
matrices. Possible concerns are that the radionuclide in routine samples is in a dif-
ferent chemical form, that the radionuclide is at much lower concentration, or that
the matrix has different constituents. The analytical and counting procedure should
be examined to determine which factors are important, and whether additional QC
tests are indicated.
Acceptable interlaboratory comparison results reinforce acceptable internal QC
results by generating confidence in analytical and measurement processes. The
information usually is required by the client and for laboratory accreditation. Par-
ticipation in these comparisons should be at least on an annual basis for every
matrix type and radionuclide that is analyzed, and would be beneficial on a more
frequent cycle if the workforce turns over frequently.
236 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

11.2.12. Documentation of Data Assessment

A significant fraction of laboratory effort must be devoted to weeding out false
results by consistent application of the QA techniques discussed in the earlier
sections of this chapter, preferably before such results are reported. The QC pro-
gram and interlaboratory comparisons are designed to identify systematic errors in
chemical analysis and radiation measurement due to problematic methods or an-
alysts. By chance, QC results instead may identify an occasional error due to lack
of attention in analysis, measurement, recording, or calculation. More commonly,
occasional errors are found during data review (see Section 10.6).
The QAP document should specify, in addition to the QC and performance
demonstration activities described in Section 11.2.11, the format for data review,
including documentation of actions and outcomes. For example, the review can
begin with matching the data summary to lists of received samples and requested
analyses, individual analytical results, and the applied calculations to look for data
gaps and inconsistencies. In the age of computers, the reviewer checks reliability
of data entry and use of codes appropriate to the calculation.
The QAP should describe responses and corrective actions when laboratory
performance levels are out of compliance. All samples identified in the report as
processed by out-of-compliance procedures, systems, or analysts must be consid-
ered potentially deficient. Appropriate deficiency report forms need to be issued
to inform the pertinent analysts, operators, supervisors, and managers. In some
instances, a sample that does not meet general QA criteria nevertheless may be
acceptable for the intended application, e.g., a measured value is well below the
permissible level although its uncertainty exceeds specifications.
Under favorable circumstances, bad data can be replaced promptly with good
data by finding misplaced samples, correcting false entries, reanalysis, remeasure-
ment, or recalculation. Less favorably, extensive studies may be required to correct
systematic errors (see Chapter 12). The deficiency forms should indicate whether
immediate corrective action is possible.

11.2.13. Audit Reports

Audits are periodic independent surveys, assessments, or site visits to ensure that
the QAP is appropriate and that radioanalytical chemistry laboratory operation
follows the QAP. Generally, a client or accreditation program sends a crew of
audit specialists, but internal audits also are performed.
The audit begins with a meeting with laboratory management, notably the QA
manager, to discuss the plan, and a review of the QAP manual. The auditors ex-
amine all aspects of the laboratory, notably its physical plant, analyst and operator
performance, analyst training, sample handling, record keeping, QA measures, QC
results, and data report preparation. Supervisors, analysts, and operators are inter-
viewed. Checks on data validity include comparing summed times recorded for
sample analyses to the time available, comparing counting times and background
count rates with those required for the estimated data uncertainty, and following
selected samples through the entire analysis process.
 11. Quality Assurance 237

The audit plan, including the schedule, the personnel involved, procedures, and
checklist is the responsibility of the audit organization. A copy of this document
should be filed with the QAP to record the extent of the audit. All information,
comments, and recommendations provided by the audit team at the exit interview
and in the visit report should be added to the filed audit plan.
Every item in the report that requires a response must be addressed. Information
must be provided for action items to describe the outcome of the action taken. All
responses should be filed with the audit plan, together with resulting documents
such as certificates of accreditation or notification that all outstanding items have
been resolved.

11.2.14. Record Retention Policies

The QAP should define record storage requirements with regard to location, time
period, security, and format. Paper records are commonly stored in lockable, fire-
resistant file cabinets. Active computer records are backed up at frequent intervals
to disks or tapes that are stored similarly. After the specified period of years,
data files are transferred to a secure storage location. The time period to ultimate
disposal may be controlled by regulations, management policy, or a specific need
for records (or lack thereof) after an extended period of time.
Old records may be needed for data mining, review, or correction by the labo-
ratory, project review by the client or regulator, or legal challenges (see Section
11.5) related to matters such as worker health or environmental degradation. A
thorough review ordinarily requires the existence of original records to support
both chain-of-custody and reported values. The information can be expected to
include the stated method, detector, calibration effort, counting time, calculations,
data uncertainty, and analyst names.
Although record storage nominally may be “permanent,” consideration must be
given to the actual time of preservation with regard to record stability, space needs,
and caretaker cost. Many data sets have no conceivable need for permanent storage
and can be discarded after several years. Nevertheless, the decision to discard
records should be reviewed at the scheduled time to consider any overlooked
potential need.

11.3. Quality Assurance in Data-Collection Activities

An agency or a laboratory charged with collection as well as analysis of a sample
set must take additional QA measures, as discussed in Section 5.10. A framework
for a radioanalytical data collection activity (DCA) must be established before the
first analysis is performed to present the purpose of the program and the needed
information, even if sample collection and analysis are separate assignments. This
integration of effort, imperative to the smooth operation of the project, includes
implementation of the DQO process and subsequent preparation of a QAPP. The
body of literature describing these activities is large; an introductory description
with references is given here.
238 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

TABLE 11.3. Individual elements of the DQO planning process

DQO step Comments
Step 1: State the Define the problem to be studied, examine budget, and determine time
Problem frame of study.
Step 2: Identify the Decide what specific decisions are to be made on the basis of the results
Decision of the study, what questions must be answered to make those
decisions, and the order of priority given to each decision.
Step 3: Clarify Inputs to Decide what information sources will be needed to answer the defined
the Decision question(s).
Step 4: Define the Study Define the study area spatially and temporally, as well as the
Boundaries environmental medium of the study.
Step 5: Develop a Determine how the data will be used to choose among alternative
Decision Rule actions, including statistical parameters to be used and logic behind
the application of these parameters.
Step 6: Specify Limits Define the implications of a decision error, i.e., the cost or health effects
on Decision Errors that result from a wrong decision.
Step 7: Optimize the Design the DCA in accordance with the needs defined in Steps 1–6 and
Design improve as needed.

11.3.1. The Data Quality Objectives Process

A DCA can be a major undertaking that requires attention to myriad details from
many different disciplines. The purpose of the DQO process is to allow planners
from different backgrounds to work together, each learning to understand the role
of the others in the overall framework of the DCA. Together, the planners can
address objectives and possible hurdles at the beginning to obtain a more realistic
concept of the scope of the project.
An illustration of the DQO planning process (EPA, 2000b) can be found online
at http://www.epa.gov/quality1/qs-docs/g4-final.pdf (December 2005). It consists
of six steps, plus the final optimization step, as indicated in Table 11.3. Each step
is separate, but all the steps are iterative. The comments pertaining to each step
in Table 11.3 are general, and do not begin to address the practical details of data
collection, which are more fully covered in Chapter 5.

11.3.2. The Quality Assurance Project Plan

The result of the DQO process is a set of informed planning elements that are
expanded into what is often called a QAPP. This document dictates the specific
elements of a DCA. The QAPP ensures that the sample collection area is logical
and well-defined, the samples collected are representative of the area with respect
to distribution, number, and frequency, the samples are collected and documented
properly, and the samples reach the radioanalytical chemistry laboratory in an un-
broken chain of custody. At the laboratory, the QAPP defines the types of analyses
that will be performed, the number of measurements per sample, the required reli-
ability and sensitivity of analytical results, and the response time. The QAPP also
 11. Quality Assurance 239

determines what will be done with the resulting data after analysis by defining the
reporting format and the distribution and/or presentation of reports.
The QAPP is not to be confused with the QAP discussed in Section 11.2. The
QAPP is distinct from a laboratory QAP in that it defines an entire project, from
start to finish, rather than governing only the handling of samples once they reach
the laboratory. Like the QAP, the QAPP should be considered a work in progress
because results of analyses are to be used to revise the sampling and analysis
program for a more effective response to the information needs of the client.

11.3.3. Example of Insufficient Data-Collection Planning:

Love Canal
In July of 1970, the Environmental Protection Agency was established. Its mandate
was to repair environmental damage that had already been done and to find ways
to change behavior (e.g., waste dumping) that would cause more environmental
problems in the future. This mandate required the agency to develop large-scale
methodologies for cleaning up contaminated areas and to establish guidelines for
future waste generation and removal. This unprecedented multidisciplinary effort
unsurprisingly experienced a number of setbacks. One of the most glaring occurred
in 1980 in the Love Canal area of upstate New York (Maney, 2002).
The Love Canal region was believed to be highly contaminated by toxic chem-
icals; the EPA was charged with assessing the level of contamination. The EPA
study would serve as the basis for determining whether the area was habitable or,
if not, what level of environmental remediation would be required to make it so.
The affected area was largely residential, which served to intensify media scrutiny
and heighten the furor over what was already a difficult situation.
The result of the EPA’s study was that low levels of contamination by hundreds of
different chemicals were evident in the area (DHHS, 1982). Almost immediately,
the agency was barraged by questions:
r Do the low levels of contamination found by the study accurately reflect reality?
r Will the contamination spread further, or become concentrated in certain areas?
r What are the health effects of the contamination?
r Is it necessary, or even possible, to clean it all up?
The Congressional Office of Technology Assessment eventually was asked to
review the EPA’s study and its findings (OTA, 1983). The OTA determined that,
on the basis of the EPA study, it was simply not possible to determine whether
unsafe levels of contamination existed in the Love Canal area. The EPA study was
deemed inadequate.
The following problems with the study were identified:
r Documentation was lacking on the handling and transfer of the samples and on
the methods used to analyze the samples.
r The sampling effort was inadequate to the task of establishing the level or pattern
of contamination. In effect, too few samples were taken and the choice of location
240 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

for those samples that were collected was not satisfactory, given the geography
of the site.
r Acceptable control sites were not established so that the data collected from the
contaminated area could not be compared with those collected from a reliably
“clean” area.
r Too few replicate samples were taken to define the statistical uncertainty of
contaminant concentration at a given sampling site.
r Insufficient effort was devoted to interpreting the data in terms of its impact on
public health. While the effect of many of the individual chemicals on humans
was well documented, no research was conducted to determine the effect—the
synergy—of many toxic chemicals in concert.
The EPA clearly needed to develop technical guidelines for environmental mon-
itoring; this included the implementation of sampling and analytical protocols, and
the establishment of acceptable techniques for the documentation and presentation
of analytical results. The alternative was to continue to produce results that were
indefensible and expensive. The EPA realized that protocols were lacking, and
published some interim guidelines on QA (EPA, 1980b), but these had not yet
been widely implemented by the time the Love Canal situation came to light.
The problems faced by the EPA were symptomatic of the entire field of chemical
and radiochemical analysis at that time. Laboratories were increasingly scrutinized
and questioned: “How can we (the recipients of the data) be sure that the data are
what you (the generators of the data) say they are?” Formal proof of the data and
the process that produced it was required to answer this question. This need led
directly to the formulation of the DQO and QAPP systems.

11.4. Cost and Benefit of Quality Assurance

Laboratory management initially considered that about 10% of their work should
consist of QC measures. However, new regulations have steadily increased that
percentage; as mentioned in Section 11.1, QA/QC efforts currently constitute 25–
30% of the analytical workload. The expansion of QA activities is certainly a
benefit in that it ensures that heightened attention is paid to achieving quality
results with a concomitant reduction in error. On the other hand, QA measures are
expensive and time-consuming. Moreover, voluminous documentation may give
a false impression of reliability, and the time devoted to preparing documents may
subtract from time better spent in training the analyst and operator. Estimates of
the costs of QA measures and the costs of correcting errors in the absence of QA
are useful to determine whether the expenditure on QA measures is justified in
light of the benefit derived.
On the basis of the appraisals of the ASQC, Ratliff (2003) has outlined four
principal categories for quality costs, as applied to a laboratory environment:
1. Prevention costs refer to the management elements that prevent “bad” data from
being generated in the first place.
 11. Quality Assurance 241

TABLE 11.4. Principal categories of QA active costs (based on Ratliff, 2003)

Cost category Elements
Prevention Preparing and maintaining the QAP
Engaging in the DQO process
Preparing and maintaining the QAPP
Document preparation and revision
Personnel training
Preventative maintenance: facility and instrumentation
Appraisal Document upkeep and filing
QC measures in procurement
Preparation of QC samples
Self-assessments
Participation in interlaboratory testing
Continuing education
Instrument calibration
Data validation
Data analysis
Preparation of QA/QC reports to management
Internal failure Disposal of defective materials and/or equipment
Extra training costs
Repeat analysis
Time spent on internal investigation
Time/revenue lost because of interruption of sample throughput
External failure Time spent cooperating with external investigation
Time/revenue lost because of interruption of sample throughput
Legal costs
Cost of corrective action
Loss of reputation

2. Appraisal costs refer to ongoing efforts to maintain and demonstrate quality

levels.
3. Internal failure costs are incurred to discover and remedy unacceptable data
before they leave the laboratory.
4. External failure costs are associated with efforts to regain customer satisfaction
and confidence if unacceptable data leave the laboratory.

Table 11.4 displays an expanded listing of costs within each category. Most
expenditure results from the time spent on QA measures, rather than from the
cost of materials. Time is of great value, and management should make efforts to
quantify the time spent on each of these QA elements. The analyst can help in this
effort by communicating with management regarding the allocation of time for
each element. For instance, it may be apparent to those in the laboratory that more
time should be spent on training or establishing a more efficient document filing
system. On the other hand, the staff may believe that some elements are consuming
too much time; perhaps too many reports are being prepared or report formats are
too elaborate. The dialogue of itself can be informative.
242 Liz Thompson, Pamela Greenlaw, Linda Selvig, and Ken Inn

Another aspect of QA made apparent by Table 11.4 is that the elements cat-
egorized as Internal and External failure are fewer, but more expensive. If QA
efforts in the Prevention and Appraisal categories are managed properly, the cost
of failure can be avoided.
The less quantifiable result of successful QA measures is the confidence they
awaken in the laboratory output of analytical results. Absent such trust, at best an
entire set of measurements may have to be repeated; at worst, the conclusion con-
cerning an entire monitoring program supported by the analyses may be rejected.
If these alternatives are unacceptable, then sufficient QA measures are worth the
effort.
Beyond reassuring regulators and stakeholders that analytical results reflect
reality, a thorough QA program described by a QAP manual has other benefits
(Ratliff, 2003):
r Some agencies, such as the NRC, require implementation of a QA system before
an operating license is granted.
r Accreditation by various groups requires that a QA system be in place. The QAP
manual serves as documentation that this has been done.
r A QAP manual serves as a historical record for the laboratory. It may be referred
to in times of investigation, or merely as an aid to memory.
r A QAP manual may be utilized as a promotional tool to emphasize ongoing
laboratory commitment to quality.
The QAP manual must, of course, reflect the efforts of effectively trained and re-
sponsible analysts and operators, QA staff, and laboratory management committed
to producing reliable results.

11.5. Defensibility of Scientific Evidence

As mentioned several times in this textbook, the regulatory environment surround-
ing radioanalytical chemistry laboratories has become increasingly restrictive. As
the level of regulation increases, so too does the pressure on a laboratory to demon-
strate that it conforms to these regulations. Therefore, the legal environment has
become just as pervasive an influence. This impact is felt in two primary situations,
both of which are becoming increasingly common in this litigious society:
1. Laboratories that pollute, renege on contracts or obligations, flagrantly disobey
the terms of their license, or fail in some other way will be sued either by their
government or by a private citizens’ group.
2. A laboratory analyst, hired as an expert witness or simply called to testify, may
need to present scientific evidence, either in defense or prosecution of these
laboratories, or pursuant to some other complaint.
In both of these situations, it is certain that the reliability of the evidence—i.e.,
the data—will be called into question to determine its admissibility. The court
will ask the laboratory and/or the analyst to defend the data. The most recent,
 11. Quality Assurance 243

broadly applicable Supreme Court ruling that addresses this issue is related to
Daubert v. Merrill Dow (1993). This ruling makes clear that data and testimony
must meet a standard that the trial judge deems reasonable to admit the testimony,
and indicated the questions that would be used to establish their admissibility. As
stated in Daubert:
Many considerations will bear on the inquiry, including whether the theory or technique
in question can be (and has been) tested, whether it has been subjected to peer review
and publication, its known or potential error rate, and the existence and maintenance
of standards controlling its operation, and whether it has attracted widespread accep-
tance within a relevant scientific community. The inquiry is a flexible one, and its focus
must be solely on principles and methodology, not on the conclusions that they generate.
(http://www.daubertontheweb.com. Accessed August 27, 2005)

These are some of the same questions that are used to evaluate a standard method
(see Sections 11.2.8 and 6.5). Accurate record-keeping in the QAP ensures that
standard laboratory procedures are documented, and accurate record-keeping in a
laboratory notebook ensures that a written record exists to indicate that said pro-
cedures were indeed followed. Although it may seem exaggerated, the importance
of instituting QA procedures and keeping QA records cannot be overemphasized.
12
Methods Diagnostics
BERND KAHN

12.1. Introduction
Ideally, a laboratory processes every sample successfully and then correctly re-
ports its results. In practice, laboratory supervisors and staff devote much time
to examining the causes of failed analyses and attempting to avoid such failures
in the future. Such efforts in the laboratory to identify and then prevent or rem-
edy problems are usually not considered as a formal topic. This chapter proposes a
systematic approach to investigating the cause of a situation, i.e., diagnosis of labo-
ratory methodologies, to solve the current problem and prevent its recurrence. The
laboratory management structure must accommodate response to current prob-
lems in terms of staff, time, and budget. It must also take measures to prevent
future incidents by instituting and adhering to a quality assurance (QA) program,
as discussed in Chapter 11.
QA is the institutional system of methods diagnosis. Implementation of a QA
program formalizes analyst training, written procedures, and the careful review of
results. This formalization demands that each staff member perform the duties of
his/her position and take direct responsibility for the result. Within this framework,
the analyst takes responsibility for the analytical process, the detector operator for
the measurement process, and the supervisor for producing the results. The effect
is to create a laboratory environment that pervasively supports reliable analysis
and dependable reporting.
The quality control (QC) tests discussed in Sections 10.5 and 11.2.9 are integral
parts of QA designed to check results. Some QC measures are prompt indicators
that warn of problem occurrence at the time of analysis; others are delayed indi-
cators that require backtracking to find when a problem first arose. Control charts
for radiation detector operation are an example of a prompt indicator of reliability.
Records of deviations from the norm in an analysis or a measurement may also
be prompt indicators if immediately considered. Periodic blank, blind, and repli-
cate analyses, especially interlaboratory comparisons, are delayed indicators for
which results may not be available for days or weeks after a problem has arisen.
Review and assessment of compiled data are delayed indicators of information
quality.

Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of

Technology, Atlanta, GA 30332

244
 12. Methods Diagnostics 245

The magnitude of the current effort devoted to QA (see Sections 10.5 and 11.4)
is a measure of the concern about problems that are encountered in radioanalyti-
cal chemistry. The goals of this chapter are twofold: first, to emphasize that QA
requirements are instituted not just to convince others of the dependability of the
laboratory but also to provide information for correcting problems and second, to
provide practical insight into laboratory problem solving.

12.2. Problem Origins in the Laboratory

The problems that present themselves in the laboratory are often not immediately
obvious. Samples decompose, lose their identity, or disappear entirely. Analytical
methods and instruments fail suddenly. Analytical results are mishandled. The
analyst may be careless, absentminded, confused, or incompetent. Laboratory su-
pervision may be at fault in issuing ambiguous directions, failing to maintain close
control over laboratory operations, or ignoring warning signs. Mistakes may be
caught quickly if the analyst recognizes common warning signs and understands
the range of problems that they may indicate. Some common problems and their
warning signs are listed in Table 12.1. Note that the listed warning signs do not
correlate to a single problem; each warning sign may signal the presence of one
or several causal problems.
Constant alertness by the supervisor, analyst, and detector operator working
together can help to recognize warning signs and respond promptly to a problem.
The aim is that few analyses will be wrong and that failed analyses are corrected.
Problem identification and correction, while necessary, are reactive solutions. It
is highly desirable that the lessons learned from previous problem occurrences be
used to prevent future problems. Future problems can be prevented by identifying
the reason that the problem occurred. General causes of laboratory problems are
the following:
r periods of change in laboratory operation, procedures, or personnel;
r sample variability;
r method instability;
r unreliable analyst behavior;
r deadline pressure;
r unreliable QC samples;
r deficient supervisory interaction.

These issues are discussed in detail in the following sections.

12.2.1. Periods of Change

Most problems tend to occur at the time of change, notably when a laboratory
begins to perform its analyses. A competent manager insists on detailed man-
uals for method performance and instrument operation, thorough staff training,
and well-tested instruments with associated computers. Written instructions must
246 Bernd Kahn

TABLE 12.1. Observed problems and warning signs in radioanalytical chemistry

Problem Warning signs
Preservation
Loss during storage Unexpected sample appearance or pH
Transfer among phases Inhomogeneous sample
Changes with time
Preparation
Loss during sample preparation Losses
Incomplete or inconsistent dissolution
Radiochemical analysis
Inaccurate aliquots of sample Separation step not fully functional
Inaccurate aliquots of reagent Strange product appearance (amount, form, color)
Inaccurate reagent preparation Low or excessively high yields
Reagent instability
Contamination
Method not followed
Inappropriate method
Incomplete purification
Counting source preparation
Sample losses Source unstable
Wrong sample dimensions Dimensions inconsistently thick or wide
Source impurity
Measurement
Detector peripherals malfunction Nonreproducible count rate
Detector malfunction Varying background
Calibration error Unusual result
Unstable background Inconsistent result
Interference
Calculation error
Data processing and reporting
Misinterpretation Unexpected result
Data mishandling Inconsistent data pattern
Data transfer error

cover all aspects of laboratory operation, as described in Section 11.1 for the QA
plan. Training is given to inculcate laboratory practices and methods application
as well as to review underlying chemical and physical principles. “Cold” (nonra-
dioactive) tests are performed in which the trainer first demonstrates the process,
then supervises every step to check analyst performance and eliminate faults, and
finally challenges the analyst to perform well independently. During this training,
the analyst should accept the responsibility to perform reliable work.
Analytical unreliability may arise when new analysts or operators, samples,
methods, or instruments are introduced. Loss of an analyst or operator and re-
placement by another is a common predictor of unreliability in analytical results,
as is the shift of an analyst or operator to a new assignment or a temporary one for
a vacation period or heavy workload. Thorough education and training can help
prevent bad analyses that are caused by unfamiliarity with a process.
 12. Methods Diagnostics 247

In the same category lies the procurement of new instrumentation and the appli-
cation of new computer software for measurement, data processing, or reporting.
If the new instrument or software is radically different than that used previously,
a class may have to be scheduled to familiarize the staff with the particulars of
the equipment. Less drastic changes, such as the updating of a model or version,
probably can be handled by reading the manuals accompanying the new product.
Even subtle changes can cause problems. Examples are preparation of new
reagents and interactions by different persons in the chain of sample preparation,
analysis, and measurement. Here again, thorough training is the greatest preventive
measure in avoiding mistakes. Analysts should be notified in writing when the lab-
oratory makes any changes in reagent concentration or identity, and these changes
should be entered in the QA manual. Changes in the procedure for sample transfer
or alteration in the chain-of-command in the laboratory should be conveyed to the
staff so that everyone knows their new duties within the management framework.

12.2.2. Sample Variability

A competent analyst realizes that a sample matrix can vary widely in kinds and
amounts of nonradioactive components and that any method must be tested for
stability with the entire range of components. Methods taken from publications
or laboratory manuals should specify the limits of applicability with respect to
potentially interfering substances. The analyst should expect that constituents be-
yond the tested range could undermine the analytical method, especially in generic
sample media such as soil, vegetation, or tissue. Analysis of samples that have con-
stituent amounts beyond the indicated limits may fail because of interference with
chemical separation or carryover with the source prepared for measurement.
The analytical problem can result in appearance of a solid where none should
appear, a clear solution instead of a precipitate, or an unusually high or low yield
for the radiation detection source. The analyst’s notes concerning the procedure
may point to the cause. Otherwise, each step of the analytical method must be
examined to find the step in which the constituent interferes and the reason for the
interference, e.g., competition for reagent, change in pH, or change in the redox
potential (pE) that is needed for the reaction.
Interference by an unanticipated radionuclide may be recognized when an un-
usually elevated value of the radiation measurement is questioned. The sample
should be subjected to detailed determination of the radionuclide decay scheme
(see Section 9.2) to attribute the measured radiation either to the radionuclide of
interest or to interfering radionuclides. Further chemical purification or spectral
analysis may confirm the presence or absence of interfering radionuclides.

12.2.3. Method Instability

Even when matrix constituents are within an acceptable range, the method may
be written in a way that permits analytical variability to the extent that the method
will fail sometimes. Factors such as mixing period, temperature, pH after reagent
248 Bernd Kahn

addition, and amount of reagent may be specified too imprecisely to achieve con-
sistently the conditions for successful analysis.
Such inconsistency may be associated with unexpectedly wide fluctuations in
yield or decontamination factor, but may be difficult to assign to the imprecise
instruction at fault unless the analyst is sufficiently observant. Without such ob-
servation, a step-by-step test of the analysis may be required.
Occasional method failure can occur before or after chemical separation. A
variable fraction of the radionuclide may be lost during storage or initial treatment
before the carrier or tracer is added, or interchange between the carrier or tracer
and the radionuclide of interest may be incomplete. During counting, instrumental
effects such as quenching may be inaccurately assessed.

12.2.4. Analyst Behavior

Some analysts are less meticulous than others. Even a careful analyst may have
moments of absentmindedness during which a reagent is not added to one sample
but is added twice to another, a step is omitted, or the wrong phase is discarded.
The incidence of such errors can be reduced by focusing on the work at hand and
reducing the opportunity for distraction, but cannot be entirely eliminated.
An odd but apparently common effect is a gradual series of unauthorized changes
in the procedure over time. Comparison of the performed procedure with the
written procedure after several years may show a subtle change in reagents, order of
addition, time of mixing, or sample volume. One can speculate that each individual
change was in response to once-only situations, but the effect is that one final
modest change disables the method. Periodic internal audits should detect these
unapproved changes before ultimate method failure.
Most unfortunate is the employment of an analyst or operator who is uninter-
ested, incompetent, or even untrustworthy. These characteristics usually are sig-
naled by aberrant behavior such as irregular attendance or lack of attention to work
in parallel with poor analytical results. Without a thorough change in behavior, only
dismissal prevents further analytical problems by such analysts or operators.

12.2.5. Deadline Pressure

Confusion caused by ad hoc changes in the work schedule adversely affects anal-
ysis and measurement reliability. Routine work scheduling considers the normal
processing time for each type of sample, the number of samples to be processed,
and all ancillary responsibilities such as reagent preparation, information record-
ing, and communication with others. Unusual deadline pressures may demand
faster analysis, adding samples to be analyzed in parallel, or accepting respon-
sibility for extraneous assignments (see Section 13.9.4). Change orders may be
verbal instead of written and may be ambiguous. When such pressures occur, as
they inevitably do, supervisors must make special efforts to give clear orders in
writing and to avoid giving assignments that confuse the analyst. Preplanning can
ameliorate deadline pressure.
 12. Methods Diagnostics 249

A source of both distraction and additional pressure is frequent interruptions

by managers with inquiries concerning the status of analyses and suggestions for
timesaving method modifications. Such interactions may be inevitable, but they
should be minimized. Analysts and operators should be shielded, as much as pos-
sible, from sources of unnecessary distraction, which only increase the likelihood
of error.

12.2.6. Quality Control Sample Reliability

Preparation of QC samples for internal testing deserves particular care because
of their impact on the perceived reliability of the analytical program. The prob-
lems listed in Table 12.2 can falsely indicate error in the analytical process that
stimulates unwarranted methods diagnostic efforts by reviewing and repeating
analyses. The QC staff must take thorough precautions in QC sample prepara-
tion to be meticulous, maintain accuracy and traceability to National Institute of
Standards and Technology (NIST) standards, prevent radionuclide contamination,
and keep reliable and detailed records. These efforts are sufficiently important to
require direct overview by the QC supervisor.
The QC program tests and provides information on the aspects of laboratory
work listed in Table 12.2. Any suggestion from QC results, e.g., a specific analysis
is unreliable, will interrupt routine analyses and cast doubt on their results until the
problem is identified. The resulting methods diagnostics should search for causes
due to either individual analysts or operators or a more general loss of laboratory
control, as categorized in Table 12.3.

12.2.7. Supervisory Interaction Needs

A minor deviation from routine response in the chemistry method or the radiation
detector can warn of an incipient problem. To recognize such signals, the analyst
and the operator must be observant of and sensitive to the process that they control,
maintain careful records, and report this information promptly to the supervisor.
For this aspect of QA to be effective, the supervisor has to have an ongoing daily
dialogue with the analyst and the operator.

TABLE 12.2. QC Sample purpose and problems

Information provided
Day-to-day consistency and precision of results
Data reliability and accuracy of results
Potential errors
Reported radionuclide concentration, uncertainty, impurities, or chemical form
Applied values of half-life, decay fractions, or energy
Calculation of radionuclide dilution or decay
Mass measurement, losses, or other aspects of source or sample preparation
Excessive delay period before use
Calculated or recorded results
250 Bernd Kahn

TABLE 12.3. Common problems that can be traced to individuals or the

general laboratory environment
Individual problems General laboratory problems
Wrong sample Sample misidentification
Inaccurate reagent preparation Reagent decomposition or instrument malfunction
Misinterpreted instructions Radionuclide cross-contamination
Unwarranted assumptions

Review of the process prompted by analyst or operator concerns should in-

volve the analyst, the supervisor, and possibly a specialist. This review may be
able to distinguish among possible causes in chemical analysis such as matrix dif-
ference, method instability, bad reagents, or analyst error. In radiation detection,
source problems, detector malfunction, and data analysis must be distinguished.
The discussion should focus on what the analyst or operator remembers about the
measurement series in question, in contrast to records for similar analyses; this
should help determine when the problem was first observed and the differences in
the process since then.

12.3. Problem Response

12.3.1. Sample Consistency
The reliability of a sample result is questioned when comparison measurements
show different results. Measurements from the following samples are usually com-
pared with measurements of the radionuclide of interest in the laboratory:
r Aliquots of replicate samples analyzed at other laboratories.
r Laboratory QC samples (see Section 11.2.9) such as replicates.
r Samples collected at related locations or times.
r Related radionuclides measured in the sample.
r Samples collected at the point of origin or in related media.
The first two bulleted items provide direct comparisons, and the other three are
guides for anticipated radionuclide levels. A measured result that differs greatly
from expectations may raise concerns that stimulate interest in reviewing the com-
pared data and perhaps repeating the analysis, but by itself is not an indicator of
erroneous measurement.
The initial response to a question of reliability should be an examination of data
uncertainty (see Section 12.3.8). Data can be concluded to be inconsistent only
if the ranges of uncertainty of the compared values do not overlap. The evaluator
must decide whether to accept an overlap by multiples of the standard deviation
such as 1σ or 2σ , and realize that occasionally and randomly a measurement lies
beyond the specified range.
Consistency in replicate samples depends on the meticulousness of replication
as well as the reliability of the compared results, hence both possible causes must be
 12. Methods Diagnostics 251

considered. Split samples of a homogeneous medium such as an aqueous solution

are readily prepared; their consistency depends on accurate pipetting or weighing in
preparing aliquots and the continuing stability of the comparison samples. Ideally,
every split sample is measured at the time of preparation to determine the mean
radionuclide concentration and the standard deviation for the distributed samples.
Measurement of a sufficiently large subset provides these values but leaves in
question the radionuclide concentration in unmeasured aliquots.
An important consideration in preparing liquid samples is assurance of radionu-
clide solubility. Radionuclides subject to hydrolysis and “radiocolloidal” behavior
(see Section 4.2) require a suitably acidic medium. Other radionuclides may re-
quire specific reagents to prevent reactions that lead to insolubility or volatility, or
a stable environment of temperature or darkness.
Preparation of aliquots from heterogeneous material such as dried soil or biota
ash requires formal sample splitting (see Section 5.10). A radionuclide may not be
uniformly distributed on particles if the solid has various components, e.g., sand
and clay in soil, that retain the radionuclide with different affinities, or if the ra-
dionuclide is incorporated in specific particles, i.e., “hot particles.” Measurements
of such solids can be compared only if the radionuclide distribution among aliquots
is demonstrably uniform.

12.3.2. Sample Preservation

A common problem in water samples is the “radiocolloidal” behavior observed in
extremely low concentration radionuclide samples, which represent the majority
of all radionuclide samples. In brief, in deionized water, many radionuclides are
sorbed on container walls and suspended material (see Section 4.4). This effect is
reduced at higher salt concentration and increased acidity, hence the EPA requires
acidification to pH 2 or less with HNO3 at sample collection for analyzing water
from public supplies.
The extent of loss from solution by surface sorption appears to be a complex
function of the chemical and physical characteristics of the radionuclide, the so-
lution, and the surface. Prevention of loss can only be inferred from studies that
replicate the samples that are being submitted and their containers. Following are
some simple tests of radionuclide loss from solution:
r Perform separate gamma-ray spectral analyses of the sample and its container.
r Insert a thin plastic bag in the container to hold the sample and subsequently per-
form separate analyses of the ashed bag and the liquid sample for radionuclides.
r Measure radionuclide sorption on the surface of a test coupon that represents
the container wall that is immersed in the solution.
r Compare radionuclide measurements of a solution to which various reagents,
such as acids, complexing agents, or carriers, had been added to prevent sorption
on container surface.
r Compare radionuclide measurements of solutions in containers with walls
pretreated with various solutions.
252 Bernd Kahn

The first two tests measure the fraction of radionuclide loss to walls, while the
others only show relative retention in solution. A deficiency in these tests for
samples with low radionuclide content is that the obtained count rate often is too
low for precise determination of loss. A radioactive tracer solution can provide
higher count rates but may not represent the conditions in the actual sample.
Another common problem is the separation of samples into two phases: solids
and liquids for solids such as sediment, vegetation, and tissue; solution and sus-
pended solids for water. For such separation
r radionuclides distribute disproportionately and nonreproducibly between the
phases;
r aliquots taken for analysis do not represent the entire sample;
r sample weight at analysis differs from weight at collection.

Freezing samples or initially filtering liquids can reduce these problems. In the
absence of freezing, a preservative should be added to prevent decomposition of
biota and accumulation of decomposition products. The analyst should plan to
minimize the problem by prompt analysis and either mix the sample before taking
an aliquot or sample the liquid and solid phases in proportion.

12.3.3. Sample Preparation

Selective collection of a radionuclide, e.g., gaseous radioiodine in air sorbed on
charcoal and particulate radioiodine in air retained on a filter, requires further
information to calculate the concentration of the radionuclide in the medium. Both
the fraction of the form in the sampled medium and the fraction retained by the
collector must be known. Information available from reported studies may not
apply precisely to the circumstances under which the samples were collected.
Another concern relates to samples in which the radionuclide is not uniformly
distributed, such as biological tissue, vegetation, or soil. The measured radionuclide
concentration may represent either the entire sample or a defined fraction, e.g., the
soluble portion, a specified particle size range, vegetation without soil (and vice
versa), or a dissected animal organ or tissue. Any separation of the defined fraction
should be performed as early as possible in the process, possibly at collection.
The separation process often is imperfect, either because of losing some of the
radionuclide to other fractions or retaining some of the other fractions. Serious
error is introduced when the radionuclide concentration in the extraneous fractions
far exceeds its concentration in the fraction of interest.
The sample may be prepared by concentrating it in processes such as
r filtering of particles from air or water;
r sorbing of airborne gases;
r sorbing ions;
r evaporating water;
r drying, ashing, and dissolving solids.
 12. Methods Diagnostics 253

One concern is partial or complete loss of the radionuclide before carrier or tracer
addition to monitor such loss. Carrier or tracer often are added after these con-
centration and dissolution processes to achieve complete interchange with the
radionuclide in solution (see Section 4.5.2). If, on the other hand, carrier or tracer
is added before dissolution, interchange of carrier or tracer with the radionuclide
of interest may not occur.
Possible loss may be anticipated and prevented by considering the known be-
havior of its chemical form. For example, if its form has relatively high vapor
pressure, either the processing temperature must be kept well below its boiling
point or its redox environment should be controlled to prevent formation of the
volatile form.
Sample nonuniformity must be avoided when a source is prepared for direct
radionuclide measurement by gamma-ray spectral analysis (see Section 7.5.1)
or gross alpha- and beta-particle counting. Some samples that were thoroughly
mixed just before measurement may remain mixed during the measurement, while
others separate into fractions by solids settling or gas emanation. For a mixture
of gas with liquid or solid, e.g., radioactive noble gases in solution or radon in
soil, the container must be filled completely to avoid a gas phase at the top and
sealed to avoid gas loss. Canning a sample is one way to retain gaseous radionu-
clides.

12.3.4. Radiochemical Analysis

Problems in purifying a sample for radionuclide measurement can be due to in-
appropriate procedures or analyst error. The circumstances can be categorized by
frequency as follows:
r The procedure suddenly is no longer successful.
r Analyses fail for a few random samples per batch.
r Analysis fails infrequently.

The first case strongly suggests a problem of recent origin: a reagent was badly
prepared, a fatal change was inadvertently introduced into the procedure, or a new
and incompletely trained analyst is at work. The second case suggests that at least
one of the separation steps is unreliable as written when the sample composi-
tion varies. Infrequent failure suggests a highly unusual sample matrix or analyst
aberration, as discussed in Section 12.2.3.
The simplest cause of failure—and the one most readily resolved—is some
temporary aberration or absentmindedness, which may be conceded by the analyst
or operator and can be confirmed by repeating the analysis or measurement. For a
more serious analyst problem, repeated analyses or measurements can be assigned
to a more experienced staff member.
Use of a reagent or carrier that is badly prepared, incorrectly labeled, in-
appropriate to the method, or no longer effective, can be pinpointed by first
repeating the analysis with a completely different set of reagents and, if success-
ful, then checking each original reagent in turn. In tandem with chemical failure,
254 Bernd Kahn

instrumental failure also must be considered. When a detector in the counting room
becomes suspect, use a second detector to check the earlier results from the first
detector.
Carrier and radionuclide tracer yields are convenient criteria of method efficacy.
Low yields suggest the action of interfering substances, but may also be caused by
analyst error, as discussed above. Yields that are unexpectedly high, especially in
excess of 100%, suggest the presence of the carrier or tracer in the original sample.
For example, some samples analyzed for radiostrontium to which strontium carrier
has been added may already contain a few milligrams of strontium. Excessively
high yields can be caused by the presence of relatively massive amounts of chemi-
cals that behave similarly to the radionuclide of interest, e.g., calcium or barium in
a process that purifies strontium in the sulfate or carbonate form, when the calcium
or barium is incompletely separated during purification. Imperfect exchange with
the radionuclide (see Section 4.5.2) is a condition that makes the yield irregular
and yield measurement, unreliable.
Sometimes the sample parameters simply do not mesh with the constraints of a
given method. Attempted analysis of a sample in which one or more constituents
exceed the amounts in the matrix for which the method was reported to be appli-
cable can result in failure. As discussed in Section 12.2.2, sample matrices such
as water, soil, and biota can be so variable that a procedure developed for one set
of samples may not be able to control interference from the different constituents
of the samples under consideration. In response, analyses can be performed with
smaller samples or more reagents.
Analysts should consider the method’s written statements of limitations with
regard to stable substances and the decontamination factor list for important in-
terfering radionuclides and compare those limitations with the characteristics of
the analyzed sample. For methods that do not have sufficiently defined limits,
the step at which the method fails will indicate the point of departure for a
study of possible interfering substances. This point can be identified by mea-
suring carrier or tracer levels at each step to find where serious loss occurs.
Tests of the inferred cause of failure include increasing reagents, inserting new
purification steps, or repeating existing steps. The presence of interfering ra-
dionuclides is shown by special measurements, e.g., for spectra or radioactive
decay.
The evaluation may show that the method is imprecisely written such that the
analyst views it differently than the writer. This oversight will become apparent
only in discussions with the analyst and must then be rectified by rewriting the
procedure.

12.3.5. Preparation of Source for Radionuclide Measurement

Problems in comparing replicates measured for radionuclide content may be related
to the type of radiation and its measurement, as discussed in Chapter 7. A sample
evaporated on a planchet for counting alpha and beta particles with a proportional
counter tends to form a nonuniform deposit with rings or off-center spots, for
 12. Methods Diagnostics 255

which measurement results of replicates are insufficiently similar. Moreover, the

dimension of the prepared source may not match that of the disk or point calibration
source.
A sample precipitated for alpha- and beta-particle counting that is collected on a
filter or slurried onto a planchet usually is replicated more consistently than when
evaporated. Precipitates should be selected to be stable chemically (for example,
not hygroscopic) and physically (no loss by flaking). Solids should be in a defined
weight range, as light as is convenient for high chemical recovery and accurate
weighing (conventionally within 0.1 mg). Filter materials should be selected for
high precipitate retention, uniform surface, flat filter, and low background radioac-
tivity (for example, glass–fiber filters contain radioactive 40 K).
For LS counting, conventionally in 20-ml vials with mixed aqueous sample
and organic cocktail, common problems are light quenching and luminescence
(see Sections 7.3 and 8.3.2). Chemical or physical separations, such as distillation
of the aqueous sample under conditions that prevent carrying of interfering sub-
stances for tritium analysis, can remove substances that suppress light emitted by
nuclear radiation or emit light stimulated by nonnuclear sources. Modern counters
have computerized data analysis that can quantify corrections for both problems.
Cocktails must be checked periodically to assure uniform counting efficiency.
If the cocktail–water mixture separates over time, the counting efficiency will
change. If a powder is counted in a scintillation gel, settling can change the detector
response.
Samples prepared for alpha-particle spectrometry are very thin and uniform. The
purification procedure should have achieved high yield and low contamination by
extraneous solids. Failure to prepare a thin sample is shown by poor peak spectral
resolution with excessive low-energy tailing. Incomplete radiochemical purifica-
tion in preceding steps is revealed by the appearance of peaks from radionuclides,
which should have been removed.
Prior separation generally is unnecessary for Ge gamma-ray spectral analysis
because the high resolution separates the characteristic gamma rays from multiple
radionuclides. Separation may be useful to remove radionuclides that obscure the
peaks of interest with a Compton continuum from gamma rays at higher energy
or with a peak at almost the same energy.
This convenience of gamma-ray spectral analysis with a Ge detector permits
counting of samples in their original shape and large volumes if detector cali-
bration and background count rates are available for that geometry. Obtaining a
radioactive efficiency calibration source in a complex geometry may be difficult,
but calibration may be performed by Monte Carlo simulation. A correction factor
must be introduced for gamma-ray attenuation when the calibration source and the
samples are large and have significantly different densities. This effect is greatest
when counting gamma rays below about 100 keV, but is noticeable even at 1 MeV.
Problems with regard to sample uniformity and stability for gamma-ray spec-
trometry were discussed in Section 12.3.3. The ease of directly placing a sample
into a container and counting should not divert the analyst from maintaining uni-
formity once a sample is prepared.
256 Bernd Kahn

12.3.6. Radionuclide Measurements

Some problems encountered with the various detectors for radionuclide measure-
ments have generic aspects related to ambient conditions, detector and ancillary
equipment functions, computer operation, and the operator. Other problems are
directly associated with the type of detector, its ancillary equipment, and its func-
tions.
Breakdown in control and stability of the immediate detector environment with
regard to cleanliness, temperature level, power supply, and radiation background
interferes with reliable radiation detector operation. Electronic components func-
tion best at a cool, constant temperature in a dust-free environment. Special
low-temperature and power-supply-stability controls are needed to stabilize the
response of gamma-ray spectrometers and liquid scintillation systems.
Radiation exposure near detectors should be both low and stable (see
Section 8.2.2). Major fluctuation in the radiation background that affects a ra-
dionuclide measurement must be recorded so that either an appropriate background
value is subtracted from the measured count rate or the count rate measurement is
repeated during a stable-background period. Periods of elevated background may
be confirmed by matching information on the time of occurrence with situations
such as
r exposure to external radiation source;
r radionuclides at elevated levels brought into the counting room;
r radionuclide contamination of detector systems within their shields;
r airborne radionuclides in the counting room.

An insufficiently shielded external radiation source can influence background mea-

surements from some distance. Airborne radionuclides typically are gaseous radon
and its particulate progeny, but after nuclear tests or major nuclear accidents may
include fission and activation products. Efforts should be made to control such
exposures for long-term radiation background stability.
Counting efficiency and background values initially are predicted from mea-
surements by the detector supplier or at other laboratories. Once acceptable levels
are confirmed, repeated measurements provide the basis for QC control charts for
the check source and background count rates (see Section 11.2.10). Problems are
indicated by subsequent QC measurements if values are beyond the control limits,
drift toward control limits, or change abruptly. The more frequent the measurement,
the sooner a problem can be recognized.
The first response to an outlying value is to repeat the measurement. If the prob-
lem is confirmed, the next step is checking the instrument and peripheral settings
for inadvertent changes that can be corrected. Next, the possibility is considered
that the background change is due to the environment or that the efficiency change
is due to a test source problem (see below). If none of these noninstrumental causes
is responsible, the detector is taken out of operation for detailed testing and repair.
Inaccurate detector calibration is one of the main problems in detector use. Con-
ventionally, detectors are calibrated with radionuclide solution standards traceable
 12. Methods Diagnostics 257

to NIST (see Section 8.2.1). These sources are used to calibrate the detection sys-
tem for counting efficiency and, for a spectrometer, energy per channel and peak
resolution. Errors occur if
r a commercial supplier provides the wrong activity value for the standard solution;
r the user prepares the source badly from the solution;
r the source location relative to the detector is not accurately reproduced;
r the measurement is recorded inaccurately;
r a calculation is erroneous.

A correct calibration source may become less reliable with time by incorrect
radioactive decay adjustment, increase in the contaminant fraction, damage in
handling, or poorer statistical power due to radioactive decay.
In some cases, efficiency and energy calibrations are obtained from a standard
source of the radionuclide of interest that has a reported uncertainty value. In other
cases, these values are interpolations between data points obtained with several
measured radionuclide standards and plotted as function of energy so that the
uncertainty of curve fitting must be considered. Monte Carlo simulation has be-
come sufficiently accurate for energy calibration at the usually attained standard
deviation of 1–2% for reliability. This approach eliminates the need for interpo-
lation, but depends for accuracy on detailed information on the dimensions of
the detector (as in Fig. 8.9), the source, and detector-to-source configuration. The
simulation should be tested with at least one measurement each at low and high
gamma-ray energy to confirm the utilized information.
Calibration errors can occur if the instrument is incorrectly adjusted, settings
are accidentally changed during operation, or components fail. Care must be taken
that the computer code is appropriate to the geometric relation of sample and de-
tector, type of sample, radionuclide, and output that is to be calculated. Utilized
constants such as type of radiation, energy, half life, and decay or ingrowth fraction
must be checked (see Section 9.2). Other pertinent information that must be con-
firmed relates to the sample dimensions, density, weight, times of measurement,
counting period, and radiation background value. Detection system effects must
be addressed such as resolving time losses and coincidence summing, as discussed
in Section 8.2.1.
The problems listed above for calibration also apply to analysis of routine sam-
ples. Consideration must be given in these calculations to the consistency of the
periodic test sources, the sample collection and processing dates, and informa-
tion that relates amount measured to mass or volume collected. Also considered
must be numerical adjustments for drying or ashing the sample, concentrating the
radionuclide, and procedure yield.
Problems that may affect proportional counters relate to control settings, the
flowing gas, the detector window, and the automatic sample changer associated
with many detector systems. Wrong control settings can affect the operating volt-
age, pulse height discrimination, anticoincidence operation, and amplification. The
count rate is affected when the wrong gas is supplied, the gas flow rate deviates
from specifications, or the gas tank is shut off or empty. The detector window
258 Bernd Kahn

may be the wrong thickness or damaged. The sample changer may have the wrong
sample location information or otherwise may malfunction.
For LS counters, successful operation depends on functional dual PMTs and their
electronic circuits. Controls must be at appropriate settings. Some systems have
spectrometer and pulse shape and timing recognition circuits to restrict background
and evaluate the extent of quenching. Luminescence from ambient light sources
can be eliminated during a period for dark-adapting the sample before counting
it. Luminescence from contaminants must be quantified with associated detector
circuits and computer software or by comparison with prepared test samples.
Alpha-particle spectrometer systems often have multiple Si detectors operated
in parallel with a single spectrometer. A vacuum must be maintained in each de-
tector cell because attenuation in air causes low-energy tailing in each peak. The
adjustable detector-to-source distance affects the counting efficiency and resolu-
tion. Power supply stability is required when low-activity sources are measured
for long counting periods, e.g., several hundred thousand seconds. Results may be
based on only a few counts collected at a peak energy region over this extended
time. At higher count rates for which actual peaks can be viewed, knowledge of
the energies and intensities of multiple peaks for each radionuclide is required to
avoid misattribution of radionuclide activity.
A computer controls gamma-ray detector and spectrometer systems because of
their complexity. The Ge detectors require extremely stable voltage and cooling
by liquid nitrogen during operation to support high resolution. Failure to replace
coolant and freezing of coolant lines are occasional problems. Settings must be
determined initially and then maintained as long as QC measurements show ac-
ceptable ranges of the background, comparison source count rate at characteristic
peaks, energy calibration, and peak resolution. Error in data interpretation can
arise when a radionuclide that is counted at high efficiency emits gamma rays
in coincidence (see Section 9.4.5). Significant differences in density between the
standard and the unknown source can cause error.

12.3.7. Computer Software Application1

Modern software is, for the most part, reliable and capable. Problems can occur
when software instructions are inappropriate, misinterpreted, or misapplied, and
when the system malfunctions. Malfunctions can affect scheduling of counting
periods and repetitive counting patterns; processing, storing, and compiling data;
accumulating QC measurements; presenting control charts; and flagging question-
able data. The most common problems are related to misinterpretation of spectral
analysis data.
Spectral analysis software can fail in the following for the indicated causes:
r Proper identification of a radionuclide due to faulty energy calibration or ra-
dionuclide energy library.

1
Contribution by Douglas Van Cleef, ORTEC, Oak Ridge, TN 37830.
 12. Methods Diagnostics 259

r Report of correct activity data due to faulty calibration or emission-fraction

library.
r Calibration for efficiency and energy due to faulty calibration sample or process.
r Calibration maintenance due to spectrum degradation by drift in applied voltage,
electronics, or high count rate.
r Low-count-rate peak detection and integration, due to inappropriate application
or nonapplication of alternatives to peak fitting such as region-of-interest or
directed fit instructions.
r Background or blank subtraction due to unsuitably high or fluctuating back-
ground.
These problems can be prevented or at least minimized by learning to understand
both the principles of spectral analysis and the capabilities and weaknesses of the
software.

12.3.8. Uncertainty Analysis

Uncertainty estimates are needed for reported data, most importantly to show
whether a measured value
r is unambiguously above or below applicable limits or guidelines;
r meets the reliability specified in the Quality Assurance Project Plan (QAPP);
r agrees with replicate measurements performed at other laboratories.

Uncertainty is commonly reported as the standard deviation, σ , or its expanded

versions, notably 2σ or 3σ (see Section 10.3.2). The standard deviation or its
multiple can be calculated according to Eq. (10.6) by replicate analyses. Although
in many instances the reported values are based only on the uncertainty of counting,
determined by propagating the uncertainty of the gross count and the background
count (see Section 10.3.1), other contributors to the uncertainty can be ignored
only if their impact is minor.
The magnitude of the estimated standard deviation is of concern if it is larger
than
r dictated by either regulation or convention for the work at hand and/or
r expected from experience with current samples and analyses.

A standard deviation smaller than needed suggests that a briefer counting period,
a less sensitive detector, or a smaller sample can be applied. A standard deviation
smaller than expected indicates a detector problem.
To improve, i.e., reduce, the standard deviation of an analytical result, the com-
ponents of the standard deviation value (see Section 10.3.1) must be evaluated to
identify for reduction the major contributors to the uncertainty of the sample mea-
surement result. Multiple measurements of the gross count and the background
count provide direct measurements of the counting uncertainty. The measurements
may indicate that longer counting periods are needed or that the background must
be reduced or stabilized. Other contributors to analysis uncertainty, such as sample
260 Bernd Kahn

losses not compensated by carrier yield determination, inaccurate measurement of

sample mass or volume, and inaccurate yield determination, should be measured
if they may be of the same magnitude as the counting uncertainty. The uncer-
tainty in analyst-related activities often is obtained by “Type B evaluations” (see
Section 10.3.1), an estimate of uncertainty based on experience or statements by
others, and may have to be checked with repeated measurements by statistical
analysis (a “Type A evaluation”).

12.4. Response Organization

The preceding discussion suggests that problems and errors in the laboratory can
be identified by an effective combination of the following activities:
r analyst and operator alertness and interaction with supervisor;
r QC measures;
r data review;
r lessons learned record.

Management must support these activities to avoid analyst error, method break-
down, and loss of laboratory control. A methods evaluation and development group
in a large laboratory, or an individual in a small laboratory, should be responsi-
ble for systematically preparing methods for laboratory application, investigating
method breakdown, and evaluating replacement methods. A group member should
be available for discussing concerns by the analyst and supervisor. The group
should review the literature to identify alternative methods of consideration and to
consider potential problems discussed by others, such as MARLAP (EPA 2004).
An independent QA group or individual is needed to assure that a QA Plan is
prepared to include all methods and training requirements, among other items.
The group prepares QC samples, inserts them in the routine sample flow, reports
results, and assures that every instance of loss of laboratory control is evaluated
and remedied.
A data review group or individual (who may be the supervisor) must consider
every value reported by the laboratory in terms of internal consistency and the
pattern of past and nearby sample values and the uncertainty of these values. The
group or individual must consider missing and questionable values, and investigate
analysis and measurement problems to identify their causes.
13
Laboratory Design and Management
Principles
CHARLES PORTER1,3 AND GLENN MURPHY2

13.1. Introduction
The primary responsibilities in managing a radioanalytical chemistry laboratory
are to perform accurate analytical measurements and report the results in a timely
manner. Fine-tuning the design elements and management practices of the labora-
tory will invariably help a laboratory to meet those responsibilities. This chapter is
designed to give students an overview of what a modern radioanalytical laboratory
looks like and how it functions. The laboratory features discussed in this chap-
ter apply directly to laboratories processing environmental and bioassay samples
with low radionuclide content, but can be extrapolated to laboratory environments
where higher level samples are processed.
The early part of this chapter discusses the design and operating practices that
support analytical processes in an environment favorable for efficient work. The
design incorporates state of the art technologies in sample flow during processing,
hood design, ventilation systems, and waste disposal. The latter part of the chapter
addresses the staffing, costs, and attitudes appropriate for a reputable laboratory.
Management and operating considerations include personnel, operating costs, and
service orientation.

13.2. Design and Operational Elements of a Radioanalytical

Chemistry Laboratory
In the past, radioanalytical chemistry laboratories processed samples resulting
from monitoring nuclear weapons development facilities, fallout from nuclear
weapon tests in the atmosphere, and nuclear power stations. At present, monitoring
cleanup of former nuclear facilities is a major source of samples, and efforts are

1
ELI Group, Inc., 3619 Wiley Rd., Montgomery, AL 36106
2
The Matrix Group & Associates, Inc., 118 Hidden Lake Dr., Hull, GA 20646
3
Charles R. Porter, ELI Group, Inc., 3619 Wiley Rd., Montgomery, AL 36106; email:
radlab@charter.net

261
262 Charles Porter and Glenn Murphy

being devoted to preparation for monitoring radiological accidents and incidents.

The laboratories are located at DOE contractor facilities, and other federal, state,
and local agencies, or are commercially operated to perform contract analyses
for government and private industry. If the demand for processing radioactive
samples increases, building new radioanalytical chemistry laboratories or mod-
ernizing old ones will become necessary. The design presented here is based on
a laboratory recently constructed in association with the authors. The original
design includes a radioanalytical chemistry section for performing a variety of
radiochemical analyses, as well as a section for the analysis of heavy metals,
hazardous chemicals, and volatile and non-volatile organics. The combination
characterizes a mixed waste laboratory designed to handle a sample stream of
both radioactive materials and hazardous chemical materials. This chapter gives
an overview of the salient features of a state-of-the-art radioanalytical chemistry
laboratory.
A primary consideration in laboratory design is the magnitude of the radioac-
tivity in the samples that the laboratory will process. The Department of Energy
has designated four hazard levels for radiological samples:

1. High Level (Type A)

2. Intermediate Level (Type B)
3. Low Level (Type C)
4. Environmental Level (Type D)

Note: These hazard levels are not to be confused with the DOE classification of
nuclear waste into high-level, low-level, mixed low-level, transuranic and 11e(2)
byproduct material categories. These nuclear waste categories are established by
DOE Order 5820.2A, which can be viewed online at http://www.directives.doe.gov
(Dec. 2005). See DOE/EM (1997) for more information on nuclear waste. To
reiterate, waste hazard levels are different than laboratory hazard levels, although
the defining terminology is similar.

Recent experience at DOE sites has shown that most of the environmental sam-
ples collected today are levels C and D. Hence, the laboratory under consideration
in this chapter is designed for the analysis of levels C and D samples. These are
environmental or bioassay samples that contain radionuclides at low concentra-
tions, i.e., approximating levels of naturally occurring radionuclides. Samples at
levels A and B generally will be analyzed in on-site government laboratories for a
variety of reasons, i.e. transportation restriction, sample assay limitations, sample
security, and national security. In Table 13.1, the authors provide their suggested
activity levels to match the four categories identified by the DOE.
Table 13.1 is merely a guide. Each laboratory should develop specific quantity
limits. In some cases, the license under which the laboratory operates will specify
the quantity limits. For instance, the NRC issues specific radioactive material
licenses to facilities, and each license specifies the maximum quantity limit for a
given radionuclide. At government owned and operated sites, the DOE facilities do
 13. Laboratory Design and Management Principles 263

TABLE 13.1. Correlation of radiological hazard levels to radionuclide

concentration levels
Guidance Level Quantity Level Quantity Level
High Level Facility mCi and above 40 MBq and up
Intermediate Level Facility μCi to mCi 40 kBq to 40 MBq
Low Level Facility nCi to μCi 40 Bq to 40 kBq
Environmental Level Facility pCi to nCi 0.04 Bq to 40 Bq

not have quantity limits imposed by any regulatory condition or license, although
specific bases for operation may establish operational limits.
Whether of a regulatory or operational genesis, the quantities established are
based on several factors. These include the radionuclides to be processed; their
physical and chemical form; their decay scheme and half life; the education, skill
and training level of the analysts; the physical design of the laboratory; and the
radiological monitoring and controls imposed. Other conditions and factors also
may justify raising or lowering the quantity limits.
Regardless of the hazard level associated with the laboratory, several physi-
cal design features are held in common. All radioanalytical laboratory facilities
consist of a receiving and initial processing area (Area A), a set of individual
analytical laboratories and radiation detection (“counting room”) areas (Area B),
and the administrative and support facilities (Area C). These core elements con-
stitute the radioanalytical chemistry arm of the facility shown in Figs. 13.1–13.3.
The three figures fit together like a puzzle; they are displayed separately to facil-
itate easier viewing. This particular facility has an overall footprint of 261 ft. by
161 ft (80 × 49 m), or about 42,000 square feet (3,900 m2 ). A few aspects of the
complete mixed-waste laboratory are included in these figures while others are
omitted.
Numbers have been added to the layout in Figs. 13.1–13.3 to indicate sample
flow through the laboratory and help identify specific laboratory components.
Samples are received at the loading dock (1) and temporarily held in a storage area
(2) until ready for sample acceptance processing. The samples are moved to the
sorting area (3) where chain of custody is verified, sample containers are opened,
and the samples are inspected for acceptance. They are logged into the Laboratory
Information Management System (LIMS) (4) and bar codes are applied. One of
the key attributes of LIMS software is the ability to track the status of an individual
sample as it moves through the facility. Every sample bar code should be scanned
when the sample leaves one area and moves into another area. Each sample thus
can be readily located during the analytical process; a technician that moves the
sample to the wrong area is notified immediately by the LIMS.
The samples are held in the storage area (5) to await sample processing. Initial
preparations begin in area (6). Chemical purification and counting source prepara-
tion are performed in the laboratory areas (7) through (11) and (19) through (25);
the laboratory spaces have different sizes to handle various procedures. Prepara-
tion and counting of tritium samples are isolated in area (12), due to its unique
264 Charles Porter and Glenn Murphy

Loading
dock

Prescreening
counting room

FIGURE 13.1. Sample receiving and pre-screening portion of the laboratory (Area A).

preparation and counting requirements. The prepared samples are moved into
area (13) for counting. Areas (14) through (18) contain laboratory space to con-
duct analysis by various instrumental techniques, such as mass spectrometers (see
Chapter 17) and atomic absorption spectrometers. Areas (26) through (29) identify
administrative and support facilities.
The results of these analyses are transferred electronically (via the LIMS) to
workstations, where the data are validated and verified, and reports are prepared.
After the data are reported and accepted by the client, sample residues are moved to
the radioactive waste storage facility for disposal (see Section 13.4.5) or returned
to the client. The design features of individual laboratory areas are described in
Sections 13.3 to 13.6.
 13. Laboratory Design and Management Principles
265

FIGURE 13.2. Laboratory and counting room space (Area B).

 266
Charles Porter and Glenn Murphy

FIGURE 13.3. Administrative and support areas (Area C).

 13. Laboratory Design and Management Principles 267

13.3. Sample Receiving and Initial Processing

The loading dock, (1) in Fig. 13.1, should adjoin the sample receiving area. The
dock should accommodate shipments in large trucks and small cargo vans by being
equipped with a height-adjustable bridge to link the dock area to a semi-trailer,
and a ramp for moving dolly loads of packages onto the dock from a small vehicle.
Ideally, the loading dock should extend across the back of the facility, but at a
minimum it must be capable of handling two cargo vans at a time.
Sample containers off-loaded onto the dock should be moved immediately into
the sample receiving area, (3). The sample receiving and storage areas must be
adequately sized to facilitate receipt of daily incoming shipments. The Chain of
Custody form (see Figure 11.1) should only be signed after all the laboratory
sample acceptance criteria have been met. Signing the bill of lading to signify
receipt of the shipment does not constitute acceptance of Chain of Custody. A rule
of thumb is to design the sample holding area to accommodate five times the design
throughput of the laboratory. Most samples are shipped in large (48-L capacity)
coolers. The design should include stainless steel or high impact rolling shelving,
walk-in coolers and freezers, seamless floors, and epoxy-based paint on the lower
walls. The design incorporates surfaces that can be readily decontaminated on the
assumption that some delivered containers may leak or have surface contamination
on arrival.
Walk-in coolers and freezers are required to preserve samples prior to sample
preparation and during the storage and analysis periods. Cold rooms and freezers
can be installed on any wall that provides direct access to a pipe chase or outside
wall so that condensate from the chillers can be drained into the waste liquid
holding tanks. The chillers can be mounted in the mechanical room penthouse or
on an outside wall.
Sample chain of custody is a critical path control tool for preparing legally
defensible data for the client. As mentioned in Section 11.3.3, some of the problems
associated with the Love Canal contamination issues resulted from an insufficiently
documented chain of custody process. Whenever possible, the chain of custody
process begins in the field with the assignment of a bar code designation, and is
subsequently carried seamlessly into and through the laboratory.
Otherwise, the sample receiving room is where the internal chain of custody
begins (see Section 11.2.5). The first step in accepting a sample is visual inspection
of the container for signs of leakage or damage. Next, the level of radioactivity of
the sample must be determined. Each container is scanned for external radiation to
confirm that shipping label and papers are accurate, and its surface must be checked
to assure absence of surface contamination. Accordingly, the receiving area must
have radiation detection instruments (located in the pre-screening counting room
(3) of Figure 13.1) to handle radiological scanning and smear counting for all
incoming containers. Portable alpha particle and beta particle/gamma-ray detectors
are used to scan the incoming packages and a low-background proportional counter
is used to count smears for alpha and beta particles.
268 Charles Porter and Glenn Murphy

After placing the package in the fume hood, it is opened and the interior should be
checked for signs of internal damage, leakage, or removable surface contamination.
If a package does not meet acceptance criteria of the laboratory or Department of
Transportation, the package must be moved to an area designated for damaged or
leaking containers and processed to avoid further leakage. The laboratory director
must decide whether the package is rejected, repackaged and returned to sender,
or transferred to the waste storage building for disposal. Samples are scanned with
a high purity germanium (HpGe) detector plus spectrometer to check whether the
sample contents that emit gamma rays agree with the sample description.
This area also serves as the preliminary sample processing facility. It may contain
muffle furnaces, drying ovens, grinding mills, evaporation bays, and fume hoods
with dedicated exhaust lines. All exhaust from this area must be treated by pre-
filters, HEPA filters, and wet scrubbing prior to discharge to the environment. Each
muffle furnace and drying oven must be covered with a canopy exhaust system to
remove the heat and fumes generated from the heated samples.
The facility should have a separate room for storing radioactive standard and
stock solutions. This room usually is located near the sample receiving and pro-
cessing area. Radioactive standards and solutions must be kept separate from other
laboratory operations to prevent cross-contamination. The room should have cold
storage capabilities and lockable cabinets. It should be designed to the same spec-
ifications as other sample preparation rooms, with a fume hood and computer
access to permit dilution and other processing of radioactive standard reference
materials and stock solutions.
Finally, samples that have been pre-screened and are prepared for processing
can be moved through the air-lock doors into the analytical laboratory. The air-
lock doors separate the analytical laboratories and counting rooms from the other
parts of the facility. This minimizes contamination inadvertently transported to or
released in the sample preparation areas.

13.4. The Radioanalytical Chemistry Laboratory

After initial sample processing, radioanalytical chemistry is performed in the indi-
vidual laboratories, numbered 7–11 and 19–25 in Figure 13.2. The model labora-
tory unit is designed to accommodate two radioanalytical chemists. The sidewalls
consist of ample processing surfaces and cabinet space, and the center island
contains a double-sided worktable, covered with a seamless laminate that can be
replaced when necessary.
Electrical outlets should be located along the center of the worktable to accom-
modate the use of mobile instruments such as pH meters. Two computer worksta-
tions are indicated for each laboratory; at least one of the personal computers (PC)
should be connected to the main server to report information for samples that are
being processed and add information to their internal chain of custody dossiers.
Each laboratory has a full complement of glassware, reagents, and storage cab-
inets. Glassware, utensils, and reagents should remain in the assigned room and
 13. Laboratory Design and Management Principles 269

not be interchanged with other laboratories. The dishwasher internals and sinks
should be stainless steel to avoid deterioration from harsh chemicals left on the
glassware.
The row of laboratories should be backed up to a physical/mechanical support
chase that supplies water, electricity, vacuum, gas, air, and drain lines. The support
chase should be large enough to accommodate maintenance personnel and equip-
ment, but access should be restricted for other facility personnel. Some designs
incorporate an overhead mechanical room or eliminate the limited-access support
chase. All air supplies and ductwork should be overhead and directly accessible
from a maintenance penthouse that holds filter banks and scrubbers.
Individual laboratories should be designed to be shut down individually and
isolated from every other room in the event of an emergency. Valves for gas, water,
and vacuum lines should be accessible through the support chase or the mechanical
room penthouse. Electrical cut-off switches should be built-in at the door. These
design features prevent a spill or contamination incident from affecting operations
in other rooms or laboratories. Ideally, the contaminated laboratory is taken off-
line, decontaminated, serviced, and brought back on line with only minor impact
on other facility operations.
The laboratory design described in the following sub-sections considers effluent
and waste control. Its aims are minimizing cross-contamination and release of
radionuclides to the environment.

13.4.1. Floors and Walls

Floor and wall coverings should be selected for ease of cleaning and decontami-
nation. Epoxy paint is recommended for walls and floors because it is relatively
impervious to most common chemicals. Its surface is smoother and tougher than
regular latex paints and will withstand the wear and tear of normal operations.
Epoxy paint is more expensive and difficult to apply but will save time, labor and
aggravation in the long run. If paint is not acceptable for floors, a seamless vinyl
floor covering may be installed. The vinyl should have a smooth surface and should
be turned up the walls for 10 cm (4 ) to serve as baseboard for the room walls.
Turned-up sides minimize the potential for spills seeping under the covering or
walls.
Laboratories must not have floor drains to prevent discharge of liquids to a waste
stream that flows directly outside the facility. Spills should be picked up with a
HEPA-filtered wet-vacuum cleaner or should be absorbed on solids. The cleanup
materials (liquid, solid, and filters) should be handled as radiological/hazardous
liquid or solid waste.

13.4.2. Airflow
The primary airflow design feature for the laboratory is single pass air. The air
is treated after it circulates through the laboratory and prior to its discharge to
the environment. Air is not re-circulated through the laboratory after treatment.
270 Charles Porter and Glenn Murphy

Each laboratory should have an average air turnover rate of 6–8 room volumes per
hour.
All air must be treated by filtration and/or scrubbing before it reaches the dis-
charge plenum. This treatment may occur in several stages. First, inexpensive
roughing filters (general collection efficiency of 50 percent or greater) are used
to remove large particles. Medium filters (collection efficiency of 80 percent or
greater) then collect smaller particle sizes, at a higher cost than the roughing fil-
ter. High efficiency particulate air (HEPA) filters (collection efficiency of 99.999
percent or greater) significantly reduce the amount of particulate materials above
1-micron particle size, but they cost considerably more. Without the roughing and
intermediate pre-filters, frequent HEPA filter changes would incur a prohibitive
cost.
Activated charcoal beds are used to capture volatile materials and delay the
discharge of radioactive noble gases. Wet-scrubbers are used to capture acids and
entrained liquids. In-line filters and wet-scrubbers should be located in the pent-
house of the facility. Scrubbers should have an average flow of 20–30 L/min and
should run continuously in any hood that evaporates mineral acids. The individual
hood washdown system should have a separate switch to wash down the entire
ductwork at the end of the day or at the end of a process.
Specific treatment combinations can be designed to match airborne contami-
nants from the physical or chemical operations, the type of sample (solid, liq-
uid, gas), and the quantity that is processed. For example, sample preparation
rooms where soil and vegetation are dried, ashed, ground and sieved require par-
ticle filter combinations but not charcoal beds and scrubbers. For treating lab-
oratory air, the multiple stage filter system should be based on the expected
maximum radionuclide concentration and airborne fraction of the processed
samples. Typical combinations include pre-filters, HEPA filters and charcoal
beds.
The individual laboratory is maintained under negative pressure relative to the
hallways and adjacent accessible areas. Negative pressure must be maintained
in each room to ensure that air is not drawn out of the fume hoods through the
room and into the corridor when a door is opened. The appropriate negative room
pressure is supported by sealing all wall joints and penetrations and by supplying
make-up air for hoods.
Each laboratory should have the capacity for four radiochemical fume hoods,
with one hood rated for perchloric acid. In Figs. 13.1 and 13.2, hoods appear as a
large X. Each fume hood must be provided with external make-up air to assist in
balancing the facility airflow. The make-up air should be at the same temperature
as the room air and provide from 50 to 75 percent of the airflow across the face of
the hood to maintain the hood face velocity at the required flow. This flow normally
is 85 to 100 linear feet per minute (26–31 meters per minute). Newer designs may
require different flow rates that should be verified by the facility industrial hygienist
or safety officer. Proper ductwork and blower motor sizes will keep the noise level
at the hood opening in the acceptable range of 85 to 92 decibels. The air discharged
from each fume hood should be wet-scrubbed and filtered, and then vented to the
 13. Laboratory Design and Management Principles 271

FIGURE 13.4. Total dissolution fume hood.

environment through a single stack (see Fig. 13.4). Normal stack heights should
be roughly six times the stack diameter. In rooms without fume hoods, the room
air should be ducted into one of the existing plenums where filtration is in-line.
New designs or major upgrades should incorporate energy conservation tech-
niques. Design and construction of “green buildings” is gaining the support of
government agencies and can reduce utility costs. The planning architect should
include these techniques in the laboratory design. Examples of energy-conserving
design elements are:
r Heat exchangers on exhaust air streams
r Low-flow fume hoods (under development)
r Night set-back controls on hoods with closed sashes
r Improved control systems to regulate air volumes and temperature on HVAC
units
r A single air discharge plenum and fan for multiple hoods
r Water reuse and rainwater collection
r Task lighting to replace general lighting.

13.4.3. Liquids
The laboratory should be designed to discharge minimal amounts of radionuclides
to the sewer. Releases must meet the limits specified in the Code of Federal Reg-
ulations Title 10, Part 20, Table 3 (see Section 14.8 for an explanation of the CFR
titles) or DOE Order 5400.5. All liquids from the facility (including laboratories,
272 Charles Porter and Glenn Murphy

change and shower facilities, rest rooms, lunch rooms, and offices) are directly
piped into one of three holding tanks. Each tank should hold at least a 5-day vol-
ume of liquid waste, or 10,000 gallons (3.8 × 104 L) for the designed laboratory.
All fume hoods should have a water wash-down system. Each hood should be
washed at the end of the day to prevent acid build-up and the resulting degradation
of air ducts. The wash-down water is collected and re-circulated several times
prior to discharge to the holding tanks. The wash-down water usually is acidic
(pH <2) and needs to be neutralized daily. Eventually, solids will accumulate in
the wash-down water container and should be flushed to the holding tanks.
The holding tanks are located in a separate bermed area that can retain 110%
of the tank volume. The discharge lines from each laboratory (i.e., sinks, hoods,
dish washers) and the holding tanks are constructed of acid- and solvent-resistant
materials. The Teflon joints are heat-sealed to prevent leaks or separation. The
immediate vicinity of the tanks is surveyed at regular intervals to check for leaks
and associated radioactivity.
The holding tanks are monitored and agitated 24 hours a day, seven days a week.
One tank is actively receiving liquid waste; the second tank holds the liquid waste
off line for agitation, sampling, treatment, and discharge; and, the third tank is held
in reserve, in the event problems occur with the other two tanks. Tanks should be
discharged daily to prevent excessive accumulation; in no case should tanks be
held for more than 7 days prior to treatment and discharge.
The liquid waste in the second tank is treated by adjusting the pH to >5.2 and
then sampled for radionuclide analysis. If the concentration is acceptable, then
the tank may be discharged to the sanitary sewage system. If the concentration
exceeds release limits, several options may be considered. The three simplest are:
r Increase the dilution in the tank to reduce the concentration to an accept-
able level for discharge (NOTE: Dilution as a disposal option is not permitted
everywhere).
r Filter the tank contents to remove particulate matter for disposal as solid waste,
then re-analyze and, if the concentration is acceptable, discharge the liquid.
r Treat the liquid by flocculation to remove dissolved materials and then treat the
liquid as described in the second bullet.

Regardless of the chosen method, all employees must adhere to government

regulations and site-specific license requirements. Because of the acids and sol-
vents used in many laboratories, some liquid waste streams should be collected
separately to avoid exothermic reactions that may lead to explosions.

13.4.4. Solid Radioactive Waste

Solid radioactive wastes are collected in each laboratory for transfer to the waste
disposal building. Sludge from the holding tanks is dried and also transferred to
the waste disposal building. The chemical and radiological contents of the wastes
are characterized to provide information for the disposal and shipment records.
 13. Laboratory Design and Management Principles 273

13.4.5. Residual Samples and Waste Disposal

Each sample that comes through the door for analysis brings with it a requirement
for final disposal of the sample, the analyte, and the process waste residue. Rarely is
a sample completely consumed in the analytical process. Several disposal options
are available:
r Contract with a licensed commercial disposal service to remove all waste.
r Return the sample residue to the client and engage a licensed commercial disposal
service for the waste residue.
r Return all samples and associated waste to the client.

Waste disposal is an integral part of the analytical process and a contributor to

analytical costs.
Sample disposal options depend on the sample type and size. Liquid samples are
mostly consumed during the analytical process and require minimal handling as
waste. Sample residue should be archived in case re-analysis is required. When the
sample results are validated and the client authorizes disposal, the sample residue
is discharged to the holding tank.
After data validation and sample release for disposal by the client, solid samples
can be returned to the client or logged into the sample disposal database and
packaged for disposal at an approved waste disposal facility. Samples must be
disposed as either radiological or mixed waste, and not commingled with sanitary
sewer or general landfill discharges.
A mixed stream sample, which contains radionuclides and a hazardous chemical
or a biological toxin, should be identified during sample receipt. Always review
the shipping papers and container markings of submitted samples to identify haz-
ardous chemical or biological contaminants. Mixed stream samples accepted for
radionuclide analysis may require long holding times and create significant waste
disposal problems.

13.5. The Counting Room

After processing, samples are moved to the counting rooms, numbered 13 and 19
in Figure 13.2. Only a small higher-level counting room area (inset to room 13) is
provided because the majority of samples processed at most laboratories are low,
environmental radiation, level samples.
Table 13.2 suggests the types and number of radiation detectors needed for
the laboratory. The first column lists the equipment to process 1000 samples per
month. For radioanalytical chemistry laboratories that processes level C and D
samples, count times per sample are in the 50–1,000 min range. If higher-level A
and B samples are processed, the brief count times require fewer detection instru-
ments. The recommended equipment number provides for counting the radiation
background, calibration standards and quality control samples as well as the actual
samples.
274 Charles Porter and Glenn Murphy

TABLE 13.2. Counting room instrumentation

High Level Counting Room Low Level Counting Room
Large Scale Minimum Large Scale Minimum
Production Design Basis Production Design Basis
Laboratory (1000 Laboratory (125 Laboratory (1000 Laboratory (125
samples/month) samples/month) samples/month) samples/month)
High Purity 2 1 16 2
Germanium
Detectors (HpGe)
Alpha Particle 16 8 164 16
Spectroscopy
Detectors
Liquid Scintillation 1 1 4 1
Counter
Low Background Gas 1 1 4 1
Proportional α-β
Counter

The second column describes a smaller laboratory that can process up to 125
samples per month. Although some laboratories start by offering limited services
and then try to grow into a full-service facility, a minimum number of samples
must be processed monthly to produce a revenue stream that will sustain operation.
This minimum must be calculated for each type and quantity of sample, number
and types of analytes requested, and facility overhead.
The counting room houses the radiation detectors and counting systems indi-
cated in Table 13.2. The counting room depicted in Figure 13.5 is a blueprint for
an actual counting room. It is approximately 6 m × 9 m in the main body, with an
extra leg of about 1.5 m × 3 m for additional detectors. A “dutch” door (a door
divided horizontally, so that the upper portion can be opened without opening the
lower portion) is used to pass samples from the sample preparation room (num-
bered 6) into the counting room (numbered 13). This door, located to the left in
Figure 13.5, allows NO foot traffic. Authorized personnel may enter through the
other doors in Fig. 13.5.
This counting room has sufficient detector capacity to analyze 1000 samples
per month. The counting room should be located near the liquid nitrogen delivery
system, which is generally installed in the gas storage room. A separate loading
dock for the gas bottle storage room can be located directly behind the counting
room, as shown in Figures 13.2 and 13.5. The counting room requires liquid
nitrogen and P-10 counting gas, and possibly other gases that are used for chemical
analyses.
Germanium detectors require liquid nitrogen for cooling, and a minimum foot-
print of 1 m by 1 m. Each Dewar should be directly connected to the liquid nitrogen
feed line through a ball-cock on/off valve and special low-temperature piping. The
Dewar is mounted on a scale to monitor liquid nitrogen content. Detectors are
housed in a steel-clad lead shield to minimize the radiation background, which
13. Laboratory Design and Management Principles 275

FIGURE 13.5. Counting room layout.

Dewar
276 Charles Porter and Glenn Murphy

includes ambient and nearby sources of radiation. The high-voltage and signal
cables can run overhead to each multi-channel analyzer. The low-level counting
room should have additional shielding in the walls that separate it from the second
room that is devoted to counting higher-level samples in the same area.
Because of the heavy weight of lead shields and steel stands, the counting room
should always be on the ground floor. Even with this stipulation, the floor should be
built of high density concrete with imbedded I-beams for increased support. Some
laboratories use a thick steel plate under the equipment footprint to distribute the
weight over a larger area.
Radiation detector shields must be of low-background materials (e.g., lead or
steel). In regions where radon gas intrusion and accumulation present a high-
background problem, the incoming air to the counting room may have to be passed
through charcoal beds to reduce the radon concentration, and room’s walls and floor
interfaces should be sealed to avoid radon emanation from the ground. The intended
location of the counting room should be checked to avoid elevated external radia-
tion levels of natural or man-made origin. Once the counting room is built, nearby
placement of facilities that will elevate external radiation levels must be prevented.
The counting room also houses alpha-particle spectrometry systems, which
require a vacuum line or pumps. These detectors can be placed anywhere within
the counting room and operated without interference from any elevated external
radiation background.
Low-background alpha/beta particle gas proportional counters can be placed
anywhere if a dedicated counting-gas cylinder is provided for each unit. If a central
supply of counting gas is used, the units should be placed in direct access to the
gas supply manifold. Liquid scintillation counters can be placed anywhere in the
counting room because they can be operated as stand-alone instruments.
All counting systems are linked directly to the laboratory computer system to
permit centralized control and for direct data transfer at the completion of counting.
The counting room requires one computer workstation to control each type of
counter, and additional terminals to support multiple users. Counting systems must
be connected to a stable uninterruptible power supply (UPS) that will provide
at least six hours of power in the event of power failure to the laboratory (see
Section 13.7.6). The UPS system should be mounted in the penthouse to minimize
equipment located on the counting room floor.

13.6. Storage
A busy laboratory utilizes large quantities of chemicals for sample preparation
and processing. The chemical storage facility should stand alone from the main
facility and be used to store and segregate hazardous materials such as flammables,
acids and bases (see Fig. 13.6). Chemicals in these different categories should
never be stored in the same room because inadvertent mixing could result in an
explosion or fire.
Radioactive waste also should be stored separately from the main facility to
avoid cross-contaminating other waste categories or supplies. Waste minimization
 13. Laboratory Design and Management Principles 277

Organic Radioactive
Acid Solvent waste waste

FIGURE 13.6. Segregation of chemicals in the hazardous materials storage building.

policy and procedures should be implemented and monitored to reduce ex-

pense.
The floor in a storage facility should be recessed at least 0.3 to 0.5 m or bermed
to this depth in divided sections. The bermed area should have sufficient volume to
collect twice the storage capacity of each section. Each section should be covered
by an acid- and solvent-resistant grate (a material that will withstand aqua regia)
and strong enough to support the weight of stored materials, moving dollies, and
personnel. All storage shelves should be made of the same corrosion-resistant
material. The building needs to be located near the loading dock to facilitate
transfer of materials, yet detached and sufficiently distant from the main facility
to minimize damage from fires or explosions.
By laboratory policy, an individual should notify the receptionist or safety staff
before entering the building. The storage building should be under continuous
surveillance by remote TV monitors, and should have direct telephone links (pick-
ing up the phone connects to the switchboard without dialing) on the outside of
the building near each door. Individuals entering the building should have direct
communication to safety support staff. Containers moved between buildings or
laboratories should be carried in rubber-wheeled carts or special carriers to mini-
mize spills because of glass breakage. All hallways and entry/egress ports should
be designed for the use of these carts.
Ventilation in the chemical storage facility should provide a minimum of five
room air exchanges per hour to minimize build-up of vapors and gases. The venti-
lation system should vent directly to the environment, well away from air intakes
for other buildings.
Separate rooms in the storage facility should be heated and cooled based on
the acceptable temperature ranges of the stored chemicals. All electrical cir-
cuits, including lights, switches, and thermostats should be made of non-corrosive
explosion-proof materials.
Individual laboratories should be equipped with cabinets for storing relatively
small amounts of chemicals that are frequently used. Separate cabinets should be
available to segregate chemicals by hazard type, i.e., flammable, acid, or base. The
cabinets should be labeled by type of contents and provided with locks.

13.7. Elements of Facility Management

A laboratory director faces a number of interrelated tasks in managing the ac-
curate and timely analysis of the many samples submitted by clients. One is to
278 Charles Porter and Glenn Murphy

place and direct competent scientists, engineers, and technicians in analytical and
technical positions, and support personnel in management, maintenance, secu-
rity, emergency response, health, and safety. A second task is to establish the
framework—facilities, equipment and supplies, budgets, and schedules—of labo-
ratory operation. A third task is to direct laboratory services to assure reliable results
that meet all client requirements and are consistent with government regulations.

13.7.1. Analytical and Technical Staff

The organizational structure of the laboratory includes the key positions of director,
quality assurance (QA) manager, radiation safety officer (RSO), radioanalytical
chemist, counting room operator, information technology (IT) manager, facility
manager, financial manager, and human resources manager. The specific staffing
needs of the laboratory are primarily based on the number and types of sam-
ples that are processed. In the smallest laboratory, some of the listed functions
are combined. A large laboratory has multiple chemists and separate radiation
safety (or health physics—HP) and industrial safety officers. Figure 13.7 suggests
the staffing requirements for a laboratory that processes about 1,000 samples per

FIGURE 13.7. Sample organization chart.

 13. Laboratory Design and Management Principles 279

TABLE 13.3. Recommended laboratory staff positions

Education/
Title Experience Responsibility
Laboratory Director MS or PhD + 10–15 years General Manager
experience
Radiation Safety Officer MS or PhD + ABHP Overall responsibility for
Certification safety and licensing
Senior MS or PhD + 10–15 years Overall responsibility for
Radioanalytical/Analytical experience sample preparation
Chemist
Counting Room Supervisor MS or PhD + 5 years General operations of counting
experience room
IT Manager BS or MS + 5 years experience Maintains computer
systems/servers and data
security/integrity
Radioanalytical/ Analytical BS or MS + 2 years experience Day to day operations in
Chemist individual laboratory
Technician High School or Associate Sample receiving, preparation,
degree counting
QA Manager MS or PhD + 5 years Overall responsibility for
experience quality assurance and
quality control
Project Manager Bachelors + 2 years Client/laboratory point of
experience contact

month. In comparison, the smallest viable laboratory processes about 125 samples
per month.
Assurance of competent staff begins by preparing position descriptions that
specify the educational requirements, operational experience, and duties of the
laboratory personnel. The qualifications of the supervisors, analysts, and instru-
ment operators who are hired must match these descriptions. Each staff position
should have a set of core training specifications (e.g., radiation safety, chemical
safety, waste minimization and disposal, quality assurance) plus specific training in
its area of responsibility. Table 13.3 provides suggested position titles, educational
requirements, and general responsibilities for the optimal laboratory staff.
The radioanalytical chemistry laboratory has some specific needs, notably that
the chemists must have specific training and experience in radioanalytical chem-
istry. The skills of the radiation safety officer must match the duties described in
Section 14.3.2. The project managers should be fully versed in the operations and
capabilities of the laboratory to serve as client’s advocate with the analytical staff
and the laboratory’s liaison with the client.

13.7.2. Support Staff

A routine facility maintenance program should keep the facility clean and in oper-
ating condition without inadvertent interruptions or discharges to the environment.
The maintenance program includes (1) general housekeeping, (2) facility support
280 Charles Porter and Glenn Murphy

features and (3) equipment. The HP staff oversees maintenance activities to min-
imize radiation exposure to personnel and spread of contamination. The HP staff
must first review the proposed activity, then survey the tasks while in progress, per-
form a final survey before resuming normal operations, and finally record radiation
exposure and radionuclide contamination levels.
Laboratories may employ either in-house or contractor janitorial staff. The
housekeeping staff needs initial training and routine re-training for areas that are
off-limits for routine cleaning or trash disposal. An individual laboratory may
have several trash receptacles to segregate normal trash, radioactive waste, and
hazardous chemical waste. Housekeeping staff should be trained to recognize the
various hazard symbols and placards and the appropriate response to each.

13.7.3. Equipment Maintenance

Mechanical support services—heat, air, electricity, water, sanitary drains, holding
tanks, scrubbing systems, and filter banks—occupy much space and require large
expenditures for support and maintenance contracts. Maintenance support and lab-
oratory staff must cooperate to minimize disruption of laboratory activities during
routine and non-routine maintenance. Conflicts and downtime can be reduced by
careful scheduling of routine maintenance for equipment such as wet scrubbers,
filter banks, and motors.
Contracts with reliable and responsive service companies for maintaining facil-
ity support equipment (i.e., HVAC systems, electrical systems) are vital for efficient
laboratory operations. Many modern systems are computer driven and so complex
that only a specialist can repair them. When laboratory integrity is threatened, e.g.,
by flooding from a burst pipe or power failure, only fast response by a maintenance
service company can protect against expensive losses and prolonged down time.
Maintenance contracts should include scheduled replacement of system compo-
nents. In the air handling systems, blower motors need regular inspection, oiling,
and belt replacement. The dust loading of the filters should drive filter replacement
frequency. Replacement of roughing filters (quarterly recommended) will prolong
the life of the HEPA filters and activated charcoal beds. The wet scrubbers need
to be drained and re-filled on a routine schedule because the pH continues to drop
when large quantities of acids are processed in the fume hoods. Scrubber water
can be collected, neutralized with a strong base (sodium carbonate or sodium hy-
droxide), and recycled back into the scrubber. A contract maintenance program
to service chillers, motor dampers, computers, and airflow measuring instruments
can assist the in-house engineering staff.
The holding tanks should be on a weekly maintenance schedule to check for
wet regions in the bermed area. These may indicate leaks, but could also be from
rain or condensate formed on exterior tank surfaces. Pumps and agitator motors
should be checked and serviced on the same schedule as sampling and discharge.
Radiation detection instruments require routine maintenance and occasional
repair. Counting room staff is expected to perform initial trouble shooting, but
specialists often are needed to diagnose problems and perform repairs. Most
 13. Laboratory Design and Management Principles 281

instruments can be obtained with maintenance and repair contracts at the time
of purchase. Operation of a small laboratory may at times be at the mercy of
a repair service for rapid turn-around in detector repair. A large laboratory may
consider comparing the cost and turn-around time of an internal repair staff with
contract repair. Rapid obsolescence of certain equipment may suggest a lease to
be more cost-effective than a purchase.

13.7.4. Laboratory Security

Security has always been a consideration, but awareness has increased since
September 11th 2001. The laboratory must be protected against illegal entry,
damage, and theft. The samples must be protected as client property. Radioac-
tive material requires heightened protection, especially if controlled material is
included.
In smaller or less secure laboratories, checkpoints are established by locking all
access doors—i.e., only those with a key may enter. In a more security-conscious
laboratory complex, each area or room in the facility is individually controlled
with a computer-activated system. The computer system can identify the person
who enters, log entry and exit times and give management the ability to restrict
unnecessary or unscheduled access (e.g., entry late at night). Each room has a list
of approved access personnel, and only those personnel will be able to enter by
use of their magnetic or bar-coded personal ID card.
Access to higher-security areas is restricted at designated boundaries. The
area of lowest security in the building—the only open and directly accessi-
ble part of the building—should be the reception area, numbered 27 in Fig.
13.3. All visitors must sign in at the reception area, and should be given tem-
porary ID cards. Visitors should be escorted throughout the remainder of the
building.
Employees will use their ID card to pass through the reception area into more
secure areas. This zone contains offices of the laboratory director and support staff,
technical staff workstations and meeting space. Bathroom, shower and change
facilities appropriate to the size and amenities of the facility should be located in
this area. Kitchen and dining space should be provided here because eating and
drinking are not allowed inside any laboratory work area.
The entire laboratory work area is the higher-security zone. Entry into the lab-
oratory should be limited on a “need to be there” basis. Direct access to and from
the laboratory should be minimized. The shower and change facility constitutes a
buffer area for transit from the main laboratory to the administrative area, as shown
in Fig. 13.3.

13.7.5. Data Security

Clients consider their submitted samples to be their property, and information
gleaned from these samples to be proprietary. If the client is a government
agency, some sample results may be considered classified information that must
282 Charles Porter and Glenn Murphy

be protected. The laboratory and specified personnel must be certified as capable

of handling such classified information, and personnel must have the appropri-
ate security classification. Facilities regulated by the Departments of Energy and
Defense have orders and guidance documents that specify the requirements for
data security, integrity and authenticity. The client may specify individual security
requirements.
Regardless of formal security requirements, all analytical data should be treated
as a valuable and irreplaceable resource. Data security begins with physical security
of the facility discussed above, and is maintained through protection of computer
hardware, software and Internet accessibility.
The laboratory computer system is the backbone of operations. It is used for sam-
ple management and tracking, data storage, reduction, and presentation, and sam-
ple archival or disposal. The system also may process personnel records, payroll,
chemical inventories, accounts receivable and payable, and waste management.
A suitable computer system, at the time of writing, can consist of two high-end
servers with a hot-swappable redundant array of independent disk (RAID) drives.
Such drives vastly increase the throughput of the system. The computers must
be equipped to perform automatic information backup with magnetic tape or by
CD read/write. Ethernet high-speed interconnectivity and ultra-high-speed Internet
connections should be available. Individual PC-style workstations with barcode
readers should be installed where samples are received, stored, processed, counted
or discarded.
The computer system should be installed in a secure limited-access room; in
Fig. 13.3 the computer system is in the room numbered 28, conveniently located
near the record storage room (26). The computer room should be equipped with a
UPS that has a six-hour rating. The system requires low humidity (<60% relative
humidity) and constant temperature (18–21◦ C).
Personnel access to the system should be controlled with a password autho-
rization system, but password protection is only as good as the type of system
implemented and maintained by IT personnel. Passwords should be changed quar-
terly, require a minimum of 8 characters and numbers, and IT personnel should
routinely test the system with password-breaking software to identify and replace
easily inferred passwords.
Commercial software (currently, Microsoft, Corel, AutoDesk, etc.) upgrades
that are issued every year or two should be evaluated for backward compatibility,
bugs, and security flaws. Lack of compatibility may prevent opening some files
and corrupt others. Replacement of specific scientific software should be evalu-
ated for consistency in aspects such as look-up tables, algorithms and calculation
routines.
Computer security depends on limiting system breaches from the Internet. Com-
mercial products are available to detect intrusion, i.e., unauthorized attempts to
penetrate the system. These products alert IT personnel when an attack is de-
tected. Commercial firewalls can minimize system penetration. Anti-virus pro-
grams can continually scan incoming material for viruses, worms, and trojans.
None of these products can replace continual vigilance by IT personnel be-
cause a determined hacker can penetrate any commercial system and disrupt
 13. Laboratory Design and Management Principles 283

laboratory information handling activities, notably sample processing and data

delivery.

13.7.6. Laboratory Emergency Support

The laboratory safety measures discussed in Chapter 14 go far toward preventing
emergency situations, but some accidents and incidents are unavoidable. Wise
management practices include planning for and preparing the facility and staff for
such eventualities.
Maintaining electrical supply is vital during any emergency. First responders
are aided in their efforts by the availability of light, communications, operating
equipment, and active security controls. Electrical supply to the laboratory should
be provided at three levels:
r Regular service is the primary feed to the entire facility. Preferably, the supply
is conditioned to reduce the impact of voltage fluctuations.
r A standby generator supplies emergency power. The emergency supply operates
emergency lights, cold rooms, freezers, refrigerators, and fume hoods. Two-
speed blowers in fume hoods should automatically drop to the lower speed when
the emergency generator is activated.
r UPS batteries and power converters to replace interrupted power for the computer
systems and counting room instruments. The batteries are continually charged
by the regular service system. During power loss, they support orderly shutdown
of computers and instruments with preservation of accumulated data and instruc-
tions for calculations. An individual system is recommended for each piece of
equipment instead of a large hard-wired system to avoid catastrophic data loss
if the hard-wired system fails.
The safety and HP staff responsible for day-to-day laboratory support should
be cross-trained as emergency responders (e.g., first aid, CPR, spill control, fire
suppression). The safety and HP office should be conveniently located near the
sample receiving and laboratory areas, as it is in the top left corner of Figure 13.1.
The office should store emergency equipment and support materials. In addition,
each hallway should have an emergency response cart equipped with a first aid kit,
spill control materials, respirators (if needed), and portable survey instruments. A
fire extinguisher should be in every laboratory and hallway. Emergency response
supplies and carts should be inspected and restocked at designated intervals that
are recorded on labels or tags at each location.
In addition, each hallway and each room should have an emergency (“panic”)
button that will set off an alarm and by-pass the computer-controlled access system
on each door. This system allows anyone to enter a laboratory when an emergency
occurs that requires immediate assistance, or to leave the laboratory. Because
pressing the button triggers a series of events and emergency responses and subverts
security, it should only be used for a true emergency.
One of the operating license criteria for commercial facilities is the preparation
and testing of an Emergency Management Plan (EMP). Arrangements must be
coordinated with the local hospital, fire department, police department, HAZMAT
284 Charles Porter and Glenn Murphy

team, and the city or county governments. The EMP must itemize the potential
hazards at the facility and specific instructions for emergency response person-
nel. For example, the Fire Department may be advised not to enter a building
without respiratory protection or to perform fire suppression from a distance.
The Police Department may be advised to enter a building only when accompa-
nied by designated site personnel. The hospital may be advised that an incoming
radionuclide-contaminated patient should be treated in a designated radiation area
of the hospital. Communications must be concise, clear and rapid during an emer-
gency, and should be limited to selected personnel. Sections 14.5 and 14.6 provide
more information on emergency planning and accident response.

13.8. Controlling Regulations and Licensing Requirements

Some of the design for, and much of the work at, radioanalytical and mixed-
waste laboratories is controlled by regulations and professional standards that
must be incorporated into the decision making process. The architect, builder and
design team members are expected to be familiar with such regulations. Laboratory
managers are advised to familiarize themselves with the regulations and criteria
pertaining to their laboratory.

13.8.1. Laboratory Design and Modification

New laboratories must be constructed in accordance with Federal, State, and local
building codes. At the date of writing, more than 100 regulations are directly
applicable to the building process. The many groups cited in Appendix B are
considered important as regulators and criteria resources, but there will be others,
depending on the nature of the facility and its location. The architectural firm
that designs the facilities should submit all pertinent regulations as part of the bid
specification package. At least one member of the laboratory planning team should
review the bid specifications to ensure that they contain the current and applicable
regulations and guides.
Design modifications for laboratories that are renovated or upgraded by replac-
ing individual components must meet current regulations because they are rarely
“grandfathered” under the old regulations. The Federal Facility Compliance Act
of 1992 includes rigorous specific requirements for the final design of the facility.
To meet these requirements, renovation of older structures may cost more than
new construction. Once an existing structure is modified, any problems with the
entire structure become the responsibility of the modifying party. One prominent
example from newspaper headlines is the widespread requirement for asbestos
remediation. In some cases, the NRC may require the facility to institute design
modifications, called “back fitting,” regulated under 10 CFR 50.109.

13.8.2. Operating Licenses

The Nuclear Regulatory Commission (NRC) regulates management and control
of radioactive materials. The Atomic Energy Act of 1954, section 274, allowed
 13. Laboratory Design and Management Principles 285

responsibility to be transferred to individual states for controlling sources of radi-

ation. One requirement for approval is that state controls be just as restrictive, if
not more so, than those of the NRC. If approved by the NRC, controlling authority
is transferred to the state, thereafter called an Agreement State.
Agreement State programs were developed to bring a modicum of control to
the state level, but they do not cover all handling of nuclear materials. For in-
stance, all import and export of nuclear materials is controlled by the NRC, as is
the construction of any nuclear material production or utilization facility. Many
waste disposal issues also fall under the aegis of the NRC. The NRC has the
ultimate authority to control all nuclear materials deemed pertinent to the de-
fense and security of the United States. More information on NRC and Agree-
ment State regulations is online at http://www.hsrd.ornl.gov/nrc/index.html (Dec.
2005).
In the U.S, either the NRC or one of the currently 33 Agreement State radio-
logical control programs issues the radioactive material license. This license is the
controlling force for the design, staffing, and operation of a radioanalytical chem-
istry laboratory. License categories include Byproduct Materials, Special Nuclear
Materials, and Generally Licensed Materials, with some sub-categories. The usual
laboratory needs a Broad Scope Byproduct Materials license and a separate Spe-
cial Nuclear Material license by the governing authority. A license application
must be submitted to and approved by the licensing agency before operation can
begin. License approval requires that the laboratory is fully organized, with a
qualified Radiation Safety Officer; laboratory staff with demonstrably appropriate
education, training, and experience; an acceptable waste management plan; spill
countermeasures; and suitable radiological instrumentation. These items and many
more must be fully described in a Radiation Protection Plan (see Section 14.2) that
can be distributed as a Radiation Safety Manual.
The laboratory also may be required to report to the EPA its radionuclide quanti-
ties or to monitor airborne emission from operating stacks and by other discharges
under the National Emission Standards for Hazardous Air Pollutants (NESHAPS).
Subpart H protects the public and the environment from radioactive materials (other
than radon) emission at DOE facilities and subpart I applies to other federal facili-
ties, including NRC licensees. The basic criteria are that the annual effective dose
equivalent to any individual
r must not exceed 10 mrem, but
r must not exceed 3 mrem from radioiodine.

The EPA provides the computer code COMPLY for calculating the radiation
dose from release of radionuclides to air. The operator must use measurements of
stack velocity/volume flow rate and radionuclide amounts to calculate radiation
doses by COMPLY. A laboratory may qualify for an exemption from monitoring
requirements if specific radionuclide amounts do not exceed limiting values.
Clients need a copy of the laboratory license to verify permission to receive
samples that contain radionuclides. In turn, the laboratory should request a copy
of the client’s license. Samples not covered by the client’s license become the
responsibility of the laboratory that accepts them.
286 Charles Porter and Glenn Murphy

13.8.3. Laboratory Decommissioning

The Agreement States and the NRC require that, as a condition of granting the
operating license, a decommissioning plan and a budget are available in the event
that a laboratory is dismantled. An alternative to a sequestered budget is a surety
bond. The decommissioning plan formulates the final release criteria, the survey
methodology, survey instrumentation, and the final status survey report. This rule
is intended to protect the public from the environmental hazards of an abandoned
and decaying laboratory site, and to protect taxpayers from bearing the cost of the
resulting cleanup.
Decommissioning of buildings, facilities and structures for government labora-
tories must be in accord with Federal regulations in 10 CFR 835, Occupational
Radiation Protection and DOE Order 5400.5, Environmental Radiation Protection
(online at http://www.directives.doe.gov as of Dec. 2005). All equipment valued
over $5,000 is inventoried and tracked, and the final disposition is directed by
regulations,
Commercial facilities are closely regulated in terms of release of facilities and
equipment for unrestricted use. The radioactive materials license has specific re-
quirements for disposal of radioactive sources and trace materials, even when the
useful life of these radioactive materials has been exceeded.

13.9. Laboratory Service

Any laboratory can survive for a brief period, but long-range sustainability depends
on the management skills of the director and the reputation of the laboratory.
While commercial laboratories have come and gone, the successful ones have the
following characteristics in common:
r A well-developed business plan with sufficient capitalization.
r Skilled managers, professional analysts, and support staff.
r State-of-the-science facilities, methods and instruments.
r Client-service orientation in analysis, scheduling, and reporting.
r A reputation for of institutional integrity.

Facility design was discussed earlier in this chapter while the other topics are
addressed below.

13.9.1. Business Planning and Capitalization

Laboratory planning must begin by identifying the clients that have a high prob-
ability of sample submission, and estimating the expected sample analysis rate.
The planner must match the sample type and analytical load to sample processing
methods, qualified staff, number and type of radiation detection instruments, labo-
ratory design, and ancillary staff and facilities. These factors must be considered in
 13. Laboratory Design and Management Principles 287

TABLE 13.4. Estimated new laboratory

construction and operating costs
General Category Cost ($1 M USD)
Design and Engineering 1.5
Land and Site Preparation 2.0
Construction 11.5
Equipment and Supplies 1.5
Personnel and Training 3.8
TOTAL $20.3

terms of location, availability, controlling regulations, operating cost, and project

capitalization.
Table 13.4 outlines project capitalization at the time of writing for a new lab-
oratory designed to analyze approximately 1,000 samples per month for both ra-
diological and hazardous chemical analyses. Initial capitalization is about twenty
million dollars to build the laboratory and begin processing samples. A definite lag
time should be expected until sufficient contracts are in place to fund operation at
or near capacity. An estimated 65 percent capacity—the approximate break-even
point in terms of net profit—may be reached after two years. Unless near-capacity
sample loads are obtained sooner, an additional amount of about nine million dol-
lars to pay two years of salaries, staff benefits, supplies, and maintenance should
be added to the initial capitalization, pushing start-up costs to over $29 million
dollars.
Not surprisingly, few new companies have the resources to construct a new fa-
cility and keep it operational until it reaches profitability. A tempting alternative
is the “bootstrap” approach of starting a much smaller laboratory—one that pro-
cesses a minimum feasible 125 samples per month—and expanding it as business
warrants. Few such attempts have been successful. The more common approach
is construction of the new, full-scale, laboratory as part of an ongoing program.
The decision to move forward on constructing a new laboratory must be fully
developed in an operational business plan, with sufficient detail to apply for and
receive adequate project funding. A large part of this effort is determining the
analytical cost structure and refining the operating plan of the laboratory.

13.9.2. Determining Analytical Costs

A set of default parameters must be estimated at the outset of operation to provide
the client with cost of analytical procedures. A sample cost list for radioanalytical
chemistry services is illustrated in Table 13.5. The earlier cost data are from a 1990
business plan developed by the authors and the 2004 cost data were extracted from
cost proposals by a commercial laboratory. The magnitude of the cost for other
analyses can be inferred from their similarity in processing to the cited methods.
Costs did not change greatly during these 14 years. The modest changes may re-
flect increased costs of instruments, salaries, and supplies, balanced by cost savings
288 Charles Porter and Glenn Murphy

TABLE 13.5. Radioanalytical chemistry costs

Estimated Cost
(USD)

Analysis Comments 1990 2004

Gamma-ray Spectral Analysis Any geometry, up to 2 h count time $73 $105
Alpha-particle Spectral Uranium or Thorium $170 $156
Analysis—Soil Plutonium $170 $166
Alpha-particle Spectral Uranium or Thorium $170 $156
Analysis—Water Plutonium $170 $166
Total Uranium $40
Liquid Scintillation Analysis Water, swipes or air samples $30 $20
Soil (tritium distillate) $67 $83
Soil (direct count) $30 $20
Screening Smears Gross alpha/beta $13
Beta (LSC, up to 2 isotopes) $20
Screening—Gross Alpha and Soil $67 $39
Gross Beta Water $67 $39
Air Filter $51 $51

with improved analytical methods. The laboratory cost list should be updated pe-
riodically to reflect such changes.
The basic cost of a radioanalytical chemical method is estimated from the person-
hours devoted to analysis from sample receipt to result submission. The associated
cost of supplies, supporting efforts, instrument use, and overhead is apportioned
among the various analyses that are performed during the same period. Personnel
cost is affected by the extent of processing needed to preserve and dissolve the sam-
ple and to remove interfering radionuclides. In essence, the simpler the process, the
lower the cost. The degree of precision and magnitude of detection level are other
cost determinants to the extent that they affect analytical skill, instrumentation,
and time requirements.
The quality assurance program, whether the normal effort or with special items
added by the client, is a major ancillary cost. The usual overhead costs of items
such as management, structures, maintenance, employee benefits, and taxes always
must be considered. Special costs may be invoked by studies to test or improve
methods of purification and measurement, explain unanticipated results, or answer
questions that arise during a project.
The above-cited factors suggest the potential benefits in achieving the most ap-
propriate cost by interacting with the client in preparing the Quality Assurance
Project Plan (see Section 11.3.2). Once the information needs are understood, an-
alytical efforts may be expanded or reduced with associated cost changes. Results
of initial screening analyses may suggest cost-reducing specification changes.
Scheduling the sample flow to the laboratory and the turnaround time for data
reporting may reduce cost by optimizing assignment of laboratory staff and
instruments.
 13. Laboratory Design and Management Principles 289

13.9.3. Minimum Detectable Activity or Concentration

A defined minimum detectable activity (MDA) or concentration (MDC), as dis-
cussed in Section 10.4, should be estimated for the result of each procedure, based
on the planned sample matrix, sample volume or mass, analytical method, count-
ing method, and counting time. If the client requests an MDA or MDC below
the default value, the laboratory needs to consider adjusting the method, with
an associated increase in cost. For example, a larger sample volume can reduce
the MDA/MDC, but the larger sample volume may increase the analytical cost
and time. A longer counting time will reduce the MDA/MDC but can impact the
turnaround time, and require more counting equipment. Judicious planning and
communication with the client can help meet individual client needs, but the labo-
ratory director may be limited in the options offered to one client by commitments
made to other clients.

13.9.4. Turnaround Time

Every client wants sample results delivered as soon as possible. The laboratory
operation plan should include a reasonable estimate of analysis completion time
for every method to present the client with the turnaround time. The specified
period begins when the sample enters the laboratory door and ends when the
analytical result is transmitted to the client. This period must include sufficient
time for ancillary activities related to the sample and for unscheduled events that
occasionally incur delay. The time requirement for the individual method then
must be placed into the context of other analyses for this and other clients.
As an example, consider a client who requires total uranium analysis of 600
samples to be delivered at the beginning of each quarter. If the analysis schedule
indicates a processing rate of 60 samples per day, then the sample load requires a
minimum of 10 days for analysis. This period must include the following activities:

r Receiving and logging 600 samples into the system

r Preparing the samples for analysis
r Preparing reagents and maintaining instruments
r Incorporating quality control samples and measurements into each batch
r Processing samples and performing measurements
r Reviewing the data
r Checking the QC samples for acceptability
r Performing additional analyses or data review as needed
r Approving results report for delivery to the client

Turn around time can be increased by situations such as workload distribution

(i.e., delay while other samples are processed), analysis difficulties, weather in-
terruptions, employee absence, and external audits. These issues may increase the
minimum time by a factor of two or three. Such eventualities must be considered
in providing your client with a reasonable turnaround time.
290 Charles Porter and Glenn Murphy

A fine line separates operating the laboratory at capacity and being over-
extended; unanticipated events often cause a laboratory to cross the line between
full capacity processing and missed sample deadlines. Each project manager must
coordinate with the laboratory director and technical staff to ensure that no one
project or event detrimentally affects another client or overall laboratory operation.
If these disruptions occur consistently, one or more of the following changes can
be made:
r Increase staff levels.
r Add off-shift work.
r Install more counting instruments.
r Reduce the project load.
r Reorganize the sample-processing schedule with selected priorities.

A longer-term option is to increase the level of automation in the laboratory. In-

troduction of automated processes can benefit the processing schedule and engage
staff in more efficient and creative activities. Automation is discussed in Chap-
ter 15.
One final caveat: samples that contain hazardous materials such as volatile
organic compounds (VOC’s) or short-lived radionuclides must be analyzed within a
limited time period. VOC’s, regulated under 40 CFR because of their high volatility,
must be analyzed within 7 days of collection. If the deadline is exceeded, the sample
is invalid. Short-lived radionuclides must be analyzed before they decay below the
detection limit.

13.9.5. Integrity
A reputation for integrity is vital to the successful operation of the radioanalytical
chemistry laboratory, as for any analytical facility. The client depends on the
validity of the reported analytical results in operating a program or performing
a task. In turn, the client provides information based on the analytical results to
various stakeholders, including workers, the public, and regulators, to demonstrate
safe operation or absence of excessive risk. Even a minor instance of unreliable
data can call into question entire reports of results and thus, the operation of the
client’s program.
Laboratory management must institute elaborate controls to guarantee data re-
liability (See Sections 10.5 and 10.6). Supervisors’ must overview the work of
analysts and instrument operators. A quality assurance program that typically in-
cludes comparisons with other laboratories, calibrations, quality control analyses
and measurements, and numerical comparisons of results should be instituted to
ensure accurate results. Periodic audits, questioning staff, and searching records
should confirm that all personnel are following written instructions. These reviews
have greatly expanded over the past several years in response to falsified results;
the cause of falsification ranged from minor errors to major incompetence and
even deceit.
 13. Laboratory Design and Management Principles 291

Institutional controls can reassure the client and other users of data that any
problems will be found and corrected, but the main quality assurance effort must
be devoted to instilling integrity in laboratory staff. Managers and supervisors
must preach integrity at every opportunity; more importantly they must demon-
strate integrity in their practices and orders. While emphasizing production and
timeliness, they must place greater value on following specified procedures and
achieving correct analytical results. Chemists, instrument operators, and support
staff must be trained to follow procedures, report any deviations from protocol or
suspicions of unreliable instrumentation, and consider a questionable result to be
unacceptable. Integrity can be defined as always doing the right thing, especially
when no one appears to be watching. The laboratory director’s responsibility is
equating a correct result to “the right thing.”
A final aspect of integrity involves the interaction of the facility with the com-
munity in which it operates. The laboratory needs to maintain good will within the
community by being open, honest, and responsive to concerns as they are raised.
Most commonly, the concerns will center on laboratory waste and effluent, but may
also involve questions about the security of hazardous material and the soundness
of the facility’s emergency planning. Unless the facility is viewed as a good neigh-
bor and partner in the community, it risks becoming a target of animosity that will
inevitably damage its operations and morale.
14
Laboratory Safety∗
ARTHUR WICKMAN1,5 , PAUL SCHLUMPER2 , GLENN MURPHY,3
and LIZ THOMPSON4

14.1. Introduction
Laboratory instruments and the chemicals used in preparing samples can create
conditions that range from relatively benign to highly hazardous. If not identified
and controlled or eliminated, these hazards can expose the laboratory worker to
injuries and illnesses and cause damage to property and the environment. When
even an experienced radioanalytical chemist begins to work in a laboratory, it
behooves the supervisor to present clearly and with emphasis on safety the docu-
mented procedures of the laboratory and the behavioral practices that are expected
of the employee. The chemist should respond with attention and concern. Each
individual in the laboratory must be given and accept the primary responsibility
for safety. The safety culture must flow from management to laboratory worker
and must be embraced by each individual.
The reader was introduced to concern for a safe working environment and re-
sponsible behavior in the laboratory at his/her first experience with laboratory
activities. This chapter is intended to build on that initial training, by providing the
reader with a comprehensive discussion of safe laboratory practice from preven-
tion measures to emergency protocols. The discussions in this chapter focus on the
radioanalytical chemistry laboratory, whether it is a government, contractor, or aca-
demic research laboratory. Safety planning is discussed first to lay the groundwork
for more specific discussions to follow. Safety officers are identified and their duties
are described, together with the safety plans they administer. Workplace hazards,

∗
This text owes the genesis of its content to Dr. Isabel Fisenne, Department of Homeland
Security. The authors would like to thank Dr. Alena Paulenova of Oregon State University
for her careful review of the text.
1
Health Sciences Branch, Georgia Tech Research Institute, Georgia Institute of Technology,
Atlanta, GA 30332
2
Safety Engineering Branch, Georgia Tech Research Institute, Georgia Institute of Tech-
nology, Atlanta, GA 30332
3
The Matrix Group, Inc., 188 Hidden Lake Dr., Hull, GA 20646
4
Environmental Radiation Branch, Georgia Tech Research Institute, Georgia Institute of
Technology, Atlanta, GA 30332
5
Arthur Wickman, Georgia Institute of Technology, Atlanta, GA 30332-0837; email:
art.wickman@gtri.gatech.edu

292
 14. Laboratory Safety 293

safety precautions, and environmental safety in these laboratories are addressed,

noting when necessary the special considerations that accompany a specific lab-
oratory environment. The regulatory environment is discussed to indicate to the
reader the laws that enforce safe work practices in the U.S.

14.2. Safety and Health Management Systems

A successful safety structure must be well-planned and well-documented. By de-
veloping and implementing an integrated safety and health management system,
supervisors are supported in ensuring a safe and healthy working laboratory. A
safety and health management system has the following main elements:
r management commitment and employee involvement;
r hazard identification and control;
r laboratory worker training; and
r periodic program review and revision.

Commitment of management and involvement of workers (in this case, lab-

oratory analysts), are essential. Management commitment to worker safety
requires:
r training workers to work safely;
r implementing radiological protection policy and practices based on the ALARA
(as low as reasonably achievable) principle for radiation workers;
r identifying pertinent Occupational Safety and Health Administration (OSHA)
standards, as per the Title 29, 1910, series of the Code of Federal Regulations
(CFR 2004);
r emphasizing individual accountability for compliance with safety standards;
r involving workers in identifying existing and potential workplace hazards and
controlling them;
r implementing a process to manage hazards through prevention, mitigation and
control;
r empowering workers to exercise their Stop Work authority in response to hazards;
r ensuring compliance with applicable requirements and regulations; and
r providing adequate resources for safety management.

Involvement by workers requires their consideration of safety and health as an

integral component of the work in the laboratory, for their own sake as well as
that of other workers in the laboratory. A safety culture is necessary to ensure that
proper procedures are followed. In addition to conducting laboratory operations
in a safe and responsible manner, the radioanalytical chemist can involve himself
in the safety culture by:
r participating in the planning process prior to selection and assignment of tasks;
r aiding in the preparation and review of job hazard analyses;
294 Arthur Wickman et al.

r making the immediate supervisor aware of the hazards and their possible controls
for proposed methods; and
r communicating constructively with safety staff by identifying concerns and re-
sponding to guidance.
Hazard identification and control are important aspects of safety in a laboratory.
Most hazards in a laboratory environment involve either unsafe conditions or
behavior. Conditions can be controlled through proper analysis and inspection of
the work environment, and implementation of controls to reduce or eliminate the
exposure to these hazards. A formal job hazard analysis, where individual tasks
are observed, broken down into their individual components, and analyzed for
existing and potential hazards is necessary for hazard identification and corrective
action. This activity must be followed by periodic formal inspections and hazard
assessments.
Unsafe behavior may be difficult to predict and control. Laboratory worker train-
ing is thus an important component of a safety and health management system.
Before they even set foot in the laboratory, workers must be aware of the potential
hazards to which they might be exposed, the protective measures they should take
regarding those hazards, and the management procedures and policies regarding
safety and health. Training is usually conducted for new workers as an orienta-
tion and then periodically in refresher courses. Training also is required whenever
a change occurs in the laboratory environment related to procedures, materials,
or equipment. Finally, training is an important aspect of corrective action, should
observation indicate that a laboratory worker is performing a task unsafely. Individ-
uals who demonstrate unsafe work practices require additional training to ensure
they have proper knowledge of safe work practices. If this remedy is ineffective,
disciplinary action must be instituted.
To close the loop in the safety and health management system, periodic assess-
ment and feedback are necessary. Indicators should be chosen that can assess the
overall performance of the laboratory with respect to safety and health. Whenever
possible, “leading” indicators such as behavioral observations should be measured
and reviewed, as well as “trailing” indicators such as the type and number of in-
juries and illnesses and loss of working time. The purpose of this assessment is to
determine the overall effectiveness of the safety and health management system
and to correct any areas of deficiency.
This safety and health management system constitutes a safety framework with
elements that should be implemented in all laboratories. More specific tools to
delegate the responsibility of safety management in the radioanalytical chemistry
laboratory are described in the following section.

14.3. Chemical Hygiene and Radiation Safety Plans

and Staffing
Day-to-day oversight for maintaining and operating a safe work environment
is delegated to two groups: the Industrial or Chemical Hygiene Office and the
 14. Laboratory Safety 295

Radiation Safety or Radiological Control Office. The industrial hygienist (IH) or

chemical hygiene officer (CHO) is concerned with the overall safety and comfort
of the laboratory workforce. The radiation safety officer (RSO) is concerned more
specifically with radioactive materials. The IH/CHO and RSO staffs function un-
der different laws and regulatory agencies, but have parallel responsibilities that
organizationally may be either combined or separate.
Management must support the efforts of these offices with necessary funding
as well as respect for their expert opinion on workplace concerns. As a corollary,
management must support rational efforts to ameliorate identified issues or areas of
concern with regard to worker/workplace safety and health. As a basic principle,
the laboratory supervisor must not countermand a decision to halt a work plan
because of health, safety, or radiation concerns by the IH/CHO or RSO, and the
work plan must be revised to meet these concerns. Descriptions of the duties and
responsibilities of the CHO and RSO, and the plans they administer, are given
below.

14.3.1. Chemical Safety Plans and Staffing

All laboratories must have a written plan—the Chemical Hygiene Plan (CHP)—
which describes the provisions that have been made for safety by the labora-
tory managers. This requirement is regulated by OSHA under standard 29 CFR
1910.1450, “Occupational Exposure to Hazardous Chemicals in Laboratories.”
The CHP sets out the specific procedures, work practices, safety equipment and
personal protective equipment that have been selected to provide employee protec-
tion for the hazards found in each laboratory. An individual university or college
laboratory will follow the CHP of the institution, which applies to all laboratories
on campus; the laboratory may also have a CHP that is specific to its individual
conditions. College students should follow the provisions of the applicable CHP
to ensure their safety. The CHP must include:
r standard operating procedures for safety and health;
r engineering controls needed to limit exposures;
r operational and maintenance criteria for engineering controls
r provisions for safety training;
r identification of hazardous chemicals; and
r any permits or special procedures required for the laboratory.

Additionally, the CHP may contain the emergency plans for the laboratory (see
Section 14.5).
The CHP must be administered by a CHO, an individual qualified by training
and experience to handle safely the chemicals and processes in the laboratory who
also has the authority to take corrective actions when needed. The CHO must be
familiar with all safety and environmental regulations that apply to the laboratory.
The CHO must understand the health and safety hazards of the chemicals in use.
The CHO must be aware of any biological monitoring required for employees
or students who use regulated chemicals. To be effective, the CHO must have
the support of managers and administrators, and must have a clear mandate to
296 Arthur Wickman et al.

enforce safety requirements. Hazardous conditions identified by the CHO must be

corrected, and the CHP must be revised to prevent any recurrence of the hazard. The
CHO must review and accept or modify any changes to chemicals or procedures
in the laboratory.

14.3.2. Radiation Safety Plans and Staffing

A radioanalytical chemistry laboratory requires a Radiation Safety Manual (RSM)
or Radiation Protection Plan (RPP) in addition to a Quality Assurance Plan (QAP).
The format is controlled by the licensing agency for the facility; the Nuclear Reg-
ulatory Commission (NRC) requires the RSM, while the Department of Energy
(DOE) requires the RPP. These plans are stand-alone documents that deal specifi-
cally with radiation safety issues and practices to set safe operating parameters in
the laboratory.
Both the NRC and DOE provide guidance documents to aid the user in the devel-
opment and implementation of the RSM or RPP. Topics concerning radionuclides
and radiation that are routinely controlled under the program include:
r ALARA principles;
r individual training requirements;
r individual dosimetry requirements;
r air concentration levels;
r residual contamination levels;
r exposure rate levels;
r instrumentation;
r routine operating procedures;
r emergency procedures;
r waste disposal options; and
r record-keeping and reports.

Each worker in the laboratory must undergo initial radiation safety train-
ing that prepares him/her to work with radioactive materials. Refresher training
must brief the employee about changes to operating procedures and regulatory
requirements.
Radiation safety staffing levels may vary widely among facilities on the basis
of facility size, radioactivity levels, and number of samples processed. Each facil-
ity is required to have an RSO that meets the education, training and experience
requirements by the NRC or DOE. The DOE protocol is for government labora-
tories while NRC and Agreement State licenses control other radionuclide-using
entities, including commercial and academic laboratories.
The NRC and Agreement States have specific license requirements for RSO
education, training and experience. To practice this specialty, the RSO must meet
the requirements of 10 CFR 35.900 series Subpart J—Training and Experience Re-
quirements. Three different sets of experience can qualify an individual to become
an RSO:
 14. Laboratory Safety 297

r certification as a Certified Health Physicist by the American Board of Health

Physics in comprehensive health physics;
r demonstration of successful completion of 200 hours of specified classroom and
laboratory training, plus one year of full-time experience at a medical institution
under the supervision of a qualified RSO; or
r work experience for an authorized user under an NRC license.

The DOE does not use the license concept, so there are no formal require-
ments to become a Radiological Control Manager (RCM)—the equivalent of
an RSO. Regardless of title and regulatory mechanism, the RSO or RCM is re-
sponsible for implementing a safety program for the use and control of radioac-
tive materials in the laboratory. The RSO/RCM is responsible for the following
activities:

r maintaining the facility license (for the NRC or the Agreement State) or the
Radiological Protection Program Plan (for the DOE);
r procuring, receiving, and delivering the radioactive material to the individual
user;
r training the user in routine handling and emergency response procedures;
r implementing an internal and external dosimetry program;
r performing routine surveys in radioactive materials use areas;
r disposing of radioactive waste; and
r controlling any other item pertinent to radiation protection (for example, a posi-
tion description may specify “5% of time spent on other duties as assigned”).

The RSO is responsible for working with the radiation laboratory worker in
planning any program that may result in radiation exposure to persons and ra-
dionuclide contamination of the work place or the environment. The radioanalyt-
ical chemist should be able to depend on the RSO for guidance in minimizing
the potential for such exposure and contamination. In brief, radioactive material
must be stored, handled, and discarded separately from and differently than the
usual laboratory reagents. The RSO must inform the regulator, management, and
the worker of personnel radiation exposure and any unusual radiation exposure
conditions.

14.3.3. Responsibility of the Radioanalytical Chemist

The newly employed analyst must read carefully the laboratory CHP and RPP to
assimilate information vital to the safe operation of the laboratory and the safety
of its workers. The analyst should make certain that the CHP and RPP reflect
any changes in procedure, and are reviewed and revised on a scheduled basis,
usually biannually, to address changes in legislation and institutional policies. The
following discussions consider aspects of worker safety that are included explicitly
in the CHP and RPP of a given laboratory.
298 Arthur Wickman et al.

14.4. Hazards in the Workplace

Laboratory work must be conducted with the realization that safe operation is only
achieved by constant focus on safety. The point is not to be afraid, but to be aware
of potential hazards and responsible for one’s behavior. Hazard identification is
crucial for accident prevention. Adverse effects can be avoided when both the
manager and the worker make safety a priority. This section identifies the primary
hazards in the radioanalytical chemistry laboratory, and ways to minimize the
possibility of accidents.

14.4.1. General Hazards and Precautions

Hazards common to any environment include tripping over an obstacle or slipping
in a spill. Floors must be kept clear of any equipment or stored chemicals. Any
spills on floors must be immediately cleaned. Any power cords or flexible hoses
that cross floors must be secured so that they do not present a tripping hazard.
The chemistry laboratory has additional hazards. Open flames, ovens and fur-
naces can cause fires and burns. Cold storage rooms and their contents, notably dry
ice and liquid nitrogen, may cause freeze burns. Corrosive acids and other chemi-
cals can cause chemical skin burns and internal damage from inhalation, ingestion
and absorption through the skin. Many chemicals are poisonous: chemicals should
not be tested by taste or smell. Compressed gases constitute inhalation, explosion
and fire hazards if handled improperly.
The radioanalytical chemistry laboratory is subject to all of these hazards with
the addition of radioactive solids, liquids and gases. The potential risks to laboratory
workers from radioactive materials include exposure to ionizing radiation, both
from external sources and from internal sources that were inhaled, ingested, or
absorbed through the skin. Contamination of the immediate work area is a safety
concern that becomes amplified if the contamination is not removed promptly and
is subsequently spread over a much wider area. Each of these hazardous situations
may result from an unintended release or from improper handling of radioactive
materials. Often, the person who is affected is not the one who caused the problem
in the first place.
Radiation hazards should be placed in perspective by distinguishing between
lower and higher levels of radionuclides in samples submitted for analysis, although
these are qualitative terms not defined by regulation (see Section 13.2). Many man-
made radionuclides in the laboratory are at levels so low that they are comparable
to naturally-occurring radionuclides in soils, rocks, biota, and urine, such as 40 K
in potassium salts and uranium and thorium isotopes and their progeny in uranium
and thorium salts. At these lower levels, the radiation hazard in the laboratory
can be viewed as minor compared to chemical hazards. Samples with significantly
higher radionuclide levels should have been described appropriately in the shipping
and/or chain of custody documents and must be processed to minimize radiation
exposure to the analyst on the basis of guidance by the RSO or RCM staff.
 14. Laboratory Safety 299

Carelessness, absentmindedness and roughhousing increase the likelihood that

a potential hazard will occur. Avoid pranks, horseplay, and other inappropriate
behavior in the laboratory. Be respectful of the presence of others in the laboratory,
and do not crowd, push, or gesture carelessly.

14.4.2. Design and Maintenance

The work environment should be designed and maintained for safety. The labora-
tory described in Chapter 13 represents the ideal: spacious, well organized and easy
to clean. In many laboratories that are older or have outgrown their space, workers
must be particularly vigilant to avoid spills and breakage in cramped operating and
storage spaces, keep aisles clear, and ensure unimpeded access to the accident re-
sponse equipment described in Section 14.6. Laboratory exits and electrical power
disconnect panels must never be blocked. In all laboratories, but especially in older
ones, management should regularly inspect the condition of plumbing, electrical
hardware, lighting fixtures and ventilation systems, and repair what is defective.
Good housekeeping augments good design in reducing the potential for acci-
dents. Analysts should keep work spaces clean and organized, have only a limited
supply of chemicals on hand and store the remainder safely, label chemicals clearly,
and dispose of waste promptly and correctly. Only the equipment needed for an ac-
tive procedure should be set up, and any unused equipment should be stored away
from the laboratory bench. At the conclusion of an assignment, all equipment
should be dismantled, cleaned, and stored.

14.4.3. Glassware and Chemical Handling

Glassware should be inspected prior to use to remove chipped, cracked, or broken
items from service and dispose them in a designated container. Soiled glassware
should be taken to the laboratory sink designated for cleaning. Clean glassware
promptly with detergent and water. Vacuum glass apparatus should be shielded
to contain contents and glass fragments in the event of an implosion. Except for
glass tubing and stirring rods, all glassware in the laboratory should be composed
of borosilicate (e.g. Pyrex).
Laboratory refrigerators must be explosion-proof and specifically designed for
chemical storage. Standard household appliances do not meet these criteria. All
containers placed in refrigerators should be labeled with the identity of the contents,
the name of the person placing the container in the refrigerator, and the date of
storage. Containers in refrigerators should be placed on storage trays with sides of
sufficient height to contain any spilled or leaking containers. No food or beverages
shall be stored in the laboratory refrigerator.
Once chemicals have been logged into the receiving area, they must be moved
immediately to a specified storage area, as described in Section 13.3. Large con-
tainers that could break during transport must be transported in a protected shipping
container or a secondary protective container large enough to hold the contents
in the event of a spill. Chemicals must be segregated by hazard classification and
300 Arthur Wickman et al.

compatibility, and stored in a secure area equipped with local exhaust ventilation.
The storage area should be continuously ventilated and under slightly negative
air pressure, well lighted, supervised, and capable of being locked to prevent un-
authorized access. Chemicals should be stored at or below eye level. Large bottles
should be no more than two feet from ground level, and not stored on the aisles
or floor. Mineral acids should be stored on acid-resistant trays, separately from
flammable and combustible materials. Acid-sensitive materials, such as cyanides
and sulfides, should be protected from contact with acids. At least annually, the
inventory of stored chemicals should be examined to identify evidence of con-
tainer deterioration or damage and to identify any chemicals to be replaced based
on length of service. Chemicals that are no longer needed should be disposed of
properly.
Chemicals taken from the storage area for use at the laboratory bench should be
transported in durable secondary containers of sufficient size and composition to
retain any spill of the chemical. Preparation or repackaging of chemicals should
take place in a designated area of the laboratory, outside of the main storage
area. Chemicals stored at the laboratory bench should be limited to the amounts
necessary for current application. Bench chemicals should be stored in an orderly
manner, away from sunlight and external sources of heat. Containers should be
protected from falling by placing them well away from the edges of counter tops.
Containers may not be stored on the floor. Chemicals may not be removed from
the laboratory without permission. Unnecessary items should not be stored at the
laboratory bench.
Material Safety Data Sheets (MSDS) are compiled and available for all chem-
icals. These documents explain the inhalation and contact health risks, as well
as the flammability and toxicity of the chemical they describe. They also outline
special handling and storage considerations. When handling a new chemical, it is
prudent to examine its accompanying MSDS.
With or without an MSDS, however, caution in chemical handling should be
observed. As a general guideline, worker and student exposures to laboratory
chemicals should be kept to a minimum. Because so many laboratory chemicals
are hazardous to humans in some way, conservative risk assessment should be
employed. Persons in the laboratory should assume that personal protection is re-
quired whenever they are working with chemicals. Chemicals of unknown toxicity
initially should be treated as toxic with respect to exposure during work performed
in the laboratory. For work with chemicals of known toxicity, appropriate precau-
tions should be taken. When working with mixtures of chemicals, the risk of the
mixture as a whole should be assumed at least to equal the risk for the most toxic
component of the mixture.
Skin contact with chemicals should be avoided. Personnel must wash all areas
of exposed skin prior to leaving the laboratory. Mouth suction for pipettes or for
starting siphons is not allowed. Prohibitions on eating, drinking, smoking, gum
chewing, or the application of cosmetics in the laboratory should be posted, and the
rules concerning these activities should be enforced by the supervisor. Hands and
face should be thoroughly washed prior to eating or drinking. Protective aprons,
 14. Laboratory Safety 301

jackets, or coats should be removed prior to entry into areas where food may be
consumed.

14.4.4. Personal Protective Equipment

All laboratory workers must wear appropriate personal protective equipment
(PPE). Safety goggles must be worn in laboratory areas. Face shields, laboratory
coats or aprons, latex, rubber or heat-resistant gloves, dust masks and safety shoes
may be required for specific work. Understanding the hazards of the materials at
hand is important for wearing the appropriate PPE, but also for shunning excessive
use of PPE. Overprotection may cause awkwardness and increase the possibility
for injury. To strike a balance between reasonable and inordinate safety measures,
the safety officer and analyst should communicate on how best to perform a given
operation. In some laboratories, there may be a HAZMAT-certified PPE technician
that can help in making these choices.
Proper dress for the laboratory includes shoes with leather or synthetic covering
over the feet. Open toed shoes, sandals, high heels, or shoes with woven fabric
coverings are not allowed. Minimize skin exposure by avoiding the use of shorts
or cutoffs; pants are preferable. Minimize the use of jewelry, remove any hanging
jewelry, and secure loose hair and clothing.
Street clothing must be protected by the use of chemically resistant laboratory
coats, jackets, or aprons. Laboratory jackets should be equipped with snaps, not
buttons, so that they can be removed quickly in case of an emergency. Protective
clothing must be fire-resistant and impervious to chemical splashes.
Gloves must be selected based on the chemical or hazard of concern, and
on the extent of expected exposure. Factors to consider in selecting protec-
tive gloves include proper fit, durability, cut resistance, chemical breakthrough
(permeation), chemical resistance, comfort, and cost. Manufacturers of protec-
tive gloves are required to test glove materials and to make data available
concerning permeation, diffusion, and chemical degradation of glove products.
Some gloves cover only the hands. In other applications, gloves should cover
the forearms or the entire arms. Certain individuals may have allergic reac-
tions to the proteins found in latex gloves, and should use suitable substitutes
made of non-latex materials. Gloves that allow permeation or diffusion of spe-
cific chemicals should not be reused. Remove gloves prior to handling ob-
jects which are not part of the laboratory procedure, and always remove them
prior to leaving the laboratory. Wash hands thoroughly whenever gloves are
removed.
Chemical splash goggles must be worn at all times by all personnel in the
laboratory. Visitors to the laboratory must don goggles prior to entry. Safety glasses,
face shields, or prescription glasses may not be substituted for goggles. Contact
lenses may be worn under goggles; they should never be adjusted or removed during
laboratory procedures except in emergency conditions. When a vigorous reaction,
implosion, or splashing is possible, a face shield should be worn in addition to
goggles.
302 Arthur Wickman et al.

14.4.5. Safety Equipment

Safety equipment includes trays, hoods, glove boxes and radiation shields. Shal-
low metal or glass trays are used for nonvolatile substances to contain potential
spills and leaks of materials such as radionuclides. They isolate specific substances
that can react in a hazardous manner with others processed on the same labora-
tory bench. Trays also serve to define work stations for specific purposes, and to
distinguish them from clean areas for notebooks and reagents.
Laboratory hoods must be used for all procedures that might release hazardous
chemical vapors or dusts. Wet chemistry for volatile substances—notably certain
organic solvents and radionuclides—should be performed in a hood to reduce
inhalation risks. Chemicals that pose a substantial risk of explosion must be handled
in a hood designed to contain explosions.
Hood fans must be operated whenever hoods are in use, and the fans should
continue to run following the end of work for a sufficient time to clear residual
air contaminants from the ductwork. Hoods should be equipped with a meter to
record air flow or air pressure, and users should confirm that hoods have proper
face velocity. The laboratory must be supplied with sufficient make-up air to satisfy
all air exhausted through the hoods. When work at the hood is concluded, the sash
should be lowered.
Hoods should not be designed to recycle exhaust air into the laboratory. “Hoods”
that do not vent to the outside, but rather filter in-laboratory air, are called laminar
flow boxes. These are important for working with dusty radioactive materials, like
soils, but are not appropriate for most other laboratory operations. Laminar flow
boxes should be clearly labeled so that laboratory policy can limit the analytical
operations that can be performed in them.
Hoods should not be used to store chemicals. Because they often are a common
workspace for several analysts, hoods tend to accumulate glassware, instruments
such as heater/stirrers, reagents, waste and unfinished samples. Extended storage
of such materials in the hood constitutes bad housekeeping that increases the
likelihood of accidental spills. Stored materials can disrupt airflow which causes
the hood to operate less efficiently. If chemicals are left temporarily in the hood,
the fan should continue to operate with the sash slightly opened.
Laboratory hoods should be inspected at least every three months. The CHO,
industrial hygienist, or designated technician should measure the face velocity of
hoods periodically to ensure that 75 to 125 feet per minute (0.4–0.6 ms−1 ) air
velocity is maintained. The sash opening height should be properly marked to
indicate the face velocity. A transport velocity in the exhaust duct of the hood of
3500 feet per minute (18 ms−1 ) is recommended. A record of the inspection should
be dated and posted at the hood. Prior to a change in chemicals or procedures, the
adequacy of the ventilation system and the suitability of the hood for the new
process must be evaluated.
While many operations with chemicals must take place in a hood, rarely would
glove boxes or shields be required in the radioanalytical chemistry laboratory dedi-
cated to processing environmental or bioassay samples. A glove box or shield may
 14. Laboratory Safety 303

be needed when an analyst is working with radioactivity standards before dilution

(which may be as much as a million times as radioactive as low-level samples), or
for an unusually radioactive sample. An environmental laboratory generally would
not accept a highly radioactive sample because of concern about contamination,
but an emergency situation may necessitate its analysis. Should this unusual situ-
ation arise, the responsibility of the laboratory analyst and RSO/RCM staff is to
insist on issuance of suitable protective equipment. Failure by the supervisor to
provide equipment for safe operations should automatically trigger a “Stop Work”
notification by the analyst. The “Stop Work” notification applies to all phases of
laboratory work, not just radionuclide analysis. The RSO/RCM or IH/CHO staff
should document such an incident in writing, and promptly submit the report to
management.

14.4.6. Flammable and Combustible Liquids

A wide variety of chemicals can be used in any laboratory environment, includ-
ing those chemicals that have the properties of being flammable or combustible.
Before discussing some of the general precautions to take in handling and storing
flammable and combustible liquids, the following definitions are necessary:
r Flashpoint: The minimum temperature at which a liquid gives off vapor within
a test vessel in sufficient concentration to form an ignitable mixture with air near
the surface of the liquid.
r Combustible liquid: Any liquid with a flashpoint at or above 100 degrees Fahren-
heit. The two classes of combustible liquids are Class II liquids with a flashpoint
between 100 and 140 degrees Fahrenheit and Class III liquids with a flashpoint
above 140 degrees Fahrenheit.
r Flammable liquid: Any liquid with a flashpoint below 100 degrees Fahrenheit.
Flammable liquids are also known as Class I liquids. There are three subcate-
gories of Class I liquids. Class IA liquids have a flashpoint below 73 degrees F
and a boiling point below 100 degrees F. Class IB liquids have a flashpoint below
73 degrees F and a boiling point at or above 100 degrees F. Class IC liquids have
a flashpoint at or above 73 degrees F but below 100 degrees F.

Basic precautions taken for flammable or combustible liquids consist of sep-

arating the liquids from sources of ignition or from oxygen. This approach can
prevent or extinguish combustion by separating from each other the fuel, source of
ignition, and oxygen. The following are a few general precautions to be considered
when storing or using flammable or combustible liquids:
r Minimizing stored quantities: According to the OSHA requirements in 29
CFR 1910.106, Class IA flammable liquids must be limited to a maximum of
25 gallons (95 L) of liquids in containers in storage in any one fire area of a
facility. Other flammable or combustible liquids must be limited to a maximum
of 120 gallons (450 L) of liquids in containers.
304 Arthur Wickman et al.

r Eliminating or isolating sources of ignition: Among the most common sources

are smoking, electrical arcing or static, flames such as Bunsen burners, and heat-
producing equipment. These potential sources of ignition should be eliminated
in areas where flammable or combustible liquids are stored or used. Smoking
must be prohibited in all laboratory areas. Electrical equipment in areas where
flammable vapor-air mixtures could occur must be suitable for these locations.
Sources of ignition such as flames or heat-producing equipment must be sep-
arated from areas where flammable or combustible liquids are stored or used.
A good rule of thumb is to have a distance of at least 20 feet (6 m) between
any source of ignition and flammable or combustible liquids, but more stringent
requirements may apply in specific situations.
r Utilizing flammable storage cabinets or rooms: Larger quantities of flammable
or combustible liquids should be stored in approved flammable liquid storage
cabinets or storage rooms. Construction specifications must be met for different
sizes of storage container and class of chemical stored therein; 29 CFR contains
the specifications for construction of flammable liquid storage buildings (29 CFR
1910.106(d)(5)(vi)) and for construction of flammable liquid storage cabinets
(29 CFR 1910.106(d)(3)).
r Separating incompatible chemicals: Chemicals that may react with each other
violently should be stored separately. For example, acids should be separated
from bases, and oxidizers should be separated from flammable and combustible
liquids.

14.4.7. Compressed Gas Cylinders

Compressed gas cylinders utilized in a laboratory environment can be hazardous.
The hazard can be posed by both the type of gas in the cylinder and the pressurized
nature of the gas. Sudden release of this pressure can result in a cylinder being
thrust into the laboratory as if it were a rocket. This can occur if the valve is broken
off when the cylinder is knocked over.
The main precaution with respect to compressed gas cylinders is to train lab-
oratory workers in their proper handling, storage, and use. Workers should re-
spect the potential for hazard of the specific gases that might escape and of
sudden pressure releases. Cylinders in storage should be securely attached to a
permanent object or wall, or kept strapped on carts that are designed for holding
cylinders. Valve protection caps, where cylinders are designed to accept them,
must be kept on cylinders when they are in storage. Flammable gas and oxy-
gen cylinders should be stored away from each other by at least 20 feet (6 m)
or be separated by a 1/2-hour-rated fire wall which is at least 5 feet (1.5 m)
high.
Cylinders should be stored in the upright position, especially flammable—gas
cylinders that are designed to be stored in this manner. Because OSHA generally
considers cylinders that are not used within a 24-h period to be “in storage,” oxygen
and acetylene cylinders that are not in daily use should be taken off the cart and
 14. Laboratory Safety 305

stored properly. As an alternative, the above-cited firewall to separate the cylinders

can be made an integral part of the cart. When moved, cylinders should always be
strapped securely to a cart designed for carrying them.

14.4.8. Electrical Hazards

The following is a list of the more likely hazards related to electricity that may be
encountered in a laboratory environment:

r Lack of a permanent and continuous ground: This hazard can result from a
broken or removed ground plug or a wiring problem at an electrical receptacle.
With an open ground, a fault in equipment can result in the path of electricity
through a laboratory worker instead of the desired ground path. Laboratory
inspections must ensure that ground pins are correctly in place for equipment
designed to have them and that receptacles are tested periodically for correct
wiring.
r Unused openings or missing covers in electrical equipment or boxes: Electri-
cal equipment can become damaged or neglected over time, with exposure to
energized electrical parts through unused openings such as knock-out closures
or missing covers. Exposure to energized electrical parts, especially parts op-
erating at over 50 Volts, must be prevented. Electrical parts should be serviced
or maintained only after de-energizing the circuit and following the established
lockout/tagout procedure. When de-energizing is not feasible, only qualified
electricians operating after training and with personal protective equipment and
insulating tools should attempt to work in the vicinity of energized electrical
parts.
r Damaged insulation: Damaged insulation must be repaired immediately by a
qualified electrician. Simply covering insulation with electrical tape will most
likely not be sufficient to protect individuals from contact. The repaired section
must have the same mechanical strength and insulating properties as the original
wiring.
r Wet or damp environments: Water can be a good conductor of electricity, hence
working with electrical equipment in a wet environment can be a hazardous
activity. Electrical equipment should be isolated from moisture. Electrical re-
ceptacles in wet or damp environments must be designed for this type of envi-
ronment, with ground-fault-circuit-interrupter (GFCI) protection. This type of
receptacle should be tested periodically in accordance with the manufacturer’s
recommendations.
r Circuit overload: Electrical equipment should not be utilized in a manner not
intended by its manufacturer. Overloaded wiring and receptacles can damage
the equipment and result in electrical shock, electrocution, and fire. Equipment
installers and users should always understand the intended use of the equipment
and follow manufacturer’s recommendations.
306 Arthur Wickman et al.

14.4.9. Unattended Operations

Unauthorized procedures and working solo in the laboratory are not permitted.
Laboratory equipment that operates unattended for continuous or overnight pro-
cessing may pose a risk of fire, explosion, or other unintended consequences. Any
persons planning to use unattended operations in the course or an experimental
series must receive written approval from the laboratory supervisor. The approval
must provide sufficient detail to show that the planned procedure will operate safely
to completion. The permit should include the anticipated completion time and the
name and telephone number (or other means of communication) of the person who
is responsible for operating the equipment and dismantling it upon completion.
Entrances to the laboratory must be posted to provide warning of the laboratory
procedure that is in progress, as well as the name and telephone number of the re-
sponsible person. Lights in the laboratory should be left illuminated. Precautions,
such as fail-safe operational mode, must be integral to the equipment being used,
so that an interruption of utility services such as water, gas, compressed air, or
electricity will still allow the process to shut down safely.

14.4.10. Monitoring
In the radioanalytical chemistry laboratory, higher- and lower-level radiation areas
must be clearly designated. When specifying the radiation level of an area, signs
akin to those in Figure 14.1 would be posted. There are many variations of this
type of sign, as they are placed anywhere radiation is used or found—including
hospitals, laboratories, chemical plants and waste clean-up sites. Each of these
locations has different radiation concerns, and a different audience viewing the
signage. What all signs will have in common, however, is the tri-foil. Depicted in

FIGURE 14.1. High-level radiation alert signage. Online at:

http://www. epa.gov/radiation/students/symbols.html (Dec. 2005).
 14. Laboratory Safety 307

FIGURE 14.2. The tri-foil. Online at:

http://www.epa. gov/radiation/students/symbols.html (Dec. 2005).

Figure 14.2, the tri-foil is the international symbol for radiation. The symbol can
be magenta or black, on a yellow (or sometimes white) background.
Access between the two appropriately labeled radiation areas should be re-
stricted, in order to avoid contaminating the low-level area. In laboratories where
both “hot” and “cold” operations are being performed in a small amount of space,
it is important to set up well-defined and clearly labeled radioactive material work
stations. In this way, radionuclide analysis is confined so that its impact on the
surrounding laboratory is minimal. The presence of radiation throughout the lab-
oratory should be monitored as discussed below.
Because the five senses are useless for detecting radiation, each facility must
have readily available portable radiation detection instruments. These instruments
should be selected to detect and quantify the three basic types of radiation: alpha
particles, beta particles, and gamma rays, as discussed in Chapter 2. In some
cases, neutron detectors may be required. The RSO/RCM generally is responsible
for calibrating the instruments at selected intervals, typically six months. The
individual user is responsible for daily operational and source checks prior to each
use.
Each radiation safety program should have an established survey program to
monitor the radiation and contamination levels in the laboratory so that the in-
dividual user can maintain a safe working environment. The measurements can
identify and measure individual radionuclides or measure radiation dose. The
dosimetry program should monitor exposures to individuals from external radi-
ation or internal—ingested or inhaled—radionuclides, and also in areas of the
laboratory.
External exposure levels generally are monitored with a thermoluminescent
dosimeters (earlier, with film badges) or electronic dosimeters. Either method
provides individual exposure information for a worker or a location in terms of ex-
posure over a pre-set monitoring period, after which the dosimeters are exchanged
for new ones. The used dosimeter is read to determine the radiation exposure dur-
ing the interval. In laboratory areas that process samples with higher radionuclide
concentrations, wrist and ring badges or direct-reading radiation dosimeters also
may be worn, but they usually are unnecessary when processing environmental
or bioassay samples. An exception may be made for persons who receive mixed
samples at the dock or handle radioactivity standards before dilution.
The internal dosimetry program generally relies on in vitro (outside) or in
vivo (inside) monitoring capabilities. In vitro programs use an external radiation
308 Arthur Wickman et al.

detection system to monitor radiation being emitted from inside the body and de-
tected on the outside of the body. In vivo programs use a sample from the body
(usually urine, but sometimes feces, blood, hair, or breath) to monitor contamina-
tion within the body.
Urine samples for bioassay are collected from workers who may have inhaled
or ingested radionuclides or hazardous chemicals, either while routinely handling
samples with higher radionuclide or chemical concentrations, or after accidental
exposure. The samples are analyzed for the radionuclides or chemicals that may
have been taken in by the worker and the measured concentrations are used to
calculate radiation exposures or to compare to radionuclide or chemical concen-
tration limits. The frequency of sampling for bioassay depends on the occurrences
of exposure and the rate of turnover of the radionuclides or chemicals in the body.
A large facility usually has an infirmary with medical staff for treating injuries
and performing physical examinations at the beginning and end of employment,
and at regular intervals. The type and frequency of routine physicals may depend on
the probability of adverse effects associated with the work environment. Employee
medical records should be maintained to track the occurrence or absence of health
effects due to the work environment.
Management must promptly report to the worker and the regulatory agency
any unusually elevated radiation exposures—both external and internal—or con-
centrations of hazardous chemicals, and institute changes in the work environ-
ment to prevent continuing exposures that reach exposure limits. The IH/CHO or
RSO/RCM staff must be instrumental in achieving such changes. The exposure
must be recorded with a description of the causes and instituted remedies.

14.5. Emergency Response

Among the various types of emergencies that can occur in a laboratory environ-
ment, occurrence of a fire is a major probability. The best way to minimize the
effect of emergencies is to prevent them. For this reason, fire prevention plans
should be instituted. If a fire or other emergency does occur, emergency plans
must be in place to protect the laboratory and its workers. These plans may be
a part of the laboratory CHP, or they may stand alone. Various issues associated
with life safety must be considerd to maximize the occupant’s ability to escape the
facility during an emergency.

14.5.1. Fire Prevention

Each laboratory facility must have a fire prevention plan, preferably in writing.
According to OSHA 29 CFR 1910.39, the fire prevention plan should include at
least the following elements:
r A list of all major fire hazards, proper handling and storage procedures for
hazardous materials, potential ignition sources and their control, and the type of
fire protection equipment necessary to control each fire hazard;
 14. Laboratory Safety 309

r Controls to prevent accumulation of flammable and combustible waste materials;

r Procedures for regular maintenance of safeguards installed on heat-producing
equipment to prevent accidental ignition of combustible materials;
r The name or title of those individuals responsible for maintaining equipment to
prevent or control sources of ignition or fires; and
r The name or title of those individuals responsible for the control of fuel source
hazards.
Laboratory workers must be aware of the fire prevention plan and must be trained
in the elements of the plan and its location within the laboratory.

14.5.2. Emergency Planning

Emergency action plans are especially necessary in laboratories. Plans should be
prepared for response to situations such as fire, explosions, flooding, severe weather
(tornadoes, hurricanes), earthquakes, medical emergencies, violent acts or threats
of violence (e.g., bombs), and release of hazardous materials. Accidents at nearby
locations such as adjacent laboratories and transportation facilities may affect the
laboratory worker. According to OSHA 29 CFR 1910.38, the minimum elements
of an emergency action plan include:
r Procedures for reporting a fire or other emergency;
r Procedures for emergency evacuation, including type of evacuation and exit route
assignments;
r Procedures to be followed by laboratory workers who remain for necessary
laboratory operations before they evacuate;
r Procedures to account for all laboratory workers after evacuation;
r Procedures to be followed by laboratory workers performing rescue or medical
duties; and
r The name or title and address (e.g., telephone or e-mail) of every individual to
be contacted by laboratory workers who need more information about the plan
or an explanation of their duties under the plan.

14.5.3. Life Safety

The National Fire Protection Association (NFPA) produces an entire text to address
the complex issues of life safety in the Life Safety Code (NFPA 101) that should
be read for detailed information. The following are general items to ensure that
laboratory workers can safely evacuate the laboratory in an emergency:
r At least two exit routes should be available to permit prompt evacuation during an
emergency. The two routes should be located as far as practical from each other in
case one is blocked. More than two exit routes may be necessary, depending on the
number of occupants, size of the building, building occupancy, or arrangement
of the workplace, while a single route may be allowed in some instances.
r Occupants must be able to open an exit route door from the inside at all times
without keys, tools, or special knowledge of door operation. Exit route doors
310 Arthur Wickman et al.

must be free of any device or alarm that could restrict emergency use of the exit
route if the device or alarm fails.
r Exit routes must support the maximum permitted occupant load of each floor
and the capacity (width) of an exit must not decrease in the direction of travel.
r The minimum exit width is 28 inches (0.7 m), but should be wider in most
circumstances.
r Exit routes must be arranged so that occupants will not have to travel toward a
high-hazard area (such as a flammable liquid storage room), unless the path of
travel is effectively shielded from the high-hazard area by physical barriers.
r Exit routes must be free and unobstructed.
r Exit routes must be adequately lighted so that an occupant with normal vision
can see along the exit route. The lighting must be impervious to power outage, so
backup lighting must be on an independent power supply (generator or batteries).
Automatic illumination by these lights should be tested periodically.
r Each exit must be clearly visible and marked by a sign reading “Exit”.
r Laboratories must install and maintain an operable alarm system that has a
distinctive signal to warn occupants of fire or other emergencies.

14.6. Accident Response and Equipment

Most safety guides emphasize hazard identification and accident prevention. When
prevention has failed, personnel must be familiar with the laboratory CHP and RPP
accident protocols and the emergency plans described in Section 14.5.2. These
plans instruct the worker about what to do and whom to notify. This information
includes the facility evacuation routes. Qualified personnel must be prepared to use
the accident response equipment placed in the laboratory to ameliorate an accident
situation or to aid in treating the injured. Examples of this equipment include fire
extinguishers, eyewash stations and safety showers.

14.6.1. Fire Extinguishers

Fire extinguishers should be accessible in every laboratory, mechanical support
area and electrical room. Several extinguishers may be warranted in a large space;
the OSHA rule is that one should not have to travel more than 75 feet (23 m) to
reach an extinguisher (29 CFR 1910.157(d){1}). Fire extinguishers also must be
located in the hallways outside the laboratory. The employer is responsible for
proper selection and distribution of fire extinguishers on the advice of the safety
officer.
Fire extinguishers must match the type and class of fire hazards associated
with the particular laboratory. Four fire hazard classes are defined by the U.S.
Department of Labor: Classes A, B, C and D.
r Class A fires result from the burning of ordinary combustible materials, such as
paper or cloth. Class A extinguishers may be water, loaded stream (a water-based
 14. Laboratory Safety 311

extinguisher that contains one of several added chemical components that in-
crease permeation of the stream into the fire), foam or multipurpose dry chemi-
cal. The numerical rating refers to the amount of water or other fire extinguishing
agent that the extinguisher holds.
r Class B fires are fueled by flammable liquids and gases, or grease. Class B
extinguishers may be Halon 1301, Halon 1211, carbon dioxide, dry chemicals,
or foam. Here, the numerical rating indicates the approximate number of square
feet of Class B fire the average user can be expected to extinguish using that
extinguisher.
r Class C fires are electrical fires, and may be extinguished with Halon 1301, Halon
1211, carbon dioxide, or dry chemical fire extinguishers. Class C extinguishers
have no additional numerical rating.
r Class D fires involve flammable metals, such as magnesium, sodium, potassium,
titanium and zirconium. Extinguishers that contain water, gas, or certain dry
chemicals cannot extinguish or control this type of fire and may actually fuel it.
Class D fire extinguishers utilize agents including Foundry flux, Lith-X powder,
TMB liquid, pyromet powder, TEC powder, dry talc, dry graphite powder, dry
sand, dry sodium chloride, dry soda ash, lithium chloride, zirconium silicate,
and dry dolomite. There is usually no numerical rating for these extinguishers.
An excellent resource for fire safety and fire extinguishers can be found at
http://www.hanford.gov/fire/safety/extingrs.htm. This website is maintained by
the Hanford Fire Department in Richland, WA, which oversees fire safety and
operations for the DOE’s Hanford facility.
The old symbols for fire extinguishers geared for each type of fire are shown in
Figure 14.3. Figure 14.4 contains the new symbols. Note that the new pictograms
have no symbol for Class D extinguishers, which are often specific to the type of
metal burning.
Among Classes A–C, crossover in the active firefighting agent results in some
extinguishers that are Class A/B (for example, foam) or Class A/B/C (some dry
chemical extinguishers). All fire extinguishers are labeled with their class hazard
rating to insure that the extinguishers are properly used. Before attempting to put
out a fire, check the label to confirm that the extinguisher is appropriate. In the old
format, the symbol for each Class of fire that the extinguisher is designed to address
will be on the extinguisher bottle. In the new format, all three pictograms are on
the extinguisher bottle, with a red strike through any Class that the extinguisher

A B C D
Ordinary Flammable Electrical Combustible
Combustibles Liquids Equipment Metals

FIGURE 14.3. The old symbols, representing Class A, B, C, and D fire extinguishers.
Online at: http://www.hanford.gov/fire/safety/extingrs.htm. (Dec. 2005).
312 Arthur Wickman et al.

Electrical
Equipment

FIGURE 14.4. The new pictograms, representing Class A, B, and C fire extinguisher. There
is no pictogram for Class D extinguishers. Online at:
http://www.hanford.gov/fire/safety/extingrs.htm. (Dec. 2005).

is NOT designed to address. One should remember that water is only useful in
putting out Class A fires. A mixed-use extinguisher will never contain water, and
the user should never improperly use a water-based extinguisher (Class A) on any
other type of fire.
Fire extinguishers must be inspected routinely to insure that they are in operating
condition. Employees must be trained in the use of fire extinguishers unless im-
mediate evacuation is the facility policy. New employees should check the CHP to
determine their responsibility (some facilities may designate only some employees
as fire officers), and inquire about the laboratory fire safety training program.
In addition to fire extinguishers, a sprinkler system in the ceiling is almost always
available. These are, of course, useful in extinguishing Class A fires. However,
these sprinklers activate automatically, in response to the presence of smoke and/or
debris in the laboratory air. The automatic addition of water to a class B, C or D fire
would be counter-productive and should be avoided. Therefore, the fire-producing
potential of material in each room should be evaluated, prior to installation of the
sprinkler system.

14.6.2. Eyewash Stations and Washdown Showers

Chemical laboratories must be equipped with a combination emergency eyewash
and shower. The unit should be located at a distance of no more than 10 seconds of
level, unobstructed travel time from anticipated exposure points. As an estimate,
100 feet (30 m) can be traveled in 10 seconds if the pathway has no obstacles. The
location of the emergency wash must be clearly marked, well lighted, and easily
accessible. No obstacles or doorways should be located on the travel path, which
should have few turns.
Initiation of the emergency wash must be accomplished by one action using
one hand. Once initiated, the flow must continue, leaving both hands free. Con-
taminated clothing should be removed as quickly as possible while the emer-
gency wash is operating. For contaminants in eyes, the unit should be operated
for 15 continuous minutes, with both hands used to hold the eyelids open. Emer-
gency washes should be flushed weekly for a minimum of 3 minutes. Bump tests
should be conducted on eye washes daily and on showers weekly. A full flow test
 14. Laboratory Safety 313

should be conducted monthly. Emergency washes should provide tepid, potable

water. Eyewashes should supply 3 gallons (12 L) of water per minute for at least
15 minutes. Showers should provide 30 gallons (120 L) per minute for 15 minutes.
Fountain heads for eyewashes should be covered to protect them from collecting
dust, debris, or chemical residues.
Eyewash stations and washdown showers should be checked periodically for
proper flow because they are infrequently used. The saline solution at stand-alone
eyewash stations should be replaced at its expiration date, or sooner if it becomes
dried out.

14.6.3. Accident Protocols

All workers must know where to go and what to do when an accident happens. The
laboratory CHP and RPP contain the emergency plan with which all laboratory
workers must be familiar. In a serious accident, workers should respond in the
following order unless a different protocol has been specified for the laboratory:

1. Stop a fire, accidental spill, or gas leak at its source, if possible. This means ex-
tinguishing or removing the flammable source of the fire, stopping the chemical
leak, or turning off the gas. If these activities are too dangerous, do not attempt
them; isolate the room and proceed to the next step.
2. Raise the alarm for those in the area to evacuate.
3. Assess injuries and assist the injured, removing them from the area for medical
attention, if possible.
4. Notify the designated response group promptly and provide as much informa-
tion as can be obtained in a brief period. The designated responder will contact
the fire department, police, and HazMat, based on the information provided. Be
as calm and accurate as possible when relaying the details of the accident.

Note: The OSHA regulations do not prescribe a specific order of operations in an

emergency, in recognition of the differences among laboratories; considerations of
location, accessibility and institutional protocol must be addressed in determining
a plan. The above recommendations, generally accepted among university CHPs,
are published online, but are not a substitute for familiarity with the accident
protocol in one’s own laboratory.

HazMat (see Section 14.9) publishes a guide to aid first responders in (1) quickly
identifying the specific or generic classification of the material(s) involved in the
incident, and (2) protecting themselves and the general public during this initial
response phase of the incident. This publication is called the Emergency Response
Handbook, and can be downloaded online. The handbook is updated every three
to four years to accommodate new products and technology. The next version is
scheduled for 2008.
314 Arthur Wickman et al.

14.7. Environmental Safety

Protecting the worker is the primary concern of the Radiation Safety and Indus-
trial Hygiene staffs, but their roles also extend to protecting the environment and
members of the public. It would seem obvious that a program that is designed to
protect the individual working in close proximity to the radioactive materials will
also protect the general public and the environment. However, more restrictive
protection levels are designated for members of the public. To ensure that these
standards are met, laboratories must control, and in some instances monitor, liquid
and airborne effluents as well as radiation levels that may affect the public.
An environmental safety program begins with the construction of the laboratory
and development of the chemical and radiation safety plans. Good pre-planning
will go a long way toward minimizing waste generation, releases to the environ-
ment, and resulting pollution levels. A chemical and radioactive material tracking
system should be used to determine where and how reagents are used and how
much waste is generated. The results should be reviewed periodically to identify
items or quantities of materials that can be reduced. Since waste disposal costs
are a significant part of the operating budget, waste minimization can help the
budget.
As a part of the operating license, both the NRC and DOE specify airborne
and liquid release quantities to the air or the sanitary sewer system, respectively.
Programs must be established that track discharges relative to the limiting con-
centrations. Effluent and waste control regulations must be understood by a staff
expert and communicated to the users of radioactive materials and chemicals.
Compliance with these regulations must be established on an institution-wide ba-
sis. Unusual or accidental discharges must be reported promptly to the supervisor
and RSO/RCM staff. Management must inform the appropriate regulator of any
discharges that reach reporting requirements. If the facility fails to comply, legal
actions can result, including fines or loss of license for on-site activities.

14.8. The Regulatory Environment

Laboratory management, the IH/CHO and RSO/RCM staffs, and those respon-
sible for receiving and shipping materials must be familiar with the regulations
and underlying laws that control the operation of the laboratory, handling of haz-
ardous and radioactive material, and release of these materials, as well as workers’
rights and responsibilities. The radioanalytical chemist should have a reasonable
understanding of regulations that are pertinent to the work, and thus should be
familiar with the Code of Federal Regulations (CFR) titles already mentioned in
this chapter.
The CFR is the comprehensive regulatory framework of the US. It is kept online,
current and in its entirety, by the National Archives and Records Administration
(NARA) at http://www.gpo.gov/nara/cfr/cfr-table-search.html#page1 (Dec. 2005).
When Congress passes a law that dictates a change of policy, the agency under
 14. Laboratory Safety 315

whose aegis the policy falls is prompted to produce a corresponding regulation.

These agencies (e.g., EPA, NRC, DOE, OSHA, DOT) write the regulations and
submit them for expert and public comment; the result is modification or approval.
Each of the 50 CFR titles contains the regulations pertaining to a particular subject
and indicates the government agency (or agencies) responsible for administering
those regulations.
For instance, Title 29 (29 CFR) concerns labor regulations and defines the
purpose and responsibilities of the Department of Labor and OSHA. Title 10
defines energy regulations and the role of the DOE and NRC. Title 40 describes
the regulations for protection of the environment and the EPA. The NARA website
above is a useful resource in researching all 50 titles; the radiochemist should
be conversant with pertinent aspects of titles 29, 10 and 40, in particular. The
following paragraphs are a brief overview of these titles.
All private sector research laboratories, whether academic or commercial,
are subject to U.S. government regulation under OSHA in 29 CFR 1910.1450,
“Occupational Exposures to Hazardous Chemicals in Laboratories.” For public
sector research laboratories, OSHA defers regulatory authority to the applicable
state government and its designee, such as the University Board of Regents. The
descending levels of regulation are state, municipal and institutional or local re-
quirements. In general, state regulations must at least meet, but may be more
stringent than, the OSHA standards. Any state program must be approved and
monitored by OSHA.
Title 10 CFR Part 20—“Standards for Protection Against Radiation”—
establishes the legal framework for protecting non-government workers from ra-
diation. The NRC uses Parts 1 through 199 to govern the use and control of
radioactive materials. The DOE uses Parts 200 through 1099 to govern the use and
control of radioactive materials. Title 10 CFR Part 835—“Occupational Radiation
Protection” establishes the legal framework for protecting government workers
from radiation.
Nuclear materials are specifically regulated under 10 CFR 1.42, administered
by the U. S. NRC, Office of Nuclear Material Safety and Security (NMSS). The
duties and responsibilities of the NMSS include protecting the public health and
safety, national defense and security by licensing, inspection, and environmental
impact assessment for nuclear facilities and activities involving nuclear materials,
and for the import and export of special nuclear materials.
Emissions from specific sources of air pollution linked to potentially serious
health problems are limited by the National Emissions Standards for Hazardous
Air Pollutants (NESHAPS) in 40 CFR 61. The NESHAPS regulations are a subset
of the EPA Clean Air Act. Hazardous air pollutants (HAPS) are identified as
chemical or radioactive hazards. The radiological NESHAPS regulations pertain to
pollutants that emit radiation from sources such as academic institutions, hospitals,
industry, vehicles, and nature (such as radon).
Solid wastes are managed under the requirements of the EPA Resources Con-
servation and Recovery Act (RCRA). The provisions of RCRA pertain to solid
chemicals as well as waste mixtures that are hazardous on the basis of ignitability,
316 Arthur Wickman et al.

corrosivity, reactivity, or toxicity. Some states regulate RCRA-related activities

under approval by EPA.
The US Department of Transportation (DOT) controls transportation of haz-
ardous and radioactive materials under the Hazardous Materials Transportation
Act by regulations in 49 CFR 171–179. Shipments of samples to the laboratory
and of waste from the laboratory must meet these regulations.

14.9. Available Guidance

Individuals responsible for monitoring safety and health in a radioanalytical chem-
istry laboratory have a variety of sources from which to obtain information. Some
of the organizations that are relevant to this topic are included in the following list.
Since the most recent information available can be found by searching online, a
web address has been included for each entity. All were viewable as of December
2005.
r The American Chemical Society (ACS)—A membership organization that con-
sists of individuals working in all fields relating to chemistry. The ACS has
developed a series of booklets to convey the essentials of the OSHA regu-
lations for chemical laboratories in a concise and readable form. To assist
in understanding the regulation, defining the responsibilities for implementa-
tion of 29 CFR 1910.1450 and guidance in audit and inspection, three ACS
booklets are available (ACS 1990, 1998 and 2000). The ACS also publishes
a guide specifically designed for academic laboratories (ACS 2002). Online at
http://www.chemistry.org/portal/a/c/s/1/home.html.
r American Industrial Hygiene Association (AIHA)—A professional member-
ship organization comprised of industrial hygienists or occupational health spe-
cialists. The AIHA publishes a protocol guide entitled “Laboratory Chemical
Hygiene” (AIHA 1995) to establish the framework to implement the OSHA
Standard. The guide presents an overall strategy for the laboratory chemical hy-
giene plan (CHP) to ensure that all individuals potentially at risk for exposure
performing laboratory tasks are informed about the hazards involved and the
means to minimize their exposures. Online at http://www.aiha.org/.
r American National Standards Institute (ANSI)—A non-governmental industrial
consensus institute responsible for developing and distributing a variety of stan-
dards, including those related to safety and health standards (see Appendix A-3
and Section 6.5). Online at http://www.ansi.org/.
r American Society of Safety Engineers (ASSE)—A professional membership
organization comprised of safety engineers. The ASSE publishes findings that
advance the knowledge base for accident reduction and illness prevention. It also
contributes to the development of standards, often working with ANSI. Online
at http://www.asse.org/.
r Environmental Protection Agency (EPA)—A branch of the United States gov-
ernment responsible for developing and enforcing regulations that ensure the
protection of the environment. Online at http://www.epa.gov/.
 14. Laboratory Safety 317

r Health Physics Society (HPS)—A professional scientific organization

whose mission is to promote the practice of radiation safety. Online at
http://www.hps.org/.
r National Fire Protection Association (NFPA)—A non-governmental profes-
sional association whose members specialize in safety and health standards
related to fire protection, including the electrical standards in the National Elec-
trical Code. Online at http://www.nfpa.org/index.asp.
r Occupational Safety and Health Administration (OSHA)—A branch of the
United States Department of Labor (or a delegated State Agency in some states),
this organization is responsible for overall regulation of safety and health as it
relates to occupational exposure. Online at http://www.osha.gov/.
r The Office of Hazardous Materials safety (HAZMAT) is a branch of the United
States Department of Transportation. Their mission is to minimize the risks to life
and property inherent in the commercial transportation of hazardous materials. To
that end, HAZMAT sponsors a variety of training courses across the U.S. These
include 1–2 day seminars and longer courses designed to certify technicians in
a wide variety of areas. Online at http://hazmat.dot.gov/training/training.htm.
In addition to the information available from the organizations listed above,
an extensive website containing Health Physics resources is available from Oak
Ridge Associated Universities (ORAU) (http://www.orau.org/ptp/infores.htm)
(Jan. 2006). The ORAU web site contains links to a number of radiation safety train-
ing programs. Louisiana State University offers an online resource guide to haz-
ardous materials and safety (http://www.lib.lsu.edu/sci/chem/guides/srs103.html)
(Jan. 2006).
Finally, the Health Physics and Radiological Health Handbook(Shleien 1998)
and the AIHA White Book (DiNardi 2003) are commonly used as safety reference
guides.
15
Automated Laboratory and Field
Radionuclide Analysis Systems
HARRY S. MILEY and CRAIG EDWARD AALSETH

15.1. Introduction
Many radiochemical analyses consist of a series of identical steps with little or
no variation from sample to sample. Operator fatigue in the execution of repeti-
tive steps leads to increased process variability and execution errors. Such tasks
are attractive candidates for automation because they can be made more efficient,
consistent, and cost-effective. This trend comes to radiochemistry at a time when
schoolchildren use computers regularly, typing or dictating homework assignments
to a word processor, with other tools such as “spell check” and “word count” to
complete the task. This digital age has also ushered in complex computers predict-
ing molecular, biological, and other highly intricate processes. While an automated
device can perform in a moment the manual computations of a scientist’s lifetime,
however, the same computer would continue computing in error the equivalent of
a thousand lifetimes, while a scientist would perceive the error and stop! Thus,
despite these tremendous information age advances, the automation of physical
and chemical processes must be done thoughtfully to avoid repetition of errors in
quality, lapses in safety, and other pitfalls that a scientist would instantly perceive
and halt.
Automated systems have several potential advantages over classical technical
labor. Analytical functions are reproduced on a programmed schedule, reducing the
turnaround time of the samples and enabling a more reliable estimate of analytical
costs. In addition to the accuracy of the equipment, costs are readily estimated,
the process is easier to scale up or down, and process parameters can be more
easily varied in an understandable way. Automated systems can produce data
from worldwide locations and may be readily standardized and compared. Manual
operation remains preferable for procedures that require considerable judgment in
the selection of alternatives.
Aspects of chemistry for which automation has been a popular and effective im-
provement over manual processes include continuous-flow separation processes,
instrumental analysis, and continuous monitoring. Radiometric measurements are
a good example, as numerous samples are counted, one after the other, each for a

Pacific Northwest National Laboratory, Richland, WA 99352

318
 15. Automated Laboratory and Field Radionuclide Analysis Systems 319

specified time period. Automation can replace manual placement of the samples
for measurement, with a large number of sequential measurements possible with-
out operator intervention. In fact, the sample collection process can be automated
in some cases and then mated to an automated measurement system.
Most radioanalytical chemistry processes may be automated to some degree,
depending on the type of physical manipulation the sample requires and the com-
plexity of the chemical process. In some cases, the essential mechanical manipula-
tion can be achieved with commercially available components originally designed
to automate manufacturing operations. For instance, the pick-and-place automa-
tion developed for the semiconductor and other industries can be readily adapted
to sample measurement. In other cases, automation relies on specially designed
commercial hardware and software tools. Examples of special computer-controlled
automated systems developed for atom-at-a-time detection and analysis of actinide
and transactinide element isotopes with half-lives as short as a few seconds are
shown in Figs. 16.4–16.6. Such techniques have been used to perform the very
first studies of the chemical properties of 104 Rf through 108 Hs.
Comprehensive knowledge of the radioanalytical chemistry procedure is needed
to devise successful mechanical substitutes for the manual steps of the process and
to ensure that the controlling software responds appropriately and predictably. As
might be expected, handling special cases and identifying problematic samples are
key elements. Examples of automated sampling, analytical processing, and mea-
surement are presented in this chapter to illustrate what may be achieved by com-
bining measurement and engineering expertise. These are illustrative of the prin-
ciples of automation and are arranged in order of increasing complexity. While the
engineering aspects of automation are beyond the scope of a radioanalytical chem-
istry text, it is useful to explore some of the ideas needed for a successful experience.

15.1.1. Automated System Control and Data Handling

One advantage of automated radionuclide analysis systems is the potential for
leveraging the benefits of rapidly developing information technology. Despite the
revolution in electronic data processing and associated technology, it is not un-
common today to find world-class measurement systems controlled manually with
handwritten logbooks for data storage, data transfer via human-originated email,
or physical transfer of storage media. The authors are familiar with a world-leading
system managed by use of paper note cards and “carbon-based automation,” known
as laboratory technicians.
By comparison, significant advantages can be obtained from rather ordinary
measurement systems properly linked and employing modern information tech-
nology. An example is the International Monitoring System (IMS), in particular
the network aspect of the automated monitoring systems of 80 stations that com-
prise the radionuclide measurement component. In the aerosol monitoring part of
the IMS network, numerous stations with good automated reporting and commu-
nications provide redundancy. This allows remarkably effective operation even in
the case of tampering, forced outages, or national withdrawal of some stations.
320 Harry S. Miley and Craig Edward Aalseth

Automation adds value to the results recorded. The IMS network provides a vast
database of signals recorded under essentially identical conditions and reflecting
regional variations in backgrounds. This allows temporal filtering to eliminate
false alerts based on the frequency of similar signatures in the past and detailed
characterization of system state-of-health, thus enabling higher confidence in ra-
dionuclide spectral data. The remainder of Section 15.1 examines aspects of control
and data handling for automated systems in more detail.

15.1.2. Spectral and Other Data Types

Several of the monitoring systems discussed in this chapter were designed to use
automated data handling. One feature of this design is the use of specific formats for
various data types. Many good standard data formats exist; a standard format for an
automation application greatly eases integration of disparate software components.
For example, the IMS formats support email transmission of several spectral data
types: normal spectra, calibration spectra, quality assurance measurement spectra,
and system blank spectra. These formats include metadata or extra data-capturing
details of the measurement such as time, calibration values, and sample number.
Other types of data useful in automated systems are state-of-health, log entry data
(automatically generated and human-originated entries), and alert messages. Alert
messages are typically limited to reporting out-of-bounds state-of-health values for
loss of power, overheated motors, detector failure, and similar urgent conditions.
Planning adequate state-of-health monitoring, such as the desired frequency of
status messages and the automatic actions initiated by state-of-health data, are
important design steps.
The recorded operating parameters captured by state-of-health data have great
utility; when a problem exists, the operator can review the most recently collected
state-of-health data to diagnose the problem and determine corrective action. In
addition, state-of-health data can be scanned automatically to attempt failure pre-
diction. Reanalysis of state-of-health data from the systems described in this chap-
ter has identified subtle hardware failures not detected by system operators until
months later.

15.1.3. Command and Control

Transmission of state-of-health information, particularly maintenance-related
data, suggests the next aspects of effective automated systems: command and
control. This is most important for remote field systems, but can make operation
of laboratory equipment easier as well.
Several scenarios exist for creating command and control systems for automated
devices or networks. Ubiquitous and reliable email systems provide a way to issue,
receive, and archive commands. Commands arrive as formatted emails; a local
software interpreter parses the email and initiates the desired function.
Interactive processes, such as setting the energy calibration of a gamma-ray spec-
trometer, do not lend themselves to email or text commands, which provide no
 15. Automated Laboratory and Field Radionuclide Analysis Systems 321

graphical feedback. The most favorable human–computer interface for highly in-
teractive operations is a graphical user interface. These may require larger commu-
nications bandwidth but provide much greater speed of user interaction. Existing
schemes for supporting such interfaces over low-bandwidth connections include
methods for compressing X-Windows protocol traffic. Another solution is to op-
erate a graphical tool on the operator’s end of the connection, with only sparse
command-response transmissions with the instrument. Well-designed graphical
user interfaces require more software development, balanced by greatly reduced
user training.
The broad categories of remote control activities are
1. system inspection commands, or data requests,
2. parameter changes and commands, including Stop and Reboot,
3. interactive sessions, such as energy calibration, and
4. software uploads.
These categories are listed in order of increasing intrusiveness of the remote con-
trol. Reckless system modification can easily result in a disabled instrument con-
figuration with, worse yet, no remote way to recover normal operation. Even in the
early days of automation, it is surprising how often this circumstance resulted in a
technician making travel arrangements. The first and second categories can be han-
dled via email commands or a low-bandwidth graphical user interface command
system. The third category requires a higher bandwidth interface and effectively
must either transmit graphics or use interactive graphical tools located on the oper-
ator’s console. The last category can be implemented most easily with a standard
and secure file transfer protocol.

15.1.4. Data Surety and Authentication

The potential for computer attack against an automated system or conflicts be-
tween poorly coordinated operations arises from allowing connection to remote
machines. An attack can be resisted by operating the system on an independent
network with no possible connection from unauthorized parties. Use of public in-
frastructure, such as public voice or data networks is more cost-effective, but open
to mischievous or malicious attack.
The threat of inappropriate remote control can be addressed by requiring a val-
idated electronic key to be transmitted with the command (Harris et al., 1999).
In the case of the first and second of the command types discussed in Sec-
tion 15.1.3, this can be done by first calculating a hash or checksum values
for the command message. A checksum or hash is a value calculated from the
transmitted data, which is sensitive to any modification of the original data.
Many such algorithms exist in the public domain (Schneier, 2000), e.g. CRC,
RMD160.
Digital signatures and encryption made possible with a public-/private-key in-
frastructure (PKI) (Schneier, 2000) also address command authentication and offer
protection and authentication for potentially sensitive data, e.g., results that could
322 Harry S. Miley and Craig Edward Aalseth

be used within a legal framework or stringent quality assurance regime. Such

techniques can allow immediate authentication that data were not changed after
original encryption or signing. This effectively addresses the problem of data or
command tampering.
The third and fourth command types discussed in Section 15.1.3 can be authen-
ticated and secured via a standard protocol such as secure shell (ssh) (Schneier,
2000). This uses PKI to secure network traffic and provide secure interactive com-
munications and file transfer.
While these concepts address the threat of data or command tampering, the threat
cannot be eliminated altogether. It can be said that the risks from unscrupulous,
mischievous, or malicious actions in automated systems can be made less than in
classical, hand generated, and authenticated laboratory data.

15.2. Manual and Automated Systems Compared: Real-Time

Aerosol Radionuclide Monitoring
To motivate the drive toward automation, consider the example of environmen-
tal radiological monitoring for aerosol fission-product debris in the atmosphere.
When performed manually, as described in Section 15.2.1, the monitoring pro-
cedure requires considerable labor. A robust long-term monitoring program can
be maintained only through procedural controls, training, and low staff turnover
over many years. In Sections 15.2.2 and 15.2.3, automation of the manual sys-
tem is described. These automated methods reduced cost and free staff for other
tasks.

15.2.1. Traditional Air-Filter Analysis

Manual analysis of aerosol fission debris began with the advent of nuclear weapons
testing and involved the collection of atmospheric aerosols on filter paper from
whole-air samples of at least 1000-m3 volume. The filters were then folded into
a good geometry for radiometric measurement, and gamma-ray spectral analy-
sis was performed with NaI(Tl) or Ge detectors. When greater sensitivity was
desired, larger volumes of air were sampled. Eventually, 10,000 m3 of air per
day was filtered for each sample. At the end of a 7-day collection period, the
filter was removed from the filter holder, carefully bagged, and stored for a pe-
riod of a few hours to a few weeks to allow short-lived gamma-ray-emitting
daughters of 222 Rn and the longer-lived progeny of 220 Rn to decay. The filter
was then hydraulically compressed into a small right-circular cylinder 0.5-cm
tall and 5 cm in diameter to provide good measurement geometry. The pre-
pared sample was placed for measurement between two large NaI(Tl) detec-
tors operating in coincidence or, more recently, against a single Ge detector.
After a measurement period selected to provide sufficient precision and sensi-
tivity, the resulting gamma-ray spectra were analyzed for fission and activation
products.
 15. Automated Laboratory and Field Radionuclide Analysis Systems 323

10 000
140B a

1000

100

10
dpm/kSCM

0.1
DL

0.01
DETONATIONS
0.001 USA
USSR
> 10 Mt CHINA
1-10 Mt Chernobyl
0 1-1 Mt
0 02-0.1 Mt
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89

FIGURE 15.1. Temporal distribution of 140 Ba measured in the atmosphere of Richland,

WA. Figure from Perkins et al. (1989). (By permission of Pacific Northwest National
Laboratory)

The concentrations of some 30 airborne radionuclides were measured between

1961 and 1989. Figure 15.1 (Perkins et al., 1989) shows the airborne concentration
of 140 Ba over this period with a detection limit (DL) of about 0.02 disintegrations
per minute (dpm) per m3 .

15.2.2. Automating Spectral Analysis

An automated method of analysis for germanium spectra was eventually adopted
with the advent of sufficient computing power in an ubiquitous platform. This
method was improved until it eventually eliminated all but periodic human QA
checks. This automated method (Gunnink and Niday, 1972) divides analysis into
five tasks:
r energy calibration;
r smooth background computation and subtraction;
r peak location and integration;
r matrix-inversion computation of radionuclide contributions;
r activity calculation/reporting.

Many versions of this basic approach exist, the most significant variation being
whether matrix inversion is used to connect a library of radionuclides to observed
peak intensities or whether a list of energies of interest is used to make key cal-
culations. Thus, the two main types today are matrix inversion and list directed.
Automated spectral analysis software is available from commercial and academic
sources with a mix of national and international quality certifications, special-
ized capabilities, and user control. Programs of this type can handle thousands of
automated analyses per day and run on most types of computers.
324 Harry S. Miley and Craig Edward Aalseth

15.2.3. Real-Time Sampler/Analyzer System

Although advances in computer software allowed for easier methods of spectral
data analysis, the actual collection and preparation of the sample remains a time-
consuming and repetitive task. The integration of collection and analysis steps
was accomplished in the design of the real-time aerosol radionuclide analyzer
collector (R-TARAC). The R-TARAC was created to conduct continuous radiation
monitoring as part of the program to search for airborne species associated with
nuclear proliferation (Smart, 1998). The R-TARAC incorporates a large (140%
relative efficiency) HPGe detector placed in the air stream directly behind an
automatically changed air filter.
This system has been used inside a research aircraft where outside air is ported
into the aircraft and through the system, as well as in an under-wing pod with direct
frontal airflow, as seen in Fig. 15.2. The system continuously measures and records
the concentration of airborne photon-emitting radionuclides. The R-TARAC stores
all collected data and can be automatically or manually controlled from a remote
console to provide instant data reporting to the flight crew and an operations center.

FIGURE 15.2. The R-TARAC, configured for under-wing pod use (top) and for fuselage- or
automobile-based use (bottom). U.S. Patent 6,184,531. (By permission of Pacific Northwest
National Laboratory)
 15. Automated Laboratory and Field Radionuclide Analysis Systems 325

The crew can use such instant data availability to maneuver the aircraft for locating
and monitoring radioactive plumes.
As with any radiation detection system, background is a concern. Ambient
airborne radionuclides (mostly radon daughters) accumulate on the filter and lessen
the detection sensitivity for the radionuclides of interest as compared to a decayed
sample. This accumulation is addressed with a filter carousel, which rotates a fresh
filter into place either at regular intervals or when the background activity level on
the filter becomes excessive.
Another potential background source is contamination of internal and external
surfaces of and near the instrument from exposure to ambient airborne radionu-
clides. These can be deposited from the air stream onto surfaces near the filter
and detector. This background can be expected to increase while the R-TARAC
is exposed to an air stream until the surface contamination reaches mechanical or
decay equilibrium. Tracer tests showed that because of the geometry of surfaces
susceptible to contamination relative to the detector, these deposits contribute
generally less than 1% of the measured radiation. A larger contribution should be
expected after the aircraft has moved through a high-radiation plume into a region
of low-radionuclide concentration.

15.3. Automated Laboratory System

Discussion and Examples
To give a flavor for the kind of features and challenges encountered in automating
laboratory systems, this section discusses the automation of some common lab-
oratory processes and collects a number of examples of automated systems that
have been developed.

15.3.1. General Automation Techniques

System automation is made easier by the availability of many subsystems that are
easily controlled by computer. Process instrumentation of all types and radiometric
measurement equipment is available with standard computer interface options.
Computer hardware and software is available both for simple and complex control
systems. Mechanical equipment for automating the handling of multiple samples
includes pumps, valves, heaters, shakers, vibrating plates, and stirring systems for
mixing samples.

15.3.2. Automated Sample Changers

for Counting Radionuclides
Automated sample-changing equipment has been available commercially for many
years. In liquid scintillation counting (LSC) systems, several hundred vials may be
placed in a train (see Section 8.5.2) for dark adaptation to allow decay of delayed
326 Harry S. Miley and Craig Edward Aalseth

FIGURE 15.3. Automated gamma-ray spectral analysis system. (By permission of Georgia
Institute of Technology)

fluorescence, and then moved, one at a time, to an elevator that brings the sample
between two photomultiplier tubes for counting. After counting, the sample is
returned to the train.
Similarly, planchets for automated gas-flow proportional-counter systems are
stacked in plastic holders. The bottom holder is moved beneath the detector for
counting and then removed to a second stack for storage or recycling.
For gamma-ray spectral analysis, a set of bulk samples can be placed on a
rotating tray that moves each sample in turn next to the massive shield that encases
the detector (see Fig. 15.3). The door in the shield opens and a mechanical arm
places the sample on top of the detector. After counting, the sample is lifted and
returned to the tray. Alpha-particle spectral analysis generally uses no automation
because the samples are counted for a long time.

15.3.3. Automated Laboratory Radionuclide

Column Separation System
Use of ion-exchange columns is a common chemical separation strategy. Elution
of radionuclides from an ion-exchange column occurs as a function of elutriant
volume. For a fixed continuous flow, radionuclides are eluted from the column
as a function of time. The output concentration of a particular radionuclide can
be described by a peak defined by the number of column volumes passed be-
tween the beginning of its elution and the end. Each radionuclide may follow at its
own column-volume parameters. These parameters are a reproducible function of
 15. Automated Laboratory and Field Radionuclide Analysis Systems 327

process characteristics such as column dimensions, flow rate, type of ion-exchange

resin, and elutriant reagent. To automate the collection of column-separated ra-
dioisotopes, an automatic changer for collection containers can be used. The
changer can be linked to an elutriant flow controller for volume-based separa-
tion. For constant-flow columns, the changer can be controlled by timing. The
result is the radionuclide or radionuclides of interest in separate containers.
Flow into columns can be controlled by computer-activated valves from reser-
voirs. One reservoir contains the dissolved sample with the radionuclides to be
separated, another holds a wash solution, and yet another holds the elutriant. First
the sample reservoir is drained into the column, next a timed or measured volume
of wash solution is admitted to the column, and finally a timed or measured vol-
ume of elutriant is added to elute the radionuclide from the column. Additional
reservoirs may hold other reagents for subsequent elution of other radionuclides
and column regeneration.

15.3.4. Automated 99 Tc Separation and Measurement by

Flow-Injection Analysis
Technetium-99 is a fission-product radionuclide that is analyzed widely in the
environment and in radioactive waste. Because of its long (213,000-y) half-life
and high fission yield (6.1%), it can remain at detectable levels after shorter-
lived radionuclides have decayed. The emitted low-energy, beta particles (0.294
MeV, maximum) are usually measured by LS counting after purification to remove
other radionuclides and interfering salts. 99 Tc also is measured with proportional
counters (see Section 6.4.1) or by ICP-MS (see Section 17.8).
Radiochemical analysis of 99 Tc is challenging because, unlike radionuclides
such as 137 Cs, it cannot be measured directly and nondestructively by gamma-ray
spectral analysis. The usual processes of dissolution, purification, and preparation
for counting can be complex for 99 Tc because it has multiple possible oxidation
states, notably cationic Tc4+ and anionic pertechnetate TcO− 4 (see Section 6.4.1).
Because of these characteristics, 99 Tc provides a good example of automated lab-
oratory analysis. Integration of modern selective chemical separation procedures,
radiation detectors, and fluid handling instrumentation in a single functional unit al-
lows the development of an automated radionuclide analyzer and radiation sensor,
as shown by Egorov et al. (2003) and by Grate and Egorov (2003), among others.
The fully automated radioanalytical chemistry system developed for rapid analy-
sis of 99 Tc in aged nuclear waste is shown in Fig. 15.4. The instrument executes
fluid handling steps that acidify the caustic sample, microwave-assisted sample
oxidation to pertechnetate Tc(VII) with peroxidisulfate, separation of 99 Tc(VII)
from radioactive interferences on an anion-exchange column, and delivery of the
purified pertechnetate to a flow-through scintillation detector.
The automated radiochemical process is performed in a single functional unit.
The instrument design incorporates advanced digital fluid handling techniques
with multiple zero dead volume digital syringe pumps and multiple valves for
sample and reagent delivery. Comprehensive multithreaded control software was
328 Harry S. Miley and Craig Edward Aalseth

FIGURE 15.4. Automated total 99 Tc analyzer based on classical radiochemical measurement

principles. (By permission of Pacific Northwest National Laboratory)

developed to enable fully automated asynchronous operation of the instrument

components as well as data processing, storage, and display.
The automated sample treatment protocol begins with sample acidification for
the first digestion step. This initial treatment ensures removal of nitrites, which are
abundant in basic waste solution matrixes and interfere with subsequent oxidation.
In addition, initial heating promotes rapid dissolution of the Al(OH)3 precipitate,
which forms during acidification of the caustic matrix. Nitric acid concentration
and volume are selected to ensure complete dissolution of Al(OH)3 upon heating
and later maintaining high pertechnetate uptake on the anion-exchange material
during sample loading. A second digestion treatment with sodium peroxidisulfate
as the oxidizing reagent converts any reduced technetium species to pertechne-
tate.
Pertechnetate is retained on a macroreticular, strongly basic, anion-exchange
resin (AGMP-1, Biorad). This anion-exchange material was selected because of its
long column life under elevated back-pressure conditions. Pertechnetate is eluted
rapidly from the anion-exchange column with strong nitric acid solution by re-
versing the direction of flow through the column. The pertechnetate separation by
anion exchange in nitric acid medium offers adequate separation selectivity from
90
Sr/90 Y and 137 Cs for determining 99 Tc. To remove interfering anionic radionu-
clides such as 106 Ru, 125 Sb, and 126 Sn, a comprehensive column wash sequence
 15. Automated Laboratory and Field Radionuclide Analysis Systems 329

with 0.2 mol/l nitric acid, 1 mol/l sodium hydroxide, 0.2 mol/l nitric acid, 0.5 mol/l
oxalic acid, and 2 mol/l nitric acid is required prior to pertechnetate elution.
A flow-through scintillation detector equipped with a lithium glass solid scin-
tillator flow cell is used to detect the eluted 99 Tc. The glass scintillator enables
an absolute detection efficiency of ∼55% and is stable in the 8 mol/l nitric acid
medium for pertechnetate elution.
An automated standard addition technique is part of the analytical protocol. An
aliquot of a 99 Tc standard solution is added to a duplicate of the sample during
acidification. The 99 Tc standard is in a nitric acid solution of the same concentration
used for sample acidification. To perform the standard addition measurement, the
sample acidification monitor instrument automatically substitutes a given volume
of the 99 Tc standard solution for an equal volume of the nitric acid. The volume
of the standard solution is selected by the software to yield an estimated threefold
higher signal relative to the analysis of an unspiked sample.
The total effective analytical efficiency (product of the recovery efficiency and
the detection efficiency) is calculated based on the difference in analytical response
obtained by the analysis of the spiked and unspiked samples. This approach pro-
vides a reliable method for remote, matrix matched, instrument calibration. Au-
tomated standard addition can be used for each sample or batch of samples. The
automated radiochemical analysis procedure is rapid, with a total analysis time of
12.5 min per sample. The total analysis time for the standard addition measurement
is 22 min (including analysis of both unspiked and spiked samples). For low-level
samples, a much longer counting time can be expected.
The analyzer instrument was successfully tested with various waste solution
samples from the US DOE Hanford site, including those with high organic content.
Quantification was verified by independent sample analysis with ICP-MS.

15.4. Automated Field System Examples

15.4.1. Aerosol Monitors: Radionuclide
Aerosol Sample Analyzer
The Caribbean Radionuclide Early Warning System and the verification technol-
ogy proposed as part of the Comprehensive Nuclear Test Ban Treaty (CNTBT)
(see http://www.ctbto.org/) (Jan 2006) presented a challenge to automation ca-
pabilities. The CNTBT features as its verification arm the International Mon-
itoring System, a network for daily monitoring of the atmosphere for fission
debris, such that the sensitivity to 140 Ba is <30 μBq (<0.0018 dpm) per m3
of air (Schulze et al., 2000). This sensitivity requirement is the primary driver
for the design of either manual or automatic technology for the 80-station
network.
For practical source–detector geometry, a compressed filter sample can improve
the detection efficiency by about a factor of five over an uncompressed filter. To
eliminate the need for sample compression, several solutions could be pursued.
330 Harry S. Miley and Craig Edward Aalseth

Increasing the size and hence the bulk efficiency of the detector is one possible but
expensive avenue. Increasing the volume of air sample is another avenue. But since
the air contains both the radionuclides of interest and obscuring background ra-
dionuclides, the improvement factor due to volume increase is only approximately
proportional to the square root of the sample volume increase. The approach ap-
plied for substantial improvement in detector–source geometry and thus counting
efficiency was automatic folding or layering of the filter material.
Several sensitivity-enhancing techniques were applied simultaneously in the
Radionuclide Aerosol Sampler Analyzer (RASA) (Miley et al., 1998). This system
employs a Ge detector with about twice the efficiency of contemporary manual
systems (90 vs. 40% relative efficiency) and twice the airflow of manual systems
(24,000 vs. 12,000 m3 per day). A simple layering mechanism provides a large
filter (0.25 m2 ) during sampling and a moderately small filter volume (∼400 cm2
and 0.5-cm thick) for measurement.
Filter volume is minimized for efficient radionuclide measurement with a seg-
mented sampling head and six independent, continuous filter rolls, as shown in
Fig.15.5. These rolls store more than a 1-year supply of daily filter changes for
drawing to and through the sample head. The six simultaneously exposed filters
(each 10 cm × 40 cm) are brought together after exposure and sealed between two
layers of sticky polyester tape that is fed from two rolls. This single filter bundle
then rests in a decay position for 24 h to eliminate gamma rays from the 222 Rn
daughters, 214 Pb and 214 Bi, and reduce those from the 220 Rn daughters (mainly
212
Pb, 212 Bi, and 208 Tl). The filter package is then pulled into a loop around a
rotating drum that circles the germanium detector. The filters remain stationary for
about 24 h for gamma-ray spectrometric analysis and are then advanced by drive
rollers forward 50 cm into the next position. Thus, in normal operation the follow-
ing processes occur: (1) today’s 0.25 m2 filter collects aerosol, (2) the previous
day’s filter is held for decay, and (3) the 2-day-old filter is being measured by the
detector within a small lead cave.
The system is automated with software modules that (1) collect gamma-ray
spectra from the sensors, (2) control relays to activate motors for advancing the

FIGURE 15.5. Segmented sample head, showing filter baffles, rolls of encapsulating
polyester strips, and the wraparound path between the detector and lead shield. U.S. Patent
5,614,724. (By permission of Pacific Northwest National Laboratory)
 15. Automated Laboratory and Field Radionuclide Analysis Systems 331

Start up Advance Filter Sample/Count Get Spectrum Save Spectrum Insert Source Count Source

0 1 2 E
4 3 5 6
A C I H J

D Abnormal Event
B G K
7
L M N

8 9 10
0 a state
Critical Error Detector Bad User Shutdown

A
a transition
Stop

FIGURE 15.6. State Machine showing states and transitions. (By permission of Pacific North-
west National Laboratory)

filters through the aerosol collection, storage, and analysis process, and (3) control
the associated processes. In all, a dozen such software devices can completely
control, monitor, and manipulate all the features of the RASA. Because the RASA
performs a sequential set of steps, from startup to shutdown, a State Machine—a
software construct that is the equivalent of a process flow diagram—was chosen
as control, as shown in Fig. 15.6.
Each decision point in the diagram corresponds to a transition from one state
to another. The State Machine most favorably is written such that a noncomputer
programmer can adjust the operation of the device by editing an English language
file that describes the transition from each state to the others. For example, the
RASA always recognizes its current state via electronic signals like pressure, tem-
perature, voltage, and timers. Upon a state change such as a loss of electrical power
or the time of sample measurement, the state is shifted and new functions, such as
filter advance and calibration, or data retrieval, are automatically performed.
An important feature of any automated system is its behavior at the application
and loss of power. The initial and final conditions need to be known to prevent, say,
loss of a sample or sample analysis data. This is easily accomplished by saving
the final condition at loss of power, but battery backup is needed to monitor and
record this state. A catastrophic condition could result if the automated system
fails to shut down or restore properly, such as system damage or bystander harm
from unexpected mechanical actions.
The spectral data collected daily by the system can confirm successful opera-
tional functioning of the RASA, including important features such as start and stop
time, detector resolution, and gain. To assure that the results are correct and can be
332 Harry S. Miley and Craig Edward Aalseth

used as legal evidence, a robust quality assurance (QA) program also is required.
This begins with system certification at the factory or production laboratory and
includes site-specific documentation of local operating procedures. The system
performs daily wide-range energy calibration, which also serves as a check on the
stability of efficiency values.
Regional laboratories that conform to national standards practices perform ex-
ternal QA measures of the network of RASA systems. An international testing
procedure assures that the laboratories remain proficient. Randomly selected fil-
ters from the network of monitoring stations, automatic and manual, are sent to
regional laboratories to determine if the station results are in control. This can be
partially accomplished by measuring the level of 7 Be (t1/2 = 53.28d), a cosmic-
ray spallation product of atmospheric nitrogen and oxygen that is always in the
atmosphere and is easily measurable on air filters.
One of the advantages of this automatic system is that the state-of-health data
recorded for the numerous subsystems, including blowers, component states, tem-
peratures, and other critical information, allow remote failure detection, diagnosis,
and possibly prevention. As an example, variation in detector temperature may
show the onset of failure of a mechanical cooler. Remote diagnostics are used to
schedule repair trips and minimize down time.

15.4.2. Gas Monitors: Real-Time 133 Xe and 135 Xe

Sampler/Analyzer
The Automatic Radioxenon Sampler/Analyzer (ARSA) (McIntyre, 2001) was de-
signed to measure radioxenon produced in nuclear explosions. Observation of
radioxenon has other uses, including the ability to indicate releases from a reactor
or a medical facility. Gaseous radioxenon is an important indicator of leakage from
underground nuclear tests because gases are more likely to escape than aerosols.
One potential problem is measuring the relatively low gamma-ray energies of the
two radionuclides, which can be obscured by Compton continua for higher-energy
gamma rays.
Stable xenon comprises only 80 ppb of the atmosphere or 0.08 cc/m3 of air
at STP. A series of chemical and physical processes are thus required to remove
contaminants, including the major gases N2 , O2 , and CO2 , the other rare gases
(especially radon), as well as atmospheric moisture. Separation of radioactive gases
in a series of traps held at the condensation temperature for each gas had earlier been
applied manually to analyze fission-produced xenon and krypton radionuclides
in air (Momyer, 1960). The first trap is cooled to a temperature at which one
elemental gas is sorbed or condensed while the others pass through, and then
heated to release the trapped gas for further purification and radiation detection.
The gases that passed through the first trap are pumped to a second trap, held at
a temperature suitable for trapping a second gas but not the other gases, and the
second gas is then released by heating the trap and processed for counting. The
ARSA design is an improvement on these early techniques because it uses selective
 15. Automated Laboratory and Field Radionuclide Analysis Systems 333

FIGURE 15.7. ARSA process diagram. Some duplicate process lines have been omitted for
clarity. (By permission of Pacific Northwest National Laboratory)

condensation with dryers, molecular sieve traps, and charcoal beds, as shown in
Fig. 15.7.
In Area 1 of Fig. 15.7, both trapping and regeneration occur; regeneration steps
are shown by the dashed line. Air is first forced though a pair of pressure-swing
dryers that consist of powdered alumina. While one is drying the incoming sample
air, the second dryer is being regenerated with dry waste air from elsewhere in the
process. These dryers switch from drying to regeneration every few minutes on a
timer, or more frequently, on detection of break-through moisture. The dried air is
then chilled.
The total flow rate is controlled by a commercial mass flow controller (MFC),
which contains an internal servo mechanism that links a mechanical valve to a
resistance thermal device (RTD). The RTD measures mass flow (rather than gas
velocity) by the change of electrical resistance in a sensing wire heated by an
adjacent hot wire. Because this measurement is affected by the specific heat of the
gas, the MFC must be calibrated for each individual gas. The desired MFC flow is
set by applying a voltage to the MFC that corresponds to the voltage generated by
the RTD at that flow. A comparator in the MFC opens or closes the internal valve
to balance the RTD and the applied voltage.
The cold sample gas flows at 100 l/min into a 0.2-l charcoal trap that is cooled
to –90◦ C to adsorb radon. Xenon flows through the trap and is then collected on
the 2-l main charcoal trap at –120◦ C. Nitrogen, oxygen, and the lighter rare gases
helium, neon, argon, and krypton pass through this trap.
 334 Harry S. Miley and Craig Edward Aalseth

Gas inlet/outlet
PMT PMT
Source transfer tube

Scintillation cell
Optical window

PMT PMT

Nal Nal

PMT PMT

Optical isolation

FIGURE 15.8. Xenon detector and internal gas cell. (By permission of Pacific Northwest
National Laboratory)

Xenon is next eluted in Area 2. At the end of the sampling period, the main trap
is valved off from the system and heated to 200◦ C to release the xenon. The outflow
from the main trap is carried by ultrapure nitrogen (no CO2 ) through a MFC at
0.24 l/min. Desorption is timed to release most xenon; flow is diverted when the
remaining radon is expected to desorb. This slow flow takes the desorbed xenon gas
with N2 carrier gas though one of a series of disposable chemical traps (NaOH +
Al) to remove most CO2 . The now purified gas is collected on a 0.2-l 5 Å molecular
sieve trap at –40◦ C, while the carrier N2 escapes. The trap is then heated to 200◦ C
to desorb xenon and transferred to a tiny cold trap that is cooled to –120◦ C to sorb
xenon in preparation for loading the counting cell. This step is necessary to transfer
the gas into a scintillation cell, as shown in Area 3 of Fig. 15.7 and also in Fig. 15.8.
The tiny final trap, loaded with the equivalent of 10 cc (at STP) Xe/CO2 product,
is heated to 200◦ C and the product gas is allowed to expand into one of four counting
cells. Transfer efficiency is boosted by use of a syringe pump, as shown in Area 3
of Fig. 15.7. The exact ratio of the Xe and CO2 mix is determined with a thermal
conductivity device (TDC). This is an important measurement because the xenon
separation efficiency depends on various factors, e.g., the conditioning of the traps
and the ambient temperature. The ARSA typically produces a product gas that is
about 50% xenon.
The ARSA system is designed to operate continuously in remote areas with
minimal maintenance and consumables limited to replacement bottles of nitrogen
carrier gas and CO2 removal traps. Dryers and gas traps were chosen that could
be automatically regenerated in the field. The system has many components and a
complex path for the analyte gas that traverses the system. The numerous valves
(∼100) for controlling the gas flow and trap regeneration processes require a control
system about one order of magnitude more complex than that for the RASA system
described in the preceding section, if the number of parts is taken as a measure of
 15. Automated Laboratory and Field Radionuclide Analysis Systems 335

10000

1000 30 keV
Gamma Singles
81 keV
133Xe (32 mBq/m3)
Counts

100

250 keV
135Xe (3.7 mBq/m3)
10
Beta Gated

1
0 50 100 150 200 250 300 350 400 450 500
Channel Number

FIGURE 15.9. Measurement of Xe-135 in the atmosphere, made possible by automated

sampling and analysis. (By permission of Pacific Northwest National Laboratory)

complexity. In addition, because of the use of elevated temperatures, particularly

in regenerating charcoal traps, the ARSA has a module specifically designed to
monitor safety parameters.
The sample should be processed quickly because of the short half-life of 135 Xe
(9.10 h) compared to 133 Xe (5.234 d). The ARSA performs three sample analyses
per day. The fair match of this frequency to the resolving time (6 h) of meteoro-
logical measurements made worldwide facilitates coordination with atmospheric
transport models. The rapid analysis capability allowed the measurement of 135 Xe
in the Earth’s atmosphere, see Fig. 15.9 (Bowyer et al., 1999).
This system suppresses radon by several orders of magnitude, but the fact that
every atom of radon is radioactive means that a specialized radiation coincidence
detection apparatus is still required to resolve radioxenons from radon. Moreover,
the radiations emitted by the two xenon radioisotopes must be resolved by spectral
analysis. This approach yields an impressive sensitivity of about 100 μBq (0.006
dpm or 3,000 atoms of 135 Xe) per m3 of air based on sampling of about 40 m3 of
air sample and a 24 h radiometric measurement.
The ARSA system has a significant advantage over manual methods because
large numbers of technicians would be required for the time-intensive gas sepa-
ration steps. Prior manual radioxenon monitoring efforts attempted to circumvent
this problem by sampling for longer collection periods. This came at the cost of
much poorer time resolution and a loss of ability to measure the shorter-lived
135
Xe. Measurement of 135 Xe is useful for determining the ratio of 135 Xe to 133 Xe
to distinguish between incidental formation in nuclear weapons and continuous
formation in nuclear reactors.
336 Harry S. Miley and Craig Edward Aalseth

FIGURE 15.10. Conceptual diagram of FDTAS, showing process components from sampling
to analysis. (By permission of Pacific Northwest National Laboratory)

15.4.3. Tritium in Water

Tritium is a major radionuclide in effluent at many nuclear facilities. In the form
of HTO, it moves essentially as water does in groundwater, surface waters, and
airborne moisture. The radionuclide emits only low-energy beta particles. Tritium
is typically measured by mixing purified water with liquid scintillation cocktail
and performing LS counting (see Section 6.4.1). Water that contains tritium at
concentrations not far below regulatory limits and can reach drinking water sup-
plies should be analyzed frequently. Automated sampling, analysis, and reporting
address this concern.
The field deployable tritium analysis system (FDTAS) (Hofstetter et al., 1999)
samples water, purifies the sample, and performs Compton-suppressed LS counting
to achieve laboratory-quality analyses. The system described by Fig. 15.10 has
remote reporting and control capability.
The sampling and analysis aspects of the system follow earlier continuous mon-
itors of tritium in water (Brown et al., 1993). Typically, two photomultiplier tubes
in a lead shield face the sample cell between them. On a cyclical basis, a sample
is collected from the body of water and injected into the sample cell. An equal
volume of LS cocktail is mixed with the sample in the cell. The sample and cocktail
are then counted and discharged. The cell is flushed with water. The next sample,
which had been collected during the counting period, then is injected into the cell.
The counting data are stored, processed to calculate the count rate, and reported.
Periodically, pure water is mixed with the cocktail and counted to provide a
background count rate. Factors related to energy distribution, counts in non-tritium
 15. Automated Laboratory and Field Radionuclide Analysis Systems 337

channels, and pulse shape can be recorded to correct for the presence of other
radionuclides and the effects of quenching or luminescence (see Section 8.3.2). The
counting period is selected to achieve the required detection limit. Measurements
may be at the rate of 1 per 100 min or less frequent if longer counting times are
required.
The FDTAS design adds water purification and improved detection sensitivity.
Laboratory tests of the various methods of purifying environmental samples appli-
cable to unattended field deployment led to selection of a single use, commercially
available, mixed bed resin column for water purification. By employing active
background suppression and pulse shape discrimination with bismuth germanate
(BGO) guard scintillators, lead shielding, and low-background cell components,
the FDTAS achieves tritium detection sensitivity that almost rivals laboratory LS
instruments. The combination of a shield, an active BGO guard detector, and low-
background materials attains a routine background of ∼1.5 count/min in the tritium
energy channel. A special low background quartz counting cell, containing 11 ml
of a 50:50 mixture of the sample and the liquid scintillation counting fluid, yields
a tritium detection efficiency of ∼25%.
The combined low background, high detection efficiency, and 5.5-ml sample
volume lead to a detection limit of 10 Bq (600 dpm) per liter for a 100-min count
(95% confidence limit). These results are achieved routinely in field tests of the
FDTAS at monitoring wells, surface streams, ground water remediation facilities,
sewage treatment plants, and the Savannah River. Results have been confirmed by
parallel sampling and laboratory analyses.
The automation of the FDTAS is of a classic and cost effective type that mostly
uses commercial hardware and support from manufacturers to create sophisticated
custom radiation detection apparatus. QC capabilities are built-in to validate the
results. Nearly real-time reporting allows operational authorities a cushion of time
to end a major tritium release and mitigate its effects.

15.5. Summary
A variety of radiochemical measurements have been successfully automated, both
in the laboratory and in the field. It might appear to the student that the effort
involved in automating a process is more trouble than the manual measurement of
the activity. To an experienced practitioner, it is exactly this economic and qual-
ity decision point that makes the selection of automation or manual approaches
interesting. When the expected variations in sample type and process parameters
allow, automation should be evaluated for ongoing capabilities and field measure-
ments. But it should also be said that the interactions of scientists, technicians, and
engineers on an automation project can be truly rewarding in many ways, as the
fusion of disciplines tends to accelerate the productivity of all.
16
Chemistry Beyond the Actinides
DARLEANE C. HOFFMAN

16.1. Introduction
Humans have been fascinated ever since ancient times with trying to understand
the composition of the terrestrial world around them and even the stars beyond
their reach. In the 4th century B.C., Aristotle proposed that all matter could be
described in terms of varying proportions of four “elements”—air, earth, fire, and
water. Elements, including gold, silver, and tin, that are found relatively pure in
nature were isolated and used over the period of the next several hundred years.
The alchemists of medieval times isolated and discovered additional elements and
used secret formulas and rituals in the attempt to find the “philosopher’s stone” and
attain their goal of transmuting lead into gold. The development of experimental
science and the scientific method in the 18th century accelerated the pace of the
discovery of new elements, but uranium, discovered in 1789 by Klaproth in pitch-
blende from Saxony, Germany, remained the heaviest known chemical element for
150 years.
After the discovery by Becquerel in 1896 that uranium salts blackened photo-
graphic plates due to radiations from the natural decay chains of uranium, Marie
Curie and Pierre Curie (Curie and Curie, 1898) began extensive studies of the new
phenomenon that they dubbed “radioactivity.” They were successful in isolating
and identifying the first radioactive elements Po (84) and Ra (88) in pitchblende.
They shared the Nobel Prize in Physics with Becquerel in 1903 for their studies
of radioactivity. Marie Curie continued her investigations of radioactive species
and isolated macro quantities of Ra to be used for medical purposes. She received
the Nobel Prize in Chemistry in 1911 for the isolation of pure radium and the
determination of its molecular weight. She is recognized as the founder of the sub-
discipline of “radiochemistry,” which can be defined as the study of radioactivity
and radioactive elements; she could rightfully be acknowledged as the founder of
nuclear medicine as well.
In the mid-1930s, the new breed of “nuclear” scientists, including both chemists
and physicists, became intrigued with the possibility of actually synthesizing new
“artificial” elements not found in nature. The discovery of “artificial” radioactivity
by Joliot–Curies in 1934, the invention of the cyclotron by E. O. Lawrence in

Nuclear Science Division, Lawrence Berkeley National Laboratory, Department of Chem-

istry, University of California, Berkeley, CA 94720

338
 16. Chemistry Beyond the Actinides 339

1929, and the subsequent operation of larger cyclotrons at Berkeley facilitated

production of artificial elements. The new developments rather quickly culminated
in the realization of the ancient alchemists’ dream of transmutation when the first
“artificial” element technetium (Z = 43) was identified in 1937 in the products of
deuteron bombardment of molybdenum in the cyclotron (Perrier and Segrè, 1937).
Then astatine (85) was identified in 1940 in cyclotron bombardments of Bi with
He ions (Corson et al., 1940).
Shortly thereafter, the first transuranium element Np (93) was identified (McMil-
lan and Abelson, 1940) in experiments conducted to investigate the newly re-
ported phenomenon of nuclear fission in neutron irradiations of U. The production
and chemical identification of element Pu (94) in the form of 238 Pu produced in
deuteron bombardments of uranium in the cyclotron at Berkeley followed in 1941,
but it was not published until 1946 (Seaborg et al., 1946). The discoveries were
voluntarily kept secret because it was recognized that the highly fissionable 239 Pu
produced at about the same time (Kennedy et al., 1946) might have potential mil-
itary uses. The artificial element Pm (61) was separated and positively identified
(Marinsky et al., 1947) among the fission products of uranium in 1945 during the
World War II Plutonium Project at the Oak Ridge, Tennessee Laboratory. Although
Tc, Pm, At, and Pu are often called “artificial” elements, all of them are found in
nature in exceedingly small amounts, either as uranium fission or capture products
or from natural decay chains. The very long-lived 244 Pu was also reported in 1971
to exist in nature in minute quantities. Thus the distinction between “artificial” and
“natural” elements is in itself somewhat artificial.
During what might be called the “golden age of discovery” between 1940 and
1955, all the transuranium elements from Np through Md (101) were discovered.
They were produced by either neutron or alpha bombardments of suitable tar-
gets and were identified with little controversy, probably because their half-lives
were long enough to permit chemical separation and identification. Beginning with
No (102), identification of new elements became ever more difficult and contro-
versial as the half-lives became shorter and the production rates even smaller. The
first positive identifications of the elements beyond Md were based on physical
techniques rather than on chemical techniques, and many controversies ensued.
However, with the discovery of longer lived isotopes of No, Lr (103), and Rf
(104), it was possible to confirm Seaborg’s hypothesis (Seaborg, 1945) that the
actinide series ends with Lr and that the chemistry of Rf will be quite differ-
ent from that of the actinides. More complete discussions of these discoveries,
the experimental methods involved, and some of the ensuing controversies can
be found elsewhere (Seaborg, 1967; Hoffman and Lee, 1999; Hoffman et al.,
2000).
The transactinide elements are defined simply as those elements with atomic
number greater than 103 (Lr), which completes the actinide series with the com-
plete filling of the inner 5f shell. The schematic periodic table of 2004 given in
Fig. 16.1 shows the lanthanide (4f) series, the actinide (5f) series, and the trans-
actinides as a 6d transition series that begins with element 104 (rutherfordium,
Rf). According to results of atomic relativistic calculations (Pershina and Fricke,
340 Darleane C. Hoffman

FIGURE 16.1. 2004 periodic table showing placement of transactinides through element
154.

1999), this transition series ends at element 112 with the filling of the 6d shell.
The p shells are filled in elements 113 through 118, which are expected to be the
heaviest members of the noble gas group. The 8s shell is expected to be filled
in elements 119 and 120, making them homologs of groups 1 and 2. Based on
relativistic calculations, element 121 should have an 8p electron in its ground-
state configuration in contrast to the 7d electron expected by simple extrapola-
tion from the elements in group 3 of the periodic table. A 7d electron would
then be added in element 122, giving it the configuration [118]8s2 7d8p, differ-
ent from that expected from its homolog thorium, which has the configuration
[Rn]7s2 6d2 . The situation becomes even more complicated beyond element 122
because the energy spacings of the 7d, 6f, and 5g levels as well as those of
the 8p, 9s, and 9p levels are so close that the chemical properties of these ele-
ments will be nearly impossible to characterize on the basis of currently known
properties.

16.1.1. Transactinide Elements and Controversy

Throughout the period from 1960 to 1977, controversies were associated with
claims to priority of discovery and proposals for the naming of elements 104
and 105, the first of the transactinides, by research groups working at Lawrence
Berkeley Laboratory (LBL) in the United States and the Joint Institutes for Nuclear
 16. Chemistry Beyond the Actinides 341

Research (JINR) in Dubna, the USSR. In 1974, the International Union of Pure
and Applied Chemistry (IUPAC) and the International Union of Pure and Applied
Physics (IUPAP) appointed an ad hoc committee of neutral experts to consider
the competing claims and to facilitate cooperation between the groups in reaching
agreement on these issues. However, the committee never met as a whole, and it
was finally disbanded in 1984, although a few smaller meetings with some of the
Berkeley and Dubna researchers were sponsored, observers were exchanged, and
some reports were issued (Hyde et al., 1987).
Pending settlement of the competing claims to discovery and approval of names,
the Commission on Nomenclature of Inorganic Chemistry (CNIC) of the IUPAC
mandated (Chatt, 1979) the use of three-letter “systematic names” based on 0=nil,
1=un, 2=bi, 3=tri, 4=quad, 5=pent, 6=hex, 7=sept, 8=oct, and 9=enn for el-
ements with Z > 100. Thus, elements 104 and 105 officially became unnilqua-
dium (unq) and unnilpentium (unp), although these names never found common
usage among heavy element researchers. In the meantime, the names kurcha-
tovium (Ku, 104) and nielsbohrium (Ns, 105) and rutherfordium (Rf, 104) and
hahnium (Ha, 105) continued to be used by the Dubna and Berkeley groups, re-
spectively.
Among the reasons for the controversies over discovery of elements 104 and
105 are their short half-lives and small production rates that made it necessary to
study and identify them “online” at the accelerators where they were produced on
the basis of their decay properties rather than on “conventional” radioanalytical
methods. New radioanalytical techniques had to be developed for unequivocally
establishing that a new element had, indeed, been produced.
The Berkeley group developed the α–α correlation technique in which the un-
known element’s α-decay is correlated with that of the α-decay of known daughter
and/or granddaughters. In some cases, the daughters can be identified by radio-
chemical separations. The measurement of the characteristic X-rays of a new ele-
ment is another definitive method. It was used at Oak Ridge National Laboratory
by Bemis et al. (1973, 1977) to confirm the Berkeley group’s discoveries of ele-
ments 104 and 105. However, this method requires detection of relatively “large”
numbers of events in order to obtain statistically significant measurements of the
X-ray energies. In contrast to these methods, the Dubna group relied primarily
on the detection of spontaneous fission (SF) decay. This is an extremely sensitive
technique, but detection of only the fragments from the fission process makes it
extremely difficult to deduce the identity of the fissioning species, especially based
on detection of only a few events.
Indirect methods such as half-life systematics, excitation functions for the pro-
duction reactions, and cross bombardments have been used to reinforce this in-
formation. In order to positively identify the atomic number of a spontaneously
fissioning nuclide from detection of the fragments, the atomic numbers of both
primary fragments from the same SF event must be determined in coincidence
and added together to determine the Z of the new, unknown fissioning nuclide.
Detection of only SF decay has resulted in much controversy concerning discovery
and identification of the transactinide elements.
342 Darleane C. Hoffman

16.1.2. Naming of Elements 104 Through 109

After the reported discoveries of elements 107 through 109 between 1981 and 1984
(Münzenberg et al., 1981, 1982, 1984) by researchers using the Separator for Heavy
Ion Reaction Products at the Gessellschaft für Schwerionenforschung mbH (GSI)
in Darmstadt, Germany, the IUPAP in 1986 decided to appoint a Transfermium
Working Group (TWG) to examine the priority of discovery of elements 101
through 109 so that official names could be proposed and approved for these
elements. The IUPAC asked that they also be represented on the Committee because
naming of the chemical elements was clearly within their historical jurisdiction.
The study consisted of two phases: (1) establishment of criteria for discovery
of new chemical elements and (2) application of these criteria in practice to the
transfermium elements.
Phase 1 of the committee’s report was accepted by both the IUPAP and the
IUPAC and published by Barber et al. (1991) in the IUPAC journal in late
1991. The judgmental phase was quickly approved by the IUPAC Bureau in
August 1991 and the IUPAP Council in September 1991, and both Phase 1 and
Phase 2 conclusions were published in the IUPAP journal in 1992 (Barber et al.,
1992).
Because the U.S. groups were not given an opportunity to view and comment on
the accuracy of any draft reports from the TWG after their visit to Berkeley in 1998
and their subsequent visit to Dubna in 1999, they were totally taken aback by this
rapid publication of the TWG conclusions without the “iterative” process that they
had understood would take place with the involved groups at LBL in the United
States, GSI in Darmstadt, Germany, and JINR in Dubna, USSR. In an attempt to
counteract this criticism, responses from the Berkeley, Dubna, and GSI groups
were invited and published in the IUPAC journal immediately following the TWG
“Discovery” article of 1993 (Barber et al., 1993). The IUPAC then stated that this
TWG report could now be considered by the CNIC, which had the responsibility
for recommending names. However, it was pointed out that the TWG report had
not been subjected to the external and internal review required by the IUPAC prior
to publication.
Another long period of controversy ensued in which the CNIC at one point
unanimously declared that an element should not be named after a living person.
The name “seaborgium” for element 106 had been proposed by its undisputed
discoverers at the March 1994 National Meeting of the American Chemical Society
(ACS) and had been quickly accepted by the ACS Committee on Nomenclature and
approved by the Board of Directors in November 1994. But at that time Glenn T.
Seaborg was obviously alive and the CNIC of the IUPAC declared that the name
“seaborgium” could not be used. Furthermore, they decided that the names for
elements 104–109 would be chosen from among those suggested by the three
groups involved, but not necessarily for the elements for which they had been
originally suggested! They proposed the following names: 104, dubnium; 105,
joliotium; 106, rutherfordium; 107, bohrium; 108, hahnium; 109, meitnerium.
This resulted in complete confusion and the decoupling of the names from those
 16. Chemistry Beyond the Actinides 343

TABLE 16.1. CNIC/IUPAC compromise recommendation for names of

transfermium elements. Approved by IUPAC, August 30, 1997, Geneva,
Switzerland
Element Name Symbol Year of discovery
101 Mendelevium Md 1955
102 Nobelium No 1958
103 Lawrencium Lr 1961

Transactinides
104 Rutherfordium Rf 1969
105 Dubnium (Hahnium)a Db (Ha)a 1970
106 Seaborgium Sg 1974
107 Bohrium Bh 1981
108 Hassium Hs 1984
109 Meitnerium Mt 1982
a Many publications of chemical studies prior to 1997 use hahnium (Ha) for element 105.

suggested by the discoverers for the elements they claimed to have discovered and
which had been in common use.
A worldwide storm of protest and criticism resulted and the IUPAC con-
vened a series of meetings to try to agree upon a compromise set of names.
The claims to discovery and the subsequent naming controversies surrounding
elements 104–109 are discussed in detail in Chapters 9, 10, and 13 of Hoff-
man et al. (2000). Finally, after several unacceptable naming proposals, the
IUPAC backed down on their edict disqualifying living persons, and the so-
called compromise list of names and symbols shown in Table 16.1, including
“seaborgium, Sg” for element 106, was approved in August 1997 (CNIC, 1997).
This was only a year before Glenn Seaborg suffered the stroke and fall at the
1998 Boston ACS meeting, which ultimately resulted in his death on February 25,
1999.

16.1.3. Elements Beyond Mt

Discoveries of element 110 have been reported by groups at LBNL (Ghiorso
et al., 1995a,b), GSI (Hofmann et al., 1995a), and a combined Dubna/Lawrence
Livermore National Laboratory (LLNL) group working at Dubna. Discovery of
elements 111 and 112 was reported by the GSI group in 1995 and 1996 (Hofmann
et al., 1995b, 1996) and in 2002 (Hofmann et al., 2002) reported confirmatory
experiments. A procedure for naming of new elements was then proposed by
the IUPAC to try to avoid future naming controversies and the use of different
names by competing research groups (Koppenol, 2002). A joint working party of
the IUPAC/IUPAP (Karol et al., 2001) examined the various claims to discovery
and gave credit for discovery of element 110 to the GSI group and invited them
to propose a name. They proposed the name Darmstadtium with symbol Ds for
element 110 after the place in Germany, where the discovery experiments were
344 Darleane C. Hoffman

conducted. The IUPAC Commission on the Nomenclature of Inorganic Chemistry

(CNIC) considered the proposal and recommended to the IUPAC Bureau that it be
accepted (Corish and Rosenblatt, 2003). It was officially approved by the IUPAC
Council in Ottawa, Canada, in August 2003.
In mid-2003, the joint working party assigned credit for discovery of ele-
ment 111 to the GSI group and asked them to propose a name (Karol et al.,
2003), but the evidence for discovery of element 112 was still deemed insuffi-
cient. The name roentgenium, symbol Rg, in honor of Wilhelm Conrad Roent-
gen, who discovered X-rays in 1895, was proposed for element 111 by the
discovery group. In May 2004 a provisional recommendation for approval of
the proposal (Corish and Rosenblatt, 2004) was sent to the IUPAC Bureau and
Council, and it was approved in November 2004.
Evidence for elements 112 through 116 has been published between 1999 and
early 2004 by a Dubna/LLNL group and a Dubna/international group (Oganessian
et al., 1999a,b,c, 2000a,b, 2002, 2004a,b,c; Lougheed et al., 2000; Oganessian,
2001, 2002), who conducted experiments at Dubna. These reports have not yet
been confirmed by other groups, and the elements are shown in italics in the
periodic table given in Fig. 16.1. Positive mass and atomic number assignments
are extremely difficult because the reported decay chains for these elements do not
connect to previously known nuclides.
An LBNL group reported discovery of element 118 in 1999 (Ninov et al., 1999),
but the claim was retracted when the original discovery could not be confirmed
(Ninov et al., 2002) and Gregorich et al. (2002) have set limits on the production
cross section of the originally reported 118 decay chain of less than a picobarn.
The isotopes of Rf through Mt and of elements 110 through 116 reported as of
early 2003 are shown in Figs. 16.2 and 16.3, respectively.

FIGURE 16.2. Isotopes of Rf through Mt (2003).

 16. Chemistry Beyond the Actinides 345

FIGURE 16.3. Isotopes of elements 110 through 116 (2003).

16.1.4. Challenges Involved in Studies of Chemical

Properties of the Transactinides
As mentioned earlier, the ever shorter half-lives and decreasing production rates
of the transactinides, the presence of a plethora of unwanted reaction products,
the necessity for production at accelerators with high-intensity beams, and the
need for special radiochemistry laboratories and detection facilities have posed
formidable challenges to studies of their chemical properties. These demands have
stimulated the development of new radioanalytical methods suitable for use in
“atom-at-a-time” studies in which the production and detection rates range from
a few atoms per minute for Rf to only an atom per week for Hs. The isotope to
be used must have unique decay characteristics if it is to be positively identified
as belonging to the element whose chemistry is being studied on an atom-at-
a-time basis. The kinetics of the chemical procedures to be used must be fast
enough to be accomplished in times comparable to the half-lives of the isotopes
involved and must give the same results for one atom as for a large number of
atoms. For example, precipitation reactions that depend on exceeding a given
solubility product are obviously not suitable. Chromatographic methods in which
a single atom undergoes many identical chemical interactions, such as in two-phase
extraction systems with rapid kinetics that reach equilibrium quickly, have been
shown to be valid.
Theoretical studies (Adloff and Guillaumont, 1993; Guillaumont et al., 1989,
1991) have shown that it is valid to combine the results from many identical experi-
ments in which only a single atom was present to obtain information about chemical
behavior although it may be necessary to repeat the identical experiment hundreds
of times to obtain statistically significant results. Thus, computer-controlled au-
tomated systems become especially attractive. Although they are not necessarily
346 Darleane C. Hoffman

TABLE 16.2. Isotopes used in first chemical studies of elements Rf through Hs

Isotope (half-life) Production reaction Cross section (nb) Year
261 Rf (75 s) 248 Cm + 18 O → 261 Rf + 5 n ∼5 1970a
262 Ha (35 s) 249 Bk + 18 O → 262 Ha + 5 n ∼6 1988b
265,266 Sg (21 s,7 s) 248 Cm + 22 O → 265,6 Sg + 5 n, 4 n ∼0.3 1997c
267 Bh (17 s) 249 Bk + 22 Ne → 267 Bh + 4 n ∼0.06 2000d
269,270 Hs (2–7 s, 14 s) 248 Cm + 26 Mg → 269,270 Hs + 5 n, 4 n ∼0.005 2002e

a Silva
(1970); b Gregorich (1988); c Schädel (1997a); d Eichler (2000); e Düllmann (2002a,b); Kirbach
(2002).

faster than manual operations, they usually give more reproducible results and are
nearly essential for around-the-clock experiments lasting weeks at a time.
To avoid dissolution of the highly radioactive and rare targets listed in Table 16.2
that must often be used to produce the isotopes of interest, foils sometimes are
placed directly behind the target to catch the activities recoiling from the target
after irradiation and then they are processed. Now it is more common to attach
the recoiling activities to aerosols in a flowing stream of inert carrier gas such as
He in which they can be rapidly transported outside the irradiation site within the
accelerator shielding to the site of the chemical system to be used.

16.2. Radioanalytical Techniques

16.2.1. Gas-Phase Chemistry
Both solution and gas-phase studies of the chemical properties of the transac-
tinides have been conducted. Gas-phase studies are especially useful for short-
lived isotopes because they avoid the lengthy process of evaporating the liquid
samples produced in most aqueous chemistry procedures prior to detection of
SF and α-decay with solid-state detectors. Both gradient gas chromatographic
(GGC) and isothermal gas chromatographic (IGC) methods have been used. In
early GGC studies pioneered by Zvara et al. (1972, 1974, 1976), a negative tem-
perature gradient was established along a chromatographic column (usually quartz)
through which a gas stream containing volatile species of the isotopes of interest
was conducted. These species deposited on the surface of the column according
to their volatilities and later the deposition zones and temperatures were deter-
mined from fission tracks registered in detectors positioned along the column.
The deposition temperatures were then correlated with the adsorption enthalpy.
The advantage is that the process can be applied to half-lives as short as a few
seconds.
The disadvantages of these early GGC techniques are that the deposition posi-
tions are determined only after the experiment is finished and only SF is detected.
Real-time observation of the nuclear decay is not possible. Therefore, the half-
lives cannot be determined and since it is necessary to correct for the half-lives of
the different isotopes involved, interpretation of the results is extremely difficult.
 16. Chemistry Beyond the Actinides 347

Cm
Mg beam
targets

He/O2 Quartz
column Thermostat
LN

Gas Oven
chamber
IVO COLD System Detector Pairs

FIGURE 16.4. Schematic of IVO-COLD system for study of gas-phase properties of HsO4
and lighter homologs. Adapted from Düllmann et al. (2002b).

There also is the usual problem of positively determining the atomic number of
the fissioning nuclide whose chemistry was being studied.
Recently, these problems were solved in a very imaginative manner by forming
the chromatographic column from opposing (Si) photodiode detectors upon which
the volatile radioactive species were directly deposited. In this way, both the radi-
ations (α and SF), and their deposition positions as a function of the temperature
gradient along the Si column, were determined simultaneously and continuously
recorded and stored via a computer system. A gradient cryogenic version of this
system, the Cryo On-Line Detector (COLD), was used to perform the first success-
ful chemical studies of Hs (108) with the α-emitting isotopes 269,270 Hs, which were
identified by their decay to known α-emitting daughters. The COLD system was
used together with the In situ Volatilization and On-line (IVO) detection system
to study the volatile oxides of Hs and Os. A schematic diagram of the IVO-COLD
system (Düllmann et al., 2002a,b) is shown in Fig. 16.4. The study indicated that
Hs formed a volatile oxide similar to that of the tetroxide of Os, its lighter group 8
homolog.
Online isothermal chromatographic systems, such as the On-Line Gas Analyzer
(OLGA) developed by Gäggeler and coworkers (Gäggeler, 1994) and the Heavy
Element Volatility Instrument (HEVI) developed by Kadkhodayan et al. (1996),
have been used to study volatile halides and oxyhalides of Rf, Db (Ha), Sg, and
Bh. The α-emitting isotopes shown in Table 16.2 were used in these studies. In
these systems, the entire length of the chromatographic column (usually quartz)
is kept at a constant temperature and the volatile species pass through the column
in a carrier gas such as He and undergo numerous sorption/desorption steps. The
yield through the column is determined by measuring the α-radiations from the
exiting species that are either reattached to aerosols and transported to a detection
device or condensed directly on a detector system.
Experiments are conducted at a series of isothermal temperatures to study the
chemical yields of the volatile species as a function of temperature. Retention
time is indicative of the volatility at the given isothermal temperature. It is equal
348 Darleane C. Hoffman

FIGURE 16.5. Schematic diagram of the HEVI system.

to the half-life of the isotope at the temperature at which 50% of the plateau
or steady-state yield is obtained. This T50% retention time can then be used as
a relative measure of volatility. A Monte Carlo program taking account of all
the experimental parameters, such as gas flow rate, length of column, and the
half-lives of the isotopes in the volatile species, is used to deduce the adsorption
enthalpies for the volatile species. A schematic diagram of the HEVI system, used
for isothermal gas-phase studies, is shown in Fig. 16.5. The recoiling products from
the reaction are attached to either KCl, KBr, or MoO3 aerosols, and transported
in He gas to the entrance to HEVI where they are deposited on a quartz wool
plug and halogenated at 900◦ C. The volatile products are transported through the
quartz column in flowing He gas and are again attached to aerosols in the recluster
chamber and transported via another gas-jet to the Merry Go-around (MG) system.
They are then deposited on thin polypropylene films placed around the periphery
of its horizontal wheel, which is rotated to position the foils successively between
pairs of surface barrier detectors for α and SF spectroscopy.
Photos of HEVI, OLGA, and the MG rotating wheel and detection system are
shown in Fig. 16.6(a)–(c); parts (d) and (e) of Fig. 16.6 will be discussed in the
following section.

16.2.2. Solution Chemistry

Examples of automated computer-controlled systems developed for studies of solu-
tion chemistry of the actinides and transactinides are the Automated Rapid Chem-
istry Apparatus (ARCA) and the microcentrifuge system SISAK (Special Isotopes
Studied by the AKUFE technique). They are quite different in that although ARCA
can be used to perform rapid, repeated, high-pressure liquid chromatography col-
umn experiments on the seconds-time scale to determine distribution coefficients,
the collected liquid samples must be dried prior to the measurement of α-particle or
SF decay with high-resolution solid-state detectors that limits detection to nuclides
with half-lives of about 30 s or longer. However, with suitably designed separation
 16. Chemistry Beyond the Actinides 349

FIGURE 16.6. (a) B. Kadkhodayan with HEVI (1992); (b) J. Kovacs and H. W. Gäggeler with
OLGA (1988); (c) D. C. Hoffman and D. M. Lee with MG rotating wheel system; (d) J. V.
Kratz and M. Schädel with Automated Rapid Chemistry Apparatus (1988); (e) Schematic
diagram of a typical SISAK liquid–liquid extraction configuration with Berkeley Gas-filled
Separator as a preseparator.
350 Darleane C. Hoffman

procedures, longer lived daughter products can be detected and used to deduce
the properties of the parent element. The ARCA shown in 16.6(d) has been used
successfully in separations of Rf through Sg. Several reviews of the results have
been published (Hoffman, 1994; Schädel, 1995; Hoffman and Lee, 1999; Kratz,
1999a; Kratz, 1999b).
The SISAK system, which had previously been used to perform liquid–liquid
extraction studies of γ-emitting nuclides with half-lives as short as a second,
was adapted for use with α-emitters by coupling it to a flowing liquid scintillation
system. This provides continuous separation and measurement of α–α correlations
and detection of SF decay of nuclides with half-lives of only a few seconds. This
permits chemical studies of short-lived nuclides but the energy resolution is not as
good as with solid-state detectors. Nevertheless, a recent experiment by Omtvedt
et al. (2002) showed that rapid preseparation in the Berkeley Gas-filled Separator
(BGS) achieved sufficient decontamination from the extremely high background
of unwanted activities. Detailed information was obtained about the chemical
properties of Rf for 4.7-s 257 Rf produced in the 208 Pb(50 Ti,n) reaction. Similarly,
the chemistry of Db (Ha) can be studied with 4.4-s 258 Db and still heavier elements
can be investigated although the production rates are steadily decreasing. Another
limitation is the requirement to choose extraction systems that have kinetics that are
rapid enough that equilibrium can be attained in the short extraction contact times
associated with the SISAK system. A schematic diagram of the SISAK system as
configured for these experiments is shown in Fig. 16.6(e).

16.2.3. Relevance and Applications to Other Fields

These exotic, frontier studies of the heaviest elements attract many undergrad-
uate and graduate students to nuclear and radiochemistry. Research in this field
provides excellent education and training for future contributions and careers in
radioanalytical chemistry and a wide variety of other applied areas and research
fields. Graduates find employment at U.S. national, federal, and state laboratories,
regulatory agencies, private industry, and universities.
The rapid, high-yield, computer-controlled automated techniques that have been
developed for separation, detection, data storage, and analysis of hundreds of
identical experiments of short-lived species, using both gas-phase and solution
chemistry, are described in this chapter. These systems can certainly be adapted
for use in radioanalytical chemistry and a wide variety of other applied fields, as
well as other research areas, when large numbers of samples must be analyzed
with reproducible results for long-lived as well as short-lived species (see also
Chapter 15). Applications in the following areas might be envisioned: (1) environ-
mental studies including long-term monitoring, modeling, and prediction of the
behavior of actinides and other radionuclides in the environment; (2) studies of
nuclear waste isolation and remediation of both radioactive and other toxic sites;
(3) treatment, processing, and minimization of wastes; (4) ultrasensitive analy-
ses and instrumentation; (5) computer-controlled automated remote processing
for hazardous materials; (6) studies of radionuclide chemistry in nuclear reactors
 16. Chemistry Beyond the Actinides 351

and in the nuclear fuel cycle; (7) surveillance of clandestine nuclear activities and
evaluation of terrorist threats using gas-analysis techniques; (8) radioisotope pro-
duction and separation; (9) radiopharmaceutical synthesis, and detection systems
for both diagnostic and therapeutic nuclear medicine procedures; (10) biochemical
and agricultural research; and (11) geochemical dating studies.

16.3. Results of Experimental Studies of Transactinides

Chemical studies of the transactinides from Rf (104) through Hs (108) have now
been performed although Bh (107) and Hs have been studied only in the gas phase.
Studies of Mt (109) await discovery of longer lived isotopes than 40-ms 268 Mt,
currently the longest known isotope of Mt. Predictions of the half-lives of 270,271 Mt
range from tenths of a second to a few seconds based on Smolańczuk’s calculations
(Smolańczuk, 1997). Initial attempts to study element 112 have been unsuccessful
so far. Placement of the transactinides in the periodic table has been based on
comparison of the behavior of the individual transactinide with that of its lighter
homologs in a given group in the periodic table. Several papers (Kratz, 1999a,b;
Pershina and Hoffman, 2003) that give detailed reviews of the experimental and
theoretical studies of the transactinides are available.

16.3.1. Chemistry of Elements 104(Rf) and 105(Db/Ha)

The first chemical experiments on the transactinides were designed to determine
their placement in the periodic table and whether the 5f actinide series actually
ended with Lr as predicted by Glenn Seaborg (Seaborg, 1945). Then elements 104
and 105 should have different chemistry from the actinides, and as members of
the 6d transition series should be placed as the heaviest members of groups 4 and
5 and have similar properties.
Soon after the discovery of element 104, Silva et al. (1970) studied its solution
chemistry for the known isotope 75-s 261 Rf. Positive identification was made by
measuring its known half-life and decay characteristics. Several hundred elutions
with α-hydroxyisobutyrate solutions from cation exchange resin columns were
performed to compare the behavior of 261 Rf with those of No2+ , trivalent actinides,
Hf4+ , and Zr4+ . The behavior of Rf was similar to those of Zr and Hf, which did
not sorb on the column and entirely different from No2+ and the trivalent actinides
that did sorb on the column at pH 4.0. These studies showed unequivocally that
Rf should be placed under Zr and Hf in the periodic table.
The first computer-controlled automated system for performing very rapid solu-
tion chemistry experiments on an atom-at-a-time basis was used in later pioneering
experiments (Hulet et al., 1980). Results showed that the anionic–chloride com-
plexes of Rf were clearly similar to Hf and much stronger than those of the trivalent
actinides, again confirming the position of Rf in group 4 of the periodic table.
In 1972, Zvara et al. (1972), in the earliest gas-phase chemical studies of element
104, reported that its tetrachloride volatility was similar to HfCl4 and greater than
352 Darleane C. Hoffman

Adsorption enthalpy on SiO2

-60

-70
ZrCl4
RfCl4
-80
ΔH ads(kJ/mol)
RfBr4

Volatility
-90

HfCl4
o

-100
ZrBr 4
-110

HfBr4
-120

-130
30 40 50 60 70 80 90 100 110

Atomic number

FIGURE 16.7. Adsorption enthalpy values and relative volatilities on SiO2 for Zr, Hf, and
Rf chlorides and bromides.

actinide and Sc chlorides. They concluded that element 104 should be placed in
group 4 of the periodic table. The results were not considered conclusive (Kratz,
1999b) because only SF events were measured, making it impossible to identify
positively the element being studied.
After these initial studies, a rather long time elapsed before additional exper-
iments were conducted in the late 1980s. These were motivated by theoretical
calculations, suggesting that relativistic effects might alter the chemical properties
of Rf and Ha from those expected by simple extrapolation of properties within a
given group in the periodic table. Subsequent studies of the gas-phase chemistry
of Rf, Zr, and Hf chlorides (Kadkhodayan et al., 1996; Türler et al., 1996) and
bromides (Sylwester et al., 2000) were conducted with OLGA and HEVI at the 88-
Inch Cyclotron at LBNL. Contrary to expectations based on simple extrapolation
of group 4 volatilities, these studies showed (Türler et al., 1996) that the Rf halides
were more volatile than those of Hf. In agreement with relativistic calculations
(Pershina and Fricke, 1999), the bromides were all found to be less volatile than
their respective halides as shown in Fig. 16.7
Many additional studies to compare the solution chemistry of Rf with Zr, Hf,
Th(IV), and Pu(IV) have been conducted and include both manual column and ex-
traction studies, and automated studies using ARCA. Extractions from aqueous so-
lutions into triisooctylamine, tributylphosphate, and thenoyltrifluoroacetone have
been studied in considerable detail. They have shown that, although Rf behaves
generally like a group 4 element, its behavior varies with different complexing
agents.
 16. Chemistry Beyond the Actinides 353

Zvara et al. (1974, 1976) reported early gradient thermochromatographic studies

comparing the volatilities of the chlorides and bromides of element 105, using 1.8-s
261
105 with Nb and Hf. They concluded that the volatility of the 105 bromide was
less than that of the bromide of Nb and about the same as Hf. Again, only SF
events were detected so it is not certain that the properties of element 105 were
actually measured.
Studies of the chemistry of element 105 languished until the very first studies
of its behavior in solution were reported by Gregorich et al. (1988). These were
undertaken simply to determine its most stable oxidation state in aqueous solution.
It had been postulated (Keller, 1984) that because of relativistic effects Ha might
have a 7s2 6d7p2 valence configuration that could result in a 3+ oxidation state if
the 7s2 electrons were sufficiently stabilized by relativistic effects. But the 6d3 7s2
was expected by analogy to Ta, which has a 5d3 6s2 configuration and a most stable
state in aqueous solution of 5+ . The reaction shown in Table 16.2 was used to make
35-s 262 Ha. It was identified by measurements of the energy and time distribution
of the α-decay and of time-correlated pairs of α-decays from 262 105 and its 4-s
258
Lr daughter. The recoiling products from the production reaction were attached
to aerosols and transferred via He jet to a collection site in a hood located outside
the radiation area where manual “glass” chemistry was performed with only a few
microliters of solutions.
The sorption of Ha on glass cover slips was compared with that of tracers of
the group 5 elements Nb and Ta produced online under similar conditions. These
group 5 elements are known to sorb on glass surfaces after fuming with nitric
acid while the group 4 elements and actinides do not. About 800 such manual
extractions taking 50 s each were conducted and a total of 26 α-decays and 26 SFs
were detected. Element 105 was found to sorb on the glass as did Nb(V) and Ta(V)
while Zr, Hf, and the trivalent actinides did not, thus demonstrating that element
105 should be placed in group 5 of the periodic table. However, in extractions
into methylisobutyl ketone from mixed HNO3 /HF solutions, Ta extracted but Ha
remained behind with Nb, again indicating that details of complexing behavior
cannot be predicted on the basis of simple extrapolation from group trends.
These experiments provided the impetus for future explorations of the complex
behavior of element 105, using ARCA II, an improved version of ARCA, to carry
out the many thousands of experiments required to get statistically significant re-
sults. A collaboration of groups from GSI/Mainz, PSI/Bern, and LBL/UCBerkeley
subsequently carried out many experiments at LBL and GSI to study the chem-
istry of element 105 in more detail. Guidance from theoretical chemists (Pershina,
1998a,b; Pershina and Bastug, 1999) has been invaluable in designing ap-
propriate experimental conditions for elucidating the influence of relativistic
effects.
An international group (Gäggeler et al., 1992; Türler, 1992) used OLGA II
and 35-s 262 Db to perform the first online isothermal gas chromatography of the
bromides of element 105. The volatile species were deposited on a moving tape
and transported in front of six large area passivated implanted planar Si detectors
for measurement of α-particles and SFs. The adsorption enthalpies were nearly
354 Darleane C. Hoffman

the same for Nb and Ta, but Db appeared to be significantly less volatile and it was
postulated that an oxybromide might have formed from traces of oxygen present.
It is predicted to be less volatile than the pentabromide (Pershina et al., 1992).
Later experiments in which the partial pressure of oxygen was varied showed that
both volatile and less volatile species were formed and additional measurements
are needed to try to produce the pure pentabromide.

16.3.2. Chemistry of Element 106 (Sg)

The first discovered isotope (Ghiorso et al., 1974), 0.9-s 263 Sg, remained the longest
lived known isotope of Sg for 20 years. In 1994, a Dubna-LLNL collaboration
(Lazarev et al., 1994; Lougheed et al., 1994) reported production of the longer lived
isotopes 266 Sg and 265 Sg (Fig. 16.2) with maximum cross sections of approximately
0.08 and 0.26 nb, respectively. Sg is expected to be the heaviest member of group 6
in the periodic table. Considerations based on both relativistic molecular–orbital
calculations (Pershina and Fricke, 1996) and empirical relationships (Eichler et al.,
1999) predict that SgO2 Cl2 should be the most stable of the Sg oxychlorides and
that the order of volatility of the dioxydichlorides of the group 6 elements should
be Mo > W > Sg.
Attempts were made at Dubna in 1996 (Timokhin et al., 1996; Yakushev et al.,
1996) to use the 0.9-s 263 Sg produced in the 249 Cf (18 O, 4n) reaction to study the
volatility of the oxychloride in experiments with the online quartz gradient ther-
mochromatographic column (TC) technique discussed earlier. They attributed 29
fission tracks found in a temperature zone of 150–250◦ C (close to the 166 W de-
position temperature) to decay of the very small SF branch of 263 Sg, and claimed
that this was the first chemical identification of Sg. However, no positive iden-
tification based on the much more abundant, known α-decay chain of 263 Sg,
was made and the detected fissions could have originated from other elements
such as 104 and 105. Kratz (1999b) reviewed these reports in detail and con-
cluded that the data did not support these claims to the first chemical identification
of Sg.
The behavior of the oxychlorides of short-lived isotopes of Mo and W produced
with a reactive gas mixture of O2 , Cl2 , and SOCl2 was investigated using OLGA
III (Gärtner et al., 1997) in preparation for studies of Sg. The adsorption enthalpies
of –90 kJ mol−1 and –100 kJ mol−1 deduced for Mo and W from the isothermal
yield curves were in agreement with the predictions. Subsequently, the first chem-
ical experiments on Sg were performed (Schädel et al., 1997a) with OLGA III
by an international collaboration in 1995 and 1996 at the UNILAC at GSI using
266,265
Sg produced as shown in Table 16.2. The volatile species exiting OLGA
III were again attached to aerosols and transported and deposited sequentially on
thin polypropylene foils placed in 64-collection positions around the periphery of
a rotating wheel system. It was stepped every 10 s to place the collected activ-
ity between pairs of passivated ion-implanted planar silicon (PIPS) detectors for
measurement of α- and SF activities. Times, energies, and positions of all events
were registered in list mode with a computer system. A mother–daughter stepping
 16. Chemistry Beyond the Actinides 355

mode was used to avoid interference from the ubiquitous 212 Pom 8.8–MeV α-
activity. Three events attributable to 265 Sg (7 s) were identified: two in daughter
mode and one triple correlation. One α–α correlation between 266 Sg (21 s) and
its 262 Rf daughter was detected. This gas-phase study of SgO2 Cl2 constituted the
first chemical study in which the detected Sg activities were positively identified
as belonging to Sg.
In a later experiment (Türler et al., 1999), the isothermal temperature was varied
to obtain the adsorption enthalpy for Sg. The yield of short-lived W isotopes from
reactions on Gd incorporated in the target was also measured at the same time. An
adsorption enthalphy of –96 ± 1 kJ mol−1 was obtained for the dioxydichloride of
W in agreement with the previous measurement. Based on 11 events attributable
to Sg, an adsorption enthalpy of –100 ± 4 kJ mol−1 or 96 ± 1 kJ mol−1 was
obtained depending on the Sg half-life used in the Monte Carlo calculations. Thus,
the experimental results confirmed the theoretically predicted volatility sequence
of Mo > W > Sg. From an analysis of the decay properties of the correlated decay
chains detected in these experiments, Türler et al. (1998) were able to recommend
better half-lives and cross sections of 7.4 + 3.3/ − 2.7 s and approximately 0.24
nb for 265 Sg and of 21 + 20/ − 12 s and approximately 0.025 nb for 266 Sg for 22 Ne
beam energies between 120 and 124 MeV.
Schädel et al. (1997a,b) also reported the first successful studies of the aqueous
chemistry of Sg in the same series of collaborative experiments conducted at the
UNILAC at GSI. In order to favor production of the longer lived 266 Sg formed by
the 4n out reaction, 121-MeV 22 Ne projectiles were used. A helium gas-transfer
system was used to transport the activities from the irradiation site to the ARCA
II system, where 3,900 collection and elution cycles were performed. A solution
of 0.1 M HNO3 /5 × 10−4 M HF was used to elute the activity sorbed initially
on columns (1.8 mm i.d. × 8 mm long) filled with the cation exchange resin
Aminex A6. This eluant was chosen because previous online tracer experiments
had shown that the formation of neutral and anionic complexes with F− ions is
characteristic of the group 4, 5, and 6 elements, but that there are distinct differ-
ences in behavior between the groups. The procedure was designed to provide
rapid decontamination from the interfering high-energy Bi and Po α-activities and
trivalent actinides also formed in the irradiation, and efficient separation of the Rf
and No daughter activities from their Sg precursors. The mean time for separation
of Rf and No from Sg was only approximately 5 s, but α-particle and SF measure-
ments were not begun until approximately 38 s after the end of collection because
of the time required to evaporate the eluted samples and manually transport them
to the PIPS detectors. The energies, times, and detector positions were recorded
in list mode on magnetic disk and tape for later data analysis. Collection times of
45 s were used for most of the cycles.
Three correlated α–α-events were identified as belonging to the 261 Rf
(78 s) →257 No (26 s) → decay sequence, thus indicating that the 7-s 265 Sg parent
had been present in the chemically separated Sg (W) fractions and decayed to Rf
and No during the time interval before measurements began. No evidence for the
longer lived 266 Sg was found, presumably because of its much smaller production
356 Darleane C. Hoffman

cross section. Later investigations were performed (Strub et al., 2000) to verify that
Rf, Th, Hf, and Zr would not elute from the cation columns under these conditions,
and it was concluded that 261 Rf could have been only in the Sg fraction as a result
of the decay of 265 Sg and that Sg, like its lighter group 6 homologs Mo and W,
formed neutral species of the type MO2 F2 .
Another similar series of more than 4,500 cycles was conducted (Schädel et
al., 1998) under the same conditions using 0.1 M HNO3 but without any flu-
oride ions to determine whether Sg would then behave in a manner different
from that of W as suggested by theoretical examination of the hydrolysis of
group-6 elements and U(VI) (Pershina and Kratz, 2001). Indeed, the Sg activ-
ity remained on the cation-exchange resin columns while W was eluted with
0.1 M HNO3 . This behavior was tentatively attributed to the weaker tendency
of Sg to hydrolyze in dilute HNO3 so that it remained as a 2+ or 1+ complex
while hydrolysis of W (and Mo) resulted in the neutral species MO2 (OH)2 . In
the earlier experiments containing fluoride ion, Sg may have been eluted from
the cation-exchange columns as neutral or anionic fluoride complexes of the type
SgO2 F2 or [SgO2 F3 ]− rather than as [SgO4 ]2− . Theoretical calculations of com-
plex formation in HF solutions (Pershina and Hoffman, 2003) indicate that com-
plex formation competes with hydrolysis in aqueous solutions and the dependence
on pH and HF concentration is extremely complicated and reversals in trends
may occur. The theoretical predictions provide experimenters with valuable guid-
ance for planning future experiments. These studies illustrate the fruitful syner-
gism that can result from close interactions and iterations between theory and
experiment.

16.3.3. Chemistry of Element 107 (Bh)

The discovery of 17-s 267 Bh (Wilk et al., 2000) produced via the 249 Bk(22 Ne,
4n) reaction provided an isotope long enough for the first chemical studies of
Bh. Online IGC experiments were conducted with OLGA at the PSI cyclotron
to compare the volatility of the oxychloride of Bh with those of Re and Tc, its
expected homologs in group 7 of the periodic table. The recoiling reaction products
were transported to OLGA on carbon particles suspended in flowing He gas. The
volatile chlorides were produced by treatment with HCl and oxygen, and the species
passing through the quartz chromatographic column were reattached to aerosols
and transported to a rotating wheel system for measurement of α-particles and
SFs. Over a period of about a month of irradiation time, six decay chains were
attributed to the decay of 267 Bh from observation of its decay to known daughter
activities.
A Monte Carlo program based on a microscopic model of the adsorption pro-
cess was used to deduce an adsorption enthalpy of –75 kJ mol−1 with a 68%
confidence interval of –66 to –81 kJ mol−1 for BhO3 Cl, predicted to be the most
probable oxychloride under these conditions. The values for Tc and Re were –51
and –61 kJ mol−1 , respectively. These results were in agreement with the calcu-
lations (Pershina and Bastug, 2000) that predicted a volatility sequence for the
 16. Chemistry Beyond the Actinides 357

trioxychlorides of Tc > Re > Bh. Thus Bh can properly be placed in group 7 of

the periodic table.

16.3.4. Chemistry of Element 108 (Hs)

Chemical separation procedures for Hs were developed on the basis of its expected
similarity to the group 8 elements Ru and Os. Fully relativistic density functional
calculations have been performed for the tetroxides of these elements (Pershina
et al., 2001). The calculated adsorption enthalpies for the tetroxides were −36.7 ±
1.5, −38 ± 1.5, and −40.4 ± 1.5 kJ mol−1 for Hs, Os, and Ru, respectively, giving
the volatility sequence Hs ≥ Os > Ru.
Prior to experiments with Hs, a cryo-thermochromatographic separator (CTS)
was developed with a channel formed by two facing rows of silicon PIN (Positive
Implanted N -type silicon) diode α-particle detectors at Berkeley by Kirbach et al.
(2002). They performed online studies of the tetroxides of α-emitting Os activities
produced at the LBNL 88-in. cyclotron. It was operated with a negative temperature
gradient ranging from 247 K at the entrance to 176 K at the exit of the channel. The
volatile species condense on the surface of the detectors at a characteristic position
along the channel. Temperatures are determined from thermocouples placed at
intervals along the channel and from measured resistances of the PIN diodes. Data
are stored continuously on disk or tape via a computer system for future offline
data analysis. The radioactive nuclides in the volatile species are identified from
their measured α-decay energies and half-lives.
The BGS was used as a preseparator to remove unwanted transfer reaction
products and scattered beam particles. The separated Os ions were stopped in a gas
mixture of 90% He and 10% O2 and then transferred in a flowing He/O2 mixture to
a quartz tube heated to 1,200 K where OsO4 was produced and then transported to
the CTS. An adsorption enthalphy of –40.2 ± 1.5 kJ mol−1 was obtained for OsO4
on the quartz (SiO2 ) surface from Monte Carlo fits to the measured adsorption
distributions.
The IVO-COLD system shown earlier in Fig. 16.4 was used in the Hs experi-
ments that were conducted at the GSI/UNILAC. Three 248 Cm targets on a rotating
wheel were used to optimize the production rate. The α-emitting isotopes 269,270 Hs
were produced as shown in Table 16.2 and identified by measurement of their de-
cay to known α-emitting daughters. The detection efficiency was about 77% for a
single α-particle. The yield of short-lived Os isotopes was checked by online mea-
surements before and after the end of the Hs experiments. The products recoiling
from the nuclear reaction were thermalized and stopped in the recoil chamber of
IVO containing He gas, while the high-energy beam particles passed on through
to the beam stop and thus removed from the thermalized recoil products. The re-
coiling products were transported to a quartz wool plug where Hs and Os were
oxidized to the tetroxides by treatment with oxygen at 600◦ C. They were then
transported by the carrier gas to the COLD TC, which is formed by 12 pairs of
opposing Si PIN photodiodes positioned so the gas flow is confined to the active
detector surfaces coated with an outermost layer of Si3 N4 (Düllmann et al., 2002b).
358 Darleane C. Hoffman

60

50

40
Relative Yield [%]

30

20

10

0
–20 –40 –60 –80 –100 –120 –140 –160
Deposition Temperature [∞C]

FIGURE 16.8. Merged thermochromatogram of HsO4 and OsO4 . Adapted from Düllmann
et al. (2002b).

A negative temperature gradient from –20◦ C to –170◦ C was established along the
TC and measured at five positions.
Three α–α correlated decay chains detected in the 64-h experiment were as-
signed to the decay of 269 Hs. Two others possibly due to a new nuclide 270 Hs or an
isomer of 269 Hs were also detected. The merged thermochromatograms for OsO4
and HsO4 are shown in Fig. 16.8. Indicated are the relative yields of HsO4 and
OsO4 as a function of the deposition temperature. Measured values are represented
by bars—HsO4 : light grey; OsO4 : dark grey. The distributions of the seven decay
chains attributed to Hs isotopes are indicated by the pattern—269 Hs: blank; 270 Hs:
hatched; Hs (isotope unknown): cross-hatched. For Os, the distribution of 1·105
events of 172 OsO4 is given. The maxima of the deposition distributions were evalu-
ated as (−44 ± 6)◦ C for HsO4 and (−82 ± 7)◦ C for OsO4 where the uncertainties
indicate the temperature range covered by the detector that registered the maxi-
mum of the deposition distribution. Solid lines represent results of a Monte Carlo
simulation of the migration process of the species along the column with standard
adsorption enthalpies of −46.0 kJ mol−1 for 269 HsO4 and −39.0 kJ mol−1 for
172
OsO4 .
As noted above, the best Monte Carlo fits to the Os data gave an adsorption
enthalpy of −39 ± 1 kJ mol−1 for OsO4 on the Si3 N4 surface in good agreement
with previous measurements on SiO2 . Using only the three events assigned to
269
Hs and a half-life value of 11 s in the Monte Carlo analysis resulted in an
adsorption enthalpy of −46 ± 2 kJ mol−1 for HsO4 . This is considerably lower
than the calculated value of −36.7 ± 1.5 kJ mol−1 and suggests that it may be
less volatile than OsO4 . Additional experiments to measure more 269,270 Hs decay
 16. Chemistry Beyond the Actinides 359

chains and to clarify the nuclear properties are clearly needed. However, the study
does show that Hs forms a volatile oxide similar to that of the tetroxide of Os, and
should be placed in group 8 of the periodic table.
No studies of Bh or Hs in solution have yet been attempted, but they should be
quite interesting and could furnish valuable information about most stable oxida-
tion states in aqueous solution and redox and complexing reactions in different
media. They might be expected to show oxidation states ranging from 3 to 7 for
Bh and 1 to 8 for Hs, as do their homologs in groups 7 and 8. A summary of the
chemical properties predicted for Bh and Hs was given in a review by Seaborg and
Keller (1986).

16.4. Future
16.4.1. Prospects for Chemical Studies of Mt (109)
Through Element 112
Very little attention has been paid to theoretical predictions of the chemical prop-
erties of Mt through element 120 since the 1986 summary of Seaborg and Keller.
Based on the positions of Mt and element 112 in the periodic table, they should be
noble metals like Pt and Au, and volatile hexafluorides and octafluorides might be
produced and used in chemical separation procedures. Early relativistic molecular
calculations (Waber and Averill, 1974; Rosen, 1998) suggested that 110F6 should
be similar to PtF6 .
Because the maximum of relativistic effects in the ns shell in group 11 is at
element 111, there has been considerable interest in the electronic structures of its
compounds. Theoretical studies (Seth et al., 1996; Liu and van Wüllen, 1999) of
the simplest molecule 111H show that the bonding is considerably increased due
to relativistic contraction of the 7s orbital. Theoretical studies (Seth et al., 1998a,b)
of the stability of higher oxidation states support earlier predictions that the 3+
and 5+ oxidation states will be more common for element 111 than they are for
Au and that the 1+ oxidation state may be difficult to prepare.
Element 112 is the most interesting of the heaviest elements from the chemical
point of view because the maxima of both the relativistic effects on the 7s shell in
the entire 7th row and within group 12 occur at element 112. The strong relativistic
contraction and stabilization of the 7s orbitals and its closed shell configuration
should make it rather inert, and the predicted rather small interatomic interaction
in the metallic state may even lead to high volatility as seen in the noble gases
(Pitzer, 1975). A simple extrapolation of the sublimation enthalpies of its lighter
group 12 homologs, Zn, Cd, and Hg, leads to a similar conclusion.
Recent relativistic molecular orbital calculations (Pershina et al., 2002) con-
firmed that element 112 can form rather strong metal–metal bonds 112M and
deposit on some transition metals as 112M in which M = Cu, Pd, Ag, Pt or Au.
The stability of the 2+ state is predicted to decrease from Hg to element 112,
and 112F2 may tend to decompose but 112F4 may be stable. The species 112F− 3
360 Darleane C. Hoffman

and/or 112F− 5 may form in solution in an appropriate polar solvent, but they will
probably quickly hydrolyze and bromides or iodides may be more stable and bet-
ter for experimental studies of solution chemistry. A recent review of the results
of theoretical studies of the properties of Mt through element 112 has been pub-
lished (Pershina and Hoffman, 2003) and should be consulted for more detailed
and up-to-date information and complete references.
Some preliminary chemical experiments on element 112 have been reported
(Yakushev et al., 2001) using the spontaneously fissioning nuclide 283 112 (∼3
min) reportedly formed in the 238 U(48 Ca, 3n) reaction with a cross section of about
5 pb. The experiments were designed to test whether element 112 behaved more
like Hg, which had been shown to deposit on Au- or Pd-coated silicon surface-
barrier detectors or was a noble gas like Rn and stayed in the gas phase. Eight
SF events were detected in the gas phase, which would indicate Rn-like behav-
ior. However, the results cannot be considered definitive since it was not shown
that the SFs belonged to element 112 nor has the original report of the approxi-
mately 3-min SF activity attributed to 283 112 been confirmed. Additional exper-
iments are planned, but use of an α-decaying isotope of 112, such as 284 112 (10
to 70 s) reported previously (Oganessian, 2001) to be the daughter of 288 114,
may be required to ascertain whether or not element 112 was actually being
observed.
Chemical studies of Mt depend on the discovery of longer lived isotopes than
42-ms 268 Mt, currently the longest known isotope of Mt, shown in Fig. 16.2. Longer
lived isotopes might be expected around the deformed nuclear shell at 162 neutrons.
The half-life of the 162-neutron nuclide 271 Mt can be estimated from interpolation
of Smolańczuk’s calculations (Smolańczuk, 1997) for even-proton, even-neutron
nuclides to be a few seconds and should decay primarily by α-emission. However,
270
Mt with only 161 neutrons may be only tens of microseconds, too short for
chemistry. Possible production reactions for 271,270 Mt include 238 U(37 Cl, 4n,5n)
and 249 Bk(26 Mg, 4n,5n). The cross sections for these reactions may be a few pb
and only tenths of pb, respectively. The BGS at LBNL is capable of separating
and positively identifying such a new nuclide based on measurement of the known
α-decay chain of its daughters even if the 271 Mt half-life is as short as tenths of
seconds. Rotating, multiple targets of 238 U could be used to increase the production
yields for the first reaction but use of multiple targets of the highly radioactive 249 Bk
would be extremely difficult.
The half-lives of 7.5 s and 1 min reported for 280,281 110 produced as granddaugh-
ters of the 288,289 114 decay chains are long enough for chemical studies. However,
the reports need to be confirmed and the yields via this decay route must be high
enough to permit statistically significant atom-at-a-time studies. Currently, there
appear to be no suitable reactions for direct production of these very neutron-rich
isotopes of element 110.
Some isotopes of element 111 should have half-lives of seconds or more. Promis-
ing production reactions need to be investigated first using an online separator and
detection system to measure half-lives and production cross sections before chem-
ical studies can be contemplated.
 16. Chemistry Beyond the Actinides 361

16.4.2. SuperHeavy Elements

The experimental discovery that the elements Bh through element 112 decay pri-
marily by α-emission, rather than SF as predicted earlier, sparked interest in re-
newing the quest for the long-sought island of stability originally predicted to be
around atomic number 114 and neutron number 184. Although recent theoretical
calculations (Smolańczuk, 2001a,b) predict that isotopes with half-lives of mi-
croseconds or longer will exist all along the route to the region of SuperHeavy
Elements (SHEs), the half-lives predicted for this spherical SHE region have de-
creased dramatically since the 1970s (Hoffman and Lee, 2003). Smolańczuk has
recently predicted that the doubly magic spherical SHE nuclide 298 114 will decay
predominantly by α-emission with a half-life of only about 12 min, but that 292 110
will be more stable and α-decay with a half-life of about 50 years. Others have
predicted that the strongest spherical shell effects might be at atomic numbers of
124 or 126 and neutron number of 184.
It is unclear how many more elements can exist, but it now appears probable that
many more nuclides with half-lives long enough for chemical studies can exist. The
question is not only what reactions to use to produce and positively identify them,
but how to enhance their production rates so chemical studies will be feasible.
New target arrangements to take advantage of higher beam intensities must be
developed. Dedicated beam lines for use by international groups of scientists at
suitable accelerators with preseparators or other special instrumentation prior to
continuous automated, online systems for chemical studies might be envisioned.
If longer lived new elements are discovered, then methods for “stockpiling” them
for offline chemical studies by producing them as by-products of other experiments
might be devised. It is tantalizing to consider the possibility of extending chemical
studies beyond Hs in order to fully explore the influence of relativistic effects and
to further define the architecture of the periodic table—or perhaps even be forced
to abandon the current form of the periodic table. Much ingenuity, dedication, and
perseverance will be required to explore this exciting new frontier.
17
Mass Spectrometric Radionuclide
Analyses
JOHN F. WACKER1,2 , GREGORY C. EIDEN1 , and SCOTT A. LEHN1

17.1. Introduction
Like most analytical instrumentation, the mass spectrometer (MS) initially found
use as a research system. These complex laboratory-built research instruments
were subsequently refined for commercial sale and widespread use, and are now
available for routine sample assay.
In recent years, the MS has become a useful alternative to the radiation detector
for measuring longer-lived radionuclides. It provides a second and completely
independent method of measurement to confirm results; it also gives the analyst
a choice, because some measurements are easier or more reliable by one method
than the other. The MS has been used to measure radioactive atoms with half-
lives greater than 10 years because the number of these atoms relative to their
decay rate is proportional to the half life; for half lives greater than 109 y, even the
conventional measurements of analytical chemistry are applicable.
Development of new MS systems focuses on the ability to measure smaller
samples accurately, either for greater sensitivity or shorter-lived radionuclides.
Advances in both construction and design have been applied, with the effect that
the modern MS operator can measure ever-smaller numbers of atoms. A related de-
velopment is the ability to individuate those samples more thoroughly by increasing
MS resolution to decrease interference from isobars—atoms and molecules with
the same atomic mass number but with minute differences in mass.
The distinction in sensitivity between measurements of radiation and atoms (as
ions at a given mass-to-charge ratio by MS) can be demonstrated by comparing
intensity in terms of the radioactive decay Eqs. (2.4) and (2.7) for a radionuclide
with an atomic mass of 100 amu and a half-life of 10 y (3.16 × 108 s). If the
sample under assay emits beta particles at the relatively low rate of 1 × 10−2 d/s,
this corresponds to 4.6 × 106 atoms or, for a 100 amu radionuclide, 7.6 × 10−14 g.
Alpha particles can be measured at an approximately 1,000-fold lower decay rate,
for which the number of atoms and the mass correspondingly are 1,000-fold lower.

1
Pacific Northwest National Laboratory, Richland, WA 99352
2
To whom correspondence should be addressed. John F. Wacker, MS P8-01, 902 Battelle
Blvd., Richland, WA 99354; email: john.wacker@pnl.gov

362
 17. Mass Spectrometric Radionuclide Analyses 363

The radiation of this test isotope can be measured at these levels but the number of
atoms (4.6 × 103 ) falls below the detection limits for quantification of most mass
spectrometers.
The efforts in applying mass spectrometry to the measurement of radionuclides
are devoted to achieving detection of small numbers of atoms. The crucial question
is whether an isotope that has, say, 106 atoms in a gram of solid or liquid can be
detected and quantified by MS in the presence of possibly 1021 atoms and molecules
of other isotopes, including many with nearly the same mass. For some elements
(e.g., Pu), detection limits below 106 atoms have been demonstrated for a variety of
sample matrices (soil, water) using several mass spectrometric techniques (Beasley
et al., 1998b; Oughton et al., 2004; Sahoo et al., 2002). Instrumental detection
limits, in terms of counts (i.e., detected ions) per atom, have exceeded 5 counts
per hundred atoms under optimum conditions.
This chapter provides an overview of mass spectrometer function and operation.
It describes specific instrument types with demonstrated or potential application
for measuring radionuclides and surveys the application of these instruments to
radionuclide detection. Finally, it discusses the circumstances under which use
of mass spectrometers is advantageous, the type of mass spectrometer used for
each purpose, and the conditions of sample preparation, introduction and anal-
ysis. Its perspective is from a national laboratory active in environmental and
non-proliferation monitoring. It emphasizes isotope ratio measurements, but mass
spectrometric measurements also provide isotope mass information. Several re-
cent books describe elemental and isotope ratio mass spectrometry in far greater
detail than is presented here (Barshick et al., 2000; De Laeter, 2001; Montaser,
1998; Nelms, 2005; Platzner, 1997; Tuniz et al., 1998). High-resolution mass spec-
trometry forms the basis of the mass scale used for elemental and isotopic masses
(Coplen, 2001), but this application of MS falls outside the scope of this chapter.

17.2. Instrumentation and Analytical Procedures

Mass spectrometers analyze gaseous ions by separating ions with differing mass-
to-charge ratios (m/z; where z is the ion charge) and detecting these separated ions.
The data produced are in the form of ion intensity at specified ion mass.
The first step in this analysis is the preparation of the sample. The point is to
put the analyte into a form that is both compatible with the instrument and free of
impurities. Next, a system is needed to introduce the sample into the instrument.
The mass spectrometer itself is composed of the following equipment, through
which the sample is moved in order to achieve the analysis:

r an ion source to volatilize and ionize the analyte of interest

r a mass analyzer to perform the mass separation of the ions
r an ion detector to detect the separated ions
r a vacuum chamber with associated vacuum pumps and hardware to house in-
strumental components
364 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

Each of the instrumental components has electronic control for operation and
data acquisition.

17.2.1. Sample Preparation and Insertion

Almost all types of samples require preparation to convert the analyte of interest
into a form that is readily inserted into a mass spectrometer for isotopic assay.
Although gaseous samples are injected directly into the source for the instrument,
purification and separation with reactive metals (also called getters), gas chro-
matography, and cryogenic separation means may be required. A few relatively
pure liquids and volatile solids can be vaporized directly, but most liquids and
solids are processed by chemical means to prepare solutions for assay. For solids,
steps must be taken to dissolve the sample; ashing may also be required to remove
organic material. Finally, reducing the sample mass may be desirable to avoid
instrumental problems, such as perturbing the plasma in an inductively-coupled
plasma mass spectrometer (ICP/MS).
In addition to the preparation of the physical form of the sample, it must often be
chemically processed to remove impurities. In traditional radiochemistry that uses
radiation detection methods, the term “impurity” refers to the presence of another
radionuclide that emits radiation of a type and energy that overlap emission from
the radionuclide of interest. When using MS as the detection method, an impurity
is defined as a material that has a mass close to that of the analyte of interest. These
isobaric interferences obfuscate the identification of characteristic mass peaks and
the measurement of the number of ions from the analyte of interest at these peaks.
Interferences can be caused by isotopes of other elements (e.g., 238 U interferes
with 238 Pu) or molecular species (e.g., both 208 Pb27 Al and 200 Hg35 Cl can interfere
with 235 U). Isobaric interferences are particularly serious for assaying radionu-
clides because of their (usually) low abundance relative to the interfering species.
Thus, the process of chemically preparing the sample stresses purification and
contamination minimization; this means controlling both the analyte in question
as well as other, often non-radioactive, species.
Chemical separations may be specific for the analyte of interest (see Chapter
3), such as liquid or gas chromatography, or scavenging (such as by precipitation)
to remove the major interfering substances. Addition of carrier, as practiced in
radioanalytical chemistry to assist in purifying radionuclides, usually is not ap-
propriate for mass spectrometric analysis. Such addition undermines the isotopic
ratio measurements that are often at the heart of this procedure, and also overloads
the system for ion generation and peak resolution (but carrier addition is used for
accelerator mass spectrometry). Addition of tracers, known as isotope dilution,
is often employed for yield determination (see Section 17.2.9). Interferences are
distinctly different in radiometric and MS analyses of radionuclides, and may be
the deciding factor in selecting one method versus the other.
The need for extensive sample purification depends in part on the instrument.
Typically, the inductively-coupled plasma mass spectrometer (ICP/MS) does not
require the degree of separation needed in thermal ionization mass spectrometry
(TIMS). TIMS offers greater detection sensitivity at the cost of more elaborate
 17. Mass Spectrometric Radionuclide Analyses 365

sample preparation and instrument control. An instrument such as an ICP/MS,

which is operated at higher mass resolution, may require less sample preparation.
Operation at high mass resolution, however, generally comes at the expense of
overall sensitivity. For most radioanalytical analyses, sensitivity is crucial, hence
high resolution analysis is rarely used. The collision/reaction cell interference
reduction method recently introduced in ICP/MS does not suffer from the trade-
off between interference reduction and sensitivity (see Section 17.4.5).
Cleanliness in the laboratory where samples are prepared is critical to main-
taining a low background for high sensitivity analysis. Radioanalytical chemists
are used to assessing chemistry blanks for radiation detection, but in MS analysis,
both radioactive and non-radioactive isotopes can add to the background against
which the sample atoms are measured.
The prepared sample must then be introduced into the MS. Liquid samples,
usually weakly acidified solutions, can be converted into an aerosol by aspiration
into the ICP/MS carrier gas in a mixing tube and inserted as a vapor. Alternatively,
the prepared solutions can be applied to a TIMS filament and dried (see Section
17.5.1). The filament is subsequently inserted into the mass spectrometer. Direct
volatilization from a furnace is an alternative for some elements. Ablation of a
solid surface by a laser followed by direct injection of the aerosol is discussed in
Section 17.7.2.

17.2.2. Ion Source

Because mass spectrometers operate by separating ionized atoms and molecules,
they require an efficient means of producing ions. The ion source ionizes the sample
atoms or molecules and accelerates the ions into the mass analyzer. Table 17.1 lists
the commonly used ionization methods and Table 17.2 summarizes the benefits
and problems associated with each method. Generally, the ion source produces a
well-collimated, mono-energetic ion beam that is injected into the mass analyzer.
Poor collimation degrades the ability of the mass analyzer to separate ions of
different mass. A significant spread in the kinetic energy of the ions also degrades
mass analyzer efficiency and resolution. Both positive and negative ions can be
produced, depending on the nature of the sample and the need to separate the
analyte species from potentially interfering species in the sample.

TABLE 17.1. Ionization methods used for mass spectrometry

Source Ionization mechanism Sample type
Electron impact (EI) e− collisions Gas or vapor
Thermal ionization Evaporation from a hot surface Solid on metal filament
(or emission)
Inductively Coupled Plasma RF plasma at atmospheric Solution, aerosol (includes laser
(ICP) pressure ablated solid), or gas
Glow discharge (GD) DC plasma at ∼1 Torr Solid
Ion bombardment Ion sputtering Solid
Laser ablation Plasma Solid
Resonance ionization Photons Gas, solid, or vapor
366 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

TABLE 17.2. Advantages and disadvantages of the various ionization methods

Source Advantages Disadvantages
Electron Impact (EI) Universal ionization; simple & Large energy spread, multiple
robust charged ions formed
Thermal ionization (or High sensitivity and precision Requires extensive sample chemistry,
emission) possible applicable to limited number of
elements
Inductively Coupled Universal ionization; solutions Ion-molecule reactions create
Plasma (ICP) easily analyzed interferences; ions have significant
energy spread
Glow discharge (GD) Universal ionization, simple Poor rejection of isobaric
sample preparation interferences
Ion bombardment Universal ionization; imaging Complex, ion-molecule reactions
possible, high spatial resolution create interferences; poor rejection
of isobaric interferences;
calibration is difficult
Laser ablation Universal ionization; high spatial Low precision, poor reproducibility,
resolution isotopic fractionation
Resonance Ionization Element– and isotope–specific Expensive and complex
ionization

The following is a brief description of the various ionization sources used in mass
spectrometry. Those sources most commonly used in the analysis of radionuclides
will be described in greater detail later in the chapter.
Electron impact (EI) ionization is useful for elements that are either volatile or
form volatile compounds. Typical ionization efficiencies are in the range of 0.1 to
1 percent. Suitable elements include noble gases and light elements such as C, N,
O. Other elements include those that form volatile compounds (e.g., uranium in
the form of UF6 ). Except for specialized applications (e.g., noble gas analysis), EI
is no longer widely used for elemental and isotopic analysis.
Thermal ionization is useful for elements with ionization potentials less than
about 7 electron volts (eV). The thermal ionization process forms positive ions
when an analyte is evaporated from a hot metal filament. The filament can be a
single filament, where ionization and evaporation occur together, or a double or
triple filament, where the sample is evaporated from one filament and the vapor
is ionized on a separate filament. Common filament materials include rhenium,
tungsten, and tantalum. Whatever the arrangement of the filament, it is always
located above the case plate, shown at the top of Fig. 17.1.
Ions are generated at the filament and accelerated downward towards the exit
plate. In a typical mass spectrometer, the filament is at high voltage (typical voltages
are +5,000 V for positive ions and –5,000 V for negative ions; voltages from
1,000 to 15,000 can be used) and the exit slit plate is at ground. Typical ionization
efficiencies range from 0.1 to 10 percent. Suitable elements include alkali, alkaline
earth, many transition elements, the rare earths, and actinides, as well as lead and
boron. Negative thermal ionization has become important in the past 15 years and
it has increased the range of elements available for analysis by thermal ionization.
 17. Mass Spectrometric Radionuclide Analyses 367

FIGURE 17.1. Thermal ionization source. Figure from Smith (2000), pg. 9.

As with the positive ion case, typical ionization efficiencies are 0.1 to 10 percent.
Suitable elements include the halogens and oxides of transition elements such as
Mo, Re, and Os.
Inductively–coupled plasma (ICP): An argon ICP will ionize all but a few ele-
ments (e.g., fluorine) and, therefore, is a nearly universal ion source. Figure 17.2
shows the ICP source; its ionizing coil and torch are on the right, while the vacuum
interface is shown on the left.
Samples are introduced into the plasma as an aerosol formed by nebulization of
a solution, by vaporization of solids in a furnace, or using laser ablation. Analyte
vapor or aerosol is carried by flowing argon in the central tube in the torch. Argon
also flows in the annular region of the torch. The two argon flows are balanced to
give a stable plasma, which is driven by radiofrequency induction from the load
coil. Analyte ions are formed by collisions with argon ions and other species in
the plasma; these are subsequently drawn into the mass spectrometer through a
differentially pumped gas inlet. Typical ionization efficiencies range upwards to
100%, but this figure is misleading as the transport of the ions from atmospheric
pressure to the vacuum of the mass spectrometer is inefficient. Overall efficien-
cies are typically 0.1% or less (ions detected compared to atoms introduced into
the plasma). Nominal voltages are near ground for a quadrupole mass analyzer,
whereas for a magnetic sector analyzer the entire ICP source (torch and vacuum
interface) is held at high voltage (thousands of volts).
Secondary ionization mass spectrometry (SIMS) is used mainly for the analysis
of solid compounds and materials. A focused ion beam (called the primary ion
beam) is directed against a surface containing the analyte of interest and ions are
formed by the sputtering action of the impacting ion beam. Figure 17.3 shows a
SIMS source, the actual dimensions of which are exaggerated for clarity. Note
that some of the ions from the incoming beam remain embedded in the sample.
Secondary ions are extracted by ion optics and accelerated into a mass analyzer
for measurement. The sample is normally at high voltage for efficient extraction
368 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.2. ICP source. Figure from Gray and Date (1983), Fig. 1.2.

and focusing of the secondary ions. Typical ionization efficiencies range from
<0.1 to about 1 percent. Virtually the entire periodic table can be analyzed with
this method. Both positive and negative ions can be produced, depending on the
composition and polarity of the primary ion beam.
Laser ablation can be used to volatilize samples. An intense laser pulse is focused
onto a solid at sufficient pulse power and energy (e.g., milliJoule energy in a pulse
of ∼10 nanosecond, or less, duration) to remove material from the surface. Typical
conditions result in crater formation after one or a few laser pulses. If performed
in flowing argon at atmospheric pressure, the ablated material can be fed into an
ICP for ionization. Ions may also be formed by an ablation pulse in a vacuum and
directly focused and extracted into a mass analyzer. Laser ablation/ionization is
used primarily for analysis of solid compounds and materials. Virtually the entire
periodic table can be analyzed with this method.
Resonance ionization mass spectrometry (RIMS) uses absorption of narrow
bandwidth laser light to ionize an element of interest. The wavelength is chosen
 17. Mass Spectrometric Radionuclide Analyses 369

FIGURE 17.3. SIMS source. Figure courtesy of G. Gillen, National Institute of Standards
and Technology, http://www.simsworkshop.org/graphics.htm (Jan. 2006).

to coincide with an allowed electronic transition of the atom (or isotope) of in-
terest. Single photon transitions are exploited as well as “multi-photon” transi-
tions. Isotopic selectively can be achieved with sufficiently narrow wavelength
tuning. This method offers very high selectivity compared to other ionization
methods, but usually with relatively low ionization efficiencies (<1%). Under
specialized conditions, such as specific ion source geometry, narrowband ex-
citation by continuous-wave lasers, rapid heating, pulsed laser operation, and
well-matched laser diameter to sample size, much higher ionization efficien-
cies are possible. RIMS systems usually are found only in laser research lab-
oratories because the lasers needed for resonance ionization are complex and
expensive.
A notable exception regarding cost and complexity is the method of resonant
laser ablation (RLA). In RLA, a low-fluence pulsed laser beam is focused onto
the solid sample (Eiden et al., 1994). The laser pulse first ablates or desorbs a
small amount of sample. On the timescale of the laser pulse (typically a few
nanoseconds), the atoms liberated from the surface absorb laser photons and are
ionized. By using a laser wavelength resonant with an atomic transition in the
atoms of interest, ionization is highly selective. This method has been extensively
demonstrated with relatively low-cost YAG pumped dye lasers.
The ions produced by any of the preceding sources are then detected by the
mass analyzers described below.
370 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

TABLE 17.3. Common mass analyzers

Mass range Mass resolution Electromagnetic
Analyzer (amu) (M/M) Duty cycle fields1
Sector field 1–10,000 300–10000 High to DC M or E
continuous
Quadrupole mass filter 1–1000 100–1000 Low DC & RF E
(QMF)
Ion trap (IT) 1–10,000 300–3000 Medium DC & RF E
Time-of-flight (TOF) 1–1,000,0002 100–5000 Low Pulsed DC E

1 M: magnetic, E: electric, DC: direct current (continuous), RF: radiofrequency

2 Greater limitation is usually poor response of ion detectors to heavy, slow moving ions.

17.2.3. Mass Analyzer

An MS is constructed with one or more of the mass analyzers listed in Table 17.3.
All of these analyzers separate ions based on mass, momentum, or velocity. Many,
such as the sector and TOF analyzers, require monoenergetic ions. Mass analyzers
of different types can be combined to create instruments with specific capabilities,
such as high mass resolution or abundance sensitivity (i.e., the ability to detect a
small-abundance mass peak near a large-abundance one). Combinations include
instruments with multiple magnetic and electrostatic sectors, multiple quadrupole
mass filters (QMF), quadrupole-time–of-flight (TOF) combinations, ion trap (IT) -
TOF combinations, and quadrupole-IT combinations. Mass spectra are generated
by varying an operating parameter (e.g., magnetic field for a sector field analyzer)
and observing the ion intensity as a function of this parameter.
Most mass analyzers for ICP/MS instruments are operated in a scanning mode.
These mass analyzers operate as bandpass filters, passing a single mass-to-charge
ratio at a time for detection by the ion collection system. The mass analyzers are
scanned either by “peak-hopping” or by continuous scan.
Most modern instruments employ the “peak-hopping” (also called peak switch-
ing) method; this allows the analyst to “hop” among the mass peaks of interest
during the assay. The ion intensity is integrated for a period of time at each peak.
The time spent “on peak” is maximized relative to the time between peaks. The
peak switching mode requires more sophisticated and accurate electronics but of-
fers the advantage that a maximum amount of time is spent measuring ions of
interest and minimal time is spent analyzing the unwanted parts of the mass spec-
trum between the peaks. The time spent at each peak translates into more counts;
recording the maximum possible number of counts is desirable for detecting ultra-
trace analytes.
At lesser sensitivity requirements, or if a measurement of the peak shape is
desired, a continuous scan of mass to record full peaks may be preferred. In
modern instruments, control is really a digital process, so that the continuous scan
mode is essentially peak hopping with a small mass interval—0.1 amu or less per
step. Very high dynamic range instruments, in which the peak “tails” are to be
measured, might use 50 or more steps per amu.
 17. Mass Spectrometric Radionuclide Analyses 371

FIGURE 17.4. ICP/MS mass spectrum of U and Th isotopes.

Figure 17.4 is an ICP/MS mass spectrum (with a QMF mass analyzer) of U

and Th isotopes; both scanning methods are illustrated. The small dots along the
spectrum show a continuous scan (actually consisting of many sub-amu shifts).
The large arrows show peak switching where the scan is halted and the QMF is
tuned to the centers of the mass peaks. In this figure, masses 232, 233, 235, and
238 have significant ion mass peaks.
To measure an isotopic ratio, the analyzer is set to the first mass of interest, the
signal is acquired for a selected duration or “dwell” time, and the analyzer is then
switched to the next mass of interest for signal acquisition. The optimum dwell
time is a function of many factors, but the main considerations are instrument noise
and ion counting statistics. The dwell time should be long enough to average out the
high frequency components of the noise, but not so long as to suffer from long-term
drift or low frequency noise. Longer dwell times are needed for the less abundant
isotopes to register sufficient counts for the desired level of precision. Quadrupole
systems can be scanned much faster than magnetic sectors; dwell times for the
former range from fractions of a ms to perhaps 100 ms whereas magnetic sector
dwell times range from 0.01 to 1 s. TOF and ion trap mass analyzers are scanned
in ways that are not easily characterized as “peak hopping” or continuous; these
scan modes are described in the sections on TOF and ion traps.
The magnetic sector field represents the classical mass analyzer. In operation,
a magnetic sector disperses ion masses in a manner analogous to an optical prism
creating a light spectrum. Sector angles of 60◦ and 90◦ are most common. As
shown in Figure 17.5, the ion source (labeled O) is at the top of the figure and
the ion collector (labeled C) is at the bottom. The arrangement depicted is the
most common one; the magnetic sector is designed to be symmetric, with the
source–to–magnet and collector–to–magnet distances being equal. Other design
configurations are possible.
Sectors require precisely collimated mono-energetic ion beams, which places
additional requirements on the ion source and ion optics. Given a mono-energetic
beam, the equation for a mass spectrometer with a sector of radius R and with ions
accelerated to a kinetic energy V is:

m/z = e R 2 B 2 /2V (17.1)

372 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.5. Magnetic sector analyzer. Figure from Platzner (1997), pg. 5.

where:
R = radius
m = ion mass
e = electron charge
z = ion charge (in units of electron charge e)
V = ion accelerating voltage
B = sector magnetic field
A single sector can achieve abundance sensitivity (i.e., the ability to detect a
small abundance mass peak adjacent to a large abundance one) in the range of
1 ppm (i.e., an isotope ratio of 1 × 10−6 ). Abundance sensitivity can be critical for
rare isotopes; an example is 236 U, which occurs only at extremely low abundance
(ratio relative to 238 U < 10−10 ) in natural U, but is produced by irradiation of
uranium in a reactor (236 U abundances vary from 10−10 to 10−2 compared to the
abundances of all U isotopes present). Common sector geometries include 60◦ and
90◦ angles, and common radii include 15 and 30 cm.
A larger radius leads to higher dispersion of the masses (e.g., more separa-
tion between adjacent masses). High dispersion leads to better mass resolution
and abundance sensitivity and enables the use of larger detectors and ion sources,
 17. Mass Spectrometric Radionuclide Analyses 373

which simplifies the design and construction of some component. High disper-
sion results in a larger overall instrument because the size of the instrument scales
directly with the sector radius. This is especially true of the magnet, which for
large instruments can weigh hundreds to thousands of kilograms. Large electro-
magnets are expensive and require large power supplies and control electronics
for stable operation. Modern commercial sector instruments are designed with so-
called extended geometries, which use specialized magnetic focusing to achieve
the equivalent dispersion of a large radius sector with a smaller magnet. Extended-
geometry instruments offer high dispersion at smaller size.
In its simplest form, a mass spectrum is created by scanning the magnetic field
and measuring the ion intensity. Alternatively, the magnetic field is held constant
and the ion source voltage (which controls the ion energy) is scanned. In modern
instruments, peak switching is performed by setting either the magnetic field or
the ion source voltage to values that correspond to the mass peaks of interest.
Sectors have the advantage that multiple ions can be detected simultaneously with
multiple detectors for each mass of interest or with an imaging detector. Early
mass spectrometers used a photoplate for ion detection to measure the entire mass
range at once, but photoplate ion images were difficult to measure quantitatively
for relative ion intensities. In modern instruments, electronic detectors (see Section
17.1.4) are used to measure ions directly.
Sector mass analyzers offer many advantages, including high precision iso-
topic analysis, simultaneous mass detection (with multiple detectors), and stable
operation. Disadvantages include large size, magnets that are heavy and require
sophisticated electronics to operate in a stable manner, and high voltages of more
than 10,000 V.
The quadrupole mass filter (QMF), shown in Fig. 17.6, represents an entirely
different approach to selecting ions by mass. In its typical configuration, the QMF
consists of four parallel cylindrical rods that are set on a square. The opposite
corners are connected electrically to both DC and RF voltages. Ions traverse along
the axis of the rods in a complicated motion and only ions of a selected ion mass-
to-charge ratio can successfully transit the quadrupole. The (a) portion of the
figure shows the conceptual layout, with electrodes with perfect hyperbolic cross-
sections. The (b) portion shows a schematic of the QMF in practice with electrodes
of circular cross-sections. Also shown are the ion source lens, collector, and a
generalized version of the drive electrical circuitry. The QMF uses a combination
of DC and RF electrical potentials to filter ions within a narrow range of m/z. In
effect, the QMF is operated as a bandpass filter.
The equations governing the operation of the QMF are derived from solutions
to Laplace’s equation, which for the geometry of the QMF are described by the
Mathieu equation (Duckworth et al., 1986; March and Hughes, 1989). The fol-
lowing discussion of the Mathieu equation is adapted from March and Hughes
(1989).

d 2u
+ (au − 2qu cos 2ξ )u = 0
dξ 2
374 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.6. Quadrupole mass filter (QMF). Figure from March and Hughes (1989),
pg. 3.

where

ωt
ξ=
2
8eU
au = ax = −a y = (17.2)
mω2r 2
4eV
qu = qx = −q y =
mω2r 2
 17. Mass Spectrometric Radionuclide Analyses 375

FIGURE 17.7. Mathieu stability diagram. Figure from March and Hughes (1989), pg. 50.

where:

r = inner radius between the quadrupole rods

m = ion mass
e = electron charge
V = RF voltage
U = DC voltage
ω = RF frequency
t = time
u = x or y position coordinate

The solutions to the Mathieu equation translate into a series of regions of stable
motion for various values of a and q, which depend on the parameters r , m, V ,
U , and ω. A common approach to visualizing the operational parameters of the
QMF comes from the use of a Mathieu stability diagram (March and Hughes,
1989), which plots a versus q. Figure 17.7 shows a Mathieu stability diagram,
with the operating lines for various resolutions, R (which is M/M), overlayed
on the diagram. For the line marked “R = 10,” only mass M2 is stable and will be
transmitted through the QMF; the other masses will be ejected.
The apex of the stability curve for the primary stable region occurs at a y =
0.23699 and q y = 0.70600. For typical QMF analyzers, a given mass m is analyzed
by use of set values for the parameters r and ω (r and ω typically 1 cm and 1–3 MHz,
respectively) and then varying the parameters U and V (the DC and RF voltages,
respectively). Mass selection is achieved for the desired mass resolution, R, given
376 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

Electron beam
source

Endcap electrode

Sample pin
aperture
Ring electrode

Laser aperture

VespelTM Endcap electrode

spacers

Channel electron
multiplier

FIGURE 17.8. Schematic diagram of an ion trap mass spectrometer analyzer. From Gill et al.
(1991), Fig. 1a.

in the usual form as m/m, where m is the desired mass range for transmission
through the QMF. A mass spectrum is created by scanning the DC and RF voltages
U and V , respectively, while maintaining a constant ratio of U/V . Resolution is
controlled by how close to the apex one scans the U/V ratio.
QMF analyzers offer advantages of light-weight, small size, and rapid response
(an entire elemental mass spectrum can be scanned in less than a second). They
find application in low resolution (∼1 amu), low abundance sensitivity (>1 ppm),
and low measurement precision (∼1%) applications (exception: precision of 0.1%
can be achieved with state-of-the-art ICP/MS instruments with QMF analyzers).
QMF analyzers have distinct advantages over magnetic sectors as the QMF does
not require a large magnet or high voltages, and the QMF is more compact. A
disadvantage of the QMF is that it operates as a bandpass filter and can essen-
tially measure one ion mass at a time, whereas a magnetic sector is capable of
simultaneous multi-mass detection.
The RF quadrupole ion trap mass spectrometer (ITMS) is a close relative of
the QMF and ideally can be thought of as a three-dimensional quadrupole (see
Fig. 17.8). The close relationship of these two devices is evident by the fact that
ion motion in the two devices is governed by essentially the same mathematical
equations. As with the QMF, the ITMS uses DC and RF electric fields and the
operation of the IT is described by solutions to the Mathieu equation. Unlike the
QMF, ITMS analyzers trap ions within the mass analyzer. Ions are trapped, ejected
to select the mass of interest, and then ejected in a controlled manner for detection.
 17. Mass Spectrometric Radionuclide Analyses 377

FIGURE 17.9. Time of flight MS analyzer. Figure from Larsen and McEwen (1998), pg. 20.

Modern ITMS analyzers utilize milli-Torr pressures of helium to cool ions, which
enables higher resolution operation and long (millisecond to second) storage times.
The ITMS is applied in organic and biochemical analysis, and in a limited fashion
for isotopic and elemental analysis. The ITMS in Fig. 17.8 is configured to use a
laser pulse to vaporize and ionize metal atoms inside the mass analyzer. Unique
features of ion traps include the ability to store selectively specific m/z ratios (ion
species).
Many other kinds of ion traps are known, but apart from the ITMS, only the ion
cyclotron resonance (ICR) trap has been used for elemental or isotopic analysis. In
an ICR trap, the ions are trapped by a combination of RF and DC fields applied to
the walls of a box. The box is held inside a very strong magnetic field and the com-
bined RF and magnetic fields result in ions undergoing cyclotron motion whose
frequency depends on the ion’s mass-to-charge ratio. The mass spectrum is deter-
mined by measurement of these characteristic frequencies. First, the time domain
signal associated with ion motion is measured. The mass spectrum is generated by
inverting the time domain waveform to a frequency domain representation with a
Fourier Transform. The method is thus known as Fourier transform, ion cyclotron
resonance mass spectrometry or FT-ICR-MS. A key feature of FT-ICR-MS is
extremely high resolution.
The time-of-flight (TOF) mass spectrometer utilizes a pulsed ion beam for mass
analysis. In practice, a packet of ions is accelerated to the same energy and injected
into a field-free drift tube (Fig. 17.9). At constant energy, light ions move faster
than heavy ones, and mass separation is achieved by measuring the arrival times
of ions at the ion detector. Given a constant drift distance L (the length of the drift
tube from ion injection to ion detector), the time, t that an ion of m/z requires to
travel the distance L is given by:

t = L (m/z)/2eV (17.3)
378 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

where:
m = ion mass
e = electron charge
z = ion charge (in units of electron charge e)
V = ion accelerating voltage
L = drift tube length
v = ion velocity
Figure 17.9 shows a schematic diagram of a reflectron-type TOF used for laser
ablation mass spectrometry. Ions are generated in the source region, which consists
of a sample mount between two parallel plates (the plate on the right has a small
hole for transmitting ions into the drift tube on the right). The laser generates a pulse
of ions that are quickly accelerated into the larger drift region on the right. Lighter
ions move faster than heavier ions. The reflector (also known as a reflectron, see
discussion below) diverts the ions towards the ion detector at the bottom. A mass
spectrum is produced initially as a plot of ion arrival time versus ion intensity. The
laser ablation system is well-matched to the TOF since DC voltages can be used
for ion extraction and acceleration. In TOF systems that have a continuous source
of ions, pulsed electronics must be used to generate the rapid pulse of ions for the
TOF.
The TOF analyzer is capable of measuring a wide mass range of ions, but only
on a pulsed basis. In simple designs and operating modes, the spread in ion energies
must be small (<1%) and the ions must initially be tightly packed (within 1 mm or
less), otherwise the resolution of the TOF will be severely degraded. Methods are
applied for compensating for both initial ion energy spread and spatially extended
ion sources. The duty cycle (the fraction of time that ions are being generated
and analyzed) is usually short, hence the overall throughput of a TOF is low. For
longer duty cycles, specialized high speed pulse generators and high throughput
data acquisition systems are used (such systems were used for a brief time in a com-
mercial ICP-TOFMS instrument). Because data are acquired with high-speed dig-
itizers (typically 1 × 109 samples/sec) with deep memories (>1 × 106 samples),
digitizing resolution is usually limited to 8 bits. If signal averaging can be used,
the effective resolution can be improved, but the usual 8-bit data severely limits
abundance sensitivity of the overall instrument. The abundance sensitivity is thus
much less than other mass analyzers, typically in the 100–1000 ppm range (i.e.,
isotope ratios less than 10−4 are unresolvable).
Mass resolution is dependent upon drift tube length, ion energy, extractor design,
and timing resolution. High resolution is easily achieved, but TOF analyzers are
not often used for applications that require high mass resolution. A variant of the
TOF analyzer, called the reflectron, uses a gradient electric field to reflect the ions
back to an ion detector located near the ion source. The reflectron has energy-
focusing properties that correct for the energy spread (i.e., velocity distribution)
of the ions to result in higher mass resolution.
In a TOF analyzer, a mass spectrum is created by measuring the ion intensity
as a function of time of arrival after the ions are injected into the TOF analyzer.
 17. Mass Spectrometric Radionuclide Analyses 379

Ions may be pulsed either by pulsing the ion source itself (e.g., a pulsed laser)
while the TOF-MS ion extraction electrodes are held at constant DC potential, or
by pulsing the ion extraction electrodes. Pulsing the electrodes is difficult because
the pulse rise time must be short (typically <10 nanoseconds) and the potentials
are high (typically 1–5 KV). Depending on whether the ions to be analyzed are
formed in a continuous ion source (hot filament, continuous plasma) or in a pulsed
ion source (laser, pulsed glow discharge), the TOF analyzer may detect all of
the ions formed or just a portion of the ions formed. If all the ions formed are
contained in the extraction region of the TOF-MS, as in a typical laser source,
then they can be accelerated and delivered to the detector with high efficiency. The
TOF-MS analyzer is capable, in principle, of simultaneous detection of different
isotopes. In practice, isotope ratio precision is limited by the coupling between the
ion extraction process and the continuous ion source and by the poor digitizing
resolution and analog detection schemes (i.e., not ion counting).
TOF analyzers offer advantages of compactness (they can be made very small—
cm-length analyzers are possible), simple construction (the TOF is basically a
tube), and their wide mass range (they obtain the full mass spectrum, usually
1–300 amu for elemental and isotopic analysis) for every measurement. Draw-
backs include the difficulty of efficiently coupling to various ion sources, limited
mass resolution, relatively low precision (1–10 % are typical) and the need to
use expensive high precision and high speed electronics for pulse generation, ion
detection, and data acquisition.

17.2.4. Ion Detector

Mass-analyzed ions are detected by either Faraday cup (FC) or electron multiplier
(EM) detectors. Each is described below.
The FC collects the ion charge for direct current measurement of the ion beam.
The FC is essentially a metal bucket that collects the ion beam. Figure 17.10 shows
a schematic diagram of a Faraday Cup with associated repeller plates and slits.
The ion beam enters from the left and is captured in the Faraday cup on the right.
The biased plate to the left of the cup has a negative voltage (∼100 V or less) to
repel secondary electrons back into the cup.
Modern FC detectors are deep and narrow so that ions and secondary electrons
are captured and contained within the cup. The current can be measured directly
by a picoammeter; for high sensitivity measurements, the current is measured as
a voltage across a high resistance (109 ohms or greater). High sensitivity mea-
surements often utilize an integrating circuit in which the current (or voltage) is
integrated for a period of time (e.g., 10 s) and the integrated value is used to de-
termine the total ion charge collected during the integration interval. FC detectors
are limited to ion beam currents of 10−14 Ampere or above (>105 ions/s). The
FC detector is stable and is used for high precision isotope measurements when
large amounts of analyte are available. Some quadrupole instruments and sec-
ondary ion mass spectrometers employ FC detectors, especially for high current
measurements. FC detectors are also used for multiple mass detection and high
380 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.10. Faraday Cup ion detector. Figure A from Platzner (1997), pg. 34.

precision isotopic analyses on magnetic sector analyzers. The main disadvantage

is low sensitivity and slow response.
The EM converts ions to electrons and amplifies the resultant electron signal by
factors of 106 or more. In operation, ions strike a conversion dynode or surface
at the front of the EM to generate a few electrons. The secondary electrons are
accelerated to the next dynode and each in turn generates more electrons. The
process is repeated at each dynode to produce a large number of electrons after
multiple stages. Typical EM detectors have 15 to 20 dynodes or their equivalent,
depending on the design. Figure 17.11 shows a drawing of a commercial detector.
The cutout oval at the top shows the single incoming ion beam, which is converted
to an electron signal that is multiplied as it strikes each dynode down the right side
of the instrument.
Variants of the EM include the discrete dynode, channeltron (or continuous
dynode), the microchannel plate, and the Daly detector. The EM can detect ion
currents below 1 ion/s. EM detectors are often operated in the pulse counting
mode, in which the pulse produced by a single ion is amplified and detected as
an individual events. Pulse counting offers excellent signal-to-noise ratio, high
sensitivity, and insensitivity to change in EM gain.
The EM also can be operated in the current or analog mode (these terms are
synonymous) in which the ion current is amplified and measured. EM detectors
are available that operate simultaneously in pulse counting and analog modes. Part
way along the amplifying dynode chain, a portion of the signal is diverted to an
 17. Mass Spectrometric Radionuclide Analyses 381

FIGURE 17.11. Electron Multiplier ion detector. Figure from (ETP 2005).

operational amplifier (op amp) that ultimately produces an “analog” signal. The
remainder of the electron beam in the dynode chain continues to the end of the
chain to produce a pulse counting signal. When the ion beam becomes too intense
to operate the detector in pulse counting mode (the anode current must be kept
below a critical value to avoid damaging the detector), the pulse counting part of
the detector is switched off (by switching the bias to one of the dynodes) and the
analog signal is used. The response of these two sections of the detector can be
cross-calibrated to yield an overall dynamic range in excess of 108 .
A variant of the EM is the Daly detector (Daly, 1960) in which ions are accel-
erated by ∼30 KV (this is called post-acceleration because it occurs after mass
analysis) into a conversion dynode, which generates a significant number (∼10)
of secondary electrons. These electrons are accelerated into a scintillator and con-
verted into light, which is detected with a photomultiplier. The Daly detector offers
high gain, low noise, and excellent stability. Other variants of the post-acceleration
detector exist; a simple configuration uses a metal plate to convert the ions into
electrons for detection with an EM. Another variant of the EM is the microchannel
plate (Coplan et al., 1984; Odom et al., 1990; Wiza, 1979). Microchannel plate
EM detectors have excellent sensitivity but poor gain stability. When operated in
382 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

an analog mode (measuring amplified ion current), the gain of some detectors can
drift by several percent during an analysis. Ion counting eliminates most gain-drift
problems. EM detectors have limited lifetimes due to the accumulated damage at
the ion conversion dynode from the impact of detected ions.
Besides the ability to count ions, the EM detector responds to changes in the
ion beam intensity instantaneously (on a nanosecond timescale), whereas the high
feedback resistance associated with FC detectors results in a long time constant for
the measurement (typically 1 to 10 seconds). Fast scanning mass analyzers such
as QMF, ITMS, and TOF analyzers almost always use EM type detectors because
the FC response is too slow.

17.2.5. Vacuum
Mass spectrometers require various degrees of vacuum in different parts of the
instrument. A vacuum is produced by a range of vacuum pumps, including rotary
vane, diaphragm, scroll, turbo-molecular, diffusion, ion, and getter pumps. Most
ion sources operate at low pressure, some at high vacuum, while the ICP operates
at atmospheric pressure. The mass analyzer and ion detectors of a mass spectrom-
eter almost always operate in high or ultrahigh vacuum. Because gases flow and
behave differently in these various vacuum/pressure regimes, understanding vac-
uum system design and operation requires some familiarity with the properties of
gases in such systems. An excellent reference for vacuum systems is Dushman
and Lafferty (1962).
The various mass spectrometer performance requirements drive vacuum sys-
tem design. These include correct operation of the ion source (arcing can occur
in the high voltage components if the operating pressure is too high), and good
transmission of ions through the ion optics and mass analyzer (scattering by back-
ground gases reduces transmission). Rough vacuum is required on the foreline of
high-vacuum pumps such as turbo-molecular pumps and diffusion pumps and in
the interface between atmospheric pressure and the vacuum of an ICP/MS. Most
modern instruments use oil-sealed rotary vane pumps for these applications, al-
though oil-free or so-called “dry” pumps (also known as scroll and diaphragm
pumps) are used because of their inherent cleanliness. Dry pumps do not insert
oil vapor into the vacuum system, a high-cost component of a high-performance
mass spectrometer system.

17.2.6. Data Acquisition and Instrument Control

The “user friendliness” of major MS methods for radionuclide analysis varies
widely. The commercially available ICP/MS instrument is highly automated. Min-
imally trained staff can operate it for routine work when supported by trained main-
tenance staff. The TIMS instrument also can be obtained as a highly automated
system, but greater operator skill generally is required than for ICP/MS to achieve
a comparable level of data quality. Likewise, the SIMS instrument is complex,
both from an operational aspect, as well as for calibration and interpretation of
 17. Mass Spectrometric Radionuclide Analyses 383

results. The accelerator mass spectrometer (AMS) instrumentation is still highly

specialized and requires expert operators.
Instrument control typically is limited in commercial instruments to the extent
that non-routine uses of the instrument for development purposes may be difficult.
The data acquisition software is not always optimized for isotope ratio measure-
ments. These issues should be considered when purchasing an instrument for work
that includes both production and development.

17.3. Measurement Methodologies

17.3.1. Mass Bias
All mass spectrometers experience some level of mass bias in which different
isotopes of the same element are measured with different sensitivity. The main
causes are fractionation in the ion source and mass discrimination in the ion optics
and mass analyzer. Although the effect is usually small, corrections for mass bias
(sometimes called mass fractionation or mass discrimination) are needed to obtain
correct values of isotopic compositions from mass spectrometric data (Barshick
et al., 2000; De Laeter, 2001; Montaser, 1998; Nelms, 2005; Platzner, 1997; Tuniz
et al., 1998). In most cases, mass bias is corrected by running suitable standards
for which the isotopic composition is well known. Because of the prevalence of
ICP/MS, a detailed discussion is given below.
Mass bias in ICP/MS is 10 times more severe than in other types of mass
spectrometry, on the order of 1%/amu. The observed bias is sensitive to both
sample composition and instrument ion optical tuning conditions. The source of
this bias is dominated by space charge effects in the ion beam extracted from the
ICP (Douglas and Tanner, 1998).
When the plasma from the ICP passes into the first vacuum chamber, a free jet
expansion results. In such jets, there is a mass dependence relative to the centerline
flux, with intensity at a given distance downstream varying as m−1/2 (Miller, 1988).
This mass dependence applies to uninterrupted, skimmed free jet expansions. A
weaker mass dependence is expected in ICP/MS because the free jet effect only
dominates in the field free region near the skimmer tip. The first strong electric
field of the mass analyzer ion optics is experienced by the skimmed plasma flow as
it passes through the next stage of differential pumping. This electric field focuses
ions and repels electrons. The resulting net positively charged beam typically has
a charge density many orders of magnitude greater than what can be focused by
conventional optics (the ion number density greatly exceeds the limit imposed by
the Child-Langmuir law). The result is a rapid and severe “space charge explosion”
due to the mutual electrostatic repulsion of positive ions in the beam.
This effect has a much stronger mass dependence than the gas dynamic effect
and dominates the overall bias. Anything that affects this space charge field, af-
fects the observed bias. Mass bias in ICP/MS thus depends on sample preparation
(as more sample matrix is put into the ICP, the composition and intensity of the
384 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

ion beam changes) and instrument tuning (changing the potential of an optical
element affects the total ion flux and its spatial distribution). These effects have
been recently discussed by Praphairaksit & Houk (2000).
The ICP ion source has been utilized in combination with most types of mass
analyzer (RF quadrupole ion filters and traps, time-of-flight, and magnetic sector
field), although commercial instruments utilize either quadrupoles or magnetic
sectors as mass analyzers. The observed mass bias is comparable in these differ-
ent instruments (Encinar et al., 2001). The precision attainable does not depend
(strongly) on the mass analyzer employed per se, but rather on whether the instru-
ment is a multi-collector/detector instrument or not. The attainable precision with
single detector instruments is much more limited than what can be achieved with
multi-collectors (MC). Subtle corrections are, therefore, of much greater interest in
the latter instrument type. Mass bias corrections are derived by measuring the bias
with an isotopic abundance standard material and fitting measured bias to a func-
tion. In most quadrupole instruments, a linear fit suffices. In higher-accuracy work
(MC-ICP/MS), the functional forms are non-linear (power law or exponential).
The isotopes used for mass bias correction can be measured in the same solution
as the sample (for internal bias correction) or in a separate solution (for external
bias correction) when correction isotopes and the sample isotopes of interest may
overlap.
Extrapolation of the bias function from the masses of the reference isotopes
to the masses of the analytes of interest is typically kept as small as is practical.
Ideally, the analytes of interest are bracketed by the mass bias calibration isotopes,
i.e., the bias correction is interpolated. For example, a Tl isotopic standard is used
to correct unknown Pb ratios, but Tl would not be a good choice for correction
of Pu isotopes. If appropriate isotopic abundance reference materials are available
and appropriate care is used, MC-ICP/MS is among the most versatile (applies
to most of the Periodic Table) and accurate methods for determining unknown
isotopic composition.

17.3.2. Precision and Accuracy

Precision is a measure of how close repeat measurements of the same sample
are to one another. There are two main types of precision, internal and exter-
nal. When performing a mass spectrometric analysis, several measurements of
the elements of interest are made. How close these determinations are to each
other defines internal precision. Internal precision generally is a measure of the
stability of sample introduction to the mass spectrometer and the stability of the
ion source and mass analyzer/detector. Internal precision in mass spectrometry is
often counting-statistics limited (see Section 10.3). When the mass spectromet-
ric results are Poisson distributed, the best internal precision that is achievable is
N−1/2 where N is the total number of counts. Thus, if 100 counts are registered
for some isotope of interest, then this result will only be reproducible, on average,
to 10% relative standard deviation. To achieve 0.1% precision, at least 1,000,000
 17. Mass Spectrometric Radionuclide Analyses 385

counts are required; this level or better is the precision/counts regime of most
modern mass spectrometry. External precision is determined by running duplicate
samples. External precision is a measure of internal precision combined with how
reproducibly sampling and sample preparation are performed.
Accuracy of a method of analysis generally is determined by analyzing standards
with matrices the same or similar to the samples of interest. These standards have
known isotopic abundances and/or concentrations. The limit to this approach often
is the accuracy of the reference technique that was used to characterize the standard.

17.3.3. Isotope Dilution or Traced Analysis

Quantitation by mass spectrometry or radiometric counting requires reference to
known standard material. This reference can be “internal” to the sample, wherein
the reference material is added to the sample at an appropriate stage of processing,
or it can be “external” when the response of the analyte in the sample is compared to
the response measured for the reference material. A commonly employed method,
isotope dilution mass spectrometry, is to add a known amount of an isotopically
altered tracer (sometimes called a spike) to the sample.
Isotope ratios, as measured by mass spectrometry, along with the known amount
of the tracer, are used to determine the amount of the analyte in the sample.
Isotopically altered elements (sometimes consisting of nearly pure stable isotopes)
and radioactive isotopes are commercially available for use as tracers. Nearly every
element can be purchased as a solution whose atom concentration is known and
traceable to NIST. Standard materials are also available with isotopic composition
determined to a high degree of accuracy. Radioactive tracers, such as 233 U and
244
Pu, are available for tracing actinide elements that do not have stable isotopes.
Because many of the samples analyzed in radionuclide determinations already
have altered isotopic ratios, many times it is possible to use a natural isotope as
the tracer.
Isotope dilution mass spectrometry (cf. (Heumann, 1992; Yu et al., 2002)) has
two main requirements. The first is that the element being analyzed must have
more than one isotope. The second is to have a well-characterized and pure tracer
solution that has a significantly different isotopic composition from the element
under analysis. In practice, a known amount of the tracer is added to the sample,
which is then treated by any necessary chemical separations before being inserted
into the mass spectrometer. The tracer must be isotopically equilibrated with the
sample by forcing them into a common valence state, as discussed in Section 4.7.
For elements with multiple valence states (such as uranium or plutonium) this is a
crucial requirement. Failure to achieve isotopic equilibration will lead to erroneous
results. Sample quantitation by isotope dilution can be determined by use of the
following general equation:
(Rm Bt − At )
Ns = Nt (17.4)
(As − Rm Bs )
386 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

where:
Ns = the number of atoms of analyte in the sample
Nt = the number of atoms of the tracer added to the sample
Rm = the measured ratio of isotopes A to B for the traced sample
At , Bt = the atom fractions of isotopes A and B in the tracer
As , Bs = the atom fractions of isotopes A and B in the untraced sample
All relevant isotopes of the analyte element are monitored and the amounts of the
unknown isotopes are referenced to the known amount of isotope tracer. If the tracer
isotope is an isotope of interest in the sample, both traced and untraced sample
analyses are performed, the former analysis providing elemental concentrations
and the latter providing isotope ratios.
Equation (17.4) is a general form that is applicable to any element and tracer
combination. For quantitation of radionuclides in which isotopically pure tracers
are available that are not present in the sample at significant abundance, a simplified
version of Eq. (17.5) will suffice:
N s = N t × Rm (17.5)
where:
Ns = the number of atoms of analyte isotope in the sample
Nt = the number of atoms of the tracer isotope added to the sample
Rm = the measured ratio of analyte to tracer isotope A for the traced sample
For the case in which the tracer also contains the radionuclide of interest,
Ns = Nt × (Rm − Rt ) (17.6)
where:
Ns = the number of atoms of analyte isotope in the sample
Nt = the number of atoms of the tracer isotope added to the sample
Rm = the measured ratio of analyte to tracer isotope for the traced sample
Rt = the ratio of analyte to tracer isotope in the tracer
Equations (17.5) and (17.6) are commonly used to quantify radionuclides by
isotope dilution mass spectrometry.

17.4. Inductively Coupled Plasma Mass Spectrometer

The ICP/MS is an elemental and isotopic analysis method that was first developed
in the early 1980s. The ICP had been used only as a source for emission spec-
troscopy until it was adapted for producing ions for a mass analyzer (Douglas and
French, 1981; Houk et al., 1980; Houk et al., 1981; Houk and Thompson, 1982).
Since 1983, several manufacturers have sold ICP/MS instruments that incorporate
various mass analyzer systems, such as quadrupole mass filter, magnetic sector
field, time-of-flight, Paul ion trap, and ion detection systems such as the electron
 17. Mass Spectrometric Radionuclide Analyses 387

multiplier (EM), Faraday cup (FC), and Daly-type (post-acceleration and scin-
tillation). These systems are used alone or in combinations of “multi-collector”
instruments. In addition, researchers have coupled these and other mass analyzers,
such as Fourier transform-ion cyclotron resonance, and multiple analyzers as in
quadrupole-quadrupole configurations, to the ICP.
The power of this MS technique has driven the development of methods to
interface ICP/MS instruments with various sample introduction systems. Special-
ized sample introduction systems include ion chromatography (Seubert, 2001), gas
chromatography (Vonderheide et al., 2002), and capillary electrophoresis (Costa-
Fernandez et al., 2000). Other techniques are hydride generation (used to volatilize
selected species and obtain some matrix/elemental separation) (Reyes et al., 2003)
(Bings et al., 2002) laser ablation (Gonzalez et al., 2002; Heinrich et al., 2003;
Russo et al., 2002), and electrothermal vaporization (Richardson, 2001; Vanhaecke
and Moens, 1999).

17.4.1. Instrument Types

Commercially available instruments almost exclusively use quadrupole or mag-
netic sector analysis; an ICP-ion trap instrument is sold in Japan and attempts have
been made to market an ICP-TOF-MS. Quadrupole-based instruments generally
cost less than magnetic-sector-field-based instruments and provide adequate res-
olution, stability, speed, precision, and accuracy. High-resolution magnetic sector
instruments can resolve many of the spectral interferences encountered in ICP/MS,
but are more often purchased for their high sensitivity in low-resolution modes,
as well as their ability to perform multi-isotope collection. To address the prob-
lem of isobaric interferences, an alternative to high mass-resolving power is the
collision/reaction cell (CRC), which is inserted between the ion source and mass
analyzer to chemically remove interfering species, as discussed in Section 17.4.5.
The CRC-ICP/MS method opens new application areas for ICP/MS, such as bio-
logical studies involving phosphorous, sulfur, and other elements that were difficult
to resolve from interferences.

17.4.2. The ICP and the ICP-to-Vacuum Interface

Since the ICP was introduced by Fassel in the late 1960s, a great deal of research
has been performed to optimize the inductively coupled plasma as an emission
source (Fassel, 1971; Fassel and Kniseley, 1974; Montaser, 1992; Montaser, 1998).
Variations in torch design, gas flows, gas composition, and applied power were
just some of the variables that were studied. The end result by the early 1980s was
a source that remains essentially unchanged today. The ICP has flow velocities
through the torch such that a small amount of injected sample aerosol moves
through the plasma in ∼10 milliseconds.
At this plasma intensity and timescale, sample aerosol droplets below a certain
threshold (∼10 microns) are desolvated and the remaining solid particle is vapor-
ized and atomized. The resulting atoms are ionized with an efficiency that depends
388 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.12. Plasma transfer in the ICP/MS.

on factors such as the ionization potential of the atom and the temperature of the
plasma. This source has an ionization efficiency of around 95% with high excita-
tion temperatures for inducing optical emission for the majority of the elements in
the periodic table. Because of the high efficiency of ionization, researchers such
as Houk and Fassel (1980) developed the ICP as an ion source for mass spectrom-
etry. Early attempts were unsuccessful due to formation of a boundary layer that
obstructed the ion-sampling aperture. Progress in overcoming this complication,
among other key developments, led to the powerful technique in use today.
The key to the development of ICP/MS was a technique to transfer ions from the
plasma into the high vacuum of the MS. This was accomplished with a differen-
tially pumped atmospheric-pressure-to-vacuum interface shown in Fig. 17.12. An
aerosol is injected into the central argon flow, where aerosol particles are desolvated
and vaporized. The ions created in the plasma are drawn into the MS at the vacuum
interface. The plasma density and temperature permit isotropic flow through the
first orifice (typically ∼1 mm diameter aperture), and the plasma undergoes free
jet expansion (Scoles, 1988) in the first vacuum chamber.
This free jet structure is “skimmed” as it passes through a second aperture
into the first high vacuum stage. The interface contains a skimmer cone (made of
temperature-resistant alloy) with a small orifice through which the plasma (consist-
ing of positive ions, electrons, and neutral gas) is drawn into the vacuum interface.
High capacity vacuum pumps remove the gas, while ions are electrically injected
into the mass spectrometer using ion optical elements and accelerating voltages.
The skimmed beam is accelerated and focused by ion optical elements (typ-
ically, cylindrically symmetric lenses). The strong electric field in these lenses
(∼ 1 keV/cm) leads to rapid charge separation in the skimmed beam. That is, the
 17. Mass Spectrometric Radionuclide Analyses 389

electrons are rapidly repelled by this field and the positive ions are accelerated and
focused. The intensity of the beam is such that once the electrons are removed,
the self-repulsion of the positive ions due to space charge effects (the so-called
space charge explosion) leads to significant loss of ion beam intensity, an important
source of analyte loss. Loss of ion current in this atmospheric-pressure-to-vacuum
interface is the main contributor to low efficiency in ICP/MS.

17.4.3. Analytical Performance

The virtues of ICP/MS include efficient production of positive atomic ions; ro-
bustness of the plasma (tolerance of sample matrix material in the ion source);
sample introduction at atmospheric pressure; ability to handle solid, liquid or
gaseous sample forms; and speed of analysis. Sample preparation requires much
less chemical purification than for TIMS, AMS or radiation detection by beta- or
alpha-spectrometry.
Most elements are efficiently ionized at the high temperature of the plasma. For
most metals (ionization potential <8 eV), ionization is typically greater than 90%.
Many elements have an ionization efficiency >98%. The few elements not analyzed
by ICP/MS are H, He, C, N, O, F, Pm, Ne, Ar, Kr, and short-lived radionuclides.
These exceptions have high ionization potentials, severe spectral interference, or
are better measured by radiometric counting or other mass spectrometric methods.
While a number of gases have been used for the plasma support gas, argon is the
gas of choice in nearly all instruments in use today. Argon has properties that offer
the best compromise in cost, robustness, thermal conductivity, ionization poten-
tial, interferences, reactivity, solvent load handling, and ease of plasma operation.
Argon generally provides greater ionization for a wider range of elements due to
the higher ionization potential leading to more energetic electrons.
Molecular gases such as nitrogen, oxygen, and carbon dioxide offer advantages
over argon in higher thermal conductivity and better solvent load handling, but
they also tend to have higher levels of interferences, more reactivity, and require
higher flow and applied power to sustain a stable plasma. Helium has a higher
ionization potential than argon with more energetic electrons in the discharge, but
a helium plasma cannot tolerate high solvent loads.
Sensitivity is assessed by comparing signal level and stability of the background
with analyte response. In early ICP/MS instruments, response was on the order of
106 ion counts/s per part-per-million (ug/mL) of analyte in a solution introduced
at a flow rate of 1 mL/min. This is an analyte introduction rate of 1 ug/min, or 1014
atoms/s for the analyte 100 Mo. The overall efficiency is thus 10−8 given 106 counts/s
vs. 1014 atoms/s from the sample. Research and development has led to efficien-
cies that have steadily increased over the years. State-of-the-art instruments and
high efficiency sample introduction systems can now achieve 10−3 (0.1 percent)
overall efficiency in terms of counts detected per atom in the sample. Champion
efficiencies as large as 0.5 percent have been reported (Rehkamper et al., 2001).
For comparison, TIMS champion efficiencies are ∼1 and ∼5 percent, for uranium
and plutonium, respectively.
390 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

TABLE 17.4. ICP/MS precision for various instruments and sample amounts
Isotope ratio Method∗ Precision Sample size Efficiency Reference
∗∗235 U/238 U ∼1 ICP/QMS 0.028% 255 pg Not Avail Platzner
∗∗235 U/238 U ∼1 DF-ICP/MS 0.026% 255 pg Not Avail Becker
238 U/235 U = 137.9 ICP/QMS 0.07% 663 pg 0.13% PNNL
238 U/235 U = 31.8 ICP/QMS 1.4% 307 fg 0.15% PNNL
235 U/234 U = 160.5 ICP/QMS 3.8% 9.42 fg 0.15% PNNL
236 U/234 U = 1.05 ICP/QMS 7.4% 324 ag 0.16% PNNL
∗ QMS: QMF mass analyzer, DF-ICP/MS: double-focusing sector-field mass analyzer
∗∗ From (Becker and Dietze, 2000).

The biggest performance difference in instruments is associated with the number

of detectors, not the type of ion source. Single-detector instruments yield precision
0.01% to 0.5% (relative standard deviation or RSD) for large samples. For small
samples, say where fewer than 106 counts are obtained for the minor isotope(s)
of interest, precision is limited by counting statistics. With larger samples and
multiple detector instruments, an RSD somewhat below 100 ppm can be achieved.
This “zeroth order” picture is true for both TIMS and ICP/MS, but TIMS is the
more precise and accurate method when detailed comparisons are made. Bias is
significantly different for TIMS versus ICP/MS and remains a challenge for multi-
collector ICP/MS (Becker, 2002). Obtaining accuracy that is commensurate with
precision at the sub-100 ppm range is difficult regardless of the method used, but
ICP/MS has more fundamental limits at present. Internal precision is limited for
low-level samples by counting statistics, as shown in Table 17.4.
Detection limits are most often reported as concentration detection limits be-
cause most ICP/MS applications are not limited by sample amount. Absolute mass
detection limits are more relevant when the amount of analyte available is highly
limited. The lowest mass detection limit for ICP/MS is ∼105 atoms, or ∼0.1
femtogram for actinide elements.

17.4.4. Isobaric Interference

Isobaric interference is the overlap of ion peaks with nominally the same m/z. In-
terferences occur between isotopes of different elements (e.g., 54 Cr/54 Fe, 90 Sr/90 Zr,
187
Re/187 Os), and between isotopes and molecular ions (56 Fe/ArO+ , 80 Se/Ar+ 2 ,
239
Pu/238 UH+ ). The resulting signal at the m/z of interest represents a generally
unknown combination of analyte signal and interferent signal. Molecular ions usu-
ally arise in one of two ways: they are formed in the ion source or by reaction of
the ion beam with background gases in the instrument’s vacuum chamber.
The nature of an isobaric interference is illustrated in Fig. 17.13, where element
B is the analyte of interest for which the isotope ratio indicated by the two spectral
lines is to be measured. A lighter element, A, has isotopes whose diatomic oxides
AO+ have three masses indicated as “molecular ions” in the figure, two of which
occur at the element B masses of interest. The spectrum observed at the element
B masses of interest is thus a composite of signals from the element B atomic
 17. Mass Spectrometric Radionuclide Analyses 391

FIGURE 17.13. ICP/MS spectral analysis with isobaric interference.

ions and the element A molecular ions. In this case, interference by A in B can
be corrected because an AO+ ion exists that does not overlap any of the B isotope
masses. From the intensity of the lightest isotope and the abundance of each of the
three isotopes of A, one can calculate the contribution of the other AO+ isotopes
to the signals observed at the B isotope masses.
Each of the major types of MS instruments for radionuclide measurements has
unique mechanisms by which isobaric interference are formed within them. In an
ICP/MS, the most common interference is from atomic isobars and the diatomic
molecular oxide, hydride, nitride, and argide ions. In the typical argon support gas
for the ICP, argide-based interferences abound, from 40 Ar interfering with 40 Ca
to 195 Pt40 Ar interfering with 235 U. The usual aqueous aerosol introduced into the
ICP provides oxygen and hydrogen, which form oxides and hydrides. Examples
of these interfering ions are listed in Table 17.5 with the analyte(s) of interest and
the mass resolving power needed to separate them.
Considerable effort has been expended to reduce interference without losing
analyte signal. These efforts include chemical removal of interference-causing
atoms prior to mass analysis, with sample introduction by hydride formation at
elevated temperature, and with selective removal of water vapor from the sample
aerosol (Montaser, 1998). Reduction of molecular ion interferences involving high
ionization potential (IP) atoms, such as argides, has been achieved by reducing the
plasma temperature (via lower RF power). Reduction of plasma temperature re-
duces the intensity of high-IP species much more than that of lower-IP species. This
“cold” plasma has been shown to reduce Ar+ and ArO+ faster than the reduction
in intensity of the species with which they interfere (Ca and Fe, respectively).

17.4.5. Collision/Reaction Cells

Introduction of a collision/reaction cell (CRC) into the ICP/MS is a recent tech-
nique for interference reduction (Koppenaal et al., 2004). Interfering ions are
eliminated by their gas phase reaction in a gas-filled cell inserted between the
392 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

TABLE 17.5. Isobaric Interference in ICP/MS and mass resolution

required to separate them
Interferent Isobaric analyte M/M
40 Ar 40 Ca 190,000
40 Ar14 N 54 Fe, 54 Cr 2,090; 2,030
40 Ar16 O 56 Fe 2,500
35 Cl16 O (in Cl matrix) 51 V 2,600
40 Ar35 Cl 75 As 8,000
Br 79 Se (τ1/2 = 650,000 a) 487,000
Ar2 80 Se 9,700
Kr (impurity in Ar) 84 Sr 43,600
Zr 90 Sr (τ1/2 = 28.78 a) 29,600
90 ZrO 106 Pd 27,400
59 Co40 Ar 99 Tc (τ1/2 = 213,000 a) 9,270
Xe (impurity in Ar) 129 I(τ1/2 = 1.57 × 107 a) 625,000
1/2 = 2.3 × 10 a)
Ba 135 Cs (τ 6 613,000
1/2 = 7.04 × 10 a)
195 Pt40 Ar 235 U (τ 8 2,010
1/2 = 24,100 a)
238 UH 239 Pu (τ 37,500

ion source and the mass analyzer. The reacting ions undergo a change in mass-
to-charge ratio in the cell. If the interfering ions react at a much greater rate
than the ions of interest, the analyte of interest is detected with reduced back-
ground. Typically, interferences can be reduced by three to nine orders of magni-
tude; there is some accompanying loss of analyte, from a few percent to 10-fold
loss.
Collision/reaction cells used in ICP/MS consist of an enclosure to contain the
gas, apertures at each end to allow ions in and out of the cell, and a set of electrodes
(e.g., octopole, hexapole, or quadrupole rod sets) to guide ions through the cell.
The main considerations in the use of the method are the reagent gas (whether to
use a “buffer” gas), gas number density, ion kinetic energy, cell length, multipole
design, use of auxiliary fields (in addition to the ion guiding field), and reaction
time. Other considerations include reagent gas purity, side reactions, and chemical
resistance of the sample ions (including desired analyte, interferences, and matrix
ions) to the reagent gases. The operating principles and design of RF multipole
ion guides and gas collision cells have been described in detail (Gerlich, 1992).
The dramatically different reactivities of gas phase ions provide an opportunity to
reduce or eliminate many isobaric interferences. Identification of suitable gaseous
reagents is aided by the advances in the understanding of gas phase ion-molecule
chemistry achieved in recent years (Armentrout, 2004). Hydrogen gas is the most
selective reagent gas currently in use. Other reagent gases, including NH3 , H2 O,
CH4 , and O2 , have been found to be much less selective, and usually require that
selective ion storage/reaction methods be employed to avoid generation of new
interferences.
Collision/reaction cells are typically operated at the lowest possible kinetic en-
ergy for several reasons. First, the benefits of collisional focusing are realized at low
 17. Mass Spectrometric Radionuclide Analyses 393

kinetic energy. Second, ions at lower energy remain longer in the cell and have more
opportunity to react. Last, lower energy enables control over endothermic reactions.
Most manufacturers of ICP/MS instruments now offer a model that incorporates
a collision cell. Interference reduction by sample purification, use of a collision cell
(also called chemical resolution), and mass separation—are said to be orthogonal,
i.e., gains in each method are independent and multiplicative with gains in the
others. The ultimate instrument in this regard—a collision cell equipped, high
resolution, magnetic sector MS—has not yet been built, but would enable the
nearest approach yet to detection without interference of any isotope at ultra-trace
levels.

17.5. Thermal Ionization Mass Spectrometry

Thermal ionization mass spectrometry, or TIMS, is sometimes called thermal
emission mass spectrometry or surface ionization mass spectrometry. It is one of
the mainstays of isotopic analysis. Historically, the emission of ions from a heated
surface was observed nearly 100 years ago, and was used as a mass spectrometer
ionization method in 1918 (Smith, 2000). Modern TIMS grew out of research in
the 1920s on thermionic emission from surfaces (Langmuir and Kingdon, 1925)
and has continued to be the method of choice for high precision isotopic analysis
for actinides and lanthanides, as well as transition elements. TIMS is popular for
isotopic analysis because the ionization method produces a relatively low energy
spread (∼0.5 eV), thus high performance analyses can be performed with single
stage magnetic sector instruments. By comparison, other ionization techniques
(e.g., electron impact) produce ions with higher energy spreads, as well as multiply-
charged ions, which often requires additional focusing to obtain well-separated and
well-behaved ion beams.
Langmuir and Kingdon (1925) developed the theoretical basis for thermal ion-
ization and showed that positive ion emission was well represented by:
N+ W −I
= Ae kT (17.7)
N0
where:
N+ = emission rate of positive ions
N0 = emission rate of neutral species
W = work function of the filament metal (eV)
I = ionization potential of element or molecule (eV)
k = Boltzmann’s constant (eV/K)
T = absolute temperature (K)
A = constant (see text)
This equation is now called the Saha-Langmuir equation. The constant, A, de-
pends on the quantum energy levels of the element or molecule and of the ioniz-
ing filament material. Efficient production of positive ions occurs from filament
394 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

material made of high work function metals and for analyte species with low ion-
ization potentials. Most metals have work functions in the range of 4 to 5 eV,
which effectively limits TIMS to elements (or molecular species) with ionization
potentials less than 8 eV. In practice, TIMS is generally used for elements with
ionization potentials of 6 eV or less. Rhenium is the filament material of choice
because of its high work function (4.98 eV) and high melting point (3180 C).
Commercial rhenium is available in ultra-high purity (99.999% or better), which
minimizes unwanted ion emission. Other metals commonly used for filaments in-
clude tantalum and tungsten. Although platinum has the highest work function
(5.13 eV), its relatively low melting point (1772 C) limits its utility for TIMS use.
Although primarily associated with positive ions, negative TIMS is also possible.
The governing equation (17.8) is similar:

N− E A−W
= Ae kT (17.8)
N0

where:

N− = emission rate of negative ions

N0 = emission rate of neutral species
W = work function of the filament metal (eV)
EA = electron affinity of element or molecule (eV)
k= Boltzmann’s constant (eV/K)
T = absolute temperature (K)
A= constant

Unlike the positive ion case, negative TIMS is most efficient with low work
function surfaces. Such surfaces are created by coating the ionizing filament with a
low work-function material such as LaB6 or Ba(NO3 )2 . Elements with high electron
affinities (hence easily ionized) include the halogens, selenium, and tellurium.
Oxides of some metals (e.g., Mo, Os, Re) also have high electron affinities and
also have high volatilities, especially when compared to their reduced, metallic
forms. Since the mid 1980s, negative TIMS has become the method of choice for
high precision and high sensitivity isotopic analysis of refractory Group VIb, VIIb
and VIII elements. For instance, Heumann (1995) developed and demonstrated a
viable method for analyzing the geochronometer Re-Os using negative TIMS. The
Re-Os system has also recently been analyzed by MC-ICPMS (Malinovsky et al.,
2002; Yin et al., 2001).
Whether the ionization is positive or negative, TIMS requires careful sample
preparation, often involving considerable chemical processing to separate and pu-
rify the element of interest. TIMS finds applications in geoscience, environmental
analysis, cosmochemistry, biosciences, medicine, material science, and physics.
Samples generally include soil, minerals, meteorites, and biological tissue. More
information on the specifics of the TIMS technique and its applications is in the
monograph by De Laeter (2001).
 17. Mass Spectrometric Radionuclide Analyses 395

17.5.1. Instrumentation
At the heart of the TIMS ion source are one or more hot filaments that serve
to vaporize and ionize atoms or molecules of interest. Once generated, the ions
are accelerated, focused, and directed into the mass analyzer for measurement.
The classic TIMS instrument consists of an ion source, a single magnetic sector
mass separator, and an ion detector. Such an instrument is capable of measuring
isotope ratios as small as 1 × 10−6 , sufficient for the isotopic analysis of most
elements. For radionuclide analysis, smaller isotope ratios are often encountered.
Specialized mass spectrometers include multiple magnetic and electric sectors
and sector instruments with retarding quadrupole lenses (Smith, 2000) to measure
down to the 10−9 range.
TIMS instruments tend to be table-top-sized, with total path-lengths from ion
source to detector in the 1 to 2 m range. Popular sector geometries include 60◦ and
90◦ deflection, which is the nominal deflection produced by the magnetic field.
The deflection radius is normally in the 10 to 30 cm range. Typical operating pa-
rameters for actinide analysis are 10 kV acceleration voltage and ∼0.5 to 1 Tesla
magnetic field to produce ion beams that are separated by mm-scale distances (1
amu mass difference at 238 amu ion beam mass). Ions are detected by either a
Faraday cup or an electron multiplier. Because of the small mass separation, de-
tector size can be a critical issue, especially for simultaneous detection of multiple
ion masses.
Historically, most TIMS instruments were equipped with a single ion detector.
In operation, the ion mass region of interest is either scanned or peak-stepped, by
changing either the ion accelerating voltage or the magnetic field. Peak-stepping
requires very stable electronics and magnet control (if an electromagnet is used) to
select accurately and with good repeatability a suite of ion masses. The ion beam
must also be of high quality and reproducibility.
A common requirement for TIMS instruments is to have flat-topped ion peaks.
Flat-top peaks are produced when the ion beam is focused to a fine line and
the analyzer slit (usually the slit in front of the detector) is wider than the ion
beam; therefore, the entire ion beam is captured by the ion detector. Flat-top
peaks are desired for high precision isotopic analysis because the measured ion
beam signal is insensitive to slight instabilities in the instrument magnetic field,
which causes the beam to shift position. Flat-top peaks are achieved at the cost
of reduced mass resolution, but such reductions are usually insignificant com-
pared to the gain in isotopic measurement precision. Figure 17.14 shows the
TIMS peak shape, acquired by high-resolution mass scans for several neodymium
isotopes. Note that several lines are slightly shifted from each other. The data
were acquired with an Isoprobe-T instrument and multi-collector FC detection
system.
Modern TIMS instruments are equipped with multiple ion detectors (see Fig.
17.15). In the 1980s, Faraday cup arrays became commercially available, and
these provided significant improvements in isotopic precision and sample utiliza-
tion. In the 1990s, arrays of pulse counting ion detectors with very compact EM
396 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.14. TIMS peak shape. Figure from GV Instruments.

FIGURE 17.15. A schematic diagram of a multi-collector mass spectrometer. Figure from

Smith (2000), pg. 14.
 17. Mass Spectrometric Radionuclide Analyses 397

or microchannel plate technology (De Laeter, 1996; De Laeter, 2001) were devel-
oped for high sensitivity and high precision isotopic analysis. These are becoming
commercially available.

17.5.2. Sample Preparation

Isotopic analysis of an element by TIMS is preceded by extensive chemical prepa-
ration to separate and purify the element to be assayed. Most TIMS analyses require
picogram to microgram quantities of the separated and purified element, although
modern instruments and sample preparation techniques have lowered detection
limits to the sub-femtogram-range (Dai et al., 2001; Stoffels et al., 1994). Detailed
description of sample processing techniques is beyond the scope of this discus-
sion, but most sample preparation utilizes dissolution, gross separation, detailed
purification, and preparation of the analyte into a form suitable for loading onto
a mass spectrometer filament. The separation and purification steps are usually
performed by procedures similar or even identical to those used in conventional
radiochemical techniques (ion exchange, redox extraction, precipitation, etc.). The
final purification and preparation is often an art because the chemical form on the
filament is critical to producing efficient ionization and low background (Beasley
et al., 1998b; Delmore et al., 1995; Kelley and Robertson, 1985; Perrin et al., 1985;
Rokop et al., 1982; Smith and Carter, 1981; Walker et al., 1981).

17.5.3. Performance and Application

Many applications focus on the isotopic analysis of elements in environmental
samples (water, rocks, soil, minerals), meteorites, biological tissues, and nuclear
materials (such as nuclear fuels). A relatively new application has been the use
of TIMS in nuclear safeguards and arms control treaty verification (IAEA, 2003).
Of most interest to this discussion is the high quality isotopic analysis of ac-
tinide elements, especially U and Pu. Typical performance parameters are given in
Table 17.6 for isotopic analysis of actinide elements and additional examples are
given in Section 17.8.

TABLE 17.6. Performance Parameters for uranium and plutonium analysis by TIMS
Element Measured isotope ratios Detection limit Comment
U 234 U/238 U ∼1pg Detection limit is usually limited by
235 U/238 U background U
236 U/238 U

Pu 238 Pu/239 Pu <1 fg Detection usually limited by sample

240 Pu/239 Pu quantity
241 Pu/239 Pu
242 Pu/239 Pu
398 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

17.6. Accelerator Mass Spectrometry

Accelerator mass spectrometry (AMS) was developed for analyzing 14 C in envi-
ronmental and archeological specimens. It is now used to measure isotopes span-
ning the periodic table from tritium to plutonium. AMS represents the marriage
of high-energy physics with radiochemistry and finds applications in geoscience,
environmental analysis, archeology, cosmochemistry, biosciences, medicine, ma-
terial science, and physics (Fifield, 1999; Hellborg et al., 2003; Herzog, 1994;
Hotchkis et al., 2000; Kutschera, 1994; Liu et al., 1994; Rucklidge, 1995; Skippe-
rud and Oughton, 2004; Tuniz et al., 1998).
AMS consists of a high-energy particle accelerator and an ion detector. Typical
atom ratios for AMS isotope measurements are in the range of 10−10 to 10−15 . These
ratios fall well below the 1 to 10−9 range measured with other mass spectrometers,
except for laser-based systems with resonance ionization techniques, which can
match the range of the AMS.
The accelerator began as a source of energetic ions to bombard targets. Re-
searchers (Tuniz et al., 1998) recognized that the techniques used to create, select,
and direct ion beams in the MeV energy range could be applied to measuring ex-
tremely small isotope ratios with high sensitivity and low background. The tech-
nique provides extremely high selectivity and efficient rejection of interferences,
with quantitative elimination of interference by molecular species. By comparison,
all other mass spectrometric systems described in this chapter suffer from varying
degrees of isobaric interference by ionic molecules with nearly identical mass.
High mass resolution can be applied with other MS systems to resolve molecu-
lar and isotopic ion species with small mass differences, but at the cost of lower
sensitivity and increased instrument complexity.

17.6.1. Instrumentation
The AMS consists of an ion source, an input mass or momentum selector (usually
a magnetic sector), a high–energy accelerator, a “stripper” (a low-pressure gas
or thin foil placed in the middle of the high voltage region), post accelerator
ion optics and mass selectors, and one or more detector systems. Figure 17.16
shows a generalized diagram. The high-energy accelerator is usually a tandem
accelerator, which consists of two back-to-back high-energy accelerators with a
stripper between the accelerators. The voltage increases from zero at one end
to the middle and decreases to zero at the other end. Low-energy negative ions
injected into one end of the accelerator emerge as high-energy positive ions at
the other end. This arrangement allows the ion detector region to be operated at
or near ground potential to simplify construction and electrical operation of the
instrument.
Ions typically are produced by a secondary ion or sputter source (the usual choice
for the AMS), often with a beam of Cs ions to sputter the sample to produce negative
ions. These ions are extracted, focused, and accelerated to energies of 10–100 keV.
 17. Mass Spectrometric Radionuclide Analyses 399

FIGURE 17.16. Schematic diagram of an AMS instrument. Figure from De Laeter (2001),
pg. 62.

The input mass selector directs a specific ion mass into the tandem accelerator,
which then accelerates the negative ions to MeV energies. A stripper in the central
region of the accelerator dissociates molecular ions and removes multiple electrons
from the ions to convert ingoing negative ions (elemental or molecular) into out-
going multi–charged positive ions. These positive ions are further accelerated by
the same potential and emerge from the back-end of the accelerator with energies
that can reach above 30 MeV, depending on the accelerator base voltage (or poten-
tial) and the degree of ionization. Charge states of +10 or more can be produced,
especially for heavy ions such as the actinides. After acceleration, additional ion
optics and switching magnets are used to select the ion mass and charge state and
to direct these ions into detectors for measurement.
By its nature, an AMS instrument is large. Path-lengths from ion source to
detector can be tens of meters. The high ion energy results in large differences in
ion path angles and trajectories. After selection by mass, the ions are separated
at cm-scale distances, in contrast to mm-scale separations produced in sector-
based mass spectrometers. A Faraday cup detector is a common choice for the
major isotope mass, whereas nuclear particle detectors (which include electron
multipliers) count individual ions of the rare isotope. The high ion energies allow
the use of detectors that can measure ion time-of-flight, ion energy, ion energy loss
rate, and ion charge. Measurement of the ion charge/mass ratio provides additional
selectivity for identifying the rare isotope.
400 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

In operation, the major and rare isotopes are measured sequentially by peak
switching from one isotope to the other, as with other mass spectrometers, to de-
termine atom ratios. The major isotope may produce a current in the microampere
range, whereas the rare isotope produces a signal as low as a few counts per minute.
The raw isotope ratio consists of counted pulses divided by integrated current. Cal-
ibration by measurement standards with known isotope ratios provides the factor
to convert the raw ratio into an atom ratio.
Unique features of the AMS include ion energy in the MeV range compared
to keV range or lower for other mass spectrometers, conversion of negative ions
into positive ions, and generation of multi–charged ions. The rare and major iso-
topes may be measured in different charge states, depending on the element under
analysis, the need to eliminate isobaric interference, and the configuration of the
instrument. On the other hand, the high energy poses safety and health consider-
ations. The ions in an AMS have sufficient energy to produce radionuclides and
radiation by nuclear reactions. Ions that strike various system parts (e.g., slits,
deflection plates, the inner walls of the vacuum system) can produce neutrons and
x-rays. High voltages are associated with the ion source, deflection plates, and
other parts of the instrument. These features require strict attention to radiological
and electrical safety protocols and protection measures.

17.6.2. Sample Preparation

The AMS is used to measure rare isotopes in samples that include air, water, ice,
soil, minerals, meteorites, biological tissue, and archeological artifacts. Sample
sizes generally are in the mg range to produce adequate rare isotope count rates
for analysis.
Samples usually are processed before analysis to extract and purify the target
element and its corresponding rare isotope. Some elements require the addition of
a carrier for processing and yield determination, such as iodine for 129 I analysis
in seawater, but care must be taken to measure the rare isotope background in the
carrier. By comparison, carrier addition is avoided in the analysis of 14 C because
samples have sufficient carbon for analysis and the goal is to measure the 14 C/12 C
ratio in the sample. The purified element is converted into a solid form suitable
for generating negative ions by cesium sputtering. The element of the rare isotope
under analysis can be analyzed in molecular form or in reduced elemental form,
depending on the ionization characteristics of the element. For instance, carbon
is analyzed as graphite whereas iodine is analyzed as silver iodide. The resultant
solid is pressed into a metal holder, commonly aluminum or copper (Tuniz et al.,
1998) and placed in the AMS ion source for analysis.

17.6.3. Performance and Application

The rare isotopes 14 C, 10 Be, 26 Al, 36 Cl, 41 Ca and 129 I provide the bulk of the
work performed by AMS laboratories. Recently, AMS has been applied to heavy
element analysis, especially for U and Pu. Other isotopes analyzed by AMS include
 17. Mass Spectrometric Radionuclide Analyses 401

TABLE 17.7. Experimental Parameters for AMS, after Tuniz et al. (1998)
Measured Detection
Isotope Sample form isotope ratio limit Carrier Comments
14 C Graphite 14 C/12 C 5 × 10−16 None —
10 Be BeO 10 Be/9 Be 3 × 10−15 Sometimes —
26 Al Al2 O3 26 Al/27 Al 3 × 10−15 None —
36 Cl AgCl 36 Cl/35 Cl, 1 × 10−15 Sometimes, Both ratios measured
36 Cl/37 Cl Chloride
41 Ca CaH2 or CaF2 41 Ca/40 Ca 6 × 10−16 None —
129 I AgI 129 I/127 I 3 × 10−14 KI —
U U + Nb metal 236 U/238 U 1 × 10−12 Nb metal (Berkovits, 2000)
Pu Pu + Fe–oxide 240 Pu/239 Pu, <1 fg Fe–oxide Detection limit in
241 Pu/239 Pu, atoms; multiple
242 Pu/239 Pu ratios analyzed

tritium, 59 Ni, 90 Sr, and 205 Pb. Detection limits for AMS are typically given as
limiting isotope ratios, which is the measured quantity in AMS. The overall ion
transmission of a typical AMS is such that the absolute atom detection efficiencies
are lower than either TIMS or ICPMS—on the order of 1 count per 100,000 atoms
injected into the instrument. Nevertheless, the extremely low background of the
AMS technique permits the detection and quantification at the sub-femtogram-
level, which is comparable to ICP/MS and TIMS analyses. Some experimental
parameters are given in Table 17.7 for the most commonly analyzed isotopes.
For some radionuclides, only the direct measurement is needed, based on cali-
bration with a radionuclide standard and reference to the measured sample mass.
For other radionuclides, the isotope ratio to its stable element is needed. An exam-
ple of a more complex situation is measurement of the 14 C/12 C isotope ratio of an
environmental or archeological sample in comparison to the modern atmospheric
CO2 value. This value must be adjusted for anthropogenic 14 C produced by atmo-
spheric testing of nuclear weapons and by other nuclear operations, and also for
changes in atmospheric CO2 with cosmic-ray flux fluctuations over time. For an
element such as Pu, which has no stable isotope, the total quantity is measured by
isotope dilution mass spectrometry in which the sample is traced (or spiked) with
242
Pu or 244 Pu (see Section 17.3.3).

17.7. Other Mass Spectrometers

17.7.1. Secondary Ionization Mass Spectrometer
Secondary ionization mass spectrometry (SIMS) is yet another method to analyze
solid samples directly. SIMS uses a focused beam of ions (called the primary
ion beam) to sputter atoms from the surface of a sample. A small fraction of the
sputtered atoms are ionized by the sputtering action, hence the term secondary
ionization. The secondary ions are extracted, accelerated, and analyzed by a mass
analyzer. Two main configurations exist: conventional SIMS, which uses electric
402 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

FIGURE 17.17. The Cameca IMS-6F secondary ion mass spectrometer. Figure from
De Laeter (2001), pg. 49.

and magnetic sectors to perform the mass separation, and TOF-SIMS, which uses
a time-of-flight mass analyzer. Both variants are available as commercial instru-
ments, as in the conventional SIMS instrument shown in Fig. 17.17.
The primary ion beam is formed on the left and accelerated and focused onto a
sample held at high voltage. Secondary ions are extracted and accelerated into the
analyzing mass spectrometer, which consists of an electrostatic sector (for energy
focusing) and a magnetic sector (for mass separation). Among the detectors used
to analyze the ions is an imaging detector. The ion optics on the Cameca instrument
are specifically designed to permit imaging of the sample with ions.
Production of secondary ions depends on many factors, but the composition of
the primary ion beam is vital to the efficient production of ions. Cesium or oxygen
ions (positive or negative, depending upon whether negative or positive secondary
ions are desired) are used (Cristy, 2000; De Laeter, 2001). Other primary ions
include noble gases, especially Ar, and more exotic species such as rhenium oxide
(Groenewold et al., 1997). The primary ion beam usually is mass selected with a
magnetic sector or quadrupole mass analyzer, and focused to a micrometer-sized
spot on the sample under analysis. Normally, the beam can be raster-scanned across
the sample with magnetic and electrostatic deflection in a manner similar to the
electron optics used in an electron microscope.
An advantage of SIMS is that the method provides micro-scale analysis of
the sample. The primary ion beam can be set to analyze a small area or raster-
scan across the sample. Some instruments are designed to produce ion images
analogous to images obtained in an electron microscope. The imaging capability
allows for high-resolution isotopic (and elemental) analysis of the sample, with
spatial resolution of 1 μm or better that depends on the instrument. The sputtering
 17. Mass Spectrometric Radionuclide Analyses 403

action of the primary ion beam can be controlled to permit calculation of the
depth of the material removed during the analysis. Depth resolution to a few
nanometers is possible. Because of these micro-scale and imaging capabilities, a
SIMS instrument is sometimes referred to as an ion microprobe or ion microscope.
Production of secondary ions is non-selective in that all elements can be ionized
by this method. Thus, the entire periodic table is accessible using SIMS. The
relative and absolute efficiencies for the production of ions for a given element are
highly dependent upon the composition of the sample, as well as the configuration
and operating parameters of the instrument. Generally, isotopic compositions are
easily measured, whereas elemental compositions, which are derived from the
analysis of isotopes of a given element, are more difficult to determine. Good
quantitative analysis requires the use of calibration standards that closely resemble
the sample under analysis.
Preparation of samples for analysis resembles that used to prepare samples for
electron microscopy. In general, preparation consists of polishing small specimens
and mounting them on a planchet. A common planchet consists of a cylinder
of vitreous carbon with polished ends. The planchet must provide a conductive
pathway to drain the charge imparted by the impact of the primary ion beam.
Without some means to dissipate the charge, the sample will become charged,
which can deflect the primary ion beam and can interfere with the extraction of
secondary ions.
Operation of a SIMS instrument resembles both that of an isotope ratio mass
spectrometer and an electron microscope. Most SIMS instruments include an op-
tical microscope so that the sample can be directly viewed during analysis, which
allows for accurate positioning of the area of interest on the sample. Data can be
in the standard mode used for other types of mass spectrometers in which ions are
produced and the mass spectrum is analyzed by scanning or peak-hopping. This
mode is sometimes called the microprobe mode in SIMS. Another application for
SIMS is the acquisition of ion-images. This mode is called the microscope mode
because the SIMS is operated as an ion microscope.

17.7.2. Laser Ablation Mass Spectrometer

Laser ablation (LA) is a powerful tool for mass spectrometric sampling of radionu-
clides from solids. When sufficiently intense laser light strikes a solid, material is
removed from the surface. This material can be in the form of atoms, molecules,
ions, electrons, or small particles. Several approaches can be used to analyze the
surface for these species. The light that usually also is emitted can be resolved
by wavelength to identify atomic or molecular species associated with the sample
surface. Depending upon the wavelength of the laser, the atoms and ions in the
laser plume can be excited, and the resulting emission can be detected.
In laser ionization or laser ablation mass spectrometry, the ions formed by the
interaction of the laser with solid are directly sampled into a mass spectrometer.
Another option is that the material removed from the surface can be swept into
an inductively coupled plasma (ICP) for atomic analysis either by light emission
404 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

from the ICP or by analyzing the ions produced in the ICP. This latter technique is
referred to as laser ablation ICP/MS. Numerous mass spectrometer analyzer types
have been used for LA-ICPMS including time-of-flight (TOF), quadrupole, sector
field, and ion trap. The analyzer type is determined by the application.
One of the problems that has been noted in laser ablation is fractionation between
the sample surface and the material removed. For any given sample type, certain
elements will be preferentially removed, giving these elements a higher relative
abundance in the material analyzed by the mass spectrometer. If fractionated, this
material is not representative of the sample, and biased data results. Research has
shown that LA-ICPMS with deeper UV wavelengths (e.g., 157 nm, 193 or 213 nm)
results in less fractionation. A pulsed laser with short pulse widths—on the order
of picoseconds to femtoseconds—provides even less fractionation.
The most common LA-ICP/MS work to date is with flash-lamp pumped Nd:YAG
lasers that produce a light pulse of 3–10 nanoseconds in width and are relatively
inexpensive. Shorter pulse width lasers (picosecond to femtosecond) are widely
available but more expensive. A wavelength other than IR or UV is used only for
resonant laser ionization or ablation. In this case, the goal is to excite or ionize a
specific element preferentially, and the laser wavelength is chosen to correspond
to a particular transition of an element, as discussed in Section 17.7.4.
The laser typically used for laser ablation is the frequency quadrupled Nd:YAG
(266 nm) with a pulse width of a few nanoseconds. With this laser, some frac-
tionation will occur. Unless a good external standard is used, quantification of the
elements in the sample is difficult if not impossible. This is not an impediment
to use of laser ablation in the field of radiochemistry or radionuclides. One of the
most important aspects of radionuclide analysis is determining isotope ratios of
elements of interest. All isotopes of a particular element behave the same when
removed from a solid sample by a laser pulse. Ionization of those isotopes can
depend on the laser bandwidth and polarization so care must be used when using
lasers for direct ionization of atomic species.
The purpose of LA-ICP/MS use is to remove material from the surface and trans-
port it into the ICP for ionization to obtain isotope ratios quickly with little sample
preparation. Inter-element isotope ratios are also important in certain radionuclide
applications, but are compromised by fractionation issues. Laser ablation provides
an overview of which analytes and isotopes are present and approximates the
concentration of each.
Literature reviews and primary literature are available for a more thorough
analysis (Durrant, 1999; Gunther et al., 2000; Russo et al., 2002; Winefordner
et al., 2000). (Gastel et al., 1997) measured long-lived radionuclides in a concrete
matrix using LA-ICPMS. The radionuclides investigated were 99 Tc, 232 Th, 233 U
and 237 Np. With a quadrupole mass spectrometer, detection limits on the order
of 10 ng/g were achieved while a double-focusing sector instrument was able to
deliver sub ng/g detection limits. (Gibson, 1998) used resonant laser ablation mass
spectrometry to investigate actinide oxides. Oxides of Th, U, Np, and Am were
imbedded in a copper matrix and analyzed by resonant LA-MS. Since actinides
are present as oxides in many common forms such as glass, ceramics, soils, and
 17. Mass Spectrometric Radionuclide Analyses 405

others, the study was designed to determined whether resonant laser ablation would
provide any benefit over normal LA in the analysis of these elements. Due to the
many closely spaced energy levels of these elements, a significant advantage of
resonant laser ablation over normal laser ablation was not observed, but the study
showed the ability to use laser ablation for these types of samples.
Becker and Dietze (2000) used laser ablation to measure long-lived radionu-
clides in geological samples, high-purity graphite, and concrete. An effort was
made to improve quantification by using geologic standards and synthetic stan-
dards of the graphite and concrete. Solution nebulization was also used as a cali-
bration method. The isotopes studied were 99 Tc, 232 Th, 233 U, 235 U, 237 Np and 238 U.
The detection limits that resulted were in the low pg/g range. Boulyga et al. (2003)
used LA-ICP/MS to determine plutonium and americium in mosses. To improve
quantification, isotope dilution was used. Detection limits in the single fg/g range
were demonstrated for both elements.

17.7.3. Glow Discharge Mass Spectrometer

Another method to analyze solid samples directly is glow discharge MS (GD-MS).
A glow discharge is a reduced-pressure plasma (∼1 torr) that is formed between
two electrodes. The two main categories of glow discharge are direct current (DC)
and radiofrequency (RF). The DC glow discharge requires the solid, electrically
conducting sample as one of the electrodes. An RF glow discharge can analyze non-
conducting samples. The GD has been used as a source for both atomic emission
and mass spectrometry, but our focus is on the glow discharge as an ion source for
mass spectrometry of radionuclides. The reader is directed to additional literature
for more information concerning glow discharges (Marcus and Broekaert, 2002)
(Baude et al., 2000).
Because the only commercially available GD-MS systems use a DC glow dis-
charge, the range of samples that can be analyzed has been somewhat limited,
especially in the field of radionuclide determination. Once an RF GD-MS is avail-
able to purchase, it is possible that GD-MS could become more popular for these
applications.
Betti (1996) and co-workers used GD-MS for sample screening in isotopic
measurements of zirconium, silicon, lithium, boron, uranium, and plutonium in
nuclear samples. The results obtained from the GD-MS were compared with re-
sults from thermal ionization mass spectrometry (TIMS). For boron and lithium
concentrations from μg/g to ng/g levels, isotopic ratios determined by GD-MS
were comparable to TIMS in terms of accuracy and precision. Uranium isotopic
ratios determined by GD-MS were also in good agreement with values measured
by TIMS with regards to accuracy. Chartier et al. (1999) used GD-MS to analyze
erbium and uranium in molybdenum-uranium fuel samples. The ratio of 166 Er to
238
U was then compared to numbers determined by thermal ionization mass spec-
trometry. The ratio of erbium to uranium was accurate to within 3% of the number
determined by TIMS.
406 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

Pajo et al. (2001a) used GD-MS to measure impurities in uranium dioxide fuel
and showed that these impurities could be used to identify the original source of
confiscated, vagabond nuclear materials. De las Heras et al. (2000) used GD-MS to
determine neptunium in Irish Sea sediment samples. The sediment samples were
compacted into a disk that was used with a tantalum secondary cathode in the
glow discharge. Using a doped marine sediment standard for calibration, detection
limits down to the mid pg/g level were determined.

17.7.4. Resonance Ionization Mass Spectrometer

A resonance ionization mass spectrometer (RIMS) uses a tunable, narrow band-
width laser to excite an atom or molecule to a selected energy level that is then
analyzed by MS. The selective ionization often is accomplished by absorption of
more photons from the exciting laser, but can also be effected by a second laser or
a broadband photon source. Multiple photon absorption can result in direct ioniza-
tion or in production of excited species that can then be ionized with a low-energy
photon source (IR laser) or by a strong electric field. Resonance ionization methods
have been applied to nearly all elements in the periodic table and to many radionu-
clides, including Cs (Pibida et al., 2001), Th (Fearey et al., 1992), U (Herrmann
et al., 1991), Np (Riegel et al., 1993), Pu (Smith, 2000; Trautmann et al., 2004;
Wendt et al., 2000), radioxenon and radiokrypton (Watanabe et al., 2001; Wendt
et al., 2000), and 41 Ca (Wendt et al., 1999).
Resonance ionization MS is a sensitive and accurate method for determining the
ionization potential of atoms. Erdmann et al. (1998) recently reported improved
measurements of the ionization potentials for 9 actinide elements by RIMS.

17.8. Applications
Mass spectrometry finds applications in many fields. A partial list of fields includes:
r archeology
r bioassay
r biosciences
r cosmochemistry
r geochemistry
r health physics
r environmental monitoring
r nuclear non-proliferation treaty monitoring
r nuclear science
r radiochemistry

All of these fields have requirements for sensitive analysis of radionuclides. In

general, mass spectrometry is used for measuring long-lived radionuclides at low
abundances, where radiation-counting techniques provide insufficient sensitivity.
Modern mass spectrometers can detect individual atoms, which translates into
 17. Mass Spectrometric Radionuclide Analyses 407

TABLE 17.8. Suitable mass spectrometric techniques for analyzing radionuclides

Isotope ICP/MS TIMS AMS Other
Tritium Yes SIMS, 3 He-NGMS
10 Be Yes
14 C Yes
10 Be Yes
26 Al Yes
36 Cl Yes
41 Ca Yes Yes SIMS
59 Ni Yes
90 Sr Yes Yes RIMS
99 Tc Yes Yes Yes SIMS
129 I Yes Yes Yes SIMS
135 Cs Yes Yes RIMS
U Yes Yes Yes SIMS
Pu Yes Yes Yes SIMS, RIMS
Fission products Yes RIMS
Activation products Yes RIMS

sub-femto-gram sensitivities in real-world samples. A useful rule-of-thumb is that

mass spectrometry is generally more sensitive than radiation counting for isotopes
with half-lives greater than 100 years. This rule depends on the specific radionu-
clide in question, its decay scheme, and the radiation counting method.
A survey by radionuclide group and application is given below. Table 17.8 lists
most of the radionuclides measured by various mass spectrometric techniques.
Specific examples are given below. The interested reader is directed for further
information to the many excellent reviews on specific mass spectrometric tech-
niques, including Platzner (1997), Montaser (1998), Tuniz et al. (1998), Barshick
et al. (2000), de Laeter (2001).

17.8.1. Uranium
ICP/MS and TIMS are the methods of choice for analyzing uranium isotopes. TIMS
can measure all of the long-lived isotopes of uranium, including 236 U, with high
precision (0.01 percent or better) and high sensitivity (10−12 g or less) with mul-
tiple collectors (Adriaens et al., 1992; Becker and Dietze, 1998; De Laeter, 2001;
Delanghe et al., 2002; Efurd et al., 1995; Pajo et al., 2001b; Platzner, 1997; Sahoo
et al., 2002; Smith, 2000; Stoffels et al., 1994; Taylor et al., 1998; Yokoyama
et al., 2001). The TIMS requires extensive chemical processing to isolate ura-
nium from the sample. In contrast, ICP/MS, typically with a quadrupole mass
analyzer—although multi-collector, sector-based ICP/MS instruments are becom-
ing increasingly popular—is used for various samples, often without chemically
separating the uranium. The ICP/MS can provide isotopic information at the 0.01
to 1 percent precision level for the major isotopes, 235 U and 238 U (Aldstadt et al.,
1996; Becker et al., 2002; Becker et al., 2004a; Becker and Dietze, 2000; Bellis
et al., 2001; Boulyga and Becker, 2001; Boulyga et al., 2000; Haldimann et al.,
408 John F. Wacker, Gregory C. Eiden, and Scott A. Lehn

2001; Kerl et al., 1997; Magara et al., 2002; Manninen, 1995; Platzner et al., 1999;
Schaumloffel et al., 2005; Uchida et al., 2000; Wyse et al., 1998). Recently, AMS
has been used for uranium analysis, especially for measuring rare isotopes such
as 236 U (Brown et al., 2004; Danesi et al., 2003; Fifield, 2000; Tuniz, 2001; Zhao
et al., 1997).

17.8.2. Other Actinides

As with uranium, TIMS is the system of choice for measuring actinides such
as plutonium, americium, and neptunium (Beasley et al., 1998a; Beasley et al.,
1998b; Beasley et al., 1998c; Dai et al., 2001; Kelley et al., 1999; Kersting et al.,
1999; Lewis et al., 2001; Magara et al., 2000; Poupard and Jouniaux, 1990; Shen
et al., 2003; Stoffels et al., 1994). Very high sensitivity (<10−15 g) can be achieved
for plutonium and other actinides due to the low backgrounds for these elements
in typical environmental samples. Isotopic data obtained from measurement of
several actinides (e.g., neptunium, plutonium, americium, curium) can provide
diagnostic information on the conditions under which these elements were formed.
New mass spectrometric methods have been developed for high sensitivity ac-
tinide analysis. ICP/MS is routinely used for actinide analyses in various samples
(Agarande et al., 2001; Alonso et al., 1993; Alonso et al., 1995; Becker, 2003;
Becker and Dietze, 1998; Becker et al., 2004b; Boulyga and Becker, 2002; Boulyga
et al., 2003; Chiappini et al., 1996; Henry et al., 2001; Jerome et al., 1995; Kenna,
2002; Kim et al., 1991; Kim et al., 2000; Magara et al., 2000; Moreno et al., 1997;
Muramatsu et al., 2001; Muramatsu et al., 1999; Pappas et al., 2004; Rodushkin
et al., 1999; Rondinella et al., 2000; Sturup et al., 1998; Taylor et al., 2001; Ting
et al., 2003; Wyse and Fisher, 1994; Wyse et al., 1998; Zoriy et al., 2004). Since the
late 1990s, AMS methods have been developed to measure plutonium and other
actinides (Brown et al., 2004; Lee et al., 2001; McAninch et al., 2000; Oughton
et al., 2000; Oughton et al., 2004; Priest et al., 1999; Skipperud and Oughton, 2004).
RIMS has also been used for ultrasensitive actinide analyses (Gruning et al., 2004;
Trautmann et al., 2004).

17.8.3. Fission Products

Fission-produced isotopes that are currently measured by MS are 129 I, 99 Tc, 90 Sr,
and long-lived noble gases. Both 129 I and 99 Tc are measured with negative ion-
ization TIMS, ICP/MS and AMS (Beals and Hayes, 1995; Becker and Dietze,
1997; Becker et al., 2000; Fifield et al., 2000; Hidaka et al., 1999; McAninch
et al., 1998; Probst, 1996; Raisbeck and Yiou, 1999; Roberts and Caffee, 2000;
Yamamoto et al., 1995; Yiou et al., 2004). Methods for analyzing long-lived 135 Cs
have been developed using RIMS, TIMS, ICP/MS, and AMS (Karam et al., 2002;
Lee et al., 1993; Litherland and Kilius, 1997; Meeks et al., 1998; Pibida et al.,
2001; Song et al., 2001). Strontium-90 can now be measured by a RIMS technique
that uses three lasers to resonant excitation of 90 Sr, followed by mass analysis by
 17. Mass Spectrometric Radionuclide Analyses 409

quadrupole MS (Wendt et al., 1997). Detection limits are in the 105 atom range at
90
Sr/88 Sr isotope ratios as small as 10−10 .

17.8.4. Activation Products

Radionuclides with the potential for detection by MS include tritium, 14 C, and 36 Cl.
AMS is the method of choice for 14 C and 36 Cl (Fifield, 1999; Fifield, 2000; Herzog,
1994; Hotchkis et al., 2000; Rucklidge, 1995; Skipperud and Oughton, 2004; Tuniz
et al., 1998). Tritium can be analyzed by measuring its 3 He decay product using
noble gas mass spectrometry (NGMS) (De Laeter, 2001; Eaton et al., 2004; Faure,
1986; Love et al., 2002), but high sensitivity measurements may require several
months to produce adequate levels of 3 He for detection. AMS has been successfully
developed for tritium analyses and is routinely used (Chiarappa-Zucca et al., 2002;
King et al., 1987; Love et al., 2002; Tuniz et al., 1998).

17.9. Instrumental Advancements

As mentioned in the Introduction to this chapter, development of new MS systems
has focused on the ability to accurately measure smaller samples, either for greater
sensitivity or shorter-lived radionuclides, and to more effectively individuate these
samples. In addition to these advancements, refinements of the instrumentation are
now allowing manufactures to scale down the size of their MS workstations.
The MS systems discussed in this chapter are large, laboratory-scale instruments.
Recent developments in building small mass analyzers, in miniaturized electronics,
and in compact vacuum systems, have stimulated efforts to build a portable, or
fieldable, MS. Chances are good for producing suit-case sized instruments that
can be used in a remote office, mounted on aircraft or trucks, or placed in a
remote location for environmental monitoring. Prototypes of such instruments have
been demonstrated and commercial versions are becoming increasingly available.
Several companies are producing portable, battery-powered systems that are used
for screening analyses on-site (i.e., not in the laboratory). One drawback is mass;
these instruments are still quite heavy, at 20–25 kg for an instrument the size of a
duffle bag.
An internet keyword search for “portable mass spec” reveals a great deal of
advancement in these endeavors. More is certain, as lighter and stronger materials
are found and engineered to produce more portable and rugged systems. Ongoing
improvements in battery technology will also have an impact. These efforts parallel
the remote automated systems discussed in Chapter 15, as the need for immediate
and/or ongoing analysis of samples becomes ever greater in our society.
Appendix

Appendix A1: ASTM Radiological Standard Test Methods,

Practices, and Guides (ASTM 2005)

Committee Publication Title

C26.01 C0859-92b Standard Terminology Relating to Nuclear Materi-
als
C26.05 C0696-99 Standard Test Methods for Chemical, Mass Spectro-
metric, and Spectrochemical Analysis of Nuclear-
Grade Uranium Dioxide Powders and Pellets
C26.05 C0697-98 Standard Test Methods for Chemical, Mass Spectro-
metric, and Spectrochemical Analysis of Nuclear-
Grade Plutonium Dioxide Powders and Pellets
C26.05 C0698-98 Standard Test Methods for Chemical, Mass
Spectrometric, and Spectrochemical Analysis of
Nuclear-Grade Mixed Oxides ((U, Pu)O2 )
C26.05 C0758-98 Standard Test Methods for Chemical, Mass Spectro-
metric, Spectrochemical, Nuclear, and Radiochem-
ical Analysis of Nuclear-Grade Plutonium Metal
C26.05 C0759-98 Standard Test Methods for Chemical, Mass
Spectrometric, Spectrochemical, Nuclear, and Ra-
diochemical Analysis of Nuclear-Grade Plutonium
Nitrate Solutions
C26.05 C0761-96 Standard Test Methods for Chemical, Mass Spectro-
metric, Spectrochemical, Nuclear, and Radiochem-
ical Analysis of Uranium Hexafluoride
C26.05 C0799-99 Standard Test Methods for Chemical, Mass Spec-
trometric, Spectrochemical, Nuclear, and Radio-
chemical Analysis of Nuclear-Grade Uranyl Nitrate
Solutions
C26.05 C1342-96 Standard Practice for Flux Fusion Sample Dissolu-
tion

410
 Appendix 411

C26.05 C1344-97 Standard Test Method for Isotopic Analysis

of Uranium Hexafluoride by Single-Standard
Gas Source Mass Spectrometer Method
C26.05 C1347-96a Standard Practice for Preparation and Disso-
lution of Uranium Materials for Analysis
C26.05 C1380-97 Standard Test Method for Determination of
Uranium Content and Isotopic Composition
by Isotope Dilution Mass Spectrometry
C26.05 C1413-99 Standard Test Method for Isotopic Analysis
of UF6 and UNH Solutions by Thermal Ion-
ization Mass Spectrometry
C26.05 E0219-80(1995) Standard Test Method for Atom Percent
Fission in Uranium Fuel (Radiochemical
Method)
C26.05 E0244-80(1995) Standard Test Method for Atom Percent Fis-
sion in Uranium and Plutonium Fuel (Mass
Spectrometric Method)
C26.05 E0267-90(1995) Standard Test Method for Uranium and Plu-
tonium Concentrations and Isotopic Abun-
dances
C26.05 E0318-91 Standard Test Method for Uranium in Aque-
ous Solutions by Colorimetry
C26.05 E0321-96 Standard Test Method for Atom Percent
Fission in Uranium and Plutonium Fuel
(Neodymium-148 Method)
C26.05.01 C0998-90(1995)e1 Standard Practice for Sampling Surface Soil
for Radionuclides
C26.05.01 C0999-90(1995)e1 Standard Practice for Soil Sample Prepara-
tion for the Determination of Radionuclides
C26.05.01 C1000-90(1995)e1 Standard Test Method for Radiochemical De-
termination of Uranium Isotopes in Soil by
Alpha Spectrometry
C26.05.01 C1001-90(1995)e1 Standard Test Method for Radiochemical De-
termination of Plutonium in Soil by Alpha
Spectroscopy
C26.05.01 C1163-98 Standard Test Method for Mounting Ac-
tinides for Alpha Spectrometry Using
Neodymium Fluoride
C26.05.01 C1205-97 Standard Test Method for The Radiochemical
Determination of Americium-241 in Soil by
Alpha Spectrometry
C26.05.01 C1284-94 Standard Practice for Electrodeposition of the
Actinides for Alpha Spectrometry
C26.05.01 C1387-98 Guide for the Determination of Technetium-
99 in Soil
412 Appendix

C26.05.01 C1402-98 High-Resolution Gamma-Ray Spectrometry

of Soil Samples
C26.05.02 C1108-99 Standard Test Method for Plutonium by
Controlled-Potential Coulometry
C26.05.02 C1165-90(1995)e1 Standard Test Method for Determining Plu-
tonium by Controlled-Potential Coulome-
try in H2 SO4 at a Platinium Working
Electrode
C26.05.02 C1168-90(1995)e1 Standard Practice for Preparation and Disso-
lution of Plutonium Materials for Analysis
C26.05.02 C1206-91e1 Standard Test Method for Plutonium by Iron
(II)/Chromium (VI) Amperometric Titration
C26.05.02 C1235-99 Standard Test Method for Plutonium by Ti-
tanium(III)/Cerium(IV) Titration
C26.05.02 C1267-94 Standard Test Method for Uranium by Iron
(II) Reduction in Phosphoric Acid Followed
by Chromium (VI) Titration in the Presence
of Vanadium
C26.05.02 C1307-95 Standard Test Method for Plutonium Assay
by Plutonium (III) Diode Array Spectropho-
tometry
C26.05.02 C1411-99 Standard Practice for the Ion Exchange Sep-
aration of Uranium and Plutonium Prior to
Isotopic Analysis
C26.05.02 C1414-99 Standard Practice for the Separation of
Americium from Plutonium by Ion Exchange
C26.05.02 C1415-99 Standard Test Method for Pu-238 Isotopic
Abundance by Alpha Spectrometry
C26.05.03 C1109-98 Standard Test Method for Analysis of Aque-
ous Leachates from Nuclear Waste Materials
Using Inductively Coupled Plasma-Atomic
Emission Spectrometry
C26.05.03 C1310-95 Standard Test Method for Determining Ra-
dionuclides in Soils by Inductively Coupled
Plasma-Mass Spectrometry Using Flow In-
jection Preconcentration
C26.05.03 C1317-95 Standard Practice for Dissolution of Silicate
or Acid Resistant Matrix Samples
C26.05.03 C1345-96 Standard Test Method for the Analysis of To-
tal and Isotopic Uranium and Total Thorium
in Soils by Inductively Coupled Plasma-Mass
Spectrometry
C26.05.03 C1379-97 Standard Test Method for Analysis of Urine
for Uranium-235 and Uranium-238 by Induc-
tively Coupled Plasma Mass Spectrometry
 Appendix 413

C26.05.03 C1412-99 Standard Practice for Microwave Oven Dis-

solution of Glass Containing Radioactive and
Mixed Wastes
C26.05.04 C1204-91(1996) Standard Test Method for Uranium in the
Presence of Plutonium by Iron(II) Re-
duction in Phosphoric Acid Followed by
Chromium(VI) Titration
C26.05.04 C1287-95 Standard Test Method for Determination of
Impurities in Uranium Dioxide by Induc-
tively Coupled Plasma Mass Spectrometry
C26.05.04 C1295-95 Standard Test Method for Gamma Energy
Emission from Fission Products in Uranium
Hexafluoride
C26.05.05 C1110-88(1997)e1 Standard Practice for Sample Preparation for
X-Ray Emission Spectrometric Analysis of
Uranium in Ores Using the Glass Fusion or
Pressed Powder Method
C26.05.05 C1254-99 Standard Test Method for Determination of
Uranium in Mineral Acids by X-Ray Fluo-
rescence
C26.05.05 C1255-93(1999) Standard Test Method for Analysis of Ura-
nium and Thorium in Soils by Energy Dis-
persive X-Ray Fluorescence Spectroscopy
C26.05.05 C1343-96 Standard Test Method for Determination of
Low Concentrations of Uranium in Oils and
Organic Liquids by X-Ray Fluorescence
C26.05.05 C1416-99 Standard Test Method for Uranium Analysis
in Natural and Waste Water by X-Ray Fluo-
rescence
C26.06 C1215-92(1997) Standard Guide for Preparing and Interpret-
ing Precision and Bias Statements in Test
Method Standards Used in the Nuclear In-
dustry
C26.08 C1009-96e1 Standard Guide for Establishing a Quality
Assurance Program for Analytical Chemistry
Laboratories Within the Nuclear Industry
C26.08 C1068-96e1 Standard Guide for Qualification of Measure-
ment Methods by a Laboratory Within the
Nuclear Industry
C26.08 C1128-95e1 Standard Guide for Preparation of Working
Reference Materials for Use in the Analysis
of Nuclear Fuel Cycle Materials
C26.08 C1156-95e1 Standard Guide for Establishing Calibration
for a Measurement Method Used to Analyze
Nuclear Fuel Cycle Materials
414 Appendix

C26.08 C1188-91(1997)e1 Standard Guide for Establishing a Quality As-

surance Program for Uranium Conversion Facil-
ities
C26.08 C1210-96E1 Standard Guide for Establishing a Measurement
System Quality Control Program for Analytical
Chemistry Laboratories Within the Nuclear In-
dustry
C26.08 C1297-95e1 Standard Guide for Qualification of Laboratory
Analysts for the Analysis of Nuclear Fuel Cycle
Materials
C26.10 C1030-95 Standard Test Method for Determination of Plu-
tonium Isotopic Composition by Gamma-Ray
Spectrometry
C26.10 C1133-96 Standard Test Method for Nondestructive Assay
of Special Nuclear Material in Low Density
Scrap and Waste by Segmented Passive Gamma-
Ray Scan
C26.10 C1207-97 Standard Test Method for Nondestructive As-
say of Plutonium in Scrap and Waste by Passive
Neutron Coincidence Counting
C26.10 C1221-92(1998) Standard Test Method for Nondestructive Anal-
ysis of Special Nuclear Materials in Homoge-
neous Solutions by Gamma-Ray Spectrometry
C26.10 C1268-94 Standard Test Method for Quantitative
Determination of Americium-241 in Plutonium
by Gamma-Ray Spectrometry
C26.10 C1316-95 Standard Test Method for Nondestructive As-
say of Nuclear Material in Scrap and Waste by
Passive-Active Neutron Counting Using a Cf-
252 Shuffler
D19.04 D1890-96 Standard Test Method for Beta Particle Radioac-
tivity of Water
D19.04 D1943-96 Standard Test Method for Alpha Particle Ra-
dioactivity of Water
D19.04 D2460-97 Standard Test Method for Alpha-Particle-
Emitting Isotopes of Radium in Water
D19.04 D2907-97 Standard Test Methods for Microquantities of
Uranium in Water by Fluorometry
D19.04 D3084-96 Standard Practice for Alpha-Particle Spectrom-
etry of Water
D19.04 D3454-97 Standard Test Method for Radium-226 in Water
D19.04 D3648-95 Standard Practices for the Measurement of Ra-
dioactivity
D19.04 D3649-91 Standard Test Method for High-Resolution
Gamma-Ray Spectrometry of Water
 Appendix 415

D19.04 D3865-97 Standard Test Method for Plutonium in Water

D19.04 D3972-97 Standard Test Method for Isotopic Uranium in Water
by Radiochemistry
D19.04 D4107-91 Standard Test Method for Tritium in Drinking Water
D19.04 D4785-93 Standard Test Method for Low-Level Iodine-131 in
Water
D19.04 D4922-94e1 Standard Test Method for Determination of Radioac-
tive Iron in Water
D19.04 D4962-95e1 Standard Practice for NaI(Tl) Gamma-Ray Spectrom-
etry of Water
D19.04 D5072-92 Standard Test Method for Radon in Drinking Water
D19.04 D5174-97 Standard Test Method for Trace Uranium in Water by
Pulsed-Laser Phosphorimetry
D19.04 D5411-93 Standard Practice for Calculation of Average Energy
Per Disintegration (E) for a Mixture of Radionuclides
in Reactor Coolant
D19.04 D5811-95 Standard Test Method for Strontium-90 in Water
D19.04 D6239-98a Standard Test Method for Uranium in Drinking Water
by High-Resolution Alpha-Liquid-Scintillation Spec-
trometry
D19.05 D5673-96 Standard Test Method for Elements in Water by Induc-
tively Coupled Plasma-Mass Spectrometry
D22.05 D6327-98 Standard Test Method for Determination of Radon
Decay Product Concentration and Working Level
in Indoor Atmospheres by Active Sampling on a
Filter
E01.05 E0402-95 Standard Test Method for Spectrographic Analysis of
Uranium Oxide
E10.03 E1893-97 Standard Guide for Selection and Use of Portable Ra-
diological Survey Instruments for Performing In Situ
Radiological Assessments in Support of Decommis-
sioning
E10.05 E0181-93e1 Standard Test Methods for Detector Calibration and
Analysis of Radionuclides
E10.05 E0266-92 Standard Test Method for Measuring Fast-Neutron Re-
action Rates by Radioactivation of Aluminum
E10.05 E0343-96 Standard Test Method for Measuring Reaction Rates
by Analysis of Molybdenum-99 Radioactivity From
Fission Dosimeters
E10.05 E0393-96 Standard Test Method for Measuring Reaction Rates
by Analysis of Barium-140 From Fission Dosimeters
E10.05 E0481-97 Standard Test Method for Measuring Neutron Fluence
Rate by Radioactivation of Cobalt and Silver
E10.05 E0523-92 Standard Test Method for Measuring Fast-Neutron Re-
action Rates by Radioactivation of Copper
416 Appendix

E10.05 E0692-98 Standard Test Method for Determining the Content

of Cesium-137 in Irradiated Nuclear Fuels by High-
Resolution Gamma-Ray Spectral Analysis
E10.05 E0704-96 Standard Test Method for Measuring Reaction Rates by
Radioactivation of Uranium-238

Appendix A2: Standard Methods, part 7000—Radioactivity

(Standard Methods 2005)

Method no. Contents

7010 Introduction
A. General Discussion
B. Sample Collection and Preservation
7020 Quality Assurance/Quality Control
A. Basic Quality Control Program
B. Quality Control for Wastewater Samples
C. Statistics
D. Calculation and Expression of Results
7030 Counting Instruments
A. Introduction
B. Description and Operation of Instruments
7040 Facilities
A. Counting Room
B. Radiochemistry Laboratory
C. Laboratory Safety
D. Pollution Prevention
E. Waste Management
7110 Gross Alpha and Gross Beta Radioactivity (Total, Suspended, and
Dissolved)
A. Introduction
B. Evaporation Method for Gross Alpha-Beta
C. Coprecipitation Method for Gross Alpha
Radioactivity in Drinking Water
7120 Gamma-Emitting Radionuclides
A. Introduction
B. Gamma Spectroscopic Method
7500 Cs Radioactive Cesium
A. Introduction
B. Precipitation Method
7500 I Radioactive Iodine
A. Introduction
B. Precipitation Method
 Appendix 417

7500 Ra Radium
A. Introduction
B. Precipitation Method
C. Emanation Method
D. Sequential Precipitation Method
E. Gamma Spectroscopy Method
7500 Rn Radon
A. Introduction
B. Liquid Scintillation Method
7500 Sr Total Radioactive Strontium and Strontium 90
A. Introduction
B. Precipitation Method
7500 3 H Tritium
A. Introduction
B. Liquid Scintillation Spectrometric Method

Appendix A3: Applicable ANSI Standard Procedures (ANSI

2005).

Procedure no. Description

ANSI N42.31-2003 American National Standard for measurement procedures
for resolution and efficiency of wide-bandgap semiconduc-
tor detectors of ionizing radiation
ANSI N42.25-1997 American National Standard calibration and usage of al-
pha/beta proportional counters

ANSI N42.23-1996 American National Standard measurement and associated

instrument quality assurance for radioassay laboratories
ANSI N42.22-1995 American National Standard—traceability of radioactive
sources to the National Institute of Standards and Technol-
ogy
ANSI N42.16-1986 American National Standard specifications for sealed ra-
dioactive check sources used in liquid-scintillation coun-
ters
ANSI N42.15-1997 American National Standard check sources for and verifi-
cation of liquid-scintillation counting systems
ANSI N42.14-1999 American National Standard for calibration and use of ger-
manium spectrometers for the measurement of gamma-ray
emission rates of radionuclides
ANSI N42.12-1994 American National Standard calibration and usage of
thallium-activated sodium iodide detector systems for as-
say of radionuclides
418 Appendix

Appendix A4: Applicable ISO Standard Procedures (ISO

2005).

Regulation ID Description
ISO 9696:1992 Water quality: Measurement of gross alpha activity in non-
saline water (Thick source method)
ISO 9697:1992 Water quality: Measurement of gross beta activity in nonsaline
water
ISO 9698:1989 Water quality: Determination of tritium activity concentration
(Liquid scintillation counting method)
ISO 10703:1997 Water quality: Determination of the activity concentration of
radionuclides by high resolution gamma-ray spectrometry
ISO 15366:1999 Nuclear energy: Chemical separation and purification of ura-
nium and plutonium in nitric acid solutions for isotopic and
dilution analysis by solvent chromatography

Appendix B: Regulation- and Standard-Setting Agencies

AGENCY ADDRESS CONTACT WEB SITE
American Institute Of One East Wacker Ph: 312-670-2400 E-mail: pubs@aisc.org
Steel Construction Drive, Suite 3100 Internet: http://www.aisc.org
(AISC) Chicago, IL
60601-2001
American National 1819 L Street, NW, 6th Ph: 202-293-8020 E-mail: info@ansi.org
Standards Institute Floor Washington, Internet: http://www.ansi.org/
(ANSI) DC 20036
American Society Of 1801 Alexander Bell Ph: 703-295-6300 E-mail: marketing@asce.org
Civil Engineers Drive Reston, VA 800-548-2723 Internet: http://www.asce.org
(ASCE) 20191-4400
American Society Of 1791 Tullie Circle, NE Ph: 800-527-4723 E-mail: ashrae@ashrae.org
Heating, Atlanta, GA 30329 404-636-8400 Internet: http://www.ashrae.org
Refrigerating And
Air-Conditioning
Engineers
(ASHRAE)
ASTM International 100 Barr Harbor Drive, Ph: 610-832-9500 E-mail: service@astm.org
(ASTM) P.O. Box C700 West Internet: http://www.astm.org
Conshohocken, PA
19428-2959
Builders Hardware 355 Lexington Avenue Ph: 212-297-2122 E-mail: assocmgmt@aol.com
Manufacturers 17th Floor New Internet:http://www. buildershar-
Association York, NY 10017 dware.com
(BHMA)
International Code 5203 Leesburg Pike, Ph: 703-931-4533 E-mail: webmaster@iccsafe.org
Council (ICC) Suite 600 Falls Internet: http://www.intlcode.org
Church, VA 22041
 Appendix 419

AGENCY ADDRESS CONTACT WEB SITE

National Electrical 1300 North 17th Ph: 703-841-3200 E-mail: webmaster@nema.org
Manufacturers Street, Suite 1847 Internet: http:// www.nema.org/
Association Roslyn, VA 22209
(NEMA)
National Fire 1 Batterymarch Park Ph: 617-770-3000 E-mail: webmaster@nfpa.org
Protection P.O. Box 9101 Internet: http:// www.nfpa.org
Association Quincy, MA
(NFPA) 02269-9101
National Institute For Mail Stop C-13 4676 Ph: 800-356-4674 E-mail: pubstaff@cdc.gov
Occupational Columbia Parkway Internet: http://www. cdc.gov/
Safety And Health Cincinnati, OH niosh/ homepage.html
(NIOSH) 45226-1998
National Roofing 10255 West Higgins
Contractors Road, Suite 600
Association Rosemont, IL 60018
(NRCA)
AGENCY ADDRESS CONTACT WEB SITE
Ph: 847-299-9070 Internet:
http://www.nrca.net
Sheet Metal & Air 4201 Lafayette Center Ph: 703-803-2980 E-mail: info@smacna.org
Conditioning Drive, Chantilly, VA Internet: http://www.smacna.org
Contractors’ 20151-1209
National
Association
(SMACNA)
Underwriters 333 Pfingsten Road Ph: 847-272-8800 E-mail: northbrook@us.ul.com
Laboratories (UL) Northbrook, IL Internet: http://www.ul.com/
60062-2096
U.S. Army Corps Of 3909 Halls Ferry Rd. Ph: 601-634-2664 E-mail: mtc-info@erdc.
Engineers Vicksburg, MS usace.army.mil Internet:
(USACE) 39180-6199 http://www.
wes.army.mil/SL/
MTC/handbook.htm
U.S. Environmental Ariel Rios Building Ph: 202-260-2090 Internet: http://www.
Protection Agency 1200 Pennsylvania epa.gov
(EPA) Avenue, N.W.
Washington, DC
20460

Mr. Doug Ashley, Architect, with Bullock Tice Associates, Pensacola, FL, provided the information
given in this appendix.
 Appendix C
420

PERIODIC TABLE 2006

atomic atomic
1 1.01 number weight
alkali metals 2 4.003

alkaline earth metals

H 14 28.09 He
Appendix

Hydrogen Helium
3 6.94 4 9.01
Si symbol: transitional metals 5 10.81 6 12.01 7 14.01 8 15.999 9 18.998 10 20.18

Silicon black solid

Li Be blue liquid other metals B C N O F Ne
Lithium Beryllium red gas Boron Carbon Nitrogen Oxygen Fluorine Neon
11 22.99 12 24.31
name
non metals 13 26.98 14 28.09 15 30.97 16 32.06 17 35.45 18 39.95

Na Mg noble gases AI Si P S CI Ar
Sodium Magnesium Aluminum Silicon Phosphorus Sulfur Chlorine Argon
19 39.10 20 40.08 21 44.96 22 47.90 23 50.94 24 51.996 25 54.94 26 55.85 27 58.93 28 58.70 29 63.55 30 65.37 31 69.72 32 72.59 33 74.92 34 78.96 35 79.90 36 83.80

K Ca Sc Ti
Potassium Calcium Scandium Titanium
V Cr Mn Fe Co Ni Cu Zn Ga Ge As Se Br Kr
Vanadium Chromium Manganese Iron Cobalt Nickel Copper Zinc Gallium Germanium Arsenic Selenium Bromine Krypton
85.47 38 87.62 39 88.91 40 91.22 41 92.91 42 95.94 43 (98) 44 101.07 45 102.91 46 106.42 47 107.87 48 112.41 49 114.82 50 118.69 51 121.75 52 127.60 53 126.90 54 131.30
37

Rb Sr Y Zr Nb Mo Tc Ru Rh Pd Ag Cd In Sn Sb Te
Rubidium Strontium Yttrium Zirconium Niobium Molybdenum Technetium Ruthenium Rhodium Palladium Silver Cadmium Indium Tin Antimony Tellurium
I
Iodine
Xe Xenon
132.91 56 137.33 57 138.91 72 178.49 73 180.95 74 183.84 75 186.21 76 190.23 77 192.22 78 195.08 79 196.97 80 200.59 81 204.37 82 207.19 83 208.98 84 (209) 85 (210) 86 (222)
55

Cs Ba La Hf Ta W Re Os Ir Pt Au Hg TI Pb Bi Po At Rn
Cesium Barium Lanthanum Hafnium Tantalum Tungsten Rhenium Osmium Iridium Platinum Gold Thallium Lead Bismuth Polonium Astatine Radon
Mercury
(223) 88 226.03 89 227.00 104 (261) 105 (262) 106 (266) 107 (264) 108 (269) 109 (268) 110 (271) 111 (272) (277) (284) (289) (288) (292)
87

Fr Ra Ac Rf Db Sg Bh Hs Mt Ds Rg 112 113 114 115 116 (117) (118)

Francium Radium Actinium Rutherfordium Dubnium Seaborgium Bohrium Hassium Meitnerium Darmstadtium Roentgenium

(119) (120) (121) (154)

58 140.12 59 140.91 60 144.24 61 (145) 62 150.40 63 151.96 64 157.25 65 158.93 66 162.50 67 164.93 68 167.26 69 168.93 70 173.04 71 174.97

Lanthanides Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu
Cerium Praseodymium Neodymium Promethium Samarium Europium Gadolinium Terbium Dysprosium Holmium Erbium Thulium Ytterbium Lutetium
90 232.04 91 231.04 92 238.03 93 237.05 94 (244) 95 (243) 96 (247) 97 (247) 98 (251) 99 (252) 100 (257) 101 (258) 102 (259) 103 (262)

Actinides Th Pa U Np Pu Am Cm Bk Cf Es Fm Md No Lr
Thorium Protactinium Uranium Neptunium Plutonium Americium Curium Berkelium Californium Einsteinium Fermium Mendelevium Nobelium Lawrencium

Superactinides (122-153)
Glossary

ablation: In this context, volatilization of a solid surface by localized heating, as

by a laser.
abundance: Amount of atoms or molecules present.
abundance sensitivity: A measure of the ability to detect a minor peak next to an
abundant one.
accelerator: Machine used to produce ions at high energy for nuclear experiments.
See Cockcroft–Walton accelerator.
actinides or actinide series: The fourteen elements with the atomic numbers 90
through 103, inclusive, in which filling of the 5f inner electron shell occurs. Name
is derived from the immediately preceding element actinium (Z = 89). Analogous
to the lanthanide series, atomic numbers 58–71, inclusive, in which electrons are
added to the 4f electron shell.
activity: See disintegration rate.
Agreement States: States that have programs to assume partial NRC regulatory
authority under the Atomic Energy Act of 1954. This authority includes licensing
and regulating radioactive byproduct materials; source materials (uranium and
thorium); and special nuclear materials.
ALARA: As low as reasonably achievable for controlling radiation exposure.
aliquot: A measured fraction of a solution.
alpha particle (α particle): A particle, consisting of two protons and two neutrons,
that is emitted from the nucleus of an unstable radionuclide. Equivalent to a helium
nucleus.
amu: Abbreviation for atomic mass unit or Dalton. The unified atomic mass unit
(symbol u) finds increasingly common usage, and is equivalent to the amu. See
Dalton.
annihilation (process): The process of interaction between matter and antimatter
to produce electromagnetic radiation. For example, the interaction between an
electron and a positron.

421
422 Glossary

antineutrino: In the process of negatron particle emission, the antineutrino carries

away the residual energy, momentum, and spin to conserve these quantities. This
sub atomic particle has zero charge, near-zero mass, and 1/2 spin.
aqua regia: A mixture of nitric and hydrochloric acids. The term literally means
“royal water,” because of its ability to dissolve gold and platinum, which are noble
metals.
artificial or “synthetic” elements: Elements not found in nature; sometimes even
refers to elements first synthesized in the laboratory before their discovery in
nature, e.g., technetium (Z = 43) or plutonium (Z = 94).
ashing: Burning a sample until only ash remains.
asymmetric fission: Fission process where fission products are distributed around
a light mass and a heavy mass.
Atomic Energy Act of 1954: One of the fundamental laws governing civilian uses
of nuclear materials and facilities.
atomic number (Z): The number of protons in the nucleus of an atom.
atomic orbital: A representation of the three-dimensional region in space in which
the electron is likely to be found. The orbital functions, which define the space,
are derived by quantum mechanics.
attribution: Ascribing responsibility. In the context of this textbook, attribution
studies refer to research conducted in order to determine the origin of a particular
set of detected radionuclides.
audit: The complete and methodical review of laboratory performance and results.
Auger electron: Pronounced O-jhay and named after French physicist Pierre
Auger, Auger electron emission is one possible consequence of an internal con-
version. When an inner core electron vacancy is filled by an electron from a higher
energy level, energy must be released. This energy release sometimes occurs as
the emission of an atomic electron, the Auger electron.
Avogadro’s number (Av ): The number of entities in one mol of any substance,
defined as the number of atoms in 0.012 kg of 12 C. Av = 6.02 × 1023 . NA is an
equivalent symbol for Avogadro’s number.
azeotropic distillation: Purification of tritiated water by codistillation with an
organic solvent. A fixed ratio of solvent and water are distilled at the corresponding
boiling point; on condensing, the organic solvent and water separate into distinct
phases, with minimal solubility of the organic solvent in water.
back extraction: Transfer of a substance that had previously been extracted into
a second solvent back into the first solvent.
bandpass filter: A control for blocking all signals outside a selected range of
characteristics such as frequency, energy, or mass to charge ratio.
bandwidth: The range of selected signal such as frequency, energy, or mass to
charge ratio.
barn (b): Unit for cross section that equals 10−24 cm2 .
 Glossary 423

−1
becquerel (Bq): The SI unit of activity of a radionuclide, equal to 1 s or 1
disintegration per second. One Bq = 2.703 × 10−11 curies (Ci).
berm: Man-made geographical feature; a mound of earth. In this context, a berm
is usually an earthen wall that surrounds holding tanks to contain any liquid that
might leak from them.
beta particle (β particle): Term given to an electron (β − particle) or positron (β +
particle) emitted from the nucleus.
bioassay: Analysis of biological material.
“blind” sample: Jargon; a sample of known content, but unknown to the analyst.
branching ratios: The fractional radiation decays that result in two or more prod-
ucts.
Bremsstrahlung radiation: “Braking radiation” (German). Electromagnetic radi-
ation produced when one charged particle (say, an electron) is deflected or deceler-
ated by another charged particle (say, a nucleus). Bremsstrahlung has a continuous
spectrum that increases in intensity with decreasing energy.
carrier: A stable isotope in appreciable amount which, when thoroughly mixed
with a trace of a specified radioelement, will carry the trace with it through a
chemical or physical process. The stable and radioactive isotopes must behave
identically.
chain reaction: A self-sustaining series of nuclear fissions that is propa-
gated through the absorption of neutrons produced as the product of prior
fissions.
Cherenkov radiation: Electromagnetic radiation resulting from charged particles
traveling in a medium faster than light can travel in the same medium. The de-
celeration of the particle produces this radiation, which is sometimes seen as the
characteristic “blue glow” at a nuclear reactor.
clear melt: The transparent form of melted solids.
Cockcroft-Walton accelerator: A high-voltage device inside which charged ca-
pacitors discharge to add electrons to hydrogen atoms to create negatively charged
ions. The charged particles are then driven through an accelerating tube toward a
target.
cold: Jargon; characterized by an absence of radiation or radionuclides.
colligative: A property that depends on the quantity of atoms, not their form.
colorimetry: Spectrophotometric analysis of visible electromagnetic radiation.
combined standard uncertainty: A standard uncertainty calculated by propa-
gation of uncertainty. The combined standard uncertainty of a result y may be
denoted by u c (y).
combustible liquid: Liquid with a flashpoint at or above 100◦ F. Class II com-
bustible liquids have a flashpoint between 100◦ F and 140◦ F and Class III com-
bustible liquids have a flashpoint above 140◦ F.
424 Glossary

Compton scattering: Occurs when gamma rays interact with electrons in a mate-
rial. The photon donates some of its energy to the electron. The electron is ejected
from its atom and the reduction in energy of the photon results in an increase in
its wavelength. The photon with the remaining energy is deflected.
continuum (energy): Range of particle energies, distributed in a continuous fash-
ion from E = 0 to E = maximum value.
conversion electron (CE): The inner orbital electron that receives the excess
energy of a metastable nucleus during the course of an internal conversion. Because
of its relative proximity to the nucleus, the conversion electron usually comes from
the K shell.
coprecipitation: The simultaneous precipitation of a normally soluble component
with a macro-component from the same solution by the formation of mixed crystals,
by adsorption, occlusion, or mechanical entrapment.
coulombic barrier: Charge potential experienced when one electrically charged
projectile approaches another charged object. For example, electrical charge re-
pulsion when an alpha particle approaches the nucleus of an atom.
coverage factor: The factor k by which the combined standard uncertainty of a
result, u c (y), of a measurement is multiplied to obtain the expanded uncertainty,
U.
coverage probability: The approximate probability that the interval y ± U de-
scribed by a measured result, y, and its expanded uncertainty, U , will contain the
true value of the measurand.
critical value: Threshold value to which the result of a measurement is compared
to make a detection decision (also sometimes called “critical level” or “decision
level”). The critical value can be expressed as (for example) the critical gross count,
critical net count, critical net count rate, or the critical activity.
cross section: Expression of the interaction probability for a nuclear reaction per
unit of area, in cm2 per atom. The term describes the area the nucleus presents for
a particular projectile to strike.
cross-talk: Interfering signals among parallel measurements.
crystal lattice energy: The energy holding ions or atoms together in a crystal
structure.
curie (Ci): A former unit of radioactivity, corresponding roughly to the acitivity
of one gram of the radioactive isotope 226 Ra. 1 Ci = 3.7 × 1010 decays per second
= 37 gigabequerels (GBq).
cyclotron: Circular charged particle accelerator invented by E.O. Lawrence in
1929. Pole diameters range in size from a few inches up to 236 inches with energies
from several million electron volts (MeV) up to 700 MeV for protons. Heavier
projectiles can also be accelerated.
DAC: Derived air concentration of a radionuclide that is related to an intake
limit.
 Glossary 425

Dalton: Named after English scientist John Dalton, the Dalton is defined as exactly
1/12 of the relative atomic mass assigned to 12 C. It is functionally the same as an
atomic mass unit (amu) and unified atomic mass unit (u).
daughter(s): Term given to the radioactive product of a radioactive parent, also
progeny.
decay: A term describing the emission of energy from a radionuclide during the
course of a nuclear transformation. Typically one decay mode, such as alpha par-
ticle emission, predominates for a particular nuclide. When more than one decay
mode is evident, branching ratios are used to describe the fraction of the decay
that occurs by each pathway.
decay chain: The succession of radioactive products (called daughters or progeny)
to which a radionuclide decays.
decay constant (λ): A constant (in units of reciprocal time) that is characteristic
of a particular radionuclide’s decay.
decay fraction: The fraction of a particular mode of decay when a radionuclide
decays by several modes; also termed branching ratio.
decontamination factor: The ratio of the proportion of contaminant to product
before treatment to the proportion after treatment.
de-excitation: Process in which an nucleus, atom, or molecule releases energy,
with the effect of reaching a less energetic state.
dewar: A metal or glass container designed to hold liquefied gases. The container
was invented by Scottish chemist and physicist Sir James Dewar, who produced
liquid hydrogen.
disintegration rate: Rate describing the number of nuclear transformations per
unit time, in units of bequerels (Bq). Also referred to as the activity of a radionuclide
sample.
doping: The process of adding a foreign material to an existing material in order
to create a composite material that contains either an excess of electrons (an n-type
material) or an excess of positive holes (a p-type material).
dosimetry: Relating to the practice of monitoring the degree of radiation exposure
of individuals, whether externally or internally.
electrodeposition: Transfer of ions from solution to the surface of an electrode as
part of an electric current.
electromagnetic radiation: Energy which has both an electrical and magnetic
component; when characterized as a wave, identified by its wavelength or fre-
quency; when characterized as an energy bundle, identified by its energy in electron
volts.
electron: The negatively charged subatomic particle that is found in orbital spaces
near the nucleus of the atom.
electron capture (EC): A decay mode characterized by the capture of an atomic
electron from, most probably, an inner electron shell by the nucleus. Electron
426 Glossary

capture parallels positron emission. If the energy difference between the parent
and the daughter is less than 1.022 MeV, electron capture is the sole decay mode.
electron multiplier: A device with multiple dynodes in series to increase an
electron pulse for measurement. Electrons are accelerated as they move toward
each dynode so that multiple electrons are freed at the dynode per arriving
electron.
electron rest energy: Mass of an electron at rest expressed in units of energy:
0.511 MeV.
electron volt (eV): The unit of energy commonly used in quantifying nuclear
processes, the electron volt is defined as the work done by an electron when it falls
through a potential difference of 1 V stated another way, the quantity of energy
that must be added to a standard electron (whose charge equals 1.602 × 10−19 C)
to accelerate it through 1 V; potential difference. Mathematically, 1 eV = (1.602 ×
10−19 C)(1 V) = 1.602 × 10−19 J.
elutriant: A solution that removes a previously retained substance from a sorbent,
such as an ion-exchange column.
Emax : The maximum beta-particle energy for a beta-particle group.
excitation: Term used to describe the process in which an atom, molecule, or
nucleus is elevated to an excited state.
exoergic: Releasing energy.
expanded uncertainty: An uncertainty, U , chosen so that the interval y ± U about
the measured result y is believed to have a high, numerically defined, probability
of containing the true value of the measurand.
extractant: A reagent that implements the transfer of a substance from one solvent
to another.
extraction: Transfer of a substance from one solvent to another.
face velocity: The velocity at which air flows from the laboratory into the interior
of the hood, measured at the face of a hood. This velocity must be high enough
to prevent backspill of inhalants from the hood to the laboratory atmosphere, but
not so high that it interferes with the stability of analytical reagent containers and
equipment.
Fano factor: The observed variance relative to the calculated Poisson distribution
variance, as observed in the peak width in spectral analysis.
Faraday cup: A sensitive device for collecting a stream of electrons or ions and
measuring it in terms of current.
Federal Facility Compliance Act of 1992: Written to amend the Solid Waste Dis-
posal Act, in order to clarify provisions concerning the application of requirements
and sanctions to federal facilities.
fissile material: Any nuclide which is capable of fissioning, either spontaneously
or induced, although the term mostly refers to heavy elements such as uranium or
plutonium. The term “fissionable” may also be used.
 Glossary 427

fission: Process in which a nucleus is split into smaller nuclei. Fission may be
spontaneous or induced, symmetric or asymmetric.
flammable liquid: Liquids with a flashpoint below 100◦ F, also known as Class I
liquids. Class IA liquids have a flashpoint below 73◦ F and a boiling point below
100◦ F; Class IB liquids have a flashpoint below 73◦ F and a boiling point at or
above 100◦ F; Class IC liquids have a flashpoint at or above 73◦ F but below 100◦ F.
flashpoint: The minimum temperature at which a liquid gives off vapor within a
test vessel in sufficient concentration to form an ignitable mixture with air near
the surface of the liquid.
flux (as a rate): The rate of transfer of particles or energy across a given barrier.
flux (chemical): A substance added to the analyte, with the goal of facilitating its
dissolution.
footprint: The amount of floor space that an item requires.
Frisch grid: An ionization chamber used for alpha-particle spectroscopy. The
size of an electronic pulse from the chamber is proportional to the energy of
the alpha particle that produced the pulse. Because the energies of alpha particles
are characteristic of the radionuclides that emit them, the Frisch grid chamber is
able to distinguish between different radionuclides by measuring pulse height.
full width at half maximum (FWHM): Width of a peak at 1/2 of its maxi-
mum peak height. The FWHM measurement is a means of expressing the energy
resolution of a spectrometer; also full peak at half maximum (FPHM).
functional group: A specific group of atoms within a molecule that predictably
engages in characteristic chemical reactions.
fusion (nuclear): The process of two or more nuclei joining together to form a
heavier nucleus.
fusion (chemical) : The process of liquifying solid materials by heating.
gamma ray (γ ray): Electromagnetic radiation emitted as a result of a nuclear
transformation.
gas-filled detector: A chamber filled with gas that contains a cathode and an
anode connected by electric circuit to a recording device to measure count rate or
radiation exposure.
Gaussian: Normally distributed or shaped like the familiar “bell curve.”
Gaussian distribution: A normal distribution of a population that is described by
approximations of the Poisson distribution.
Gaussian peak: A spectral peak whose shape is Gaussian.
Geiger–Mueller (G-M) counter: A gas-filled ionization detector operating in the
Geiger–Mueller applied voltage region.
getter: Jargon; term for reactive metals used to scavenge for impurities.
grab sample: A sample collected by hand.
gravimetric: Measurement by weight.
428 Glossary

ground: A large conducting body such as the earth that is used as a common return
for an electric circuit.
half-life (t1/2 ): The time required for any given amount of a radionuclide to decay
to one-half its value. t1/2 = ln 2/decay constant.
HAZMAT: Term colloquially used to refer to hazardous materials and to functions
of agents of the Office of Hazardous Materials Safety. The stated mission of the
OHM is to “promulgate a national safety program that will minimize the risks to
life and property inherent in commercial transportation of hazardous materials.”
See website at http://hazmat.dot.gov (01/15/2006).
Henry’s Law: At contant temperature, the amount of a given gas dissolved in a
given liquid is directly proportional to the partial pressure of the gas when it is in
equilibrium with the liquid.
homogeneous: Uniform throughout.
homolog: In inorganic chemistry, an element in a similar position, such as a col-
umn, in the periodic table.
hoods: Devices designed to protect the analytical chemist from excess exposure
to inhalants by continuously removing laboratory air to the outside, after the air
has passed through a filtering system.
hopcalite: A catalyst for oxidizing hydrogen in hydrocarbons to water vapor.
hot particle: Jargon for an intensely radioactive particle.
hot swap: One of the features of a RAID system, whereby the drives are connected
to the controller. A “hot swappable” RAID drive is one that can be changed out if
it fails, without shutting down the system.
hot: Jargon; characterized by the presence of radiation or radionuclides.
hygroscopic: A characteristic of a material, wherein it absorbs water.
hyperpure: Extremely pure; describes purified germanium for a detector.
ICP/ MS: Inductively coupled plasma/mass spectrometer.
immiscible: Not able to be mixed together.
induced fission: The process in which neutrons are “fired” at a fissionable source
material to cause it to split.
ingrowth: Describes the accumulation of product atoms from radioactive parent
atoms as they decay.
input estimate: In a particular measurement, a measured or imported value of an
input quantity.
input quantity: In a mathematical model of measurement, any of the quantities
whose values are measured or imported and used to calculate the value of the
output quantity Y.
interlaboratory testing: Program in which a number of laboratories participate
to demonstrate competence.
 Glossary 429

internal conversion: A radioactive decay process in which an excited nucleus de-

excites by transferring its energy to an inner orbital electron, which is consequently
emitted. See conversion electron.
ionization (secondary): Ionization that results from the collision of ions or elec-
trons through their interaction with media to form a second generation of ions.
isobar: Two or more atoms with the same mass number ( A), but different atomic
numbers (Z ) and numbers of neutrons (N ), are said to be isobaric.
isobaric interferences: In mass spectrometry, interferences from extraneous atoms
or molecules that have masses very similar to that of the analyte atom.
isothermal: A process performed at a fixed energy.
isotope dilution: Addition and complete mixing of a second isotope with an isotope
of the same element to measure the yield of a reaction.
laminar flow boxes: Similar to hoods, these boxes are designed to filter in-
laboratory air continuously through the box, rather than venting to the outside,
to isolate particulate matter inside the box.
lanthanides or lanthanide series: The fourteen elements with atomic numbers
58–71, inclusive, in which electrons are added to the 4f inner electron shell.
law of propagation of uncertainty: The general equation, based on local approx-
imation of a mathematical model by a first-order Taylor polynomial, that may be
used to calculate the combined standard uncertainty of the result of a measurement.
linear accelerator: A linear accelerator uses alternating electrodes to drive charged
particles along a linear path toward a target.
loaded solvent: The solute-loaded result of an initial extraction. If several solutes
are contained in this extract, one may be singled out and removed through back
extraction.
Lucas cell: A cylindrical container for measuring radon gas that has a quartz
window for transmitting scintillations at one end, has all other internal walls coated
with ZnS(Ag) powder that responds with scintillations to alpha particles, and has
a stopcock for connection to a pump to create a vacuum and then admitting a gas
sample.
mass analyzer: A component of the mass spectrometer that separate ions based
on mass, momentum, or velocity.
mass defect: Energy is either required or released when nucleons bind together to
form a nucleus. The mass (or its equivalent energy) of the product nucleus then
differs from the additive component masses. The mass defect, calculated as the
total nucleon mass (M) minus the nuclear mass number ( A), can be viewed as a
reflection of the stablilty of the nucleus.
mass number (A): The sum of the number of protons and neutrons in the nucleus.
Material Safety Data Sheet (MSDS): A form that contains information about the
physical and chemical properties of a substance. The MSDS is intended to provide
personnel with safe methods of working with the substance in question.
430 Glossary

matrix: The components of a sample mixture, other than the analyte.

measurand: The “particular quantity subject to measurement” (ISO 1993).
metastable nucleus: Nuclear isomer or isomeric state of an atom, wherein a proton
or neutron in the nucleus is excited. In this state, a change in the spin of the nucleus
is required before it can go to a lower state and release extra energy.
minimum detectable activity (MDA): A measure of the detection capability of
a radioanalytical measurement process, defined as an estimate of the smallest true
value of the activity (or massic activity, volumic activity, etc.), which gives a
specified probability 1 – β of detection. The definition of the MDA presupposes
that an appropriate detection criterion (i.e., the critical value), with a specified
Type I error rate α, has been defined. It is a mistake to use the MDA itself as the
detection criterion.
momentum: Mass of an object multiplied by the velocity of the object; must be
conserved for interacting processes.
mono-energetic: Particles of a single energy, as in alpha particles.
Monte Carlo simulation: A numerical analysis technique that randomly generates
values for uncertain variables. The simulation program then uses the results to
construct a model that approximates the solution to some physical or mathematical
problem.
muon: A subatomic particle.
National Institute of Standards and Technology (NIST): The federal organiza-
tion that is officially responsible for standards in the United States.
negatron: A negatively charged electron emitted during a nuclear transformation
or process.
network protocol: Designed to augment a secure channel by making it more se-
cure, a network protocol requires that all communications by that channel adhere to
a defined set of rules. The rules generally address such issues as data authentication
and error detection.
neutrino: A subatomic particle with zero charge, near-zero mass, and 1/2 spin.
The neutrino is emitted as a byproduct of positron particle emission, carrying away
the appropriate amount of energy and momentum to conserve these quantities.
neutron: Fundamental subatomic particle of the atom that has no charge and a
mass of approximately one dalton.
neutron activation: Process of activating stable nuclides by bombarding them
with neutrons, thus forming radionuclides.
neutron flux: Measure of the neutron population per given cross section of space
per unit time.
neutron moderator: A medium, such as deuterium or graphite, that slows down
the fast neutrons produced in a nuclear reactor to reduce their kinetic energy to the
thermal level. The thermal neutrons may then be absorbed by a new fuel atom and
continue the chain reaction.
 Glossary 431

n-type: Describes a material whose structure contains an excess of electrons by

which charge can move.
nuclear chemistry: The field of chemistry that studies and applies nuclear prop-
erties and reactions.
nuclear decay: See decay.
nuclear explosion: An explosion that results from nuclear processes by fission
and fusion.
nuclear isomer: Nuclides having the same mass number and atomic number, but
occupying different energy states.
nuclear reactions: Reactions taking place between the nucleus and an energetic
particle.
Nuclear Regulatory Commission (NRC): The U.S. Nuclear Regulatory Com-
mission is an independent agency established by the Energy Reorganization Act of
1974 to regulate civilian use of nuclear materials. Online at http://www.nrc.gov/.
nuclear transformation: Transformation of the nuclear state as a result of one or
more nuclear processes.
nucleon: A term referring to either or both protons and neutrons.
nucleon number: The number of nucleons in the nucleus.
nucleus: The positively charged part of an atom in which most of the mass of the
atom resides. The nucleus is composed of neutrons and protons and is the location
of nuclear processes.
nuclide: Term given to denote any atomic species characterized by a specific
atomic number (Z ) and mass number (A). Nuclear isomers may also be recognized
as distinct nuclides, provided they have half lives long enough to be observed, e.g.,
234
Pa and 234m Pa are nuclides distinct from each other.
output estimate: In a particular measurement, the calculated value of the output
quantity.
output quantity: In a mathematical model of measurement, the quantity Y whose
value is calculated from measured or imported values of the input quantities.
overvoltage: Voltage required for electrodeposition of a substance in excess of
that predicted by half cell potentials due to extraneous reaction.
parent: In a decay chain, the parent is the immediate predecessor of some specified
radionuclide; its progeny are usually referred to as daughters.
personal protective equipment (PPE): Any equipment worn in order to protect
the wearer from the effects of an accident, whether chemical or radiological.
phosphor: A material that emits light (scintillates).
photomultiplier tube (PMT): Electronic tube that amplifies signal from inci-
dent light by multiplying (cascading) electrons, with more electrons from each
successive plate.
photodisintegration: A nuclear reaction initiated by an energetic gamma ray.
432 Glossary

photoelectric effect: The emission of a free electron from an atom that interacts
with and totally absorbs a gamma ray.
photopeak: In gamma-ray spectral analysis, the photoelectric interaction peak,
but more commonly defined as the full-energy peak.
planchet: A metal disk on which a sample is deposited or fixed in preparation for
counting.
positron: Positively charged electron emitted in a nuclear transformation or pro-
cess.
precipitate: The solid product of a chemical separation, left over after the super-
natant liquid from which it came is removed.
principal quantum number (n): The energy of an electron in an atom is primarily
dependent on this quantum number, as is the size of the atomic orbital in which it
is contained. The greater the value of n for an electron, the higher the energy of
that electron. The principal quantum number may have any positive integer value.
private key: An algorithm that allows a group or individual to enter a key agree-
ment, wherein all communications will require that key before proceeding.
process stream: A part of an industrial process that transfers materials.
propagation of uncertainty: Mathematical operation of combining standard un-
certainties (and estimated covariances) of input estimates to obtain the combined
standard uncertainty of the output estimate.
proportional counter: A gas-filled ionization detector that operates in the propor-
tional region of the applied voltage, in which the output pulse energy is proportional
to the deposited energy.
proton: Fundamental subatomic particle of the atom that has a charge of +1 with
a mass of approximately one dalton.
public key: A form of encryption that allows users to communicate securely
without prior possession of a secret key.
p-type: Describes a material whose structure contains positive holes through which
charge can move.
quality assurance (QA): A program that systematically monitors and evaluates
laboratory operation to ensure that sample analysis meets standards of quality.
quality assurance plan (QAP): A planning tool that allows for the comprehensive
management of laboratory operation to ensure that standards of quality are met.
quality assurance project plan (QAPP): A planning tool that allows for the
comprehensive management of a data collection activity (DCA), from sample
collection through data reporting, to ensure that standards of quality are met.
quality control (QC): System for ensuring maintenance of proper standards in
the laboratory by use of standard and comparison samples.
quenching: In a scintillation detector, to extinguish light emanation from the
sample, to a degree that depends on the quenching agent, thus reducing the
 Glossary 433

measurement of radionuclide content. In a gas-filled detector, to terminate ion-

ization promptly after it has been detected in order to avoid extended pulse
tailing.
radioactivity: The property of an unstable nuclide that is characterized by the
emission of particles or radiation from its atomic nucleus.
radioanalytical chemistry: Application of techniques, instrumentation and
knowledge from radiochemistry and analytical chemistry for the quantitative anal-
ysis of a wide range of concentrations of both radioactive and non-radioactive
species. Applicable to investigations ranging from geochemistry and age dating
to diagnostic and therapeutic nuclear medicine procedures as well as to routine
assay of radioactivity in the environment and surveillance of clandestine nuclear
activities and terrorist threats.
radiochemistry: The subdiscipline of chemistry that studies radioactive materials
and applies them in chemical processes.
radiocolloidal: Colloid-like behavior of a radionuclide at very low concentration
in solution.
radioelements: Radionuclides of the same element.
radionuclide: Nuclide that is unstable and will de-excite spontaneously.
RAID drives: Redundant Array of Independent Discs. A RAID system is a col-
lection of hard drives joined together, for goals that can be achieved. See online
descriptions at http://bugclub.org/beginners/hardware/raid.html (01/15/2006).
range: The distance a particle travels through a specified material.
Raoult’s Law: The vapor pressure of each component in an ideal solution is
dependent on both the partial pressure and the mole fraction of each component
in the solution.
recoil energy: Kinetic energy carried by the nucleus after it emits a particle.
recoil nucleus: Nucleus that results from the decay of a radionuclide. Also called
the product nucleus, or daughter nucleus. Recoil balances the momentum associ-
ated with the emission of particles or radiation.
refractory: A solid that is stable and heat resistant at elevated temperatures.
remote access: The communication between a local and remote computer over a
data link. Access to the data link is invariably limited to those with authorization.
resin: For ion-exchange processes, an organic structure for ion-exchange sites.
resonance structures: Two or more representations of the electronic structure of
a molecule. The actual structure of the molecule may be viewed as a hybrid of all
these structures. The more structures, the greater the stability of the molecule.
scavenger: A substance that reacts with (or otherwise removes) a trace component.
scintillation cocktail: A solution that consists of a scintillating substance and an
organic solvent, which serves as the medium for counting a radionuclide sample
in a liquid scintillation counter.
434 Glossary

sector field analyzer: A spectrometer that separates ions by the curvature of their
flight path, for which the radius is inversely proportional to the magnetic field and
directly proportional to the square root of the mass to charge ratio.
secular equilibrium: A case where the activity (disintegration rate) of the parent
and progeny becomes equal after a time. For this type of equilibrium, the half-life
of the parent is at least 10 times greater than that of the daughter.
secure shell (ssh): A set of standards and accompanying network protocol that
allows for the establishment of a secure channel between a local computer and one
located remotely.
sequential analysis: Serial chemical analysis of several substances in the same
solution.
shell (as in K, L, etc): The letter designation of the principal quantum number, n.
Orbitals of the same n value are referred to as belonging to the same shell. Several
subshells may be contained in a given shell; subshell designation depends on other
quantum numbers
slurry: A watery mixture of insoluble material.
solubility product (Ksp ): Mathematically, the product of the thermodynamic ac-
tivities of the ionic components of a substance precipitated from a solution at
equilibrium.
source: In the context of radionuclide analysis, a sample prepared for radionuclide
measurement, i.e., the source of the radionuclides that are being measured.
specific activity: For a specified isotope, or mixture of isotopes, the activity (dis-
integration rate) of a material divided by the mass of the element.
spallation: To break off. In nuclear physics, the term refers to the emission of a
large number of neutrons by a heavy radionuclide after it has been bombarded by
charged particles.
spike: See tracer.
spiking: Jargon; addition to a sample of a tracer for the radionuclide of interest.
spontaneous fission: Natural decay process found in elements with Z greater than
or equal to 90 (although it is unclear whether Th (Z = 90) can fission sponta-
neously).
stable nuclide: A nuclide that is not radioactive.
standard method: A procedure that has been selected by testing and agreement
among skilled professionals.
standard reference material (SRM) or standard radioactive material (SRM):
Material, often in liquid form, that is certified to have a known concentration or
decay rate. SRM is used in instrument calibration and as primary standards for
laboratory measurements.
standard uncertainty: Uncertainty of the result of a measurement expressed as
a standard deviation (sometimes called a “one-sigma” uncertainty). The standard
uncertainty of a measured value x may be denoted by u(x).
 Glossary 435

stakeholder: In the context presented herein, a person or group that has an interest
in the results produced by a laboratory or facility.
state-of-health: During the lifetime of an instrument, its performance or “health”
tends to deteriorate. The degree of this decline can be inferred from measurements,
specified for a particular system, that define its well being. Parameters may include
battery power, calibration outputs and various hardware checks.
stopping power: The differential energy loss per travel distance, (dE/dx) of a
particle.
STP (standard temperature and pressure): Zero degrees Celsius (273◦ Abso-
lute) and one atmosphere pressure.
subatomic particle: Any fundamental particle that is smaller than the atom.
symmetric fission: Fission process in which product yields decrease on both sides
of an intermediate mass that is produced with maximum yield.
Szilard–Chalmers effect: The rupture of the chemical bond between an atom
and the molecule of which the atom was a component, as a result of a nuclear
transformation of that atom.
thermoluminescent detector (TLD): Thermoluminescent material in powder
form or small chips is placed in a badge for radiation monitoring. After a specified
period of time, the badge is collected and “read” by heating it to emit light that is
proportional to the energy deposition of the detected radiation.
thin target: Term applied to materials that are irradiated but do not significantly
attenuate the projectile beam or flux.
traceability: Documentation for a standard that allows it to be historically linked
to the laboratory of its formulation and testing, notably at NIST.
tracer: A substance that is added to a material to observe the behavior of one of
its components.
transactinides: All elements with atomic number greater than lawrencium (103),
the last of the actinide series.
transient equilibrium: A case where the ratio of the daughter activity to the parent
activity becomes constant with a value that exceeds 1.0 after several half-lives. The
half life of the parent is longer than that of the daughter.
transport velocity: The velocity at which air is flushed from the hood to the outside
air. This velocity depends on fan speed and ductwork dimensions and is important
to assure that airborne particles are transported from the hood and not deposited
in the ductwork.
Type A evaluation: “Method of evaluation of uncertainty by the statistical analysis
of series of observations” (ISO 1995).
Type B evaluation: “Method of evaluation of uncertainty by means other than the
statistical analysis of series of observations” (ISO 1995).
uncertainty propagation: See propagation of uncertainty.
436 Glossary

uncertainty: A “parameter, associated with the result of a measurement, that

characterizes the dispersion of the values that could reasonably be attributed to the
measurand” (ISO 1993, 1995).
VOC: Volatile organic chemicals that are emitted from certain liquids and solids.
volatilization: The conversion of a solid or liquid to a gas or vapor by application
of heat, by reducing pressure, by chemical reaction or by a combination of these
processes.
X-ray: One possible consequence of an internal conversion. Electromagnetic ra-
diation that is emitted when an inner shell electron of an atom is removed and an
electron from an orbital of higher energy drops into the position. It has a wavelength
of 10 pm to 10 nm or an energy from electron volts to kiloelectron volts.
yield: The fraction of a substance that is recovered after separations are performed.
zeolite: One of hydrous aluminum silicates used as ion-exchange material.
References

ACGIH 2001. Air Sampling Instruments for Evaluation of Atmospheric Contaminants, 9th
ed. Cincinnati OH: American Conference of Government Industrial Hygienists, Inc.
ACS 1990. American Chemical Society. Informing Workers of Chemical Hazards: The
OSHA Hazard Communication Standard. 2nd ed. Washington, DC: ACS.
ACS 1998. American Chemical Society. Living with the Laboratory Standard: A Guide for
Chemical Hygiene Officers. Washington, DC: ACS.
ACS 2000. American Chemical Society. Safety Audit/Inspection Manual. Washington, DC:
ACS.
ACS 2003. American Chemical Society. Safety in Academic Chemistry Laboratories. 7th
ed. Washington, DC: ACS.
Adloff, J.-P. and Guillaumont, R. 1993. Fundamentals of Radiochemistry. pp. 327–352.
Boca Raton, FL: CRC Press.
Adriaens, A. G., Fassett, J. D., Kelly, W. R., Simons, D. S., and Adams, F. C. 1992. De-
termination of Uranium and Thorium Concentrations in Soils—Comparison of Isotope-
Dilution Secondary Ion Mass-Spectrometry and Isotope-Dilution Thermal Ionization
Mass-Spectrometry. Anal Chem 64(23), 2945–2950.
AIHA 1995. American Industrial Hygiene Association. Laboratory Chemical Hygiene: An
AIHA Protocol Guide. Fairfax, VA: AIHA Press.
Alaimo, R. J. and Fivizzani, K. P. November/December 1996. Qualifications and training
of chemical hygiene officers. Chemical Health and Safety 3(6), 10–13.
Aldstadt, J. H., Kuo, J. M., Smith, L. L., and Erickson, M. D. 1996. Determination of
uranium by flow injection inductively coupled plasma mass spectrometry. Anal Chim
Acta 319(1–2), 135–143.
Alonso, J. I. G., Babelot, J. F., Glatz, J. P., Cromboom, O., and Koch, L. 1993. Applications
of a glove-box ICP-MS for the analysis of nuclear- materials. Radioc Acta 62(1–2),
71–79.
Alonso, J. I. G., Sena, F., Arbore, P., Betti, M., and Koch, L. 1995. Determination of fission-
prroducts and actinides in spent nuclear-fuels by isotope-dilution ion chromatography
inductively-coupled plasma-sass spectrometry. J Anal Atom Spectromy 10(5), 381–393.
Altshuler, B. and Pasternack, B. 1963. Statistical measures of the lower limit of detection
of a radioactivity counter. Health Phys 9, 293–298.
Amphlett, C. B. 1964. Inorganic Ion Exchangers. New York, NY: Elsevier.
Anders, E. 1960. The Radiochemistry of Technetium. National Research Council Report
NAS-NS 3021. Washington, DC: National Research Council.

437
438 References

ANSI 1995. American National Standards Institute, N42.22-1995. Traceability of Ra-

dioactive Sources to NIST and Associated Instrument Quality Control. Washington, DC:
ANSI.
ANSI 2005. American National Standards Institute. Online at http://www.ansi.org/. N42
standards may be found by clicking “eStandards Store” then searching for N42. Viewed
on September 20, 2005.
ANSI/IEEE 1995. American National Standards Institute—Institute of Electrical and Elec-
tronics Engineers Standard N42.22-1995. American National Standard Traceability of
Radioactive Sources to NIST and Associated Instrument Quality Control. Online at
www.ansi.org. Viewed on September 20, 2005.
APHA 1972. American Public Health Association, Intersociety Committee for a Manual of
Methods for Ambient Air Sampling and Analysis. Methods of Air Sampling and Analysis.
Washington, DC: APHA.
Armentrout, P. B. 2004. Fundamentals of ion-molecule chemistry. J Anal Atom Spectrom
19(5), 571–580.
ASME 1989. American Society of Mechanical Engineers, NQA-1. QualityAassurance Pro-
gram Requirements for Nuclear Facilities. New York, NY: ASME.
ASQC 1995. American Society for Quality Control, ANSI/ASQC E4-1994. Specifications
and Guidelines for Quality Systems for Environmental Data Collection and Environmen-
tal Technology Programs. Milwaukee, WI: ASQC.
ASTM 2005. American Society for Testing and Materials. Standards may be found online
at http://astm.org/. Click “Technical Committees” then search for the standard by the
committee or sub-committee that developed it. Viewed on September 20, 2005.
Bachmann, K. 1982. Separation of trace elements in solid samples by formation of volatile
inorganic compounds. Talanta 29, 1–29.
Barber, R. C., Greenwood, N. N., Hrynkiewicz, A. Z., Jeannin, Y. P., Lefort, M., Sakai, M.,
Ulehla, I., Wapstra, A. H., and Wilkinson, D. H. 1991. Criteria that must be satisfied for
the discovery of a new chemical element to be recognized. Pure Appl Chem. 63, 879–886.
Barber, R. C., Greenwood, N. N., Hrynkiewicz, A. Z., Jeannin, Y. P., Lefort, M., Sakai, M.,
Ulehla, I., Wapstra, A. H., and Wilkinson, D. H. 1992. Discovery of the transfermium
elements. Prog Part Nucl Phys 29, 453–530.
Barber, R. C., Greenwood, N. N., Hrynkiewicz, A. Z., Jeannin, Y. P., Lefort, M., Sakai, M.,
Ulehla, I., Wapstra, A. H., and Wilkinson, D. H. 1993. Naming of new elements. Pure
Appl Chem 65, 1757–1813.
Barshick, C. M., Duckworth, D. C., and Smith, D. H. 2000. Inorganic mass spectrometry:
fundamentals and applications in Practical Spectroscopy. Vol. 23, pp. 512. New York,
NY: Marcel Dekker.
Baude, S., Broekaert, J. A. C., Delfosse, D., Jakubowski, N., Fuechtjohann, L., Orellana-
Velado, N. G., Pereiro, R., and Sanz-Medel, A. 2000. Glow discharge atomic spectrometry
for the analysis of environmental samples—a review. J Anal Atom Spectrom 15(11),
1516–1525.
Baumgartner, F. and Kim, M. A. 1990. Isotope effects in the equilibrium and non-equilibrium
vaporization of tritiated water and ice. Appl Radiat Isot 41, 395–399.
Beals, D. M. and Hayes, D. W. 1995. Technetium-99, iodine-129 and tritium in the waters
of the Savannah River Site. Sci Total Environ 173(1–6), 101–115.
Beasley, T., Cooper, L. W., Grebmeier, J. M., Aagaard, K., Kelley, J. M., and Kilius,
L. R. 1998a. Np-237/I-129 atom ratios in the Arctic Ocean: Has Np-237 from western
European and Russian fuel reprocessing facilities entered the Arctic Ocean? J Environ
Radioactiv 39(3), 255–277.
 References 439

Beasley, T. M., Kelley, J. M., Maiti, T. C., and Bond, L. A. 1998b. Np-237/Pu-239 atom
ratios in integrated global fallout: a reassessment of the production of Np-237. J Environ
Radioactiv 38(2), 133–146.
Beasley, T. M., Kelley, J. M., Orlandini, K. A., Bond, L. A., Aarkrog, A., Trapeznikov,
A. P., and Pozolotina,V. N. 1998c. Isotopic Pu, U, and Np signatures in soils from
Semipalatinsk-21, Kazakh Republic and the Southern Urals, Russia. J Environ Radioactiv
39(2), 215–230.
Becker, J. S. 2002. State-of-the-art and progress in precise and accurate isotope ratio mea-
surements by ICP-MS and LA-ICP-MS—plenary lecture. J Anal Atom Spectrom 17(9),
1172–1185.
Becker, J. S. 2003. Mass spectrometry of long-lived radionuclides. Spectrochim Acta B
58(10), 1757–1784.
Becker, J. S. and Dietze, H. J. 1997. Double-focusing sector field inductively coupled
plasma mass spectrometry for highly sensitive multi-element and isotopic analysis—
Invited lecture. J Anal Atom Spectrom 12(9), 881–889.
Becker, J. S. and Dietze, H. J. 1998. Inorganic trace analysis by mass spectrometry. Spec-
trochim Acta B 53(11), 1475–1506.
Becker, J. S. and Dietze, H. J. 2000. Precise and accurate isotope ratio measurements by
ICP-MS. Fresen J Anal Chem 368(1), 23–30.
Becker, J. S., Pickhardt, C., and Dietze, H. J. 2000. Laser ablation inductively coupled
plasma mass spectrometry for the trace, ultratrace and isotope analysis of long-lived
radionuclides in solid samples. Int J Mass Spectrom 202(1–3), 283–297.
Becker, J. S., Burow, M., Boulyga, S. F., Pickhardt, C., Hille, R., and Ostapczuk, P. 2002.
ICP-MS determination of uranium and thorium concentrations and U-235/U-238 isotope
ratios at trace and ultratrace levels in urine. Atom Spectrosc 23(6), 177–182.
Becker, J. S., Burow, M., Zoriy, M. V., Pickhardt, C., Ostapczuk, P., and Hille, R. 2004a.
Determination of uranium and thorium at trace and ultratrace levels in urine by laser
ablation ICP-MS. Atom Spectrosc 25(5), 197–202.
Becker, J. S., Zoriy, M., Halicz, L., Teplyakov, N., Muller, C., Segal, I., Pickhardt, C.,
and Platzner, I. T. 2004b. Environmental monitoring of plutonium at ultratrace level
in natural water (Sea of Galilee-Israel) by ICP-SFMS and MC-ICP-MS. J Anal Atom
Spectrom 19(9), 1257–1261.
Bellis, D., Ma, R., Bramall, N., McLeod, C. W., Chapman, N., and Satake, K. 2001. Airborne
uranium contamination—as revealed through elemental and isotopic analysis of tree bark.
Environ Pollut 114(3), 383–387.
Belloni, J., Haissinsky, M., and Salama, H. N. 1959. On the adsorption of some fission
products on various surfaces. J Phys Chem 63, 881–887.
Bemis, C. E., Jr., Dittner, P. F., Silva, R. J., Hahn, R. L., Tarrant, J. R., Hunt, L. D., and
Hensley, D. C. 1977. Production, L x-ray identification and decay of the nuclide (260)105.
Phys Rev C 16, 1146–1157.
Bemis, C. E., Jr., Silva, R. J., Hensley, D. C., Keller, O. L., Jr., Tarrant, J. R., Hunt, L. D.,
Dittner, P. F., Hahn, R. L., and Goodman, C. D. 1973. X-Ray identification of element 104
Phys Rev Lett 31, 647–650.
Berkovits, D., Feldstein, H., Ghelberg, S., Hershkowitz, A., Navon, E., and Paul, M. 2000.
U-236 in uranium minerals and standards.Nuc Instrum Meth B 172, 372–376.
Betti, M. 1996. Use of a direct current glow discharge mass spectrometer for the chemical
characterization of samples of nuclear concern. J Anal Atom Spectrom 11(9), 855–860.
Bings, N. H., Bogaerts, A., and Broekaert, J. A. C. 2002. Atomic spectroscopy. Anal Chem
74(12), 2691–2711.
440 References

Blanchard, R. L., Kahn, B., and Birkhoff, R. D. 1960. The preparation of thin, uniform,
radioactive sources by surface adsorption and electrodeposition. Health Phys 2, 246–255.
Bland, C. J. 1984. Tables of the geometrical factor for various source-detector configura-
tions. Nuc Instrum Methods 223, 2–3, 602–606.
Blue, T. E. and Jarzemba, M. S. 1992. A model for radon gas adsorption on charcoal for
open-faced canisters in an active environment. Health Phys 63(2), 226–232.
Bock, R. 1979. A Handbook of Decomposition Methods in Analytical Chemistry. New York:
John Wiley.
Boni, A. L. 1966. Rapid ion exchange analysis of radiocesium in milk, urine, seawater and
environmental samples. Anal Chem 38, 89–92.
Boulyga, S. F. and Becker, J. S. 2001. Determination of uranium isotopic composition and
U-236 content of soil samples and hot particles using inductively coupled plasma mass
spectrometry. Fresen J Anal Chem 370(5), 612–617.
Boulyga, S. F. and Becker, J. S. 2002. Isotopic analysis of uranium and plutonium using
ICP-MS and estimation of burn-up of spent uranium in contaminated environmental
samples. J Anal Atom Spectrom 17(9), 1143–1147.
Boulyga, S. F., Becker, J. S., Matusevitch, J. L., and Dietze, H. J. 2000. Isotope ratio
measurements of spent reactor uranium in environmental samples by using inductively
coupled plasma mass spectrometry. Int J Mass Spectrom 203(1–3), 143–154.
Boulyga, S. F., Desideri, D., Meli, M. A., Testa, C., and Becker, J. S. 2003. Plutonium and
americium determination in mosses by laser ablation ICP-MS combined with isotope
dilution technique. Int J Mass Spectrom 226(3), 329–339.
Bowyer, T. W., Abel, K. H., Hubbard, C. W., Panisko, M. E., Reeder, P. L., Thompson,
R. C., and Warner, R. A. 1999. Field testing of collection and measurement of radioxenon
for the comprehensive test ban treaty. J Radioanal Nuc Ch 240, 109–122.
B riesmeister, J. F. 1990. MCNP-A general Monte Carlo code for neutron and photon
transport. Version 4.2. Technical Report LA-7396-M, Los Alamos National Laboratory.
Brown, T. A., Marchetti, A. A., Martinelli, R. E., Cox, C. C., Knezovich, J. P., and Hamilton,
T. F. 2004. Actinide measurements by accelerator mass spectrometry at Lawrence Liv-
ermore National Laboratory. Nuc Instrum Meth B 223–224, 788–795.
Brown, E., Skougstad, M. W., and Fishman, M. J. 1970. Methods for collection and analysis
of water samples for dissolved minerals and gases in Techniques of Water-Resources
Investigations, Book 5. Chapter A1. Reston, VA: US Geological Survey.
Brown, R. M., Workman, W. J., and Kotzer, T. G. 1993. Tritium program at chalk river
laboratories. Tritium monitoring in the environment in Transactions of the Ameri-
can Nuclear Society. Paper 1, pp. 20. Vol. 69. LaGrange Park, IL: American Nuclear
Society.
Budnitz, R. J., Nero, A. V., Murphy, D. J., and Graven, R. 1983. Instrumentation for Envi-
ronmental Monitoring. Vol. 1, Radiation. 2nd ed. New York, NY: John Wiley.
Burnett, W. C., Schultz, M. K., Corbett, D. R., Horwitz, E. P., Chiarizia, R., Dietz, M.,
Thakkar, A., and Fern, M. 1997. Preconcentration of actinide elements from soils and
large volume water samples using extraction chromatography. J Radioanal Nucl Chem
226, 121–127.
Byrnes, M. E. 2001. Sampling and Surveying Radiological Environments. Boca Raton, FL:
Lewis Publishers, CRC Press.
Carlton, W. H., Bauer, L. R., Evans, A. G., Geary, L. A., Murphy, C. E., Jr., Pinder, J. E., and
Strom, R. N. 1992. Cesium in the Savannah River Site Environment. US Department of
Energy Report WSRC-RP-92-250. Aiken, SC: Westinghouse Savannah River Company.
Cember, H. 1996. Introduction to Health Physics. New York, NY: McGraw-Hill.
 References 441

Cerrai, E. and Ghersini, G. 1970. Reversed-phase extraction chromatography in inorganic

chemistry, pp 3-189 in: Advances in Chromatography. Vol. 9. Giddings, J. C. and Keller,
R. A., eds. New York, NY: Marcel Dekker.
Cesarano, C., Pugnetti, G., and Testa, C. 1965. Separation of cesium-137 from fission
products by means of a Kel-F column supporting tetraphenylboron. J Chromatog 19,
589–593.
CFR 2004. Code of Federal Regulations. CFR is online at http://www.gpo.gov/nara/cfr/cfr-
table-search.html#page1. (Viewed on December 2005)
CGIH 2001. Air Sampling Instruments for Evaluation of Atmospheric Contaminants, 9th
ed. Cincinnati, OH: American Conference of Governmental Industrial Hygienists, Inc.
Chartier, F., Aubert, M., and Pilier, M. 1999. Determination of Am and Cm in spent nuclear
fuels by isotope dilution inductively coupled plasma mass spectrometry and isotope di-
lution thermal ionization mass spectrometry after separation by high-performance liquid
chromatography. Fresen J Anal Chem 364(4), 320–327.
Chatt, J. 1979. Recommendations for the naming of elements of atomic numbers greater
than 100. Pure Appl Chem. 51, 381–384.
Chemweb 2005. Online through Champlain St. Lawrence College at http://www.slc.qc.ca/
www/Chemweb/images/ET/sda.gif. Viewed on January 5, 2006.
Chiappini, R., Taillade, J. M., and Brebion, S. 1996. Development of a high-sensitivity in-
ductively coupled plasma mass spectrometer for actinide measurement in the femtogram
range. J Anal Atom Spectrom 11(7), 497–503.
Chiarappa-Zucca, M. L., Dingley, K. H., Roberts, M. L., Velsko, C. A., and Love, A. H.
2002. Sample preparation for quantitation of tritium by accelerator mass spectrometry.
Anal Chem 74(24), 6285–6290.
Chieco, N. A. ed., 1997. The Environmental Measurements Laboratory Procedures Manual.
U. S. Department of Energy Report HASL-300. New York, NY: Environmental Mea-
surements Laboratory. Online at: http://www.eml.doe.gov/publications/Procman.cfm.
Viewed on April 2, 2005.
Choppin, G. R., Liljenzin, J-O., and Rydberg, J. 1995. Radiochemistry and Nuclear Chem-
istry. 3rd ed. Oxford, London: Butterworth-Heinemann.
CNIC 1997. CNIC/IUPAC compromise recommendation approved August 30, 1997,
Geneva, Switzerland. Pure Appl Chem 69, 2471–2473.
Coleman, G. H. 1965. The Radiochemistry of Plutonium. National Research Council Report
NAS-NS 3058. Washington, DC: National Research Council.
Coomber, D. I. 1975. Radiochemical Methods in Analysis. NewYork, NY: Plenum Press.
Coplan, M. A., Moore, J. H., and Hoffman, R. A. 1984. Double-focusing ion mass-
spectrometer of cylindrical symmetry. Rev Sci Instrum 55(4), 537–541.
Coplen, T. B. 2001. Atomic weights of the elements 1999. J Phys Chem Ref Data 30(3),
701–712.
Corish, J. and Rosenblatt, G. M. 2003. Name and symbol of the element with atomic number
110. Pure Appl Chem 75, 1613–1615.
Corish, J. and Rosenblatt, G. M. 2004. Name and symbol of the element with atomic number
111. Pure Appl Chem 76, 2101–2103.
Corley, J. P., Denham, D. H., Michels, D. E., Olsen, A. R., and Waite, D. A. 1977. A Guide
for Radiological Surveillance at ERDA Installations. Energy Research and Development
Administration, Division of Safety, Standards, & Compliance. ERDA77-24. Washington,
DC: ERDA.
Corson, D. R., MacKenzie, K. R., Segrè, E. 1940. Artificially radioactive element 85. Phys
Rev 58, 662.
442 References

Coryell, C. D. and Sugarman, N. 1951. Radiochemical studies: the fission products. National
Nuclear Energy Series. 3 books, Vol. 9. New York, NY: McGraw-Hill.
Costa-Fernandez, J. M., Bings, N. H., Leach, A. M., and Hieftje, G. M. 2000. Rapid simul-
taneous multielemental speciation by capillary electrophoresis coupled to inductively
coupled plasma time-of-flight mass spectrometry. J Anal Atom Spectrom 15(9), 1063–
1067.
Cristy, S. S. 2000. Secondary ion mass spectrometry in Inorganic Mass Spectrometry.
Vol. 23.. Barshick, C. M., Duckworth, D. C., and Smith, D. H., eds. pp. 159–221. New
York, NY: Marcel Dekker.
Curie, P. and Sklodowska Curie, M. 1898. Sur une substance Nouvelle Radio-active, con-
tenue dans la Pechblende. Compt. Rend. 127, 175.
Currie, L. A. 1968. Limits for qualitative detection and quantitative determination—
application to radiochemistry. Anal Chem 40, 586–593.
Curtiss, L. F. and Davis, F. J. 1943. A counting method for the determination of small
amounts of radium and radon. J Res Nat Bur Stand. 31, 181–195.
Dai, M. H., Buesseler, K. O., Kelley, J. M., Andrews, J. E., Pike, S., and Wacker, J. F. 2001.
Size-fractionated plutonium isotopes in a coastal environment. J Environ Radioactiv
53(1), 9–25.
Daly, N. R. 1960. Scintillation type mass spectrometer ion detector. Rev Sci Instrum 31(3),
264–267.
Danesi, P. R., Bleise, A., Burkart, W., Cabianca, T., Campbell, M. J., Makarewicz, M.,
Moreno, J., Tuniz, C., and Hotchkis, M. 2003. Isotopic composition and origin of uranium
and plutonium in selected soil samples collected in Kosovo. J Environ Radioactiv 64(2–
3), 121–131.
Daubert v. Merrell Dow Pharmaceuticals 1993. 509 U.S. 579.
de Laeter, J. R. 1996. The role of off-line mass spectrometry in nuclear fission. Mass
Spectrom Rev 15(4), 261–281.
de Laeter, J. R. 2001. Applications of Inorganic Mass Spectrometry. New York, NY: John
Wiley.
de las Heras, L. A., Bocci, F., Betti, M., and Actis-Dato, L. O. 2000. Comparison between the
use of direct current glow discharge mass spectrometry and inductively coupled plasma
quadrupole mass spectrometry for the analysis of trace elements in nuclear samples.
Fresen J Anal Chem 368(1), 95–102.
De, A. K., Khopkar, S. M., and Chalmers, R. A. 1970. Solvent Extraction of Metals. London:
Van Nostrand Reinhold Co.
Dean, J. A. 1992. Lange’s Handbook of Chemistry. 14th ed. Table 11.22, pp. 11–61. New
York: McGraw-Hill.
Dean, J. A. 1999. Lange’s Handbook of Chemistry. 15th ed. New York, NY: McGraw-Hill.
Delanghe, D., Bard, E., and Hamelin, B. 2002. New TIMS constraints on the uranium-238
and uranium-234 in seawaters from the main ocean basins and the Mediterranean Sea.
Mar Chem 80(1), 79–93.
Delmore, J. E., Appelhans, A. D., and Peterson, E. S. 1995. A rare-earth-oxide matrix for
emitting perrhenate anions. Int J Mass Spectrom Ion Processes 146, 15–20.
DeVoe, J. R. 1962. Application of Distillation Techniques To Radiochemical Separations.
National Research Council Report NAS-NS-3108. Washington, DC: NRC.
Deyl, Z., ed. 1979. Electrophoresis: a survey of techniques and applications, Part A. Tech-
niques. J Chromatogr Libr Vol. 18. New York, NY: Elsevier.
DHHS 1982. Department of Health and Human Services. DHHS evaluation of results
of environmental chemical testing by EPA in the vicinity of Love Canal—Implications
 References 443

for human health—Further considerations concerning habitability. Washington, DC:

DHHS.
Dienstbach, F. and Bachmann, K. 1980. Determination of plutonium in soil by gas chro-
matographic separation and alpha spectroscopy. Anal Chem 52, 62624.
DiNardi, S. R. ed. 2003. The Occupational Environment: Its Evaluation, Control and Man-
agement. Fairfax, VA: AIHA Press.
DOE 1987. Department of Energy Report DOE/EH-0053. The Environmental Survey Man-
ual, Vol. 4. Appendices E to K. Washington, DC: DOE.
DOE/EM 1997. U.S. Department of Energy/Office of Environmental Management Report
No. 0319. Linking Legacies: Connecting Cold War Nuclear Weapons Production Pro-
cesses to their Environmental Consequences. Washington, DC: DOE/EM.
Douglas, D. J. and French, J. B. 1981. Elemental analysis with a microwave-induced plasma-
quadrupole mass-spectrometer system. Anal Chem 53(1), 37–41.
Douglas, D. J. and Tanner, S. D. 1998. Fundamental considerations in ICP-MS in Inductively
Coupled Plasma Mass Spectrometry. Chapter 8. Montaser, A., ed. Dordrecht: Wiley-
VCH.
Duckworth, H. E., Barber, R. C., and Venkatasubramanian, V. S. 1986. Mass Spectroscopy.
Cambridge, MA: Cambridge University Press.
Düllmann, C. E., Eichler, B., Eichler, R., Gäggeler, H. W., Jost, D. T., Piguet, D., and Türler,
A. 2002a. IVO, a device for in situ volatilization and on-line detection of products from
heavy ion reactions. Nucl Instrum Meth A 479, 631–639.
Düllmann, C. E., Brüchle, W., Dressler, R., Eberhardt, K., Eichler, B., Eichler, R., Gäggeler,
H. W., Ginter, T. N., Glaus, F., Gregorich, K. E., Hoffman, D. C., Jäger, E., Jost, D. T.,
Kirbach, U. W., Lee, D. M., Nitsche, H., Patin, J. B., Pershina, V., Piguet, D., Qin, Z.,
Schädel, M., Schausten, B., Schimpf, E., Schött, H.-J., Soverna, S., Sudowe, R., Thörle,
P., Timokhin, S. N., Trautmann, N., Türler, A., Vahle, A., Wirth, G., Yakushev, A. B., and
Zielinski, P. M. 2002b. Chemical investigation of hassium (element 108). Nature 418,
859–862.
Durrant, S. F. 1999. Laser ablation-inductively coupled plasma mass spectrometry: achieve-
ments, problems, prospects. J Anal Atom Spectrom 14(9), 1385–1403.
Dushman, S. and Lafferty, J. M. 1962. Scientific Foundations of Vacuum Technique. New
York, NY: John Wiley.
Eaton, G. F., Hudson, G. B., and Moran, J. E. 2004. Tritium-helium-3 age-dating of ground-
water in the Livermore Valley of California.in Radioanalytical Methods in Interdisci-
plinary Research. Laue, C. A. and Nash, K. L., eds. pp. 235–245. Washington, DC:
ACS.
Egorov, O. B., O’Hara, M. J., Addleman, S. R., Grate, J. W. 2003. Automation of radiochem-
ical analysis: from groundwater monitoring to nuclear waste analysis in Radioanalytical
Methods in Interdisciplinary Research: Fundamental to Cutting Edge Applications. ASC
Symposium Series 868. Washington, DC: ACS.
Efurd, D. W., Rokop, D. J., Aguilar, R. D., Roensch, F. R., Banar, J. C., and Perrin, R. E.
1995. Identification and quantification of the source terms for uranium in surface waters
collected at the rocky-flats facility. Int J Mass Spectrom Ion Processes 146, 109–117.
Ehmann, W. D. and Vance, D. E. 1991. Radiochemistry and Nuclear Methods of Analysis.
New York, NY: Wiley-Interscience.
Eichholz, G. G. and Poston, J. W. 1979. Principles of Nuclear Radiation Detection. Ann
Arbor, MI: Ann Arbor Science Publishers.
Eichholz, G. G., Nagel, A. E., and Hughes, R. B. 1965. Adsorption of ions in dilute aqueous
solutions on glass and plastic surfaces. Anal Chem 37, 863–868.
444 References

Eichler, B., Türler, A., and Gäggeler, H. W. 1999. Thermochemical characterization of

seaborgium compounds in gas adsorption chromatography. J Phys Chem A 103, 9296–
9306.
Eichler, R., Brüchle, W., Dressler, C. E., Düllmann, C. E., Eichler, B., Gäggeler, H. W.,
Gregorich, K. E., Hoffman, D. C., Hübener, S., Jost, D. T., Kirbach, U. W., Laue, C. A.,
Lavanchy, V. M., Nitsche, H., Patin, J. B., Piguet, D., Schädel, M., Shaughnessy, D. A.,
Strellis, D. A., Taut, S., Tobler, L., Tsyganov, Y. S., Türler, A., Vahle, A., Wilk, P. A.,
and Yakushev, A. B. 2000. Chemical characterization of bohrium (element 107). Nature
(Letters) 407, 63–65.
Eiden, G. C., Anderson, J. E., and Nogar, N. S. 1994. Resonant laser-ablation—
semiquantitative aspects and threshold effects. Microchem J 50(3), 289–300.
Eisenbud, M. 1987. Environmental Radioactivity. 3rd ed. New York, NY: Academic Press.
Encinar, J. R., Alonso, J. I. G., Sanz-Medel, A., Main, S., and Turner, P. J. 2001. A compari-
son between quadrupole, double focusing and multicollector ICP-MS instruments Part I.
Evaluation of total combined uncertainty for lead isotope ratio measurements. J. Anal
Atom Spectrom 16, 315–321.
England, T. R. and Rider, B. F. 1994. Evaluation and compilation of fission product yields
1993. Los Alamos National Laboratory Report LA-UR-94-3106. ENDF (Evaluated
Nuclear Data File) 349. Viewed on January, 2006.
EPA 1978.US Environmental Protection Agency Report EPA/EERF-78-1. Radon in Water
Sampling Program. Washington, DC: EPA.
EPA 1980a. US Environmental Protection Agency Report 520/1-80-012. (Also Health
Physics Committee Report HPSR-1). Upprading Environmental Radiation Data.
Washington, DC: EPA.
EPA 1980b. US Environmental Protection Agency, Office of Monitoring Systems and
Quality Assurance Publication QAMS 005/80. Interim Guidelines and Specifications
for Preparing Quality Assurance Project Plans. Washington, DC: EPA.
EPA 1984. US Environmental Protection Agency Report EPA 520/5-884-006. Eastern
Environmental Radiation Facility Radiochemistry Procedures Manual. Washington,
DC: EPA. An index to the procedures contained in this manual is available online at
www.epa.gov/region01/oarm/testmeth.pdf.
EPA 1986. US Environmental Protection Agency Report EPA/530/SW-86/055 (OSWER-
9950.1). RCRA Ground Water Monitoring Technical Enforcement Guidance Document.
Washington, DC: EPA.
EPA 2000a. US Environmental Protection Agency Manual 5360 A1. EPA Quality Manual
for Environmental Programs. Washington, DC: EPA. This document can be found online
at http://www.epa.gov/QUALITY/qa docs.html.
EPA 2000b. US Environmental Protection Agency, Office of Environmental Information
Report EPA QA/G-4. Guidance for the Data Quality Objectives Process. Washington,
DC: EPA. This document can be found online at http://www.epa.gov/quality1/qs-docs/g4-
final.pdf.
EPA 2000c. US Environmental Protection Agency Report 402-R-97-016 (also NRC Re-
port NUREG-1575). Multi-agency Radiation Survey and Site Investigation Manual
(MARSSIM). Washington, DC: EPA. May be downloaded from the EPA website at
http://www.epa.gov/radiation/marssim/. (Viewed on January 2006)
EPA 2004. US Environmental Protection Agency, Report 402-B-04-001 (also NRC Report
NUREG-1576, NTIS PB2004-105421). Multi-Agency Radiological Laboratory Analyt-
ical Protocols Manual (MARLAP). Washington, DC: EPA.
 References 445

Erdmann, N., Nunnemann, M., Eberhardt, K., Herrmann, G., Huber, G., Kohler, S., Kratz,
J. V., Passler, G., Peterson, J. R., Trautmann, N., and Waldek, A. 1998. Determination
of the first ionization potential of nine actinide elements by resonance ionization mass
spectroscopy (RIMS). J Alloys Compd 271, 837–840.
ETP 2005. Online at http://www.etpsci.com/htm/etp/tech info/tech articles/active film.asp.
Viewed on November 15, 2005.
Evans, R. D. 1955. The Atomic Nucleus. Pp. 470–510, 625 New York, NY: McGraw-Hill.
Faires, R. A. and Boswell, G. G. J. 1981. Radioisotope Laboratory Techniques. 4th ed.
Oxford, London: Butterworth-Heinemann.
Fassel, V. A. 1971. Spectroscopic properties and analytical applications of induction-
coupled plasmas. Appl Spectrosc 25(1), 148–&.
Fassel, V. A. and Kniseley, R. N. 1974 Inductively coupled plasmas. Anal Chem 46(13),
1155–&.
Faure, G. 1986. Principles of Isotope Geology. P. 399. New York, NY: John Wiley.
Fearey, B. L., Tissue, B. M., Olivares, J. A., Loge, G. W., Murrell, M. T., and Miller,
C. M. 1992. High-precision thorium RIMS for geochemistry. Inst Phys Conf Ser (128),
209–212.
Fifield, L. K. 1999. Accelerator mass spectrometry and its applications. Rep Prog Phys
62(8), 1223–1274.
Fifield, L. K. 2000. Advances in accelerator mass spectrometry. Nucl Instrum Meth B 172,
134–143.
Fifield, L. K., Carling, R. S., Cresswell, R. G., Hausladen, P. A., di Tada M. L., and Day,
J. P. 2000. Accelerator mass spectrometry of Tc-99. Nucl Instrum Meth B 168(3), 427–
436.
Finney, G. D. and Evans, R. 1935. The radioactivity of solids determined by alpha-ray
counting. Phys Rev 48, 503.
Finston, H. L. and Kinsley, M. T. 1961. The Radiochemistry of Cesium. National Research
Council Report NAS-NS 3035. Washington, DC: National Research Council.
Firestone, R. B. 1990. Table of Isotopes. 8th ed. New York, NY: John Wiley.
Firestone, R. B. 1996. Table of Isotopes. 9th ed. New York, NY: John Wiley.
Friedlander, G. and Herrman, G. 2003. History of nuclear and radiochemistry in Handbook
of Nuclear Chemistry. Vol. I. Chapter 1. Vertes, A., Nagy, S., and Klencsar, Z., eds.
Dordrecht: Kluwer Academic.
Friedlander, G., Kennedy, J. W., Macias, E. S., and Miller, J. M. 1981. Nuclear and Radio-
chemistry. 3rd ed. New York, NY: John Wiley.
Fritz, J. S. and Umbreit, G. R. 1958. Ion exchange separation of metal ions. Anal Chim Acta
19, 509–516.
FRMAC 2002. Federal Radiological Monitoring and Assessment Center Report
DOE/NV/11718—181. Monitoring and Sampling Manual. Vol. 1, Rev. 1. Washington
DC: FRMAC.
Gäggeler, H. W. 1994. On-line gas chemistry experiments with transactinide elements. J
Radioanal Nucl Chem Articles 183, 261.
Gäggeler, H. W., Jost, D. T., Kovacs, J., Scherer, U. W., Weber, A., Vermeulen, D., Türler,
A., Gregorich, K. E., Henderson, R. A., Czerwinski, K. R., Kadkhodayan, B., Lee,
D. M., Nurmia, M., Hoffman, D. C., Kratz, J. V., Gober, M. K., Zimmermann, H. P.,
Schädel, M., Brüchle, W., Schimpf, E., and Zvara, I. 1992. Gas phase chromatography
experiments with bromides of tantalum and element-105. Radiochim Acta 57,
93–100.
446 References

Garcia, R. and Kahn, B. 2001. Total dissolution of environmental and biological samples for
radiometric analysis by closed-vessel microwave digestion. J Radiological Nucl Chem
250, 85–91.
Gärtner, M., Boettger, M., Eichler, B., Gäggeler, H. W., Grantz, M., Hübener, S., Jost,
D. T., Piguet, D., Dressler, R., Türler, A., and Yakushev, A. B. 1997. On-line gas chro-
matography of Mo, W and U (oxy) chlorides. Radiochim Acta 78, 59–68.
Gastel, M., Becker, J. S., Kuppers, G., and Dietze, H. J. 1997. Determination of long-
lived radionuclides in concrete matrix by laser ablation inductively coupled plasma mass
spectrometry. Spectrochim Acta B 52(14), 2051–2059.
Gerlich, D. 1992. Inhomogeneous Rf-fields—a versatile tool for the study of processes with
slow ions. Adv Chem Phys 82, 1–176.
Ghiorso, A., Lee, D., Somerville, L. P., Loveland, W., Nitschke, J. M., Ghiorso, W., Seaborg,
G. T., Wilmarth, P., Leres, R., Wydler, A., Nurmia, M., Gregorich, K., Czerwinski, K.,
Gaylord, R., Hamilton, T., Hannink, N. J., Hoffman, D. C., Jarzynski, C., Kacher, C.,
Kadkhodayan, B., Kreek, S., Lane, M., Lyon, A., McMahan, M. A., Neu, M., Sikkeland,
T., Swiatecki, W. J., Türler, A., Walton, J. T., and Yashita, S. 1995a. Evidence for the
synthesis of element 267 110 produced by the 59 Co + 209 Bi reaction. Nucl Phys 583,
861–866.
Ghiorso, A., Lee, D., Somerville, L. P., Loveland, W., Nitschke, J. M., Ghiorso, W., Seaborg,
G. T., Wilmarth, P., Leres, R., Wydler, A., Nurmia, M., Gregorich, K., Czerwinski, K.,
Gaylord, R., Hamilton, T., Hannink, N. J., Hoffman, D. C., Jarzynski, C., Kacher, C.,
Kadkhodayan, B., Kreek, S., Lane, M., Lyon, A., McMahan, M. A., Neu, M., Sikkeland,
T., Swiatecki, W. J., Türler, A., Walton, J. T., and Yashita, S. 1995b. Evidence for the
possible synthesis of element 267 110 produced by the 59 Co + 209 Bi reaction. Phys Rev C
51, R2293–R2297.
Ghiorso, A., Nitschke, J. M., Alonso, J. R., Alonso, C. T., Nurmia, M., Seaborg, G. T.,
Hulet, E. K., and Lougheed, R. W. 1974. Element 106. Phys Rev Lett 33, 1490–1493.
Gibson, J. K. 1998. Laser ablation mass spectrometry of actinide dioxides: ThO2 , UO2 ,
NpO2 , PuO2 and AmO2 . Radiochim Acta 81(2), 83–91.
Gill, C. G., Daigle, B., and Blades, M. W. 1991. Atomic mass-spectrometry of solid samples
using laser ablation-ion trap mass-spectrometry (LAITMS). Spectrochim Acta B 46(8),
1227–1235.
Gilmore, G. and Hemingway, J. D. 1995. True coincidence summing in Practical Gamma-
Ray Spectrometry. Chapter 7. West Sussex, England: John Wiley.
Gindler, J. E. 1962. The Radiochemistry of Uranium. National Research Council Report
NAS-NS 3050. Washington, DC: National Research Council.
Gonzalez, J., Mao, X. L., Roy, J., Mao, S. S., and Russo, R. E. 2002. Comparison of 193,
213 and 266 nm laser ablation ICP-MS. J Anal Atom Spectrom 17(9), 1108–1113.
Gordon, L., Salutsky, M. L., and Willard, H. H. 1959. Precipitation from Homogeneous
Solution. New York, NY: John Wiley.
Granstrom, M. L. and Kahn, B. 1955. The adsorption of cesium in low concentration by
paper. J Phys Chem 59, 408–410.
Grate, J. W. and Egorov, O. 2003. Automated radiochemical separation, analysis, and sens-
ing in Handbook of Radioactivity Analysis. pp. 1129–1164. San Diego, CA: Elsevier.
Gray, A. L. and Date, A. R. 1983. Inductively coupled plasma source-mass spectrometry
using continuum flow ion extraction. Analyst 108(1290), 1033–1050.
Gregorich, K. E., Ginter, T., Loveland, W., Peterson, D., Patin, J. B., Folden, C. M., III,
Hoffman, D. C., Lee, D. M., Nitsche, H., Omtvedt, J. P., Omtvedt, L. A., Stavsetra, L.,
 References 447

Sudowe, R., Wilk, P. A., Zielinski, P., and Aleklett, K. 2003. Cross section limits for the
208
Pb(86 Kr,n)293 118 reaction. Eur Phys J A 18, 633–638.
Gregorich, K. E., Henderson, R. A., Lee, D. M., Nurmia, M. J., Chasteler, R. M., Hall,
H. L., Bennett, D. A., Gannett, C. M., Chadwick, R. B., Leyba, J. D., Hoffman, D. C.,
and Herrmann, G. 1988. Aqueous chemistry of element l05. Radiochim Acta 43, 223–231.
Groenewold, G. S., Delmore, J. E., Olson, J. E., Appelhans, A. D., Ingram, J. C., and Dahl,
D. A. 1997. Secondary ion mass spectrometry of sodium nitrate: Comparison of ReO4-
and Cs+ primary ions. Int J Mass Spectrom Ion Processes 163(3), 185–195.
Gruning, C., Huber, G., Klopp, P., Kratz, J. V., Kunz, P., Passler, G., Trautmann, N., Waldek,
A., and Wendt, K. 2004. Resonance ionization mass spectrometry for ultratrace analysis
of plutonium with a new solid state laser system. Int J Mass Spectrom 235(2), 171–178.
Guillaumont, R., Adloff, J.-P., and Peneloux, A. 1989. Kinetic and thermodynamic aspect
of tracer-scale and single atom chemistry. Radiochim Acta 46, 169–176.
Guillaumont, R., Adloff, J.-P., Peneloux, A., and Delamoye, P. 1991. Sub-tracer scale be-
haviour of radionuclides. Application to actinide chemistry. Radiochim Acta 54, 1–15.
Gunnink, R. and Niday, J. B. 1972. Description of the GAMANAL program, UCRL-51061
in Computerized Quantitative Analysis by Gamma-ray Spectrometry. Vol. I. Livermore,
CA: Lawrence Livermore Laboratory.
Gunther, D., Horn, I., and Hattendorf, B. 2000. Recent trends and developments in laser
ablation-ICP-mass spectrometry. Fresen J Anal Chem 368(1), 4–14.
GV Instruments 2005. Application Report AN-20. Online at http://www.gvinstruments.
co.uk/index.asp.
Gy, Pierre M. 1992. Sampling of Heterogeneous and Dynamic Material Systems: Theories
of Heterogeneity, Sampling and Homogenizing. Amsterdam, The Netherlands: Elsevier.
Hahn, O. 1936. Applied Radiochemistry. Ithaca, NY: Cornell University Press.
Haissinsky, M. 1964. Nuclear Chemistry and its Applications. Reading, MA: Addison-
Wesley.
Haldimann, M., Baduraux, M., Eastgate, A., Froidevaux, P., O’Donovan, S., Von Gunten,
D., and Zoller, O. 2001. Determining picogram quantities of uranium in urine by iso-
tope dilution inductively coupled plasma mass spectrometry. Comparison with alpha-
spectrometry. J Anal Atom Spectrom 16(12), 1364–1369.
Harris, M. K., Herrington, P. B., Miley, H. S., Ellis, J. E., McKinnon, A. D., and St. Pierre,
D. E. 1999. Data authentication demonstration radionuclide stations in 21st Seismic
Research Symposium: Technologies for Monitoring the Comprehensive Nuclear-Test-
Ban Treaty. pp. 331–337. Los Alamos, NM: Los Alamos National Laboratory.
Heimbuch, A. M., Gee, H. Y., DeHaan, A. Jr., and Leventhal, L. 1965. The assay of alpha-
and beta-emitters by means of scintillating ion-exchange resins in Radioactive Sample
Measurement Techniques in Medicine and Biology. pp. 505–519. Vienna, Austria: Inter-
national Atomic Energy Agency.
Heinrich, C. A., Pettke, T., Halter, W. E., Aigner-Torres, M., Audetat, A., Gunther, D.,
Hattendorf, B., Bleiner, D., Guillong, M., and Horn, I. 2003. Quantitative multi-element
analysis of minerals, fluid and melt inclusions by laser-ablation inductively-coupled-
plasma mass-spectrometry. Geochim Cosmochim Ac 67(18), 3473–3497.
Hellborg, R., Faarinen, M., Kiisk,M., Magnusson, C. E., Persson, P., Skog, G., and
Stenstrom, K. 2003. Accelerator mass spectrometry—an overview. Vacuum 70(2–3),
365–372.
Henry, R., Koller, D., Liezers, M., Farmer, O. T., Barinaga, C., Koppenaal, D. W., and
Wacker, J. 2001. New advances in inductively coupled plasma—mass spectrometry
448 References

(ICP-MS) for routine measurements in the nuclear industry. J Radioanal Nucl Ch 249(1),
103–108.
Hensley, J. W., Long, A. O., and Willard, J. E. 1949. Reactions of ions in aqueous solution
with glass and metal surfaces. Ind Eng Chem 41, 1415–1421.
Herrmann, G., Riegel, J., Rimke, H., Sattelberger, P., Trautmann, N., Urban, F. J., Ames,
F., Otten, E. W., Ruster, W., and Scheerer, F. 1991. Resonance ionization mass-
spectroscopy of uranium. Inst Phys Conf Ser (114), 251–254.
Herzog, G. F. 1994. Applications of accelerator mass-spectrometry in extraterrestrial ma-
terials. Nucl Instrum Meth B 92(1–4), 492–499.
Heumann, K. G. 1992. Isotope-dilution mass-spectrometry. Int J Mass Spectrom Ion Pro-
cesses 118, 575–592.
Heusser, G. 2003. Low-background gamma spectroscopy of natural decay chain activity
at the μBq/kg level. Paper presented at the national meeting of the American Chemical
Society, Division of Nuclear Chemistry and Technology; Paper No. 15. New York. Con-
tact address for Heusser: Max Planck Institut fuer Kernphysik, POB 103 980 D-69029,
Heidelberg, Germany.
Hidaka, H., Holliger, P., and Gauthier-Lafaye, F. 1999. Tc/Ru fractionation in the Oklo and
Bangombe natural fission reactors, Gabon. Chem Geol 155(3–4), 323–333.
Hindman, F. D. 1986. Actinide separations for α spectrometry using neodymium fluoride
coprecipitation. Anal Chem 58, 1238–1241.
Hoffman, D. C. 1994. The heaviest elements. Chem & Eng News 72, 24–34.
Hoffman, D. C. and Lee, D. M. 1999. Chemistry of the heaviest elements—One atom at a
time. J Chem Ed 76, 331–347.
Hoffman, D. C. and Lee, D. M. 2003. Superheavy Elements in Handbook of Nuclear
Chemistry, vol. 2. Chapter 10. pp. 397–416. Attila Vértes, S. N., Zoltán Klencsár, eds.
Norwell, Massachusetts: Kluwer Academic Publishers.
Hoffman, D. C., Hamilton, T. M., and Lane, M. R. 1996. Spontaneous Fission, in Nuclear
Decay Modes, Chapter 10. Poenaru, D. N. ed. Bristol: Institute of Physics Publishing.
Hoffman, D. C., Ghiorso, A., and Seaborg, G. T. 2000. The Transuranium People: The
Inside Story. pp. 258–298, 379–396. London: Imperial College Press.
Hoffman, D. C., Lee, D. M., and Pershina, V. 2004. Transactinide Elements and Future
Elements in: The Chemistry of the Actinide and Transactinide Elements. 3rd ed. vol. III.
Chapter 14. Katz, J. J., Morss, L. R., Edelstein, N., Fuger, J. eds. London: Kluwer
Academic Publishers.
Hofmann, S., Hessberger, F. P., Ackermann, D., Münzenberg, G., Antalic, S., Cagarda, P.,
Kindler, B., Kojouharova, J., Leino, M., Lommel, B., Mann, R., Popeko, A. G., Reshitko,
S., Saro, S., Uusitalo, J., and Yeremin, A. V. 2002. New results on elements 111 and 112.
Eur Phys J A 14, 147–157.
Hofmann, S., Ninov, V., Hessberger, F. P., Armbruster, P., Folger, H., Münzenberg, G.,
Schott, H. J., Popeko, A. G., Yeremin, A. V., Andreyev, A. N., Saro, S., Janik, R., and
Leino, M. 1995a. Production and decay of 269 110. Z Phys A 350, 277–280.
Hofmann, S., Ninov, V., Hessberger, F. P., Armbruster, P., Folger, H., Münzenberg, G.,
Schott, H. J., Popeko, A. G., Yeremin, A. V., Andreyev, A. N., Saro, S., Janik, R., and
Leino, M. 1995b. The new element 111. Z Phys A 350, 281.
Hofmann, S., Ninov, V., Hessberger, F. P., Armbruster, P., Folger, H., Münzenberg, G.,
Schott, H. J., Popeko, A. G., Yeremin, A. V., Saro, S., Janik, R., and Leino, M. 1996. Z
Phys A 354, 229–230.
Hofstetter, K. J., Cable, P. R., and Beals, D. M. 1999. Field analyses of tritium at environ-
mental levels. Nucl Instr Methods in Phys Res A422, 761–766.
 References 449

Holden, N. E. and Hoffman, D. C. 2000. Spontaneous fission half-lives for ground-state

nuclides. Pure & Appl Chem 72, 1525–1562.
Horowitz, E. P., Chiarizia, R., Dietz, M. L., Diamond, H., and Nelson, D. M. 1998. Separation
and preconcentration of actinides from acidic media by extraction chromatography. Anal
Chim Acta 281, 361–372.
Horrocks, D. L 1974. Applications of Liquid Scintillation Counting, New York, NY: Aca-
demic Press.
Hotchkis, M., Fink, D., Tuniz, C., and Vogt, S. 2000. Accelerator mass spectrometry analyses
of environmental radionuclides: sensitivity, precision and standardisation. Appl Radiat
Isotopes 53(1–2), 31–37.
Houk, R. S. and Thompson, J. J. 1982. Elemental and isotopic analysis of solutions by
mass-spectrometry using a plasma ion-source. Am Mineral 67(3–4), 238–243.
Houk, R. S., Fassel, V. A., Flesch G. D., Svec H. J., Gray A. L., and Taylor C. E. 1980.
Inductively coupled argon plasma as an ion-source for mass-spectrometric determination
of trace-elements. Anal Chem 52(14), 2283–2289.
Houk, R. S., Svec, H. J., and Fassel, V. A. 1981. Mass-spectrometric evidence for suprather-
mal ionization in an inductively coupled argon plasma. Appl Spectrosc 35(4), 380–
384.
Hulet, E. K., Lougheed, R. W., Wild, J. F., and Landrum, J. H. 1980. Chloride complexation
of element 104. J Inorg Nucl Chem 42, 79–82.
Hyde, E. K., Hoffman, D. C., and Keller, O. L., Jr. 1987. A history and analysis of the
discovery of elements l04 and l05. Radiochim Acta 42, 57–102.
IAEA 1989. International Atomic Energy Agency Technical Report Series No. 295. Mea-
surement of Radionuclides in Food and the Environment—A Guidebook. Vienna, Austria:
IAEA.
IAEA 2003. Safeguards, Techniques and Equipment. pp. 82. Vienna, Austria: International
Atomic Energy Agency.
ICRU 1972. International Commission on Radiation Units and Measurements, Report 22.
Measurement of Low-level Radioactivity. Washington, DC: ICRU.
ICRU 1979. International Commission on Radiation Units and Measurements, Report 31.
Average Energy Required to Produce an Ion Pair. Washington, DC: ICRU.
ICRU 1984. International Commission on Radiation Units and Measurements, Report 37.
Stopping Powers for Electrons and Positrons. Washington, DC: ICRU.
ICRU 1993. International Commission on Radiation Units and Measurments, Report 49.
Stopping Powers and Ranges of Protons and Alpha Particles. Washington, DC:
ICRU.
ICRU 1997. International Commission on Radiation Units and Measurments, Report 56.
Dosimetry of External Beta Rays for Radiation Protection. Washington, DC: ICRU.
IEEE 1996. Institute of Electrical and Electronics Engineers. Report No. 325-1996. IEEE
Standard Test Procedures for Germanium Gamma-Ray Detectors. Posted online at iee-
explore.ieee.org in 2002.
IEEE 2002. Institute of Electrical and Electronics Engineers. IEEE Standard for Software
Quality Assurance Plans. IEEE (draft).
ISO 1987. International Organization for Standardization, ISO 9000/ASQC Q9000 Series.
Milwaukee, WI: ASQC.
ISO 1993. International Organization for Standardization. International Vocabulary of Basic
and General Terms in Metrology. Geneva, Switzerland: ISO.
ISO 1995. International Organization for Standardization. Guide to the Expression of Un-
certainty in Measurement. Geneva, Switzerland: ISO.
450 References

ISO 2005. International Organization for Standardization online at http://www.iso.org/.

Standards in Appendix A-4 may be found by clicking “ISO Store” and searching for the
standards number. Viewed on January 10, 2006.
Jerome, S. M., Smith, D., Woods, M. J., and Woods, S. A. 1995. Metrology of plutonium
for environmental measurements. Appl Radiat Isotopes 46(11), 1145–1150.
Kadkhodayan, B., Türler, A., Gregorich, K. E., Baisden, P. A., Czerwinski, K. R., Eichler,
E., Gäggeler, H. W., Hamilton, T. M., Stoyer, N. J., Jost, D. T., Kacher, C. D., Kovacs,
A., Kreek, S. A., Lane, M. R., Mohar, M. F., Neu, M. P., Sylwester, E. R., Lee, D. M.,
Nurmia, M. J., Seaborg, G. T., and Hoffman, D. C. 1996. On-line gas chromatographic
studies of chlorides of rutherfordium and homologs Zr and Hf. Radiochim Acta 72, 169–
178.
Kahn, B., Blanchard, R. L., Kolde, H. E., Krieger, H. L., Gold, S., Brinck, W. L., Averett, W.
J., Smith, D. B., and Martin, A. 1971. Radiological Surveillance Studies at a Pressurized
Water Nuclear Power Reactor. US Environmental Protection Agency Report RD71-1.
Washington, DC: EPA.
Kahn, B., Rosson, R., and Cantrell, J. 1990. Analysis of 228 Ra and 226 Ra in public water
supplies by gamma-ray spectrometer. Health Phys 59, 125–131.
Karam, L. R., Pibida, L., and McMahon, C. A. 2002. Use of resonance ionization mass
spectrometry for determination of Cs ratios in solid samples. Appl Radiat Isotopes 56(1–
2), 369–374.
Karge, H. G. and Weitkamp J. 1998. Molecular Sieves: Science and Technology. New York,
NY: Springer-Verlag.
Karol, P. J., Nakahara, H., Petley, B. W., and Vogt, E. 2001. On the discovery of the elements
110–112. Pure Appl Chem 73, 959–967.
Karol, P. J., Nakahara, H., Petley, B. W., and Vogt, E. 2003. On the claims for discovery of
elements 110, 111, 112, 114. Pure Appl Chem 75, 1601–1611.
Keller, O. L., Jr. 1984. Chemistry of heavy actinides and light transactinides. Radiochim
Acta 37, 169–180.
Keller, J. H., Thomas, T. R., Spence, D. T., and Maech, W. J. 1973. An evaluation of materials
and techniques used for monitoring airborne radioiodine species in Proceedings of the
Twelfth Air Cleaning Conference, USA Department of Energy Report CONF-720823,
Vol. 1, p. 322. Washington, DC: DOE.
Kelley, J. M. and Robertson, D. M. 1985. Plutonium Ion emission from carburized rhenium
mass-spectrometer filaments. Anal Chem 57(1), 124–130.
Kelley, J. M., Bond, L. A., and Beasley, T. M. 1999. Global distribution of Pu isotopes and
Np-237. Sci Total Environ 238, 483–500.
Kenna, T. C. 2002. Determination of plutonium isotopes and neptunium-237 in environ-
mental samples by inductively coupled plasma mass spectrometry with total sample
dissolution. J Anal Atom Spectrom 17(11), 1471–1479.
Kennedy, J. W., Seaborg, G. T., Segrè, E., and Wahl, A. C. 1946. Properties of 239 94. Phys
Rev 70, 555–556.
Kepak, F. 1971. Adsorption and colloidal properties of radioactive elements in trace con-
centrations. Chem Rev 71, 357–369.
Kerl, W., Becker, J. S., Dietze, H. J., and Dannecker, W. 1997. Isotopic and ultratrace
analysis of uranium by double-focusing sector field ICP mass spectrometry. Fresen J
Anal Chem 359(4–5), 407–409.
Kersting, A. B., Efurd, D. W., Finnegan, D. L., Rokop, D. J., Smith, D. K., and Thompson,
J. L. 1999. Migration of plutonium in ground water at the Nevada Test Site. Nature
397(6714), 56–59.
 References 451

Kim, C. K., Seki, R., Morita, S., Yamasaki, S., Tsumura, A., Takaku, Y., Igarashi, Y., and
Yamamoto, M. 1991. Application of a high-resolution inductively coupled plasma mass-
spectrometer to the measurement of long-lived radionuclides. J Anal Atom Spectrom 6(3),
205–209.
Kim, C. S., Kim, C. K., Lee, J. I., and Lee, K. J. 2000. Rapid determination of Pu isotopes
and atom ratios in small amounts of environmental samples by an on-line sample pre-
treatment system and isotope dilution high resolution inductively coupled plasma mass
spectrometry. J Anal Atom Spectrom 15(3), 247–255.
King S. E., Phillips G. W., August R. A., Beach L. A., Cutchin J. H., and Castaneda C.
1987. Detection of tritium using accelerator mass-spectrometry. Nucl Instrum Meth B
29(1–2), 14–17.
Kingston, H. M. and Haswell, S., eds. 1997. Microwave Enhanced Chemistry: Fundamen-
tals, Sample Preparation, and Applications. Washington, DC: ACS.
Kirbach, U. W., Folden, C. M., III, Ginter, T. N., Gregorich, K. E., Lee, D. M., Ninov,
V., Omtvedt, J. P., Patin, J. B., Seward, N. K., Strellis, D. A., Sudowe, R., Türler,
A., Wilk, P. A., Zielinski, P. M., Hoffman, D. C., and Nitsche, H. 2002. The Cryo-
Thermochromatographic Separator (CTS): A new rapid separation and a detection system
for on-line chemical studies of highly volatile osmium and hassium (Z = 108) tetroxides.
Nucl Instrum Meth A 484, 587–594.
Kirby, H. W. and Salutsky, M. L. 1964. The Radiochemistry of Radium. National Research
Council Report NAS-NS 3057. Washington, DC: National Research Council.
Kleinberg, J. and Cowan, G. A. 1960. The Radiochemistry of Fluorine, Chlorine, Bromine
and Iodine. National Research Council Report NAS-NS 3005. Washington, DC: National
Research Council.
Knoll, G. F. 1989. Radiation Detection and Measurement. 2nd ed. New York, NY:
Wiley.
Knoll, G. F. 2000. Radiation Detection and Measurement. 3rd ed. New York, NY: John
Wiley.
Kocher, D. C. 1977. Nuclear Decay Data for Radionuclides Occurring in Routine Re-
leases from Nuclear Fuel Cycle Facilities. Oak Ridge National Laboratory Report
ORNL/NUREG/TM-102. Oak Ridge, TN.
Kolthoff, I. M. 1932. Theory of coprecipitation: the formation and properties of crystalline
precipitates. J Phys Chem 36, 860–881.
Kolthoff, I. M. and Elving P. J. 1971. Treatise on Analytical Chemistry. Pt. 1, Vol. 9. New
York, NY: John Wiley.
Kolthoff, I. M. and Lingane, J. J. 1939. The fundamental principles and applications of
electrolysis with the dropping mercury electrode and Heyrovsk’s polarographic method
of chemical analysis. Chem Rev 24, 1–94.
Koppenaal, D. W., Eiden, G. C., and Barinaga, C. J. 2004. Collision and reaction cells in
atomic mass spectrometry: development, status, and applications. J Anal Atom Spectrom
19(5), 561–570.
Koppenol, W. H. 2002. Naming of new elements (IUPAC Recommendations 2002). Pure
Appl Chem 74, 787–791.
Korenman, I. M. 1968. Analytical Chemistry of Low Concentrations. Jerusalem: Israel
program for scientific translations ltd.
Korkisch, J. 1989. Handbook of Ion Exchange Resins: Their Application to Inorganic
Analytical Chemistry. Boca Raton, FL: CRC Press.
Kovach, J. L. 1968. Adsorbents; review and projection in Proceedings of the tenth AEC
air cleaning conference. First, M. W. and Morgan, J. M., Jr., eds. pp. 149–166.US DOE
452 References

Report CONF-680821. Springfield, VA: Clearinghouse for federal scientific and technical
information.
Kratz, J. V. 1999a. Fast chemical separation procedures for transactinides in Heavy Elements
and Related New Phenomena. Greiner, W. and Gupta, R. K. eds. pp. 43–63. Singapore:
World Scientific.
Kratz, J. V. 1999b. Chemical properties of the transactinide elements in Heavy Elements
and Related New Phenomena. Greiner, W. and Gupta, R. K. eds. pp. 129–193. Singapore:
World Scientific.
Kraus, K. A. and Nelson, F. 1956. Anion exchange studies of the fission products in
Proceedings of the 1st International Conference on Peaceful Uses of Atomic Energy.
Vol. 7. A/CONF.8/7. pp. 113–125. New York, NY: United Nations Publishers.
Kusaka, Y. and Meinke, W. W. 1961. Rapid Radiochemical Separations. National Research
Council Report NAS-NS 3104. Washington, DC: National Research Council.
Kutschera, W. 1994. Atom counting of long-lived radionuclides. Nuclear Instruments &
Methods in Physics Research Section a-Accelerators Spectrometers Detectors and Asso-
ciated Equipment 353(1–3), 562–562.
Ladd, M. F. C. and Lee, W. H. 1964. Elementary Practical Radiochemistry. London: Cleaver-
Hume Press.
Laitinen, H. A. and Watkins, N. H. 1975. Cathodic stripping coulometry of lead. Anal Chem
47, 1352–1358.
Langmuir I. and Kingdon K. H. 1925. Thermionic effects caused by vapours of alkali metals.
P Roy Soc A 107, 61–79.
Larsen, B. S. and McEwen, C. N. 1998. Mass Spectrometry of Biological Materials. pp. 469.
New York, NY: Marcel Dekker.
Laue, C. A. and Nash, K. L., eds. 2003. Radioanalytical Methods in Interdisciplinary
Research, ACS Symposium Series 868. Washington, DC: American Chemical Society.
Lazarev, Y. A., Lobanov, Y. V., Oganessian, Y. T., Utyonkov, V. K., Abdullin, F. S.,
Buklanov, G. V., Gikal, B. N., Iliev, S., Mezentsev, A. N., Polyakov, A. N., Sedykh,
I. M., Shirokovsky, I. V., Subbotin, V. G., Sukhov, A. M., Tsyganov, Y. S., Zhuchko, V.
E., Lougheed, R. W., Moody, K. J., Wild, J. F., Hulet, E. K., and McQuaid, J. H. 1994.
Discovery of enhanced nuclear stability near the deformed shells N=162 and Z=108.
Phys Rev Lett 73, 624–627.
Lazarev, Y. A., Lobanov, Y. V., Oganessian, Y. T., Utyonkov, V. K., Abdullin, F. S., Polyakov,
A. N., Rigol, J., Shirokovsky, I. V., Tsyganov, Y. S., Iliev, S., Subbotin, V. G., Sukhov,
A. M., Buklanov, G. V., Gikal, B. N., Kutner, V. B., Mezentsev, A. N., Subotic, K., Wild,
J. F., Longheed, R. W., and Moody, K. J. 1996. Alpha decay of (273)110 - shell closure
at N=162. Phys Rev C 54, 620–625.
Lederer, C. M., Hollander, J. M., and Perlman, I. 1967. Table of Isotopes, 6th ed. New York,
NY: John Wiley.
Lee, S. H., Gastaud, J., La Rosa, J. J., Kwong, L. L. W., Povinec, P. P., Wyse, E., Fifield,
L. K., Hausladen, P. A., Di Tada, L. M., and Santos, G. M. 2001. Analysis of plutonium
isotopes in marine samples by radiometric, ICP-MS and AMS techniques. J Radioanal
Nucl Ch 248(3), 757–764.
Lee, T., Ku, T. L., Lu, H. L., and Chen, J. C. 1993. 1st detection of fallout Cs-135 and
potential applications of Cs-137/Cs-135 ratios. Geochim Cosmochim Ac 57(14), 3493–
3497.
Lewis, D., Miller, G., Duffy, C. J., Efurd, D. W., Inkret, W. C., and Wagner, S. E.
2001. Los Alamos National Laboratory thermal ionization mass spectrometry results
from intercomparison study of inductively coupled plasma mass spectrometry, thermal
 References 453

ionization mass spectrometry, and fission track analysis of mu Bq quantities of Pu-239

in synthetic urine (LA-UR-001698). J Radioanal Nucl Ch 249(1), 115–120.
Libby, W. F. 1940. Reactions of high energy atoms produced by slow neutron capture. J Am
Chem Soc 62, 1930–1943.
Lieser, K. H. 2001. Nuclear and Radiochemistry: Fundamentals and Applications. 2nd ed.
New York, NY: John Wiley.
Lindsey, A. J. 1964. Electrodeposition in: Comprehensive analytical chemistry. Vol. IIA.
pp. 7–64. Wilson, C. L., Wilson, D. W., and Strouts, C. R. N., eds. New York, NY:
Elsevier.
Litherland, A. E. and Kilius, L. R. 1997. Neutral injection for AMS. Nucl Instrum Meth
Phys B 123(1–4), 18–21.
Liu, Y. F., Guo, Z. Y., Liu, X. Q., Qu, T., and Xie, J. L. 1994. Applications of accelerator
mass-spectrometry in analysis of trace isotopes and elements. Pure Appl Chem 66(2),
305–334.
Liu, W. and van Wüllen, C. 1999. Spectroscopic constants of gold and eka-gold (element
111) diatomic compounds: The importance of spin–orbit coupling. J Chem Phys 110,
3730–3735.
Lockhart, L., Patterson, R., and Anderson, W. 1964. Characteristics of Air Filter Media Used
for Monitoring Airborne Radioactivity. Naval Research Laboratory Report NRL-6054.
Washington, DC: US Department of Defense.
Lougheed, R. W., Moody, K. J., Wild, J. F., Hulet, E. K., McQuaid, J. H., Lazarev, Y. A.,
Lobanov, Y. V., Oganessian, Y. T., Utyonkov, V. K., Abdullin, F. S., Buklanov, G. V.,
Gikal, B. N., Iliev, S., Mezentsev, A. N., Polyakov, A. N., Sedykh, I. M., Shirokovsky, I. V.,
Subbotin, V. G., Sukhov, A. M., Tsyganov, Y. S., and Zhuchko, V. E. 1994. Observation
of enhanced nuclear stability near the 162 neutron shell. J Alloys and Comp 213, 61–66.
Love, A. H., Hunt, J. R., Roberts, M. L., Southon, J. R., Chiarappa-Zucca, M. L., and Dingle,
K. H. 2002. Use of tritium accelerator mass spectrometry for tree ring analysis. Environ
Sci Tech 36(13), 2848–2852.
Lucas, H. F. 1957. Improved low-level alpha scintillation counter for radon. Rev Sci Instrum
28, 680–683.
Magara, M., Hanzawa, Y., Esaka, F., Miyamoto, Y., Yasuda, K., Watanabe, K., Usuda, S.,
Nishimura, H., and Adachi, T. 2000. Development of analytical techniques for ultra trace
amounts of nuclear materials in environmental samples using ICP-MS for safeguards.
Appl Radiat Isotopes 53(1–2), 87–90.
Magara, M., Sakakibara, T., Kurosawa, S., Takahashi, M., Sakurai, S., Hanzawa, Y., Esaka,
F., Watanabe, K., and Usuda, S. E. 2002. Isotope ratio measurement of uranium in
safeguards environmental samples by inductively-coupled plasma mass spectrometry
(ICP-MS). J Nucl Sci Technol 39(4), 308–311.
Malinovsky, D., Rodushkin, I., Baxter, D., and Ohlander, B. 2002. Simplified method for
the Re-Os dating of molybdenite using acid digestion and isotope dilution ICP-MS. Anal
Chim Acta 463(1), 111–124.
Maney, J. P. 2002. Optimizing data collection design. Environ Sci Technol 39 (19), 383A–
389A.
Manninen, P. K. G. 1995. A rapid method for uranium isotope ratio measurement by
inductively-coupled plasma-mass spectrometry. J Radioan Nuc Ch Le 201(2), 71–80.
March, R. E. and Hughes, R. J. 1989. Quadrupole storage mass spectrometry in Chemical
Analysis. Vol. 102. Winefordner, J. D., ed. pp. 471. New York, NY: John Wiley.
Marcus, R. K. and Broekaert, J. A. C. 2002. Glow Discharge Plasmas in Analytical Spec-
troscopy. New York, NY: John Wiley.
454 References

Marcus, Y. and Kertes, A. S. 1969. Ion Exchange and Solvent Extraction of Metal Complexes.
New York, NY: Wiley Interscience.
Marinsky, J. A., Glendenin, L. E., and Coryell, C. D. 1947. J Am Chem Soc 69, 2781.
Martin, B., Ockenden, D. W., and Foreman, J. K. 1961. The solvent extraction of plutonium
and americium by tri-n-octylphosphine oxide. J Inorg Nucl Chem 21, 96–107.
Martin, M. J. and Blichert-Tolf, P. H. 1970. Radioactive atoms. Nucl Data Tables A8, 1–198,
1970.
McAninch, J. E., Hamilton, T. F., Brown, T. A., Jokela, T. A., Knezovich, J. P., Ognibene,
T. J., Proctor, I. D., Roberts, M. L., Sideras-Haddad, E., Southon, J. R., and Vogel J. S.
2000. Plutonium measurements by accelerator mass spectrometry at LLNL. Nucl Instrum
Meth A 172, 711–716.
McAninch, J. E., Marchetti, A. A., Bergquist, B. A., Stoyer, N. J., Nimz, G. J., Caffee,
M. W., Finkel, R. C., Moody, K. J., Sideras-Haddad, E., Buchholz, B. A., Esser, B. K.,
and Proctor, I. D. 1998. Detection of Tc-99 by accelerator mass spectrometry: preliminary
investigations. J Radioan Nucl Ch 234(1–2), 125–129.
McDowell, W. J. 1992. Photon/electron-rejecting alpha liquid scintillation (PERALS) spec-
trometry. A review. Radioactivity and Radiochemistry 3, 26–53.
McIntyre, J. I., Abel, K. H., Bowyer, T. W., Hayes, J. C., Heimbigner, T. R., Panisko, M. E.,
Reeder, P. L., and Thompson, R. C. 2001. Measurements of ambient radioxenon levels
using the automated radioxenon sampler/analyzer (ARSA). J Radioan Nucl Ch 248 (3),
629–635.
McMillan, E. M. and Abelson, P. H. 1940. Radioactive element 93. Phys Rev 57, 1185.
Meeks, A. M., Giaquinto, J. M., and Keller, J. M. 1998. Application of ICP-MS radionuclide
analysis to “Real World” samples of Department of Energy radioactive waste. J Radioan
Nucl Ch 234(1–2), 131–135.
Miley, H. S., Bowyer, S. M., Hubbard, C. W., McKinnon, A. D., Perkins, R. W., Thompson,
R. C., and Warner, R. A. 1998. A description of the DOE radionuclide aerosol sam-
pler/analyzer for the Comprehensive Test Ban Treaty. J Radioan Nucl Ch 235 (1–2),
83–87.
Miller, D. R. 1988. Free jet sources in Atomic and Molecular Beam Methods. Vol. 1.
Chapter 2. Scoles, G., ed. Oxford: Oxford University Press.
Moghissi, A. A., Bretthauer, E. W., and Compton, E. H. 1973. Separation of water from
biological and environmental samples for tritium analysis. Anal Chem 45, 1565–1566.
Momyer, F., Jr. 1960. The Radiochemistry of the Rare Gases. National Academy of
Sciences—National Research Council Report NAS-NS 3025. Washington, DC: National
Research Council.
Montaser, A. 1992. Inductively Coupled Plasmas in Analytical Atomic Spectrometry. New
York, NY: John Wiley.
Montaser, A. 1998. Inductively Coupled Plasma Mass Spectrometry. pp. 964. New York,
NY: Wiley-VCH.
Moreno, J. M. B., Betti, M., and Alonso, J. I. G. 1997. Determination of neptunium
and plutonium in the presence of high concentrations of uranium by ion chromatog-
raphy inductively coupled plasma mass spectrometry. J Anal Atom Spectrom 12(3), 355–
361.
Münzenberg, G., Armbruster, P., Folger, H., Hessberger, F. P., Hoffman, S., Keller, K.,
Poppensieker, K., Reisdorf, W., Schmidt, K.-H., and Schott, H.-J. 1984. The identification
of element 108. Z Phys A, 317, 235–236.
Münzenberg, G., Armbruster, P., Hessberger, F. P., Hofmann, S., Poppensieker, K., Reisdorf,
W., Schneider, J. H. R., Schneider, W. R. W., Schmidt, K.-H., Sahm, C.-C., and Vermeulen,
 References 455

D. 1982. Evidence for element 109 from one correlated decay sequence following the
fussion of 58 Fe with 209 Bi. Z Phys A 315, 145.
Münzenberg, G., Hofmann, S., Hessberger, F. P., Reisdorf, W., Schmidt, K. H., Schneider,
J. H. R., Armbruster, P., Sahm, C.-C., and Thuma, B. 1981. Identification of element 107
by α correlation chains. Z Phys A 300, 107.
Muramatsu, Y., Hamilton, T., Uchida, S., Tagami, K., Yoshida, S., and Robison, W. 2001.
Measurement of Pu-240/Pu-239 isotopic ratios in soils from the Marshall Islands using
ICP-MS. Sci Total Environ 278(1–3), 151–159.
Muramatsu, Y., Uchida, S., Tagami, K., Yoshida, S., and Fujikawa, T. 1999. Determination
of plutonium concentration and its isotopic ratio in environmental materials by ICP-
MS after separation using ion-exchange and extraction chromatography. J Anal Atom
Spectrom 14(5), 859–865.
Murthy, G. K. and Campbell, J. E. 1960. A simplified method for the determination of
Iodine131 in milk. J Dairy Science 8, 1042–1049.
Murthy, G. K., Coakley, J. E., and Campbell, J. E. 1960. A method for the elimination of
ashing in strontium-90 determination of milk. Dairy Sci 43, 151–154.
NAS-NRC 1960a. National Academy of Sciences—National Research Council Reports
NAS-NS 30xx. The Radiochemistry of ELEMENT. Washington, DC: Volumes in the
series published by the National Research Council, 1960–1988.
NAS-NRC 1960b. National Academy of Sciences—National Research Council Reports
NAS-NS-31xx: Nuclear Science Series. Radiochemistry Technique. Available from
NTIS, Springfield, VA. Volumes in the series published 1960–1988.
Navratil, O., Hala, J., Kopunec, R., Macasek, F., Mikulaj, V., and Leseticki, L. 1992. Nuclear
Chemistry. New York, NY: Ellis Horwood, PTR Prentice Hall.
NCRP 1976a. National Council on Radiation Protection and Measurements, Report No. 47.
Tritium Measurement Techniques. Bethesda, MD: NCRP.
NCRP 1976b. National Council on Radiation Protection and Measurements, Report No. 50.
Environmental Radiation Measurements. Bethesda, MD: NCRP.
NCRP 1979. National Council on Radiation Protection and Measurements,Report No. 62.
Tritium in the Environment. Bethesda, MD: NCRP.
NCRP 1983. National Council on Radiation Protection and Measurements Report No. 75.
I-129: Evaluation of Releases from Nuclear Power Generation. Bethesda, MD: NCRP.
NCRP 1985a. National Council on Radiation Protection and Measurements, Report No. 81.
Carbon-14 in the Environment. Bethesda, MD: NCRP.
NCRP 1985b. National Council of Radiation Protection and Measurements Report
No. 58. A Handbook of Radioactivity Measurement Procedures. 2nd ed. Bethesda, MD:
NCRP.
NCRP 1987a. National Council on Radiation Protection and Measurements, Report No. 94.
Exposure of the Population in the United States and Canada from Natural Background
Radiation. Bethesda, MD: NCRP.
NCRP 1987b. National Council on Radiation Protection and Measurements, Report
No. 87. Use of Bioassay Procedures for Assessment of Internal Radionuclide Deposition,
Bethesda, MD: NCRP, 1987.
Nelms S. 2005. Inductively Coupled Plasma Mass Spectrometry Handbook. Victoria,
Australia: Blackwell Publishing.
Nichols, M. 2006. Monte Carlo Simulation of Beta Particle Detection. Ph.D. thesis. Atlanta,
GA 30332: Georgia Institute of Technology.
Ninov, V., Gregorich, K. E., Loveland, W., Ghiorso, A., Hoffman, D. C., Lee, D. M., Nitsche,
H., Swiatecki, W. J., Kirbach, U. W., Laue, C. A., Adams, J. L., Patin, J. B., Shaughnessy,
456 References

D. A., Strellis, D. A., and Wilk, P. A. 1999. Observation of superheavy nuclei produced
in the reaction of 86 Kr with 208 Pb. Phys Rev Lett 83, 1104–1107.
Ninov, V., Gregorich, K. E., Loveland, W., Ghiorso, A., Hoffman, D. C., Lee, D. M., Nitsche,
H., Swiatecki, W. J., Kirbach, U. W., Laue, C. A., Adams, J. L., Patin, J. B., Shaughnessy,
D. A., Strellis, D. A., and Wilk, P. A. 2002. Retraction of 1999 published paper. Phys
Rev Lett 89, 039901–1.
NIST 1994. National Institute of Standards and Technology, Technical Note 1297. Guide-
lines for Evaluating and Expressing the Uncertainty of NIST Measurement Results.
Gaithersburg, MD: NIST.
Odom, R. W., Strathman, M. D., Buttrill, S. E., and Baumann, S. M. 1990. Nondestructive
imaging detectors for energetic particle beams. Nucl Instrum Meth B 44(4), 465–472.
Oganessian, Y. T. 2001. The synthesis and decay properties of the heaviest elements. Nucl
Phys A 685, 17c–36c.
Oganessian, Y. T. 2002. Synthesis and properties of even-even isotopes with Z = 110—116
in 48 Ca induced reactions. J Nucl Radiochem Sci 3, 5–8.
Oganessian, Y. T., Yeremin, A. V., Popeko, A. G., Bogomolov, S. L., Buklanov, G. V.,
Chelnokov, M. L., Chepigin, V. I., Gikal, B. N., Gorshkov, V. A., Gulbekian, G. G., Itkis,
M. G., Kabachenko, A. P., Lavrentev, A. Y., Malyshev, O. N., Rohac, J., Sagaidak, R. N.,
Hofmann, S., Saro, S., Giardinas, G., and Morita, K. 1999a. Synthesis of nuclei of the
superheavy element 114 in reactions induced by 48 Ca. Nature 400, 242–245.
Oganessian, Y. T., Utyonkov, V. K., Lobanov, Y. V., Abdullin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Gobomolov, S. L., Gikal, B. N.,
Mezentsev, A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Buklanov, G. V., Subotic,
K., Itkis, M. G., Moody, K. J., Wild, J. F., Stoyer, N. J., Stoyer, M. A., and Lougheed,
R. W. 1999b. Synthesis of superheavy nuclei in the 48 Ca + 244 Pu reaction. Phys Rev Lett
83, 3154–7.
Oganessian, Y. T., Yeremin, A. Y., Gulbekian, G. G., Bogomolov, S. L., Chepigin, V. I.,
Gikal, B. N., Gorshkov, V. A., Itkis, M. G., Kabachenko, A. P., Kutner, V. B., Lavrentev,
A. Y., Malyshev, O. N., Popeko, A. G., Rohac, J., Sagaidak, R. N., Hofmann, S.,
Münzenberg, G., Veselsky, M., Saro, S., Iwasa, N., and Morita, K. 1999c. Search for
new isotopes of element 112 by irradiation of 238 U with 48 Ca. Eur Phys J A 5, 63–
68.
Oganessian, Y. T., Utyonkov, V. K., Lobanov, Y. V., Abdullin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Bogomolov, S. L., Gikal, B. N.,
Mezentsev, A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Ivanov, O. V., Buklanov,
G. V., Subotic, K., Itkis, M. G., Moody, K. J., Wild, J. F., Stoyer, N. J., Stoyer, M. A., and
Lougheed, R. W. 2000a. Synthesis of superheavy nuclei in the Ca-48+Pu-244 reaction:
(288)114 - art. no. 041604. Phys Rev C 62, 041604 (R) 1–4.
Oganessian, Y. T., Utyonkov, V. K., Lobanov, Y. V., Abdulin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Bogomolov, S. L., Gikal, B. N.,
Mezentsev, A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Ivanov, O. V., Buklanov,
G. V., Subotic, K., Itkis, M. G., Moody, K. J., Wild, J. F., Stoyer, N. J., Stoyer, M. A.,
Lougheed, R. W., Laue, C. A., Karelin, Y. A., and Tatarinov, A. N. 2000b. Observation
of the decay of 292 116. Phys Rev C 63, 011301(R) 1–2.
Oganessian, Y. T., Utyonkov, V. K., Lobanov, Y. V., Abdullin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Bogomolov, S. L., Gikal, B. N.,
Mezentsev, A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Ivanov, O. V., Buklanov,
G. V., Subotic, K., Voinov, A. A., Itkis, M. G., Moody, K. J., Wild, J. F., Stoyer, N. J.,
Stoyer, M. A., Lougheed, R. W., and Laue, C. A. 2002. Eur Phys J A 15, 201–204.
 References 457

Oganessian, Y. T., Utyonkoy, V. K., Lobanov, Y. V., Abdullin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Bogomolov, S. L., Mezentsev,
A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Voinov, A. A., Buklanov, G. V., Subotic,
K., Zagrebaev, V. I., Itkis, M. G., Patin, J. B., Moody, K. J., Wild, J. F., Stoyer, M. A.,
Stoyer, N. J., Shaughnessy, D. A., Kenneally, J. M., and Lougheed, R. W. 2004a. Phys
Rev C 69, 021601(R).
Oganessian, Y. T., Yeremin, A. V., Popeko, A. G., Malyshev, O. N., Belozerov, A. V.,
Buklanov, G. V., Chelnokov, M. L., Chepigin, V. I., Gorshkov, V. A., Hofmann, S., Itkis,
M. G., Kabachenko, A. P., Kindler, B., Munzenberg, G., Sagaidak, R. N., Saro, S., Schott,
H.-J., Streicher, B., Shutov, A. V., Svirikhin, A. I., and Vostokin, G. K. 2004b. Eur Phys
J A 19, 3–6.
Oganessian, Y. T., Utyonkoy, V. K., Lobanov, Y. V., Abdullin, F. S., Polyakov, A. N.,
Shirokovsky, I. V., Tsyganov, Y. S., Gulbekian, G. G., Bogomolov, S. L., Gikal, B. N.,
Mezentsev, A. N., Iliev, S., Subbotin, V. G., Sukhov, A. M., Voinov, A. A., Buklanov,
G. V., Subotic, K., Zagrebaev, V. I., Itkis, M. G., Patin, J. B., Moody, K. J., Wild, J. F.,
Stoyer, M. A., Stoyer, N. J., Shaughnessy, D. A., Kenneally, J. M., and Lougheed, R. W.
2004c. Phys Rev C 69, 054607.
Omtvedt, J. P., Alstad, J., Breivik, H., Dyve, J. E., Eberhardt, K., Folden III, C. M., Ginter,
T., Gregorich, K. E., Hult, E. A., Johansson, M., Kirbach, U. W., Lee, D. M., Mendel, M.,
Nahler, A., Ninov, V., Omtvedt, L. A., Patin, J. B., Skarnemark, G., Stavsetra, L., Sudowe,
R., Wiehl, N., Wierczinski, B., Wilk, P. A., Zielinski, P. M., Kratz, J. V., Trautmann, N.,
Nitsche, H., and Hoffman, D. C. 2002. SISAK liquid-liquid extraction experiments with
pre-separated 257 Rf. J Nucl Radiochem Sci 3, 121–124.
Ostlund, H. G. and Mason, A. S. 1985. Atmospheric tritium in: Tritium Laboratory Data
Report No. 14. Rosenstiel School of Marine and Atmospheric Science. Miami, FL:
University of Miami.
OTA 1983. Office of Technology Assessment, NTIS order #PB84-114917. Habitability of
the Love Canal area: An analysis of the technical basis for the decision on the habit-
ability of the emergency declaration area. Washington, DC: U.S. Government Printing
Office.
Oughton, D. H., Fifield, L. K., Day, J. P., Cresswell, R. C., Skipperud, L., Di Tada,
M. L., Salbu, B., Strand, P., Drozcho, E., and Mokrov, Y. 2000. Plutonium from Mayak:
Measurement of isotope ratios and activities using accelerator mass spectrometry. Environ
Sci Technol 34(10), 1938–1945.
Oughton, D. H., Skipperud, L., Fifield, L. K., Cresswell, R. G., Salbu, B., and Day, P. 2004.
Accelerator mass spectrometry measurement of Pu-240/Pu-239 isotope ratios in Novaya
Zemlya and Kara Sea sediments. Appl Radiat Isotopes 61(2–3), 249–253.
Pajo, L., Schubert, A., Aldave, L., Koch, L., Bibilashvili, Y. K., Dolgov, Y. N., and
Chorokhov, N. A. 2001a. Identification of unknown nuclear fuel by impurities and phys-
ical parameters. J Radioanal Nucl Ch 250(1), 79–84.
Pajo L., Tamborini G., Rasmussen G., Mayer K., and Koch L. 2001b. A novel isotope anal-
ysis of oxygen in uranium oxides: comparison of secondary ion mass spectrometry, glow
discharge mass spectrometry and thermal ionization mass spectrometry. Spectrochim
Acta B 56(5), 541–549.
Pappas, R. S., Ting, B. G., and Paschal, D. C. 2004. Rapid analysis for plutonium-239 in
1 ml of urine by magnetic sector inductively coupled plasma mass spectrometry with a
desolvating introduction system. J Anal Atom Spectrom 19(6), 762–766.
Parrington, J. R., Knox, H. D., Breneman, S. L., Baum, E. M., and Feiner, F. 1996. Chart
of the Nuclides. 15th ed. San Jose, CA: GE Nuclear Energy.
458 References

Pasternack, B. S. and Harley, N. H. 1971. Detection limits for radionuclides in the analysis
of multi-component gamma ray spectrometer data. Nucl Inst and Methods 91, 533–540.
Penneman, R. A. and Keenan, T. K. 1960. The Radiochemistry of Americium and Curium.
National Research Council Report NAS-NS 3006. Washington, DC: National Research
Council.
Percival, D. R. and Martin, D. B. 1974. Sequential determination of radium-226, radium-
228, actinium and thorium isotopes in environmental and process waste samples. Anal
Chem 46, 1742–1749.
Perkins, R. W., Robinson, D. E., Thomas, C. W., and Young, J. A. 1989. IAEA-SM-306/125.
Comparison of nuclear accident and nuclear test debris in Proceedings of the Interna-
tional Symposium on Environmental Contamination Following a Major Nuclear Accident.
pp. 111–139. Vienna: IAEA, October 16–20, 1989.
Perrier, C. and Segrè, E. 1937. Some chemical properties of element 43. J Chem Phys 5,
712.
Perrin, R. E., Knobeloch, G. W., Armijo, V. M., and Efurd, D. W. 1985. Isotopic analysis of
nanogram quantities of plutonium by using a sid ionization source. Int J Mass Spectrom
Ion Processes 64(1), 17–24.
Perrin, D. D. 1964.Organic Complexing Reagents: Structure, Behavior, and Application to
Inorganic Analysis. New York: Interscience Publishers.
Pershina, V. 1998a. Solution chemistry of element 105-Part I: Hydrolysis of group 5 cations:
Nb, Ta, and Pa. Radiochim Acta 80, 65–73.
Pershina, V. 1998b. Solution chemistry of element 105-part II: Hydrolysis and complex
formation of Nb, Ta, Ha and Pa in HCl solutions. Radiochim Acta 80, 75–84.
Pershina, V. and Bastug, T. 1999. Solution chemistry of element 105-Part III: Hydrolysis
and complex fomation of Nb, Ta, Db and Pa in HF and HBr solutions. Radiochim Acta
84, 79–84.
Pershina, V. and Bastug, T. 2000. The electronic structure and properties of group 7 oxychlo-
rides, MO3 Cl, where M=Tc, Re, and element 107, Bh. J Chem Phys 113, 1441–1446.
Pershina, V. and Fricke, B. 1996. Group 6 dioxydichlorides M2 Cl2 (M=Cr, Mo, W, and
element 106, Sg)—the electronic structure and thermochemical stability. J Phys Chem
100, 8748–8751.
Pershina, V. and Fricke, B. 1999. Electronic structure and chemistry of the heaviest elements
in Heavy Elements and Related New Phenomena. vol. 1. pp. 184–262. Greiner, W. and
Gupta, R. K., eds. Singapore: World Scientific.
Pershina, V. and Hoffman, D. C. 2003. The chemistry of the heaviest elements in Theoretical
chemistry and physics of heavy and superheavy elements in the series, “Progress in
Theoretical Chemistry and Physics”. Chapter 3. pp. 55–114. Kaldor, U. and Wilson, S.,
eds. Dordrecht, The Netherlands: Kluwer Academic Publishers.
Pershina, V. and Kratz, J. V. 2001. Solution chemistry of element 106: Theoretical predic-
tions of hydrolysis of group 6 cations Mo, W and Sg. Inorg Chem 40, 776–780.
Pershina, V., Sepp, W.-D., Bastug, T., Fricke, B., and Ionova, G. V. 1992. Relativistic effects
in physics and chemistry of element 105. III. Electronic structure of hahnium oxyhalides
as analogs of group 5 elements oxyhalides, J Chem Phys 97, 1123–1131.
Pershina, V., Bastug, T., Fricke, B., and Varga, S. 2001. The electronic structure and prop-
erties of group 8 oxides MO4 , where M=Ru, Os, and Element 108, Hs. J Chem Phys
115, 792–799.
Pershina, V., Bastug, T., Jacob, T., Fricke, B., and Varga, S. 2002. Intermetallic compounds
of the heaviest elements: The electronic structure and bonding of dimers of element 112
and its homolog Hg. Chem Phys Lett 365, 176–183.
 References 459

Pibida, L., Nortershauser, W., Hutchinson, J. M. R., and Bushaw, B. A. 2001. Evaluation of
resonance ionization mass spectrometry for the determination of Cs-135/Cs-137 isotope
ratios in low-level samples. Radiochimica Acta 89(3), 161–168.
Pitzer, K. S. 1975. Are elements 112, 114, and 118 relatively inert gases? J Chem Phys 63,
1032–1033.
Platzner, I. T. 1997. Modern Isotope Ratio Mass Spectrometry. New York, NY: John
Wiley.
Platzner, I. T., Becker, J. S., and Dietze, H. J. 1999. Stability study of isotope ratio mea-
surements for uranium and thorium by ICP-QMS. Atomic Spectroscopy 20(1), 6–12.
Porter, C. R. and Kahn, B. 1964. Improved determination of strontium-90 in milk by an
ion-exchange method. Anal Chem 36, 676–678.
Porter, C. R., Kahn, B., Carter, M. W., Rehnberg, G. L., and Pepper, E. W. 1967. De-
termination of radiostrontium in food and other environmental media. Env Sci Tech 1,
745–750.
Poupard, D. and Jouniaux, B. 1990. Determination of picogram quantities of americium
and curium by thermal ionization mass-spectrometry (Tims). Radiochimica Acta 49(1),
25–28.
Praphairaksit, N. and Houk, R. S. 2000. Reduction of mass bias and matrix effects in
inductively coupled plasma mass spectrometry with a supplemental electron source in a
negative extraction lens. Anal Chem 72, 4435–4440.
Priest, N. D., Pich, G. M., Fifield, L. K., and Cresswell, R. G. 1999. Accelerator mass
spectrometry for the detection of ultra-low levels of plutonium in urine, including that
excreted after the ingestion of Irish sea sediments. Radiat Res 152(6), S16–S18.
Probst, T. U. 1996. Studies on the long-term stabilities of the background of radionuclides
in inductively coupled plasma mass spectrometry (ICP-MS)—A review of radionuclide
determination by ICP-MS. Fresen J Anal Chem 354(7–8), 782–787.
Raisbeck, G. M. and Yiou, F. 1999. I-129 in the oceans: origins and applications. Sci Total
Environ 238, 31–41.
Ratliff, T. A. 2003. Laboratory Quality Assurance System. 3rd ed. pp. 4, 31–32. Hoboken,
NJ: John Wiley.
Rehkamper, M., Schonbachler, M., and Stirling, C. H. 2001. Multiple collector ICP-MS: In-
troduction to instrumentation, measurement techniques and analytical capabilities. Geo-
standard Newslett 25(1), 23–40.
Reyes, L. H., Gayon, J. M. M., Alonso, J. I. G., and Sanz-Medel, A. 2003. Determination
of selenium in biological materials by isotope dilution analysis with an octapole reaction
system ICP-MS. J Anal Atom Spectrom 18(1), 11–16.
Richardson, S. D. 2001. Mass spectrometry in environmental sciences. Chem Rev 101(2),
211–254.
Riegel, J., Deissenberger, R., Herrmann, G., Kohler, S., Sattelberger, P., Trautmann, N.,
Wendeler, H., Ames, F., Kluge, H. J., Scheerer, F., and Urban, F. J. 1993. Resonance
ionization mass spectroscopy for trace analysis of neptunium. Appl Phys B Laser O
56(5), 275–280.
Rieman, W. and Walton, H. F. 1970. Ion Exchange in Analytical Chemistry. New York, NY:
Pergamon Press.
Roberts, M. L. and Caffee, M. W. 2000. I-129 interlaboratory comparison: Phase II results.
Nucl Instrum Meth Phys B 172, 388–394.
Rodushkin, I., Lindahl, P., Holm, E., and Roos, P. 1999. Determination of plutonium con-
centrations and isotope ratios in environmental samples with a double-focusing sector
field ICP- MS. Nucl Instrum Meth Phys A 423(2–3), 472–479.
460 References

Rokop, D. J., Perrin, R. E., Knobeloch, G. W., Armijo, V. M., and Shields, W. R. 1982.
Thermal ionization mass spectrometry of uranium with electrodeposition as a loading
technique. Anal Chem 54(6), 957–960.
Rondinella, V. V., Betti, M., Bocci, F., Hiernaut, T., and Cobos, J. 2000. IC-ICP-MS applied
to the separation and determination of traces of plutonium and uranium in aqueous
leachates and acid rinse solutions of UO2 doped with Pu-238. Microchem J 67(1–3),
301–304.
Rosen, A. 1998. Twenty to thirty years of DV-X α calculation: A survey of accuracy and
application. Adv Quant Chem 29, 1–30.
Rosson, R., Jakiel, R., Klima, S., Kahn, B., and Fledderman, P. 2000. Correcting tritium
concentrations in water vapor monitored with silica gel. Health Phys 78, 68–73.
Rucklidge, J. 1995. Accelerator mass-spectrometry in environmental geoscience—A re-
view. Analyst 120(5), 1283–1290.
Russo, R. E., Mao, X. L., Liu, H. C., Gonzalez, J., and Mao, S. S. 2002. Laser ablation in
analytical chemistry—A review. Talanta 57(3), 425–451.
Sahoo, S. K., Yonehara, H., Kurotaki, K., Fujimoto, K., and Nakamura, Y. 2002. Precise
determination of U-235/U-238 isotope ratio in soil samples by using thermal ionisation
mass spectrometry. J Radioanal Nucl Ch 252(2), 241–245.
Schädel, M. 1995. Chemistry of the transactinide elements. Radiochim Acta 70/71, 207–
223.
Schädel, M., Brüchle, W., Dressler, R., Eichler, B., Gäggeler, H. W., Günther, R., Gregorich,
K. E., Hoffman, D. C., Hübener, S., Jost, D. T., Kratz, J. V., Paulus, W., Schumann, D.,
Timokhin, S., Trautmann, N., Türler, A., Wirth, G., and Yakuschev, A. 1997a. Chemical
properties of element 106 (seaborgium). Nature Lett 388, 55–57.
Schädel, M., Brüchle, W., Schausten, B., Schimpf, E., Jäger, E., Wirth, G., Günther, R.,
Kratz, J. V., Paulus, W., Seibert, A., Thorle, P., Trautmann, N., Zauner, S., Schumann, D.,
Andrassy, M., Misiak, R., Gregorich, K. E., Hoffman, D. C., Lee, D. M., Sylwester, E. R.,
Nagame, Y., and Oura, Y. 1997b. First aqueous chemistry with seaborgium (element 106).
Radiochim Acta 77, 149–159.
Schädel, M., Brüchle, W., Jäger, E., Schausten, B., Wirth, G., Paulus, W., Günther, R.,
Eberhardt, K., Kratz, J. V., Seibert, A., Strub, E., Thörle, P., Trautmann, N., Waldek,
W., Zauner, S., Schumann, D., Kirbach, U., Kubica, B., Misiak, R., Nagame, Y., and
Gregorich, K. E. 1998. Aqueous chemistry of seaborgium (Z=106). Radiochim Acta 83,
163–165.
Schaumloffel, D., Giusti, P., Zoriy, M. V., Pickhardt, C., Szpunar, J., Lobinski, R., and
Becker, J. S. 2005. Ultratrace determination of uranium and plutonium by nano-volume
flow injection double-focusing sector field inductively coupled plasma mass spectrometry
(nFI-ICP-SFMS). J Anal Atom Spectrom 20(1), 17–21.
Schmitt, G., Markell, C., and Hagen, D. 1990. Membrane approach to solid-phase extrac-
tions. Anal Chim Acta 236, 157–164.
Schneier, B. 2000. Secrets & Lies: Digital Security in a Networked World. New York, NY:
John Wiley.
Schulze, J., Auer, M., and Werzi, R. 2000. Low level radioactivity measurement in support
of the CTBTO. Appl Radiat Isotopes 53, 23–30.
Scoles, G. 1988. Atomic and Molecular Beam Methods. Vol. 1. Oxford: Oxford University
Press.
Seaborg, G. T. 1945. The chemical and radioactive properties of the heavy elements. Chem
Eng News 23, 2190.
Seaborg, G. T. 1967. Actinides reviews. Actin Rev 1, 3–38.
 References 461

Seaborg, G. T., and Keller, O. L., Jr. 1986. Future elements in The Chemistry of the Actinide
Elements. 2nd ed. Katz, J. J. Seaborg, G. T. and Morss, L. R. eds. pp. 1635–1643. London:
Chapman and Hall.
Seaborg, G. T., Wahl, A. C., and Kennedy, J. W. 1946. Radioactive element 94 from
deuterons on uranium. Phys Rev 69, 367.
Sekine, T. and Hasegaw, Y. 1977. Solvent Extraction Chemistry: Fundamentals and Appli-
cations. New York, NY: Marcel Dekker.
Seth, M., Schwerdtfeger, P., Dolg, M., Faegri, K., Hess, B. A., and Kaldor, U. 1996. Large
relativistic effects in molecular properties of the hydride of superheavy element 111.
Chem Phys Lett 250, 461–465.
Seth, M., Cooke, F., Schwerdtfeger, P., Heully, J.-L.,and Pelissier, M. 1998a. The chemistry
of the superheavy elements. II. The stability of high oxidation states in group 11 elements:
Relativistic coupled cluster calculations for the di-, tetra- and hexafluoro metallates of
Cu, Ag, Au, and element 111. J Chem Phys 109, 3935–3943.
Seth, M., Faegri, K., and Schwerdtfeger, P. 1998b. The stability of the oxidation state +4
in group 14 compounds from carbon to element 114. Angew Chem Int Ed Engl 37,
2493–2496.
Seubert, A. 2001. On-line coupling of ion chromatography with ICP-AES and ICP-MS.
Trac-Trend in Anal Chem 20(6–7), 274–287.
Shen, C. C., Cheng, H., Edwards, R. L., Moran, S. B., Edmonds, H. N., Hoff, J. A., and
Thomas, R. B. 2003. Measurement of attogram quantities of Pa-231 in dissolved and
particulate fractions of seawater by isotope dilution thermal ionization mass spectroscopy.
Anal Chem 75(5), 1075–1079.
Sherwin, C. W. 1951.Vacuum evaporation of radioactive materials. Rev Sci Instrum 22 (5),
339–341.
Shleien, B. ed. 1992. The Health Physics and Radiological Health Handbook. Silver
Springs, MD: Scinta.
Shleien, B., Slaback, L., Jr., and Birky, B. 1998. Handbook of Health Physics and Radio-
logical Health. 3rd ed. Baltimore, MD: Lippincott.
Shuping, R. E., Philips, C. R., and Moghissi, A. A. 1970. Krypton-85 levels in the envi-
ronment determined from dated krypton gas samples. Radiol Health Data R 11, 667–
672.
Shuter, E. and Teasdale, W. E. 1989. Application of drilling coring, and sampling techniques
to test holes and wells in Techniques of Water-Resources Investigations. Chapter F1.
Book 2. Reston, VA: US Geological Survey.
Sill, C. W. and Olson, D. G. 1970. Sources and prevention of recoil contamination of
solid-state alpha detectors. Anal Chem 42, 1596–1607.
Sill, C. W. and Williams, R. L. 1969. Radiochemical determination of uranium and the
transuranium elements in process solutions and environmental samples. Anal Chem
41(12), 1624–1632.
Sill, C. W., Puphal, K. W., and Hindman, F. D. 1974. Simultaneous determination of alpha-
emitting nuclides of radium through californium in soil. Anal Chem 46, 1725–1737.
Silva, R., Harris, J., Nurmia, M., Eskola, K., and Ghiorso, A. 1970. Chemical separation of
rutherfordium. Inorg Nucl Chem Lett 6, 871–877.
Skipperud, L. and Oughton, D. H. 2004. Use of AMS in the marine environment. Environ
Int 30(6), 815–825.
Smart, J. E. 1998. Real-time airborne radiation analysis and collection (RTARAC) searching
for airborne species characteristic to nuclear proliferation. J Radioanal Nucl Ch. 235,
105–108.
462 References

Smith, D. H. 2000. Thermal ionization mass spectrometry in Inorganic Mass Spectrometry:

Fundamentals and Applications. Vol. 23. Barshick, C. M., Duckworth, D. C., and Smith,
D. H., eds. pp. 1–30. New York, NY: Marcel Dekker.
Smith, D. H. and Carter, J. A. 1981. A simple method to enhance thermal emission of metal
ions. Int J Mass Spectrom and Ion Processes 40(2), 211–215.
Smolańczuk, R. 1997. Properties of the hypothetical spherical superheavy nuclei. Phys Rev
C 56, 812–824.
Smolańczuk, R. 2001a. Production and decay of element 120. Phys Lett B 509, 227–
230.
Smolańczuk, R. 2001b. Formation of superheavy elements in cold fusion reactions. Phys
Rev C 63, 044607-1-8.
Song, M., Probst, T. U., and Berryman, N. G. 2001. Rapid and sensitive determination
of radiocesium (Cs-135, Cs-137) in the presence of excess barium by electrothermal
vaporization-inductively coupled plasma-mass spectrometry (ETV- ICP-MS) with potas-
sium thiocyanate as modifier. Fresen J Anal Chem 370(6), 744–751.
Standard Methods 2005. Standard Methods for the Evaluation of Water and Wastew-
ater. American Public Health Association (APHA), American Water Works As-
sociation (AWWA) and Water Environment Federation (WEF), 2005. Online at
http://www.standardmethods.com/store/. Click “Browse Standard Methods.” Viewed on
January 2006.
Steinberg, E. P. 1962. Counting methods for the assay of radioactive samples in Nuclear
Instruments and Their Uses. Chapter 5. Snell, A. H., ed. New York, NY: John Wiley.
Stevenson, P. C. and Nervik, W. E. 1961. The Radiochemistry of the Rare Earths: Scandium,
Yttrium, and Actinium. National Research Council Report NAS-NS 3020. Washington,
DC: National Research Council.
Stoffels, J. J., Wacker, J. F., Kelley, J. M., Bond, L. A., Kiddy, R. A., and Brauer, F. P. 1994.
Environmental monitoring of Hanford nuclear facility effluents by thermal ionization
mass-mpectrometry. Appl Spectrosc 48(11), 1326–1330.
Strub, E., Kratz, J. V., Kronenberg, A., Nähler, A., Thörle, P., Zauner, S., Brüchle, W., Jäger,
E., Schädel, M., Schausten, B., Schimpf, E., Zongwei, L., Kirbach, U., Schumann, D.,
Jost, D., Türler, A., Asai, M., Nagame, Y., Sakara, M., Tsukada, K., Gäggeler, H. W., and
Glanz, J. P. 2000. Fluoride complexation of rutherfordium (Rf, element 104). Radiochim
Acta 88, 265.
Sturup, S., Dahlgaard, H., and Nielsen, S. C. 1998. High resolution inductively coupled
plasma mass spectrometry for the trace determination of plutonium isotopes and isotope
ratios in environmental samples. J Anal Atom Spectrom 13(12), 1321–1326.
Sulcek, Z. and Povondra, P. 1989. Methods of Decomposition in Inorganic Analysis. Boca
Raton, FL: CRC Press.
Sunderman, D. N. and Townley, C. W. 1960. The Radiochemistry of Barium, Calcium and
Strontium. National Research Council Report NAS-NS 3010. Washington, DC: National
Research Council.
Sylwester, E. R., Gregorich, K. E., Lee, D. M., Kadkhodayan, B., Türler, A., Adams, J. L.,
Kacher, C. D., Lane, M. R., Laue, C. A., McGrath, C. A., Shaughnessy, D. A., Strellis,
D. A., Wilk, P. A., and Hoffman, D. C. 2000. On-line gas chromatographic studies of Rf,
Zr, and Hf bromides. Radiochim Acta 88, 837–843.
Takiue, M. and Ishikawa, H. 1978. Thermal neutron reaction cross section measurements
for fourteen nuclides with a liquid scintillation spectrometer. Nucl Instrum Methods 148,
157–161.
 References 463

Talvitie, N. A. 1972. Electrodeposition of actinides for alpha spectrometric determination.

Anal Chem 44, 280–283.
Taylor, R. N., Croudace, I. W., Warwick, P. E., and Dee, S. J. 1998. Precise and rapid
determination of U-238/U-235 and uranium concentration in soil samples using thermal
ionisation mass spectrometry. Chem Geol 144(1–2), 73–80.
Taylor, R. N., Warneke, T., Milton, J. A., Croudace, I. W., Warwick, P. E., and Nesbitt, R. W.
2001. Plutonium isotope ratio analysis at femtogram to nanogram levels by multicollector
ICP-MS. J Anal Atom Spectrom 16(3), 279–284.
Thiers, R. E. 1957. Separation, concentration, and contamination in Trace Analysis. Yoe,
J. H. and Koch, H. J., eds. pp. 637–666. New York, NY: John Wiley.
Timokhin, S. N., Yakushev, A. B., Xu, H. G., Perelygin, V. P., and Zvara, I. 1996. Chemical
identification of element 106 by thermochromatography. J Radioanal Nucl Ch Lett 212,
31–34.
Ting, B. G., Pappas, R. S., and Paschal, D. C. 2003. Rapid analysis for plutonium-239 in
urine by magnetic sector inductively coupled plasma-mass spectrometry using Aridus
desolvation introduction system. J Anal Atom Spectrom 18(7), 795–797.
Tolgyessy, J. and Bujdoso, E. 1991. CRC Handbook of Radioanalytical Chemistry. Boca
Raton, FL: CRC Press.
Tolgyessy, J., Varga, S., and Krivan, V. 1971. Nuclear Analytical Chemistry. 5 books.
Baltimore, MD: University Park Press.
Trautmann, N., Passler, G., and Wendt, K. D. A. 2004. Ultratrace analysis and isotope ratio
measurements of long-lived radioisotopes by resonance ionization mass spectrometry
(RIMS). Anal Bioanal Chem 378(2), 348–355.
Tuniz, C. 2001. Accelerator mass spectrometry: Ultra-sensitive analysis for global science.
Radiat Phys Chem 61(3–6), 317–322.
Tuniz, C., Bird, J. R., Fink, D., and Herzog, G. F. 1998. Accelerator Mass Spectrometry:
Ultrasensitive Analysis for Global Science. Baton Rouge, FL: CRC Press.
Türler, A., Gäggeler, H. W., Gregorich, K. E., Barth, H., Brüchle, W., Czerwinski, K. R.,
Gober, M. K., Hannink, N. J., Henderson, R. A., Hoffman, D. C., Jost, D. T., Kacher,
C. D., Kadkhodayan, B., Kovacs, J., Kratz, J. V., Kreek, S. A., Lee, D. M., Leyba,
J. D., Nurmia, M. J., Schädel, M., Scherer, U. W., Schimpf, E., Vermeulen, D., Weber,
A., Zimmermann, H. P., and Zvara, I. 1992. Gas phase chromatography of halides of
element-104 and Element-105. J Radioanal Nucl Ch Ar 160, 327–339.
Türler, A., Eichler, B., Jost, D. T., Piguet, D., Gäggeler, H. W., Gregorich, K. E., Kad-
khodayan, B., Kreek, S. A., Lee, D. M., Mohar, M., Sylwester, E., Hoffman, D. C., and
Hübener, S. 1996. On-line gas chromatography studies of chlorides of rutherfordium and
homologs Zr and Hf. Radiochim Acta 73, 55–66.
Türler, A., Dressler, R., Eichler, B., Gäggeler, H. W., Jost, D. T., Schädel, M., Brüchle, W.,
Gregorich, K. E., Trautmann, N., and Taut, S. 1998. Decay properties of 265 Sg(Z=106)
and 266 Sg(Z=106). Phys Rev C 57, 1648–1655.
Türler, A., Brüchle, W., Dressler, R., Eichler, B., Eichler, R., Gäggeler, H. W., Gärtner,
M., Glatz, J. P., Gregorich, K. E., Hübener, S., Jost, D. T., Lebedev, V. Y., Pershina, V.
G., Schädel, M., Taut, S., Timokhin, S. N., Trautmann, N., Vahle, A., and Yakushev, A.
B. 1999. First measurement of a thermochemical property of a seaborgium compound.
Angew Chem Int Ed 38, 2212–2213.
Uchida, S., Garcia-Tenorio, R., Tagami, K., and Garcia-Leon, M. 2000. Determination of
U isotopic ratios in environmental samples by ICP-MS. J Anal Atom Spectrom 15(7),
889–892.
464 References

Underhill, D. W. 1996. The adsorption of argon, krypton and xenon on activated charcoal.
Health Phys 71 (2), 160–166.
UNSCEAR 2000a. United Nations Scientific Committee on the Effects of Atomic Radiation,
Annex B. Exposures from Natural Radiation Sources. Vienna: UN.
UNSCEAR 2000b. United Nations Scientific Committee on the Effects of Atomic Radiation,
Annex C. Exposures from Man-made Sources of Radiation. Vienna: UN.
Vanhaecke, F. and Moens, L. 1999. Recent trends in trace element determination and speci-
ation using inductively coupled plasma mass spectrometry. Fresen J Anal Chem 364(5),
440–451.
Vertes, A., Nagy, S., and Klencsar Z., eds. 2003. Handbook of Nuclear Chemistry. 5 books.
Dordrecht: Kluwer Academic.
Vonderheide, A. P., Montes-Bayon, M., and Caruso, J. A. 2002. Development and appli-
cation of a method for the analysis of brominated flame retardants by fast gas chro-
matography with inductively coupled plasma mass spectrometric detection. J Anal Atom
Spectrom 17(11), 1480–1485.
Waber, J. T. and Averill, F. W. 1974. Molecular orbitals of PtF6 and E110 F6 calculated by
the self-consistent multiple scattering X α method. J Chem Phys 60, 4460–4470.
Wahl, A. C. and Bonner, N. A. 1951. Radioactivity Applied to Chemistry. New York: John
Wiley.
Wakat, M. A. 1971. Catalogue of γ -rays emitted by radionuclides. Nucl Data Tables 8,
445–666.
Walker, R. L., Carter, J. A., and Smith, D. H. 1981. A bulk resin bead procedure to obtain
uranium and plutonium from radioactive solutions for mass-spectrometric analysis. Anal
Lett Pt A 14(19), 1603–1612.
Walton, H. F. and Rocklin, R. D. 1990. Ion Exchange in Analytical Chemistry. Boca Raton,
FL: CRC Press.
Wang, Z., Kahn, B., and Valentine, J. D. 2002. Efficiency calculation and coincidence
summing correction for germanium detectors by Monte Carlo simulation. IEEE T Nucl
Sci 49, 1925–1931.
Watanabe, K., Iguchi, T., Ogita, T., Uritani, A., and Harano, H. 2001. Development of failed
fuel detection and location technique using resonance ionization mass spectrometry. J
Nucl Sci Technol 38(10), 844–849.
Weast, R. C. 1985. CRC Handbook of Chemistry and Physics. Boca Raton, FL: CRC
Press.
Wendt, K., Bhowmick, G. K., Bushaw, B. A., Herrmann, G., Kratz, J. V., Lantzsch,
J., Muller, P., Nortershauser, W., Otten, E. W., Schwalbach, R., Seibert, U. A., Traut-
mann, N., and Waldek, A. 1997. Rapid trace analysis of Sr-89, Sr-90 in environmental
samples by collinear laser resonance ionization mass spectrometry. Radiochim Acta 79(3),
183–190.
Wendt, K., Blaum, K., Bushaw, B. A., Gruning, C., Horn, R., Huber, G., Kratz, J. V., Kunz,
P., Muller, P., Nortershauser, W., Nunnemann, M., Passler, G., Schmitt, A., Trautmann, N.,
and Waldek, A. 1999. Recent developments in and applications of resonance ionization
mass spectrometry. Fresen J Anal Chem 364(5), 471–477.
Wendt, K., Trautmann, N., and Bushaw, B. A. 2000. Resonant laser ionization mass spec-
trometry: An alternative to AMS? Nucl Instrum Meth B 172, 162–169.
Wilk, P. A., Gregorich, K. E., Türler, A., Laue, C. A., Eichler, R., Ninov, V., Adams, J.
L., Kirbach, U. W., Lane, M. R., Lee, D. M., Patin, J. B., Shaughnessy, D. A., Strellis,
D. A., Nitsche, H., and Hoffman, D. C. 2000. Evidence for new isotopes of element 107:
266
Bh and 267 Bh. Phys Rev Lett 85, 2697–2700.
 References 465

Williams, R. L. and Grothaus, G. E. 1984. Determination of actinides at the radiological

and environmental sciences laboratory. Nucl Instrum Meth A 223 (2–3), 200–203.
Wilson, C. L., Wilson, D. W., and Strouts C. R. N. 1968. Physical separation methods in
Comprehensive Analytical Chemistry. Volume IIB. New York, NY: Elsevier.
Winefordner, J. D., Gornushkin, I. B., Pappas, D., Matveev, O. I., and Smith, B. W.
2000. Novel uses of lasers in atomic spectroscopy. J Anal Atom Spectrom 15(9), 1161–
1189.
Winn, W. G. 1993. A capability for monitoring effluent tritium with liquid scintillation/
robotics. Tritium monitoring in the environment.in Transactions of the American Nuclear
Society. Vol. 69. Paper 8. pp. 26. LaGrange Park, IL: American Nuclear Society.
Wiza, J. L. 1979. Microchannel plate detectors. Nucl Instrum Methods 162(1–3), 587–
601.
Wood, W. W. 1976. Guidelines for collection and field analysis of ground-water samples for
selected unstable constituents in Techniques of Water-Resources. Book 1. Chapter D2.
Reston, VA: US Geological Survey.
Wyse, E. J. and Fisher, D. R. 1994. Radionuclide bioassay by inductively-coupled plasma-
mass spectrometry (ICP/MS). Radiat Prot Dosim 55(3), 199–206.
Wyse, E. J., MacLellan, J. A., Lindenmeier, C. W., Bramson, J. P., and Koppenaal, D. W.
1998. Actinide bioassays by ICPMS. J Radioanal Nucl Ch 234(1–2), 165–170.
Yakushev, A. B., Timokhin, S. N., Vedeneev, M. V., Xu, H. G., and Zvara, I. 1996. Compar-
ative study of oxochlorides of molybdenum, tungsten and element 106. J Radioan Nucl
Ch Ar 205, 63–67.
Yakushev, A. B., Buklanov, G. V., Chelnokov, M. L., Chepigin, V. I., Dmitriev, S. N.,
Gorshkov, V. A., Hübener, S., Lebedev, V. Y., Malyshev, O. N., Oganessian, Y. T., Popeko,
A. G., Sokol, E. A., Timokhin, S. N., Türler, A., Vasko, V. M., Yeremin, A. V., and
Zvara, I. 2001. First attempt to chemically identify element 112. Radiochim Acta 89, 743–
745.
Yamamoto, M., Syarbaini, K., Kofuji, K., Tsumura A., Komura, K., Ueno, K., and Assinder,
D. J. 1995. Determination of low-level Tc-99 in environmental samples by high-resolution
ICP-MS. J Radioan Nucl Ch Ar 197(1), 185–194.
Yin, Q. Z., Jacobsen, S. B., Lee, C. T., McDonough, W. F., Rudnick, R. L., and Horn, I.
2001. A gravimetric K2 OsC16 standard: Application to precise and accurate Os spike
calibration. Geochim Cosmochim Ac 65(13), 2113–2127.
Yiou, F., Raisbeck, G., and Imbaud, H. 2004. Extraction and AMS measurement of carrier
free I-129/I-127 from seawater. Nucl Instrum Meth B 223–224, 412–415.
Yokoyama, T., Makishima, A., and Nakamura, E. 2001. Precise analysis of U-234/U-238
ratio using UO2+ ion with thermal ionization mass spectrometry for natural samples.
Chem Geol 181(1–4), 1–12.
Yu, L. L., Fassett, J. D., and Guthrie, W. F. 2002. Detection limit of isotope dilution mass
spectrometry. Anal Chem 74(15), 3887–3891.
Zhao, X. L., Kilius, L. R., Litherland, A. E., and Beasley, T. 1997. AMS measurement of
environmental U-236 preliminary results and perspectives. Nucl Instrum Meth B 126(1–
4), 297–300.
Zoriy, M. V., Pickhardt, C., Ostapczuk, P., Hille, R., and Becker, J. S. 2004. Determina-
tion of Pu in urine at ultratrace level by sector field inductively coupled plasma mass
spectrometry. Int J Mass Spectrom 232(3), 217–224.
Zumwalt, L. R. 1950. Atomic Energy Commission Report AECU-567. Absolute beta count-
ing using end-window Geiger-Mueller counters and experimental data on beta-particle
scattering effects. Washington, DC: AEC.
466 References

Zvara, I., Belov, V. Z., Domanov, V. P., Korotkin, Y. S., Chelnokov, L. P., Shalayevsky,
M. R., Shchegolev, V. A., and Hussonois, M. 1972. Chemical isolation of Kurchatovium.
Sov Radiochem 14, 115–118.
Zvara, I., Aikhler, V., Belov, V. Z., Zvarova, T. S., Korotkin, Y. S., Shalayevsky, M. R.,
Shchegolev, V. A., and Hussonois, M. 1974. Gas chromatography and thermochromato-
graphy in the study of transuranium elements. Sov Radiochem 16, 709–715.
Zvara, I., Belov, V. Z., Domanov, V. P., and Shalayevsky, M. R. 1976. Chemical isolation of
nielsbohrium as ekatantalum in the form of the anhydrous bromide; II experiments with
a spontaneously fissioning isotope of nielsbohrium. Sov Radiochem 18, 328–334.
Index

absorption Auger electrons, 28, 108

alpha particle, 63 automation, 6, 85, 86, 108, 290, 318, 319, 325,
beta particle, 25 329, 336
accelerator mass spectrometry (AMS), 364, 383, in sampling, 318–319
398–401 in separation, 318–319
accelerators sample changers, 325–326
Cockcroft-Walton, 17, 421 spectral analysis, 323
accelerators, 1, 341, 398 avalanche, 30
accident, 2, 94, 256, 298, 431 Avogadro’s number, 13, 422
equipment, 301–302 azeotrope, 58
protocol, 310, 313–314
response, 284 back scattering, 125, 139
actinides, 96, 338–341, 393 background radiation, 35, 142–147
activation, 3, 93, 256 compensating for, 142–147
neutron, 103 sources, 142–147
products, 322, 407 Bateman equation, 14
chemical, 117 beta particle, 95, 307, 414
radiation, 409, 415 spectrometry, 151
adsorption, 49, 346, 424 efficiency tracing, 155
Agreement States, 286, 296, 421 beta particle emission, 8, 116
alpha particle energy of, 123, 125–126
counting, 37 range, 233
range, 20, 124, 398 decay, 172, 178–179
alpha particle emission, 8, 11, 425 spectrometry, 151, 165
energy of, 9, 66, 123 binding energy,10, 11, 27, 28, 160
analyst, 7, 93, 95, 101, 120–121, 129, 211, 293, bioassay samples, 91, 100, 261, 302
362, 414 blank, 81, 92, 101, 190, 203, 205–207, 210,
as expert witness, 242 211–212, 230, 259, 320, 358, 365
behavior, 248 branching ratio,
anion exchange, 47, 48, 52, 85 Bremsstrahlung radiation, 142, 423
antineutrino, 9, 422
aqua regia, 70, 277, 422 calibration, 37, 123, 202, 320, 366, 400
Atomic Energy Act, 284, 421 carrier, 36, 98, 211, 334, 346, 400
attenuation, 21, 123, 255 isotopic, 73, 75, 101, 121–122
attribution, 92, 220, 422 non-isotopic, 74–75, 107, 122
audit reports, 222, 236–237 carrier-free, 54, 63, 74, 76
Auger electron emission cation exchange, 47, 99, 351
competing with X ray emission, 11, 422 cesium, 34, 108, 400

467
468 Index

chain of custody, 78, 186, 222, 298 in Ge detectors, 140, 168, 258, 322
chelation, 52 in LS, 34, 37–38, 97, 100, 105–106, 108–109,
Chemical Hygiene Officer (CHO), 295–296, 117, 127–128, 130, 134, 137, 145,
302–303, 308, 314 151–153, 158, 165–166, 183–184, 255,
Chemical Hygiene Plan (CHP), 295, 297, 308, 327
310, 312–313, 316 in organic scintillators, 34
chemical hygiene, 294–295, 316 in surface barrier detectors, 32, 157, 348, 360
staffing, 295–296 practical counting efficiency, 128, 135, 137,
Cherenkov radiation, 18, 23, 130, 152, 423 190
chromatography, 46, 54, 348, 353, 364, 387, 418 counting
gas, 353, 364, 387 alpha-particle, 124, 126, 150
liquid, 348 beta-particle, 37, 75, 95, 109, 125, 127, 151,
paper, 46 180, 182, 184, 253, 255
clear melt, 423 gamma-ray, 161, 180, 201
Code of Federal Regulations (CFR), 271, 284, gross alpha/beta, 95, 98, 100, 126–127,
286, 293, 295–296, 303–304, 308–310, 182–183, 253
314–316 liquid scintillation (see liquid scintillation)
explained, 314–316 statistics, 37, 198–199
coincidence, 34, 141, 145 counting efficiency, 13, 34, 36, 37, 103, 106,
combustible liquids, 303–304, 423 110
command and control, 320–321 critical value, 189, 204
calculation of, 205
complex-forming, 48
and MDA, 204
Comprehensive Nuclear Test Ban Treaty
cross-section, 45, 47, 373
(CNTBT), 329
Curie, Marie, 338
compton scattering, 23–25, 28, 140, 146, 160,
161, 424
Data Quality Objectives (DQOs), 238
contaminants, 39–40, 43–44, 48, 50–51, 57, 105,
108, 121, 128, 132, 145, 164, 171, 203, data
210, 224, 230, 258, 270, 273, 302, 312, 332 assessment, 222, 236
control chart, 208–211, 231–232 authentication, 321–322, 430
conversion electrons, 142 presentation of, 6, 190, 216–219
emission of, 163, 166 review, 190, 213–216
co-precipitation, 3, 42, 66-68, 74, 109, 110, 112, security, 279, 281–283
113 surety, 321–322
correction validation, 241, 273
energy, 15, 140 verification, 220
mass, 133 Daubert v. Merrill Dow, 243
cosmic rays, 1, 14, 103, 144 dead time (see also resolving time)
cost decay constant, 12, 176, 178, 191, 209,
laboratory, 288 425
of analysis, 99, 222, 290 decay fraction, 116, 137, 142, 162, 166–168,
counters (see also detectors) 175, 179, 190, 425
Geiger-Mueller, 29, 139, 148, 427 decay schemes
ionization chambers, 29–30, 148, 162 positron emission and electron capture, 181
maintenance alpha particle and gamma ray, 181
proportional, 97, 100, 105–106, 109, 116, beta particle, 179
117, 128, 134, 137, 139–140, 145, 148, 151 beta particle and electron capture, 179–180
and source thickness, 123–124 beta particle and gamma ray, 180
calculation of, 123–126 conventions, 172
for alpha particles, 124, 126, 150 decommissioning, 94, 220–221, 286
for beta particles, 37, 75, 95, 109, 125, 127, decontamination factor (DF), 39, 95, 102, 171
151, 180, 182, 184, 253, 255 defensibility of results, 242–243
for gamma rays, 131–133 depletion layer, 33
 Index 469

detector (see also counters) Fano factor, 36, 426

cadmium zinc telluride, 162 Faraday Cup, 379, 380, 387, 395, 399
geometry, 124, 136, 138, 150, 190, 329–330 Federal Facilities Compliance Act, 426
germanium, 105, 109, 131, 132, 138, fire extinguishers, 310–312
140–141, 160–161, 166, 168, 179, 181, hazard classes, 310–311
183, 255, 258, 322, 330 location, 310
lanthanum chloride or bromide, 162 maintenance, 311
liquid scintillation (see scintillation detectors) fission fragments (FF), 8
NaI(Tl), 33–34, 36–37, 132, 158–161, 183, fission
322, 415 asymmetric, 11, 422
surface barrier, 32–33, 157, 348, 360 induced, 11, 17, 428
solid state, 31–33, 129, 140, 158–159, 346, spontaneous, 8, 11, 341, 360, 434
348 symmetric, 11, 435
disintegration rate, 1, 12–14, 16, 37 flammable liquids, 303, 311
dissolution, 69–70 fume hoods
distillation, 39, 56–59 airflow, 270
azeotropic, 58–59, 104 corrosion of, 70
distribution coefficient, 44, 48, 348 maintenance, 270
distribution washdown, 270
Gaussian, 200, 427 fusion
normal, 19, 196, 206, 427 chemical, 71
Poisson, 137, 199–201, 207, 426–427
gamma rays, 10, 25, 212
efficiency gamma-ray emission
counting (see counting efficiency) energy of, 8–9, 19, 417
electrodeposition, 61–63, 67, 75, 111, 113, 185, gas cylinders
411 handling of, 304
electron capture, 11, 66, 105, 175, 179, 181 Gaussian distribution, 200
competing with positron emission, 181, error curve, 200
425–425 Geiger-Meuller counter, 29, 139, 167, 427
electron-hole pair glow discharge MS (GD-MS), 365–366, 379,
electron impact ionization, 365–366 405–406
emanation, de-emanation (of radon), 110, 276 Guide to the Expression of Uncertainty in
emergency Measurement (GUM), 192–194, 196
planning, 182, 283–284, 291, 295, 309
protocol, 292 half life, 13, 94, 165, 183
response, 182, 187, 225, 278 Hanford, 311, 329
support, 283 hazards
Environmental Protection Agency (EPA), chemical, 298
419, 4, 239, 316 electrical, 305
equilibrium gas cylinders, 304–305
absence of, 14, 176, 178 radiation, 298
secular, 14, 155, 176–177, 179, 434 HAZMAT (see Office of Hazardous Materials
transient, 14, 177, 435 Safety)
error (see uncertainty) health physics, 79, 152, 278, 297, 317, 406
excited state, 10, 18, 142 Henry’s Law, 57, 428
expert witness, 242 hoods (see fume hoods)
extraction
liquid phase, 50–54 impurities, 32, 34, 95, 119, 127, 134, 186, 249,
solid phase, 54 363–364, 406
inductively-coupled plasma (ICP), 190, 327,
facility 329, 364–365, 370, 387–389
cost to build, 287 inductively-coupled plasma mass spectrometry
decommissioning, 94, 220–221, 286 (ICP/MS), 364–365, 370, 376, 391–392
470 Index

ingrowth, 3, 14, 95, 98, 106, 110 cocktail, 34, 75, 97, 127, 129, 130, 152–156,
integrity 183, 336
in business practices, 286–287 counter, 18, 23, 104, 151–156, 276
interference, 28 system, 256, 350
isobaric, 390–391 Love Canal, 239–240, 267
radionuclide, 93, 95, 142 low-concentration behavior, 66–68
interlaboratory testing, 241, 428 Lucas cells, 98, 110, 84, 199, 429
internal conversion, 10, 422, 424, 429, 436 luminescence, 33, 127, 153, 156, 255
International Union of Pure and Applied
Chemistry (IUPAC), 341–344 maintenance
intralaboratory testing (see quality control) equipment, 280–281
iodine, 33, 37, 41–42, 83, 97, 101, 108, 128, mass defect, 429
213, 400, 415–416 mass excess, 15
ion cyclotron resonance (ICR), 377 mass spectrometer
ion trap mass spectrometer (ITMS), 376–377 accelerator, 364, 398–401
ion-exchange applications, 406–409
resins, 44–48, 99, 185, 327 ion detection, 373, 379–382, 386
separation, 44–49 ion source, 365–369
ionization, 18–19, 28, 30, 32, 36–37, 84, 105, mass analyzer, 370–379
137, 147–148, 162, 365–369, 388–389, 393 vacuum chamber, 382
chamber, 29–30, 147–148, 162, 427 mass spectrometry
isobar, 9, 362, 391, 429 of actinides, 338–361
isotope, 1, 3, 7–11, 73 of activation products, 409
isotope dilution, 3, 72–73, 364, 385 of fission products, 408–409
isotope enrichment, 65 of uranium, 407–408
isotope separation, 39, 65, 113, 363 method
isotopic exchange, 74, 253 instability of, 245
IVO-COLD, 347, 357 modification of, 119–120
development of, 118–120
laboratory Minimum Detectable Activity (MDA), 287
license, 285 molecular sieves, 46, 60
notebooks, 228–229, 243 monitoring, 2, 18, 83, 91, 93–94, 147, 182, 187,
access, 224, 269, 281 220–221, 240, 261–263, 306–308
airflow in, 269–271 Monte Carlo
design, 261–291 evaluation, 355–356
discharge, 129, 268–272, 314, 336 programs, 248, 356
flooring, 269 simulation, 125–126, 133, 135–136, 255, 257
hazard levels, 262 multi-channel analyzer, 276
management, 261–291
operating procedures, 222, 224–225 NaI(Tl) detector, 34, 37, 132, 160, 183
operation, 6, 119, 222, 224, 236, 245–246, 293 negatrons
safety, 292–317 emission of, 9, 422
staffing, 78 neutrino, 9, 145
laminar flow boxes, 302, 429 neutron activation, 16, 82, 84, 103–104, 108,
laser ablation, 365–369, 378, 387, 403–404 111, 114, 430
legislation neutrons
Atomic Energy Act, 284, 421 epithermal, 17
RCRA, 315–316 fast, 17, 430
regulatory, 217, 219, 295, 421 high-energy, 17
license resonance energy, 17
radioactive materials, 284–285 sources, 108
liquid-liquid extraction, 50–54, 349–350 thermal, 16, 430
liquid scintillation (see also scintillation and normal distribution, 19, 196, 206
scintillation detectors) nuclear reaction, 431
 Index 471

Occupational Safety and Health Administration radiation

(OSHA), 293, 295, 303, 313 exposure, 162, 256, 297
defined, 293 shielding, 144–145
Office of Hazardous Materials Safety radiocolloid
(HAZMAT) behavior, 64–65, 68–69, 74, 101–102, 251,
defined, 317, 428 433
Emergency Response Handbook, 317 radionuclide purity, 51, 66, 101
online radionuclide decontamination, 39, 40, 95,
references, 4 102
resources, 4 radionuclides
actinides, 6, 8, 11, 43, 52, 73, 96, 338–361
pair formation, 19, 23, 28, 42 americium, 113, 130
PERALS, 129, 136 carbon, 34, 52, 57, 75, 83, 108, 128, 400,
performance 403
assessment, 294 cesium, 402
demonstration, 222, 232 cobalt, 415
Personal Protective Equipment (PPE), 295, 301, cosmic ray produced, 94, 118, 169, 187
305, 431 curium, 130
photodisintegration, 28, 431 europium, 109
photoelectric effect, 23, 26–28, 432 iodine, 33, 41, 83, 97, 101, 108, 213, 400
photoelectric peak, 160 iron, 105, 113
photopeak (see also photoelectric peak) lanthanides, 8, 52, 54
photomultiplier tube, 18, 33, 98, 110, 127, 152, neptunium, 130
326, 336, 431 plutonium, 130
plateau, 31, 150, 348 potassium, 45, 298
plutonium, 63, 67, 101, 112–113, 168, 385 promethium, 73, 101, 122
positrons radium, 130
emission of, 168 radon, 131
precipitate, 40–41, 73 ruthenium, 107, 128
precipitation separation, 40–44, 75, 101, 105 separation of
proportional counter, 29–30, 37, 75, 81, 84, 97 extremely long-lived, 96, 103–114
protective equipment, 303 separation of
long-lived, 14, 94, 96, 103–117
QA/QC efforts separation of
benefits of, 240 short-lived, 18, 87, 94, 117, 290, 389
time and expense of, 240 sodium, 45, 108, 311
quadrupole mass filter (QMF), 370–371, 373, strontium, 105
375–376 super heavy elements, 361
Quality Assurance Project Plan (QAPP), 221, technetium, 73, 101, 107, 122, 128, 215,
259, 432 327–328, 339
quality control (QC) thorium, 130, 298
background, 231 transactinides, 3, 8, 11, 73, 339–340,
to check results, 244 345–346, 348, 351, 435
instrumental, 207–208, 214, 231 tritium, 58, 69
Interlaboratory comparision, 212, 232, uranium, 130, 298
235–236, 244 xenon, 37, 57, 82, 84, 97–98, 105, 168, 332,
Internal QC, 229, 235 333–335
in sample collection, 91–92 yttrium, 99
sample types, 91, 95 zirconium, 46, 47, 216, 100, 311, 405
quenching, 30–31, 104, 128, 130 radium, 57, 73, 84, 101, 103
radon, 57, 60, 84
Radiation Protection Plan (RPP), 285, 296 range
Radiation Safety Manual (RSM), 285, alpha particle, 19, 20, 123, 124, 150
296 beta particle, 10, 20, 123, 167
472 Index

Raoult’s Law, 57, 433 solid media, 33, 34

Rayleigh scattering, 28 systems, 35
recoil energy, 9, 66, 163 screening, 115, 127, 130, 182
records Seaborg, Glenn, 339, 342–343, 351
maintenance, 228 secondary ionization mass spectrometry (SIMS),
notebooks, 228–229, 243 367, 382, 401, 403
retention, 237 security
refractory species, 70, 100 data, 281–283
regulations laboratory, 292–317
CFR, 271, 284, 286, 290, 293, 295–296, 303, self-absorption, 65, 105, 127–128, 133
314–316 slurry, 122, 126, 183, 434
DOE, 271, 314–316 smears, 100, 129, 183, 267
EPA, 109, 239–240, 285, 314–316 software
labor, 314–316 concerns, 258–259
OSHA, 295, 314–316 procurement, 225, 247
Resistance Thermal Device (RTD), 333 reliance of automation on, 319
resolution, 35–38, 74, 103, 129–131, 138, 148, upgrades, 282
157, 159, 168, 208, 217, 231, 255, solid-phase extraction, 54–56
resolving time, dead time, 140–141 solid-state detector, 31–33, 35, 37, 168, 170, 346
resonance ionization mass spectrometry solubility, 40–42, 50
(RIMS), 368–369, 406, 408 solvent extraction, 3, 39, 51, 54, 66, 108, 117, 124
Resources Conservation and Recovery Act sorption, 47, 60–61
(RCRA), 315–316 source
form of, 123–125
safety geometry of, 139, 190, 330, 369
chemical, 279, 295–296 preparation of, 131, 254
laboratory, 292–317 self-absorption, 65, 131
radiation, 295 spectral analysis
sample loss, 68, 122, 246 alpha, 35, 38, 112–114, 128, 130, 156–158,
sample types 168–169, 255, 258, 274
atmospheric beta, 37
concerns, 80, 94, 115 gamma, 35, 101, 105, 110, 131–133, 144,
milk, 88 158–162, 167–169, 183, 191, 199, 255,
hazard levels, 262 258, 274, 323–326, 331–335
meat and aquatic species, 88–89 General (for radiation, not mass), 35–38
sediment, 90 spectrometry
soil alpha, 73, 76, 112, 130, 145, 198, 201, 202,
concerns, 411 255, 276, 389, 411
vegetation beta, 389
crops, 88, 99 gamma, 145, 159, 191, 199
vegetation and crops spike, 210
concerns, 88, 99 staff
water, 42, 68, 75, 96, 129–130 educational requirements of, 279
preservation, 64, 69 maintenance, 382
variability, 245, 247 training, 223–224
saturation region, 29 standard certificates, 222, 235
scattering, 23–25, 28, 125, 136–137, 139–140, standard deviation, 35, 189, 192–194
146, 160–161, 190, 382, 424 standard methods, 118, 229
scavenging, 40, 42, 48, 74 standard procedures, 222, 229
scintillation, 18 Standard Reference Material (SRM), 227, 232,
cocktail, 34, 75, 97 268, 434
scintillation detectors, 33, 34, 35, 84, 102 strontium, 43, 67, 105–107, 115–117, 126, 254,
inorganic media, 33, 34 408, 415, 417
organic media, 33, 34 Szilard-Chalmers reaction, 66
 Index 473

Table of Isotopes, 164, 167 standard, 192–193, 195–196

thermal ionization, 364, 366 uranium, 1, 96, 102, 110–112, 407–408
thorium, 11, 63, 96, 103, 128, 130,
157 validation
time-of-flight (TOF), 370–371, 377 of data, 241, 273
traceability, 227, 229, 249, 417, 435 verification
tracer, 102 of compliance, 220
training, 223 of data, 220
transactinides, 3, 8, 11, 73, 339, 340 volatilization, 363, 368, 387
transuranium, 85, 113, 128, 339 thermal, 131
tritium, 58, 60, 66, 69, 75, 82, 96–97, 255, voltage
263, 396–397, 401 critical, 61
turnaround time, 78, 288–290, 318 overvoltage, 62, 67, 431

uncertainty analysis, 259–260 waste

uncertainty disposal, 261, 272–273, 285
counting, 153 effluent, 94, 269, 396
of curve fitting, 257 liquid, 272
evaluation, 192–196, 201–202 mixed, 262–263, 273, 413
alpha spec, 201–202 solid, 269, 272, 315, 426
gamma spec, 201–202 storage, 264, 268
Type A, 193–194, 198, 203, 260, 262,
435 X rays, 10–11, 23, 28, 37, 105, 108, 112, 140,
Type B, 193, 197–199, 260, 262, 435 142, 146, 151, 159–160, 166–169, 175,
expanded, 196 179, 180–181, 184, 341, 344, 436
measurement uncertainty, 6, 171, 192–204
propagation of, 194–195, 197, 423, 429, yield
432, 435 chemical, 95, 105, 142
sources of, 192, 202 fission, 18, 109, 115, 117, 327