ICRU REPORT No.

88

MEASUREMENT AND REPORTING OF

RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

THE INTERNATIONAL COMMISSION ON
RADIATION UNITS AND
MEASUREMENTS
(Published December 2015)

Journal of the ICRU Volume 12 No 2 2012

Published by Oxford University Press
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Report Committee
W. Hofmann (Chairman), University of Salzburg, Salzburg, Austria
H.S. Arvela, Radiation and Nuclear Safety Authority – STUK, Helsinki, Finland
N.H. Harley, New York University Medical Center, New York, New York, USA
J.W. Marsh, Public Health England, Chilton, UK
J. McLaughlin, University College of Dublin, Dublin, Ireland
A. Röttger, Physikalisch-Technische Bundesanstalt (PTB), Braunschweig, Germany
S. Tokonami, Hirosaki University, Hirosaki, Aomori, Japan

Consultants

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Z. Daraktchieva, Public Health England, Chilton, UK
Xi. Detao, University of South China, Heugyang, China

ICRU Sponsors
E. Fantuzzi, ENEA, Istituto di Radioprotezione, Bologna, Italy
H.-G. Menzel, Euorpean Organization for Nuclear Reserach (CERN), Geneva, Switzerland
H.-G. Paretzke, Helmholtz Zentrum München, German Research Center for Enviornmental Health,
Neuherberg, Germany
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv019
Oxford University Press

The International Commission on Radiation Units and

Measurements

Introduction range of applicability. Situations can arise from time

to time for which an expedient solution of a current
The International Commission on Radiation Units
problem is required.
and Measurements (ICRU), since its inception in

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The ICRU invites and welcomes constructive com-
1925, has had as its principal objective the develop-
ments and suggestions regarding its recommenda-
ment of internationally acceptable recommenda-
tions and reports. These may be transmitted to the
tions regarding:
Chairman.
(1) quantities and units of ionizing radiation and
radioactivity,
(2) procedures suitable for the measurement and
application of these quantities in clinical radi- Current Program
ology and radiobiology, and
The Commission recognizes its obligation to
(3) physical data needed in the application of these
provide guidance and recommendations in the areas
procedures, the use of which tends to assure uni-
of radiation therapy, radiation protection, and the
formity in reporting.
compilation of data important to these fields, and to
The Commission also considers and makes similar scientific research and industrial applications of ra-
types of recommendations for the radiation protec- diation. Increasingly, the Commission is focusing on
tion field. In this connection, its work is performed the problems of protection of the patient and evalu-
in cooperation with the International Commission ation of image quality in diagnostic radiology and ra-
on Radiological Protection (ICRP). diation oncology. These activities do not diminish
the ICRU’s commitment to the provision of a rigor-
Policy ously defined set of quantities and units useful in a
very broad range of scientific endeavors.
ICRU endeavors to collect and evaluate the latest The Commission is currently engaged in the for-
data and information pertinent to the problems of mulation of ICRU Reports treating the following
radiation measurement and dosimetry and to recom- subjects:
mend the most acceptable numerical values for
physical reference data and techniques for current Bioeffect Modeling and Biologically Equivalent Dose
use. Concepts in Radiation Therapy
The Commission’s recommendations are kept Key Data for Measurement Standards in the
under continual review in order to keep abreast of Dosimetry of Ionizing Radiation
the rapidly expanding uses of radiation. Monitoring and Assessment of Radiation Releases to
The ICRU feels that it is the responsibility of na- the Environment
tional organizations to introduce their own detailed Operational Radiation Protection Quantities for
technical procedures for the development and main- External Radiation
tenance of standards. However, it urges that all Prescribing, Recording, and Reporting Brachytherapy
countries adhere as closely as possible to the inter- Cancer of the Cervix
nationally recommended basic concepts of radiation Prescribing, Recording, and Reporting Ion-Beam
quantities and units. Therapy
The Commission maintains and develops a system Prescribing, Recording, and Reporting Stereotactic
of quantities and units and concepts (e.g., for radi- Treatments with Small Photo Beams
ation therapy) and guidance for measurement proce- Retrospective Assessment of Individual Doses for
dures and techniques having the widest possible Acture Exposures to Ionizing Radiation

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 THE INTERNATIONAL COMMISSION ON RADIATION UNITS AND MEASUREMENTS

Small-Field Photon Dosimetry and Applications in International Radiation Protection Association

Radiotherapy International Union of Pure and Applied Physics
United Nations Educational, Scientific and Cultural
The Commission continually reviews progress in Organization
radiation science with the aim of identifying areas
in which the development of guidance and recom- The Commission has found its relationship with
mendations can make an important contribution. all of these organizations fruitful and of substantial
benefit to the ICRU program.
The ICRU’s Relationship with Other
Organizations Operating Funds
In addition to its close relationship with the ICRP, Financial support has been received from the fol-
the ICRU has developed relationships with national lowing organizations:
and international agencies and organizations. In
these relationships, the ICRU is looked to for American Association of Physicists in Medicine
primary guidance in matters relating to quantities, Belgian Nuclear Research Centre

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

units, and measurements for ionizing radiation, and Canadian Nuclear Safety Commission
their applications in the radiological sciences. In Federal Office Public Health, Switzerland
1960, through a special liaison agreement, the ICRU Helmholtz Zentrum München Hitachi, Ltd.
entered into consultative status with the International Radiation Protection Association
International Atomic Energy Agency (IAEA). The International Society of Radiology
Commission has a formal relationship with the Ion Beam Applications, S.A.
United Nations Scientific Committee on the Effects Japanese Society of Radiological Technology
of Atomic Radiation (UNSCEAR), whereby ICRU MDS Nordion
observers are invited to attend annual UNSCEAR Nederlandse Vereniging voor Radiologie
meetings. The Commission and the International Philips Medical Systems, Incorporated
Organization for Standardization (ISO) informally Radiological Society of North America
exchange notifications of meetings, and the ICRU is Siemens Medical Solutions
formally designated for liaison with two of the ISO U.S. Environmental Protection Agency
technical committees ICRU is a member of U.S. Nuclear Regulatory Commission
Consultative Committee for Units (CCU) – BIPM Varian Medical Systems
and Consultative Committee for Ionizing Radiation
(CCRI(I) – BIPM and Observer to CCRI(II) and In addition to the direct monetary support pro-
CCRI (III). ICRU also enjoys a strong relationship vided by these organizations, many organizations
with its sister organization, the National Council on provide indirect support for the Commission’s
Radiation Protection and Measurements (NCRP). program. This support is provided in many forms,
In essence, ICRU and NCRP were founded concur- including, among others, subsidies for (1) the time of
rently by the same individuals. Presently, this long- individuals participating in ICRU activities, (2)
standing relationship is formally acknowledged by travel costs involved in ICRU meetings, and (3)
a special liaison agreement. ICRU also exchanges meeting facilities and services.
reports with the following organizations: In recognition of the fact that its work is made
possible by the generous support provided by all of
Bureau International des Poids et Mesures the organizations supporting its program, the
European Commission Commission expresses its deep appreciation.
International Council for Science
International Electrotechnical Commission Hans-Georg Menzel
International Labour Office Chairman, ICRU
International Organization for Medical Physics Geneva, Switzerland
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv018
Oxford University Press

Measurement and Reporting of Radon Exposures

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

1.1 Indoor Radon . . . . . . . . . . . . .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. . 21

1.2 Outdoor Radon . . . . . . . . . . . .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. . 22
1.3 Thoron . . . . . . . . . . . . . . . . . . .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. . 23
1.4 Protection Against Radon . .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. . 23
1.5 Aim of the Present Report .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. .. .. . .. . 24

2. Health Effects of Radon Exposure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3. Radon and Radon Progeny Inhalation and Resultant Doses . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.1 Lung Dose Assessment Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.2 Radon versus Radon Progeny Doses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2.1 Lung Doses due to Inhalation of Radon Gas and Thoron Gas . . . . . . . . . . . . . . . 30
3.2.2 Lung Doses due to Inhalation of Short-lived Radon Progeny . . . . . . . . . . . . . . . 31
3.2.3 Lung Doses due to Inhaled Thoron Progeny. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3 Lung Doses versus Other Organ Doses. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.3.1 Doses to Internal Organs Arising from Inhalation of Radon Progeny . . . . . . . . 32
3.3.2 Doses to Internal Organs Arising from Inhalation of Radon Gas . . . . . . . . . . . . 33
3.3.3 Skin Dose from Deposited Radon Progeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.3.4 Ingestion of Radon in Water. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
3.4 Sensitive Target Cells in Bronchial Epithelium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.5 Personal and Environmental Parameters Affecting Lung Dosimetry . . . . . . . . . . . . . . 37
3.6 Dependence of Doses on Physical Activities (Breathing Parameters) and Age . . . . . . 39
3.6.1 Dependence on Physical Activities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.6.2 Dependence on Age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.7 Dependence on 222Rn Progeny Absorption Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.8 Dependence on Radon Progeny Related Aerosol Parameters . . . . . . . . . . . . . . . . . . . . . 43
3.8.1 Radon Progeny Aerosol Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.8.2 Sensitivity of Lung Dose from Inhalation of 222Rn Progeny to Aerosol
Parameter Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.9 Variability and Uncertainty of Individual Lung Doses . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.9.1 Comparison of Different Lung Dosimetry Models . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.9.2 Intra- and Intersubject Variability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

3.9.3Inhomogeneity of Surface Activities and Resulting Doses within Bronchial 49

Airways . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.9.4 Comparison of Bronchial Doses between Non-smokers and Smokers . . . . . . . . 50
3.9.5 Contribution of Sensitive Target Cells in Bronchial Epithelium to Lung
Cancer Risk. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
3.10 Human versus Experimental Animal Doses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.10.1 Animal Inhalation Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.10.2 Animal Dosimetry Models. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

4. Characteristics and Behavior of Radon and Radon Progeny . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.1 Radon Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55

4.2 Radon Decay Schemes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
4.3 Behavior of Radon and Radon Progeny in Indoor Environments . . . . . . . . . . . . . . . . . 57
4.3.1 Steady-state Activity Concentrations of Radon and Thoron Gases in

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Indoor Air . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.3.2 Steady-state Activity Concentrations of Radon Progeny in Indoor Air . . . . . . . 59
4.3.3 Radon Progeny Parameters Affecting Lung Dosimetry . . . . . . . . . . . . . . . . . . . . 60
4.4 Airborne Radon Activity Concentrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.4.1 Radon in Homes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4.4.2 Radon in Workplaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.4.3 Comparison of Radon in Homes and Indoor Workplaces . . . . . . . . . . . . . . . . . . . 63
4.4.4 Thoron in Homes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.5 Equilibrium Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
4.6 Attached and Unattached Fractions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
4.6.1 Unattached Fraction, fp, for Radon (222Rn) Progeny . . . . . . . . . . . . . . . . . . . . . . 66
4.6.2 Correlation between Equilibrium Factor, F, and Unattached Fraction, fp, 67
for 222Rn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6.3 Unattached Fraction, fp, for Thoron (220Rn) Progeny . . . . . . . . . . . . . . . . . . . . . . 68
4.7 Aerosol Size Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

5. Principles of Radon and Radon Progeny Detection Systems and Measurements . . . . . . . . 71

5.1 Radon and Radon Progeny Metrology and Quality Assurance of Measurements . . . . 71
5.1.1 General Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.1.2 Comparisons of Radon Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
5.2 Radon Gas (222Rn, 220Rn) in the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.1 Radon Measurement Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
5.2.1.1 Airborne Radon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
5.2.1.2 Waterborne Radon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.1.3 Soilborne Radon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
5.2.2 Radon Detection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
5.2.3 Thoron Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
5.3 Radon and Thoron Progeny Activity Concentrations and Particles Size
Distributions in the Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.3.1 Radon and Thoron Progeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
5.3.2 Radon Progeny Measurement Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.3.3 Measurement of the Unattached Fraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.3.4 Radon Progeny Particle Size Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
5.3.4.1 Number Size Distribution Measurements . . . . . . . . . . . . . . . . . . . . . . . . 86
5.3.4.2 Direct Activity Size Distribution Measurements . . . . . . . . . . . . . . . . . . 86
5.4 Retrospective Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
5.4.1 Surface Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
5.4.2 Volume Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
5.4.3 In-vivo Measurements of 210Pb. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
 Contents

5.5 Personal Monitoring for Radon and Radon Progeny . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93

5.5.1 Personal Monitoring for Radon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.5.2 Personal Monitoring for Radon Progeny. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94

6 Strategies for Radon and Radon Progeny Measurements and Surveys . . . . . . . . . . . . . . . . . 95

6.1 Objectives: Areal and Individual Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

6.2 Radon versus Radon Progeny Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.3 Areal Surveys and Mapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.3.1 Goals of Radon Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
6.3.2 Sampling and Survey Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.2.1 Random Sample Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.2.2 Stratified Sampling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.2.3 Choice of the Strategy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
6.3.2.4 Period of Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

6.3.2.5 Detector Choice and Deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.3.2.6 Examples of Survey Practices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
6.3.3 Use of Volunteer Data and Large Radon Mapping Data . . . . . . . . . . . . . . . . . . . . 101
6.3.4 Radon Maps. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.3.4.1 Geological Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.3.4.2 Aerial Gamma Radioactivity Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
6.3.4.3 Radon in Soil Gas Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.4.4 Indoor Radon Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
6.3.4.5 Combined Maps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.3.5 Lognormal Modeling of Indoor Radon Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
6.3.6 Mapping the Proportion of Dwellings above Reference Level. . . . . . . . . . . . . . . 104
6.4 Long-term versus Short-term Areal Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.4.1 Integrating versus Time-resolved Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . 104
6.4.2 Predicting the Annual Average Using Short-term Measurements . . . . . . . . . . . 105
6.4.3 Predicting the Past Thirty Years of Radon Exposure from Annual Radon
Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
6.4.4 Using Short-term Measurements to Make Action Decisions . . . . . . . . . . . . . . . . 107
6.5 Homes and Workplaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
6.6 Individual Exposure Assessment: Time-resolved Measurements . . . . . . . . . . . . . . . . . . 110
6.6.1 Comparison of Areal and Personal Exposure Assessment at Workplaces . . . . . 111
6.6.2 Comparison of Integral and Time-resolved Personal Measurements . . . . . . . . . 111

7. Interpretation of Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

7.1 Variations of Areal and Local Radon Activity Concentrations . . . . . . . . . . . . . . . . . . . . 113

7.1.1 Worldwide Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.1.2 Spatial Variation within a House . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
7.2 Diurnal and Seasonal Variations of Radon Activity Concentrations . . . . . . . . . . . . . . . 115
7.3 Physical Processes Affecting Indoor Radon Activity Concentrations . . . . . . . . . . . . . . 116
7.3.1 Pressure Difference and Air Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
7.3.2 Radon Entry from Soil and Building Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
7.3.3 Effect of Wind on Radon Entry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
7.3.4 Comparison of Driving Forces. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
7.3.5 Seasonal Correction Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
7.3.6 Atypical Seasonal Variations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
7.3.7 Long-term Variation in Annual Average Radon Activity Concentrations . . . . . 122
7.4 Thoron Interference on Radon Detection Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.4.1 Time Integrating Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.4.2 Continuous Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
7.4.3 Mathematical Analysis of Radon/Thoron Atmospheres using Nuclear Track
Detectors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

7.5 Variation of Aerosol Parameter Values for Radon Progeny . . . . . . . . . . . . . . . . . . . .. . 125

7.5.1 Equilibrium Factor, F, and Unattached Fraction, fp, for 222Rn . . . . . . . . . . . .. . 125
7.5.2 Particle Size Distributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . 127
7.6 Estimation of Missing Exposure Data and Uncertainties Involved . . . . . . . . . . . . . .. . 132

8. Variabilities and Uncertainties of Radon and Radon Progeny Exposures and Dosimetry 135

8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

8.1.1 The Meaning of Uncertainty in Metrology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
8.1.2 Variability of Long-term Average Radon Gas Exposures . . . . . . . . . . . . . . . . . . . 135
8.1.3 Classification of Uncertainties in Exposure Assessment for Epidemiological
Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
8.2 Uncertainty Evaluations: from the Realization of the Unit to the Field
Measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
8.2.1 Radon Gas Activity Concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

8.2.1.1 Rn-220 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
8.2.1.2 Rn-222 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
8.2.1.3 Determination of an Average Activity Concentration in a Room . . . . . 139
8.2.2 Radon and Thoron Gas Exposures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
8.2.2.1 Non-direct Reading Devices: Rn-220 Exposure Calibration . . . . . . . . . 141
8.2.2.2 Non-direct Reading Devices: Rn-222 Exposure Calibration . . . . . . . . . 141
8.3 Other Sources of Uncertainties in Assessment of the Annual Average Radon Activity
Concentration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.3.1 Uncertainties due to Spatial Variation of Indoor Radon Activity
Concentration in Dwellings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.3.2 Uncertainties in Extrapolating a Short-term Measurement to an Annual
Average. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.3.3 Uncertainties due to Long-term Variation in Annual Average Radon Activity
Concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
8.3.4 Combined Uncertainty in the Estimation of Long-term Average Radon
Activity Concentration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
8.3.5 Uncertainties Associated with the Estimate of Individual Exposure Obtained
with Areal Measurements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
8.4 Uncertainties of Radon Progeny Measurements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
8.4.1 Measurand and Derived Quantities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
8.4.2 An Example for the Determination of Derived Quantities . . . . . . . . . . . . . . . . . 145
8.5 Uncertainties of Dosimetric Results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8.5.1 Application of Different Dosimetric Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8.5.2 Uncertainties of Model Parameter Values Used in Dose Calculations . . . . . . . 147
8.5.2.1 Sensitive Target Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
8.5.2.2 Radon Progeny Size Distributions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
8.5.2.3 Apportionment Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
8.5.2.4 Radiation Weighting Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
8.5.3 Intersubject Variability. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
8.5.4 Summary of Uncertainties of Dose Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . 149
8.6 Effect of these Uncertainties on the Analysis of Epidemiological Studies . . . . . . . . . . 149

9. Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

9.1 Good Practice Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. .. . .. . 151

9.2 Recommendations Regarding Measurement Strategies . . . . . . . . . . . . . . . . . .. .. . .. . 152
9.3 Recommendations Regarding Measurement Techniques . . . . . . . . . . . . . . . . . .. .. . .. . 155
9.4 Recommendations Regarding Recording and Reporting of Measurements . .. .. . .. . 155
 Contents

Appendices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Appendix A (Section 5) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Appendix B (Section 7) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161

Appendix C (Section 8) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv003
Oxford University Press

Preface

One of the principal objectives of the International control of exposure. These reports recommend the
Commission on Radiation Units and Measurements use of action or reference values as operational ra-
(ICRU) is to provide recommendations and guidance diation protection tools which have been implemen-
on performing and reporting radiation measure- ted within national regulatory frameworks. In
ments. This Report on Measurements and Reporting 2010, ICRP published a report (115) on Lung

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

of Radon Exposures presents the most recent recom- Cancer Risk from Radon and Progeny which
mendations for measurements in the field of reviewed epidemiological studies on residential and
radiation protection. Earlier ICRU publications on occupational exposures. One important conclusion
radiation protection measurements include Report of this report was that the nominal risk coefficient
20 (1971), Radiation Protection Instrumentation for exposure to radon should be taken to be twice
and its Application, Report 53 (1994), Gamma-ray that previously assumed. As a consequence, refer-
Spectrometry in the Environment, Report 56 (1997), ence values, in terms of radon activity concentra-
Dosimetry for External Beta Rays for Radiation tion in air, Bq m23, were lowered proportionally in
Protection, Report 69 (2003), Direct Determination of many countries in accordance with recommenda-
Body Content of Radionuclides and Report 75 (2006), tions by international organizations. The public
Sampling for Radionuclides in the Environment. interest in and concern for radon exposure
Epidemiological studies have demonstrated that increased substantially and the need for reliable,
inhalation of radon and its short-lived decay pro- reproducible radon measurement procedures and
ducts can cause lung cancer. Increased lung cancer techniques became more obvious.
incidence in workers in uranium and other mines The objective of this report is, therefore, to provide
has been known since the nineteenth century but it conceptual and practical guidance for radon mea-
was only in the middle of the last century that in- surements in air and in water. The recommenda-
halation of radon and its progeny was recognized as tions include guidance for the choice of strategies for
the cause. The influence of radon on lung cancer radon and radon progeny measurements and
risk to the general public was established even surveys and for interpreting and reporting measure-
more recently. A large number of studies on radon ment results, appropriate for the goal of the mea-
activity concentrations in dwellings and mines surements. The report also addresses methods to
worldwide were published in the second half of last determine and reduce uncertainties associated with
century and the United Nations Scientific these measurements and resulting dosimetric esti-
Committee on the Effects of Atomic Radiation mates. It describes the state-of-the-art of radon
(UNSCEAR) summarized the results in several measurement techniques which is expected to be of
reports. There has also been an increase in epi- relevance in view of the reduced reference levels in
demiological studies on lung cancer incidence dwellings and in the workplace as well as for epi-
related to radon exposures in dwellings. demiological studies.
The International Commission on Radiological The recommendations in this report are aimed at
Protection (ICRP) has addressed the issue of authorities planning radon surveys, at experts per-
radiation protection against radon exposure and forming measurements and at scientists involved in
published a number of reports providing recom- epidemiological studies on lung cancer risk due to
mendations and guidance for the assessment and radon inhalation.

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv002
Oxford University Press

Glossary: Definitions, Quantities, and Units

In contrast to other radiation measurements, the metrology of radon involves several rather sophisticated
definitions, quantities, and units. Since these are rather uncommon, this glossary gives an overview about
these special terms used in international standards and recommendations, technical descriptions, and scien-
tific papers. General terms of statistics, metrology, and physics are not part of this glossary.
The following definitions, quantities, and units are used in agreement with ICRU 85a, ICRP 103 (definition

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

given in Annex B of ICRP 103, not from the glossary), ICRP 32, IEC 61577, and ISO 11665. In the special case
that the definitions given in the above documents are not totally consistent or have need of further specifica-
tion, e.g., modernization, this ICRU report aims to attempt to do so.
The quantities describing the movement of air from the environment toward buildings are used in compli-
ance with ISO 9972 and EN 13829:2000.
Several more quantities, terms, or definitions are taken from ISO Nuclear Energy Vocabulary (Parts 1 and 2),
the IAEA Glossary, the IEC’s International Electrotechnical Vocabulary (IEC 60050), the BIPMs International
Vocabulary of Metrology, and also the ISO 17025.
Nuclear data are taken from Monographie BIPM-5, while fundamental constants are based on CODATA
evaluations.
Note: The symbol Rn is used in the following text to refer to both, 222Rn and 220Rn. If a special isotope is
named, it is done by purpose and the definition is only valid in this configuration.

Absorbed Dose
Absorbed dose, D, is defined as the quotient of d1, by dm, where d1 is the mean energy imparted by ionizing
radiation, to matter of mass dm thus

d1
D¼ :
dm

The unit of absorbed dose is J kg2l. The special name for the unit of absorbed dose is gray (Gy).

Absorbed Dose to Radon Exposure Conversion Coefficient

The absorbed dose to radon exposure conversion coefficient defines the relationship between the absorbed
dose to an organ/tissue or region and the exposure to inhaled short-lived radon progeny. The exposure can be
expressed in terms of potential alpha energy exposure (Jm23 h) or exposure to radon (Bq m23 h) with a given
equilibrium factor F.
Note: There are published data in Gy WLM21 unit. Although the WLM is a not an SI unit, it is still used for
the characterization of the radon progeny exposure, particularly to understand historical publications.

Activity
Number dN of spontaneous nuclear transitions or nuclear disintegrations of a radionuclide of amount N
produced during a short time interval dt.

dN
A¼ :
dt

The unit of activity is Bq.

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Activity Concentration
Activity A per unit volume V of the respective isotope.

A
C¼
V

Relevant activity concentrations can be marked by an index, for example, Cdeep is used as radon activity con-
centration in deep soil air. The unit of activity concentration is Bq m23.

Activity Median Diameter (AMD)

The activity median diameter AMD is the median of the activity distribution of diameters of unit density
(kg m23) spheres. The unit of activity median diameter is m.

Activity Median Aerodynamic Diameter (AMAD)

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The activity median aerodynamic diameter AMAD is the median of the activity distribution of diameters of
unit density (kg m23) spheres that have the same terminal settling velocity in air as the aerosol particle of
interest.

Activity Median Thermodynamic Diameter AMTD

The activity median thermodynamic diameter AMTD is the median of the activity distribution of diameters
of spherical particles that have the same diffusion coefficient in air as the aerosol particle of interest.

Activity Size Distribution

The activity size distribution of short-lived radon progeny represents the differential distribution of the frac-
tions of attached or unattached activity concentrations as a function of thermodynamic or aerodynamic par-
ticle diameter.

Aerodynamic Diameter
See activity median aerodynamic diameter.

Aerosol
A suspension of solid or liquid particles in a gaseous medium. Airborne particles can have a wide range of
sizes; typically from 0.5 nm to 10 mm.
Note: The particle size of unattached radon progeny is of the order of magnitude of nm. If there is need for a
more precise definition, then this report proposes 5 nm diameter as the upper limit for the unattached
progeny (i.e., cluster carrying progeny).

Aerosol Particle Size Distribution

The aerosol particle size distribution is defined as a function corresponding to several partial concentrations
(number of particles of a defined range of diameter per unit volume of air). The unit of the distribution is m23.

Air Change Rate at Reference Pressure

Air leakage rate per internal volume at the reference pressure difference across the building envelope. The
unit of air change rate is h21.
Note: The reference pressure is usually 50 Pa. It is abbreviated as n50 or ACH50.

Air Leakage Rate

The air leakage rate is the air flow rate across the building envelope. The unit of air leakage rate m3 h21.
Note: This movement includes flow through joints, cracks, and porous surfaces, or a combination thereof,
induced by the air-moving equipment specified in the standard ISO 9972.

4
 Glossary

Apportionment Factors
To take account of potential differences in radiation sensitivity between regions of the lung, the equivalent
dose to the bronchial, bronchiolar, and the alveolar regions are weighted by apportionment factors, which re-
present their relative contribution to the total radiation detriment of the lung.

Attached Fraction
The fraction of the potential alpha energy concentration of short-lived radon progeny that is attached to the
ambient aerosol particles.
Note 1: The attached progeny may have a tri-modal activity size distribution, which can be approximated by
a combination of three lognormal distributions (Porstendörfer, 2001). These consist of the nucleation mode
with AMD values between 10 and 100 nm, the accumulation mode with AMD values of 100 –450 nm, and a
coarse mode with an AMD . 1 mm. Generally, the greatest fraction of the potential alpha energy (PAE) is in
the accumulation mode.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Note 2: If there is need for a more precise definition of attached progeny, then this report proposes 5 nm
diameter as the lower limit for the attached progeny (i.e., aerosols carrying progeny).
Note 3: The sum of attached and unattached fractions is equal to 1.

Attachment Rate
The attachment rate X expresses the adsorption velocity of the unattached radon progeny to the atmospher-
ic aerosol:

X¼bZ

where b is the attachment coefficient (m3 s21), and Z the aerosol number concentration (m23). The unit of
attachment rate is s21.

Breathing Frequency
The breathing frequency is the number of breaths per unit time. The unit of breathing frequency is s21.

Breathing Rate
The breathing rate is the volume of air inhaled to the lung per unit time i.e., tidal volume multiplied by re-
spiratory frequency. The unit of breathing rate is m3 s21.

Calibration
Is an operation that, under specified conditions, in a first step, establishes a relation between the quantity
values associated with measurement uncertainties provided by measurement standards and corresponding
indications with associated measurement uncertainties and, in a second step, uses this information to estab-
lish a relation for obtaining a measurement result from an indication.
Note 1: A calibration may be expressed by a statement, calibration function, calibration diagram, calibration
curve, or calibration table. In some cases, it may consist of an additive or multiplicative correction of the indi-
cation with associated measurement uncertainty.
Note 2: Calibration should not be confused with adjustment of a measuring system, often mistakenly called
“self-calibration,” nor with verification of calibration.

Coarse mode
Aerosol particles that are larger than 2 mm in diameter

Confounder
The presence of another independent variable associated with an exposure that accounts wholly or partially
for the disease.

5
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Deposition Velocity
The deposition velocity, vg, is defined as:

wðdÞ
vg ¼
Zðz; dÞ

where w(d) is the number of particles with diameter d deposited per unit surface area and time and Z(z,d) is
the concentration of particles with diameter d at height z above a surface. Note that the deposition velocity
has the dimension of a velocity, but is not a velocity in a physical sense.

Effect Modifier
A variable that differentially ( positively or negatively) modifies the observed effect of a risk factor on disease
status. The effect of the factor may be different for different groups.

Effective Dose

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The effective dose, E, is defined by a weighted sum of tissue equivalent doses, wTHT, as:
X X X
E¼ w T HT ¼ wT wR DT;R
T T R
P
where wT is the tissue weighting factor for tissue T with wT ¼ 1 and wR is the radiation weighting factor
T
(see definition) and DT,R is the mean absorbed dose from radiation type R in tissue T. The sum is performed
over all organs and tissues of the human body considered to be sensitive to the induction of stochastic effects.
The wT values are chosen to represent the contributions of individual organs and tissues to the overall radi-
ation detriment from stochastic effects. The unit of effective dose is J kg21. The special name for the unit of
effective dose is sievert (Sv).
Note: The unit is the same for equivalent dose and effective dose as well as for some operational quantities.
Care must be taken to ensure that the quantity being used is clearly stated.

Effective Dose to Radon Exposure Conversion Coefficient

The conversion coefficient gives the effective dose due to inhaled short-lived radon progeny. The exposure
can be expressed in terms of potential alpha energy exposure (J m23 h) or exposure to radon activity concen-
tration (Bq m23 h) with a given equilibrium factor F. The unit of effective dose to radon exposure conversion
coefficient is Sv (J m23 h)21 or Sv (Bq m23 h)21.
Note: there are published data in Sv WLM21 unit. Although the WLM is a not an SI unit, it is still used for
the characterization of the radon progeny exposure, particularly to understand historical publications.

Effective Leakage Area (ELA)

ELA was developed by Lawrence Berkeley Laboratory (LBL) and is used in their infiltration model. The ef-
fective leakage area is defined as the area of a special nozzle-shaped hole (similar to the inlet of a Blower Door
fan) that would leak the same amount of air as the building does at a pressure difference of 4 Pa. The unit of ef-
fective leakage is m2.
Note 1: Once the leakage rate for the building has been measured, it can be used to estimate the cumulative
size of all leaks or holes in the building’s air barrier. The estimated leakage areas not only provide a way to
visualize the physical size of the measured holes in the building, but they are also used in infiltration models
to estimate the building’s natural air change rate (i.e., the air change rate under natural weather conditions).
Note 2: Equivalent leakage area (EqLA): EqLA is defined by Canadian researchers at the Canadian
National Research Council as the area of a sharp-edged orifice (a sharp round hole cut in a thin plate) that
would leak the same amount of air as the building does at a pressure difference of 10 Pa. The EqLA is used in
the AIM infiltration model.

Emanation Coefficient
The emanation coefficient is defined as the fraction of radon atoms released into a rock or soil pore space
from a radium-bearing grain.

6
 Glossary

Equilibrium Equivalent Activity Concentration

The activity concentration of radon, CRn, in radioactive equilibrium with its short-lived decay products that
has the same potential alpha energy concentration Cp as the non-equilibrium mixture to which the Ceq refers:

Ceq;Rn-222 ¼ kPo-218 CðPo-218Þ þ kPb-214 CðPb-214Þ þ kBi-214 CðBi-214Þ þ kPo-214 CðPo-214Þ

Ceq;Rn-220 ¼ kPo-216 CðPo-216Þ þ kPb-212 CðPb-212Þ þ kBi-212 CðBi-212Þ þ kPo-212 CðPo-212Þ:

The weighting coefficients k are calculated by nuclear data and given in Table 1.

Table 1. Coefficients for the calculation of the equilibrium equivalent concentration from measured activity concentrations of radon
progeny

kPo-218 u(kPo-218) kPb-214 u(kPb-214) kBi-214 u(kBi-214) kPo-214 u(kPo-214)

0.106 0.002 0.513 0.010 0.381 0.009 5.2  1028 1  1029
kPo-216 u(kPo-216) kPb-212 u(kPb-212) kBi-212 u(kBi-212) kPo-212 u(kPo-212)
6.684  1026 0.223  1026 0.9133 0.0001 0.0866 0.0001 8.05  10212 6  10214

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Since kPo-214 ,, 1, kPo-216 ,, 1, and kPo-212 ,, 1, the corresponding activity concentration can be omitted.

Ceq;Rn-222 ¼ kPo-218 CðPo-218Þ þ kPb-214 CðPb-214Þ þ kBi-214 CðBi-214Þ

Ceq;Rn-220 ¼ kPb-212 CðPb-212Þ þ kBi-212 CðBi-212Þ:

The unit of equilibrium equivalent activity concentration is Bq m23.

Note 1: For Rn-222, the following conversion is valid: Ceq ¼ Cp/[5.57(10)  1029 J Bq21] or Ceq ¼ Cp/
[3.47(7)  1010 eV Bq21] .
Note 2: For Rn-220, the following conversion is valid: Ceq ¼ Cp / [7.565(8)  1028 J/Bq] or Ceq ¼ Cp/
[4.722(5)  1011 eV Bq21].

Equilibrium Factor
The equilibrium factor is the ratio of equilibrium equivalent activity concentration Ceq and the radon
activity concentration CRn.

Ceq
F¼
CRn

Note: In the case of 220Rn, the relatively long half-life of 212Pb may lead to cases where 220Rn completely dis-
appears before 212Pb grows in; in this case, the quantity is not defined.

Equivalent Dose
The equivalent dose to an organ or tissue, HT, is defined by
X
HT ¼ wR  DT;R
R

where DT,R is the mean absorbed dose from radiation type R to tissue T, and wR is the radiation weighting
factor for radiation R. The sum is performed over all types of radiations involved. The unit of equivalent dose
is J kg21. The special name for the unit of equivalent dose is sievert (Sv).
Note 1: The unit Sv is the same for equivalent dose and effective dose as well as for some operational dose
quantities. Care must be taken to ensure that the quantity being used is clearly stated.
Note 2: Values of equivalent dose to a specified tissue from any type(s) of radiation can be compared directly.

Equivalent Dose to Radon Exposure Conversion Coefficient

This conversion coefficient is the equivalent dose to an organ per unit exposure to radon progeny. The expos-
ure can be expressed in terms of potential alpha energy exposure (J m23 h) or exposure to radon activity con-
centration (Bq m23 h) with a given equilibrium factor F. The unit of equivalent dose to radon exposure
conversion coefficient is Sv (J m23 h)21 or Sv (Bq m23 h)21.

7
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Note: There are published data in Sv WLM21 unit. Although the WLM is not an SI unit it is still used for
the characterization of the radon progeny exposure, particularly to understand the historical publications.

Exposure to Radon
The time-integral of the activity concentration during a defined period of time.
ð
PRn ðC; DtÞ ¼ CRn  dt
Dt

The unit of exposure to radon is Bq m23 h.

Exposure to Radon Progeny

See “Potential alpha energy exposure. ”

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Friction Velocity
The friction velocity is defined as the square root of the ratio of wall shear stress to the fluid density.

Functional Residual Capacity

The functional residual capacity is the air volume of the lung at the end of normal expiration. The unit of
functional residual capacity is m3.

Internal Volume
Heated, cooled, or mechanically ventilated space within a building or part of a building subject to the meas-
urement, generally not including the attic space, basement space, and attached structures. The unit of intern-
al volume is m23.

International System of Units (SI)

System of units, based on the International System of Quantities, their names, and symbols, including a
series of prefixes and their names and symbols, together with rules for their use, adopted by the General
Conference on Weights and Measures (CGPM).

K-factor
The ratio of the equivalent dose to the lung per unit potential alpha energy exposure in homes for a given
population group to that for a miner exposed in mines.

Mean Absorbed Dose

The mean absorbed dose in the region of an organ or tissue T, DT, is defined by
Ð
Dðx; y; zÞ rðx; y; zÞ dV
T ¼ T
D Ð
rðx; y; zÞdV
T

where V is the volume of the tissue region T, D the absorbed dose at a point (x, y, z) in that region, and r the mass
density at this point. In practice, the mean absorbed dose in an organ or tissue T; D  T , is usually written DT. The
21
unit of mean absorbed dose is J kg . The special name for the unit of mean absorbed dose is gray (Gy).
Note: When using the quantity absorbed dose in practical radiation protection applications, doses are aver-
aged over tissue volumes. It is assumed that, for low doses, the mean value of absorbed dose averaged over a
specific organ or tissue can be correlated with radiation detriment for stochastic effects in that tissue with an
accuracy sufficient for the purposes of radiological protection. The averaging of absorbed doses in tissues or
organs and the summing of weighted mean doses in different organs and tissues of the human body comprise
the basis for the definition of the protection quantities which are used for limiting stochastic effects at low

8
 Glossary

doses. This approach is based on the Linear-No-Threshold (LNT) model and therefore allows the addition of
doses resulting from external and internal exposure.

Measurement
A measurement is a process of experimentally obtaining one or more quantity values that can reasonably be
attributed to a quantity.
Note 1: Measurement implies comparison of quantities or counting of entities.
Note 2: Measurement presupposes a description of the quantity commensurate with the intended use of a
measurement result, a measurement procedure, and a calibrated measuring system operating according to the
specified measurement procedure, including the measurement conditions.

Measurand
The particular quantity that is subject to the measurement.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Measurement Error
The error of measurement is the measured quantity value minus a reference quantity value.
Note 1: The concept of “measurement error” can be used both (a) when there is a single reference quantity
value, which occurs if a calibration is made by means of a measurement standard with a measured quantity
value having a negligible measurement uncertainty or if a conventional quantity value is given, in which case
the measurement error is known, and (b) if a measurand is supposed to be represented by a unique true quantity
value or a set of true quantity values of negligible range, in which case the measurement error is not known.
Note 2: Measurement error should not be confused with production error or mistake.

Measurement Result
The result of a measurement is a set of quantity values being attributed to a measurand together with any
other available relevant information.
Note 1: A measurement result generally contains “relevant information” about the set of quantity values,
such that some may be more representative of the measurand than others. This may be expressed in the form of
a probability density function (PDF).
Note 2: A measurement result is generally expressed as a single measured quantity value and a measure-
ment uncertainty.

Measurement Traceability
The property of a measurement result whereby the result can be related to a reference through a documen-
ted unbroken chain of calibrations, each contributing to the measurement uncertainty.

Metrology
Metrology is the science of measurement and its application.
Note: Metrology includes all theoretical and practical aspects of measurement in any field of application.

Potential Alpha Energy (PAE)

The potential alpha energy, 1p, is the total alpha energy emitted during the decay of a progeny atom along
the decay chain up to 210Pb or 208Pb, respectively, for the decay chains of the 222Rn and 220Rn.
The potential alpha energy 1p (X2A) of a progeny is calculated by the following equations according to the
decay chains of 222Rn and 220Rn.
The values for the transition probability p as well as the uncertainties assigned to the nuclear data are
taken from the Monographie-5 of BIPM and the conversion from eV to J is by the data of CODATA (Tables 2
and 3).
222
Rn:
1p (Po-218) ¼ Sipi1i (Po-218) þ Skpk1k (Po-214),
1p (Pb-214) ¼ Skpk1k (Po-214),

9
 Table 2. Potential alpha energy per atom for 222Rn progeny including standard uncertainty

Potential Standard Potential Standard Potential Standard Potential Standard

alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1)

1p (Po-218) u(1p) (Po-218) 1p (Pb-214) u(1p) (Pb-214) 1p (Bi-214) u(1p) (Bi-214) 1p (Po-214) u(1p) (Po-214)
13688.9 keV 0.6 keV 7687.9 keV 0.5 keV 7687.9 keV 0.5 keV 7686.7 keV 0.5 keV
2.19321  10212 J 0.00009  10212 J 1.23174  10212 J 0.00008  10212 J 1.23174  10212 J 0.00008  10212 J 1.23155  10212 J 0.00008  10212 J

10
Table 3. Potential alpha energy per atom for 220Rn progeny including standard uncertainty

Potential Standard Potential Standard Potential Standard Potential Standard

alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1) alpha-energy uncertainty (k ¼1)
MEASUREMENT AND REPORTING OF RADON EXPOSURES

1p (Po-216) u(1p) (Po-216) 1p (Pb-212) u(1p) (Pb-212) 1p (Bi-212) u(1p) (Bi-212) 1p (Po-212) u(1p) (Po-212)
14 582.7 keV 5.1 keV 7804.2 keV 5.1 keV 7804.2 keV 5.1 keV 8785.2 keV 4.4 keV
2.33641  10212 J 0.00081  10212 J 1.25036  10212 J 0.00081  10212 J 1.25036  10212 J 0.00081  10212 J 1.40754  10212 J 0.00071  10212 Jh

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

 Glossary

1p (Bi-214) ¼ Skpk 1k (Po-214),

1p (Po-214) ¼ Skpk1k (Po-214).
220
Rn: (including the branching of the decay of 212Bi, pS ¼ Skpk):

1p (Po-216) ¼ Sipi1i (Po-216) þ Skpk1k (Bi-212) þ (12 pS) ffi Smpm1m (Po-212),

1p (Pb-212) ¼ Skpk1k (Bi-212) þ (12pS) ffi Smpm1m (Po-212),
1p (Bi-212) ¼ Skpk 1k (Bi-212) þ (12 pS) ffi Smpm1m (Po-212),
1p (Po-212) ¼ Skpk1k(Bi-212).

The potential alpha energy is a quantity for characterizing radon progeny atmospheres, not radon atmo-
spheres. The index refers to the radon isotope and the decay chain.
Since 1p(Pb-214) ¼ 1p(Bi-214)  1p(Po-214) and 1p(Pb-212) ¼ 1p(Bi-212), the equations are rather simple:

1p,Rn-222 ¼ 1p(Po-218) NPo-218 þ1p (Pb-214) (NPb-214 þ NBi-214 þ NPo-214)

1p,Rn-220 ¼ 1p(Po-216) NPo-216 þ1p(Pb-212) (NPb-212 þ NBi-212) þ1p(Po-212) NPo-212.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

where N is the number of the respective atoms. Since the value is not directly connected with a measurand, in
contrast to the potential alpha energy concentration, it should be used for theoretical work (modeling and
simulation) only. The unit of potential alpha energy is J.

Potential Alpha Energy Concentration (PAEC)

The concentration of any mixture of short-lived radon decay products in air in terms of the alpha energy
released during complete decay through Pb-210 for Rn-222 progeny or through Pb-208 for Rn-220 progeny.
Since 1p(Pb-214) ¼ 1p(Bi-214)1p(Po-214) and 1p(Pb-212) ¼ 1p(Bi-212), the equations are rather simple:
 
CðPo-218Þ CðPb-214Þ CðBi-214Þ CðPo-214Þ
Cp;Rn-222 ¼ 1p ðPo-218Þ þ þ þ 1p ðPb-214Þ
lPo-218 lPb-214 lBi-214 lPo-214
 
CðPo-216Þ CðPb-212Þ CðBi-212Þ CðPo-212Þ
Cp;Rn-220 ¼ 1p ðPo-216Þ þ þ 1p ðPb-212Þ þ 1p ðPo-212Þ
lPo-216 lPb-214 lBi-214 lPo-212

The unit of potential alpha energy concentration is J m23.

Note: Due to the short half-lives of 216Po and 212Po, these isotopes are in activity equilibrium with their parent
nuclide: C (Rn-220) ¼ C (Po-216) and C (Bi-212) (1 2 pS) ¼ C (Po-212) with the transition probabilities pk for
the a-decays of 212Bi resulting to pS ¼ Skpk, where C is the measurand, that is the activity concentration of the
respective progeny.

Potential Alpha Energy Exposure

The time integral of the potential alpha energy concentration in air Cp over a given time period Dt.
ð
PðCp;Rn ; DtÞ ¼ Cp;Rn ðtÞ dt
Dt

The unit of potential alpha energy exposure is J m23 h.

Pressure and Pressure Difference

Pressure p and pressure difference Dp from a reference pr. The unit of pressure and pressure difference is
Pascal (Pa).

Progeny
The term progeny includes the whole set of short-lived decay products of a specified radon decay chain. A
particular isotope is indicated by its chemical symbol followed by its mass number. The term progeny of
222
Rn refers to 218Po, 214Pb, 212Bi, and 214Po, while the term progeny of 220Rn refers to 216Po, 212Pb, 212Bi,
212
Po, and 212Tl.

11
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Radiation Weighting Factor

A factor by which the organ or tissue absorbed dose is multiplied to reflect the higher biological effectiveness
of high-LET radiations compared with low-LET radiations. It is used to derive the equivalent dose from the
absorbed dose averaged over a tissue or organ. See also Equivalent dose and Effective dose in this glossary.
The radiation weighting factors represent consensus values of the maximum RBE values for a given radi-
ation for radiation protection purposes and do not represent true RBE values.

Radon Concentration
The radon concentration, cRn,is defined as the amount of a constituent nRn divided by the volume of the
mixture V
nRn
cRn ¼
V

The unit of radon concentration is mol m23.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Note: Radon concentration is often used instead of radon activity concentration. This can cause confusion
and should be avoided.

Radon Entry Rate

The radon entry rate is the radon activity entering the house per unit time.
The flow rate of soil gas entering a house is caused by a combination of different processes (diffusion, exhal-
ation, or convection). The radon entry rate covers all of these processes. The unit of radon entry rate is Bq s21.

Radon Leakage Area (RLA)

Leakage area for flow of radon, that is analogous to the effective leakage area used for building envelope air
leakage. The unit of radon leakage area is m2.

Reference Level
A national reference level for radon represents the maximum accepted radon activity concentration in a resi-
dential dwelling and is an important component of a national program. For homes with radon activity concen-
trations above these levels, remedial actions may be recommended or required. When setting a reference level,
various national factors such as the distribution of radon, the number of existing homes with high radon activity
concentrations, the arithmetic mean indoor radon level, and the prevalence of smoking should be taken into
consideration.

Relative Biological Effectiveness (RBE)

The ratio of a dose of a low-LET reference radiation to a dose of the radiation under consideration, that gives
an identical biological effect. RBE values vary with dose, dose rate, and biological endpoint considered. In
radiological protection, the maximum RBE for stochastic effects at low doses (RBEM) is of particular interest.
Note: Radiation weighting is based mainly on an evaluation of the relative biological effectiveness (RBE) of
the different radiations with respect to stochastic effects. The RBE is used in radiobiology for characterizing
the different biological effectiveness of radiations. RBE values are given as the ratio of the absorbed doses of
two types of radiation producing the same specified biological effect in identical irradiation conditions (dose
value of a reference radiation divided by the corresponding dose value of the considered radiation which
causes the same level of effect). RBE values for a specific radiation depend upon the conditions of exposure in-
cluding the biological effect investigated, the tissue or cell type involved, the dose and the dose rate, and the
dose fractionation scheme; therefore, for a given type and energy of radiation, there will be a range of RBE
values. The RBEs reach maximum values (RBEM) at low doses and low dose rates. RBEM is therefore of par-
ticular interest for defining radiation weighting factors for use in radiological protection. The weighting
factors are taken to be independent of the dose and dose rate in the low-dose region.

Risk
Risk relates to the probability or chance that an outcome, e.g., lung cancer, will occur.

12
 Glossary

† Excess absolute risk

An expression of risk based on the assumption that the excess risk from radiation exposure adds to the
underlying (baseline) risk by an increment dependent on dose but independent of the underlying natural or
background risk. The lifetime excess absolute risk is the risk cumulated by an individual up to a given age (typ-
ically 90 years).
† Relative risk
The ratio of the incidence rate or the mortality rate from the disease of interest, e.g., lung cancer, in an
exposed population to that in an unexposed population. The excess relative risk is defined as the relative risk
minus 1.

† Detriment-adjusted risk
The probability of the occurrence of a stochastic effect, modified to allow for the different components of the
detriment in order to express the severity of the consequences.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Seasonal Correction Factor
A multiplying factor applied to a measurement with duration of one or more months in order to derive a
meaningful annual average radon activity concentration.

Shape Factor
The aerodynamic shape factor is a dimensionless constant used to relate the drag force experienced by an
irregularly shaped particle moving in air relative to the particle’s equivalent volume diameter.

Temperature and Temperature Difference

Absolute temperature T and temperature difference DT (here: indoor2outdoor difference). The unit of
temperature and temperature difference is K.

Thermodynamic Diameter
See activity median thermodynamic diameter.

Tidal Volume
The tidal volume is the air volume inhaled in a single breath for any given physical activity. The unit of tidal
volume is m3.

Tissue Weighting Factor

This is a factor wT by which the equivalent dose to a tissue or organ is weighted to represent the relative con-
tributions of that tissue or organ to the total radiation detriment from stochastic effects resulting from
uniform irradiation of the body.
It is defined such that:
X
wT ¼ 1
T

Total Lung Capacity

The total lung capacity is the air volume of the lung at the maximum inspiratory level. The unit of total lung
capacity is m3.

Type A Evaluation of Measurement Uncertainty

Type A evaluation of a component of measurement uncertainty is carried out by a statistical analysis of mea-
sured quantity values obtained under defined measurement conditions.

13
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Type B Evaluation of Measurement Uncertainty

Type B evaluation of a component of measurement uncertainty is determined by means other than a Type A
evaluation of measurement uncertainty.

Unattached Fraction
A fraction of progeny may not become attached to airborne particles and this quantity is often referred to as
the free or unattached fraction. The unattached fraction is defined as the fraction of the potential alpha energy
concentration of short-lived radon progeny that is not attached to the ambient aerosol.
Note 1: The particle size concerned is of the order of magnitude of nanometer. If there is need for a more
precise definition of unattached progeny, then this report proposes 5 nm diameter as an upper limit for the un-
attached progeny (i.e., clusters carrying progeny).
Note 2: The sum of attached and unattached fraction is equal to 1.

Uncertainty Budget

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The uncertainty budget of a measurement uncertainty is the statement of the components of that measure-
ment uncertainty, and of their calculation and combination.
Note: An uncertainty budget should include the measurement model, estimates, and measurement uncer-
tainties associated with the quantities in the measurement model, covariances, type of applied probability
density functions, degrees of freedom, type of evaluation of measurement uncertainty, and any coverage
factor.

Uncertainty of Measurement
A parameter associated with the result of a measurement that characterizes the dispersion of the values
that could reasonably be attributed to the measurand.
Note 1: The parameter may be, for example, a standard deviation (or a given multiple of it), or the half-width
of an interval having a stated level of confidence.
Note 2: Uncertainty of measurement comprises, in general, many components. Some of these components
may be evaluated from the statistical distribution of the results of series of measurements and can be charac-
terized by experimental standard deviations. The other components, which can also be characterized by stand-
ard deviations, are evaluated from assumed probability distributions based on experience or other
information.
Note 3: It is understood that the result of the measurement is the best estimate of the value of the measur-
and and that all components of uncertainty, including those arising from systematic effects such as compo-
nents associated with corrections and reference standards, contribute to the dispersion.

Volume Flow Rate

The volume flow rate is a quantity equal to the infinitesimal volume dV of a substance crossing a given
surface during a time interval with infinitesimal duration dt, divided by this duration, thus

dV
qV ¼
dt

The unit of volume flow rate is m3 s21.

Note: In the special case of the volume flow rate of the lung, this quantity is the tidal volume divided by the
inhalation time.

Volume
Referred volume of the respective calculation or measurement. The unit of volume is m23.

Working Level (WL)

Working level is a historical unit of potential alpha energy concentration. 1 WL ¼ 20.8 mJ m23.
Note 1: Although the WL is not an SI unit, it is still used for the characterization of the radon progeny activ-
ity concentration, particularly to understand the historical publications.

14
 Glossary

Note 2: 1 WL was originally defined as the concentration of potential alpha energy associated with the
radon progeny in equilibrium with 100 pCi l21 (3700 Bq l21). However, in ICRP Publication 65, it was rede-
fined as: Any combination of short-lived progeny of radon in 1 l of air that will result in the emission of 1.30 
105 MeV of potential alpha energy (both definitions are identical within a few percent).

Working Level Month (WLM)

Working Level Month is a historical unit of potential alpha energy exposure. 1 WLM ¼ 3.54 mJ h m23. It cor-
responds to the cumulative exposure from breathing an atmosphere at a concentration of 1 WL for a working
month of 170 h.
Note: The WLM is not an SI unit, but it is important to understand the historical publications (see Working
Level).

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

15
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
 Abstract

Lung cancer risk caused by the inhalation of radon (222Rn) and its short-lived progeny is related to lung
dose, which cannot be directly measured. The only measurable parameters which allow the determination of
lung doses are the radon and radon progeny activity concentrations and related size distributions. Although
lung cancers are caused by the inhaled short-lived radon progeny and not by the radon gas, it is the radon gas
which is commonly measured and not its progeny. Since radon gas measurements are much easier to carry

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

out, require less expensive equipment and are especially suited for long-term measurements, the report
focuses on the measurement of the radon gas for specific exposure conditions in homes and workplaces.
The first objective of this report is to provide information on how to measure radon, covering measurement
techniques of radon in air and water, currently available detection systems, and measurement strategies most
appropriate for the desired goal of a measurement campaign. Critical measurement strategy decisions are the
selection of the measured radionuclide (i.e., radon gas or radon progeny and related size distributions), choice
of the measurement period (i.e., short-term or long-term measurements), the choice of detector and its deploy-
ment, the type of measurement (i.e., areal or personal measurements), the survey strategy (i.e., integral or
time-resolved measurements), or the strategy to accomplish the specific goal of a survey (i.e., measurements
describing the current status or retrospective measurements). The choice of a specific strategy depends on the
purpose of the survey, and differs therefore between the demands of a nation-wide indoor radon survey or an
epidemiological study.
The second objective of this report is how to interpret and report the results of these measurements, the
associated uncertainties, and the resulting dosimetric estimates. Care should be taken when reporting and
interpreting radon measurements because measured radon activity concentrations exhibit significant spatial
variations (i.e., local and areal), and temporal variations (i.e., diurnal, seasonal, and annual). Consequently,
estimates of the average annual radon activity concentrations are typically used for radon surveys and are
compared with reference levels for radiation protection purposes. Other factors that may affect the interpret-
ation of radon measurement results and the related dose estimates include thoron (220Rn) interference on
radon detection systems, variations of aerosol parameters, equilibrium factor, duration of exposure (i.e., occu-
pancy times in a building or location) and breathing rates. Often encountered problems are the uncertainties
in extrapolating short-term measurements carried out at different locations within a building, or at different
times during a year or in different years to statistically reasonable average values.
Finally, the third objective of this report is to provide recommendations on optimal measurement strategies,
measurement techniques, recording and reporting of measurements for different measurement objectives,
such as individual exposure, average population exposure in a region or country, epidemiological studies or
compliance with reference levels in radiation protection.
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv004
Oxford University Press

Executive Summary

There are several isotopes of radon, but the most inhalation of radon and its progeny, most notably
important ones from a radiation protection perspec- lung cancer risk, are briefly discussed in Section
tive are 222Rn (historical name: radon) and 220Rn 2. Since lung cancer risk is related to the dose deliv-
(historical name: thoron). Since radon activity con- ered by alpha particles to sensitive target cells of the
centrations in homes are much higher than those of bronchial epithelium, Section 3 is devoted to dosi-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

thoron in most locations, this report focuses on recom- metric aspects of radon and radon progeny inhal-
mendations regarding measurements and reporting ation, focusing on lung dosimetry in order to
of inhaled 222Rn gas and its short-lived progeny. establish relationships between lung doses and
Lung cancers are caused by the inhaled short-lived radon and radon progeny measurements.
radon progeny and not by the radon gas. However, Although radiation doses in lung tissue cannot be
radon gas measurements are much easier to carry measured, lung doses are related to radon and radon
out, require less expensive equipment, and are espe- progeny exposure parameters in terms of dose per
cially suited for long-term measurements. Thus, it is unit exposure, e.g., in mSv (Bq h m23)21 for radon
the radon gas that is commonly measured and not measurements or in mSv WLM21 (Working Level
the short-lived radon progeny. Using a typical value Month) for radon progeny measurements. To assess
of the equilibrium factor F appropriate to specific lung cancer risk for specific exposure conditions
exposure conditions, measured radon activity concen- in homes and workplaces, we have to measure the
trations can be used to estimate the contribution relevant radon exposure parameters, such as radon
of radon progeny to lung dose. It is not possible, activity concentrations, radon progeny activity con-
however, using measured radon activity concentra- centrations, and related size distributions. Thus, the
tions to obtain information on unattached fractions radiological and aerosol characteristics of radon and
and size distributions of radon progeny. However, its progeny and their behavior in indoor environ-
despite these limitations, radon gas can be regarded ments are described in detail in Section 4, thereby
as a reasonable surrogate for radon progeny for providing the necessary information as to which
typical radon progeny exposure conditions. exposure parameters have to be measured to charac-
The objectives of this report are to provide guid- terize the exposure conditions.
ance to organizations planning a measurement After having identified the relevant exposure
campaign and to individuals conducting such a parameters, currently available detection devices
survey of how to measure radon and its short-lived and measurement techniques to measure these
progeny, how to report the results of these measure- parameters are presented in Section 5. These
ments, and what are the uncertainties associated include experimental detection systems that provide
with these measurements. To accomplish this goal measurements of radon gas in air and water, radon
and to provide relevant information on radon and progeny activity concentrations, and activity size
radon progeny behavior in homes, workplaces, and distributions. The measurement techniques for per-
outdoors, the report is subdivided into nine sections, sonal monitoring of radon and its progeny are also
supplemented by three appendices, each addressing discussed as well as retrospective measurements
a specific issue related to radon and radon progeny and personal monitoring for radon and its progeny
measurements and their statistical interpretations. for risk assessment purposes, e.g., in epidemiologic-
Section 1 sets the stage for the objectives of this al studies.
report, providing basic information on radon levels Before starting a measurement campaign, it is
typically encountered in dwellings and outdoors and imperative to select a measurement strategy, which
discussing reference levels for the protection of the is appropriate for the desired aim of the campaign.
population against radon. To understand why it is Different measurement strategies for radon and radon
necessary to measure radon levels in homes and progeny measurements are compared and evaluated
workplaces, potential health effects caused by the in Section 6, comprising sampling strategies, detector

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

deployment, time of measurement, short-term or long- dosimetry-related uncertainties affect the analysis of
term measurements, and areal or individual measure- epidemiological studies.
ments. The choice of a specific strategy depends on the Based on the previous sections, the final Section 9
purpose of the survey, and differs therefore between provides recommendations on measurement strat-
the demands of a nation-wide indoor radon survey or egies, measurement techniques and reporting of
an epidemiological study. measurements for different measurement objec-
In general, measured radon activity concentrations tives, such as individual exposure, average popula-
exhibit significant spatial variations (i.e., local and tion exposure in a region or country, epidemiological
areal) and temporal variations (i.e., diurnal, seasonal, studies, or compliance with reference levels in radi-
and annual). Therefore, great care should be taken ation protection.
when interpreting radon and radon progeny mea- For the benefit of the reader, the concepts of me-
surements. The interpretation of results of radon and trology and quality assurance relevant for radon
radon progeny measurements, discussed in Section 7, and radon progeny measurements as well as exam-
must consider the effects of spatial and temporal ples for the analyses of uncertainties in the calibra-
variations, thoron (220Rn) interference on radon tion by a primary radon activity standard and in

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

detection systems, variations of aerosol parameter interlaboratory comparisons are summarized in
values, equilibrium factor, duration of exposure (i.e., Appendix A (related to Section 5). To give practical
occupancy times in building), and individual breath- advice for the analysis of measurement results, two
ing rates. detailed examples are given in Appendices B
The variabilities and uncertainties of radon and (related to Section 7) and C (related to Section 8).
radon progeny exposures and methods to derive Appendix B describes a method to analyze the
meaningful average values from these measurements results of the nuclear track detector exposure at the
are further explored in Section 8. Often encountered Physikalisch-Technische Bundesanstalt (PTB) in
problems are the uncertainties in extrapolating short- Braunschweig, Germany, considering the cross-
term measurements carried out at different locations sensitivity to radon and thoron and the determin-
within a house, or at different times during a year or ation of the associated decision threshold and the de-
in different years to statistically reasonable average tection limit. A measurement method using solid-
values. In addition to these measurement uncertain- state nuclear track detectors, calibrated in the PTB
ties, there are also uncertainties in dosimetric results radon reference chamber, for the determination of
based on these measurements, such as the application a calibration coefficient and how the information
of different dosimetry models and the uncertainties derived can be used in field measurements is
in model parameter values. Both measurement and described in Appendix C.

20
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv006
Oxford University Press

1. Introduction

Why is radon such an important issue? First of Typically, uranium is present at concentrations of
all, epidemiological studies have demonstrated that between 1 and 3 parts per million ( ppm) in most
inhalation of radon and its short-lived progeny can rocks and soils. The uranium content of a soil will be
cause lung cancer, i.e., radon is a potent carcinogen. similar to the uranium content of the rock from
Secondly, radon is a ubiquitous natural radionuclide which it was derived. Ra-226, which is the immedi-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

which can be found everywhere in the world, i.e., ate parent of radon, is a decay product of uranium
everybody in a population is exposed to radon and not (Section 4.1). In general, the higher the uranium
only a selected group, such as smokers, in the case of content and the gas permeability of a soil, the
lung cancer, or more general, radiation workers. greater the radon activity concentration in soil gas
In recent years, several reports related to radon and the probability that houses built on such soil
have been published, addressing primarily health will have high levels of indoor radon. The radon pro-
effects aspects, particularly lung cancer risk (EPA, duced by the decay of radium in the soil subjacent to
2003; ICRP, 2010; NA/NRC, 1991; 1999a; WHO, a house is usually the main source of indoor radon.
2009). It is not the intention of this report to dupli- Soil gas containing radon may enter a house by
cate these extensive reports, but rather to focus on pressure-driven flow through the foundations. This
issues not specifically addressed there, such as the is because the air in a house is generally warmer
optimal planning of measurements, their experi- and at a lower pressure than the subjacent soil gas
mental realization and final reporting, and the un- (Mäkeläinen et al., 2001). In soil gas, radon activity
certainties associated with such measurements and concentrations typically range from less than 10 000
resulting dosimetric estimates. Although it does not up to 100 000 Bq m23. Less than 1% of the indoor air
deal with the risk aspect of radon inhalation, the in a house usually originates in the soil; the remain-
issues addressed in this report do have the potential der coming from outdoor air, which is generally
for improving such risk estimates. quite low in radon activity concentration. Houses
There are numerous isotopes of radon (Firestone with poorly sealed foundations, built on high perme-
and Shirley, 1999), but the most important ones ability ground and with several entry points for soil
for radiation protection are 222Rn (historical name: gas, may draw more than 10% of their indoor air
radon) and 220Rn (historical name: thoron). However, from the soil. Thus, even if the soil gas has only mod-
since radon activity concentrations in homes are erate levels of radon, the activity concentration of
much higher than those of thoron in most locations, radon inside such houses may be relatively high.
with the exception of thorium-rich areas, this report The radon exhaled from building materials in most
focuses on recommendations regarding measure- cases does not significantly contribute to indoor radon
ments and reporting of inhaled radon gas and its levels. The radium content of building materials will
short-lived progeny. be similar to the rock or clay from which they are
The comparison between radon and radon progeny made, which is generally low. However, some building
doses in the bronchial region of the lung indicates materials which may have high concentrations of
that radon progeny doses are about two orders of radium: alum shale concrete and building materials
magnitude higher than corresponding radon doses. made of volcanic tuff, by-product phosphogypsum,
This clearly illustrates that the dose to the lungs and and some industrial waste materials are examples of
hence the resulting lung cancer risk arises mainly such materials (Keller et al., 2001).
from the inhalation of the short-lived progeny. An increment to indoor radon levels might also
come from water supplies. Surface reservoir water
supplies and rivers usually contain very little radon.
1.1 Indoor Radon
But groundwater may contain high activity concen-
Uranium and thorium occur naturally in soil and trations of radon depending on the uranium/radium
rocks and provide a continuous source of radon. content of the aquifer formation. Public water works

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

using groundwater and private domestic wells often radon levels in the water, and industrial buildings
have closed systems with short transit times that do with specific work practices and ventilation condi-
not remove radon from the water or permit it to tions. More information on radon in workplaces will
decay. The radon is out-gassed from the water to the be given in Section 4.4.2.
indoor air when the water is used for washing,
cooking, and other purposes in a house. The areas
where groundwater radon is most likely to make a
1.2 Outdoor Radon
significant contribution to indoor air are areas
that have high levels of uranium in the underlying Land masses are the main sources of outdoor
rocks. Radon activity concentrations as high as radon, while sea waters, having very low radium
several thousand Bq l21 have been found in water concentrations, can be considered as radon sinks.
from drilled wells in regions with granite rock or Consequently, outdoor air radon levels are much
other uraniferous rocks and soils (NA/NRC, 1999b; lower (about 0.1 Bq m23) over oceans and seas than
UNSCEAR, 2008). A summary of all published data over a continental land mass such as mainland
concluded that the contribution of radon in domestic Europe (Chevillard et al., 2002). National data on

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

water supplies to indoor air radon is about 1 Bq m23, average outdoor radon levels are quite limited.
if water with 10 000 Bq m23 of radon is being used in Averages appear to lie between 5 and 20 Bq m23
a house (NA/NRC, 1999b). (UNSCEAR, 2008). The ratio of the radon activity
The distribution of indoor radon gas levels in concentration in outdoor air to the mean indoor
dwellings in many countries has been determined radon activity concentration in European countries
both by national surveys and in other investigations. would appear to be in the range of about 7% (Czech
Figure 1.1 illustrates the significant variability of Republic) to 20% (UK). In the USA, EPA has con-
measured mean indoor radon activity concentrations ducted measurements of outdoor radon in all 50
among 50 countries around the world (UNSCEAR, states. This survey and other state measurements
2008), ranging from only a few Bq m23 in Cyprus to were summarized in NA/NRC (1999b). A mean for the
about 140 Bq m23 in the Czech Republic. USA from 437 measurements is 14.8 + 5.3 Bq m23.
Indoor workplaces include, for example, schools, Radon levels in outdoor air are determined mainly
hospitals, post offices, jails, shops, cinemas, office by the soil characteristics (uranium/radium content,
buildings, and common workshops. The primary porosity, and the consequent radon exhalation rate),
workplaces where radon may cause health problems local topology, and meteorological conditions. In
are underground mines, in particular uranium some conditions, such as atmospheric temperature
mines, waterworks in the case of sufficiently high inversions in valleys with high radon fluxes from the

Figure 1.1. Mean indoor activity concentrations from 222Rn surveys in 50 countries (UNSCEAR, 2008).

22
 Introduction

soil, short-term, elevated outdoor radon levels have level is expressed as an average annual radon activity
been observed. Although high outdoor radon levels concentration and represents a level where action
are rare, they could however be of local health sig- would almost certainly be warranted to reduce expos-
nificance in communities in areas such as former ure. In its 2007 recommendations, ICRP recom-
uranium mining districts where elevated radon ex- mended an upper reference level of 600 Bq m23 for
halation from tailing ponds combined with meteoro- dwellings and 1500 Bq m23 for workplaces (ICRP,
logical and topological conditions could give rise to 2007). However, as ICRP now recommends a nominal
high outdoor radon levels of seasonal duration. risk coefficient for radon which is twice the previous
A direct proportionality in risk between indoor value, the upper reference level for dwellings has
and outdoor radon exposures based simply on the been reduced to 300 Bq m23 (ICRP, 2010).
radon activity concentration should not, however, be It is the responsibility of national authorities to
assumed. This is because factors that influence lung establish their own national reference levels, taking
dose such as the equilibrium factor between radon into account the prevailing economic and societal
and its progeny and also aerosol characteristics can circumstances and specific local exposure conditions
differ considerably between indoors and outdoors. in their country. For dwellings, the World Health

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Organization (WHO) recommends a national refer-
ence level of 100 Bq m23, but states that if this
1.3 Thoron cannot be implemented under the prevailing
country-specific conditions, then the chosen value
There are many isotopes of radon most of which
should not exceed 300 Bq m23 (WHO, 2009). This is
have very short (,1 s) half-lives (Firestone and
consistent with ICRP’s recommendations (ICRP,
Shirley, 1999). As a consequence of these short half-
2010). A survey of 36 countries carried out by WHO
lives, very little of these isotopes can migrate from
found that almost all countries have reference levels
their source to contribute to the activity in indoor
between 200 and 400 Bq m23 for existing housing.
air. Apart from 222Rn, the only other radon isotope
Some countries have different reference levels for
that can occur indoors in significant amounts is
220 new and existing buildings with lower values for
Rn, commonly referred to as thoron. It is a
new houses (WHO, 2009). For comparison, the US
member of the 232Th decay series and its immediate
Environmental Protection Agency (EPA) (EPA,
parent is 224Ra (Section 4.1). There has been an in-
2003) proposes a desired action level of 74 Bq m23
creasing interest in indoor thoron and its progeny in
(or 2 pCi l21) and recommends mitigation if the
recent years. Due to its short half-life, thoron in the
radon activity concentration exceeds 148 Bq m23 (or
soil gas beneath a building cannot survive long
4 pCi l21).
enough, in most situations, to enter a building and
In its Statement on Radon (ICRP, 2010), the
thereby contribute significantly to the level of
Commission also recommended a level of 1000 Bq m23
thoron in indoor air. Indoor thoron is generally due
as an entry point for applying occupational radio-
to the exhalation of thoron from thorium that may
logical protection requirements in existing exposure
be present in the materials of the internal surfaces
situations, replacing the 1500 Bq m23 upper refer-
of the building. There are some building materials,
ence level for workplaces. In its recent publication on
such as volcanic tuff in Italy, that have been found to
“Radiological Protection against Radon Exposure”
have a high rate of thoron exhalation (Nuccetelli and
(ICRP, 2014), the Commission retains the upper
Bochicchio, 1998). While, in general, indoor thoron
reference level value of 300 Bq m23 for dwellings
levels are low, uncommon situations have been iden-
and recommends the same value of 300 Bq m23 for
tified in recent years, such as cave dwellings in
all mixed-use buildings (i.e., with access for both
China, where airborne thoron progeny concentra-
members of the public and workers) and workplaces.
tions can contribute significantly to radiation doses
This includes, for example, schools, hospitals, post
received by the occupants. In some cases, they have
offices, jails, shops, cinemas, office buildings and
been found to exceed those from the radon (222Rn)
common workshops.
progeny in the same location (Tokonami et al., 2004).
A specific graded approach is recommended by
ICRP (2014) for workplaces: in workplaces where ex-
posure to radon is not considered as occupational,
1.4 Protection against Radon
the first step is to reduce the activity concentration
ICRP’s protection policy against radon is based on of radon as low as reasonably achievable below the
setting reference levels and applying the principle same reference level as set for dwellings. If difficul-
of optimization to reduce exposures as low as reason- ties are met in the first step, a more realistic ap-
ably achievable. For indoor radon, the reference proach is recommended as a second step, consisting

23
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

of optimizing protection using the actual parameters 1.5 Aim of the Present Report
of the exposure situation, such as occupancy, to-
It is important to note that lung cancers are
gether with a reference level of 10 mSv annual dose.
caused by the inhaled short-lived radon progeny and
If despite all reasonable efforts, individual doses
not by the radon gas. Thus, it is necessary to deter-
remain above 10 mSv, then the workers are consid-
mine levels of the short-lived radon progeny.
ered as occupationally exposed and the relevant
However, what is routinely measured is the radon
requirements for occupational exposure would
gas and not its progeny. This raises the question: is
apply. However, for some workplaces, such as under-
radon gas an appropriate surrogate for short-lived
ground miners, national authorities may consider
radon progeny? As will be discussed in his report,
from the outset that workers’ exposure to radon is
radon gas measurements are much easier to carry
occupational. Occupational Health and Safety is a
out, require less expensive equipment, and are espe-
priority now in both developed and developing coun-
cially suited for long-term measurements.
tries. Radon activity concentrations are increasingly
Lung cancer risk caused by the inhalation of radon
monitored at workplaces and mitigation actions are
and its short-lived progeny is related to lung dose,
implemented when required.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

which cannot be measured in humans. What can be
Radon reference levels in homes suggested at the
measured, however, are radon and radon progeny ac-
national and international level are based on epide-
tivity concentrations and related size distributions,
miologically obtained lung cancer data. Both lung
which then serve as input data for dose calculations.
cancer risk at low exposure levels as well as related
Thus, the aim of this report is to advise:
radon activity concentrations are subject to signifi-
cant statistical uncertainties. Thus, the question † how to measure radon and its progeny,
arises whether the recommended reference levels † how to report the results of these measurements
can be justified on statistical grounds. In other appropriate for the goal of a measurement,
words, how accurate are the exposure (or dose) – and
effect relationships upon which these reference † how to determine and reduce the uncertainties
levels are based. First of all, the effect of radon expo- associated with these measurements and result-
sures on lung cancer risk at low radon exposures is ing dosimetric estimates
blurred by the presence of many confounding factors
and effect modifiers which may act in a synergistic The first objective is to give advice (i) to authorities
or antagonistic fashion. Secondly, the reconstruction planning radon surveys on measurement strategies
of the past exposure history in epidemiological and types of measurements required for a radon
studies is extremely difficult as it is commonly based measurement program or an epidemiological study
on a small number of measurements. Due to a (Section 6), and (ii) to the personnel actually carrying
latency period for bronchial tumors of about 5 years, out such measurement programs on the type of mea-
the determination of statistically significant average surements that should be carried out to characterize
radon concentrations over a period of many years is radon and radon progeny exposures, such as aerosol
affected by local and temporal variations. In add- size distributions and potential alpha energy (PAE)
ition, all radon measurement devices are subject exposures in homes and workplaces (Sections 5 and
to statistical errors, due to instrumental errors or 7). Both comprise measurements required for a radon
incorrect calibration. Since bronchial tumors are measurement program to check that radon levels are
produced by inhaled radon progeny and not by the below the reference levels both at home and in the
radon gas, uncertainties exist about the conversion workplace, measurements required for dose assess-
of radon activity concentrations to radon progeny ac- ment, particularly if radon levels are close to reference
tivity concentrations, as well as about the aerosol levels, and measurements required for epidemiologic-
size distributions to which inhaled radon progeny al studies to be related to lung cancer risk.
are attached. Furthermore, different lung dosimetry The second objective is to analyze all potential un-
models produce a range of doses in sensitive target certainties involved in the determination of radon
cells for the same exposure conditions (Section exposures and related bronchial doses since the un-
3.8.1). Finally, uncertainties of individual doses per certainties associated with the exposure or dose axis
unit exposure to radon and its progeny are caused of the exposure (dose) –effect relationship will affect
by inherent inter-subject variability of lung para- the validity of any chosen reference level (Section 8).
meters and related breathing patterns. Although Finally, the third objective is to provide recom-
the statistical errors associated with current lung mendations on (i) the proper measurement and com-
cancer risk estimates at low radon exposures are plete recording of the results of measurements of
also quite significant, these uncertainties are not radon, thoron, and their progeny, and (ii) on the
considered in this report. correct reporting of these measurements (Section 9).

24
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv005
Oxford University Press

2. Health Effects of Radon Exposure

Radon has long been identified as a cause of lung In the joint analysis of underground miners
cancer at higher activity concentrations and it was described in the BEIR VI report (NA/NRC, 1999a), 6
recognized as a human lung carcinogen by the out of the 11 cohorts had some smoking information.
National Institute for Occupational Safety and Health The analysis of these data showed that the relative
(NIOSH, 1971), the World Health Organization risk (RR) of lung cancer with cumulative exposure to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(WHO, 1986; 2009), and by the National Research radon was linear for lifelong non-smokers and for
Council (NA/NRC, 1999a). The main source of infor- current and ex-smokers. Assuming a RR model, the
mation on risks of radon-induced lung cancer has excess relative risk (ERR) per unit increase in radon
been epidemiological studies of underground miners exposure was higher among lifelong non-smokers
(ICRP, 1993a), and more recent studies have provided compared with current and ex-smokers, although
informative data on risks at lower levels of exposure the confidence intervals overlapped. This suggests
(Darby et al., 2005; 2006; Hunter et al., 2013; Lubin sub-multiplicative interaction between radon and
et al., 1997; NA/NRC, 1999a; Tomášek et al., 2008a; smoking in causing lung cancer (i.e., less than the
Walsh et al., 2010). These studies have shown signifi- product of the individual risks from the two agents
cant associations between cumulative radon exposure but more than the sum of the risks). However, the ab-
and lung cancer mortality at lower radon activity con- solute risk of lung cancer per unit increase in radon
centrations found in homes. exposure is much greater in smokers than in non-
In the BEIR VI report (NA/NRC, 1999a), risk smokers as smokers have much higher rates of lung
models that take account of effect modifying factors cancer than non-smokers in the absence of radon ex-
such as time since exposure, age, and exposure rate posure.
have been derived from the joint analysis of 11 Recently, a joint analysis of European epidemio-
cohorts of miners from China, Czech Republic, USA, logical studies on uranium miners with smoking
Canada, Sweden, Australia, and France. Also more information was carried out (Hunter et al., 2013;
recently, a risk model has been derived from the Leuraud et al., 2011). As expected, the carcinogen-
joint analysis of the French and Czech miner cohorts ic effect of radon exposure was confirmed even
associated with low levels of exposure (Tomášek after adjustment for smoking. The results from
et al., 2008a). Tomášek et al. (2008b) used these risk analyzing the joint effects of radon and smoking
models to calculate the lifetime excess absolute risk indicated a sub-multiplicative interaction; the
(LEAR) for reference populations defined by the ERR WLM21 was greater for non-smokers com-
International Commission on Radiological Protection pared with current or ex-smokers, although there
(ICRP, 2010). Considering a chronic exposure during was no statistically significant variation in the
adulthood, recent estimates of LEAR are significantly ERR WLM21 associated with smoking status. This
greater (by about a factor of 2) compared with previ- is in agreement with the BEIR VI analyses (NA/
ous estimates. As a result, ICRP now recommends NRC, 1999a) and with the results from an updated
a detriment-adjusted nominal risk coefficient for a analysis of the Colorado Plateau miner cohort
mixed adult population of non-smokers and smokers (Schubauer-Berigan et al., 2009). In contrast, a
of 8  10210 per Bq h m23 for exposure to 222Rn in recent nested case –control study of Czech uranium
equilibrium with its progeny, i.e., 5  1024 per WLM miners indicated that the combined effect from
or 14  1025 per mJ h m23 (ICRP, 2010). This new radon and smoking was closer to an additive than to
value is approximately double the previous nominal a multiplicative interaction (Tomášek, 2013). This
risk coefficient given in ICRP Publication 65 (ICRP, was shown only when a modifying effect of time
1993a). It should be noted, however, that the LEAR es- since exposure was used. If the interaction is addi-
timate is dependent upon the background lung cancer tive as opposed to multiplicative then this would
rates assumed for the reference population and this lead to higher estimates of lifetime risks for the non-
strongly depends on the prevalence of smoking. smoking population.

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

In addition to epidemiological studies of under- of lung cancer with increasing domestic radon activ-
ground miners, case – control studies of lung cancer ity concentration, considering exposures over a
and residential radon exposures have been con- period of 30 years preceding the diagnosis of cancer.
ducted. The results from 23 residential case – control There was evidence of a risk of lung cancer even for
studies and 6 pooled or meta-analyses are shown in those exposed to an activity concentration below 200
Figure 2.1 (UNSCEAR, 2008). In particular, four Bq m23 (Darby et al., 2006). The estimates of the RR
joint analyses have been carried out based on data of lung cancer per unit activity concentration of
from Europe (Darby et al., 2005; 2006), North radon in the four joint analyses were close to each
America (Krewski et al., 2005a, 2006), China (Lubin other. The combined estimate for Europe, America,
et al., 2004), and Germany (Wichmann et al., 2005). and China was 1.09 per 100 Bq m23 during a 30 year
Each joint analysis demonstrated an increased risk exposure (UNSCEAR, 2008). However, when the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 2.1. Residential radon risk estimates from individual or pooled studies (UNSCEAR, 2008). Note that all references for this figure
are listed in the UNSCEAR Report.

26
 Health Effects of Radon Exposure

changed to become more like those of men and there-

fore using the data for males were more appropriate
to estimate future risks among women.
On the basis of results from the joint European
residential studies (Darby et al., 2005; 2006), the
absolute risk of lung cancer by age 75 years for life-
long non-smokers was estimated as 0.4%, 0.5%, and
0.7% for lifetime average residential radon activity
concentrations of zero (theoretical non-exposure
situation), 100 and 400 Bq m23, respectively (Darby
et al., 2006). For current smokers (of 15–24 cigar-
ettes d21), the corresponding estimates were about
25 times greater (10%, 12%, and 16%).
Although comparisons between residential studies
and miner studies are complex, appropriate compari-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 2.2. Relative risk (RR) of lung cancer versus long-term sons of lung cancer risks estimates from miner
average residential radon activity concentration in the European studies and indoor studies show good consistency
pooling study (Darby et al., 2005; 2006). Corrections for the
(Hunter et al., 2013; ICRP, 2010; Tomášek et al.,
year-to-year variability in the radon exposure were made. A
best-fitted straight line with 95% confidence intervals is shown; 2008a; UNSCEAR, 2008). For example, based on con-
ERR per 100 Bq m23 increase ¼ 0.16 (95% CI: 0.05–031). Risks versions from WLM to time-weighted average radon
are relative to the extrapolated risk at 0 Bq m23. Adopted from activity concentration, the ERR estimates from the
Darby et al. (2005; 2006). joint European studies on uranium miners were in
agreement with those from the joint analysis of the
European residential radon studies (Hunter et al.,
year-to-year variability in the radon exposure was 2013).
considered in the European study, the estimated RR It was estimated for Europe that lifelong exposure
increased to 1.16 (1.05 – 1.31) per 100 Bq m23. This to radon in homes currently accounts for about 9%
was considered by ICRP as a reasonable estimate of of deaths from lung cancer and hence about 2% of
the risk associated with low prolonged exposures in all cancer deaths (Darby et al., 2005). This was
homes (ICRP, 2010). Figure 2.2 shows the RR versus calculated assuming a population-weighted average
the long-term average residential radon activity con- radon activity concentration of 59 Bq m23 for
centration obtained from the joint European study Europe (UNSCEAR, 2000) with an RR of 1.16 per
after correcting for random uncertainties in the 100 Bq m23. Using a modified version of the BEIR
radon measurements (Darby et al., 2005; 2006). VI risk model (NA/NRC, 1999a) with an estimated
The joint European residential study showed a average residential radon activity concentration of
statistically significant increasing trend in lung 46 Bq m23, the US Environmental Protection
cancer risks with domestic radon exposure for both Agency estimated that about 13% of the lung cancer
smokers and lifelong non-smokers (Darby et al., deaths (i.e., 21 100 deaths) in the USA in 1995 were
2005; 2006). The estimated value of ERR did not radon-related (EPA, 2003). Hence, radon is consid-
differ significantly by smoking status. Consequently, ered to be the second leading cause of lung cancer
Darby et al. (2006) assumed the same ERR for after smoking.
smokers and non-smokers (i.e., they assumed that The miner studies mainly considered exposures
the interaction of the risks from smoking and radon during adulthood; it was only the Chinese tin miner
is multiplicative). Because the background lung study where there were data on exposures in child-
cancer rates in current smokers (of 15 –24 cigarettes hood and adolescence. There was no clear indication
d21) is about 25 times greater than that in lifelong that the ERR depends on age at first exposure, but
non-smokers, the absolute value of risk of lung the data are sparse (NA/NRC, 1999a). In the resi-
cancer per unit increase in radon is also about 25 dential studies, the exposure period considered was
times greater in current smokers compared with 30–35 years prior to diagnosis of lung cancer with
lifelong non-smokers. However, for ex-smokers who an assumed lag time of 5 years from lung cancer in-
gave up smoking more than 10 years ago, the lung duction to diagnosis. As most of the lung cancer
cancer rates are only about 5 times greater than that cases occurred over the age of 50 years, adulthood
for lifelong non-smokers (Darby et al., 2006). These exposures were mainly considered. Therefore, reli-
estimates of the effect of smoking were based on able information of risks of lung cancer arising from
male data only. The reason for this is that in recent exposures to radon and its progeny during childhood
years, European women’s smoking habits have is currently not available.

27
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Studies of underground miners generally have not using a model based on geographical region, soil
shown any excess of cancer other than lung cancer type, and house characteristics rather than by
to be associated with radon exposure (Darby et al., direct measurements in the homes in question.
1995; NA/NRC, 1999a; UNSCEAR, 2008). There ICRP concluded that the review of the available epi-
have been some associations of extra-pulmonary demiological evidence to date shows no consistent
cancers with radon exposure suggested in individual evidence for an association between radon activity
studies, but they have not been replicated in other concentrations and cancer other than that of the
studies and no consistent pattern has emerged. For lung (ICRP, 2010).
example, excesses or trends with radon exposure In sharp contrast to the harmful effects of radon
were noted for leukemia (Rericha et al., 2006), exposure, natural radon-rich thermal water and
kidney (Vacquier et al., 2008), liver, and stomach vapor have been used for decades in radon spas, such
cancers (Kreuzer et al., 2008), but were not con- as Badgastein (Austria), Bad Schlema (Germany), or
firmed by other studies. Several large-scale residen- Misasa (Japan), for the treatment of various rheumat-
tial case – control studies were unable to confirm an ic diseases, such as ankylosing spondylitis (Tempfer
association between radon exposure and leukemia et al., 2010). Some therapeutic benefits have been

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

risk (CCSI, 2002; Lubin et al., 1998; Steinbuch reported for carefully selected patient groups and
et al., 1999). However, a recent residential study in pathologies (Deetjen et al., 2005). However, knowl-
Denmark suggested a significant positive association edge of the physiological and molecular mechanisms
between radon activity concentrations and acute and their time sequence triggered by the radon expos-
lymphocytic leukemia (Raaschou-Nielsen et al., 2008). ure is still too fragmentary to understand why these
A weakness of this study, acknowledged by the reported beneficial effects can be produced by these
authors, was that the radon exposures were estimated relatively small radon doses.

28
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv010
Oxford University Press

3. Radon and Radon Progeny Inhalation and Resultant Doses

3.1. Lung Dose Assessment Models The human respiratory tract is composed of three
functional regions: (i) the extrathoracic region, consist-
Radon progeny doses to the bronchial region of the
ing of the nose, mouth, and pharynx, acting primarily
lung are about two orders of magnitude higher than
as a filter to protect the lungs, (ii) the tracheobronchial
corresponding doses produced by radon gas. Thus, cur-
or conductive region, consisting of bronchial and bron-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

rently used lung dosimetry models focus exclusively
chiolar airways, whose primary task is to distribute
on the prediction of bronchial doses by inhaled radon
inhaled air to the gas-exchange region, and (iii) the
progeny, neglecting the contribution by the inhaled
alveolar–interstitial (pulmonary) region, where the gas
radon gas.
exchange between lung and blood takes place via the
Following the inhalation of radon and its short-lived
alveoli, but where also inhaled radon and progeny may
progeny, the highest doses are received by the bron-
be transferred to the other organs of the human body
chial region of the human lung produced by alpha
via the bloodstream. Bronchial and alveolar regions also
particles emitted from short-lived radon progeny
differ in their clearance mechanisms, with relatively
deposited on bronchial airway surfaces. Hence, the
fast mucociliary clearance in the bronchial airways and
primary health effect due to radon inhalation is the for-
relatively slow macrophage-mediated transport in the
mation of bronchogenic carcinomas predominantly in
alveolar region. For a detailed description of the struc-
bronchial airways. Since it is not possible to measure
ture and function of the human respiratory, the reader
bronchial doses directly, the primary dosimetric issue
is referred to the ICRP (1994) report on the Human
is the development of appropriate dosimetric models
Respiratory Tract Model.
for the calculation of doses to sensitive target cells in
From a modeling point of view, lung dosimetry
bronchial tissue.
models comprise five submodels:
Radon progeny were not recognized as the rele-
vant contributors of dose from radon until the work
of Bale (1980). Harley and Fresco (1951) and Harley (1) a morphometric lung model, detailing the ana-
(1980) showed that radon progeny retention in the tomical structure of the lung in terms of the
lung was about 50% of the inhaled radon progeny ac- number of airway generations and their charac-
tivity. In evaluating the exposure of workers mining teristics, such as diameter, length, branching,
or processing radium-bearing materials, the hazard and gravity angles;
from the inhalation of radon progeny was concluded (2) a respiratory physiology model, which defines
to be much more serious than that from radon itself. the breathing patterns related to different phys-
This was contrary to Evans and Goodman (1940) ical activities;
who had based the early guidelines for mines on the (3) a particle deposition model, which comprises
assumption that the important radiation dose to extrathoracic, bronchial, and pulmonary depos-
bronchial epithelium was the alpha radiation from ition efficiencies, physical deposition mechanisms
radon gas. Bale (1980) estimated that the absorbed in cylindrical airways, and related analytical
bronchial dose from radon progeny could be up to 8 deposition equations for specific flow patterns;
times the dose from radon gas. (4) a bronchial clearance model comprising muco-
The challenges in lung dosimetry of radon progeny ciliary clearance and transport into the blood;
result from the non-uniformity of the radon progeny (5) a dosimetry model, which specifies the geometry
deposition within the bronchial region and among of alpha particle interactions with sensitive target
bronchial airway surfaces, the short range of alpha cells at different depths in bronchial epithelium.
particles in relation to the non-uniform distribution
of sensitive target cells in bronchial epithelium, and Two different modeling approaches are currently
the relationship between energy deposition (stopping used to calculate doses to the lungs following the in-
power) and alpha particle range. halation of short-lived radon decay products:

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

(1) Models in which the absorbed dose to sensitive Table 3.1. Annual equivalent lung dose coefficients following
target cells in bronchial airway generations is cal- continuous inhalation of radon gas and thoron gas
culated, such as the deterministic and stochastic
Radon Annual equivalent References
airway generation models proposed by Haque and isotope lung dose coefficient
Collinson (1967), Harley and Pasternack (1972; [mSv (Bq m23)21]
1982), Hofmann (1982a), Hofmann et al. (2010),
Jacobi (1964), Jacobi and Eisfeld (1980), James 222
Rn 1.1 Pohl and Pohl-Rüling (1977)
222
(1988), Winkler-Heil and Hofmann (2002), and Rn 7.5 Jacobi and Eisfeld (1980)
220
Zock et al. (1996). These models are sometimes Rn 5.8
222
Rn 6.2 Peterman and Perkins (1988)
called anatomical or biological models. 220
Rn 2.9
(2) Semi-empirical compartment models, such as the 222
Rn 5.9 Khursheed (2000)
ICRP (1994) Human Respiratory Tract Model
(HRTM), combine many airway generations into
compartments to simplify dose calculations. These constant activity concentration of 1 Bq m23 in the
models are sometimes called pharmacokinetic or inhaled air, Jacobi and Eisfeld (1980) calculated dose

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

biokinetic models. equivalent rates to different tissues of the human
While the ICRP (1994) model can be applied to the body. For the inhalation of 222Rn, they reported an
inhalation of any radionuclide, the generation-based annual equivalent dose coefficient to the lung of
models were specifically developed for inhaled radon 7.5 mSv (Bq m23)21 for inhaled radon and 5.8 mSv
progeny. (Bq m23)21 for inhaled thoron (Table 3.1).
In general, the structure of the ICRP (1994) com- A multicompartment model to simulate the dy-
partment model is the same as that of the airway gen- namics of inert radioactive gases in the human body
eration models. There are, however, a few significant has been developed by Peterman and Perkins
differences: (1) the lung structure consists of only three (1988), utilizing the blood flow and solubility data
compartments, the large bronchial airways (BB), the reported by Nussbaum and Hursh (1957). In this
smaller bronchiolar airways (bb), and the alveolar– model, it is assumed that an inert gas is transported
interstitial region (AI), instead of single airway genera- through the body to various organs via the blood
tions; (2) deposition fractions in tracheobronchial (TB) stream. This model was used to determine equiva-
(TB ¼ BB þ bb) and AI compartments are obtained by lent doses resulting from inhalation of radon and
appropriate fits to human experimental data and thoron. Annual equivalent lung dose coefficients
expressed as functions of particle and flow parameters were 6.2 mSv (Bq m23)21 for radon inhalation and
(hence the name “semi-empirical”) instead of ab initio 2.9 mSv (Bq m23)21 for thoron inhalation (Table 3.1).
physical deposition calculations; (3) clearance path- A refined dynamic model for the retention of inert
ways and related half-times refer to the whole compart- gases in the body for the calculation of inhalation dose
ment and not to single airway generations; and (4) coefficients has been published by Khursheed (2000).
doses to bronchial target cells are computed for the The annual equivalent dose coefficient for the lungs for
total alpha activity in a given compartment rather the inhalation of radon gas was 5.9 mSv (Bq m23)21
than for steady-state surface activities in individual (Table 3.1), which is very similar to the dose value pre-
cylindrical airway generations. dicted by Peterman and Perkins (1988).
Annual lung equivalent dose coefficients for con-
tinuous inhalation of radon and thoron are listed in
Table 3.1. Doses due to inhaled thoron are consist-
3.2 Radon versus Radon Progeny Doses ently lower than those for radon.
3.2.1 Lung Doses due to Inhalation of Radon
and Thoron Gas 3.2.2 Lung Doses due to Inhalation of
Short-Lived Radon Progeny
Based on experimental observations of the radon
concentrations in different organs and tissues of the While the annual equivalent lung doses due to
human body, Pohl and Pohl-Rüling (1977) derived inhaled radon listed in Table 3.1 were expressed in
organ doses following continuous inhalation of terms of inhaled radon activity concentrations, bron-
radon. For the human lung, they reported an annual chial equivalent doses produced by inhaled short-lived
equivalent dose coefficient of 1.1 mSv (Bq m23)21 for radon progeny are commonly expressed in terms of cu-
the inhaled radon gas alone (Table 3.1) (recalculated mulative exposures, e.g., in mSv (Bq m23)21 or in mSv
from published absorbed doses). WLM21. The most important airborne radon progeny,
Assuming steady-state conditions for the specific from the perspective of radiation dose to lung tissue as
activities of 222Rn and 220Rn in body tissues at a a consequence of inhalation, are the alpha-emitting

30
 Radon and Radon Progeny Inhalation and Resultant Doses

nuclides 218Po (half-life 3.07 min, Ea ¼ 6.11 MeV) and Table 3.3. Comparison of radon progeny absorbed dose
214 coefficients in bronchial (BB), bronchiolar (bb) and alveolar–
Po (half-life 162 ms, Ea ¼ 7.83 MeV). Doses to sensi-
interstitial (AI) airways arising from the exposure to 1 WLM
tive basal and secretory cells for a variety of exposure
predicted by three lung dosimetry models (Winkler-Heil et al.,
conditions have been published in the past and are 2007)
discussed in more detail in Section 3.9.1. Here, only a
few model predictions are presented to illustrate the Region Mode Absorbed dose per WLM (mGy WLM21)
magnitude of bronchial doses relative to the doses
RADEP/ RADOS IDEAL-DOSE
delivered by the radon gas. To facilitate comparison
IMBA
with the dose predictions for the radon gas, reported
radon progeny dose-exposure conversion coefficients BB Unattached 76.5 81.1 76.7
for mine atmospheres (in mSv WLM21) were con- Attached 7.9 6.1 7.0
verted to annual equivalent dose coefficients in homes bb Unattached 25.0 10.4 4.9
[in mSv (Bq m23)21] by assuming an equilibrium Attached 5.6 3.3 3.3
AI Unattached 0.01 0.001 0.003
factor of 0.4 and a K-factor of 1 (Table 3.2).
Attached 0.4 0.3 0.3
The comparison between radon and radon

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

progeny doses in the bronchial region of the lung
listed in Tables 3.1 and 3.2 demonstrates that radon
progeny doses are about two orders of magnitude Inspection of Table 3.3 reveals that the doses to
higher than corresponding radon doses. This clearly the alveolar-interstitial region are between one and
indicates that the dose to the lungs mainly arises two orders magnitude smaller than those to the
from the inhalation of the short-lived progeny. bronchial and bronchiolar regions. These results
While the above dose calculations have shown also demonstrate that the unattached fraction con-
that radon progeny doses from alpha particles are tributes to the total dose only in the bronchial and
primarily responsible for radiobiological effects in bronchiolar regions, but not in the alveolar region.
the lungs, their distribution throughout the lung is, Since bronchial doses are much higher than alveolar
however, relatively inhomogeneous, with signifi- doses, the obvious target region relevant for lung
cantly higher doses to the bronchial (BB) and bron- tumors is the bronchial region.
chiolar (bb) airways than to the alveolar–interstitial In addition to alpha particles, short-lived radon
region (AI). For comparison, regional dose distribu- progeny also emit beta particles. Calculations by
tions per unit exposure in WLM predicted by three Markovic et al. (2011) produced dose-exposure con-
lung dosimetry models for a reference worker version coefficients due to beta radiation of 0.21 mSv
are compiled in Table 3.3, differentiating between WLM21 for 222Rn progeny and 0.06 mSv WLM21 for
220
the attached, i.e., radon progeny attached to the Rn progeny.
ambient aerosol, and the unattached fractions. The
three models applied here are (i) the RADEP/IMBA
3.2.3 Lung Doses due to Inhaled Thoron
code (Marsh and Birchall, 2000), based on the ICRP
Progeny
Human Respiratory Tract Model (ICRP, 1994), a de-
terministic regional compartment model, (ii) the Except for the dual decay between 212Bi and 212Po,
RADOS model (Winkler-Heil and Hofmann, 2002), a the structure of the thoron progeny decay scheme is
deterministic airway generation model, and (iii) the similar to that for the short-lived radon progeny. Only
IDEAL-DOSE model (Hofmann et al., 2010), a sto- radioactive half-lives, alpha particle energies and
chastic airway generation model. related energy-range relationships have to be replaced
by the corresponding values for the thoron progeny.
The most important airborne thoron progeny, from
Table 3.2. Comparison of bronchial annual equivalent dose the perspective of radiation dose to lung tissue as a
coefficients for inhaled short-lived radon progeny predicted by consequence of inhalation, is 212Pb (half-life 10.64 h).
different lung models Lead-212 itself is a beta particle emitter, but when it
decays, it gives rise to the alpha-emitting thoron
References Annual equivalent lung dose
progeny 212Bi (half-life 60.5 min, Ea ¼ 5.5–6.1 MeV)
coefficient [mSv (Bq m23)21]
and 212Po (half-life 3  1027s, Ea ¼ 8.68 MeV).
Jacobi and Eisfeld (1980) 144 A summary of effective dose conversion coefficients
James (1988) 436 for thoron progeny is given in Table 3.4. Values range
Harley et al. (1996) 326 from 1.5 to 5.7 mSv WLM21, depending on activity
Porstendörfer (2001) 395 size distributions, unattached fractions, and the
Marsh et al. (2005) 439
dosimetric lung models employed. Consistent with
Winkler-Heil et al. (2007) 268
earlier dose calculations for inhaled radon progeny

31
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 3.4. Summary of calculated effective dose conversion transport to the alimentary tract. In comparison,
coefficients (mSv WLM21) for thoron progeny indoors based on doses to systemic organs and gastrointestinal tract
adult male breathing conditions (ICRP, 2010; Li et al., 2010)
regions are low and can generally be ignored in the
References Effective dose conversion
calculation of effective doses. The equivalent dose to
coefficient (mSv WLM21) the extrathoracic region of the respiratory tract is
similar to that of the lungs. However, its contribution
Harley and Pasternack (1973) 2.0 to the effective dose is generally quite small as it is
Jacobi and Eisfeld (1980) 1.5 considered to be 1 of the 13 “remainder organs” and
ICRP (1987) 1.8 consequently has a small tissue weighting factor
James (1988) 3.5
Marsh and Birchall (1999a; 1999b) 3.8
(ICRP, 2007). Typically, the lung dose contributes
UNSCEAR (2000) 1.9 about 95% or more to the effective dose following
Porstendörfer (2001) 2.4 the inhalation of 222Rn or 220Rn progeny. The longer
Ishikawa et al. (2007) 5.4 radioactive half-life of the 220Rn decay product 212Pb
Kendall and Phipps (2007) 5.7 (10 h) compared with the 222Rn progeny (,30 min)
Li et al. (2010) 3.8
means that a greater fraction of 212Pb is absorbed in

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Hofmann et al. (2014) 4.6
blood before decay takes place in the lung.
Doses to the lung, extrathoracic region, systemic
(Winkler-Heil et al., 2007), dose conversion coeffi- organs, and the gastrointestinal tract regions arising
cients based on the ICRP (1994) Human Respiratory from the inhalation of radon progeny have been cal-
Tract Model are higher than those obtained by dosi- culated by implementing the ICRP dosimetric and
metric airway generation models. biokinetic models (Kendall and Smith, 2002; 2005;
A comparison between dose conversion coefficients Marsh et al., 2012, for 222Rn progeny calculations;
in terms of mSv per WLM for inhaled radon progeny Ishikawa et al., 2007; Kendall and Phipps, 2007;
(Rn) with those for inhaled thoron progeny (Tn) indi- Tschiersch et al., 2007, for 220Rn progeny calcula-
cates that the ratio Rn:Tn is about 3 (Ishikawa et al., tions). These models include the ICRP Publication 66
2007). However, if expressed in terms of doses per Human Respiratory Tract Model (HRTM) (ICRP,
unit equilibrium equivalent activity concentration of 1994), the ICRP Publication 30 Gastrointestinal (GI)
radon/thoron exposures (in Bq h m23), then the tract model (ICRP, 1980), the ICRP Publication 67
thoron progeny conversion coefficients are greater by biokinetic models for polonium and lead (ICRP,
about a factor 4 than those for short-lived radon 1993b), and the ICRP Publication 30 biokinetic
progeny. This arises because 1 WLM corresponds to model for bismuth (ICRP, 1980).
4.68  104 Bq h m23 equilibrium equivalent activity For illustrative purposes, Table 3.5 gives annual
concentration of thoron and to 6.37  105 Bq h m23 equivalent doses to organs arising from the inhal-
equilibrium equivalent activity concentration of ation of 222Rn progeny calculated by Kendall and
radon over a period of 1 working month (170 h). Smith (2002). These calculations apply to an adult
continuously exposed to an indoor radon (222Rn) con-
centration of 200 Bq m23 (with equilibrium factor
3.3 Lung Doses versus Other Organ Doses F ¼ 0.4, an indoor occupancy of 22 h d21, and an
In this section, the equivalent dose to the lung average breathing rate of 0.9 m3 h21). As can been
and to other organs are considered and compared. seen from Table 3.5, the calculated doses to systemic
Doses arising from the inhalation of radon progeny organs and gastrointestinal tract regions are at least
and from the inhalation of radon gas are discussed two orders of magnitude less than the lung dose,
separately. The external dose to the skin from radon with the kidney receiving the highest organ dose
progeny that have been deposited on the skin outside the respiratory tract.
surface is also considered as well as doses from the The organ doses from the inhalation of thoron
ingestion of radon in water. progeny calculated by Kendall and Phipps (2007)
show that the lung dose is about 30 times or more
greater than the doses to the systemic organs. The
3.3.1 Doses to Internal Organs Arising from
bone surfaces and the kidney receive the next two
Inhalation of Radon Progeny
highest doses after the respiratory tract.
As described above, following exposure to radon
and its progeny, the dose to the lung mainly arises
3.3.2 Doses to Internal Organs Arising from
from the inhalation of the short-lived progeny.
Inhalation of Radon Gas
Because of the short half-lives of the progeny, dose is
delivered to the lung tissues before clearance can take Radon gas is soluble in water, body fluids, and
place, either by absorption into blood or by particle tissue. Volunteer studies in which subjects have

32
 Radon and Radon Progeny Inhalation and Resultant Doses

Table 3.5. Annual equivalent doses to organs arising from the of several minutes, whereas fatty tissues with a poor
inhalation of radon gas and radon progeny at an indoor 222Rn blood supply and high radon solubility have half-
concentration of 200 Bq m23 with equilibrium factor F ¼ 0.4
times of several hours or more. Following continuous
Organ/tissue Annual equivalent doses to organs (mSv)
exposure to radon, equilibrium activity concentrations
are reached typically within an hour for tissues with a
Inhaled radon Inhaled radon Total low retention half-time of a few minutes, whereas for
progenya gasb fatty tissues with a longer retention half-time, it can
take several days to reach equilibrium.
Lung 159 1.2 160
For continuous inhalation of 222Rn or 220Rn gas,
Extrathoracic region 70.9 0.42 71
Stomach 0.08 0.06 0.14 Jacobi and Eisfeld (1980) calculated equilibrium
Small intestine 0.05 0.06 0.11 dose rates to organs and tissues, including the lung,
Colon 0.02 0.05 0.07 liver, kidney, spleen, red bone marrow, bone sur-
Red bone marrow 0.03 0.65 0.68 faces, and body fat. Organ doses have also been cal-
Bone surface 0.17 0.03 0.20
culated with pharmacokinetic models for radon gas
Liver 0.05 0.09 0.14
Breast 0.02 0.42 0.44 based on blood flow rates and radon solubility coeffi-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Kidney 0.54 0.05 0.59 cients (Peterman and Perkins, 1988, for 222Rn and
220
Gonads 0.02 0.05 0.07 Rn; Khursheed, 2000, for 222Rn). The radon solu-
Brain 0.02 0.06 0.08 bility of a tissue is represented by a tissue-to-blood
Bladder 0.02 0.05 0.07
partition coefficient, defined as the ratio of the con-
Muscle 0.02 0.05 0.07
Skinc,d 25.1 — — centration of radon gas in tissue and blood at equi-
Effective dose 19.7 0.28 20 librium. Values of these partition coefficients were
mainly derived from the in vivo data in rats of
a
Doses from inhaled radon progeny were calculated by Kendall Nussbaum and Hursh (1957). The rat data showed
and Smith (2002) for an adult male sedentary worker with an that radon is significantly more soluble in omental
indoor occupancy of 22 h d21 and a breathing rate of 0.9 m3 h21
(Table B.16B of ICRP Publication 66, ICRP, 1994). It was also
fat compared with other tissues. The annual organ
assumed that the radon progeny were moderately soluble in the doses calculated by Khursheed (2000) for a continu-
lung (i.e., Absorption Type M). ous exposure of 200 Bq m23 of 222Rn gas are given in
b
Doses from inhaled radon gas were calculated by Khursheed Table 3.5. The organ receiving the highest dose from
(2000) assuming 100% indoor occupancy. the inhaled 222Rn gas is the lung with the next two
c
External dose to skin taken from Eatough and Henshaw (1992).
d
External dose to skin taken from Harley and Robbins (1992).
highest organs being red bone marrow (RBM) and
breast due to their fat content. The lung dose from
inhaling the radon gas, which was calculated from
inhaled radon or ingested radon-rich water have the decays in the lung air, is about 100 times lower
shown that radon is absorbed into the bloodstream than the dose to the lung from inhaling radon
via the lung or the GI tract and is retained in tissues progeny (Tables 3.1 and 3.2).
with half-times varying from minutes to several Khursheed (2000) calculated an annual equivalent
hours or more (Gosink et al., 1990; Harley and dose to the RBM of about 0.65 mSv at a 222Rn activity
Robbins, 1992; Harley et al., 1994; Hursh et al., concentration of 200 Bq m23 (Table 3.5). It was
1965). The long-term retention half-times were assumed that the RBM consisted of 40% fat cells and
assumed to be associated with retention in body fat. that the fat cells are uniformly distributed throughout
Retention half-times were also shown to decrease the haematopoietic tissue. Similar annual doses have
with exercise (Gosink et al., 1990). been calculated by Richardson et al. (1991) for the
Following inhalation, most of the radon is same radon activity concentration, 0.75 mSv for a 40
exhaled, but some of it is absorbed in blood from the year old and 1.0 mSv for a 70 year old. In these calcula-
lungs whence it moves rapidly within the body. tions, they modeled the RBM as a mixture of spherical
Radon gas absorbed in pulmonary blood is distribu- fat cells distributed uniformly in the marrow and con-
ted in arterial blood to capillaries in tissues and sidered the variation of the fat content with age. The
organs and is then transferred from tissue to venous dose to the haematopoietic tissue from alpha particles
blood. The gas is again carried in the venous blood originating in the fat cells was calculated and the “self-
to pulmonary blood where some of it is exhaled, absorption” in the fat molecules, which reduces the
while the rest returns to arterial blood and the cycle dose, was taken into account. However, they assumed
continues. The extent to which radon is retained in a a 30% higher partition coefficient for fat than
tissue depends mainly on the relative solubility of Khursheed (2000), based on solubility measurements
radon in the tissue and the blood flow rates to the of human extracted fat samples (Nussbaum and
tissue. Tissues with a rich blood supply and a low to Hursh, 1958). These calculations were later revised fol-
moderate solubility of radon have retention half-times lowing analysis of fat fractions and sizes of fat cells in

33
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

the marrow cavity of adult ribs. The calculations gave “wrist watch” dosimeters (Eatough et al., 1999).
annual doses at 200 Bq m23 of 0.60–1.6 mSv to the However, this skin dose will only contribute a small
RBM (Allen et al., 1995). In contrast, a lower value of amount to the overall effective dose as the tissue
0.26 mSv was calculated for fatty marrow by Harley weighting factor for the skin is 0.01 (ICRP, 2007).
and Robbins (1992) as they assumed a lower partition Sevcova et al. (1978) reported that in Czech miners,
coefficient for fat. In all these calculations, the actual the excess basal cell carcinomas were on the forehead
spatial distribution between fat cells and haematopoie- and cheek.
tic tissue was not considered; a homogeneous mixture Skin deposition per unit activity concentration of
222
was assumed for simplicity, which would tend to over- Rn was measured to be significantly higher out-
estimate the dose. Harley and Robbins (1992) also cal- doors than for indoor exposure (Fews et al., 1999).
culated doses to soft tissue, fatty tissue, bone surfaces, This is likely because of the greater air movement
blood, and T lymphocytes. and also because of the possibility of wet deposition
from rainfall. The external equivalent skin dose to
the face was estimated to be 95 mSv yr21 for a con-
3.3.3 Skin Dose from Deposited Radon
tinuous outdoor exposure of 7 Bqm23 of 222Rn (Fews

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Progeny
et al., 1999).
Radon progeny in the ambient air can deposit on
surfaces including human skin. The alpha particles
3.3.4 Ingestion of Radon in Water
emitted will deliver a dose to the outer layers of the
skin in areas exposed to the atmosphere such as the Surface waters contain relatively low activity con-
neck and the face, whereas skin protected by cloth- centrations of dissolved radon, typically less than
ing and hair will in general receive a minimal dose. about 4000 Bq m23 (NA/NRC, 1999b). However, water
It has been assumed by many authors that the from ground water systems can have relatively high
basal cell layer of the epidermis is the target cell levels of dissolved radon, with activity concentrations
layer for the induction of skin cancer. In places, of 10 000 000 Bq m23 or greater (NA/NRC, 1999b).
where the skin is thin, such as the face, these target Because radon can easily be released by agitation
cells lie within the range of the alpha particles from in water, many uses of water release radon into the
the radon progeny on the skin surface. However, indoor air, which then contributes to the total indoor
Charles (2004) noted that there are existing animal airborne radon activity concentration and thus to
data that imply that the target cells are in the the inhalation pathway. Ingestion of water is also
underlying dermis, in which case they may lie too thought to pose a direct health risk through irradi-
deep to receive any significant dose from radon ation of sensitive cells in the gastrointestinal tract
progeny on skin surfaces. He concluded that cur- and in other organs once it is absorbed into the
rently there is no definitive answer to the location bloodstream. Thus, radon in drinking water can po-
and the identity of the target cells in the skin that tentially produce adverse health effects in addition
play a dominant role in the induction of skin cancer. to lung cancer.
A number of authors have calculated external doses Because of the relatively small volume of water
to the basal layer of the skin from radon progeny used in homes, the large volume of air into which
(Eatough and Henshaw, 1992; Harley and Robbins, radon is dispersed, and the exchange of indoor air
1992; Sevcova et al., 1978). The dose depends mainly with the ambient atmosphere, radon in water typical-
on the deposition velocity of the radon progeny, which ly adds only a small increment to the indoor air activ-
in turn depends on particle size and air movement. ity concentration. For example, a typical pattern of
Eatough and Henshaw (1992) estimated an average use of water containing radon at about 10 000 Bq m23
external skin dose of 25 mSv yr21 (range 17–170 mSv will on average increase the air radon activity concen-
yr21) for domestic exposure of 200 Bq m23 of 222Rn. tration by only about 1 Bq m23, based on a transfer
These values relate to the basal cell layer of the face coefficient from water to indoor air of about 1.0  1024
and neck with an assumed epidermal thickness of (NA/NRC, 1999b). Since there is always radon in
50 mm. The estimated range mainly reflects the un- indoor air from the emanation of soil gas into indoor
certainties in the deposition velocity of the radon air, only very high activity concentrations of radon in
progeny. Similar values were calculated by Harley water will make a significant contribution to the
and Robbins (1992) (10–200 mSv yr21). Based on ex- indoor airborne activity concentration. The transfer
perimental measurements of deposition rates of 218Po coefficient of radon in water to radon in air measured
and 214Po, Fews et al. (1999) estimated an external in a large bathroom in an energy-efficient home with a
skin dose for the face of about 100 mSv yr21 for an private well was 4.3  1024 (Harley et al., 2014).
indoor exposure of 200 Bq m23. This value was con- The stomach is the port of entry of ingested radon
sistent with the reported values found on personal into the body and thus is of particular concern for

34
 Radon and Radon Progeny Inhalation and Resultant Doses

risk assessment. Alpha particles emitted by radon The equivalent dose to the stomach wall was 8.4 
and its short-lived progeny within the contents of the 1028 Sv Bq21. The assumption that radon equili-
stomach cannot penetrate the mucus layer lining of brates between the stomach contents and the wall
the epithelium and therefore cannot reach the stem (i.e., saturated diffusion) was made.
cells at risk in the stomach wall. Thus, the dose to the For the ingestion pathway, NA/NRC (1999b) calcu-
wall depends heavily on the extent to which radon lated an age- and gender-averaged stomach cancer
diffuses from the contents into the wall. Once radon risk from lifetime ingestion of radon dissolved in
has entered the blood, through either the stomach or drinking water of 1.9  1029 for a radon activity con-
the small intestine, it is distributed among the centration of 1 Bq m23. Estimates for the inhalation
organs according to the blood flow and the relative of radon released from water yielded a lung cancer
solubility of radon as described in Section 3.3.2. risk due to lifetime exposure to radon in water at 1
Radon dissolved in the blood that enters the lung will Bq m23 of about 1.6  1028, based on a transfer coef-
readily be removed from the body by exhalation. ficient of 1  1024, which is about an order of magni-
Doses to the stomach estimated in different studies tude higher than the risk contributed by the
are summarized in Table 3.6 (NA/NRC, 1999b). ingestion pathway. Thus, most of the cancer risk

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

These estimates, exhibiting a wide range of values, posed by radon in drinking water arises from the
differ primarily by the extent to which diffusion into transfer of radon into indoor air and the subsequent
the stomach wall is considered. For example, Hursh inhalation of radon decay products, and not from the
et al. (1965) assumed that ingested radon diffuses ingestion of water.
through the stomach wall to blood at a rate such that
the activity concentration in the stomach wall is the
same as that in the stomach contents (i.e., saturated
3.4 Sensitive Target Cells in Bronchial
diffusion). By contrast, Harley and Robbins (1994)
Epithelium
assumed on the basis of the structure of the stomach
wall and the counter-current flow of fluid from the Sensitive target cells in bronchial epithelium are
stomach wall into the lumen that ingested radon defined as the cells receiving energy from ionizing ra-
cannot diffuse into the stomach wall with a sufficient diation that lead to the development of lung cancer.
depth to irradiate radiosensitive cells. However, NA/ Among the several cell types in bronchial tissue, basal
NRC (1999b) assumed on the basis of a radon diffu- cells have been selected as target cells by several
sion model that the time-integrated activity concen- authors since they are considered to be the progeni-
tration of radon at the depth of the stem cells in the tors of the ciliated and goblet cells in the epithelium
adult stomach wall is 30% of the activity concentra- and have a long lifetime (Ford and Terzaghi-Howe,
tion in the lumen. The resulting estimate of equiva- 1992; NCRP, 1984; Robbins et al., 1990). Secretory or
lent dose to the stomach wall was 2.4  1028 Sv Bq21 mucous cells have also been proposed as target cells
of 222Rn ingested. The corresponding effective dose for carcinogenesis (Johnson et al., 1990; McDowell
was 3.5  1029 Sv Bq21 of ingested radon. Assuming et al., 1985). Thus, it is current practice to identify
either no diffusion or saturated diffusion gave values both cell types as the relevant target cells for cancer
of effective dose of 2.1  1029 and 3.8  1028 Sv per induction, assuming that the cell nucleus is the most
Bq ingested radon, respectively (NA/NRC, 1999b). likely actual target region.
Khursheed (2000) calculated an effective dose of The locations of sensitive basal and secretory cell
1.0  1028 Sv Bq21 for ingestion of 222Rn of which nuclei within the bronchial epithelium, labeled as BB
97% is contributed by the dose to the stomach wall. in ICRP (1994), are illustrated in Figure 3.1. Radon
progeny deposited on the top of the mucus layer and
cleared from the initial deposition site by mucociliary
Table 3.6. Summary of estimates of equivalent dose coefficients to action are usually assumed to be uniformly mixed
the stomach per unit activity of 222Rn ingested (NA/NRC, 1999b) within the viscous mucus gel layer, forming a uni-
formly distributed alpha particle surface source.
Studies Equivalent dose coefficient Depending on the irradiation geometry (i.e., the dis-
(Sv Bq21)
tance between the alpha particle emission site and
Von Doebeln and Lindell (1965) 1.1  1027
the location of target cell nuclei), a certain fraction of
Hursh et al. (1965) 1.1  1027 emitted alpha particles will actually reach basal and
Suomela and Kahlos (1972) 1.3  1027 secretory cell nuclei. Since secretory cell nuclei are
Crawford-Brown (1989) 3.0  1027 closer to the epithelial surface than basal cell nuclei,
Brown and Hess (1992) 8.8  1028 with an assumed mean depth of 25 mm, secretory
Harley and Robbins (1994) 1.6  1029
Sharma et al. (1997) 8.2  1028
cells receive a higher dose than the basal cells with
an assumed mean depth 42 mm (NA/NRC, 1999a).

35
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 3.1. Schematic model of the location of sensitive target cell nuclei (basal and secretory cells) throughout the bronchial epithelium
(NA/NRC, 1991).

The cell depths assumed in NA/NRC (1999a) and Table 3.7. Depth (mm) of target cell nuclei for lung cancer
ICRP (1994) were based primarily on the study of induction below the epithelial surface (Robbins et al., 1990)
Mercer et al. (1991).
The bronchiolar region, labeled as bb in ICRP Population Basal cell nuclei (N)a Mucous cell nuclei (N)a
(1994), is characterized by a thinner epithelial tissue,
Male smokers 28 + 1.8 (23) 22 + 1.4 (28)
which contains only secretory cells as the sensitive Male 26 + 1.6 (10) 18 + 1.3 (10)
target cells (ICRP, 1994; NA/NRC, 1999a). non-smokers
Measurements made by Robbins et al. (1990) based Male ex-smokers 25 + 2.1 (15) 18 + 1.4 (15)
on 10 000 electron micrographs of bronchial tissue Female smokers 27 + 1.6 (28) 20 + 1.6 (29)
samples from over 100 persons show that the cell Female 27 + 1.5 (22) 17 + 1.0 (24)
non-smokers
depths are somewhat smaller than assumed by NA/ Female 30 + 2.0 (17) 20 + 1.3 (17)
NRC (1999a). Robbins et al. (1990) measured values of ex-smokers
19 and 27 mm for secretory and basal cell nuclei, re-
a
spectively. These tissue samples were dissected by Cell nuclei depth averaged over airway generation 3–6. The
airway generation so no ambiguity concerning specific nucleus depth is from the midpoint of the nucleus to the free
epithelial surface. N is the number of subjects and the
generations exists. These values are also smaller than
uncertainty term is the standard error of the mean.
the published values of Gastineau et al. (1972).
Baldwin et al. (1991) reported similar nuclear depths
as in the Robbins et al. (1990) study; however, the nuclei in the peripheral bronchiolar (airway genera-
airways were characterized by diameter which does not tions 9–15) epithelium, while basal cells are rarely
permit exact identification of airway generation as a found in the bronchioles (Mercer et al., 1991). In the
range of diameters exists for each generation. The data ICRP (1994) tissue model, it is implicitly assumed
of Robbins et al. (1990) based on a total number of 9954 that basal and secretory cell nuclei are uniformly
basal cell nuclei and 8958 mucous cell nuclei are shown distributed within the defined regions. However,
in Table 3.7. Mercer et al. (1991) found that cell nuclei were not
The ICRP (1994) model of target cell nuclei, i.e., uniformly distributed, but exhibited a distinct
secretory and basal cells, in the bronchial (airway maximum within the reported ranges.
generations 1-8) wall, based primarily on the histo- Comparison of Tables 3.7 and 3.8 reveals that the
logical measurements of Mercer et al. (1991), pro- secretory and particularly the basal cell nuclei
vides information on the ranges of nuclear depths depths reported by Robbins et al. (1990) are smaller
from the surface of the epithelium. Moreover, ICRP than the average nuclear depths proposed by ICRP
(1994) presents depth ranges for secretory cell (1994). Since bronchial doses decrease with

36
 Radon and Radon Progeny Inhalation and Resultant Doses

Table 3.8. Range of depths of secretory and basal cell nuclei in parameters can be determined on site by experimen-
bronchial (airway generations 1– 8) and bronchiolar (airway tal methods for given exposure conditions, personal
generations 9– 15) epithelium of the human lung (ICRP, 1994)
parameters are commonly based on average values
Range of depths in
obtained from anatomical and physiological studies.
epithelial tissue (mm) The most important personal factors affecting
lung dosimetry are (1) extrathoracic, i.e., nasal and
Secretory cells Basal cells oral, geometry, (2) anatomical structure and linear
airway dimensions of the lung, (3) physical activities
Bronchial 10– 40 (25)a 35 –50 (42.5)
and related breathing parameters, (4) bronchial
Bronchiolar 4 – 12 (8) n.a.
clearance velocities, (5) spatial distribution and
a
Calculated average values, assuming a uniform distribution frequency of sensitive target cells in bronchial epi-
within the epithelium. thelium, (6) human subject age, and (7) smoking
status.
Individual variations of the structure of the nasal
increasing depth in bronchial epithelium, cell nuclei
and oral passages in volunteers have been reported

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

located at different depths within the reported
by Cheng et al. (1996), revealing significant fluctua-
ranges will receive different doses. For example, the
tions of the nasal cross-section and the shape of the
dose at 26 mm depth is about a factor of 2 higher
nasal passages. Indeed, intersubject variations of
than the dose at a depth of 42 mm. This highlights
measured deposition fractions could be attributed to
the significance of the target cell depth in bronchial
corresponding fluctuations of these two anatomical
dose modeling that attempts to account for alpha
parameters, thereby affecting the fraction of inhaled
particle traversal of cell nuclei from radon decay pro-
particles entering the lung.
ducts on the epithelial surface. Hence improved pre-
Since lung volumes can be correlated with body
cision in bronchial dose modeling is possible with
weight and height (ICRP, 1994), airway dimensions
the use of more comprehensive data sets, i.e., global
exhibit significant intersubject variations (Hofmann
estimates of biological parameters.
et al., 2002), which affect deposition of inhaled radon
Regarding the depth distribution of target cell
progeny and, in consequence, bronchial doses. In add-
nuclei among the bronchial and bronchiolar airways,
ition to this volumetric variability, also structural vari-
Baldwin (1994) reported a decrease in basal cell
ability in terms of number of airways was observed
depths with penetration into the airway system, con-
(Hofmann et al., 2002).
sistent with the corresponding reduction in airway
Physical activities and related breathing para-
diameters. Likewise, Mercer et al. (1991) measured
meters, such as breathing frequency and tidal volume,
the depth distributions of basal and secretory cells,
determine the amount of inhaled activity per unit
among other cell types, for large and small bronchi,
time as well as deposition fractions in a single breath,
bronchioles, and terminal bronchioles, showing a de-
depending on the velocity of the airflow through the
crease from trachea to terminal bronchioles for both
bronchial passages (see Section 3.6.1). Since daily ac-
cell types. Based on these measurements, Hofmann
tivity patterns may be quite different among workers
et al. (1996) defined the average depths of secretory,
or members of the population at large, differences in
Ds, and basal, Db, cell nuclei as functions of the bron-
breathing parameters may lead to a wide range of
chial airway diameter, d:
bronchial doses for identical exposure conditions.
Ds ¼ 173:14 d2 þ 20:494 d þ 3:577 Mucociliary clearance velocities and transit times
ð3:1Þ in bronchial airways are related to airway diameters
ðmaximum depth ¼ 25 mmÞ and lengths and therefore reflect their intersubject
variations (Asgharian et al., 2001). Indeed, Yeates
and
et al. (1975) found significant intersubject variations
Db ¼ 90:38 d2 þ 131:41 d  3:809 of tracheal mucus velocities among a group of adult
ð3:2Þ volunteers. Since bronchial mucus clearance veloci-
ðmaximum depth ¼ 44 mmÞ ties determine the distribution of nuclide-specific
surface activities among bronchial airways, inter-
subject variations of bronchial clearance parameters
3.5 Personal and Environmental Parameters
will affect local doses.
Affecting Lung Dosimetry
The thickness of the bronchial epithelium and result-
Absorbed doses in sensitive target cells in bron- ing depths of sensitive target cells are related to airway
chial epithelium depend on several personal and en- diameters and thus exhibit corresponding intersubject
vironmental factors, which will be discussed in more variations (Mercer et al., 1991). Since doses to target
detail in subsequent sections. While environmental cell nuclei are a function of their distance from the

37
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

airway surface, different target cell depths lead to dif- operating physical deposition mechanisms are func-
ferent bronchial doses (see Section 3.4). tions of the particle diameter. By the same token,
Subject age affects not only the anatomical struc- the relative frequencies of the attached and un-
ture of the human lung but also breathing para- attached fractions determine the total fraction of
meters (ICRP, 1994). While these changes are most deposited radon progeny and hence resulting bron-
apparent in the developing lung (Ménache et al., chial doses (see Section 3.8).
2008) (Section 3.6.2), they have also been observed Deposition fractions of inhaled radon progeny
in an aging population, where a less efficient bron- depend exclusively on aerosol properties and breath-
chial clearance may further enhance the effect of ing parameters and not on their activity concentra-
age on lung dosimetry. tions. Thus, for defined aerosol parameters and
Epidemiological studies have demonstrated that inhalation conditions, ambient radon progeny con-
bronchial tumors occur preferentially in smokers. centrations affect bronchial doses in a proportional
Thus, the smoking status is another personal factor fashion, providing a simple relationship between
to be considered in lung dosimetry in addition to the bronchial dose and activity concentration.
well-documented carcinogenic effect of cigarette

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

smoke. Exposure to inhaled cigarette smoke can in-
crease the thickness of the bronchial epithelium, 3.6 Dependence of Doses on Physical
thereby reducing bronchial dose, but can also slow Activities (Breathing Parameters) and Age
down mucus clearance velocities, change breathing
3.6.1 Dependence on Physical Activities
parameters and the histology of the bronchial epi-
thelium, thereby increasing bronchial doses (Baias Radiation doses to cells and tissues of the human
et al., 2010) (Section 3.9.4). respiratory tract are determined to a large extent by
Albert and Lippmann (1971) and Albert et al. breathing parameters, such as the tidal volume, i.e.,
(1973), based on studies with inhaled radioactively the air volume inhaled in a single breath, and the
labeled particles, showed that clearance in normal breathing frequency, i.e., the number of breaths per
subjects consisted of two discrete phases of bronchial unit time, and the ventilation rate as the product of
clearance. The first was completed within 1–2 h and both parameters. Both tidal volume and breathing
the second within 4 – 10 h. The 90% clearance time, frequency are functions of the physical activity. The
i.e., the time after which 90% of the inhaled particles equivalent physical analogue normally used in de-
were cleared from the lungs, were the same for both position calculations is the flow rate, i.e., the tidal
smokers and non-smokers. For comparison, Sanchis volume inhaled during the inspiration time (m3 s21).
Aldas et al. (1971) measured an overall faster clear- For radiation protection purposes, ICRP (1994)
ance of inhaled particles from the lung in smokers. has defined four typical physical activities, i.e.,
They stated, however, that the clearance rate in the resting (sleeping), sitting awake, light exercise, and
large proximal airways was slower than in non- heavy exercise, and has assigned typical parameter
smokers (T1/2 ¼ 2.3 h in smokers; T1/2 ¼ 42 min in values for tidal volume and breathing frequency for
non-smokers), but faster in more peripheral airways. each physical activity and for seven different age
Several types of clearance abnormalities were and gender classes (Table 3.9) (Note: The values for
observed in some subjects who smoked cigarettes or ages 3 months and 1 year were not included into this
who had demonstrable lung disease. These included: table because the dependence on physical activity is
(1) an extended delay in the onset of clearance or not very meaningful in small infants as they cannot
between clearance phases, (2) a spasmodic type of control their activities).
clearance with intermittent tracheal blockage, and (3) The effect of physical activity and related breathing
an extended period of clearance arrest with retro- parameters on bronchial doses reveals two specific fea-
grade movement of particles from the hilar region to tures which act in an antagonistic fashion: The higher
more distal lower lung regions (Albert and Lippmann, ventilation rate at higher physical activities relative to
1971). resting increases the inhaled amount of radon progeny
The most important environmental parameters in a single breath, thereby increasing the resulting
affecting lung dosimetry are (1) attached and un- doses. On the other hand, a higher flow rate at
attached size distributions, (2) attached and unattached enhanced physical activities decreases the diffusion
fractions, and (3) radon progeny activity concentrations deposition efficiency of submicron radon progeny by
and their nuclide-specific distributions. decreasing the residence time in an airway, and thus
The size distributions of inhaled attached and un- decreasing the resulting doses (note: although depos-
attached radon progeny determine their deposition ition of larger particles by inertial impaction would
efficiencies in all airways of the human respiratory increase with rising flow rate, Brownian motion domi-
tract, from the nose to the alveolar airways, as all nates the deposition behavior of radon progeny).

38
 Radon and Radon Progeny Inhalation and Resultant Doses

Table 3.9. Reference respiratory values for a general Caucasian population at different levels of physical activity: resting (sleeping),
sitting awake, light exercise, and heavy exercise. Breathing parameters are: Tidal volume VT (in l), breathing frequency f (in min21), and
ventilation rate B (in m3 h21) as the product of tidal volume and breathing frequency (ICRP, 1994)

Age/gender Resting (sleeping) Sitting awake Light exercise Heavy exercise

VT f B VT f B VT f B VT f B

5 year 0.174 23 0.24 0.213 25 0.32 0.244 39 0.57 N/A N/A N/A
10 year
Male 0.304 17 0.31 0.333 19 0.38 0.583 32 1.12 0.841 44 2.22
Female 0.667 46 1.84
15 year
Male 0.500 14 0.42 0.533 15 0.48 1.0 23 1.38 1.352 36 2.92
Female 0.417 14 0.35 0.417 16 0.40 0.903 24 1.30 1.127 38 2.57
Adult
Male 0.625 12 0.45 0.750 12 0.54 1.25 20 1.50 1.923 26 3.00
Female 0.444 12 0.32 0.464 14 0.39 0.992 21 1.25 1.364 33 2.70

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

N/A, not applicable.

Bronchial target cell doses for different physical ac- Table 3.10. Effective dose conversion coefficients (mSv WLM21)
tivity levels compiled in Table 3.10 were computed with for different levels of physical activity under typical indoor
the stochastic lung dosimetry model IDEAL-DOSE exposure conditions
(Hofmann and Winkler-Heil, 2011; Hofmann et al.,
Age/ Resting Sitting Light Heavy
2010). The comparison of doses produced by typical
gender (sleeping) awake exercise exercise
uranium mine exposure conditions (Winkler-Heil et al.,
2007) has indicated that the IDEAL-DOSE code pre- 5 year 8.5 12.4 27.8 N/A
dicts slightly lower dose values than the IMBA code, 10 year
based on the ICRP Human Respiratory Tract Model Male 6.0 7.6 31.7 80.6
(ICRP, 1994). Aerosol size distributions and unattached Female 6.0 7.6 31.7 63.2
15 year
fractions used for the physical activity calculations Male 4.8 5.7 21.7 57.7
refer to typical indoor exposure conditions without Female 4.5 5.1 23.5 58.2
smoking (Marsh et al., 2005) (Note: smoking affects the Adult
activity size distribution of the attached radon progeny Male 4.4 5.5 18.7 46.2
as well as the unattached fraction). Female 3.6 4.3 19.2 52.1
The results listed in Table 3.10 exhibit a consist-
N/A, not applicable.
ent trend: For all ages, doses increase with rising
physical activity from resting to heavy exercise by
about one order of magnitude. This further indicates childhood to adulthood have been reported. Phalen
that the effect of increased ventilation rates clearly et al. (1985) made limited measurements of about 25
surpasses the effect of reduced deposition efficien- airways per cast provided by Mortensen et al. (1983)
cies. However, with increasing physical exercise, and used these measurements to develop their model
most humans switch from nasal breathing to partly of conducting airway dimensions as a function of age.
oral breathing (ICRP, 1994), which slightly reduces Using equations developed in Phalen et al. (1985) for
the filtration efficiency of the extrathoracic region. airway length and diameter as a function of height,
Furthermore, the strong dependence of lung doses Phalen and Oldham (2001) examined the deposition
on breathing rate emphasizes the importance of report- of particles in the bronchial and pulmonary regions
ing the physical activity related to a given radon meas- of the lung for different ages.
urement site. Thus, information on the radon or radon Mortensen et al. (1983; 1989) prepared silicon
progeny activity concentration without reporting the rubber casts of the lungs of a large number of
associated physical activity pattern is insufficient for children and made complete measurements of the
the correct assessment of individual lung doses. conducting airways. Although limited by the lack
of any airway measurements distal to the 10th
generation, this database is unique in having com-
3.6.2 Dependence on Age
plete length, diameter, and branching angle infor-
Only a few measurements of the anatomical dimen- mation on approximately 1000 airways in the first
sions for the normal developing human lung through 10 generations.

39
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Ménache et al. (2008) developed single-path whole- Based on the ICRP Human Respirator Tract
lung models and lobar models of the lungs of 11 chil- Model (ICRP, 1994), Marsh et al. (2005) calculated
dren between 3 months and 21 years of age based on dose-exposure conversion coefficients for typical
a combination of the Mortensen et al. (1983; 1989) home exposure conditions (Marsh et al., 2002) for
data and published information on distal airway ages ranging from 3 months to adulthood. The
dimensions. As of today, these morphometric models reported effective dose coefficients refer to two
represent the most detailed airway geometries for indoor exposure conditions: (i) typical home without
particle deposition modeling purposes. smokers, and (ii) typical home with smokers. Values
Since the growth of tracheobronchial and pulmon- for the house with smokers were calculated assum-
ary regions of the human lung has not been systemat- ing that subjects were in a smoke-filled living room,
ically measured, Hofmann (1982a; 1982b) pursued a except while they were sleeping in a smoke-free
different modeling approach: he assembled many bedroom.
measurements of the upper airways from different For the dose calculations presented in Table 3.11
authors, extrapolated the systematic features identi- (Marsh et al., 2005), the following parameters were
fied there to the lower airways, and incorporated them assumed to be age-dependent:

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

together with limited data on the pulmonary airway
into a mathematical description of the growing lung (1) Mass of the target tissue.
(Hofmann et al., 1989a, 1989b; Martonen et al., 1989). (2) Respiratory frequency and tidal volume for each
This age-dependent model is based on the structure of level of physical activity (see Table 3.9).
Weibel’s (1963) Model A, which describes the tracheo- (3) Average time spent at each level of exercise.
bronchial tree by 17 symmetrically dividing genera- (4) Lung capacities, such as functional residual cap-
tions. For comparison, the bronchial airway model acity (FRC), regional dead spaces, and diameters
published by Phalen et al. (1985) is based on the rela- of the trachea and of generations 9 and 16.
tively few complete bronchial pathway measurements
made in a single lobe and the findings extrapolated to
all other lobes, based on the bronchial airway struc- The effect of smoking on the exposure conditions
ture of the Yeh and Schum (1980) lung model. and resulting lung doses in homes should be consid-
While the above airway generation models are ered in epidemiological studies, involving non-
partly based on measured morphometric data, par- smoking members of the family living in the same
ticle deposition for different ages in the ICRP Human house as smoking members.
Respiratory Tract Model (ICRP, 1994) is modeled by Harley (1984) calculated the bronchial dose in
applying age-specific modifying factors in the alge- mines for the adult male, and in residences for the
braic deposition efficiency equations for each lung adult male, female, 10, and 1 year old. This calculation
compartment. These modifying factors were specified used Yeh and Schum (1980) lung dimensions with
for the following ages: 3 months, 1, 5, 10 and 15 years modifying (scaling) factors of 1.0, 0.76, 0.5, and 0.33,
(male and female), and adult (male and female) respectively. Calculated dose conversion coefficients in
The effect of morphometric and physiological
changes of the lung on doses to tracheobronchial
and alveolar compartments was first described by Table 3.11. Effective dose-exposure conversion coefficients (mSv
WLM21) for different ages under typical home exposure
Hofmann et al. (1979). This model was subsequently conditions, differentiating between the effects of unattached and
refined by replacing the compartmental system attached radon progeny (Marsh et al., 2005)
of the lung by the airway generation structure
of Weibel’s Model A (1963) and bronchial genera- Effective dose conversion coefficient (mSv WLM21)
tion doses to bronchial basal cells were computed for
Age/gender Home without smokers Home with smokers
typical age-specific physical activity patterns
(Hofmann, 1982a; Hofmann et al., 1989a). Unatt. Att. Total Unatt. Att. Total
Height and weight can be used to scale physiologic-
al, respiratory, and morphometric lung parameters by 3 months 2.6 7.0 9.6 1.7 6.5 8.2
allometric equations (ICRP, 1994). For example, allo- 1 year 3.4 7.9 11.3 2.4 7.4 9.8
5 year 3.7 7.6 11.3 2.3 7.1 9.4
metric relationships have been published for the total 10 year 4.9 8.6 13.5 2.4 8.0 10.4
lung capacity (TLC) as function of height and age, 15 year
and tidal volume (VT) as function of weight (ICRP, Male 4.1 7.3 11.4 2.3 6.7 9.0
1994). Examples of variations of respiratory values Female 4.1 7.4 11.5 2.0 6.8 8.8
for ethnic groups, such as Japanese, Chinese, Indian, Adult
Male 4.8 8.1 12.9 2.4 7.5 9.9
African American, Senegalese, and Zimbabwian, are Female 4.7 8.0 12.7 2.0 7.2 9.2
also listed in the ICRP (1994) report.

40
 Radon and Radon Progeny Inhalation and Resultant Doses

residences were 8, 10, and 18 mSv WLM21 for daily material specific. The model assumes that the rate
active and resting breathing cycles of 16 and 8 h for of absorption is the same in all regions of the re-
males, females, and the 10 year old. The calculated spiratory tract except in the anterior nose (region
bronchial dose for the 1 year old was 14 mSv WLM21 ET1) where none occurs. The HRTM treats absorp-
assuming a constant daily breathing rate. tion to blood as a two-stage process:
These dose calculations reveal that doses to sensi-
tive bronchial and bronchiolar cells do not vary appre- † the dissociation of the particles into a material
ciably with age, exhibiting a slight maximum at the that can be absorbed into blood (i.e., dissolution),
age of about 10 years. For comparison, previous calcu- and
lations with an airway generation model revealed a † the uptake of material dissolved from particles, or
dose maximum at the age of about 4 years (Hofmann, material deposited in a soluble form.
1982a). This difference is primarily caused by the def-
inition of age-specific physical activity patterns, To represent time-dependent dissolution, it is
which affect the age-dependency of dose calculations assumed that a fraction (fr) dissolves rapidly at a rate
(Hofmann et al., 1989b). sr while the remaining fraction (1 2 fr) dissolves more

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The age-dependency of bronchial doses for defined slowly at a rate ss. Uptake is usually assumed to be
average age-specific physical activities can also be instantaneous, but for some elements, a fraction of
seen in Table 3.10, based on computations with the the dissolved material is absorbed more slowly as a
stochastic IDEAL-DOSE model. There, doses de- result of binding to the respiratory tract components.
crease monotonously with rising age for each physic- To represent time-dependent uptake, a fraction, fb, of
al activity, indicating that children might receive the dissolved material is assumed to be retained in a
higher doses than adults when exposed to the same bound state, from which it is transferred into blood at
exposure characteristics. This emphasizes the im- a rate sb, while the remaining fraction (1 2 fb) trans-
portance of radon measurements in schools. Note fers to blood instantaneously. Because the bound state
that the relatively weak dependence on age shown is considered to represent the interaction of an element
in Table 3.11, as opposed to the distinctly decreasing in dissociated (ionic) form with cells forming the lining
relationship with age illustrated in Table 3.10, is of the respiratory tract, the bound fraction fb and
caused by the application of typical age-specific uptake rate sb are assumed to be element-specific.
daily physical activity patterns, which are charac- Bound material is not subject to particle transport.
terized by higher physical activities in adults when In the late 1960s experiments in which volunteers
compared with children. inhaled 212Pb ions, unattached or attached, to con-
In conclusion, the age-dependency of lung doses densation nuclei gave overall absorption half-times
for defined physical activities predicted by different from lungs to blood of about 10 h (Booker et al.,
dosimetric models depends on assumptions regard- 1969; Hursh and Mercer, 1970; Hursh et al., 1969;
ing lung morphology, breathing parameters, muco- Marsh and Birchall, 1999a; 1999b). For simplicity
ciliary clearance velocities, and target cell depths and for dosimetry purposes, calculations were typic-
for the different ages, as illustrated by the applica- ally performed assuming that the 222Rn progeny
tion of two different dosimetry models in Tables 3.10 (218Po, 214Pb, and 214Bi) are absorbed with a half-
(IDEAL-DOSE model) and 3.11 (ICRP model). time of 10 h. Sensitivity analysis showed that the
Thus, documented variations in the morphology of lung dose is sensitive to the absorption rates of the
the developing lung and related respiratory para- radon progeny if the assumed absorption half-times
meters introduce some uncertainties in radon are less or comparable with their radioactive half-
progeny lung dosimetry, which may be comparable lives (Marsh and Birchall, 2000; Zock et al., 1996).
with the inherent biological intersubject variability For example, it has been shown that the equivalent
(see Section 3.9.2). dose to the lung is reduced by more than 10% if the
absorption half-time is less than 8 min for 218Po or
less than 2 h for 214Pb or 214Bi compared with the as-
sumption of a 10 h absorption half-time (Marsh and
3.7 Dependence on 222Rn Progeny
Birchall, 2000).
Absorption Parameters
Marsh and Bailey (2013) reviewed animal and vol-
Particles deposited in the respiratory tract are unteer data to determine lung-to-blood absorption
cleared by two competitive processes: absorption to rates for radon progeny. As stated above early experi-
blood and particle transport to the alimentary tract ments in which volunteers inhaled 212Pb ions, gave
and lymphatics. In the HRTM (ICRP, 1994), it is overall absorption half-times from lungs to blood of
assumed that particle transport rates are the same about 10 h. Animal experiments designed to investi-
for all materials, whereas absorption into blood is gate the clearance kinetics of lead ions in detail

41
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

showed two phases of absorption: about 10% is Table 3.12. Values of the absorption parameters of the HRTM for
absorbed with a half-time of about 10–15 min, the inhaled radon progeny proposed by the ICRP Task Group on
Internal Dosimetry (INDOS)
rest with a half-time of about 10 h. Most studies in
which the volunteers inhaled unattached 212Pb or Inhaled radon Dissolution parameter Uptake
212
Pb attached to condensation nuclei suggest that if progeny values parameter
there is a rapid absorption component then this is values
likely to be only 5%. However, the volunteer data of
fr sr (d21) ss (d21) fb sb (d21)
Butterweck et al. (2002) suggest that unattached lead
is more soluble; about 30% is rapidly absorbed to blood
Polonium 1 3 — 0 —
with a faster rate (sr . 100 d21). Lead 0.1 100 1.7 0.5 1.7
Experimental evidence was found supporting the Bismuth 1 1 — 0 —
existence of a bound fraction for lead. Lead ions
deposited on nasal or bronchial epithelium of rats Rapid dissolution fraction fr, rapid dissolution rate, sr, slow
and rabbits cleared more slowly than insoluble par- dissolution rat, ss, bound fraction, fb.
ticles deposited simultaneously. Similar half-times

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

associated with slow uptake of lead in different ionic the analysis of the volunteer data of Hursh et al.
forms (e.g., nitrate, hydroxide, chloride) suggest that (1969) (Marsh and Birchall, 1999a; 1999b). In this
this is associated with binding (a characteristic of study, volunteers were exposed to an aerosol of 212Pb
the element) rather than dissolution (a characteris- and 212Bi attached to “natural” particles in room air
tic of the chemical form of the element). Estimated and the amount of 212Bi excreted in urine was
values of bound fraction from animal and volunteer reported for one of the volunteers.
studies ranged from 0.2 to 0.8 (Marsh and Bailey, The equivalent dose to lung and the effective dose
2013). calculated with the absorption parameter values
The volunteer studies of Booker et al. (1969) indi- given in Table 3.12 are only a few percent greater
cated that unattached and attached 212Pb have similar than the value obtained assuming a single absorp-
absorption characteristics. This suggests that the tion half-time of 10 h with no binding (Marsh and
attached 212Pb rapidly separates from its host follow- Bailey, 2013).
ing deposition in the respiratory tract. Rapid separ-
ation could occur because of alpha recoil and/or
physiochemical interactions with the lung fluid. 3.8 Dependence on Radon Progeny-Related
Booker et al. (1969) noted that “The attachment of Aerosol Parameters
212
Pb to nuclei seems to be irreversible in air, but auto-
3.8.1 Radon Progeny Aerosol Parameters
radiographs have shown (Heard, 1968) that the activ-
ity is largely desorbed in aqueous media” (Heard, 1968 The size distribution of an inhaled aerosol is one
personal communication). Based on these data and of the factors that determine the fraction of the
other experimental evidence, Marsh and Bailey (2013) intake that is deposited in each region of the respira-
assumed that the unattached and attached radon tory tract. Deposition depends on the particle’s size
progeny have the same absorption characteristics. as well as the subject’s breathing pattern and the
In the forthcoming Occupational Intakes of geometry of the respiratory tract. A further consid-
Radionuclide (OIR) document of the ICRP, for the eration is that some of the ambient aerosols, to
purposes of dosimetry, specific absorption parameter which radon progeny attach, are unstable in satu-
values will be given for polonium, lead, and bismuth rated air and grow very quickly on inhalation due to
inhaled as a decay product of radon. In the prepar- the high humidity in the respiratory tract (NA/NRC,
ation of this document, the ICRP Task Group on 1991; Sinclair et al., 1974). For simplicity, this is
Internal Dosimetry (INDOS) has proposed absorp- generally modeled by assuming that the attached
tion parameter values for radon progeny, which are aerosol instantaneously grows by a given factor as
given in Table 3.12. The parameter values chosen the aerosol enters the nose or the mouth, while the
for lead as a decay product of radon are consistent size of the unattached progeny is assumed to remain
with the values suggested by Marsh and Bailey constant in the respiratory tract (NA/NRC, 1991).
(2013) based on their review. Neglecting particle The equivalent dose to the lung arising from the
transport, about 5% of the deposit [i.e., fr(1 2 fb)] is inhalation of short-lived radon progeny is directly
absorbed to blood rapidly with a half-time of 10 min, proportional to the amount deposited in the bron-
while the remainder is absorbed with a half-time of chial (BB) and bronchiolar (bb) regions of the lung.
about 10 h. The fraction of the dissolved material, fb However, despite the higher deposition fractions of
which is assumed to be bound is 0.5. The absorption attached radon progeny in alveolar airways, the
parameter values chosen for bismuth were based on dose to the BB and bb regions is much greater than

42
 Radon and Radon Progeny Inhalation and Resultant Doses

that of the AI region because the mass of the target deposition

p ffiffiffiffiffiffiffiffiffiffiffiffiffiffi by diffusion increases with the quotient
tissue in the BB and bb region are small (a few tR =dth (NA/NRC, 1991).
grams) compared with the mass of the alveolar– Diffusion batteries measure dth, whereas low-
interstitial (AI) region (1 kg) (ICRP, 1994). pressure cascade impactors measure dae (see Section
To understand how the deposition in BB and bb 5.3.4.2). The two diameters, dth and dae, are related via
depends on the aerosol parameters, it is necessary the particle density (r) and the shapepffiffiffiffiffiffiffiffifactor (x). For a
to describe the mechanisms of aerosol deposition in first approximation, dth ¼ dae x=r. However, for
the respiratory tract. There are three major mechan- small particles (dth, 1 mm) whose size approaches the
isms, namely gravitational sedimentation, inertial mean free path of air molecules, the viscous drag is
impaction, and diffusion. reduced and therefore the settling velocity increases.
In gravitational sedimentation, the particle To take account of this, the slip correction (often called
falling under gravity experiences a viscous resistive the Cunningham factor) needs to be applied to the
force of the air, which increases as the particle accel- above equation (Hinds, 1982; ICRP, 1994; Willeke,
erates. As a result, the particle reaches a constant 1976) as follows:
velocity (i.e., the settling velocity) when the viscous

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

force of the air is equal and opposite in direction to sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
the gravitational force. The magnitude of the set- x Cðdae Þ
dth ¼ dae ð3:3Þ
tling velocity depends on the physical size of the par- r Cðdth Þ
ticle, its shape, and density. This settling can lead to
deposition on respiratory surfaces and the probabil-
ity of deposition increases with tR d2ae , where tR is where C(dae) is the slip correction for a particle of aero-
the residence time and dae is the aerodynamic diam- dynamic diameter, dae, and C(dth) is the slip correction
eter (NA/NRC, 1991). The aerodynamic diameter is for a particle of thermodynamic diameter, dth. Note:
defined as the diameter of a unit density sphere that examples of size distributions are given in Section 4.7.
has the same terminal settling velocity in air as the Porstendörfer (2001) and his co-workers mainly
particle of interest. Generally, gravitational sedi- carried out activity size measurements of the
mentation is important for particles with dae greater attached progeny with a low-pressure cascade im-
than about 0.5 mm. pactor. However, some measurements in closed
When an airstream carrying a particle is forced to rooms without additional aerosol sources were
change direction because of an obstacle, the inertia carried out with both a low-pressure cascade impact-
or momentum of the particle resists the change of or and a diffusion battery, which measures the
direction. If the momentum is high enough, the par- thermodynamic diameter (Reineking et al., 1988;
ticle will continue in its original direction and 1992a). Comparisons of these measurement results
deposit on the obstacle. High linear velocities and show that the value of the activity median thermo-
abrupt changes in the direction of airflow occur in dynamic diameter (AMTD) was similar to that of the
the nasal passages, pharynx, and at central airway activity median diameter (AMD) measured with the
bifurcations. The probability of deposition by iner- low-pressure cascade impactor with the differences
tial impaction increases with the product of d2ae and being less than about 10%. Such differences are
the respired flow rate. Generally, inertial impaction small compared with the uncertainties of the data
is important for particles with dae greater than evaluation procedure and the normal variation in
about 2 mm (NA/NRC, 1991). the size distribution under realistic working or
Diffusion (i.e., Brownian motion) is the random living conditions. For indoor measurements, it can
motion of an aerosol particle caused by collisions with therefore be assumed that the values of AMD deter-
gas molecules. This motion can lead to contact with mined with low-pressure cascade impactors are good
and deposition on respiratory surfaces. Diffusion is approximations to the corresponding values of
the dominant mechanism of deposition in the airways AMTD.
for small particles of less than 0.5 mm and is therefore Some authors have used Equation (3.3) to calculate
important for radon progeny aerosols. Unlike gravita- the corresponding values of dth from the dae of the
tional sedimentation and inertial impaction, diffusion attached progeny measured with low-pressure
is independent of particle density. This means that cascade impactors. For such purposes, values of 1.4 g
dae is not a direct measure of diffusion and therefore cm23 and 1.1 were assumed for r and x, respectively,
for small particles, the size is best expressed in terms under normal conditions in homes and buildings
of the thermodynamic diameter. The thermodynamic (Marsh et al., 2002; Reineking et al., 1988). The
diameter, dth, is defined as the diameter of a spherical density and shape factor of the unattached progeny
particle that has the same diffusion coefficient in air are generally assumed to be unity for dosimetry pur-
as the particle of interest. The probability of poses.

43
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

To take account of the hygroscopic growth of

attached progeny in the respiratory tract, it is gener-
ally assumed that dth increases by a factor between
1.0 and 2.0 as the particle enters the nose and
mouth (James et al., 2004; Marsh et al., 2002; 2005;
NA/NRC, 1991; Porstendörfer and Reineking, 1999).
Sinclair et al. (1974) found that atmospheric parti-
cles in their laboratory increased in diameter by
about a factor of 2 when the relative humidity
increases from 0 to 98%. The ambient aerosol origi-
nated from an industrial area close to the sea and
the authors expected it to consist of a mixture of
NaCl and (NH4)2SO4 salts with a mixture of acids
(HNO3, H2SO4, and HCl). Based on this work, a Figure 3.2. Regional deposition as a function of particle size
panel of experts from the National Research Council calculated with the HRTM for a standard worker: blue continuous

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

assumed for dosimetry purposes a hygroscopic line, bronchial (BB) region; continuous line, bronchiolar (bb)
region; dashed line, 1/2 ET equals half of the deposition in
growth factor of 2 and that the density and the
the extrathoracic (ET) region, approximating deposition during
shape factor of these hygroscopically enlarged parti- the inhalation phase (note: filtration in the ET region affects the
cles were both unity (NA/NRC, 1991). Li and Hopke fraction of particles entering the lung). Unit density and unit
(1993) measured hygroscopic growth factors of shape factor were assumed.
indoor combustion aerosols including cigarette
smoke, incense smoke, candle flame, natural gas
is the dominant deposition mechanism in this range.
flame, and propane fuel flame. The average growth
Initially, deposition in BB and bb increases between
factors ranged from 1.5 to 1.9. In contrast, radon
0.6 nm and about 4 nm as more of the aerosol passes
progeny attached to diesel particles or to an aerosol
through ET. Deposition in BB and bb then decreases
produced from cooking oil, for example, are hydro-
with particle sizes for dth , 400 nm as less is depos-
phobic (Dua and Hopke, 1996; Dua et al., 1999;
ited by diffusion. As can be seen from Figure 3.2, the
Weingartner et al., 1997).
amount deposited in BB and bb is most sensitive to
Computed growth curves of NaCl particles reveal
particle diameters in the unattached range (0.3 –
that hygroscopic particles below about 500 nm
5 nm) and in the nucleation range (10 –100 nm). In
reach their equilibrium diameter already within the
comparison, it is less sensitive to particle diameters
mouth/nose and the large bronchial airways (Ferron
in the accumulation range (100 –500 nm). Note:
et al., 1988; Winkler-Heil et al., 2014). This suggests
examples of activity size distributions are given in
that the hygroscopic growth of unattached and
Section 4.7.
attached radon progeny can be modeled by a con-
The dependence of the effective dose per WLM on
stant equilibrium growth factor throughout the
particle size is displayed in Figure 3.3, calculated
whole respiratory tract. In other words, deposition
with the HRTM for a standard worker. As expected,
of hygroscopic radon progeny can be calculated for
the equilibrium diameter, thereby greatly simplify-
ing deposition calculations for hygroscopic particles.
For illustrative purposes, Figure 3.2 shows the re-
gional deposition, i.e., the fraction of inhaled parti-
cles deposited in a given region, as a function of
particle diameter calculated with the HRTM (ICRP,
1994). Unit density and shape factor was assumed
for simplicity. To model deposition in the HRTM
semi-empirical equations based on experimental de-
position data were used. However, a theoretical
model of gas transport and particle deposition was
used to calculate the fractional deposition in BB, bb,
and AI regions of the lung. This was evaluated by
considering aerodynamic (gravitational sedimenta-
tion, inertial impaction) and thermodynamic (diffu-
sion) processes acting competitively. The deposition Figure 3.3. Effective dose per WLM as a function of particle size
in the extrathoracic (ET) region decreases with par- calculated with the HRTM for a standard worker. Unit density
ticle size in the range 0.6 – 100 nm because diffusion and unit shape factor was assumed.

44
 Radon and Radon Progeny Inhalation and Resultant Doses

it has a similar shape to Figure 3.2, which gives the fraction (fp) was approximated by:
fractional deposition in the central airways (BB and
bb regions) as a function of particle size. Effective dose per WLM ¼ 11:35 þ 43 fp mSv WLM1
ð3:4Þ
3.8.2 Sensitivity of Lung Dose from
Inhalation of 222Rn Progeny to Aerosol For fp ranging from 0.04 to 0.2, the effective dose
Parameter Values per WLM varies from 13 to 20 mSv WLM21 [3.7 –5.6
Marsh and Birchall (2000) performed a sensitivity mSv per (mJ h m23)].
analysis with the HRTM to identify those aerosol Equation (3.4) can also be expressed in terms of
parameters that have most influence on lung dose dose per unit radon (222Rn) gas exposure:
arising from the inhalation of 222Rn progeny under
the conditions found in houses. In this analysis, it Effective dose per ðBq h m3 Þ
was assumed that the size distribution of the  
¼ 1:57  106 F 11:35 þ 43 fp mSv ðBq h m3 Þ1
attached progeny had been measured with a low-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

pressure cascade impactor and was therefore given ð3:5Þ
in terms of the aerodynamic diameter. The aerosol
parameters that most affected the equivalent dose to Equation (3.5) follows from 3.4 as 1 WLM approxi-
the lung per WLM were the unattached fraction, the mately equals (6.37  105/F) Bq h m23 of 222Rn gas
activity median aerodynamic diameter (AMAD) of exposure.
the nucleation mode, and the fraction of the attached The activity size distribution of the radon progeny
Potential Alpha Energy Concentration (PAEC) asso- depends on the exposure conditions. Results of cal-
ciated with the nucleation mode. In contrast, the culations performed with the HRTM to approximate
parameters that had little effect on the lung dose the effective dose per unit exposure to radon
per WLM include parameters related to the coarse progeny for different exposure conditions are given
mode, the density and shape factor of the un- in Table 3.13. The variation in doses for the home
attached and accumulation mode particles, and the environment [10–21 mSv WLM21; 2.8 –5.9 mSv per
equilibrium factor. For an adult at home, the effect- (mJ h m23)] and for the workplace [11–26 mSv
ive dose per WLM as a function of unattached WLM21; 3.1 –6.3 mSv per (mJ h m23)] is due to

Table 3.13. Values of effective dose per unit exposure to 222Rn progeny calculated with the HRTM for different exposure scenarios

Exposure scenario Publication Effective dose per unit exposurea

mSv WLM21 mSv (mJ h m23)21

Adults at homeb
Home without cigarette smoke Marsh and Birchall (2000) 15 4.2
James et al. (2004) 21 5.9
Marsh et al. (2005) 13 3.7
Home with cigarette smoke James et al. (2004) 16 –18 4.5–5.1
Marsh et al. (2005) 10 2.8
Workplaceb
Indoors Harrison and Marsh (2012) 21 5.9
Tokonami et al. (2003) 20 5.6
With air conditioning on 22 –24 6.2–6.8
With air cleaner on 56 16
Mines Solomon et al. (1994) 16c – 26d 4.5c –7.3d
James et al. (2004)e 18 –21 5.1–5.9
Marsh et al. (2005) 12.5 (9c –13.5f ) 3.5 (2.5c –3.8f )
Harrison and Marsh (2012) 11c 3.1c

a
1 WLM ¼ 3.54 mJ h m23.
b
An average breathing rate of 0.78 m3 h21 was assumed for an adult at home and a value of 1.2 m3 h21 for a standard worker.
c
For diesel-powered mines with diesel engines in operation.
d
For non-diesel areas of mine.
e
Calculations were performed with a time-weighted mean activity size distribution obtained from measurements in four different work
areas of diesel-powered mines (NA/NRC, 1999a).
f
Without working actions.

45
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

different assumptions of the aerosol parameter range of values from 5.7 to 71 nGy (Bq h m23)21, with
values. For an indoor workplace with an air cleaner an average value of 17.5 nGy (Bq h m23)21 illustrates
in operation, the unattached fraction was high (fp ¼ model-specific differences in assumptions regarding
0.56) resulting in a calculated effective dose of 56 unattached fraction, breathing rates, bronchial clear-
mSv WLM21 or 16 mSv per (mJ h m23). Although, ance mechanisms, and target cell depths as well as
there was a reduction in the radon progeny activity on the application of different morphometric models.
concentration by the use of an air-cleaning device, Effective dose coefficients proposed by modeling
the increased fp value showed that the air cleaner efforts published after the publication of the
was not effective for dose mitigation (Tokonami UNSCEAR (2008) report are listed in Table 3.15 for
et al., 2003). uranium mining exposure conditions. Here, effective
The calculations of James et al. (2004) used the dose conversion coefficients are expressed in terms of
reference activity size distributions for homes and the cumulative exposure to radon progeny in WLM.
mines given in the BEIR VI report (NA/NRC, Dosimetry models comprise generation-based models
1999a). For the home, this was based on measure- and the compartmental ICRP (1994) model as well as
ments carried out by Hopke et al. (1995) in six deterministic and stochastic models. Compared with

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

houses in the USA and Canada. Size distributions the wide range of dose factors listed in Table 3.14, the
were recommended for homes with and without cig- effective dose conversion coefficients shown in
arette smoke. The resulting estimates of effective Table 3.15 are relatively similar to each other.
dose per unit exposure [16-21 mSv WLM21; 4.5–5.9 In summary, the various parameters included in
mSv per (mJ h m23)] were higher than those calcu- the models rely on relatively well-documented ana-
lated by Marsh et al. (2005) [10– 13 mSv WLM21; tomical values such as airway dimensions, breath-
2.8 – 3.7 mSv per (mJ h m23)] based on activity size ing rates, mucous clearance rates, and target cell
measurements carried out in Europe (Table 3.13). depths. Other physical parameters that estimate ex-
Porstendörfer (2001), using an airway generation posure conditions such as unattached fraction,
model developed by Zock et al. (1996), calculated the aerosol particle diameters are assumed conditions
effective dose per unit exposure as a function of fp and vary significantly.
for different workplace scenarios and for a home. One uncertainty calculation derived from pub-
For example, the effective dose per WLM for an lished bronchial dose model parameter values is
adult at home was given as: reported in NCRP Report No. 164 (NCRP, 2009).
This Monte Carlo uncertainty derivation assumed
Effective dose per WLM ¼ 6:1 þ 42 fp mSv WLM1 that all model parameter values have a lognormal
ð3:6Þ distribution. The analysis (NCRP, 2009) resulted in
an overall arithmetic mean of 10 mSv WLM21 with
This gives a value of 10 mSv WLM21 [2.9 mSv per SD of 5, and a geometric mean of 9 mSv WLM21
(mJ h m23)] with fp ¼ 0.1, which is lower than the with GSD of 1.6. The dose conversion coefficient
corresponding value of 16 mSv WLM21 [4.4 mSv per derived from modeling parameter uncertainties has
(mJ h m23)] calculated with the HRTM [Equation a relatively large error, thus the potential for error
(3.4)]. term reduction exists.
Marsh et al. (2002) carried out a parameter uncer-
tainty analysis with the HRTM to calculate the
probability distribution of the equivalent dose to the
3.9 Variability and Uncertainty of Individual lung (wlung Hlung) per unit exposure to radon
Lung Doses progeny in the home. It was assumed that the
HRTM is a realistic representation of the physical
3.9.1 Comparison of Results from Different
and biological processes, and that the parameter
Lung Dosimetry Models
values are uncertain. The parameter probability dis-
Deterministic and stochastic models have been tributions used in the analysis were based upon pub-
published in the past which permit the calculation of lished data and on expert judgment. Parameters
radon progeny doses to bronchial airway generations, considered include: (i) aerosol parameters, (ii)
the pulmonary region, and the dose to specific lung subject-related parameters such as breathing rate,
cancer target cells in each region. UNSCEAR (2008) fraction breathed through the nose, and particle
summarized 13 principal bronchial dose models for transport rates, (iii) target cell parameters such as
radon progeny, published between 1956 and 1998, to depth of basal and secretory cell layer, and (iv) ab-
determine a central value (Table 3.14). Dose factors sorption rates of attached and unattached radon
are expressed in this table in terms of the equilibrium progeny. The resulting distribution was approxi-
equivalent radon activity concentration. The wide mately lognormal with a geometric mean of 14 mSv

46
 Table 3.14. Doses from deposited radon progeny derived from principal dose models (UNSCEAR 2008). For the individual references listed in this table, the reader is referred to the
UNSCEAR report

Year Investigator Parameter values Target region Model type Dose factora
[nGy(B q hm23)21]
Unattached Breathing rate
fraction (m3 h21)

1956 Chamberlain and Dyson [C1 ] 0.09 1.2 Average in 45 mm epithelium Cast of trachea and bronchi 11
1959 ICRP [14] 0.1 1.2 Mean TB region Deposition retention assumptions 6.7
1964 Jacobi [Jl] 0.25 Basal cells (30 mm) Findeisen/Landahl 6-region anatomical model 24
1964 Altshuler et al. [A3] 0.085 0.9 Basal Cells (22 mm) Findeisen/Landahl 6-region anatomical model 32
1967 Haque and Collinson [H3] 0.35 Basal Cells (30 mm) Weibel dichotomous airway model 71
1972 Harley and Pasternack [H5] 0.04 0.9 Basal Cells (22 mm) Weibel dichotomous airway model 5.7
1980 Jacobi and Eisfeld [J2] 0.1 1.2 Mean epithelium Weibel dichotomous airway model, correction 8.9

47
for upper airway turbulent diffusion [M2]
1980 Janies et al. [J6] 0.1 1.2 Mean epithelium Yeh-Shum anatomical model [Y2] 14
1982 Harley and Pasternack [H6] 0.07 1.1 Basal Cells (22 mm) Same as Jacobi and Eisfeld [J2] 6.4
1982 Hofmann [H10] 0.2 0.9 Mean epithelium Same as Jacobi and Eisfeld [J2] 11
1991 National Research Council [N10] 0.16 1.2 Basal Cells (35–50 mm) Yeh-Shum anatomical model [Y2], correction 21
for upper airway turbulent diffusion
1996 Harley et al. [H4] 0.1 1.2 Basal Cells (27 mm) Nikiforov and Schlesinger [N12] anatomical 9
model, airway deposition from empirical data
from human airway casts
1998 Marsh and Birchall [M2] 0.08 0.8 Bronchial cells: Basal (35 –50 mm), ICRP lung model [11] 8.5
secretory (10–40 mm) Bronchiolar cells: 19
secretory (4–12 mm) 14
Radon and Radon Progeny Inhalation and Resultant Doses

a
Per unit 222Rn concentration (EEC). WLM converted to Bq h m23 using 0.27 1023 WL (Bq m23)21 and 170 h per working month.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 3.15. Comparison of effective doses per WLM obtained with of Falk et al. (1997; 1999). However, the lung dose
different lung dosimetry models for uranium mining exposure arising from exposure to radon progeny is not very
conditions
sensitive to the bronchial clearance rates as the
Authors Effective dose conversion
progeny are short-lived compared with these half-
coefficients (mSv WLM21) times (Marsh and Birchall, 2000). Therefore, lung
doses calculated with the revised model will not be
Harley (1984) 6.9 (working þ diesel) significantly different for radon progeny. Effective
Winkler-Heil et al. (2007) 8.3 (RADOS) doses per unit exposure to radon progeny calculated
8.9 (mean) (IDEAL-DOSE) with the revised model for home and mine exposures
7.8 (median) (IDEAL-DOSE)
Marsh et al. (2005) 12.5 (HRTM; ICRP 1994)
were about 13 and 11 mSv WLM21, respectively
Marsh and Bailey (2013) 10.6 (revised HRTMa) (Marsh and Bailey, 2013).
Porstendörfer (2002) 11.5 (without sources)
11.2 (with sources)
5.7 (plus coarse aerosol) 3.9.2 Intra- and Intersubject Variability
Porstendörfer and Reineking 9.0 (working þ diesel)
In radiation protection, bronchial doses are routinely

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(1999) 6.7 (without working)
Average value 8.9 calculated for a standard or reference man by assuming
defined average values for all anatomical and physio-
a
Calculated with the revised Human Respiratory Tract Model logical parameters involved in radon lung dosimetry,
(HRTM) (Bailey et al., 2009). A full description of this model will therefore providing a single dose value for specified ex-
be given in the ICRP Publication on “Occupational Intakes of
posure conditions. In reality, however, these anatomical
Radionuclides, Part 1.”
and physiological parameters can vary significantly
within a given subject (intra-subject variability) or
among different subjects (inter-subject variability),
WLM21 and a geometric standard deviation of 1.5. hence predicting a range of bronchial doses for the
The limits of the 95% confidence interval corre- same exposure conditions. For example, sources of
sponded to 6.3 – 31 mSv WLM21. Birchall and James intra-subject variability are the asymmetry and vari-
(1994) also carried out a parameter uncertainty ana- ability of linear airway dimensions in bronchial and
lysis with the HRTM to calculate the distribution of alveolated airways (Koblinger and Hofmann, 1985;
effective doses per unit exposure to radon progeny in 1990), mucociliary clearance velocities and transit
a mine and obtained similar results for the statistic- times in bronchial airways (Hofmann and Sturm,
al parameters of the dose distributions. 2004) and the location of target cells throughout the
To obtain an indication of the uncertainty caused bronchial epithelium (Harley et al., 1996; Mercer et al.,
by the structure of the human respiratory tract 1991). Sources of inter-subject variability are breathing
model, Winkler-Heil et al. (2007) compared pre- parameters for defined physical activities (ICRP, 1994),
dicted effective doses for radon progeny inhalation daily physical activity patterns (ICRP, 1994), size and
obtained using the HRTM, a deterministic airway structure of nasal and oral passages (Cheng et al.,
generation model and a stochastic airway generation 1996), size of the lung (FRC) (Hofmann et al., 2002),
model. The same input parameter values were tracheobronchial mucociliary clearance rates (Yeates
assumed for mine conditions. The three models et al., 1975), and thickness of bronchial epithelium
yielded similar results, ranging from 8.3 to 11.8 mSv (Mercer et al., 1991).
WLM21 (2.3 – 3.3 mSv per mJ h m23). The authors Hofmann et al. (2010) carried out an uncertainty
noted that one of the important issues affecting the analysis with a stochastic airway generation model to
comparison is the averaging procedure for the doses derive the frequency distribution of bronchial doses
calculated in airway generation models. per WLM taking account of inter- and intra-subject
ICRP has recently updated the ICRP Publication variability, while aerosol parameter values were fixed
66 Human Respiratory Tract Model (ICRP, 1994) to for a mine atmosphere. In the stochastic dose model,
take account of more recent data, mainly from volun- inter-subject variability of bronchial doses was defined
teer studies (Bailey et al., 2009). Changes relate to as the effect of morphological and physiological param-
particle transport from the nasal passages, bron- eter variations on bronchial doses among a group of
chial tree, and alveolar region. In particular, the subjects for defined exposure conditions, where each
slow particle clearance component with a 23 d half- subject is characterized by a dose distribution due to
time assumed for the bronchial and bronchiolar intra-subject variations. The primary biological para-
regions has been revised. It is now assumed that meters contributing to the intra- and inter-subject
slow particle clearance from the bronchial tree only variability of bronchial doses were the variability of
occurs in the bronchiolar region with a half-time of the extrathoracic and thoracic airway structure and
about 3.5 d, which is based on the volunteer studies airway dimensions, random variations of breathing

48
 Radon and Radon Progeny Inhalation and Resultant Doses

parameters, individual mucociliary clearance veloci- inhomogeneous dose distributions within airway
ties, and variations of the thickness of the bronchial bifurcations. To illustrate this dose inhomogeneity,
thickness and related depths of target cells (Mercer doses were calculated for three selected sites within
et al., 1991). Calculations of inter-subject variability an asymmetric bifurcation: T (carinal ridge), R1 (cy-
indicated that the asymmetry and variability of the lindrical section), and R2 (curved transition zone).
airway geometry is the most important factor, followed For a cumulative exposure of 20 WLM, typical for
by the filtering efficiency of nasal passages and by the residential radon exposures, resulting doses in se-
diameter-related thickness of the bronchial epithelium cretory cells, located at a depth of 20 mm, ranged
(Hofmann et al., 2010). from 5.14 Gy at T to 4.33  1022 Gy at R1, with an
Resulting bronchial dose distributions were approxi- intermediate value of 2.04  1021 Gy at R2 (Fakir
mated by lognormal distributions; BB: median ¼ et al., 2005). In general, about 10% of the bifurcation
3.2 mGy WLM21, GSD ¼ 2.3; and bb: median ¼ 2.3 surface area receives a dose roughly 10 times higher
mGy WLM21, GSD ¼ 4. The results showed that the than the average bifurcation dose. At the tip of the
inter-subject variations were significantly higher in carinal ridge, this dose ratio may be about hundred
the peripheral bronchiolar airways than in the larger times higher, although for an even smaller number

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

bronchial airways. These dose distributions were also of cells.
much wider than the distributions obtained with the The crucial question is then whether such local
HRTM (Marsh et al., 2012). In deterministic dosimetry inhomogeneities of radon progeny surface activities
models, such as the HRTM, each individual is charac- are more carcinogenic than the same total activity
terized by a single dose value. Thus, stochastic dosim- uniformly distributed. For example, limited epi-
etry models will, by definition, produce wider, more demiological data on lung cancer mortality following
realistic dose distributions. occupational inhalation of plutonium aerosols and
Available morphometric models of the human tra- the incidence of liver cancer and leukemia in thoro-
cheobronchial tree are based on measurements from trast patients suggest a moderate enhancement
a few laboratories of a small number of individuals. factor for inhomogeneous versus uniform radio-
In order to determine the degree of inter-subject nuclide distributions of the same total activity
variability, analyses of airway lengths, diameters, (Charles et al., 2003). Furthermore, histological
and branching angles from accident victims were studies have revealed that neoplastic and preneoplas-
performed using solid silicone casts of the upper tic lesions are preferentially observed at carinal
bronchial tree from eight human lungs (Nikiforov ridges in the human lung (Garland et al., 1962).
and Schlesinger, 1985). The results indicated that Although no pertinent information is currently avail-
there are significant differences among subjects. able for inhaled radon progeny, both findings suggest
The coefficient of variation was least for diameters that carinal ridges in bronchial airway bifurcations
(29%) and greatest for branching angles (78%), and are indeed the initial sites of bronchial tumor occur-
intermediate for lengths (42%). rence, consistent with the predicted local dose distri-
butions.
3.9.3 Inhomogeneity of Surface Activities
and Resulting Doses within Bronchial 3.9.4 Comparison of Bronchial Doses
Airways between Non-Smokers and Smokers
In lung dosimetry models, it is commonly Changes of several anatomical, physiological, and
assumed that radon progeny activities are uniformly histological parameters, such as decreased or
distributed within the mucous-serous layer along increased mucus velocity, increased thickness of the
the whole surface area. Consequently, each cell at a mucus layer, increased mucus viscosity, decreased
given depth in a given generation receives the same lung volumes, increased breathing frequency,
dose. However, experimental and modeling studies smaller tidal volumes, bronchial airway obstructions
revealed a strong inhomogeneity of radon progeny and destruction of the alveolar architecture, and
deposition, with significantly enhanced deposition basal cell hyperplasia as a result of cigarette con-
at the carinal ridge, together with an impaired sumption have been reported (Baias et al., 2010).
mucus transport around the dividing wedge of the Based on these observations, smokers were subdi-
two daughter branches (Farkas and Szöke 2013; vided into four exposure categories by varying cigar-
Hofmann et al., 1990). The resulting inhomogeneous ette consumption and duration of exposure: Light
activity distributions, i.e., local accumulations of ac- short-term, light long-term, heavy short-term, and
tivity at carinal ridges due to enhanced deposition heavy long-term smokers. It is important to note
and reduced mucociliary clearance, together with a that published physiological and morphological data
non-uniform irradiation geometry lead to vary widely among the different sources, so that

49
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

these categories represent only rough approxima- Table 3.16. Comparison of basal and secretory cell doses per unit
tions. The results of bronchial dose calculations exposure and their combination by applying different weighting
procedures for uranium mining conditions (Marsh et al., 2005;
(Baias et al., 2010) are only briefly summarized
Winkler-Heil et al., 2007)
here: (1) Calculated doses for light short-term
smokers deviate only minimally from the doses in Bronchial doses (mGy WLM21)
non-smokers as only small changes of morphological
and physiological parameters have been reported; Basal Secretory Equal Weighted by
weighting density
(2) for light long-term and heavy short-term
smokers, doses will be reduced due to a thickening of
BB (1 –9) 2.77 6.23 4.50 3.34
the mucus layer and increased mucus velocities; and bb (10 –16) 2.30 2.33 2.05 1.85
(3) doses for heavy long-term smokers can increase BB þ bb 2.54 4.28 3.27 2.60
by up to a factor of 2 relative to non-smokers, caused
by impaired mucus clearance, higher breathing fre-
quencies, reduced lung volume, and airway obstruc- dosimetric point of view, non-targeted cells do not
tions (see Section 3.5 for human studies of lung affect the resulting doses listed in Table 3.16. Due to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

clearance in smokers and non-smokers). the columnar structure of the bronchial epithelium,
the depth distributions of epithelial cells are the
same as that of the secretory cells (Mercer et al.,
3.9.5 Contribution of Sensitive Target Cells
1991) and hence doses based on scenario 1 are the
in Bronchial Epithelium to Lung
same as the secretory cell doses. If the initial
Cancer Risk
damage is produced in all epithelial cells, then the
At present, basal and secretory cells are considered probability that this damage will occur in basal and
to be the primary target cells in the bronchial epithe- secretory cells (scenario 2) is proportional to their
lium (ICRP, 1994). Robbins et al. (1990) measured the relative volumetric densities and thus equal to the
volume density of basal, mucous, and indeterminate weighted bronchial cell doses.
cells in four bronchial airway generations in smokers For the assessment of an average lung dose to be
and non-smokers. A comparison of volume density of related to lung cancer risk, an additional weighting
basal and mucous cell nuclei between smokers and procedure has been introduced through apportion-
non-smokers showed no statistically valid differences. ment factors for the BB, bb, and AI regions. Equal
However, the volume density of mucous cell nuclei weighting (ABB:Abb:AAI ¼ 0.333:0.333:0.333) (ICRP,
was considerably less than that of basal cell nuclei in 1994) leads to an average lung dose of 2.36 mGy
both smokers and non-smokers. WLM21, while application of the apportionment
Because of lack of more pertinent information, it factors (ABB:Abb:AAI ¼ 0.80:0.15:0.05) proposed by
is assumed that both cell types equally contribute Porstendörfer (2002) yields an average lung dose of
to bronchial tumor induction. Hence bronchial 4.07 mGy WLM21.
doses are commonly expressed as the average of 50%
basal and 50% secretory cell doses, except for the
IDEAL-DOSE model, where basal and secretory
3.10 Human versus Experimental
doses are weighted by their relative nuclear volu-
Animal Doses
metric densities (Winkler-Heil and Hofmann, 2005),
based on the measurements of Mercer et al. (1991). Laboratory animals have been used as human sur-
As opposed to these direct effects, non-targeted rogates to supplement epidemiological studies for the
mechanisms, where cells not hit at all exhibit a assessment of lung cancer risk following exposure to
radiobiological response, may play an important radon and its short-lived progeny. The primary ad-
role, particularly at low doses, where only a small vantage of laboratory animal studies is the possibility
fraction of cells is actually hit (Brenner et al., 2001; to conduct inhalation experiments under controlled
Fakir et al., 2009). In terms of cellular dosimetry, exposure conditions and to investigate the effects of
this would imply that all epithelial cells should be exposure characteristics, such as the equilibrium
considered as initial target cells (scenario 1), al- factor or the unattached fraction, and the contribu-
though only basal and secretory cells may still be tion of concomitant carcinogenic pollutants and non-
the primary recipient cells (scenario 2). radiological cancer-related factors, such as cigarette
Cell-specific and weighted doses to different sensi- smoke. Indeed, the uncertainty associated with the
tive target cells are presented in Table 3.16. These post-exposure assessment of radon and radon
results indicate that the choice of different target progeny exposure parameters in uranium miner and
cells or any combination thereof will lead to a range indoor radon cancer surveys is one of the major defi-
of values varying by about a factor of 2. From a ciencies of epidemiological studies. Moreover, for low

50
 Radon and Radon Progeny Inhalation and Resultant Doses

level radon exposures in homes, the radiological cellular and tissue level in both animals and
radon and radon progeny effect may be masked by humans. To further explore this issue, dosimetry
the concomitant exposure to other non-radiological models for different animal species will be discussed
co-carcinogens, which may act in a synergistic or an- in the following section.
tagonistic fashion, and hence cannot reliably be sepa-
rated from the effects of these additional exposures.
3.10.2 Animal Dosimetry Models
Animal studies have been conducted primarily with
3.10.1 Animal Inhalation Experiments
rats; however, dogs, hamsters, and mice were also used
Several types of laboratory animals, primarily as experimental animals. To compare species-specific
adult males, have been used in radon inhalation risk estimates and thus to make risk extrapolations
experiments. These ranged from small rodents, such from animals to humans, detailed dosimetric models
as CAF strain mice (Morken and Scott, 1966), are needed to account for the various dissimilarities of
Sprague–Dawley (Monchaux, 2005), and Wistar rats morphometric and physiological parameters, which
(Cross and Monchaux, 1999), and Syrian Golden may impact on the resulting doses to sensitive target

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

hamsters (Desrosiers et al. 1978), to beagle dogs cells.
(Cross et al. 1982). Among these animals, the most ex- The first dosimetry model for an animal lung was
tensive inhalation experiments were carried out with developed by Desrosiers et al. (1978), who calculated
rats (Cross and Monchaux, 1999). A detailed report of airway doses for three reference mine atmospheres
the earlier animal inhalation experiments carried out in the Syrian Golden hamster lung, ranging from
in the USA can be found in the NCRP Report No. 78 about 0.3 mGy WLM21 to about 14 mGy WLM21.
(NCRP, 1984), while the results of the more recent in- These calculations indicated that all reference atmo-
halation studies with rats in the USA and in France spheres produced radiation doses in the hamster
are summarized in the papers of Cross (1988a, lung which are lower in the upper bronchial airways
1988b), Cross and Monchaux (1999), and Monchaux than those predicted in humans.
(2005). The frequencies of non-respiratory neoplasms For inhalation experiments using dogs at the
such as bone, liver, and breast cancers were elevated, University of Rochester (UR) (Morken, 1973), doses
although no excess of tumors other than respiratory to the trachea, the major bifurcation, and the whole
have been observed in the human underground miner lung were estimated by measuring the related radon
studies. progeny activity concentrations. Although these
Tobacco smoke was observed to enhance the estimates were not based on a specific dosimetric
number of lung tumors by a factor of 2–4 when model, they ranged from 3.6 to 207 mGy WLM21 for
smoke exposure was given following 222Rn exposure selected bifurcations and about 50 mGy WLM21 for
(Cross, 1988b; Monchaux et al., 1994). Smoke expos- the whole bronchial tree, which is about an order of
ure decreased the tumor latency period. NCRP (1984) magnitude higher than that estimated for the
summarized the risk models developed in rat studies human bronchial region for the same exposure con-
and derived an estimated excess lifetime lung tumor ditions (NCRP, 1984).
risk coefficient of 3  1024 per WLM. This value is Another experimental inhalation study with
similar to the ICRP (2010) recommended estimate beagle dogs was conducted at the Battelle Pacific
and to the UNSCEAR (2008) estimate of 5  1024 per Northwest Laboratories (PNL) (Cross et al., 1982).
WLM for human underground miner data as well as To understand possible mechanistic differences
to the recently published estimate of 5  1024 per between the responses of the human and dog lung to
WLM for the human lifetime excess absolute risk for radon progeny exposure in terms of radiation doses,
the ICRP reference population consisting of smokers Harley et al. (1992) developed a dosimetric model for
and non-smokers (ICRP, 2010). the dog lung. The computed alpha dose per unit ex-
These animal radon inhalation studies, particu- posure to basal cell nuclei in the upper bronchial
larly those with rats, revealed that the lung cancer airways ranged from 2 to 7 mGy WLM21 depending
risk per unit exposure observed is consistent with upon the exposure protocols used in the UR and
that reported for uranium miners over a wide range PNL studies, respectively. The dose to the alveolar
of radon exposure levels (Cross 1988a; Harley, 1988). tissue in both studies was 3 mGy WLM21. For com-
This observation raises the questions whether the parison, model predictions for the human lung
similarity of lung cancer incidences is due to the under the same exposure conditions were 9 mGy
similarity of bronchial doses. An alternative inter- WLM21 for the bronchial airways and 0.5 mGy
pretation could also be that doses are different, but WLM21 for the alveolar tissue.
that these differences are compensated by compar- In the following years, practically all experimental
able differences in the carcinogenic response at the animal radon inhalation studies were carried out

51
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

with rats. In particular, two major inhalation studies particle diameters of the attached radon progeny
with radon progeny have been conducted in the USA (0.12 or 0.5 mm AMD) and unattached fractions
at the Battelle Pacific Northwest Laboratories (PNL) (1.3% or 9.5%), doses to bronchial target cells ranged
(Cross, 1988a, 1988b; Cross and Monchaux, 1999; from 1.8 to 10.7 mGy WLM21, averaged over all
Cross et al., 1982; 1984; Gilbert et al., 1996), and in bronchial generations (except the trachea)
France at the Compagnie Generale des Matieres (Table 3.17). For comparison, the alpha dose to the
Nuclaires (COGEMA) (Chameaud et al., 1984; alveolar region ranged from 1.8 (0.5 mm AMD) to
Monchaux, 2004; 2005; Monchaux and Morlier, 2002; 3.6 mGy WLM21 (0.12 mm AMD). The average bron-
Monchaux et al., 1994; 1999). These studies were chial dose value over all exposure conditions of
complemented and augmented by rat inhalation 5.6 mGy WLM21 is very similar to the value for both
experiments carried out in the UK at Harwell, focus- occupational and environmental exposures for humans
ing on cellular radiobiological effects (Collier et al., of about 5 mGy WLM21.
1999; 2005). The wealth of information produced by The dosimetric model proposed by Hofmann et al.
these studies underscored the necessity to develop (1993) utilized the morphometric model of Yeh et al.
detailed dosimetric models for the rat bronchial tree (1979) for the Long Evans rat and applied it both to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(Fakir et al., 2008; Harley, 1988; Hofmann et al., the Sprague – Dawley and Wistar rats due to their
1993; Winkler-Heil et al., 2014). similarity (Mercer and Crapo, 1989). Since the ana-
A complete bronchial morphometric model of tomical dimensions of the Yeh et al. (1979) model re-
the tracheobronchial tree for a female Long Evans portedly refer to total lung capacity (TLC), linear
black and white hooded rat has been reported by airway dimensions were scaled down to an FRC
Raabe et al. (1976) and Yeh et al. (1979). On the of 40% TLC, assuming that diameters and lengths
other hand, the rats used in the US and French in- vary with the cube root of lung volume. Relative
halation studies were male Wistar rats (Cross et al., volumetric fractions of basal and secretory cell
1984) and male Sprague – Dawley rats (Chameaud nuclei as functions of depth in bronchial epithelium
et al., 1981). A comparison of morphometric data for were taken from Mercer et al. (1991). Doses were
tracheobronchial airways between a Long Evans rat computed to basal and secretory cell nuclei in all
(Koblinger and Hofmann, 1988; Raabe et al., 1976; bronchial airway generations for both the Battelle
Yeh et al., 1979) and a Sprague – Dawley rat (Mercer (0.3 mm AMTD) and COGEMA (assumed 0.15 mm
and Crapo, 1989) indicated that the anatomical AMTD) aerosol parameters and different equilib-
structures of both rat strains were very similar. rium factors and unattached fractions. Computed
Therefore, it is a reasonable assumption for model- mean bronchial dose ranged from 7.1 to 10.9 mGy
ing purposes that the morphometric data provided
by Raabe et al. (1976) and Yeh et al. (1979) are a Table 3.17. Model-derived dose conversion coefficients of the
valid representation for all rat strains used in the in- bronchial airways for different animal species and the human lung
halation studies, although differences in breathing
Animal Dose conversion Reference
parameters as a function of body weight have to be coefficient
taken into account. Based on rigorous statistical (mGy WLM21)
analyses of the morphometric data provided by
Raabe et al. (1976) for the Long Evans rat, Syrian hamster 0.3–1.4 Desrosiers et al. (1978)
Koblinger and Hofmann (1988) developed a stochas- Beagle dog 2–7 Harley et al. (1992)
tic model of the rat tracheobronchial tree, later sup- Balb/c mouse 8.6–31.4 Sakoda et al. (2013)
Rat (Sprague–Dawley, 5.6 (1.8–10.7) Harley (1988)
plemented by a corresponding stochastic model of Wistar, Fisher)
the alveolar region based on measured data in a Rat (Long-Evans, 8.8 (7.1–10.9) Hofmann et al. (1993)
Sprague – Dawley rat (Koblinger et al., 1995). Wistar, Sprague–
The first dosimetric model for radon progeny in- Dawley)
halation in the rat lung was developed by Harley Rat (Long Evans, Wistar, 8.1 (7.2)a Fakir et al. (2008)
Sprague–Dawley)
(1988), based on the morphometric data for the Rat (Long Evans) 8.6–18.3 Sakoda et al. (2013)
female Long Evans black and white hooded rat pro- Rat (Long Evans, 7.8 Winkler-Heil et al.
vided by Raabe et al. (1976) and Yeh et al. (1979) and Sprague–Dawley) (2014)
strain-specific breathing parameters for the Long Human 6.0 Harley et al. (1996)
Evans, Wistar or Sprague – Dawley, and Fisher 334 3.6–5.0 Hofmann et al. (1993)
6.3 Fakir et al. (2008)
rats. Tracheobronchial doses were computed for the 5.8 Winkler-Heil et al.
reference atmosphere reported by Cross (1988a) for (2014)
the Battelle inhalation experiments and for the
a
natural ambient aerosol reported by Chameaud Bronchial dose without crossfire from nuclide sources in alveolar
et al. (1984) for the French studies. Depending on tissue.

52
 Radon and Radon Progeny Inhalation and Resultant Doses

WLM21 (Table 3.17), while corresponding alveolar the range from 51% up to 100%. Since both esti-
doses ranged from 1.2 to 4.9 mGy WLM21. For com- mates are based on modeling assumptions, it cannot
parison, bronchial cellular doses for the human lung be decided on scientific grounds which values are
for typical indoor exposure conditions are approxi- more realistic.
mately 5.0 mGy WLM21, i.e., average doses to the The results from the dosimetric analyses of radon
rat bronchial tree are slightly higher than those to progeny inhalation experiments with rats can be
the human bronchial cells by about a factor 1.5 –2. summarized as follows:
More recently, Fakir et al. (2008) analyzed rele-
vant microdosimetric quantities and investigated
(1) Absorbed doses and microdosimetric quantities
the contribution of crossfire alpha particles emitted
are slightly higher in rat bronchial airways than
from radon progeny deposited in the alveolar region
in corresponding human airways. This confirms
to bronchial absorbed doses. Based on the dosimetric
the a priori assumption in rat inhalation experi-
model of Hofmann et al. (1993) described above,
ments that the rat lung is a suitable surrogate
hit frequencies, absorbed doses, and critical micro-
for the human lung.
dosimetric quantities were calculated for basal and

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(2) The doses to the alveolar region relative to those
secretory cell nuclei located at different depths in epi-
to the bronchial airways are significantly higher
thelial tissue for each bronchial airway generation
in the rat than in the human lung. This may
and for defined indoor exposure conditions. Absorbed
partly explain the experimental findings that
doses, considering the effect of crossfire, and cellular
approximately 70% of the observed tumors in the
hit frequencies were slightly higher in rat airways
rat lung are bronchogenic carcinomas and 30% of
than in corresponding human airways. Average
bronchioalveolar origin (Cross, 1988), contrary to
absorbed doses, including alpha particle emissions
the cancer distribution reported for uranium
from the mucus layer and the alveolar region and
miners with about 84% in bronchial–bronchiolar
averaged over the entire bronchial tree (without
airways and 16% in the alveolar region (Ellett
trachea), were 8.1 mGy WLM21 for the rat and 6.3
and Nelson, 1985; Saccomanno et al., 1996).
mGy WLM21for the human lung (Table 3.17). While
Doses to the bronchiolar airways are further
the contribution of crossfire alpha particles is insig-
increased by high-energy transfer of crossfire
nificant in the human lung, it can reach 33% in per-
alpha particles emitted from the alveolar region
ipheral bronchiolar airways of the rat lung.
(this effect can be neglected in the human lung).
Based on the stochastic model of the rat lung
(Koblinger et al., 1995), Winkler-Heil et al. (2014) re-
cently performed dose calculations for typical indoor In conclusion, the animal studies performed in the
exposure characteristics. Since studies by Hofmann USA and France have been entirely supportive of the
et al. (1999) have demonstrated that airway dia- human epidemiology. With the exception of tumor
meters are more appropriate morphometric scaling type (i.e., a greater prevalence of solid alveolar
parameters to classify local deposition patterns tumors and bronchiolar–alveolar carcinomas in
across different species than the conventionally animals), the lung cancer response is reasonably con-
used airway generations, dose distributions as func- sistent with that in human exposures. The slightly
tions of airway diameter classes were computed for different occurrences of bronchial carcinomas in the
the human and rat lung. Assuming an AMTD of human and rat bronchial airways following exposure
0.3 mm for the attached fraction and 5 nm diameter to radon and its short-lived progeny might be
for an unattached fraction of 3%, the mean bronchial explained either by differences of doses to sensitive
doses were 7.8 mGy WLM21 for the rat and 5.8 mGy target cells in the bronchial epithelium or by differ-
WLM21 for the human lung, respectively. ences in radiosensitivities of target cells in both
Corresponding dose calculations for the human species. Cross (1988) explained these observed differ-
bronchial airways for the exposure conditions used ences in tumor incidences by assuming that rats are
already in the dose calculations for the rat lung about twice as sensitive to lung cancer induction as
(Fakir et al., 2008; Harley, 1988; Hofmann et al., humans for the same cumulative exposure. Dosimetric
1993) are also listed in Table 3.17. calculations (Hofmann et al., 1993; Winkler-Heil et al.,
In the human lung, alveolar doses are typically 2014) suggest, however, an alternative interpretation,
of the order of 10% or less of bronchial doses namely that differences in cancer risk might also be
(Hofmann, 1982c), while alveolar doses in the rat related to differences in bronchial doses, while both
lung range from about 16% up to 50% (Hofmann species have similar radiobiological sensitivities
et al., 1993), depending on modeling assumptions. for the range of doses produced in the inhalation
For comparison, Harley (1988) reported values in experiments.

53
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv007
Oxford University Press

4. Characteristics and Behavior of Radon and Radon Progeny

4.1 Radon Sources

Therefore, there is a large variation of indoor 222Rn
Radon is a naturally occurring radioactive gas, activity concentrations depending upon heating,
which has no taste, smell, or color. It is an inert ventilation rates, and meteorological conditions as
noble gas that is encountered in elemental form well as the geology of the area.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

either as a gas, or dissolved, usually in water. Buildings with concrete floors often have cracks
There are a number of isotopes of radon (Firestone around edges and gaps around service entries, such
and Shirley, 1999), but the most important isotopes for as water supply, electricity, or sewage pipes. Cracks
radiation protection are 222Rn and 220Rn. Radon-222 is and gaps also permit soil gas entry. If buildings have
a member of the uranium (238U) natural decay series suspended timber floors, then the gaps between the
and 220Rn is a member of the thorium (232Th) natural floorboards are the main route of entry.
decay series (see Figures 4.1 and 4.2). Because of their In building areas where 222Rn levels are low, such
origins, the isotopes 222Rn and 220Rn are commonly as on upper floors of a block of flats, the main sources
known as radon and thoron, respectively. Rn-222 is a of radon may be building materials and outdoor air.
direct decay product of 226Ra and 220Rn is a direct Exposure to radon in buildings may also arise in
decay product of 224Ra. areas contaminated with radium from past industrial
Uranium, radium, and thorium occur naturally in activities.
soil and rocks and provide a continuous source of Radon is soluble in water. Radon-222 dissolved in
radon. Radon can escape from the earth’s crust water can de-gas and escape to indoor air leading to an
either by molecular diffusion or by convection and additional source of exposure via inhalation. This is an
as a consequence is present in the air outdoors and important source of exposure for those employed in
in buildings. Radon escaping from ground to outdoor water works where ground water with a high radon ac-
air is diluted to low activity concentrations, with the tivity concentration is treated or stored (NA/NRC,
amount of dilution dependent on the atmospheric 1999b; Schmitz and Nickels, 2001; Trautmannsheimer
stability and the presence of wind and level of turbu- et al., 2003). Other examples include those working in
lence. However, radon activity concentrations can indoor areas of thermal spa facilities (Geranios et al.,
reach high levels within enclosed spaces, such as 2004; Soto and Gómez, 1999) and those members of
underground mines, caves, and buildings. the public using drilled wells. In the latter case, inges-
In general, the problems posed by radon (222Rn) tion of water should also be considered as a route of
are much more widespread than those posed by intake for radon. However, during showering, dish
thoron (220Rn). Because thoron has a short half-life washing, or if boiled, most if not all the radon is ex-
(56 s), it is less able than radon to escape from the pelled from the water. Moreover, geological factors
point where it is formed. As a consequence, building which lead to high activity concentrations of radon in
materials are the most usual source of indoor thoron water may independently give rise to high levels of
exposure. In contrast, radon, which has a half-life of radon in indoor air.
3.8 d, can diffuse in soil more than a meter from the Because radon is inert, nearly all of the gas that is
point where it is formed. As a result, the ground inhaled is subsequently exhaled. However, 222Rn
underneath buildings is usually the main source of decays into a series of solid short-lived radioisotopes,
indoor radon. Generally, high 222Rn levels in build- most of which attach to the ambient aerosol. A propor-
ings are caused by pressure-driven convection flow tion of the inhaled radon progeny deposits in the
because, as a result of warm indoor air, the air pres- respiratory airways of the lung. Because of their rela-
sure at ground level in most buildings is slightly tively short half-lives (,30 min), the radon progeny
lower than the soil air pressure. This causes a flow of decay mainly in the lungs before clearance, either by
soil air into the building carrying radon with it. absorption into blood or by particle transport to the
Other factors such as ventilation rates and meteoro- alimentary tract, can take place. Two of the short-
logical conditions also affect the convection flow. lived progeny, 218Po and 214Po, emit alpha particles

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 4.1. Natural decay series: 238U.

and it is the energy from these alpha particles that from 218Po to 214Po with half-lives of the order of
dominates the dose to the lung and the associated risk minutes and the long-lived progeny from 210Pb to the
of lung cancer. stable 206Pb with half-lives of the order of days or
years. Because of their long half-lives, resulting air-
borne activities are orders of magnitude smaller than
4.2 Radon and Thoron Decay Schemes those of their short-lived precursors. Thus, only the
inhalation of short-lived radon progeny will be con-
The decay schemes of the uranium (238U) natural sidered for the assessment of bronchial doses and
decay series and of the thorium (232Th) natural resulting lung cancer induction. Half-lives and
decay series are illustrated in Figures 4.1 and 4.2 emitted radiations of radon and its short-lived
(Be et al., 2004). Both natural decay series have progeny are listed in Table 4.1 (Be et al., 2004).
three characteristic features: Likewise, thoron progeny from 216Po to 208Tl have
† The first nuclide in the decay series has a very half-lives ranging from seconds to hours, before
long half-life, comparable to the age of the earth ending up at the stable 208Pb. Since the thoron decay
(109 – 1010 years) scheme does not contain any long-lived progeny in
† The final nuclide of the decay series is a stable contrast to the radon progeny, no distinction between
lead isotope short-lived and long-lived progeny is necessary.
† Each decay series contains a radioactive noble gas Half-lives and emitted radiations of thoron and its
progeny are listed in Table 4.2 (Be et al., 2004).
From a dosimetric point of view, the decay products of Concern for the potential hazards resulting from
222
Rn can be divided into two groups depending on exposure to the isotopes of radon is commonly cen-
their radioactive half-lives: the short-lived progeny tered upon 222Rn and 220Rn. However, there is a

56
 Characteristics and Behavior of Radon and Radon Progeny

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 4.2. Natural decay series: 232Th.

third natural radioactive decay series originating decay, interactions with aerosols, deposition on sur-
from 235U, historically known as the actinium series, faces, and ventilation. Here, particular emphasis is
which contains another radon isotope, namely 219Rn placed on the behavior of radon and its progeny in
or actinon. Because of its short half-life of only 3.96 s comparison to that of thoron and its progeny.
and the relatively low abundance of 235U in natural Starting with Jacobi in 1972, a number of similar
uranium, the 219Rn activity concentration in air is room models have been developed to describe the be-
negligible when compared with that of 222Rn. Thus, havior of these species in the indoor environment
actinon will not be discussed in the present report. (Jacobi, 1972; Porstendörfer, 1994; 2001). In these
room models, it is assumed that 222Rn and its
progeny are uniformly mixed in the indoor air. It is
also assumed that there are no thermal gradients
4.3 Behavior of Radon, Thoron and Their
present and that the air is of uniform humidity
Progeny in Indoor Environments
throughout the room volume. In what follows in this
The activity concentrations of radon, thoron, and section, unless otherwise specified, all activity con-
their progeny in indoor air arise as a result of the centrations both for the parent gases and their
interplay of a number of complex processes. The progeny are expressed in activity concentrations per
most important of these processes are radioactive unit volume of air (Bq m23).

57
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

222
Table 4.1. Main radioactive decay properties of Rn and its (4.2) may be simplified to
short-lived progeny
CiRn ¼ ERn =v þ CoRn ð4:3Þ
Radionuclide Half-life Main energies (MeV) and intensities

Alpha Beta Gamma Similar equations apply to thoron, but in this case,
the radioactive decay constant for the gas is lTn ¼
222
Rn 3.823 d 5.59 (100%) — — 45.4 h21 corresponding to a radioactive half-life of
218
Po 3.07 min 6.11 (100%) — — 56 s. Therefore, exhalation of thoron from indoor sur-
214
pb 26.9 min — 0.67 (47%) 0.053 (15%) faces is generally the principal source of indoor
0.73 (41%) 0.295 (27%)
thoron. In addition, because lTn is much greater than
1.02 (9%) 0.351 (46%)
214
Bi 19.8 min — 1.540(17%) 1.120 (15%) the range of v values under steady-state conditions
1.894 (7%) 1.764 (15%)
3.270 (20%) CiTn ¼ ETn =lTn ð4:4Þ
214
Po 162 ms 7.83 (100%) — —
Due to its short half-life, the activity concentration of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

thoron gas falls off rapidly with distance from its main
220
Table 4.2. Main radioactive decay properties of Rn and its source, the internal surfaces of the rooms. Therefore,
short-lived progeny
in Equation (4.4), C iTn should be considered as a mean
Radionuclide Half-life Main energies (MeV) and intensities
value of the thoron activity concentration in the room
air rather than a uniform activity concentration
Alpha Beta Gamma throughout the room volume. Assuming the room sur-
faces are the main source of indoor thoron then the
220
Rn 55.8 s 6.41 (100%) — 0.550 (0.12%) fall off of the thoron activity concentration with dis-
216
Po 0.15 s 6.91 (100%) — —
212 tance x from a room surface may be written as:
Pb 10.64 h — 0.331 (82%) 0.239 (82%)
0.567 (13%) p
212
Bi 60.5 min 6.17 (25%) 2.252 (55%) 0.040 (26%) CTn ðxÞ ¼ ETn = ðlTn Da Þexpðx=La Þ
6.21 (10%) 0.727 (7%) ¼ CTn ð0Þ expðx=La Þ ð4:5Þ
212
Po (64%) 300 ns 8.95 (100%) — —
208
Tl (36%) 3.06 min — 1.291 (24%) 0.511 (25%)
1.524 (22%) 0.583 (87%) where CTn(0) is the thoron activity concentration at
1.801 (49%) 0.860 (13%) the room surface (x ¼ 0), Da is its diffusion coefficient
2.614 (100%) in air, and La ¼ (Da/lTn)1/2 is the diffusion length of
thoron in air. La has been estimated to range from
about 3 to 20 cm depending on the effective value of
4.3.1 Steady-State Activity Concentrations of Da used. In any case at 1 m from a wall, the calculated
Radon and Thoron Gases in Indoor Air thoron activity concentration will be less than 1% of
its value at the wall surface. However, measurements
Assuming the sources contributing to the radon ac- of Doi et al. (1994), Harley et al. (2010), and Reddy
tivity concentration in a room are constant, the et al. (2012) do not support such a rapid decrease. In
indoor radon activity concentration may be described their studies, thoron activity concentrations decreased
by the following equation: only to 10–40% at 1 m distance from the source, most
likely due to circulating air currents.
dCiRn =dt ¼ ERn þ vCoRn  ðlRn þ vÞCiRn ð4:1Þ

where C iRn is the indoor activity concentration of

radon, CoRn the outdoor activity concentration of 4.3.2 Steady-State Activity Concentrations
radon, lRn the decay constant of radon, v the ventila- of Radon Progeny in Indoor Air
tion rate, ERn the total volume-specific entry rate of Radon progeny in a room may be present in three
radon to the indoor spaces from soil gas, exhalation of forms. These are:
building materials, and release from water.
Under steady-state conditions where a balance (a) Unattached radon progeny: Immediately after
exists between the supply and removal rates of radon formation following alpha decay of radon, the
progeny so formed are mainly positively
CiRn ¼ ðERn þ vCoRn Þ=ðlRn þ vÞ ð4:2Þ charged. They rapidly form charged or neutral
clusters with water vapor molecules and trace
In dwellings in temperate climate, v is generally in gases that are present in the air. These un-
the range of 0.1–1.5 h21, values that are much attached radon progeny are nanoparticles in the
greater than lRn ¼ 7.6  1023 h21. Thus, Equation size range 0.5 –5 nm (Sections 4.6 and 7.5.2).

58
 Characteristics and Behavior of Radon and Radon Progeny

They play an important role in lung dosimetry Under steady-state conditions, we therefore obtain
(Section 3.8). for the unattached and attached progeny activity con-
(b) Attached radon progeny: Unattached radon centration, respectively:
progeny will attach to aerosol particles present
in the air, forming attached radon progeny. The Cj u ¼ ðlj Cj1u þ lj Rj1 Cj1 a Þ=ðlj þ b Z þ qu þ vÞ
size distribution of attached progeny depends on ð4:8Þ
the aerosol particle size distribution, the trace
gases, and the attachment coefficient. and
(c) Deposited radon progeny: Both unattached and
attached radon progeny deposit on room sur- Cj a ¼ ðvCj a;o þ ð1  Rj1 Þlj Cj1 a
faces. Depending on the characteristics of the þ b Z Cj1 u Þ=ðlj þ qa þ vÞ ð4:9Þ
surfaces on which they deposit subsequent
progeny produced by alpha decay on the surface with C0a ¼ 0 and C0u ¼ C0 (radon gas activity concen-
may give rise to alpha recoil of implanted radon tration).
progeny in the surface. The presence of such Here, Rj21 is the recoil factor of the j21th

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

implanted progeny activities in glass forms the attached radon progeny. This is the desorption frac-
basis for one type of retrospective radon meas- tion of the jth progeny so formed from the aerosol
urement technique (Section 5.4). particle surface following an alpha decay. In other
The different physical mechanisms affecting the be- words, Rj21 is the probability that an attached radio-
havior of radon progeny in a room are illustrated in active atom, j –1, desorbs from its host following
Figure 4.3. alpha decay. For typical indoor aerosols from consid-
In the steady state, well-mixed room model, the erations of recoil energy and particle size, Rj21 can
following equations may be used to describe the be expected to be close to 1, but the experimentally
behavior of both unattached and attached radon determined value of Rj21 ¼ 0.8 is generally used
progeny j with decay constant lj (see Figure 4.3). (Mercer, 1976). For beta decay, the recoil factor is
The activity concentration of the jth radon progeny negligible, i.e., R2 ¼ R3 ¼ 0. The term b Z in
in the unattached state, Cju, may be written as Equations (4.8) and (4.9) is the attachment rate (X)
of radon progeny to aerosols where b is the mean
dCj u =dt ¼ lj Cj1u þ lj Rj1 Cj1a  ðlj þ X þ qu þ vÞCj u attachment coefficient and Z is the number concen-
tration of the aerosol particles (McLaughlin, 1972)
ð4:6Þ
[see Equation (4.14)]. The parameter q a and q u re-
present the attached and unattached radon progeny
The activity concentration of the jth radon progeny in
deposition rates to room surfaces, respectively. The
the attached state, Cja, may be written as
parameter Cja,o is the activity concentration of
dCj a =dt ¼ vCj a;o þ ð1  Rj1 ÞlCj1 a þ XCj1u the attached radon progeny j in outdoor air. It is
assumed that the unattached radon progeny in
 ðlj þ qa þ vÞCj a ð4:7Þ outdoor air is removed by plate-out (surface depos-
ition) during ventilation.
In rooms, deposition of radon progeny on walls
and furniture is a major determinant of steady-state
activity concentrations in indoor air. Deposition can
be characterized by the deposition velocity vg:

vg ¼ wðdÞ=Zðz; dÞ ð4:10Þ

where w(d) is the number of particles with diameter d

deposited per unit surface area and time and Z(z,d) is
the concentration of particles with diameter d at
height z above a surface. In the literature, the depos-
ition velocity is often normalized by the friction vel-
ocity u*: vg þ ¼ vg/u*, where u* depends on the air
velocity profile and the roughness of the surface.
Normalized deposition velocities as functions of par-
Figure 4.3. Schematic representation of the behavior of radon ticle diameter are displayed in Figure 4.4 for different
progeny in an enclosed space. Adopted from NA/NRC (1991) and rough and smooth surfaces (Porstendörfer, 1994).
Porstendörfer (1994). Two observations can be made: (i) deposition

59
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

velocities increase with increasing roughness of the

surface, and (ii) the minimum of deposition lies
between particle diameters of 0.2 and 0.7 mm for all
surfaces.
For a well-mixed room air, the deposition rates q for
attached and unattached radon progeny [Equations
(4.6–4.9)] are related to the deposition velocities vg
[Equation (4.10)] via the surface to volume ratio S/V
by (Porstendörfer, 1994):

q ¼ vg S=V ð4:11Þ

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

4.3.3 Radon Progeny Parameters Affecting
Lung Dosimetry
As described in more detail in Section 3 of this
report, estimates of lung doses due to the inhalation of
short-lived radon progeny are strongly dependent on Figure 4.4. Deposition velocity vg normalized by the friction
velocity u* as a function of aerosol particle diameter for different
the choice of input parameters and other model smooth and rough surfaces: grass (solid line), filter paper (dot-dash
assumptions. This leads to some uncertainty in esti- line), and Al foil (broken line). Taken from Porstendörfer (1994).
mated absorbed doses. In this regard, two of the most
important parameters of airborne indoor air progeny
are the equilibrium factor, F (Section 4.5), and their
activity fraction. Attached activity deposition in the
size distributions (Sections 3.7 and 4.7). Measurements
bronchial tree is typically only a few percent, whereas
in indoor air in several countries have shown F values
about 40% of the inhaled unattached progeny deposit
in dwellings to lie generally between 0.2 and 0.8. In
in the bronchial tree.
contrast to this, for a number of physical reasons,
the F-value in outdoor air is usually higher and in
the range 0.6–0.8. Indoor F values outside a range of
0.2–0.8 are very rare and would only arise in either ex- 4.4 Airborne Radon Activity Concentrations
tremely low or high aerosol concentrations, respectively,
4.4.1 Radon in Homes
or under other unusual indoor air conditions not com-
monly met in most dwellings. Based on actual measure- For epidemiological and health risk reasons,
ments, both ICRP (1993a) and UNSCEAR (2008) have radon activity concentrations were measured in sig-
adopted a typical worldwide F-value of 0.4 for indoor nificant numbers of dwellings on all continents, in-
air and 0.8 for outdoor air. As most large-scale surveys volving almost 5 billion people in 67 countries
of radon exposure are based on measurements of radon (Chambers and Zielinski, 2011). Currently, the data-
gas and not of its progeny, this value of 0.4 is generally base on radon activity concentrations in dwellings is
used for practical reasons in estimating lung doses to growing. A summary of these data was presented in
the general population. UNSCEAR and WHO publications (UNSCEAR,
The size distributions of both unattached and 2008; WHO, 2009). In the conclusions of UNSCEAR,
attached indoor radon progeny and their partitioning the worldwide geometric mean value of radon activ-
between these two modes are also parameters of im- ity concentration is 37 Bq m23 (GSD: 2.2). The
portance in dose estimation. This is because of the minimum value is below 10 Bq m23 (Egypt, Cyprus,
different deposition patterns of these modes in the re- Australia), while the higher values of 243 Bq m23
spiratory tract. The unattached (sometimes called (Spain) and of 2745 Bq m23 as an arithmetic mean
ultrafine) mode lies in the 0.5–5 nm size range, while for Iran are measured in high background areas. In
the attached mode, reflecting the indoor aerosol size surveys of some thousands of dwellings monitored,
distribution, is generally found in the 100-300 nm sites above 1000 Bq m23 are expected, although
size range (Sections 3.7 and 4.7). Although the frac- their frequency is less than 1% (Hamori et al., 2006;
tion of unattached activity is usually low, neverthe- Tomášek et al., 2001; Žunic et al., 2007).
less because of its high deposition efficiency in Annual mean indoor radon activity concentrations
human airways, it makes a disproportionate contri- depend on a large number of factors (Section 7). If
bution to lung dose in comparison to the attached these factors act multiplicatively and independently,

60
 Characteristics and Behavior of Radon and Radon Progeny

Table 4.3. Long-term indoor radon surveys in some European Countries from National Summary Reports

Country and No. of dwellings Measurement period and Mean value Geom. Mean Percent over Max
population (millions) sampled approx. duration (Bq m23) (Bq m23) 200 Bq m23 (Bq m23)

Czech Republic (10.2) .150 000 1984–present 1 year 140 110 12–18 25 000
Denmark (5.5) 3120 1995–1996 1 year 64 53 2.9 590
Finland (5.2) 2866 2006–2007 1 year 96 62 10.4 33 000
Germany (82.4) .50 000 1978–2003 1 year 49 37 1.6 .10 000
Ireland (4.2) 11 319 1992–1999 1 year 89 57 7.5 1924
Italy (58) 5361 1989–1998 1 year 70 52 4.1 1036
UK (61) 450 000 1980–2005 3–12 months 20 14.9 0.5 17 000

Source: http://rem.jrc.ec.europa.eu/RemWeb/publications/EUR_RADON.pdf

then the distribution of radon activity concentra- Table 4.4. Indoor radon activity concentrations in some

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

tions can be approximated as lognormal (Miles, Non-European countries
1998). Some publications suggest that measured
Country (pop.in Mean Geom. Geom. Maximum
radon data fit a lognormal distribution (Gunby et al., millions) value mean STD (Bq m23)
1993; Nero et al., 1994; White et al., 1992), but in (Bq m23) (Bq m23) (Bq m23)
other literature, this was not confirmed (Goble and
Socolow, 1990; Kies et al., 1996). It now seems that Argentina (39) 35 25 2 211
the “log-normal mysticism” (Tóth et al., 2006) Australia (18) 11 8 2.1 420
Canada (30) 34 14 3.6 1720
remains an open question in environmental statis-
China (1316) 44 34 — 586
tics, despite many endeavors aimed at clarification India (945) 57 42 2.2 210
(Bossew, 2010) (see Section 6.3.5). Japan (125) 16 13 1.8 310
A summary of indoor radon surveys carried out in a South Korea (49) 53 43 1.8 1,350
number of European countries is given in Table 4.3 USA (250) 46 25 3.1 Not listed
(Dubois, 2005). It is important, however, to note that
Source: UNSCEAR (2000).
the survey designs were not the same for each
country. Dwellings were selected in some countries on
the basis of population density. Where this approach were carried out in areas where high radon levels
was used, more measurements were made in large were expected on the basis of geological characteris-
centers of population than in low population rural tics. It should be noted that representative radon
areas. In this approach, estimates can be made of the surveys have not yet taken place in countries with
collective exposure and health risk of the general the largest populations such as China and India.
population in a country. Such information is useful to
the relevant authorities for the development of nation-
4.4.2 Radon in Workplaces
al radon control strategies. National surveys in other
countries were made on a geographical basis where Indoor workplaces include, for example, schools,
the strategy was to achieve the same density of dwell- hospitals, post offices, jails, shops, cinemas, office
ing sampling per unit area irrespective of the popula- buildings, and common workshops. The primary
tion density distribution. The database from a workplaces where radon may cause health problems
carefully designed survey can be used in conjunction are underground mines, in particular uranium mines,
with other national databases, to produce both waterworks in the case of sufficiently high radon
population-weighted data and geographically based levels in the water, and industrial buildings with spe-
data. Notwithstanding the differences in European cific work practices and ventilation conditions. More
survey designs, the data presented in Table 4.3 give a information on radon in workplaces will be given
reasonably accurate overview of average radon activity in Section 6.5. For a significant proportion of the
concentrations in contemporary European dwellings. population, the time spent in buildings is increased
In Table 4.4, a summary is given of indoor radon by a modern life style. Therefore, public buildings
data for a number of large non-European countries with a high density of people (e.g., kindergartens,
(UNSCEAR, 2000). The maximum radon activity schools, hospitals) have been monitored for radon.
concentration values quoted in Table 4.4 are the Table 4.5 gives some typical values of radon activity
maximum values found in the national surveys. concentrations measured at selected workplaces in
Much higher indoor radon activity concentrations different countries for non-uranium mines, caves, and
are often found in targeted surveys because they spas. In several workplaces, the radon activity

61
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 4.5. Examples of radon activity concentrations at some selected workplaces in the world taken from UNSCEAR (2008) and other
sources

Country Radon (Bq m23) Interval Site Device Ref

Mine
Hungary 817 (575– 957)a One year Manganese PADC Kávási et al. (2009)
Kosovo Metohia (200–800) Lead/Zinc PADC Jakupi et al. (1997)
Pakistan 192 (121– 408) 30 d, 6 Coal CN Qureshi et al. (2000)
Argentina 4800 (1800– 6000) Summer Gold/Touristic PADC Anjos et al. (2010)
Iran 3500 (1000– 10 400) 6–8 h Turquoise Pulse ionization Fathabadi et al. (2006)
2370 (580–4700) Bauxite chamber
990 (500– 1770) Lead
220 (40– 590) 2 Coal
137 (15– 630) 2 Manganese
150 (50– 390) Poshpate
19 (11– 34) Lead/Zinc

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

5 (2– 10) Chromites
Turkey 172; 340b (50 –587) 2 months 3 Lignite PADC Çile et al. (2010)
Brazil 64; 4365b (18.9– 19 351) 7–47 h Copper Active Malanca and Gaidolfi
(1997)
UK 65 613 (27– 1244) 2–82 weeks 5 Coal Scintillation cell Page and Smith (1992)
India 5.3 Coal Scintillation cell Mishra and Subba
1244.7 Copper SSNTD Ramu (1988)
19 Gold
186.1 Lead/Zinc
38.6 Manganese
78.8 Mica
Slovenia 1419 (590–76 900) Grab sampling Mercury Scintillation cell Kobal et al. (1990)
658 (185– 1050) Lead
256 (30– 465) 5 Coal
Caves
UK (27– 7800) 1 year Creswell Crags PADC Gillmore et al. (2002)
Venezuela (100–80 000) 1–6 months Guacharo, PADC, CN Sajó-Bohus et al. (1997)
AlfredoJahn, so on
Spain 3562 (186–7120) 1 year Altamira Scintillation cell Lario et al. (2005)
Saudiarabia 74– 451 6 months Al-Somman Platea PADC Al-Mustafa et al. (2005)
Greece 88 000 Petralona SSNTD Papastefanou et al.
(2005)
Australia 500; 795b (26330) 6 months Winter/spring 57 caves SSNTD Solomon et al. (1996)
Turkey 1918.8 (20– 5833) 2 months Summer Gökgöl SSNTD Aytekin et al. (2006)
657 (304– 885) Cehennemağzı
Czech 4000– 5000 1 year SSNTD Thinová and Burian
(2008)
Hungary 4600 (500–12 400) 1 year Hospital PADC Somlai et al. (2007a)
Brazil 5300 (1900– 8400) 214 d Santana SSNTD Alberigi et al. (2011)
Baths
Croatia 40.3 (10.9– 109) 5 months 9 spas CN Radolić et al. (2005)
Greece 149; 2380b (27-8263) 2–3 d 5 spas Pulse ionization Vogiannis et al. (2004a)
chamber
Hungary 94; 1511b (54– 2040) 1 year 1 spa PADC Kávási et al. (2011)

a
Arithmetic mean (minimum – maximum).
b
Minimum arithmetic mean; maximum arithmetic mean (minimum–maximum).
PADC, poly-allyl diglycol carbonate; CN, cellulose nitrate; SSNTD, solid-state nuclear track detector.

concentration is below 100 Bq m23 (lead/zinc mine, The radon activity concentrations in caves can
Iran; coal, gold, manganese, and mica mines, India; reach high levels because of the low air exchange
spa, Croatia), but for some caves and mines, the radon rate due to poor ventilation. The caves are natural
activity concentration is more than 1000 Bq m23. For formations with a special microclimate and to pre-
example, radon activity concentrations of about serve this microclimate, forced ventilation cannot be
80 000 Bq m23 have been measured in a mercury applied. In mines, forced ventilation is compulsory
mine (Slovenia) and in caves (Venezuela and Greece, due to occupational health and safety requirements,
Petralona). and as a result, a significant reduction in radon

62
 Characteristics and Behavior of Radon and Radon Progeny

activity concentrations can be achieved. However, in relative to residential exposure. Application of the
areas with inefficient ventilation, high radon levels Finnish occupancy factors to the Italian results
can be expected. would give a value of 15%.
Forced ventilation is very important when operat- A representative survey in both workplaces and
ing a bath in natural spas, due to the high relative homes, based on random population sampling, was
humidity. The area of a bath is much smaller than carried out in Finland (Mäkeläinen et al., 2005). The
that of a mine or cave, and the source of the radon, survey included simultaneous measurements in the
primarily the spa, is localized and therefore, the homes and workplaces of the 520 participants. In
ventilation efficiency is high, resulting in relatively addition, 123 participants had a personal radon
low radon activity concentrations. monitor in their pocket or handbag during the
survey. The questionnaire included a section on time
spent at work, at school and outdoors, and possibly
4.4.3 Comparison of Radon in Homes and
time spent at a summer residence (recreational
Indoor Workplaces
dwelling) during the previous year. The measure-
Radon exposures in homes have been monitored ments covered a 2-month period, starting from 15

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

in numerous studies and extensive national surveys February to 9 April 2001. Table 4.7 shows the key
(Section 6.3.2.6). Radon progeny activity concentra- results of the workplace survey. The assessment of
tions in mines are subject to regulatory control and radon activity concentration during the daily
are well measured. Surveys in workplaces are working hours would be biased if only day-and-night
sparse and radon activity concentrations in typical integrating alpha-track dosimeters were used.
workplaces are not well known. Only in the special Therefore, the radon activity concentration in the
cases of schools, spas and caves have representative workplace during working hours was obtained by
surveys been carried out. Section 4.4.2 describes multiplying the radon activity concentration mea-
examples of workplace surveys. sured with alpha-track dosimeters by a correction
Table 4.6 shows example results of occupancy- factor of 0.5, defined as the ratio of the average
weighted average radon activity concentrations from radon activity concentration during working hours
national or regional workplace surveys, including a to that of the whole week (Annanmäki et al., 1996).
comparison with residential surveys. National The arithmetic and geometric means of the dwell-
surveys aiming at a good representativeness in the ings, 104 and 68 Bq m23, are in good agreement
whole country have been carried out in Japan and with the earlier national survey. The arithmetic and
Finland. The Italian survey is based on random geometric mean workplace activity concentrations
samples of workplaces and dwellings in the Tuscany of the participants based on integrating detectors
area. were 30 and 20 Bq m23, which is about 30% of the
The results of Table 4.6 indicate that the national corresponding activity concentration in dwellings
average long-term radon activity concentration at (Table 4.7).
workplaces may be higher (Italy and Japan) or lower A 1-week continuous radon monitoring with 1 h
(USA, Finland) than the corresponding residential measurement period was carried out in 13 randomly
exposures by a factor of 3. These national average selected workplaces in order to determine the ratio
radon activity concentrations are affected by the na- of the average radon activity concentration during
tional occupancy times at workplaces and the radon working hours to that of the whole week. A later
activity concentration during working hours. For analysis of these data gave a mean ratio of 0.55. This
example, in the case of Finland, occupancy times of is in agreement with the factor of 0.5 used in the
0.73 at home and 0.14 at workplaces, together with study. The minimum and maximum ratios were 0.08
the average radon activity concentration listed in and 0.87, the median was 0.58 and the 25% and 75%
Table 4.6, result in a workplace exposure of 5.5% quartiles were 0.3 and 0.8. The mean proportion of

Table 4.6. Workplace mean (N), median (Bq m23) and dwellings mean (N), median (Bq m23) in national or regional radon surveys in
worplaces and dwellings (N is the number of measured sites).

Country Type of workplace Workplace mean (N), Dwellings mean (N), Reference
median (Bq m23) median (Bq m23)

USA, NM Offices 24.3 (65), 18.5 75.0 (47), 55.5 Whicker and McNaughton (2009)
Japan All 20.8 (940), 15.3 15.5, (705) 11.7 Oikawa et al. (2006) Sanada et al. (1999)
Italy, Tuscany All —, 43 (1159) —, 32 (1541) Bucci et al. (2011)
Ireland Schools 93 (4000), — 91 (91019), — Long and Fenton (2011)
Finland All 30 (333), — 104 (520), — Mäkeläinen et al. (2005)

63
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 4.7. Basic statistics of the radon activity concentrations of the Finnish national home and workplace survey, measured using
2-month alpha-track detectors (Mäkeläinen et al., 2005)

Living Bed-room Dwelling Work-place Personal Occupancy-weighted

room mean monitor radon activity concentration

Number 447 492 520 333 339 309

Arithmetic mean (Bq m23) 99 103 104 30 85 88
Standard deviation (Bq m23) 151 165 159 40 122 118
Geometric mean (Bq m23) 64 65 68 20 58 62
Geometric standard 2.4 2.4 2.4 2.6 2.2 2.2
deviation

the time Finns spent at home or at work/school/

public buildings or outdoors was 0.73, 0.14, and

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

0.09, respectively. Figure 4.5 exhibits the radon ac-
tivity concentration measured using the personal
monitor versus the occupancy-weighted calculated
radon activity concentrations.

4.4.4 Thoron in Homes

In most buildings, doses from thoron (220Rn)
progeny are a small proportion of those from radon
(222Rn) progeny due to the generally low thorium ac-
tivity concentrations of the building materials.
However, recent measurement campaigns reported
increased thoron activity concentrations in homes Figure 4.5. Radon activity concentrations measured by personal
worldwide (McLaughlin, 2010). Indeed, potential monitors versus occupancy-weighted mean radon activity
alpha-energy concentration levels from 220Rn and its concentration calculated using the participant’s occupancy factors
progeny were found to be comparable with those from from the questionnaire study and the radon activity concentrations
measured at her/his dwelling and workplace in the Finnish national
radon progeny (Bi et al., 2010). Table 4.8 gives a
workplace survey, based on 123 observations (Mäkeläinen et al.,
summary of some long-term measurements of indoor 2005).
thoron gas and its progeny (McLaughlin, 2010).
concentration (PAEC) as the existing non-equilibrium
mixture. It is estimated from
4.5 Equilibrium Factor
EEC ¼ F  222 Rn gas activity concentration ð4:12Þ
Because radon progeny in the air can be removed
by deposition on surfaces and ventilation, the activ-
Many of the published determinations of the equilib-
ity concentrations of the short-lived radon progeny
rium factor were based on short-term samples and
in the air are not in equilibrium with that of the
high sampling air flow that may have affected the
radon gas. This is quantified by the equilibrium
equilibrium in interior spaces. The determination of
factor, F, which is a measure of the degree of disequi-
the equilibrium factor requires measurement of both
librium between the radon gas and its progeny.
radon gas and its progeny. Radon can be measured
The inhaled decay products and not radon gas
using passive detectors, but the progeny usually
deliver the majority of the alpha particle dose to the
require electrically operated equipment (see Sections
bronchial airways (see Section 3). Thus, the equilib-
5.2 and 5.3). If real-time measurement of the Working
rium factor is of dosimetric importance because it is
Level (WL) is made along with radon gas measure-
used to estimate the progeny activity concentration in
ments, the equilibrium factor F can be estimated from
air when measurements of radon and not progeny are
made. The equilibrium factor, F, is defined as the ratio F ¼ ðWL  3700Þ=222 Rn gas activity
of the equilibrium equivalent activity concentration ð4:13Þ
(EEC) to the radon gas activity concentration. concentration ðBq m3 Þ;
The EEC is the activity concentration of radon
gas, in equilibrium with its short-lived progeny, based on the original definition of the WL, i.e., 1 WL
which would have the same potential alpha energy equals the PAEC associated with radon progeny in

64
 Characteristics and Behavior of Radon and Radon Progeny

Table 4.8. Recent long-term measurements of indoor thoron and its progeny

Country Number of dwellings Thoron activity conc. (Bq m23) Mean (Min–Max) EETCa (Bq m23) Mean (Min–Max)

Korea (2007) 450 40 (?–731) 0.89 (? –5.82)

Canada (2007) Ottawa 93 53 (5 –924)
Canada (2009) Winnipeg 117 34 (5 –297)
Hungary (2007) 72 98 (1 –714)
China (2004) Shanxi 193 153 (10–865) 1.6 (0.3–4.9)
China (2006) 102 351 (0–1471) 2.77 (0.8–5.7)
Gansu 49
Gejiu Yunnan 29 3297 (39 –7908) 10.2 (2–23.9)
Ireland (2009) 205 22 (0 –174) 0.47 (0.1–3.7)
Serbia (2006) 137 160 (2–945)

a
EETC is the equilibrium equivalent thoron concentration.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Table 4.9 Summary statistics for measurements of the equilibrium factor F mean and (range), taken from Harley et al. (2012a)

Measurements F

Reineking and Porstendörfer (1990) 79 measurements in 10 rooms 0.30 + 0.1 (0.15, 0.49)
Hattori and Ishida (1994) 4500 measurements, 2 nuclear power plants (high ventilation) 0.30 + 0.1 (0.1, 0.6)
Hopke et al. (1995) 143 samples in 2 houses with a smoker 0.48 + 0.11 (0.25, 0.80)
Hopke et al. (1995) 422 samples in 5 non-smoking houses 0.38 + 0.17 (0.11, 0.97)
Clouvas et al. (2003) 4-h measurements for 29 weeks in a lab 0.62 + 0.09 (0.46, 0.82)
Clouvas et al. (2003)4-h measurements in 25 apartments 0.47 + 0.09 (0.2, 0.7)
Chen and Marro (2011) Grab samples in 12 576 houses 0.54 + 0.15 (0.20, 0.82)
Harley et al. (2012a) 3 month measurements in 2 labs and 6 houses 0.75 + 0.12 (0.59, 0.95)

equilibrium with 100 pCi l21 of 222Rn gas. The potential dependent relationship between thoron gas and its
alpha energy is rounded to 1.3  105 MeV l21 of air progeny (Tokonami, 2010).
(2.08  105 J m23).
Some selected values of F are given in Table 4.9.
UNSCEAR (2008) selected central values for F of 0.4
(indoors) and 0.6 (outdoors). The indoor value was 4.6 Unattached Fractions
mainly based on measurements in dwellings in the
After decay of radon gas, the freshly formed radio-
USA (Hopke et al., 1995) and in India (Ramachandran
nuclides react rapidly with water vapor and possibly
and Subba Ramu, 1994). These selected average
H2SO4 to grow from 0.5 nm to clusters of about
values may be subject to change as new measurements
1.2 nm diameter (Andreae, 2013; Kulmala et al.,
are made. If dose is calculated from radon gas mea-
2013; UNSCEAR, 2006). Growth of the particle
surements, better data (large scale measurements)
increases to about 2 nm with ammonia, organic
regarding the equilibrium factor will reduce the uncer-
amines, and oxidized organic molecules and from bio-
tainty in the dose estimates.
genic hydrocarbons stabilizing the growing clusters;
In rooms with additional aerosol sources, such as
the clusters then can grow to 5 nm (Andreae, 2013;
cigarette smoke, values of the equilibrium factor are
Kulmala et al., 2013; Porstendörfer, 2001). These are
usually higher than in clean rooms. At higher aerosol
referred to as the unattached progeny. The un-
particle concentrations, the unattached fraction is
attached progeny diffuses rapidly, attaches readily to
lower as more radon progeny are attached. Since
other aerosols and surfaces and deposits very effi-
attached radon progeny deposit with a much smaller
ciently (100%) in the respiratory tract if inhaled.
probability on room surfaces than unattached
Approximately 60% of the inhaled unattached
progeny, higher particle concentrations lead to higher
progeny deposits in the extrathoracic region and
equilibrium factors (Porstendörfer, 1994).
about 40% in the bronchial tree.
For thoron progeny, an F-value is a less useful
The degree of attachment to aerosol particles
quantity, since thoron activity concentrations in air
depends on the ambient aerosol concentration:
vary significantly with position in a room due to its
very short half-life (55.8 s), leading to a position- X¼bZ ð4:14Þ

65
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

where X is the attachment rate (s21), b the attach- 4.6.1 Unattached Fraction, fp, for Radon
ment coefficient (m3 s21), and Z the aerosol number (222Rn) Progeny
concentration (m23) (Porstendörfer, 1994).
The unattached fraction, fp, is defined as the frac-
The radon decay product 218Po is formed as a posi-
tion of the PAEC of the short-lived progeny that is
tively charged atom from radon decay. A fraction of
218 not attached to the ambient aerosol (note: the un-
Po atoms are neutralized and as stated above, all
218 attached fractions f218 and f214 discussed in the pre-
Po rapidly form molecular clusters of about 0.5 to
vious section refer to the individual 218Po and 214Pb
5 nm diameters with water vapor or other constituent
activity concentrations). The magnitude of fp pri-
molecules. Within 1–100 s, the unattached 218Po may
marily depends on the number concentration of par-
attach to the local aerosol particles. Subsequently,
ticles of the ambient aerosol, Z, and can be
unattached 214Pb atoms may be formed by the decay
estimated with the semi-empirical equation given by
of unattached 218Po or by the recoil of a fraction of
218 Porstendörfer (2001):
Po from attached 218Po due to the alpha decay
(note: since 214Pb is a beta emitter with a half-life of
414
26.9 min, no unattached 214Bi and 214Po are formed). Radon ð222 RnÞ progeny : fp ¼ ð4:17Þ

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Z ðcm3 Þ
El-Hussein (1996) determined an average 218Po
attachment rate of 0.025 s21in three rooms with dif-
ferent aerosol sources and ventilation rates. This Porstendörfer (2001) measured fp using a single
leads to a half-life of a 218Po atom in the unattached screen diffusion battery with 50% penetration for
state of 27 s. 4 nm diameter particles (Section 5.3.3). A condensa-
The steady-state activity concentration of un- tion nuclei counter was used to measure Z for par-
attached 218Po and 214Pb is mainly dependent on the ticle diameters greater than 5 nm. Equation (4.17)
aerosol concentration and can change significantly agrees fairly well with data for 2000 , Z , 7  105
due to normal aerosol concentration fluctuations cm23 (Porstendörfer, 2001). At lower particle con-
(El-Hussein, 1996). Raabe (1969) calculated the at- centrations (Z , 400 cm23), the agreement with the
tachment coefficient and reported a value of about data is poor (Cheng et al., 1997). Further, the above
1026, while El-Hussein (1996) measured an average equation may underestimate fp in situations where
value of 1025. the radon progeny are far from equilibrium, as is the
Neglecting the removal processes of ventilation case in some modern mines, which are ventilated at
and surface deposition, it can be shown that the a high rate to reduce radon activity concentrations
steady-state fraction of unattached 218Po activity is (Cavallo et al., 1999).
(Porstendörfer, 1994; Raabe, 1969): The fp values are between 0.03 and 0.08 for “normal”
indoor air quality with aerosol particle concentration
f218 ¼ l218 =½b Z þ l218  ð4:15Þ in the range (5–15)  103 cm23 (Porstendörfer, 2001).
Measurements of fp for 222Rn progeny in indoor work-
and of 214Pb is places such as schools and offices show a wide range of
values, typically between 0.03 and 0.15 and with some
values greater than 0.20 (Hattori and Ishida, 1994;
f214 ¼ l214 ðl218 þ R1 b ZÞ=½ðl218 þ b ZÞðb Z þ l214Þ 
ð4:16Þ

where, f218 is the ratio of 218Po (unattached)/total

218
Po; f214 the ratio of 214Pb (unattached)/total 214Pb;
Z the aerosol particle concentration (cm23); b the at-
tachment coefficient for any species, charged or un-
charged (s21) (1025); b Z the attachment rate X
(s21); l218, 214 the decay constant for 218Po and
214
Pb; R1 the recoil factor for 218Po.
For example, using El-Hussein coefficients, for an
aerosol particle concentration of 103 particles cm23,
f218 ¼ 0.00379/(0.01 þ 0.00379) ¼ 0.27 or about 27%
of the 218Po activity concentration. If the aerosol con-
centration were to increase to 2  103 cm23, the
steady-state unattached fraction of 218Po would be Figure 4.6. Unattached fractions as a function of aerosol
0.16 or about 16%, a reduction by a factor of almost 2 concentration. Measured attachment coefficients for 218Po and
214
in unattached activity concentration (Figure 4.6). Pb are taken from El-Hussein (1996).

66
 Characteristics and Behavior of Radon and Radon Progeny

Hattori et al., 1995; Porstendörfer, 2001; Tokonami variation between F and fp based on measurements in
et al., 1996a; Vaupotič, 2008a; Yu et al., 1998). Similar indoor air (Marsh et al., 2002). This negative correl-
values have also been measured in dwellings (Chen ation between F and fp has also been observed in a
et al., 1998; El-Hussein, 2005; Hopke et al., 1995; Huet tourist cave (Vaupotič, 2008a). The correlation can be
et al., 2001a; Kojima and Abe, 1988; Kranrod et al., explained as follows for conditions where the ventila-
2009; Mohamed, 2005; Reineking and Porstendörfer, tion rate is not too high: when the aerosol particle con-
1990; Tokonami et al., 1996b; Vargas et al., 2000; Yu centration is high, the unattached fraction is low, and
et al., 1996). In working places with additional aerosol the equilibrium factor is relatively high as more of the
sources due to technical processes, combustion and radon progeny are attached and stay in the air. More
human activities, the particle concentration can be stay in the air because plate-out rates (i.e., deposition
high (. 4  104 cm23) and, as a consequence, the fp rates) for the aerosol-attached nuclides are significant-
value is low (around 1% or less) (Porstendörfer, 2001). ly lower than those for the unattached nuclides
However, fp is greater than 0.10 for poorly ventilated (Figure 4.4, Porstendörfer, 1994). This is also illu-
rooms (ventilation rate , 0.5 h21) without additional strated in Figure 4.8, which shows the relationship
aerosol sources, rooms with an operating air cleaner, between F and fp as a function of the attachment rate

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

and poorly ventilated underground caves. In such (X) and the particle concentration (Z) (Porstendörfer,
places, the particle concentration may be less than 4  1994; Porstendörfer and Reineking, 1992).
103 cm23. For example, El-Hussein (2005) measured fp Taking account of this negative correlation
and particle concentrations in 25 rooms of different between F and fp, it has been shown that for indoor
houses with low ventilation rates (, 0.3 h21). The air, the radon gas concentration is a better index of
particle concentrations ranged from 1.2  103 to 5  dose than the PAEC under a range of aerosol condi-
103 cm23 and the corresponding fp values ranged from tions normally encountered (James et al., 1988;
0.02 to 0.22 with a mean of 0.09. Marsh and Birchall, 1998; Vanmarcke et al., 1989;
The relative activity ratios of unattached radon Vargas et al., 2000). UNSCEAR (2008) reported a
progeny (218Po:214Pb:214Bi) in indoor air have been similar range of dose factors with the values of fp
measured to be approximately 1:0.1:0 (Kojima and normally observed. On this basis and because of
Abe, 1988; Reineking and Porstendörfer, 1990). some practical considerations, such as less complex
Thus, most of the unattached activity is associated and expensive equipment, radon gas measurements
with the 218Po. More recently, El-Hussein et al. are generally carried out in homes and indoor work-
(1999) measured a higher activity ratio of about places. However, in mines with forced ventilation, a
214
Pb/218Po ¼ 0.5 for unattached progeny in closed consistent correlation between F and fp is unlikely,
room air. For the attached progeny, the measure- so control of radon exposure in mines is typically in
ments gave mean values around 214Pb/218Po ¼ 0.7 terms of PAE exposure.
and 214Bi/218Po ¼ 0.5. Although the relative activity The actual relationship between F and fp depends
concentrations of the individual radon progeny will primarily on the ratio of the deposition rates of the
vary with environmental conditions of exposure, the attached and unattached radon progeny (q a/q u). The
equivalent dose to the lung per WLM is relatively in- deposition rate q depends on the surface to volume
sensitive to these ratios (Marsh and Birchall, 2000).
In contrast, the dose per WLM is very sensitive to
the unattached fraction.
Although the unattached fraction is small com-
pared with the attached fraction, it has a dispropor-
tionately large effect on the bronchial dose because
of its greater deposition efficiency in the bronchial
region.

4.6.2 Correlation Between Equilibrium

Factor, F, and Unattached Fraction,
fp, for 222Rn
For 222Rn and its progeny in indoor air, F is negative-
ly correlated with the unattached fraction, fp (Chen
et al., 1998; Huet et al., 2001a; Marsh et al., 2002; NA/
Figure 4.7. Variation of unattached fraction with equilibrium
NRC, 1991; Tokonami et al., 1996b; Vanmarcke et al., factor in indoor air. Adopted from Marsh et al. (2002).
1989; Vargas et al., 2000; Vaupotič, 2007; Vaupotič and Measurements were carried by Huet et al. (2001a; 2001b),
Kobal, 2006). As an example, Figure 4.7 shows the Reineking and Porstendörfer (1990), and Vargas et al. (2000).

67
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

compared with that of the radon progeny. Therefore,

the fp value for the thoron progeny is lower than
that for the radon progeny under the same condi-
tions.
The semi-empirical equation of fp for thoron
progeny derived by Porstendörfer and his colleagues
is given by Equation (4.18) (Porstendörfer, 2001).
Reasonable agreement was obtained between
Equation (4.18) and the data of Tschiersch et al.
(2007), for 900 , Z , 3  104 cm23.
150
Thoron ð220 RnÞ progeny : fp ¼ ð4:18Þ
Z ðcm3 Þ

The unattached fraction of thoron progeny (220Rn)

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

for “typical” indoor air with aerosol particle concen-
tration of (5 –15)  103 cm23 is between 0.01 and
0.03 (Porstendörfer, 2001).
From measurements of outdoor and indoor
212
Pb/220Rn ratios, UNSCEAR (2008) has assumed
average equilibrium factors of 0.003 for outdoors
and 0.02 for indoors.

4.7. Aerosol Size Distributions

Aerosol particle size for radon decay products is
defined as their activity median diameter in air.
Specific definitions include thermodynamic diam-
Figure 4.8. The unattached fraction (fp) and the equilibrium
eter, the diameter of a spherical particle that has the
factor (F) in rooms with different aerosol sources as a function of
the attachment rate (X) and particle concentration (Z). Taken same diffusion coefficient in air as the particle of
from Porstendörfer (1994). interest and aerodynamic diameter which is ap-
proximately equal to the physical (volume equiva-
lent) diameter of the particle times the square root
ratio (S/V) and the deposition velocity (vg) [Equation of its “effective density.” The effective density is the
(4.11)], which in turn depends on the particle size ratio of the density to its shape factor. The particle
and the roughness of the surface (Figure 4.4). Since size distribution for radon progeny has predomin-
the ratio S/V is the same for both attached and un- antly two modes, the unattached diameter from 0.5
attached radon progeny, the ratio of the deposition to 5 nm and the attached from 100 to 450 nm (accu-
rates varies with surface roughness, as unattached mulation mode), with medians of about 1–2 and
radon progeny may experience a rougher surface 100 –300 nm, respectively. Sinclair et al. (1974) were
than the larger attached radon progeny. Thus, the the first to measure activity size distributions in
roughness of the surface areas of the room as well as indoor and outdoor air. NCRP (1984) summarized
the particle size distribution are factors that affect particle size distributions measured in New York
the relationship between F and fp. The ventilation and New Jersey residences and showed unattached
rate and the radon entry rate also affect this and attached median diameters of 1 and 125 nm, re-
relationship. spectively. Other attached modes are sometimes
observed; these are the nucleation mode, with dia-
meters of 10s of nanometer and the coarse mode,
4.6.3 Unattached Fraction, fp, for Thoron
with diameters of a few 1000s of nanometer. These
(220Rn) Progeny
are typically introduced when there is a specific
There are fewer published measurements of fp for source such as small aerosols released in cooking or
thoron progeny compared with radon progeny. large aerosols from dispersion activities. However,
However, because of the relatively long radioactive the nucleation mode has also been measured for an
half-life of the thoron decay product 212Pb (10 h), aged aerosol (i.e., without additional aerosols) in a
more of the 212Pb is likely to become attached moderately ventilated room and in closed rooms of

68
 Characteristics and Behavior of Radon and Radon Progeny

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 4.9. Relative size distribution of the PAEC of radon progeny in indoor air in closed rooms (Porstendörfer, 1996).

houses (Porstendörfer, 1996; Reineking et al., 1994)

(Section 7.5.2).
As an example of a typical measurement in a
typical home, Figure 4.9 shows the relative size dis-
tribution of the PAEC of radon progeny in indoor air
in closed rooms of homes, i.e., without additional
aerosol sources (Porstendörfer, 1996). For compari-
son, Figure 4.10 displays the relative activity size
distributions of the short-lived radon and thoron
progeny 214Po and 212Po in outdoor air, averaged
over a three-week measurement campaign (Gründel
et al., 2005).
Aerosol size distributions or radon progeny activ-
ity size distributions for specific atmospheres are
summarized in NA/NRC (1999a). Aerosol size distri-
butions were measured in rooms with a gas stove or
side-stream cigarette smoke (Li and Hopke, 1993),
218
Po and214Po activity size distributions in closed
rooms, with or without aerosol sources (Reineking
and Porstendörfer, 1986), and 218Po distributions in
rooms with cooking, cigarette smoke, or a kerosene
heater (Tu and Knutson, 1988). These and other
published activity size measurements are summar-
ized by Marsh et al. (2002) including those measure-
ments carried out in rooms with aerosols produced Figure 4.10. Relative activity size distributions of 214Po and 212Po
in outdoor air, averaged over a 3-week measurement campaign
by smoking, cooking, gas combustion, tiled stove
(Gründel et al., 2005).
heating, fumigating sticks, candle burning, and by
electric heaters.
Although the published activity size distributions codes have changed in the past two decades due in
to date appear to have similar activity median dia- part to radon reduction and energy-efficient techni-
meters (Section 7.5), some effort should be made to ques and new construction materials.
obtain recent measurements over a wide range of Volunteer studies and research studies with casts
conditions, especially in new buildings. Building of the human lung demonstrate that the particle size

69
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

of an inhaled aerosol along with breathing rate are assumed aerosol particle size distribution. Better pre-
the major factors controlling the site and amount cision in dosimetric modeling will be obtained with
of deposition in the respiratory tract (Section 3.7). more global information on particle size distribution
Therefore, bronchial dose models rely mainly on the in a variety of residential conditions.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

70
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv008
Oxford University Press

5. Principles of Radon and Radon Progeny Detection

Systems and Measurements

5.1 Radon and Radon Progeny Metrology certificates of the respective devices used in the meas-
and Quality Assurance of Measurements urement, and from international standards.
The economic consideration in the choice of a
5.1.1 General Aspects

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

measurement principle should include the costs of
Performing a measurement that is able to contrib- operation, man-power for the analysis of the data,
ute to a scientific or legal purpose implies that the and also quality assurance.
measurement should be suited to the purpose with The concepts of metrology and quality assurance
regard to relevant for radon and radon progeny measurements
as well as examples for the analyses of uncertainties
(1) its capability (e.g., measured quantity, sampling in the calibration by a primary radon activity stand-
type, long- or short-term measurement), ard and in interlaboratory comparisons are summar-
(2) its physical properties (e.g., maintained accur- ized in Appendix A.
acy, traceability, uncertainty, detection limit,
range of application), and
(3) cost efficiency (e.g., instrument cost, man power, 5.1.2 Comparisons of Radon Measurements
quality assurance expenses). Until now, numerous radon comparisons have been
performed worldwide. In the 1980s, the rising
Before it can be used as a basis for a further study, concern about radon-induced lung cancer triggered
for example, radon mapping or epidemiology, exist- the start of global comparison programs based on a
ing data should be assessed with regard to criteria 1 common radon atmosphere in which multiple radon
and 2. Data that will be assessed in the future must and radon progeny detection systems (active as well
be characterized precisely to choose an optimal as passive ones) were exposed. The Organisation for
method with regard to all three criteria. Economic Co-operation and Development (OECD)/
The design of a study has to yield all information Nuclear Energy Agency and the Commission of the
that is required for criterion 1, in particular, the us- European Communities (CEC) ran the “Programme
ability of the measurand or measurands, which can on radon and thoron dosimetry”, starting in 1983
be activity concentration, potential alpha energy (OECD, 1985). The responsibility for managing the
concentration, equilibrium factor, unattached and program was shared by the former Australian
attached fraction, exposure to radon, or exposure to Radiation Laboratory (ARL) for the Pacific region, the
radon progeny. Depending on the aim of the study, US Department of Energy (DOE), the former
each of these quantities alone or in combination can Environmental Measurements Laboratory (EML),
be an appropriate choice. and the US Bureau of Mines for North America, as
The sampling type can be either grab, continuous, well as the former National Radiological Protection
or integrating. This is normally a characteristic Board (NRPB) (now Public Health England) for
quality of a special type of device and sometimes Europe. Following this line of work, the International
linked to the duration of a measurement. Thus, for Atomic Energy Agency (IAEA) drew up an
example, solid-state nuclear track detectors (SSNTD) “International Radon Metrology Programme” together
are integrating devices used for long-term measure- with the CEC, beginning in 1992. Measurements were
ments (normally weeks or months). In the case of high performed at EML in 1990, 1992, 1994, 1995, and
levels of radon activity concentration, short-term mea- 1996 and at the US Environmental Protection Agency
surements with these devices can also be appropriate. (EPA) in 1994. These international programs were
The physical properties of a measurand must be flanked by multiple national efforts.
identified from the data of the producer of the device For example, the Environmental Measurements
(data from qualification or type tests), from calibration Laboratory of the US Department of Energy developed

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

a 2.82  2.82  2.4 m calibration chamber and per- compared with the reading of the system under test
formed calibrations based on a National Institute of for the calculation of the calibration coefficient.
Standards and Technology (NIST) Standard Reference After calibration, the system under test can now be
Material (SRM) 226Ra solution. The chamber calibra- used as a transfer or secondary standard. A calibra-
tions were performed using 2 l pulse ionization cham- tion by a secondary standard is based on a compari-
bers (Fisenne et al., 1990). son of the system under test to a reference
To summarize the many comparisons, it can be instrument (secondary standard) which was cali-
noted that these programs provided: brated in or traceable to a reference atmosphere in
the past. The radon activity concentration chosen
- a large database on different radon and radon for the point of calibration is established, and both
progeny measuring systems in different applica- systems are exposed to it. The reading of the refer-
tions, ence instrument and the system under test are
- a basis for the new development of detectors and observed simultaneously for the calculation of the
analytical methods, and calibration factor.
- routines for quality assurance. The second approach of a common reference at-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

mosphere for the conduct of the comparison has the
In view of the many radon quantities measured in advantage of providing the CRV in a shorter time,
parallel in these comparisons, they were not able— but it does not achieve uncertainties as small as the
and also did not intend—to provide modern metro- first approach.
logical information like uncertainty budget, trace- In the BIPM key comparison database, an example
ability information, and correlation analysis (ISO of each approach can be found: the European
17025) (ISO, 2005). Association of National Metrology Institutes
Due to the “Arrangement on the mutual recognition (EUROMET) has EUROMET.RI(II)-S1.Rn-222 (iden-
of the equivalence of national standards and of cali- tical to Euromet project 657) (Röttger et al., 2005;
bration certificates issued by national metrology insti- 2006) for the first approach using 1, 3, and 10 kBq
tutes (MRA)” in 1999, the situation for metrological m23 radon activity concentrations for the point of
comparisons changed drastically. Since then a com- calibration and the Euro-Asian Cooperation of
parison has to deal with a single quantity, the result National Metrology Institutions (COOMET) has
of each participant has to be given in the form of a COOMET.RI(II)-S1.Rn-222 (2009) for the second
value with an assigned uncertainty. Furthermore, the approach.
uncertainty budget for the calculation of the result The Euromet comparison demonstrated the
has to be included, as well as the information about ability of all 12 participants to perform a calibration
the traceability of each input quantity to national or of the radon activity concentration with an uncer-
international standards. These restrictions are funda- tainty below 12% for a level of confidence of 95%
mental for the assessment of the result of each partici- (Figure 5.1), indicating a satisfactory degree of
pant with regard to the comparison reference value equivalence. The smallest uncertainty was below
(CRV). 3% for a level of confidence of 95%.
In the case of radon, there is an additional charac- The results of most participants of a radon com-
teristic to be taken into account: the respective parison are correlated due to the common traceabil-
quantities can only be realized by use of a reference ity to one single radon gas standard producer. This
atmosphere which cannot be transported (Honig makes a careful correlation analysis necessary to
et al., 1998). Therefore, a comparison of the radon achieve an appropriate CRV.
activity concentration can only be performed by ap- The Coomet comparison (COOMET.RI(II)-
plying a transfer standard (the CRV is the calibra- S1.Rn-222, 2009) demonstrated the ability of six par-
tion factor of the transfer standard, i.e., a secondary ticipants to measure the radon activity concentration
standard) or by using a common reference atmos- simultaneously in the range of 75 Bq m23 to 12 kBq
phere (the CRV is the activity concentration). m23 in a stable reference atmosphere created at the
The first approach has the advantage of compar- National Scientific Center, Institute of Metrology
ing the quality of the realization of reference atmo- (NSC IM) Kharkiv, Ukraine. The degree of equiva-
spheres in a realistic way: Each participant works at lence was found to be satisfactory, although it has
his radon chamber according to a well-established to be mentioned that in the lower activity range (,1
quality system. Moreover, the usage of a primary kB qm23), the uncertainties were much higher than
standard is possible (Paul et al., 2002). The system the typical uncertainties (4–15% at a level of con-
under test and the 222Rn gas activity standard are fidence of 95%) reached in the other activity ranges.
enclosed in the chamber. The activity concentration In addition to comparisons of national institutes,
chosen for the point of calibration is calculated and there are also interlaboratory radon comparisons

72
 Radon and Its Progeny Detection Systems and Measurements

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 5.1. Results of the Euromet comparison 657 at an activity concentration of 10 kBq m23 for the obtained calibration factor k of the
participants identified by code: i ¼ 1 . . . N (1, PTB; 2, BfS; 3, STUK; 4, BEV; 5, ARCS; 6, Inte-UPC; 7, IRSN; 8, SUJCHBO; 9, PSI; 10, SSI;
11, HPA; 12, MPA).

taking place. As a rule, comparisons are conducted by 5.2 Radon Gas (222Rn, 220Rn) in the
recognized reference laboratories. Interlaboratory Environment
radon comparisons help to ensure a uniform quality
5.2.1 Radon Measurement Methods
standard and will preferably be organized for passive
radon instruments (nuclear track detectors, activated Because radon is ubiquitous and soluble in water,
charcoal detectors, electrets). The comparisons it is present in air, water, and soil. The final quantifi-
are often used for the purposes of determining the cation of radon activity concentration usually occurs
performance and surveillance of approved radon in an air sample as radon is a gas under atmospheric
services. pressure and temperature. To measure the activity
Regular annual radon intercomparisons are con- concentration of radon in water, either the radon is
ducted by various test institutions. The intercom- extracted by bubbling air through the water or the
parisons are designed for instruments using solid radon is measured directly using liquid scintillation
state nuclear track detectors, electrets, or activated techniques where the water sample is mixed with a
charcoal detectors and run with similar procedures: scintillation cocktail (Section 5.2.1.2).
radon services submit a sufficient number of instru- Both active and passive methods can be used to
ments of the same type to the provider of an inter- measure radon. For the active method, an air
comparison. Depending on the applied test scheme, sample is forced by pressure into the measuring
devices are randomized and grouped according to chamber, while for the passive method, movement
the number of exposures. The number of instru- due to natural diffusion takes place.
ments to be submitted depends on the number of In terms of the sampling time, the measurement
exposures and the need for additional transfer methods can be categorized into three different
instruments being used for measurement effects methods: grab sampling, short-term continuous
during storage and delivery. After exposure, the sampling, and time-integrating sampling (long-
instruments are returned to radon services in order term) (NCRP, 1988). For the grab sampling method,
to determine the exposures to radon and report the the sampling duration is several seconds, minutes,
results to the provider of the intercomparison. Finally, or hours and the result reflects the radon activity
the provider prepares a report with the measure- concentration at the time of the measurement. For
ments and reference data. short-term continuous sampling, the sampling
Radon services interested in participating in a duration is several hours or days and the radon
radon intercomparison can get further information activity concentration is registered typically for 30,
about providers and organizational conditions from 60 min, or 2 h intervals. In this case, the temporal
the European Information System on Proficiency fluctuation of the radon activity concentration can
Testing Schemes (eptis) available via the Internet. be detected. For time-integrating sampling, the

73
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

sampling duration may be either days, months, or 1 is applied which admits radon gas but inhibits the
year. The result is given as the integrated average entry of its progeny into the measuring chamber.
and therefore provides no information about the Both direct ( pulse ionization chamber) and indirect
temporal change in radon activity concentration (electric field collection) detection methods can be
within the duration of the measurement. used.
During the active sampling in the electric field
5.2.1.1 Airborne Radon collection method, radon enters the measuring
Grab Sampling. Most radon grab sample techni- chamber via a filter medium. Positively charged
218
ques use an alpha scintillation cell which was intro- Po is formed from the decay of radon in the sam-
duced in the 1950s by Damon and Hyde (1952), pling air of the chamber (Dua et al., 1983). The posi-
Lucas (1957), and Van Dilla and Taysum (1955). tive 218Po ions are collected electrostatically on the
There are two methods for collecting a representa- negative electrode of a semiconductor detector. The
tive air sample. In the first method, an air sample is advantage of the continuous sampling device is that
drawn directly into the alpha scintillation cell by a a temporal variation of radon activity concentration
vacuum pump. The second method is one of the can be observed. The disadvantage is the cost, the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

older and simpler methods of radon sampling, which noise of the pump, and the requirement of electric
is performed with an airtight, radon proof collaps- power if the operation is longer than about 2 weeks.
ible bag to sample air over the desired sampling A commonly applied passive method uses an acti-
time period (Pohl and Pohl-Rüling, 1976; Sill, 1977). vated charcoal collector, where the collector allows
After the sampling period, the activity concentration continuous adsorption on the active sites of the
of radon in the bag is transferred to a scintillation carbon beds (George, 1984; Kappel et al., 1993). The
cell. The main purpose of the collapsible bag is to most useful configuration has a diffusion barrier to
avoid the variation in pump flow rate due to separate the charcoal from the ambient air which
build-up of back pressure in a container. The pump improves the uniformity of response to variations of
flow rate is not critical as long as it is suitable for the radon activity concentration with time. In addition,
size of the bag and the sampling duration. However, it is well known that charcoal is also a very good ab-
variation of the flow rate over the collection time sorber of water vapor, which can reduce the adsorp-
period of the sample will affect the accuracy of the tion efficiency for radon, thus requiring a moisture
measurement, thus requiring pumps with controlled correction (Iimoto et al., 2004). During the measure-
flow rates. ment period (typically 2–7 d because the half-life of
To measure the radon activity concentration, the radon is only 3.8 d), the adsorbed radon undergoes
scintillation cell is sealed after air sampling, either radioactive decay. A device commonly used by
by direct collection or via collection in a bag. The several groups consists of a circular container filled
inner surface of the cell is coated with zinc sulfide with activated charcoal (George, 1984; Kappel et al.,
(ZnS), which emits pulses of light (scintillations) 1993) or a specially designed plastic scintillation
when struck by an alpha particle. To count these vial with a small, porous cartridge, containing char-
scintillations, the cell, which is fitted with a clear coal, fixed near the top of the vial (Kappel et al.,
window, is optically coupled with a photomultiplier 1993; L’Annunziata, 2003; Passo and Floeckher,
tube. The scintillations resulting from alpha parti- 1991).
cles emitted by the radon and the radon progeny in After exposure, the device is tightly sealed to
the air sample are counted. After a few hours, ap- maintain maximum sensitivity and returned to a la-
proximate secular equilibrium is reached and the boratory for the analysis of the quantity of radon
pulse rate is proportional to the radon activity con- adsorbed by using gamma spectrometry or liquid
centration in the cell. scintillation (Carnoba et al., 1999).
Advantages of these techniques include their sen-
sitivity and the rapidity with which the results can Time-Integrating Sampling. This method is often
be obtained. Some disadvantages include the fact used in national radon surveys for human exposure
that radon levels can show large spatial and tem- or dose estimation and in epidemiological studies.
poral variations. The detection limit of these techni- Integration over a long period of time has the advan-
ques depends on the size of the scintillation cell, the tage to average out short-term fluctuations of radon
background level, or sensitive volume, and ranges levels due to diurnal and seasonal variations.
from 1 to 37 Bq m23 for a 30 min counting interval The sample collection occurs in a passive way, by
(George, 1996). diffusion, using a diffusion container with a radon de-
tector, e.g., a solid-state nuclear track detector
Short-term Continuous Sampling. Independent (SSNTD) or an electrically charged teflon disk inside
of the short-term sampling method, a filter medium it. In the container, the temperature gradients are

74
 Radon and Its Progeny Detection Systems and Measurements

small, which reduce the effect of convection inside

the chamber and hence plate-out on the SSNTD or
electric charging of the teflon disk (Frank and
Benton, 1973). Some designs use a plastic bag as a
filter for the whole detecting device (Durrani and Ilic,
1997; Nikezic and Yu, 2004; Tommasino et al., 1986).
For measurement of indoor radon activity concentra-
tions, exposures of 3 months to 1 year are required,
because this technique is not sensitive to low level
radon.
A new type of personal exposure meter for radon
gas has been developed for the purpose of individual
monitoring (Karinda et al., 2008). Since this meter
Figure 5.2. Membrane tube system for continuous measurement
is based on a passive technique, it is applicable to (Surbeck, 1996).
indoor measurements over a long period of time and

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

is thus particularly suitable for epidemiological
studies (see Section 5.5.1).

5.2.1.2 Waterborne Radon. Radon in water

can be determined by measuring the radon released
by bubbling air through the water. This measure-
ment is a typical grab sampling method where many
active devices can be applied (Cosma et al., 2008;
Somlai et al., 2007a; Todorovic et al., 2011; Zmazek
et al., 2002). Continuous measurements are also
available by using gas transfer membranes for
radon and water separation (Figure 5.2) (Surbeck, Figure 5.3. Pulse height spectrum of radon and its decay
1996). Passive measurements are carried out with products obtained in a typical radon-in-water measurement by
SSNTDs and gas transfer membranes (Tommasino LSC (Yasuoka et al., 2004).
et al., 2012).
An alternative passive grab sampling method is
the use of a liquid scintillation counter (LSC), where
the water sample containing radon is mixed with a
scintillation cocktail, e.g., Toluene (Prichard et al.,
1992; Salonen, 2010; Yasuoka et al., 2004; Yokoyama
et al., 2011). Figure 5.3 shows the pulse-height spec-
trum in an LSC produced by radon and its progeny
in water.
For long-term measurements, SSNTDs are also
used as cheap and simple detection materials
(Figure 5.4) (Marques et al., 2004; Vásárhelyi et al.,
1997). In this radon measurement device, radon dif-
fuses from the water sample into the airspace above.
The fiberglass filter reduces vapour entry and avoids
thoron entry into the detection volume.
Figure 5.4. Device for the long-term measurement of radon in
water (grab sampling) using an SSNTD detector (Marques et al.,
5.2.1.3 Soilborne Radon. When radium 2004).
decays in soil grains, a fraction of the resulting atoms
escape from the mineral grains to air-filled pores— Exhalation Measurements (In situ). Radon exhal-
this is referred to as emanation. The radon gas is sub- ation from the ground surface affects both indoor and
sequently transported through the pores of the outdoor radon activity concentrations. Therefore, it is
material and some of it reaches the surface before important to clarify the exhalation process of radon
decay. The amount of activity released per unit from the soil to the atmosphere. Moreover, measure-
surface area and per unit time is the exhalation rate. ment of the radon exhalation rate is applied to

75
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

research fields such as health physics, environmental where IRn is the production rate of radon (s21), lRn
science, and geosciences (Ishimori and Maruo, 2005; the decay constant of radon (s21), V the volume of
Lawrence et al., 2009; Sahoo et al., 2010; Somlai the accumulation chamber (m3), and v the sampling
et al., 2006b). In environmental science, radon can be flow rate (m3 s21). The radon atom concentration in
used as a tracer for the evaluation of the environmen- the accumulation chamber is obtained with Equation
tal behavior of air pollutants (Iida et al., 1996). Data (5.3), where C is the constant of integration,
on radon exhalation rate from a soil surface are
needed for a calculation model of atmospheric trans- IRn C
port (Sakashita et al., 2004). In geosciences, the NRn ¼ þ eðlRn þðv=V ÞÞt
ðlRn þ ðv=V ÞÞV lRn þ ðv=V Þ
relationship between the behavior of exhaled radon
ð5:3Þ
and earthquakes has been studied for more than
30 years. Increases in radon activity concentration in
soil, ground water, and the atmosphere as precursor For the initial condition (t ¼ 0, NRn ¼ 0) in
phenomena of earthquakes have been reported by Equation (5.3), the constant of integration C is calcu-
many researchers (Igarashi et al., 1995; Yasuoka and lated by Equation (5.4),

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Shinogi, 1997).
IRn
C¼ ð5:4Þ
Continuous and Grab Sampling (Active). For a V
continuous measurement system, a ventilation-type
accumulation chamber is used for radon exhalation When Equation (5.4) is substituted into Equation
rate determination. A diagram of the flow-through (5.3), Equation (5.5) is obtained,
exhalation rate measurement system is shown in
Figure 5.5 (Hosoda et al., 2011). The accumulation IRn
chamber is set on the ground surface. The air inlet NRn ¼ f1  eðlRn þv=VÞt g ð5:5Þ
ðlRn þ ðv=VÞÞV
fixed at 2 m above the ground surface is connected
by a tube to the accumulation chamber inlet. The air
The radon activity concentration ARn (Bq l21) is
in the accumulation chamber is continuously drawn
obtained with Equation (5.6) using Equation (5.5),
into the scintillation or Lucas cell (Lucas, 1957). A
manometer connected to the accumulation chamber
lRn IRn
outlet measures the pressure difference between ARn ¼ f1  eðlRn þv=VÞt g ð5:6Þ
inside and outside the accumulation chamber. ðlRn þ ðv=VÞÞV
The radon atom concentration NRn (m23) in the
accumulation chamber is evaluated with Equation The generation rate of radon activity ERn (Bq s21) is
(5.2), then evaluated with Equation (5.7) using Equation
(5.6)
VdNRn ¼ IRn dt  lRn VNRn dt  NRn vdt ð5:2Þ
 v
lRn þ VARn
ERn ¼ lRn IRn ¼ V v ð5:7Þ
 lRn þ t
1e V

Thus, the exhalation rate of radon JRn (Bq m22 s21)

can be estimated with Equation (5.8),
 v
lRn þ VARn
JRn ¼ V v ! ð5:8Þ
 lRn þ t
S 1e V

where S is the surface area under the accumulation

chamber (m2).
Grab sampling measurement is also possible
using a ventilation-type accumulation chamber.
The radon exhalation rate JRn by grab sampling
Figure 5.5. A flow-through exhalation rate measurement system. is estimated with Equation (5.9), obtained from

76
 Radon and Its Progeny Detection Systems and Measurements

Figure 5.6. Diagram of the system used for simultaneous in situ

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

measurement of radon and thoron exhalation rates from the
ground.

Figure 5.7. Schematic diagram of the experimental system used

Equation (5.8).
to evaluate the emanation coefficient of radon.
lRn VARn
JRn ¼ ð5:9Þ
Sð1  elRn t Þ
Emanation (Ex situ). The emanation coefficient is
Grab Sampling (Passive). Radon and thoron ex- the fraction of radon formed in the soil grains that
halation rate measurements can be accomplished escapes into the pores. In other words, the ratio
using a special scintillation cell system (Figure 5.6) between the radon that escapes into the pore spaces
(Sägusa et al., 1996; Shimo et al., 1994). It is com- to the total amount of radon generated (equivalent
posed of the accumulation chamber (skirt section) to the radium activity concentration in the case of
that covers the ground, scintillation detector with secular radioactive equilibrium between radon and
aluminized Mylar sheet, light guide, photomultiplier radium). An accumulation method has commonly
tube, pulse counter, and timer. A large-area acrylic been used to calculate the radon emanation coeffi-
sheet is coated with a ZnS(Ag) scintillator. This cients from soils, rocks, and building materials
ZnS(Ag) scintillator is connected to a photomultiplier (Chao et al., 1997; Hosoda et al., 2009; Tuccimei
tube via a tapering light-guide. The purpose of the et al., 2006). An airtight accumulation chamber
skirt section is to collect radon and thoron gases that equipped with a scintillation cell monitor is used to
are exhaled from the ground surface. Measurements measure the emanation coefficient. Each sample is
can be recorded over consecutive 30 s intervals enclosed in an accumulation chamber as shown in
during a total recording period of 30 min. Figure 5.7 for 1– 3 d. The emanated radon is trans-
The thoron exhalation rate JTn is obtained by the ferred from the accumulation chamber to the scintil-
following formula: lation cell.
In general, the radon activity concentration in
JTn ¼ ðN10  Nb ÞCFT ð5:10Þ the accumulation chamber will increase gradually
where N10 is the count rate 10 min after the start of until it reaches a secular radioactive equilibrium
the measurement (cpm), Nb the count rate of the activity concentration (Aeq) after about 30 d. This
background (cpm), and CFT is the calibration coeffi- equilibrium activity concentration in the accumula-
cient for 220Rn. Since the count rate of 220Rn and its tion chamber is considered to be equal to the radon
decay product (216Po) reaches equilibrium 7–8 min activity concentration in the pore space amongst
after starting the measurement, the count rate at 10 the solid grains. The growth curve can be expressed
min is used for N10. On the other hand, the count rate as
of 222Rn and its decay products (218Po and 214Po) is
based on the count rate after 30 min. The radon exhal-
At ¼ Aeq ð1  exp ðltÞÞ ð5:12Þ
ation rate JRn is obtained by the following formula:

JRn ¼ ðN30  N10 ÞCFR ð5:11Þ where At is the activity concentration of radon mea-
where N30 is the radon count rate 30 min after the sured at time t (Bq m23), t the accumulation time of
start of the measurement, and CFR is the calibration the sample (s), and Aeq the radon activity concentra-
coefficient for 222Rn. tion in equilibrium (Bq m23). The radon emanation

77
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

coefficient can be defined as (Morawska, 1989): depth, but deploying them in a pipe is also possible
using an automatic exchanger (Chavez et al., 1997).
Aeq V
f ¼ ð5:13Þ
ARa M 5.2.2. Radon Detection Systems
where f is the radon emanation coefficient; ARa
the radium activity concentration of the sample For radon measurements the a-particle detection
(Bq kg21), which can be determined by gamma spec- method is the primary one as radon, during its
trometry; V the gas empty volume (the chamber decay, emits a-particles exclusively, while the short-
volume minus the sample volume, m3); and M the lived decay products of radon emit a-, b-particles
sample mass (kg). and g-rays, depending on the individual radio-
nuclide (see Section 4.2).
The method, when detection focuses on a-particles
from radon and its progeny, is called the direct
Soil Gas Measurement (In situ). The measure- method. For the indirect method, only a- or b-particles
ment of soil gas radon can use either active or or g-rays emitted from radon progeny are detected by

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

passive methods. These measurements are usually the two-filter method (Thomas and Leclare, 1970) or
made for geological purposes. the electrostatic collecting method (Iimoto et al., 1998)
As it is an in situ measurement, grab-sampling is and the radon activity concentration is recalculated
the most widespread method using a scintillation from these data (Figure 5.9).
cell system or other measuring systems (Buzinny
et al., 2009; Genrich, 1995; Neznal et al., 2004a; Scintillation Detector. The different types of the
Papastefanou, 2007; Shweikani and Hushari, 2005). scintillators can be divided into two large groups:
For the sampling, a sampling probe is used, collect-
ing soil gas from a depth of 50– 100 cm. Continuous (1) Inorganic scintillators
measurement is also possible using a soil air flow (2) Organic scintillators.
system (Figure 5.8) (Froňka et al., 2008).
For radon measurements, the most frequently
In the case of the passive method, the SSNTDs
used inorganic scintillator material is the ZnS(Ag).
are the preferred method with maximum 1– 2 weeks
A typical application is where the inner part of an
exposure time as radon activity concentrations with
alpha scintillation cell, apart from its window, is
some hundreds kBq m23 occur in the soil gas, which
coated with ZnS(Ag) (alpha scintillation cell) and is
due to track density saturation effects may exceed
optically coupled to a photomultiplier tube. The
the high level detection limit (Mazur et al., 1999;
most common detector based on this procedure for
Tanner, 1991). The track detectors are usually
radon measurement is the alpha scintillation cell.
placed in a protective chamber and buried at some
The device has become known as the Lucas cell as it
was developed by Lucas (1957). The flask designed
by Lucas has the shape of a right circular cylinder
with a hemispherical cap. Its diameter and volume
are 5 cm and 100 cm3, respectively. Typically, the de-
tection efficiency is 75–80% and its background
count rate is about 0.1 cpm. Under these conditions,
the uncertainty in measuring a sample containing 10
Bq m23, using a 3 h measurement and background-
count intervals, is about 30%. Improved efficiency
may be achieved by extracting radon from a larger
volume of air and transferring it to the cell (Ingersoll
et al., 1983; Lucas, 1957).
Gas-Filled Detector. The various types of gas-filled
detector include:

(1) Ionization chambers

(2) Proportional counters
(3) Geiger Mueller (GM) counters.

For radon measurements, the pulse ionization

Figure 5.8. Continuous soil radon monitoring system (Froňka chamber (PIC), i.e., an ionization chamber operated
et al., 2008). in the pulse mode, is frequently used. In this case,

78
 Radon and Its Progeny Detection Systems and Measurements

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 5.9. Radon detection methods according to the detected radiation.

no charge multiplication takes place and the output In addition, silicon PIN photodiodes are often
signal is proportional to the particle energy dissi- used as a relatively cheap detector (Ui et al., 1998).
pated in the detector. Since the signal from an ion-
ization chamber is not large, only strongly ionizing
particles such as alphas, protons, fission fragments, Solid-State Nuclear Track Detectors. The most
and other heavy ions are easily detected by such widely used technique for integrating measurements
detectors. of radon activity concentration is based on plastic or
polymeric materials (Durrani and Ilic, 1997; Nikezic
Semiconductor Detector. The following different and Yu, 2004). These materials are often called solid-
types of semiconductor detectors may be used: state nuclear track detectors (SSNTD) or alpha-track
(1) Surface-barrier detector detectors. The most common materials in use for
(2) Diffused-junction detector radon detection are cellulose nitrate (CN) film,
(3) Silicon lithium-drifted detector poly-allyl-diglycol carbonate (PADC), and polycar-
(4) Germanium lithium-drifted detector bonate (PC) plastic. The passage of an alpha particle
(5) Germanium or high-purity germanium detector through an SSNTD produces a narrow primary
(6) CdTe, CdZnTe, and HgI2 detectors. damage trail or latent track along the length of its
path in the material (typically 20–70 mm), which can
For radon measurements, surface-barrier detectors be made visible by chemical or electrochemical
are most commonly used in two-filter measurement etching (Table 5.1). The use of such alpha track detec-
systems (Brunke et al., 2002; Tokonami et al., 1996c; tors for passive long-term integrating measurements
Whittlestone and Zahorowski, 1998) or in electrostat- of indoor radon is very popular for large-scale surveys
ic collection (Iida et al., 1991; 1996; Iimoto et al., (Alter and Oswald, 1983; 1987).
1998) indirect measurement methods as it can Several designs of measuring devices have been
provide a high resolution spectrum of particle ener- used for indoor radon surveys, for example, bare tech-
gies. Due to the energy discriminative ability of this nique, diffusion technique, etc. Details of these devices
detector, thoron analysis is also possible (Durridge, can be found in recent publications (Azimi-Garakani
2000; Iimoto et al., 1998). et al., 1988; Bartlett et al., 1986a, 1986b; Mellander

79
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 5.1. Etching conditions for several types of SSNTD

Type of SSNTD Etching conditiona Reference

CN NaOH solution Cherouati and Djeffal

2.5 N at 608C for 130 min (1988)
PADC NaOH solution Kenawy and Morsi
thickness 6 N at 60 8C for 6 h (1991)
0.5–0.7 mm KOH solution Langroo et al. (1991)
6.25 N at 60 8C for 6 h
PADC NaOH solution Pahapill et al. (1996)
thickness 1 mm 26 N at 68 8C for 17 h
KOH solution Bochicchio et al.
25.4 N at 608C for 3 h, followed with electrochemical etching at 30 8C for 5 h, (1996)
with 30 kV cm21 at 2 kHz
PC Chemically etching used a mixed solution of 8 mol l21 of C2H5OH and KOH (ratio 4:1) Gomez et al. (1993)

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

and electro-chemical etching in 6 N KOH solution (containing 20% by volume of alcohol):
applying a high voltage of 800 at 2 kHz for 3 h at room temperature.

and Enflo, 1992; Stoop et al., 1997; Tokonami et al., etching, usually in alkaline solutions. The etching
2005; Zhuo et al., 2002). The devices described in these process used depends on the type of SSNTD material,
reports are generally based on the use of SSNTDs. the alkaline solution concentration, its temperature,
For the bare technique, the SSNTD material is and time period for etching. Some etching conditions
mounted open-faced or bare on a wall in a building for various SSNTDs are shown in Table 5.1.
to record directly alpha particles from the airborne After etching, the number of tracks can be deter-
radon. However, this technique shows that the effect mined using optical microscopy by manual scanning
of radon progeny plate-out on the detector is obvi- and counting (CN, PC, and PADC) (Tokonami et al.,
ously more significant than that of the radon gas. In 1996d), automatic counting system with special soft-
measurements made under natural conditions, the ware such as image analyzer for PADC (Tokonami
effect of plate-out becomes very large due to me- et al., 2003), spark counter for CN (Huang et al.,
teorological factors, such as wind velocity and tem- 1986), and microfiche reader for PC (Baixeras et al.,
perature differences (Porstendörfer, 1994) and the 1997). Most radon alpha track detectors are counted
simultaneous plate-out of aerosol particles and over a small area (cm2).
thoron progeny on the detector surface (Abu-Jarad The radon activity concentration can be estimated by
and Fremlin, 1982). In addition, many researchers using a calibration coefficient. Every SSNTD-based
noted that this technique was unstable and difficult radon detector design should be calibrated in a radon
to calibrate (Abu-Jarad et al., 1981; Hadler et al., calibration chamber at least once a year. Determination
1991; Mäkeläinen, 1986). In spite of these difficul- of the calibration coefficient requires exposure of the
ties, this technique has been used in some major epi- SSNTD to a known radon activity concentration in a
demiological studies (Darby et al., 2005; Krewski radon exposure chamber. These exposures are used to
et al., 2005a). For some track detectors (especially obtain or verify the calibration coefficient between net
those of the open variety), possible interference from track production rate per unit area and radon activity
thoron and its progeny must be taken into account concentration. Based on the surveyed literature, it
(see Section 7.4). would seem that the calibrations were performed at
The diffusion technique is widely used for radon radon activity concentrations in the range of about
surveys, with the advantage that it is not affected by 2000–10 000 Bq m23 (Eappen and Mayya, 2004; Khan
the decay products in ambient air. Here, the SSNTD et al., 1990; Langroo et al., 1991; Subba Ramu et al.,
material is mounted inside a small, almost airtight, 1988). If radon detectors are well designed to measure
222
closed container. Air containing radon can enter the Rn, the radon activity concentration can be simply
container by diffusion, but not the radon progeny calculated according to the following equation,
which are prevented from penetrating into the con-
tainer by an effective diffusion barrier. Alpha parti- G  BG
CRn ¼ ð5:14Þ
cles from radon that enter the chamber and its FT
in-grown progeny are detected by the SSNTD. where CRn is the radon activity concentration (Bq m23);
After exposure, the tracks due to alpha particles G and BG are the gross and background track density
are made visible by chemical or electrochemical (track cm22), respectively; F is the calibration

80
 Radon and Its Progeny Detection Systems and Measurements

coefficient (tracks cm22 Bq21 m3 h21); and T the expos- and A and B are the constants for a particular elec-
ure time (h). The background track density is obtained tret configuration.
using unexposed detectors during the same time period
as the exposed detectors. The detection limit for this 5.2.3 Thoron Measurements
technique for a 3 month long exposure is 5–10 Bq m23,
depending on the size of the scanned detector area The theory and methodology of thoron measure-
(George, 1996). Since several studies have revealed that ments are the same as for radon measurements.
some of the detectors are affected by thoron interfer- Application of the Lucas cell system as a direct
ence, Equation (5.14) cannot be applied in this case. measurement is the preferred way to measure
Their evaluation is described in detail in Section 7.4. thoron activity concentrations for both grab (Eappen
et al., 2008; Knutson et al., 1994; Sumesh et al.,
2012) and continuous sampling methods (Eappen
Electret Detectors. An electret passive environ- et al., 2007; Falk et al., 1992; Iimoto et al., 1998).
mental radon monitor is a continuously integrating Electrostatic collection using a semiconductor de-
radon detector consisting of a small chamber having tector (Figure 5.10) is also a frequently used indirect

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

a charged Teflon disk (electret) at the bottom and a method and used for many commercial monitors.
filtered inlet at the top (Kotrappa et al., 1988; 1990). In many large-scale radon and thoron surveys,
After radon gas passively enters into the chamber radon –thoron discriminative detectors are used
through the filter, alpha particles emitted from the (Tokonami et al., 2005). Figure 5.11 shows overviews
decay of radon and its progeny ionize the air mole- of a radon –thoron discriminative detector. The de-
cules and these ions are collected by the electret. tector consists of two different diffusion chambers, a
This causes the voltage level of the electret to de- low and a high air-exchange-rate chamber. Each
crease. The radon activity concentration can then be chamber is made of an electro-conductive plastic.
obtained by the following equation (Kotrappa et al., PADC is used as the detecting material and a piece
1990): of it is attached to the bottom of the chamber with
adhesive putty. Radon in air can penetrate into the
Vi  Vf
RnC ¼  BG ð5:15Þ low air-exchange-rate chamber through an invisible
TCF
air gap between its lid and bottom by means of diffu-
BðVi þ Vf Þ sion. Since this air gap functions as a very effective
CF ¼ A þ ð5:16Þ diffusion barrier, thoron can scarcely enter the
2
chamber through such a small pathway due to its
where RnC is the radon activity concentration very short half-life (55.8 s), compared with that of
(Bq m23); T the exposure period (d); Vi and Vf are radon (3.82 d). In order to detect thoron more effect-
the measured initial and final electrets voltages (V), ively, some holes are opened at the side of the other
respectively; CF is the conversion coefficient (V per chamber and are covered with an electro-conductive
Bq m23 d); BG the radon activity concentration sponge. This chamber is referred to as the high
equivalent of the natural g background radiation; air-exchange-rate chamber.

Figure 5.10. One possible construction of the continuous 220Rn monitor using the electrostatic collection method (Iimoto et al., 1998).

81
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 5.11. Overviews of 222Rn– 220Rn discriminative detectors (Tokonami et al., 2005).

It is not easy to remove the lid in this detector and Sathish, 2011), consisting of two cylindrical
unless a cutting tool is used. This design feature acts cups, one cup recording tracks from both 222Rn and
220
as a protection against unwelcome human interfer- Rn and the other cup tracks only from 222Rn.
ence in the measurements. Due to its small size, this
detector can be put into most post-boxes and thus
permits easy and cost-effective transportation. 5.3 Radon and Thoron Progeny Activity
Following exposure of the detectors, the PADC Concentrations and Particle Size
plates are taken out of the chamber and are chem- Distributions in the Environment
ically etched and alpha tracks are counted with a
track reading system. Using two alpha track dens- 5.3.1 Radon and Thoron Progeny
ities (number of tracks cm22) from the low and high The dose to the lungs is predominantly caused by
air-exchange-rate chambers (NL and NH), radon and the deposition of radon progeny in bronchial airways
thoron activity concentrations can be obtained by (Aurand et al., 1956; Jacobi, 1964; 1984). In radiation
solving the following two equations: protection, however, for reasons of simplicity and
costs, the measurement of radon is the preferred
NL ¼ XRn CFRn1 T þ XTn CFTn1 T þ BG ð5:17Þ method for the estimation of human exposure to its
progeny. To interpret these measurements, a value
NH ¼ XRn CFRn2 T þ XTn CFTn2 T þ BG ð5:18Þ for the equilibrium factor, F, is required to estimate
the radon progeny activity concentration in air. Since
where XRn and XTn are the mean activity concentra- the lung dose also depends on the unattached frac-
tions of radon and thoron during the exposure period in tion, fp, the separation and measurement of the
Bq m23 ; CFRn1 and CFTn1 are the radon and thoron attached and unattached fraction of the radon
calibration coefficients for the low air-exchange-rate progeny is of interest. It is achieved by splitting the
chamber in number of tracks cm22 Bq21 m3 h21; CFRn2 equilibrium-equivalent activity concentration into an
and CFTn2 are the radon and thoron calibration coeffi- attached and an unattached equilibrium-equivalent
cients for the high air-exchange-rate chamber in activity concentration. For 222Rn and its progeny,
number of tracks cm22 Bq21 m3 h21; T the exposure mean values of F and fp measured in dwellings and
time in hours; and BG the background alpha track indoor workplaces range from 0.2 to 0.7 (Sections 4.5
density on the CR-39 detector in tracks cm22. and 7.5.1) and from 0.03 to 0.2 (Sections 4.6.1 and
A similar method to measure 222Rn and 220Rn in 7.5.1), respectively. Under conditions where the ven-
situations where both gases are present is the tilation rate is not too high, measurements have
SSNTD-based twin cup detector (Ramachandran shown that F is negatively correlated with fp (Section

82
 Radon and Its Progeny Detection Systems and Measurements

4.6.2). Therefore, fundamental studies concerning set up at the German radon reference chamber in the
the correlation of F, fp, and the environmental para- Physikalisch-Technische Bundesanstalt (PTB). This
meters such as aerosol particle concentration will be is worldwide the only primary standard for traceable
of interest. calibrations of 222Rn or 220R progeny activity concen-
Both the equilibrium factor F and the unattached trations.
fraction fp are relative measures of the amount of Radon progeny size distributions are covered in
short-lived radon progeny in air, either in sum or for Sections 4.7, 5.3.3, and 7.5.2.
the unattached part. It is important to note that
several measurements of F and fp in typical environ- 5.3.2 Radon Progeny Measurement Methods
ments (rooms used for normal purposes, outdoor,
mines, etc.) (Huet et al., 2001a; 2001b; Porstendörfer, The radon progeny activity concentration is
2001; Reineking and Porstendörfer, 1990; Vargas defined as the activity concentration (Bq m23) of the
et al., 2000) provide a set of data to estimate the ex- specific decay products 218Po, 214Pb, and 214Bi – 214Po
posure caused by radon progeny on the basis of a in air. The measurement of individual nuclides is
radon activity concentration measurement. rare but not a difficult one (Scott, 1981; Tsivoglou

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Naturally, these results have quite large systemat- et al., 1953).
ic uncertainties, but taking into account the quite The first measurements of radon progeny used a
comparable statistical and systematic uncertainties charged wire for collection with progeny measure-
of a radon activity concentration measurement with ment in an electroscope (Elster and Geitel, 1902).
a passive device, the application of typical F and fp Later, measurements in mines were made to deter-
values is a reasonable method in most cases of radi- mine the new unit of exposure, the Working Level
ation protection. Nevertheless, this method is highly (WL), which was assumed to be related to risk
dependent on the availability of data for F and fp in (Kusnetz, 1956). These consisted of a short-term fil-
different environments, which have to be measured tered air sample followed by a single measurement
by a portable, for example, active system that has to with a scintillation probe and it was stated that the
be calibrated under well-defined conditions by a WL could be estimated within 13% accuracy.
much more accurate method. Individual progeny could be calculated using the
Thoron (220Rn) has a relatively short half-life of same sampling method followed by 3 count intervals
55.8 s. For this reason, the exhalation of thoron from (Tsivoglou et al., 1953).
the ground is usually of less significance, compared All types of electronic instruments that are based
with 222Rn. This is, however, not always valid to the on grab sampling, continuous sampling, and inte-
same extent for building materials. Already simple grating measurement methods can be used for
measurements of radon progeny in normal living radon progeny activity measurements in air under
spaces or workspaces show that—in addition to the certain conditions.
progeny of radon (218Po, 214Pb, and 214Bi– 214Po)— A radon progeny measurement has to yield at
also the progeny of thoron (212Pb– 212Bi and 212Po) least one of the following results:
sometimes occur in relevant activity concentrations. (a) the activity concentration of one or more short-
This is, among other things, due to the half-lives of lived radon progeny,
the 220Rn progeny, which are considerably longer (b) the potential alpha energy concentration or ex-
than those of 222Rn. To evaluate the relevance for ra- posure of short-lived radon progeny,
diation protection, it is important to note that at (c) the equilibrium equivalent activity concentra-
identical activity concentrations of radon and tion.
thoron, the potential alpha energy concentration for
thoron progeny has a value which is 14 times higher In order to do this, the instrument has to have a
than that of radon progeny at equilibrium. For the sampling assembly, a radiation detection assembly,
determination and evaluation of radiation exposures and a data processing and recording unit
by natural radionuclides, it is, therefore, in special (Figure 5.12).
cases, also necessary to correctly measure thoron The international standards IEC 61577-1 (IEC,
and its progeny, in addition to radon. 2006), IEC 61577-2 (IEC, 2000), IEC 61577-3 (IEC,
For the measurement of the individual short-lived 2011), and IEC 61577-4 (IEC, 2009) and ISO
radon and thoron progeny, a precise method is 11665-2 (ISO, 2012a) and ISO 11665-3 (ISO, 2012b)
required for the separation and measurement of each define the conditions under which such a system can
short-lived progeny activity concentration in air. For be operated, what technical requirements exist, and
calibration purposes, a special sampling system to- what kind of calibration is required. It has to be
gether with a measuring system by simultaneous a- emphasized that there are quite a number of proto-
g-spectrometry (Paul et al., 1999) was developed and type devices available, as well as a small number of

83
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Figure 5.12. Sampling assembly, radiation detection assembly, and data processing and recording unit are mandatory for radon progeny
measurements. The detector can be a surface barrier detector (PIPS) performing the measurement on- or offline, a photomultiplier
associated with a sensitive scintillation surface such as ZnS(Ag), a HP Ge-detector, or even SSNTDs.

The process of taking a sample on a target (filter or

screen) of the short-lived radon progeny from a radon
atmosphere is governed by a set of differential equa-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

tions (Bateman, 1910). These equations work equally
well for 222Rn and 220Rn. These equations describe
the build-up of the radon progeny activities, assum-
ing a constant collection rate CRnP. Initial condition
for the number of particles is N(RnP; t ¼ 0) ¼ 0 for
all isotopes. N(RnP) is the absolute number and lRnP
is the decay constant of the respective radon progeny
(RnP). During the time interval of t[ [0 : ts] with ts as
sampling time, the number of short-lived progeny
Ns,RnP(t) for an ideal collection on a target is given by:

cPo
Ns;Po ðtÞ ¼ ð1  expðlPo tÞÞ ð5:19Þ
lPo
Figure 5.13. Sampling techniques for PAEC measurements:
sampling unit and activity measurement are separated (a) or
together in one unit (b) (Porstendörfer, 1996).
 
cPo cPb
Ns;Pb ðtÞ ¼ þ ð1  expðlPb tÞÞ
lPo lPb
commercial devices. The inherent quality of radon cPo
 ðexpðlPb tÞ  expðlPo tÞÞ
progeny measurements performed with these lPo  lPb
instruments is by nature not to be compared with ð5:20Þ
the quality of radon gas measurements performed
with well-tested radon devices qualified as transfer  
cPo cPb cBi
standards in comparisons. This does not limit the Ns;Bi ðtÞ ¼ þ þ ð1  expðlBi tÞÞ
lPo lPb lBi
applicability of radon progeny measurement instru-  
ments, but makes a careful handling of the quality cPo cPb
 þ ðexpðlPb tÞ
assurance necessary. lBi  lPb lBi  lPb
 
Sampling techniques for the radon progeny meas- cPo lPb
urement can be divided into two groups (Figure 5.13):  expðlBi tÞÞ 
ðlBi  lPb ÞðlPo  lPb Þ
(a) sampling and activity measurement are sepa-
 ðexpðlPb tÞ  expðlBi tÞÞ
rated, i.e., the measurement of the activity is per-  
formed after completion of a collection cycle, or (b) cPo lPb
þ
sampling system and activity detection are combined ðlPo  lBi ÞðlPo  lPb Þ
in one unit, i.e., the measurement of the activity is  ðexpðlBi tÞ  expðlPo tÞÞ ð5:21Þ
performed during sampling (Porstendörfer, 1996).
There are quite a number of algorithms that can be
used in combination with measuring set-ups incorp- To shorten the equations, the following abbrevia-
orating filters or screen/filter combinations together tions are used: Po for 218Po, Pb for 214Pb, and Bi for
214
with gross a-counting or a-spectrometry. Sometimes, Bi. The calculation for 214Po has been omitted,
b or g counting is used to supplement or replace a because it is in activity equilibrium for all practical
counting. purposes with 214Bi due to its very short half-life of
164 ms.

84
 Radon and Its Progeny Detection Systems and Measurements

The collection rate CRnP(s21) is defined as The determination of the respective U can be
:
performed by time analysis of subsequent gross a, or
CðRnPÞ Vfl S a-b-counting or a-spectrometry measurements with
CRnP ¼ ð5:22Þ different delay times or by a single a-g-spectrometry
lRnP
measurement.
where C(RnP) is the: airborne RnP activity concen- One widely used analysis (with numerous modifi-
tration (Bq m23); Vfl is the air flow through the cations and additions in the literature) is the
sample (m3 s21); and S the dimensionless collection so-called Thomas method (Thomas, 1972) that bases
efficiency of the target. The activity A of the respect- the analysis of the progeny activity collected on a
ive isotopes building up during the sampling time ts filter by a sampling system with subsequent gross
and the measured decays (integral of the decay func- a-counting by a detector comprising a photomultiplier
tion in a given time interval tm after the sampling associated with a sensitive scintillation surface in
was stopped) is given in Figure 5.14. ZnS(Ag). A detailed instruction including the analysis
In the case of grab sampling, a delay time td has to of counts, and the uncertainty analysis and conditions
be included to allow for the delay between the end of for sampling is given in ISO 11665-3 (ISO 2012b).

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

sampling and the start of the measurement. The detection limit and decision threshold have to
The number of short-lived radon progeny on the be calculated according to standards ISO 11929
target after the collection is finished, Nd,RnP(t . ts), (ISO, 2010).
is given by The techniques to measure the PAEC are broadly
the same as those for the individual determination,
Nd;Po ðtÞ ¼ Nc;Po ðts Þ expðlPo tÞ ð5:23Þ
but are simpler to execute and to analyze. For
example, the method originally proposed by Kusnetz
(1956) is based on a single gross alpha count of a
Nc;Po ðts Þ lPo filter sample. The basis of this method is that, about
Nd;Pb ðtÞ ¼ Nc;Pb ðts Þ expðlPb tÞ þ
lPo  lPb 1 h after sampling, the counts from 214Po formed on
 ðexpðlPb tÞ  expðlPo tÞÞ ð5:24Þ the filter by decay of the precursor nuclides increase
and the count rate corresponds reasonably well to the
PAEC at the time of sampling and is almost inde-
Nc;Pb ðts Þ lPb pendent of the progeny equilibrium. A potential
Nd;Bi ðtÞ ¼ Nc;Bi ðts Þ expðlBi tÞ þ alpha energy concentration integrated measuring
lBi  lPb
system for short-lived radon progeny is shown in ISO
 ðexpðlPb tÞ  expðlBi tÞÞ 11665-2 (ISO 2012a).
 
Nc;Po ðts ÞlPo lPb
þ ðexpðlPb tÞ 5.3.3 Measurement of the Unattached
ðlBi  lPb ÞðlPo  lPb Þ
  Fraction
Nc;Po ðts ÞlPo lPb
 expðlBi tÞÞ þ In principle, two methods exist for the determin-
ðlPo  lBi ÞðlPo  lPb Þ
ation of the unattached fraction (Porstendörfer,
 ðexpðlBi tÞ  expðlPo tÞÞ ð5:25Þ 1996; Tu and Knutson, 1988): (1) estimation of the
unattached fraction from a number concentration
The activity on the target increases during the measurement, and (2) direct measurement of the
collection time t e [0 : ts] and decreases after the end unattached fraction based on diffusion methods.
of the collection due to the radioactive decay (see Based on the measurement of the number concen-
Figure 5.14). The integral activity in the time inter- tration by means of a condensation nuclei counter
val t e [td : td þ tm] yields the sum of decays U of (CNC) and the semi-empirical relationship between
the chosen isotope on the target during this time the unattached fraction of the PAEC and the
interval: number concentration [Equation (4.17) in Section
4.6.1], the unattached fraction can be determined
ðm
td þt
with an error of 10 –20% (Tu and Knutson, 1988).
URnP ðtc ; td ; tm Þ ¼ Nd;RnP ðtc Þ lRnP dt The most frequently used methods for the direct
td measurement of the unattached fraction of the
radon progeny are based on their diffusion proper-
ðm
td þt
ties. Because of their small size, unattached progeny
¼ ARnP ðtc Þ dt: ð5:26Þ have higher diffusivities than the attached progeny
td and thus diffuse more rapidly to surfaces, such as

85
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure
: 5.14. Collection and decay of short-lived radon progeny for an activity concentration C(RnP) ¼ 1 Bq m23 and a volume air flow of
Vfl ¼ 1023m3 min21. In this example, the characteristic times are chosen to be ts ¼ 1200 s, td ¼ 60 s and tm ¼ 300 s.

Figure 5.15. Two measurement methods for the determination of the unattached fraction of the PAEC: (a) the difference method, and (b)
the direct method (Porstendörfer, 1996).

the walls of tubes or the wires of a wire screen. For the direct determination of the unattached
These collection devices are called therefore diffu- fraction by means of diffusion methods two methods
sion batteries (see Section 5.3.5.2). Since a fraction of can be used (Figure 5.15) (Porstendörfer, 1996). The
the attached progeny is also deposited on the screens, difference method is based on two parallel measure-
though with a much smaller probability, the mea- ments of the PAEC, one measurement in connection
sured activities have to be corrected, e.g., by using with a diffusion battery (DB) for removal of the un-
alpha spectroscopy (Reineking and Porstendörfer, attached progeny, and subsequent measurement of
1990). the activities deposited on the two filters by alpha

86
 Radon and Its Progeny Detection Systems and Measurements

spectrometry. For the direct method, radon progeny scanning mobility particle sizer in connection with a
collected on a screen are directly determined by CNC (Tu and Knutson, 1988). For particle size mea-
means of alpha spectrometry. After screen filtration, surements in the diameter range from about 100 nm
the non-collected fraction of the radon progeny is to 5 mm, an optical aerosol spectrometer is the pref-
sampled on a membrane filter and registered with a erable choice because of its considerably higher size
second alpha spectrometer unit. The direct meas- resolution when compared with a cascade impactor.
urement technique is more difficult to calibrate, but By means of these measurement systems, the size
has a higher accuracy for small fp values than the fractionated particle number concentration can be
difference method. registered at given time intervals.

5.3.4 Radon Progeny Particle Size 5.3.4.2 Direct Activity Size Distribution
Distributions Measurements. There are two measurement tech-
niques suitable for the direct measurement of activ-
As already discussed in the previous section for ity size distributions (Reineking et al., 1988): (i) the
unattached radon progeny, indirect and direct wire screen diffusion battery, and (ii) the cascade im-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

methods can be also be used to determine the activ- pactor. The screen diffusion batteries can be used for
ity size distributions of the attached radon progeny particle diameters between 0.5 and 200 nm, i.e., for
(Porstendörfer, 1996, Reineking et al., 1992a): unattached and attached radon progeny, and the
cascade impactor is operated in the size range from
(1) the indirect approach utilizing the measurement
60 to 10 000 nm, i.e., only for attached progeny.
of the number size distribution of the carrier
Therefore, the combination of both systems is an ex-
aerosol, and
pensive but complete measurement technique for
(2) the direct approach of measuring the activity
classifying the entire size spectrum of the radon
size distribution.
progeny.

5.3.4.1 Number size distribution measure- Diffusion Batteries. The physical principle of a
ments. The decay products of radon attach quickly diffusion battery is the removal of particles on the
to ambient aerosol particles. The activity size distri- walls of a tube or the wires of a wire screen,
bution C(d) of the aerosol, which is to be determined, arranged as a series of tubes or wire screens, by
and the number size distribution Z(d), which can be Brownian motion (diffusion). Due to the inverse re-
measured, are different because the attachment lationship between particle diameter and the diffu-
probability b(d) is a function of particle diameter sion coefficient, the smaller the particle the higher
d. The relation between both size distributions is is the deposition on the tube walls or wire screens.
given by: For size distribution measurements with the wire
screen diffusion battery technique, several screen
CðdÞ ¼ C=X bðdÞZðdÞ ð5:27Þ
stages with different penetration characteristics
with (“graded” wire screens) between 0.5 and 100 nm are
needed (Figure 5.16). A wire screen diffusion battery
ð1
may consist of two distinctly different configurations
C¼ CðdÞ dd ð5:28Þ that may be termed as “series” or “parallel” (Holub
0
et al., 1988; Hopke et al. 1992; Knutson et al., 1984;
where C is the radionuclide activity concentration 1988; Porstendörfer, 1996). The penetration P, the
and X the attachment rate expressing the adsorp- ratio of the particle or activity concentrations C/Co,
tion velocity of the decay product to the aerosol with depends on the diffusion coefficient D(d), which is a
the number particle concentration Z(d). Values for X function of particle diameter d, the flow velocity vo,
are given in Section 4.6. and a screen parameter, which characterizes the
The procedure is to measure the number size distri- mesh density and the configuration of the system. A
bution Z(d) and to calculate the activity size distribu- detailed account of the historical development of dif-
tion C(d) by means of the attachment coefficient b(d). fusion batteries, the principal physical mechanisms,
This method of determination includes the inaccur- their different designs, and applications can be
acy of the attachment coefficient derived theoretically found in Knutson (1999).
and experimentally confirmed only for spherical The series system consists of a number of individ-
particles. ual wire screens with different 50% cut-off penetra-
The number size distribution in the size range tion values operated sequentially, thus yielding as
from a few nanometer up to about 200 nm can be many stages as wire screens. After sampling, the ac-
measured with an electrostatic classifier or a tivity collected on each wire screen and on the

87
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

size distribution using either the Twomey (1975), the

expectation-maximization (Maher and Laird, 1985),
or the Simplex algorithms (Nelder and Mead, 1965),
while the penetration characteristics of the wire
screen stages can be calculated using the Cheng and
Yeh (1980) penetration theory. For example, using
the Simplex algorithms, the relative activity size dis-
tribution of a given radon progeny can be approxi-
mated by a sum of log-normal distributions.
Problems associated with the wire screen method,
such as the resuspension of deposited progeny by
recoil effects, the limited knowledge in the charac-
terization of the screen diffusion batteries in the
small diameter range, or the weakness in the math-
ematical algorithms used in the data deconvolution,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

prompted Michielsen et al. (2005) and Michielsen
and Tymen (2007) to develop an annular diffusion
channel (ADC) battery, which allows a continuous
measurement. It consists of five annular diffusion
channels of different lengths plus a reference filter,
operating in parallel. The alpha particles emitted by
the 218Po and 214Po collected, or formed on the filter,
Figure 5.16. Principle of a parallel wire screen diffusion battery are detected by an alpha detector, placed in the
for the measurement of radon progeny activity size distributions.
inner tube of the diffusion channel opposite the
Adapted from Porstendörfer (1996).
filter. Each unit of the diffusion battery is character-
ized by a particle penetration curve. The slope of the
channel penetration curve is steeper than that of a
backup filter is measured simultaneously with a set screen, which indicates a higher selectivity of the
of parallel gross alpha counters, such as ZnS(Ag) ADC relative to the wire screen method. A similarly
scintillation detectors or by alpha spectroscopy constructed cylindrical diffusion tube (CDT) has
(Holub and Knutson, 1987). Since the ratio of activ- been reported by Vargas et al. (2005).
ity collected on the front and back faces of a wire Another method to measure radon progeny size
screen depends on the nature of the activity size dis- distributions was proposed by Johansson et al.
tribution as well as on the screen parameters, this (1984) who used a combination of an electrical
dependence must be considered in the data evalu- aerosol analyzer (EAA) and alpha spectrometry.
ation (Holub and Knutson, 1987; Solomon and Ren, Through a modification of the EAA, each size frac-
1992). For the determination of unattached size dis- tion was collected on a filter and subsequently mea-
tributions, finer meshes are used than for the sured by alpha spectrometry. A miniature
attached fraction. integrating particle size sampler was developed by
The parallel configuration system consists of a Harley et al. (2012a), consisting of an impactor, four
number of wire screen stages operated in parallel, with graded screens, and a backup filter. Sampling is for
each stage containing a specific set of wire screens and extended periods (months) and the measurement is
a backup filter (Figure 5.16). In connection with simul- the deconvolution of the 210Pb, 210Po measurement
taneous sampling and alpha-spectroscopic counting on the six filtration stages.
during and after sampling, a substantial improvement
in sensitivity and accuracy associated with measure- Cascade Impactors. An impactor operates under
ments of low, ambient radon progeny activity concen- the principle that if a stream of particle-laden air is
trations can be obtained (Ramamurthi and Hopke, directed towards a surface, particles of sufficient
1991; Reineking and Porstendörfer, 1986). Such a par- inertia will impact upon the surface, while smaller
allel system of wire screen diffusion batteries for a high particles with less inertia will follow the air stream
sampling flow rate of 2 m3 h21 makes it possible to lines and thus will not be collected (see Figure 5.17).
measure size distributions of activity concentrations as By operating several impactor stages at different flow
low as 5 Bq m23 (Reineking and Porstendörfer, 1986). conditions, the aerosol particles can be classified into
The observed activity concentration of the radon several size ranges from which the size distribution
progeny deposited on a given screen then allows the can be determined. The single stages are usually
reconstruction of the corresponding activity-weighted operated in a series (cascade) arrangement, also

88
 Radon and Its Progeny Detection Systems and Measurements

are removed from the entrance air by a tube diffu-

sion battery mounted in front of the impactor system
(Gründel et al., 2005; Reichelt et al., 2000).
In general, the activity impacted on the various
impactor stages is measured by alpha spectroscopy,
using either solid-state detectors or scintillation
counters. However, by replacing the impactor by a
surface barrier alpha detector, the individual radon
progeny can directly be measured by alpha spectros-
copy during sampling (Figure 5.17). Such a low-
pressure computer-controlled online alpha impactor
with eight stages was developed to measure semi-
continuously the size distribution of the radon
daughter aerosol over longer time periods (Kesten
et al., 1993). This online impactor with a sampling

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

rate of 5 m3 h21 makes it possible to measure the ac-
tivity size distribution also for low activity concen-
tration levels of radon progeny.
An interesting alternative method to measure the
alpha activity deposited on impactor plates was
developed by Iida et al. (2008). Since most of impact-
or/alpha spectrometry devices are expensive, too
large, less portable, and cumbersome to conduct in
the environment, they proposed to use imaging
plates as a reusable sensor for the detection and
storage of ionizing radiation energy in photo-
stimulant phosphor crystals. Thus, the imaging
Figure 5.17. Low-pressure computer-controlled online alpha plate detects and stores the image of alpha particles
impactor with eight stages developed for measuring and these images of alpha spots are subsequently
semi-continuously the size distribution of the radon progeny
aerosols over longer time periods (note: only two stages of the
analyzed by a computer program.
eight-stage unit are shown here) (Kesten et al., 1993).
Comparison of Activity Size Distributions
Measured by Diffusion Battery and Cascade
known as the cascade impactor. The aerosol stream is Impactor. Reineking et al. (1988) measured the ac-
passed from stage to stage with continually increas- tivity size distributions obtained by both a high
ing velocities and decreasing particle cut-off sizes. volume diffusion battery and a low pressure cascade
The most important characteristic of an impactor impactor in connection with alpha and gamma spec-
is the collection efficiency curve, which gives the troscopy (Figure 5.18). In order to compare the
fraction of particles of a given size collected from the results of the diffusion batteries and of the impactor,
incident stream as a function of particle size. Ideally, activity median aerodynamic diameters (AMAD)
an impactor should collect all particles larger than a were calculated to a first approximation from diffu-
certain cut-off size upon the plate, while all other sion equivalent activity median thermodynamic dia-
particles follow the gas flow out of the impaction meters (AMTD) by multiplying them with the
region. In reality, however, the efficiency curve of a square root of the particle density. Assuming a
typical impactor stage spans over a range of particle density of 1.4 g cm23 yielded mean AMADs of the
sizes, although impactors are normally designed to attached fraction of 214 nm (218Po) and 234 nm
have sharp cut-off characteristics, i.e., steep effi- (214Bi/214Po). These values are slightly higher than
ciency curves (Reineking et al., 1984). the AMADs measured with the impactor. In other
The size range of an impactor can be lowered to words, the comparison of these measurement
smaller sizes by increasing the value of the results shows that the value of the AMTD measured
Cunningham correction, i.e., by applying low pres- with the diffusion batteries was similar to that of
sures in the impactor. Such low-pressure impactors the AMAD measured with the impactor with the dif-
can be used to separate particles down to 50 nm ferences being less than about 10%.
(Hering et al., 1978; 1979). Since impactors are used
to measure the size distribution of attached radon Combined Diffusion and Impaction Methods. A
progeny, the unattached clusters of radon progeny combination of a cascade impactor associated in

89
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Knutson (1988) compared the direct measurement

data obtained by a diffusion battery with data simul-
taneously measured by an electrical aerosol size ana-
lyzer (EAA), which sizes particles by means of their
electrical mobility. The radon progeny attachment
theory (Porstendörfer, 1994) was then used to calcu-
late 218Po particle size distributions from the number-
weighted particle size distribution measured by the
EAA. Although indirect measurements agreed very
closely with direct measurements, a systematic differ-
ence was observed, i.e., geometric mean diameters
from the direct measurements were larger than those
from the indirect measurements, which may be attrib-
uted to the approximations made in the indirect
measurements.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

In general, the primary advantage of the indirect
method is the much finer size distribution relative to
the rather coarse distributions obtained from diffu-
sion batteries or cascade impactors. On the other
hand, the main problem associated with the applica-
tion of the indirect methods is the necessary conver-
sion of the measured number size distribution to an
activity size distribution through attachment theory.
Figure 5.18. Comparison of activity size distributions measured Thus, a reliable determination of the activity size dis-
in closed rooms without additional aerosol sources with the tribution of the short-lived radon and thoron daugh-
diffusion battery technique, yielding diffusion equivalent
diameters (upper panel) and with the cascade impactor, providing
ters is only possible through direct measurements.
aerodynamic equivalent diameters (lower panel) (Reineking
et al., 1988).

5.4 Retrospective Measurements

series with a granular bed diffusion battery was A number of factors limit the accuracy of recon-
developed by Boulaud and Chouard (1992) and structing the historical radon exposure of subjects in
Tymen et al. (1992), covering a broad range of par- residential epidemiological studies. The most import-
ticle diameters from 0.0075 to 15 mm. The inertia ant of these are: the residential history of the sub-
unit comprises an eight-stage cascade impactor with jects, radon exposures elsewhere, and changes that
an operational cut-off diameter in the 0.35 –7.5 mm may have occurred in the radon levels in current and
range. The diffusion unit consists of six parallel previous residences. Most people change residence a
pipes 20 cm long and 4 cm in diameter containing number of times during their lifetime. Contemporary
granular beds of different depths. The diameters of radon measurements should, therefore, be made in
the beads vary between 1 and 5 mm depending on as many previous residences as possible if a reason-
the desired collection efficiency. The sixth pipe is able estimate of the time-weighted cumulative expos-
empty and serves as a reference filter. The six filters ure to radon is to be made. As radon levels in a
collect particles that have passed through the im- residence may change over the years due to changes
pactor and the different granular beds. After sam- in lifestyle, energy conservation practices and in the
pling, all filters and collecting plates are analyzed fabric of a dwelling the contemporary radon activity
by alpha spectroscopy. concentration may differ substantially from those in
the years of most relevance for the induction of lung
Comparison between Direct and Indirect Methods. cancer. In dealing with subject residence mobility, it
The direct measurement using the diffusion battery should also be noted that as indoor radon levels in
method involves tedious, complicated, and time- most countries are log-normally distributed persons
consuming procedures and detection of radon progeny in high radon dwellings on moving are more likely to
activity concentrations using this method is very diffi- move to dwellings with lower rather than higher
cult for low radon progeny levels. Hence the indirect radon levels. High mobility, therefore, reduces the
method via the measurement of the number size dis- variability of the exposure in the study subjects and
tribution may indeed present a viable alternative. To may necessitate an increase in the size of the
compare both direct and indirect methods, Tu and required study sample in the interests of maintaining

90
 Radon and Its Progeny Detection Systems and Measurements

an adequate study power. This is of most relevance to complex. It is dependent on parameters such as the
studies carried out in the USA and Canada, which characteristics of the room aerosol, room geometry, air
are societies with high residence mobility. movement patterns, and ventilation rate (Cornelis
In order to address these difficulties, there is a et al., 1993; Walsh and McLaughlin, 2001). Under
need for alternative approaches to determine radon typical indoor conditions, a radon exposure of approxi-
exposure which are not based on the measurement mately 1 kBq m23 yr might be expected to give rise
of contemporary radon in present and past resi- to a 210Po surface implanted activity of approximately
dences. A number of such approaches have been 1 Bq m22. Po-210 emits a 5.3 MeV alpha particle and
developed. These are largely based on the measure- in a thin surface layer of glass can be accurately mea-
ment of the long-lived radon progeny 210Po or its pre- sured using surface barrier detectors and PICs. Their
cursor 210Pb, originating from radon in the indoor use for this purpose has been largely confined to la-
air and which have built up in a variety of household boratory calibration work. For practical and economic
objects or even in the skeleton of exposed persons. reasons in large-scale surface trap field studies, the
210
These approaches, in principle, make it possible to re- Po activity is measured in a dwelling by mounting
construct the radon exposure of persons over past SSNTDs on the surface of chosen glass objects.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

decades. Three such techniques involving these long- Protocols have been prepared to assist in the selection
lived radon progeny are discussed here: 210Po surface of suitable glass artefacts for such purposes
traps; 210Po volume traps; and skeletal 210Pb. (McLaughlin, 1998). A variety of SSTND configura-
Generically, these techniques are called retrospective tions and methods of analysis are in use in which
radon techniques. They are time-integrating methods. alpha tracks from the surface 210Po alpha activity are
distinguished from those due to the intrinsic alpha ac-
tivity of the glass (Falk et al., 1996; Fitzgerald and
5.4.1 Surface Traps
McLaughlin, 1996; McLaughlin, 1998; Trotti et al.,
It has been known for over 100 years that expos- 1996). Using modified versions of the Jacobi (1972)
ure to radon could give rise to an “active deposit” of room model together with either standard room para-
both its short-lived and long-lived progeny on sur- meters or best estimates, it is possible to estimate the
faces such as glass (Crookes, 1903). The activity mean radon activity concentration to which the glass
measured on the surface of the glass was found to was exposed. If, for example, 210Po is measured on a
have two main components. One component could personal glass object of known history, such as the
be easily removed by simple cleaning, while the glass covering a family photograph of known age, this
other was permanently implanted and could only be measurement can be used to estimate the radon ex-
removed by abrasion of the glass itself. The perman- posure of an individual over past decades, even if
ently implanted component arises as a result of changes of residence have occurred. It should be
alpha recoils following the decays of 218Po and 214Po noted that this method is non-destructive as it does
and is found typically up to a maximum depth of ap- not damage the glass surfaces.
proximately 100 nm into the surface of the glass. The behavior of radon progeny at the air–glass
When a glass object such as a mirror is exposed to surface interface and of factors such as the mass
indoor air containing radon, its short-lived progeny loading of dust deposits and routine cleaning pro-
activity deposited on its surface will over time give cesses can influence the efficiency of alpha recoil im-
rise, due to the alpha recoil implantation process, to plantation. These topics have been the subject of
a build-up of long-lived progeny such as 210Pb considerable study both by computer simulation and
within the glass (Cornelis et al., 1993). experimentally (Roos and Samuelsson, 2005; Roos
It was first proposed in the late 1980s that surface and Whitlow, 2003). The findings of such studies
implanted 210Po could be used as a retrospective have been of importance to the development of
monitor for radon exposure (Lively and Ney, 1987; surface trap field protocols and in the interpretation
Lively and Steck, 1993). The half-life of 210Pb of 22 of the implanted activity measurements. An import-
years controls the rate of growth for the build-up of ant and critical assumption in the surface trap
this nuclide and its descendant, the alpha emitter method is that 210Pb resulting from recoil implant-
210
Po, in the exposed glass. Glass objects in a dwelling ation remains effectively immobile in the surface
exposed to radon and its airborne progeny can be layer of glass over exposure periods comparable to
therefore considered to constitute a 210Po “surface the lifetime of individuals exposed to radon in their
trap.” This surface trap can be considered as a form of homes. By means of ion beam implantation, the dif-
record of the integrated historical activity concentra- fusion of 209Pbþ ions in soda-lime glass has been
tions of radon and its short-lived progeny in the studied under conditions that mimic the alpha recoil
indoor air. The relationship between the radon con- implantation of 210Pb (Ekman et al., 2006). No stat-
centration in a room and surface implanted 210Po is istically significant loss of 209Pb from the glass was

91
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

observed for annealing temperatures as high as volume traps. These may be divided into two categor-
600oC. Extrapolating to room temperature, this ies. Included in the first category are spongy and
work implies that alpha recoil implanted 210Pb in porous household materials, such as mattresses and
glass from the decay of 222Rn will be very effectively other soft furnishings found in all dwellings. Radon
retained in the approximately 80 nm surface layer of gas may freely diffuse into these porous materials
the glass in which it is located. and will decay within them. The radon progeny 218Po
A number of studies involving the use of glass so formed will deposit on the surfaces of the pores.
210
Po surface traps as retrospective radon monitors This will give rise to a build-up within the porous
have taken place (Mahaffy et al., 1993; Steck and volume trap of long-lived radon progeny such as
210
Field, 1999). They have, for example, been used in Pb and 210Po (Samuelsson and Johansson, 1994).
Europe in the former uranium mining districts of By means of radiochemical methods, the 210Po in
the east of Germany, in a Swedish epidemiological small samples of such volume trap material can be
case – control study of non-smokers, and in high measured (Oberstedt and Vanmarcke, 1996), assum-
natural radiation areas in Yugoslavia (Falk et al., ing uniform distribution within the volume trap. For
2001; Žunić et al., 1999). In the Swedish study, the a typical volume trap, such as the sponge filling of a

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

surface trap measurements were used to provide an sofa, an inbuilt 210Po specific activity of approximate-
alternative estimate of individual radon exposure ly 0.05 mBq cm23 (kBq m23 yr) would be obtained.
which could be compared with that based on contem- Protocols have been developed to reduce background
porary radon measurements. This study also effects from radon progeny carried into the pores by
allowed the usefulness of the retrospective tech- household dust. The principal advantage of the
nique to be assessed. Results from 315 measure- volume trap method is that it is a direct monitor of
ments on 165 persons were evaluated and in most the radon gas activity concentration in the dwelling
cases, retrospective measurements were made on in the past. In comparison to surface traps, another
two glass objects associated with the same individ- advantage of the volume trap method is that the
210
ual. It is estimated for this study that the precision Po concentration is independent of aerosol condi-
of the exposure assessment by this technique is ap- tions and other room parameters, which directly
proximately 20%. influence the deposition of radon progeny onto sur-
A residential epidemiological study in Missouri faces. There are, however, a number of practical dis-
showed significantly increased lung cancer risks advantages with this method. The principal one is
when long-term radon exposures were estimated on that it is a destructive method requiring the permis-
the basis of glass surface trap measurements. sion of the dwelling occupant to remove samples of
Exposure estimates for the same subjects based on material from household furnishings. Due to the rela-
contemporary radon in air measurements showed no tively high costs associated with radiochemical ana-
significant increased risks. This suggests that the lysis, the cost of using it in large-scale surveys might
glass surface trap estimates are a more relevant ex- prove prohibitive in comparison to the non-
posure proxy than contemporary air-based estimates destructive surface trap method. This volume trap
for relating past radon exposure to lung cancer risk method has been used in surveys of high radon dwell-
(Alavanja et al., 1999; Lagarde et al., 2002). ings in Norway, Germany, and Serbia. In these field
In this context, it should be noted that it is not surveys, it has proved to be a viable field retrospective
radon gas but its short-lived radon progeny, in par- technique (Paridaens and Vanmarcke, 1999).
ticular the unattached fraction that delivers the The second category of volume trap methods
most significant doses to the bronchial epithelium. exploits the property of the solubility of radon gas in
Glass surface trap measurements correlate mainly commonly used polycarbonate materials such as CDs
with the concentration of unattached radon progeny and DVDs (Dimitrova et al., 2011; Pressyanov, 2012;
in the air which plate out on surfaces, such as Pressyanov et al., 2001). The polycarbonate material
glass (Cornelis et al., 1992; Lagarde et al., 2002; of a CD will absorb radon which will subsequently
Walsh and McLaughlin, 2001). Therefore, surface decay within the material. The latent tracks pro-
trap measurements should in principle be more duced in the polycarbonate material by the alpha par-
relevant to dose and risk quantification in residen- ticles following the decay of the absorbed radon and
tial radon epidemiological studies than radon gas its progeny will be directly proportional to the mean
measurements. activity concentration of the radon in the air of a
dwelling. These tracks can be made visible for count-
ing by electrochemical etching. To eliminate a back-
5.4.2 Volume Traps
ground contribution to the track density from
Another approach to the retrospective assessment non-absorbed radon and its progeny in the air above
of radon exposure is the use of what are termed 210Po the CD, a surface layer of approximately 80 mm is

92
 Radon and Its Progeny Detection Systems and Measurements

removed by etching. The track density below that time interval and at a specific location. Several
depth is proportional only to the absorbed radon and measuring techniques are currently available.
its ingrown progeny. This second category of a volume For example, eyeglass lenses were used for the es-
trap method also has the disadvantage of being a timation of the personal exposure (Fleischer et al.,
destructive method, but the content of CDs and 2001). They are composed of poly-allyl-diglycol car-
DVDs can, if required, be saved prior to the etching bonate (PADC). They were calibrated in a radon
procedure. chamber and tested by being worn for various
periods from 1 to 5 years. Average radon activity con-
5.4.3 In vivo Measurements of 210Pb centrations for wearers ranged from 14 to 130 Bq
m23 (Fleischer et al., 2001).
Estimates of long-term radon exposures can also, A monitor for personal exposure measurement in
in principle, be made by the in-vivo measurement of residences was reported by Harley et al. (1991), util-
210
Pb in the human skeleton in a low-level shielded izing a solid-state-nuclear-track detector (SSNTD)
gamma counting chamber. The feasibility of such an as the radon detector and a CaF2 chip as a gamma-
approach was investigated by researchers at BfS detector. The device was tested in 52 homes in the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Berlin. In this work, measurement of the 210Pb skull Chicago area with 84 occupants wearing the
activity of a small number of volunteers, who had detectors.
been exposed to high levels of radon, was carried out In China, Detao and Fuqi (1997) developed a
using low energy germanium detectors in a shielded passive radon personal dosimeter with electrostatic
chamber. A major background problem with this ap- collection by electret which greatly improved the
proach is that only approximately 2% of skeletal sensitivity of radon monitoring.
210
Pb is estimated to be due to the inhalation and Two versions of this personal dosimeter are cur-
subsequent decay of indoor air short-lived radon rently available, a “sensitive” and a “less sensitive”
progeny. On average, 86% of the 210Pb is due to in- “miner radon personal dosimeter”. The sensitive
gestion and approximately 12% as a result of direct “miner radon personal dosimeter” was used in
inhalation of atmospheric 210Po. Smoking and the copper, lead, and zinc mines, where radon dose was
consumption of some alcoholic beverages can add surveyed for 100 different exposure scenarios of
substantially to the body burden of this nuclide workers. While this dosimeter type is suitable for
(Salmon et al., 1998). For human subjects with a monitoring individual cumulative doses for several
high exposure to radon over many years, such as 20 days, the “less sensitive” type “miner radon personal
years of exposure to 2000 Bq m23 of radon, a meas- dosimeter” is suitable for monitoring over periods of
urable contribution may be made to skeletal 210Pb. more than 1 month.
This assumption is based on the known correlation An electronic radon dosimeter based on deposition
between radon exposure and skeletal 210Pb in of radon progeny on a semiconductor detector coupled
uranium miners. This approach to long-term radon with an alpha spectrometer offers some advantages
exposure assessment in humans will for the foresee- compared with passive dosimeters (Streil et al., 2002).
able future only be realistic for subjects exposed to In that dosimeter, approximately the size of a small
very high levels of radon. Measurements of 210Pb in mobile phone, radon gas diffuses through a mem-
subjects living in the same areas and with a similar brane into the measurement chamber, with a semi-
dietary intake may be used to distinguish those in conductor detector placed opposite to the entry
dwellings with high radon levels from those in dwell- window. Charged radon progeny produced by decay
ings with low radon levels. Similar work in the USA inside the chamber are collected at the detector
uses the measurement of skull 210Pb as a means to surface due to the electric field applied between the
estimate lung exposure from the inhalation of radon detector and the chamber wall. The system detects
progeny, but in this case, this technique seems to be alpha decays from both the collected radon progeny
only realistic for individuals living in very high and the radon gas. A multi-channel-analyzer (MCA)
indoor radon environments for long durations processes all incoming events. An integral spectrum
(Laurer et al., 1999). and a record of five peak-areas (each assigned to a
single nuclide) at every time step are stored for com-
puting activity concentration and dose values. To cal-
5.5. Personal Monitoring for Radon and culate the dose values, the equilibrium factor and the
Radon Progeny dose conversion coefficient (dependence on particle
size and lung model) must be known. These para-
5.5.1 Personal Monitoring for Radon
meters can be changed by the user or transferred to
For personal monitoring of radon, radon activity the dose management system to be adapted to local
concentrations should be recorded over a certain exposure conditions. The computed dose and the

93
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

radon activity concentrations are displayed online measurement of the activity concentration of radon
during the exposure time and pre-established dose while applying an equilibrium factor, a personal ex-
limits can be watched by an alert function. posure to progeny measuring system might be a so-
A new electronic personal exposure meter for lution. This condition might be met in the case of
radon gas was reported by Karinda et al. (2008). The mining. The application of such a device will be
exposure meter consists of a radon diffusion subject to the national approving authority.
chamber and two silicon detectors with 200 mm2 The first recording of personal radon progeny ex-
active area each. Radon gas diffuses through 10 posure data began in some mines in the 1960s.
holes (4 mm in diameter) into the detection Personal monitors worn by individuals provide more
chamber. The holes are covered by a filter in order to information than the combination of radon or radon
prevent radon progeny in the outside air to enter the progeny activity concentrations at fixed locations
chamber. Thus, the device is independent of aerosol plus records of where time is spent. A number of per-
concentration and humidity. Alpha detectors inside sonal monitors were developed and tested for
the chamber register the alpha radiation emitted by uranium miners, but many did not survive the
the decay of 222Rn and its progeny (218Po and 214Po). rugged environment and rough treatment by the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The exposure meter is optimized with respect to miners. One such type still in use is the one first
short-term (days) and long-term (.1 year) measure- described by Duport et al. (1980) for the French
ments of indoor radon activity concentrations and uranium mines. This active device uses a pump to
personal radon exposure, while the low power con- collect the radon progeny, which then pass through
sumption allows long-term measurements for more collimators with different filters for energy discrim-
than a year without recharge interruptions. The ex- ination and are subsequently recorded on a polycar-
posure meter records measured activity concentra- bonate track-etch detector.
tion levels in adjustable time intervals allowing a Commercial real-time personal samplers and
time-resolved analysis. The low weight (150 g) and working level monitors are sometimes used in
small casing (113  29  62 mm) allows the expos- mining operations, mainly to document decay
ure meter to be carried comfortably on a person. Its product activity concentrations for dosimetric appli-
advantages over the film badge dosimeters are: cations. For example, Su (2007) developed a kind of
online information on exposure available (without passive radon/gamma personal dosimeter in China.
interrupting the measurement), time-resolved ex- There are also some miner radiometers (or person-
posure monitoring, and lower measurement uncer- al exposure measuring systems) available, which all
tainties achievable. include a sampling assembly with an active pump,
as given in Figure 5.12. They might include a radi-
ation detection assembly and data processing unit.
5.5.2 Personal Monitoring for Radon
Until now, there appears to be only one patent
Progeny
(though it has different numbers EP0021081A2,
In environments where the activity concentration EP0021081B1, US4385236) dating from 1980, for a
of (or exposure to) radon progeny cannot be judged portable instrument for selectively detecting alpha-
(or not judged with the required accuracy) by a particles derived from radon without an active pump.

94
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv011
Oxford University Press

6. Strategies for Radon and Radon Progeny

Measurements and Surveys

There are multiple reasons for measuring radon objective is to determine airborne activity concentra-
and multiple methods are available for its measure- tions for specific individuals, such as in epidemio-
ment. Motivations to measure radon activity concen- logical studies, then individual exposure assessment

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

trations are: is needed. However, if the primary objective is to de-
termine whether the radon activity concentration in
– to establish the range, mean, and distribution of a specific house or room is below a certain limit, then
radon activity concentrations in homes to inform areal measurements are the appropriate choice.
overall exposure estimates and national strategies, If the spatial and temporal exposure conditions
– to identify “radon prone” or similar areas to remain relatively stable, then individual exposures
support decisions on surveys, public advice, build- may reasonably be approximated by areal measure-
ing control, etc., ments. The objective of area monitoring for radon in
– to determine the indoor radon activity concentra- dwellings and workplaces is the determination of a
tion in specific premises to support decisions long-term average radon activity concentration rep-
about remediation or other radon control, and to resentative of the exposure of residents and workers.
confirm the immediate and ongoing effectiveness These measurements are preferably carried out over
of remediation, a long-term period to cover daily and seasonal var-
– to inform epidemiological investigations of the iations. In the case of short-term measurements,
health impact of radon exposure, appropriate corrective measures must be taken in
– to determine radon exposures in workplace situa- order to correct measurements which are not suffi-
tions where exposures might be high and where ciently representative. However, an individual as-
area-based measurements are inappropriate, sessment of radiation exposure is required, when a
– to investigate random and systematic variations in given individual frequently changes exposure sites
radon activity concentrations, e.g., short- and long- with different exposure conditions or when the ex-
term, to ensure that other measurements can be posure conditions at a given site are subject to con-
properly interpreted in relation to reference levels siderable spatial and temporal fluctuations.
or to support site-specific dose assessments, In the case of occupational exposures, areal mea-
– to estimate historic exposures in a specific location surements are commonly performed to investigate
or building, and whether exposures are below a certain established
– to investigate relevant indoor air parameters, e.g., reference level. If that level is exceeded, then further
F-value, to establish default parameter values for measurements may be required to demonstrate com-
general use in appropriate circumstances or to de- pliance with annual dose limits. In this case, indi-
termine site-specific values where defaults are vidual measurements might be required.
not appropriate. In terms of measurement equipment and meth-
These different reasons for measuring radon deter- odology, areal and individual measurements have
mine the appropriate measurement strategy and to meet different needs. In the case of areal mea-
impact on the choice of an appropriate measurement surements, measurements are taken with station-
method. ary devices which must be installed at positions
at which the activity concentrations and the activ-
ity size distribution of radon progeny are represen-
6.1 Objectives: Areal and Individual
tative of those to which individuals are exposed.
Measurements
In order to derive individual exposures from areal
The choice between individual and areal radon measurements, individual occupancy times and
and radon progeny measurements depends on the related physical activities must be considered
objective of the intended measurements. If the primary (Section 3.6.1).

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

An individual assessment of the exposure to radon practice, however, to assume typical values for equi-
and short-lived progeny can be achieved by a meas- librium factors derived from measurements made
urement device carried by the monitored person, under similar exposure conditions, e.g., in homes or
preferably worn outside on the upper part of the in mines. Hence, the conversion procedure contains
trunk. In the case of varying physical activities, an inherent element of uncertainty.
breathing rates must be monitored to assess lung While radon progeny activity concentrations can
doses. be converted into radon activity concentrations or
For practical reasons, areal measurements of vice versa, at least in principle, radon progeny size
radon and radon progeny activity concentrations are distributions for attached and unattached fractions
normally used to assess individual exposures. cannot be related to radon activity concentrations. A
strong dependence of bronchial doses on the size dis-
tribution of the inhaled aerosols has been observed
(NA/NRC, 1991). For example, the dose-exposure
6.2 Radon versus Radon Progeny
conversion coefficient of unattached radon progeny
Measurements
(less than 5 nm in diameter) is about an order of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Epidemiological studies of lung cancer risk follow- magnitude higher than the corresponding value for
ing exposure to radon and its short-lived progeny attached progeny (about 200 nm in diameter)
have been carried out for two defined population (Section 3.8.1). This effect of radon progeny size dis-
groups: workers exposed to radon in uranium mines tributions on dose cannot be captured by the radon
and the population at large exposed to radon in activity concentration, which adds another element
homes (Lundin et al., 1971; NA/NRC, 1999a). The of uncertainty to the relationship between radon ac-
measurement of all physical parameters relevant for tivity concentrations and bronchial doses and, in
radon lung dosimetry and related risk analysis is further consequence, lung cancer risk.
the primary objective of the assessment of the expos- Presently, the recognized lung cancer risk in resi-
ure to inhaled radon and radon progeny. The rele- dences or mines is based upon epidemiological
vant dose for the induction of bronchial tumors is studies using measurements of radon gas or radon
the dose to sensitive bronchial epithelial cells pro- progeny without regard to physical atmospheric
duced by inhaled radon progeny. Hence, the physical parameters. The first lung cancer risk estimate in
parameters required for these dose calculations are an epidemiological study was determined in under-
the activity concentrations of the short-lived radon ground mines using radon progeny (WL) measure-
progeny and their related aerosol size distributions, ments (NIOSH, 1971). Presently, the broad similarity
distinguishing between attached and unattached of lung cancer risks from either residential or mining
radon progeny. Although all radon progeny quan- studies suggests that radon with a central estimate of
tities are measurable, they are not suitable for equilibrium factor is an adequate surrogate for radon
large-scale surveys because of the relatively sophis- progeny. Both radon gas and WL are atmospheric
ticated, and therefore expensive, equipment requi- quantities and thus surrogates for the relevant car-
red. Therefore, the measurement of radon progeny cinogenic bronchial dose.
characteristics is commonly replaced by the meas- It should be noted that measurements have shown
urement of radon activity concentrations, such as in that the equilibrium factor, F, is negatively corre-
home and workplace radon surveys. For example, lated with the unattached fraction, fp, for conditions
radon progeny activity concentrations in uranium where the ventilation rate is relatively low (Section
mines were traditionally expressed in terms of the 4.3.3). Taking into account of this negative correl-
Working Level (WL) or the Working Level Month ation between F and fp, it has been shown that for
(WLM). This leads to the fundamental question whe- indoor air, the radon gas activity concentration is a
ther radon activity concentrations are an adequate more robust indicator of dose than the potential
surrogate for radon progeny activity concentrations. alpha energy concentration (PAEC) under a range of
If the equilibrium factor, F, between the individual aerosol conditions normally encountered (Section
radon progeny activity concentrations and the radon 4.6.2).
activity concentration for a specific exposure situat- Radon progeny (WL) measurements provide the
ion is known, then radon progeny activity concentra- most direct information for calculating dose. For this
tions can be converted to an equilibrium equivalent reason, occupational exposures, such as in some
radon activity concentration and vice versa. In both underground mines, require WL measurements
cases, the correct conversion from radon progeny because of legal requirements for documentation of
activity concentrations to radon activity concentra- dose. However, uncertainty in calculating the lung
tions hinges upon the accurate determination of dose remains unless the associated aerosol particle
the equilibrium factor (Section 4.5). It is current size distribution is measured or well known.

96
 Strategies for Radon and Radon Progeny Measurements and Surveys

Thus, in conclusion, radon gas measurements can identified in the Netherlands, then this might be con-
be an adequate surrogate for radon progeny and for sidered a high radon area. Then again, in countries
dose calculations. However, uncertainty remains like Finland or the Czech Republic where the arith-
unless two factors are well known, the equilibrium metic mean radon value is 96 Bq m23, an area with a
factor and the particle size distribution. mean value of 100 Bq m23 would not be so consid-
ered. Some countries such as Ireland and the UK
have specific but different definitions of a high radon
area. In Ireland, a High Radon Area is defined as any
6.3 Areal Surveys and Mapping 10 km  10 km grid square where it is predicted on
6.3.1 Goals of Radon Surveys the basis of the national survey that 10% of the
homes will exceed the Radon Reference Level of 200
Radon surveys form an essential initial step in the Bq m23. In the UK, a Radon Affected Area is defined
establishment of a national or regional radon as any 1 km  1 km grid square with a probability
program aiming to reduce the population risk. WHO that 1% or more of present or future homes will be
(2009) reviews the goals and organization of national above the Radon Action Level of 200 Bq m23. The

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

radon programs as well as the role of radon surveys European Union (EU) Directive on Basic Safety
and radon mapping. Font (2009) provides a review of Standards for Ionizing Radiation states that “Member
the goals, design, and quality assurance of radon States shall identify areas where the radon concentra-
surveys. When targeting the residential and work- tion (as an annual average) in a significant number of
place exposure, the key objectives of the surveys are: buildings is expected to exceed the relevant national
(1) Obtaining the distribution of the annual average reference level” (EU, 2014). Such definitions have
radon activity concentration to which the popu- practical implications for the development of a nation-
lation is exposed in a country or in an area. al radon action plan.
(2) Exploring the seasonal variation and correction The EU Directive on Basic Safety Standards for
factors so that radon measurements can be inter- Ionizing Radiation (EU, 2014) and the International
preted to determine the annual average radon Atomic Energy Agency Basic Safety Standards
activity concentration. (IAEA, 2014) include a requirement that EU and
(3) Finding the geographic areas where high radon IAEA Member States should have national radon
exposures are most likely (radon-prone areas). action plans.
(4) Exploring the radon activity concentration and Similarly, the WHO Handbook on Indoor Radon
exposures of workers at workplaces and in public (2009) gives guidelines for national radon programs.
buildings. School surveys and exposure of chil- The Directive includes a list of elements that should
dren is an important target area. be considered in such action plans. This list includes
(5) Targeted surveys for exploring the exposure of a strategy for radon surveys and information to es-
special population or housing groups. For tablish radon-prone areas. The WHO Handbook
example, the results of radon prevention mea- also states that national radon programs should aim
sures can be studied by measuring a random to reduce the overall population risk and the individ-
sample of new buildings. This is a new and im- ual risk for people living with high radon activity
portant aspect because the radon activity con- concentrations.
centration in new constructions determines the If the ultimate objective of a national radon action
likely future exposure to radon. plan is to reduce the national health burden due
to radon exposure, then caution needs to be exer-
As stated in (2), one of the common goals of radon cised when considering identified high radon areas.
surveys is to identify areas having elevated percen- In particular, the population density distribution
tages of dwellings with radon activity concentra- should be considered as an important input into the
tions above the national reference level. In the radon development of any strategy used to deal with high
literature, the terms “high radon area,” “elevated radon areas. In an average low radon area, but with
radon area,” “radon-prone area,” or even “radon a high population density, there may be more citi-
affected area” are used in this context. There is, zens at a high level of radon exposure than in an
however, as yet no universal definition for these average high radon area of low population density.
terms and they may be perceived and defined in a dif- The strategy to be adopted in dealing with such con-
ferent way in each country. For example, a country trasting situations should be developed on the basis
such as the Netherlands with an arithmetic mean of objective cost-effectiveness analyses rather than
indoor radon activity concentration of 13.5 Bq m23 is on the basis of automatic prioritization and target-
considered not to have a high radon area. If, however, ing of high radon areas for remedial and preventa-
an area with a mean value of 100 Bq m23 were to be tive action.

97
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

6.3.2 Sampling and Survey Methods approaches discussed in that report are also applic-
able to the sampling of radon activity concentrations.
It is clear from the published literature that no
standardized measurement protocol for indoor
radon surveys exists. The actual protocol used in an 6.3.2.1 Random Sample Surveys. Most
individual survey depends on many factors such as random sample surveys are based on a representa-
the objective of the survey and the resources avail- tive sample from the housing stock. The addresses
able. In the case of national surveys, as distinct from may be sampled, e.g., by using the postal addresses
regional or local surveys, a general consensus has obtained from the national postal office, the tele-
been developing over recent years on how such a phone register, customer listings of utilities, or the
survey should be conducted. The principal objective electoral registers. When estimating the mean popu-
of a national survey usually is to obtain the popula- lation exposure, the need for relevant correction
tion distribution of annual average radon activity factors should be taken into account, such as type of
concentrations. The database from a national survey house, building material, or population density.
can then be used as an information platform to plan A representative random sample from the popula-
tion provides the best basis for estimating the popu-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

and develop national strategies to deal with radon
exposures of the population. lation exposure. Population registers or census data
The following general aspects are important in can be used when choosing the participants. Persons
survey planning (Font, 2009; Särndal et al., 1992). under the age of consent can be replaced by the eldest
of the parents. Normally corrections are needed due
– Specification of the objective to variation in participation rate in different areas or
– Target population, parameters to be estimated house types.
– Inventory of resources, budget, staff, data process-
ing, and detectors 6.3.2.2 Stratified Sampling. In stratified sam-
– Choice of radon detector, installation, and collec- pling, the population is divided in subgroups called
tion of detectors “strata.” Each stratum is then sampled as an inde-
– Requirements as to time schedule and accuracy pendent subgroup. Stratified sampling is a powerful
required and flexible method that is widely used in practice.
– Data collection method, questionnaire design The method provides the following benefits:
– Information security and confidentiality (i.e., im-
plementation of guidelines on data protection) (1) Flexibility when a specified precision is wanted
– Sampling design, sample selection mechanism, for a specified subpopulation.
and sample size determination (2) Since practical aspects related to response, mea-
– Data processing methods, including editing and surements, and information material may differ
imputation greatly from one subpopulation to another, the
– Specification of formulas for statistical quantities choice of sampling can be made differently in dif-
and measures of precision ferent subpopulations to increase the efficiency
– Training of personnel, organization of field work of the survey.
– Allocation of resources to different survey operations (3) For administrative reasons, the survey organization
– Allocation of resources to control and evaluation can be divided into several geographic districts.

Preliminary or pilot surveys can be used to test the 6.3.2.3 Choice of the Strategy. In the surveys
survey strategy, detectors, exposure times, and proced- of exposure of the population to indoor radon, either
ure to install and collect the detectors. Questionnaire random sample surveys or stratified sampling can
forms and the questions play an important role in be utilized. Choosing the dwellings is based, for
getting data regarding factors affecting radon activity example, on population or dwelling registers. Many
concentration. Question design is affected by the goals biases can distort the results, so statistical expertise
of the survey and, for example, housing and occupancy is needed in the design. The accuracy of a random
characteristics in the country. Questionnaires should sample survey is highest in high population density
be pre-tested by a representative test group in order to areas. Using stratified sampling, the varying popu-
get feedback. lation densities and special conditions of different
A strict ongoing quality assurance program is regions can be taken into account.
required for the radon measurements including The strategies for workplace surveys are variable.
traceability to a qualified primary standard. A high-quality representativeness regarding build-
Although ICRU Report No. 75 (ICRU, 2006) refers ings has been achieved in surveys where all schools,
to the sampling of radionuclides in environmental day-care centers, or public buildings in the country
media, the statistical principles of the sampling or region have been measured. The number of

98
 Strategies for Radon and Radon Progeny Measurements and Surveys

measurements in the buildings and the choice of the surveys. Even though it is possible to manage tech-
rooms measured are important factors in determin- nical problems due to the effect of humidity and heat
ing the exposure to workers or citizens staying in with charcoal detectors, the main problem with their
the building. A strategy has often been to choose use in national surveys arises from the 3.82 d half-life
work spaces with the highest risk, for example, of radon. In effect, because of the half-life of radon,
underground spaces and work rooms on the lowest charcoal detectors have a radon “memory” of less
level in high rise buildings, an approach which may than 2 weeks and therefore cannot be used to deter-
yield biased results. mine the long-term (.3 months) average radon con-
A marked difference between radon activity con- centration in a dwelling. Charcoal detectors are, on
centration during working hours and the diurnal the other hand, quite useful as screening devices and
average may set special requirements for workplace have in the past also been used in connection with
surveys. The diurnal variation may be marked and survey results in the USA (Price and Nero, 1996).
time-resolved radon monitor measurements may be In a dwelling, it is recommended that radon be
needed for accurate exposure determinations (see measured in at least two rooms. High occupancy
Section 7.2). Also, the effect of seasonal variation rooms, such as the main living room and principal

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

may be different in workplaces compared with resi- bedroom, are preferred. As most of the indoor radon
dential buildings. in a dwelling comes from the ground subjacent to
the building, the lowest inhabited level of the build-
6.3.2.4 Period of Measurement. In a national ing is recommended as one of the measurement loca-
survey, the measurement period should ideally be 1 tions. Positioning of the detectors in a room is
year. For practical reasons, this may not be possible, important both to obtain a representative value of
in which case a measurement period of at least 3 the long-term average radon activity concentration
months is recommended. Both short-term and ann- in the room and for protecting the detector from con-
ual radon measurements were made in 158 homes ditions likely to affect it negatively. For these
in a radon prone area in Iowa (Barros et al., 2014). reasons, the detectors should not be placed close to
In this case, short-term and annual measurements doors and windows in order to reduce the effect of
were highly correlated (r ¼ 0.87). Section 6.4 deals the intake of outdoor air or of air from other parts of
with the accuracy of short- and long-term measure- the dwelling. The location at which the devices are
ments as predictors of the annual average. Using installed should be representative of the average
seasonal correction factors, the radon activity con- ventilation in the room. Distortions of strong sources
centrations determined for measurement periods of heat such as direct sunlight, ovens or radiators,
between 3 months and 1 year can be converted into and heat-emitting facilities on windowsills should
annual values. If a national survey is carried out in be avoided, as should conditions of high humidity as
a phased fashion (say four sequential 3-month meas- may be found in bathrooms and cooking areas. The
urement periods or three 4-month periods) then in detectors should be deployed at least 10 cm from
principle the results could be used to generate spe- walls, should be approximately in the normal
cific seasonal correction factors. Seasonal correction breathing zone (1 –2 m height) and should be in-
factors will be discussed further in Section 7.3.5. accessible to children and pets. The surface where
Apart from the use of seasonal correction factors, the device lies should be non-masonry. The detectors
other temporal correction factors may be needed de- should be used under normal living and ventilation
pending on the detector type used. For example, in conditions and not with the dwelling purposely
the case of nuclear track detectors, it may be neces- sealed or closed during the measurement period.
sary to correct for detector background, for the aging The dates of the commencement and cessation of
and fading of tracks and for track saturation effects measurement must be recorded. Spatial variation
(Section 8.2.2.2). within a house is reviewed in Section 7.1.2.
Because of spatial variation of indoor radon, more
6.3.2.5 Detector choice and deployment. than one detector may be required when carrying
Passive, etched-track (nuclear track) detectors are out areal monitoring of indoor workplaces (Section
the detectors of choice for large-scale national 6.5.1). For example, Public Health England gives a
surveys but electret ionization chambers can be con- guide to employers for the number of detectors
sidered as a suitable alternative. As national surveys required for areal monitoring of radon in a work-
usually require radon to be measured in some hun- place. See website: www.ukradon.org/information/
dreds of dwellings, the use of active electronic radon workplace (accessed January 2015)
monitors would be impractical, primarily on the basis
of cost. The use of charcoal canisters to measure 6.3.2.6 Examples of Survey Practices.
radon is also not recommended for long-term national Reviews of radon surveys and mapping in the USA

99
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

and Europe are available (Chambers and Zielinski, random sample, which would have involved an un-
2011; Dubois, 2005, Dubois et al., 2010; EPA, 1993). feasibly large fraction of the approximately 8000
The WHO Handbook on Indoor Radon (2009) gives a Italian towns. Therefore, a simple random sampling
brief summary and guidance on radon surveys. was used only for the 50 “large towns” (i.e., over 100
Example results from surveys in European and 000 inhabitants), whereas smaller towns were com-
Non-European countries are given in Tables 4.3 and bined into randomly selected 150 clusters. The final
4.4. Section 1.1 reviews the UNSCEAR (2008) total numbers of sampled dwellings and towns that
summary of the worldwide radon database. resulted were slightly higher than the foreseen 5000
An overview of radon surveys in 32 European and 200, respectively.
countries (Dubois, 2005) shows that in the majority The UK and Austrian surveys represent random
of countries, random selection of dwellings is basic sample surveys based on dwelling sampling. In the
to the survey. In many of these cases, the sampling UK, it was estimated that the sample reflected the
frequency has been proportional to population population distribution without corrections (Wrixon
density and special emphasis has been given to et al., 1988). In the Austrian (Friedmann, 2005)
radon-prone areas. In some countries, the surveys survey, the measurements were made mainly during

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

were based on multiple types of previous survey spring and autumn. Homes were selected at random
results (France, Germany, Switzerland) or the from the telephone register. The chosen number of
survey material was taken from the first stage, non- measurements in an area was proportional to the
representative local surveys. Census data ( popula- number of inhabitants in the area. Finally, 1 in 200
tion registers)-based sampling has been used only in homes (1 in 700 inhabitants) were selected. The
few cases (Finland, Portugal, Norway). sample size was reduced in larger cities (multistor-
The national radon surveys in the USA ied houses) by concentrating on ground floor homes
(Marcinowski et al., 1994) and Italy (Bochicchio and homes in suburbs for the estimate of radon risk
et al., 2005) are good examples of the stratified ap- from ground sources. Later corrections were made in
proach. In the US study, a stratified, three stage order to obtain the mean exposure in larger towns.
sampling procedure was used to identify differences Approximately 40 000 measurements were per-
in radon levels across EPA’s (US Environmental formed in about 16 000 rooms, with several detectors
Protection Agency) 10 regions, and to ensure ample in each house.
coverage for areas that were expected to have highly The national radon survey in Ireland was geo-
variable radon levels, resulting in 22 regional strata. graphically based using the 10-km grid squares of
In each region, counties were assigned into one of the national grid as the unit area (Fennell et al.,
three radon potential categories: high, medium, or 2002). Radon measurements were carried out in
low. The division was carried out using previous 11 319 houses throughout the country.
radon survey results and geographical data. A multi- The latest French survey is based on previous na-
stage process was used to select the housing units, tional surveys complemented with more recent data,
including Census Bureau information on urbaniza- yielding a total of 12 261 measurements (Billon
tion, residence heating, and ventilation characteris- et al., 2005). When estimating the exposure of the
tics. The total sample size was 11 423. The survey population, corrections for season of measurements,
data were collected through personal interviews housing characteristics, and population density
with respondents and by placing radon measure- have been implemented. According to a national oc-
ment devices in their homes for 1 year, with at least cupancy study, an indoor occupancy factor of 0.9 was
one detector on every floor. The interview covered 77 used.
questions related, for example, to characteristics of For Switzerland and Germany, the distributions
the resident’s home and time spent in different of radon activity concentrations are based on nation-
levels of the home. To generate an unbiased popula- al radon data from different sources (Menzler et al.,
tion estimate, it was necessary to use sampling 2008). In the Swiss study, measured radon values
weights to reflect the unequal probabilities of select- were corrected for seasonal effects, adjusted for floor
ing the sample. A special Quality Assurance Project level, averaged by dwelling, and weighted with the
was developed, including procedures for data collec- population size for each community. The estimate of
tion, sample custody, detector analysis and calibra- the radon distribution in Germany was based on a
tion, and data processing. weighted selection of indoor radon studies. The em-
Italy (Bochicchio et al., 2005) has used a two-stage phasis on high-risk houses was taken into account.
stratified sampling scheme. In the survey, each of 21 In the latest Japanese study, 3900 homes were
regions were subdivided into the two strata of large selected and the number of homes in each prefecture
and small towns, giving a total of 42 strata. The was allocated by the Neyman allocation method to
door-to-door approach prevented the use of a simple reduce the variance in population-weighted radon

100
 Strategies for Radon and Radon Progeny Measurements and Surveys

activity concentration (Suzuki et al., 2010). The Burke et al. (2010) in Ireland, the weighting tech-
method utilizes the size of population and the SD of nique for correction of the bias due to over-sampling
the radon activity concentration in each prefecture of high radon areas was not effective when regional
based on a former national survey. weights were used. To be effective, it was necessary
Andersen et al. (2001) have used Bayesian statis- to apply weights at a localized level. Volunteer data
tics in the Danish survey order to get the best esti- are valuable in decision-making at the regional level
mate of the percentage of houses exceeding the due to adequate measurement density.
reference value. In the UK, it was also found that volunteers from
In the Indian survey, altogether 1500 measure- radon measurement campaigns and householders
ments were carried out in 25 regions over a span of 3 who were more willing to have a radon measurement
years (Ramachandran and Sathish, 2011). The had higher radon levels than those less willing
period of measurement was 90 d. A similar region- (Miles, 2001). More willing householders were likely
based survey has been carried out as a coordinated to be in a higher socioeconomic group living in a
project in 7 Arab countries with altogether 1426 detached house with good level of heating, and less
measurements (Al-Azmi et al., 2012). The survey likely to live in a block of apartments. In compari-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

aimed at a density of 1 – 2 detectors in each 1 km2. son, houses with a lower level of heating or upper
Cities from different parts of the countries were floors of apartments are more likely to have lower
chosen. Varying approaches were needed to achieve radon levels.
public participation in the survey.
Finland is one of the few countries where the na- 6.3.3.2 Large radon mapping data. In a
tional surveys are based on random sampling from Finnish study of a database of 92 000 houses, weight-
the population register. In this approach, corrections ing was applied by calculating the parameters in
are needed due to differences in participation rate in 1 km2 cells and weighting the cells by the number of
areas or house types (e.g., apartments and low-rise dwellings in the cell (Valmari et al., 2011). Such
houses) with different radon activity concentrations. random sample surveys provide the best basis for
In the latest Finnish random sample survey in decision-making and for exploring the future trends in
2006, these corrections decreased the population- national indoor radon activity concentrations.
weighted national average from 109 to 96 Bq m23
(Mäkeläinen et al., 2009). 6.3.4 Radon Maps
Reducing radon activity concentrations in new
buildings is an important challenge. Taking a Radon maps using a variety of data have been
random sample from houses that have received developed to evaluate the radon potential of an area,
building permission is an efficient tool to explore the i.e., an indication of the radon activity concentration
effect of preventive measures carried out at the na- to be present in this area. These data may include
tional level. The Finnish new construction survey in parameters, such as indoor radon measurements,
2009, in comparison with the previous national uranium or radium content of the soil, permeability
radon survey (Mäkeläinen et al., 2009), showed that and moisture of the soil, soil radon gas, and the
radon activity concentrations were reduced by 30% gamma signal from 214Bi. The radon potential is
due to new regulations and practical guidelines estimated indirectly from these parameters, and
(Arvela et al., 2012). then mapped. The maps are named after the main
Examples of results on workplace surveys have parameter used to estimate the radon potential.
been presented in Section 4.4.2. Some radon potential maps are a combination
of several parameters.

6.3.3 Use of Volunteer and Large Radon

6.3.4.1 Geological Maps. Geological maps of
Mapping Data
the type of topsoil mineral soils together with
6.3.3.1 Volunteer data. Many countries have uranium or radium concentration of the mineral
collected indoor radon data on a voluntary basis in soils and the underlying rocks form an important
connection with radon campaigns or from other pri- basis for radon predicting activity concentrations in
vately organized radon measurements. These kinds a specific area where no indoor radon measurements
of volunteer measurement tend to be biased due to are available. It is important to know the radon ac-
over-sampling of high radon areas, i.e., the average tivity concentrations of the near-surface materials,
indoor radon level computed from volunteer data as this may influence the indoor radon, which
will probably be artificially higher than the overall usually comes from the upper several meters of the
average indoor radon level computed for a country earth’s surface. Geological maps are useful for under-
from more representative data. In the study of standing the physical properties such as permeability

101
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

of the materials at the surface. Permeability of the 6.3.4.3 Radon in Soil Gas Maps.
soil and subfoundation filling materials is the most Measurements of radon gas in soil air can be used as
important factor affecting leakage flow of radon- a predictor of indoor radon. The radon activity con-
bearing soil into dwellings. Therefore, radon activity centration of soil air can be measured by passive or
concentrations are higher in areas of coarse top-soil active devices, following different protocols (Section
types such as gravel and sand and lower in areas of 5.2.1.3). The passive device provides a better esti-
less permeable soil types such as clay. mate of the soil radon because of long-term measure-
The most radon-prone areas of UK, Devon and ments. However, it has to be protected from the soil
Cornwall, lie on granites intruded into folded sedi- moisture, which affects the precision of the measure-
mentary rocks. The granites are characterized by ments. The active method involves measuring the
moderate to high levels of uranium. The radon poten- radon in the sample of soil air gas, collected from a
tial is a function of uranium concentration, mineral- probe driven into the ground. This method provides
ogy, and permeability. However, if purely uranium data quickly, but these short-term measurements
concentration or radium content of rock is used to es- may vary greatly due to daily, weekly, and seasonal
timate the number of homes above a threshold, they changes in soil and atmospheric conditions that are

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

will provide an inaccurate result, because the rela- averaged out during long-term measurements.
tionship between these parameters and radon po- Radon activity concentration in soil gas surveys
tential appears to vary between different geological using active devices on a national or large-scale
formations. For example, high permeability can level were carried in several European countries.
result in high radon potential even where the bed- The methods applied for mapping of the soil gas
rock uranium concentrations are moderate, as Ball radon were described in Barnet et al. (2005), Kemski
and Miles (1993) showed. Therefore, pure geological et al. (2005), Neznal et al. (2004), and Scheib et al.
maps should be used only in conjunction with indoor (2006). In national mapping practices based on soil
measurements to provide reliable measures of radon gas radon activity concentration, the permeability of
in homes. the topsoil mineral soil type is normally included in
the assessment procedure, because variations in
6.3.4.2 Aerial Gamma Radioactivity Maps. permeability of soil cause a greater variation in
Radioactivity maps are developed from the mea- indoor radon activity concentrations than variations
sured aerial gamma radioactivity data over a certain in soil gas radon activity concentration. Soil gas
region. The data are then used to quantify and de- radon activity concentration or corresponding
scribe the radioactivity of rocks and soils. The major- uranium concentration results are valuable, espe-
ity of the gamma-ray signal is derived from the cially when exploring areas with elevated natural
upper 20– 25 cm of surface materials. A gamma-ray radioactivity in soils. There is a lack of qualified
detector is mounted in an aircraft that is flown over studies on the success of the soil gas radon mapping,
an area at a certain altitude, which is usually 120 – e.g., in predicting houses exceeding the reference
150 m. The 214Bi signal is used to trace back the level in areas classified as high and low risk.
238
U equivalent, assuming that the uranium and its
decay products are in secular equilibrium. There is a 6.3.4.4 Indoor Radon Maps. Indoor radon ac-
good match between areas identified on aero- tivity concentration is highly variable; it is estab-
radioactivity maps as having high levels of surface lished that the average indoor radon activity
uranium and areas for which high levels of indoor concentration varies by more than an order of mag-
radon have been reported. However, the signal can nitude between different areas. Radon maps provide
be blocked by the water in the surface layer and information about the spatial variation of indoor
therefore the uranium content will be underesti- radon activity concentrations.
mated on the maps. Phillips et al. (1993) used the Radon maps based on direct measurements of
aerial gamma radioactivity data to predict areas indoor radon activity concentration form the least
with elevated radon potential in the USA and error-prone basis for residential radon maps. House
Appleton et al. (2008) used aerial gamma radioactiv- radon data have been obtained by national or target
ity data to predict areas with elevated radon poten- surveys (Section 6.3.2.6). The national surveys
tial in Northern Ireland. They compared the maps are better designed for statistical analysis because
obtained from geological and indoor radon data with of the selected sampling methodology-population
the maps modeled from the airborne radiometric weighting, but they contain too few measurements
and soil geochemical data. Although the analysis per sampling area. The data from target surveys
showed good correlation, the conclusion was that the undertaken to identify homes with high radon levels
airborne radiometric maps should be validated by in areas of known elevated radon potential contain
indoor measurements. many more results. These surveys are deliberately

102
 Strategies for Radon and Radon Progeny Measurements and Surveys

biased because of their purpose, but they still can the geology, the aerial radioactivity, the soil perme-
be used for mapping purposes. The long-term mea- ability, and the foundation type. The radon potential
surements are preferable for mapping purposes in was defined by the authors as low, medium, and
order to average the short-term radon variations high. Ielsch et al. (2010) developed a methodology to
(Section 6.4). derive a map of the geogenic radon potential in
There are some uncertainties in magnitude of the France. They determined the capacity of the geo-
mapped parameter arising from both measured data logical units to produce radon based on their ura-
and mapping contributions. The uncertainties in nium content. This initial map was then improved
measured data are due to the choice of measured by taking into consideration the major fault lines
homes, the duration of measurement, the method of and underground mines, which control the preferen-
measurement, and the temporal (diurnal, seasonal) tial pathways of radon through the ground.
variation. The uncertainties in the mapping are due
to the modeling distribution of activity concentra-
tions, the working assumptions about spatial vari- 6.3.5 Lognormal Modeling of Indoor
ation and the grouping of data points into the wrong Radon Data

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

geological category. Studies by Cohen (1986), Gunby et al. (1993),
The parameter mapped may be the mean indoor Hamori et al. (2006), Kim et al. (2003), Marcinowski
radon activity concentration, the proportion of et al. (1994), Miles (1994; 1998), Nero et al. (1986),
homes above a reference level, or the non-numerical and White et al. (1992) showed that the distribution
risk categories low, medium, or high. However, for of indoor radon activity concentrations in many
practical applications of policy options for reducing countries can be approximated by a lognormal dis-
the risk of radon exposure to the population, the tribution. This means that the logarithm of the
maps of the proportion of houses exceeding a refer- radon activity concentration follows a Gaussian
ence or action level are most useful. function. The reason why the radon activity concen-
Various types of area boundaries are used in the tration follows this distribution can be understood in
analysis of data and the presentation of maps. terms of multiplicative factors affecting the relation-
Boundaries can be administrative, geological, or ship between radium in the ground and radon in the
arbitrary such as grid squares. Geological and arbi- indoor air.
trary divisions are common choices due to straight- Miles (1994) showed that the measured indoor
forward determination of an area and simplicity radon activity concentration, Ri, in any home can be
of data analysis. Nevertheless, as pointed out by expressed by
Dubois et al. (2010), there are uncertainties of the
indoor radon measurements inside the area bound- Ri ¼ Ro þ Rsource  A  B  C  . . . ð6:1Þ
aries mainly due to the true variability of the radon
activity concentration within the cell, the number of where Ro is the outdoor radon activity concentration,
observations in the cell, and the uncertainties Rsource a term depending on the radium content in
related to the seasonal variability of indoor radon ac- the ground; and A, B, C. . . are terms depending on
tivity concentration. factors such as permeability of the ground beneath,
number and size of entry routes, under-pressure in
6.3.4.5 Combined Maps. Miles and Appleton the building, and ventilation of the building.
(2005) developed a method, which combines the The equation can be rewritten as:
results of indoor radon measurements by grid
squares and geological units in houses in order to lnðRi  Ro Þ ¼ lnðRsource Þ þ lnðAÞ þ lnðBÞ
produce Radon Affected Area maps of the UK. The þ lnðCÞ þ . . . ð6:2Þ
land area is first divided using a combination of
bedrock and superficial geological characteristics The distribution of ln(Ri – Ro) is expected to be
derived from geological maps. Then each of the indi- normal if there are a sufficient number of independ-
vidual indoor radon measurements is allocated to ent and randomly distributed terms. If ln(Ri –Ro) is
the appropriate bedrock– superficial geological com- normally distributed, then (Ri – Ro) is said to be log-
bination underlying it. The combined method allows normally distributed. The distribution of (Ri – Ro) in
for a better estimation of the number of homes above the UK within large areas, 5 km grid squares, and
the action level than that obtained from either within geological units were approximately log-
method separately. normal (Gunby et al., 1993; Miles, 1994).
Gundersen and Schumann (1996) developed a A lognormal distribution can be characterized by
method to derive a radon potential map of the USA its geometric mean (GM) and its geometric standard
based on the available indoor radon measurements, deviation (GSD). The probability density function

103
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

f(y) of normal distribution y ¼ ln(x) with mean m and This allows national and local governments to iden-
standard deviation s is tify the most affected areas, and to estimate the
number of homes exceeding the reference level in
1 ðy  mÞ2 each area. Different approaches and methods have
fð yÞ ¼ pffiffiffiffiffiffiffiffiffiffiffi e ð6:3Þ
2ps2 2s2 been used to calculate the proportion of homes above
reference levels.
The probability density function F(x) of a lognormal Assuming a lognormal distribution of indoor
distribution is radon activity concentrations, the proportion of the
distribution above the reference level (RL) can easily
1 ðlnx  mÞ2 be calculated. The procedure developed by Miles
FðxÞ ¼ pffiffiffiffiffiffiffiffiffiffiffi e ;x . 0 ð6:4Þ
x 2ps2 2s2 (1998) involves subtracting the outdoor radon activ-
ity concentrations from the measured indoor values
The arithmetic mean m of normal distribution f (y) is and taking the natural logarithm, i.e., ln(Ri 2 Ro).
the logarithm of the geometric mean (GM) of log- The arithmetic mean and the standard deviation of
normal distribution F(x): these corrected indoor values are then calculated.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi The proportion of homes above the reference level
GM ¼ N x1 x2 x3 . . . xn ð6:5Þ
RL, NRL, for a normal distribution y ¼ ln(Ri – Ro),
1 with mean m and standard deviation s, can be calcu-
lnðGMÞ ¼ ðlnx1 þ lnx2 þ lnx3 . . . þ lnxn Þ ð6:6Þ
n lated using the cumulative standard normal distri-
lnðGMÞ ¼ m ð6:7Þ bution function F, as follows:
ð lnðRLRo Þ
The standard deviation s of a normal distribution 1 ðy  mÞ2
NRL ¼ 1  pffiffiffiffiffiffiffiffiffiffiffi e dy
f(y) is the logarithm of the geometric standard devi- 1 2ps2 2s2
ation (GSD) of a lognormal distribution F(x):  
ln(RLRo Þm
¼1F ð6:9Þ
ln(GSD) ¼ s ð6:8Þ s

The GM is equal to the median of the lognormal dis- Other techniques have been developed to reduce the
tribution and the 68% confidence interval is given influence of the extreme values on the sample mean
by (GM/GSD; GM  GSD). and the sample standard deviation. Miles (1994)
Although the distribution of indoor radon activity applied a “sort technique” to calculate GM and GSD
concentrations usually follows a lognormal distribu- and found that this was a more accurate method
tion, this should be checked in each survey. For compared with using standard formulae, if sufficient
example, small areas with elevated radon activity data are available. Miles (1994; 2002) developed
concentration may affect the distribution. The statis- mapping methods to estimate GM and GSD for each
tical parameters calculated should include at least grid square when the data are sparse. Miles and
the arithmetic mean, the standard deviation, the Appleton (2005) showed that the Bayesian estimate
geometric mean, and geometric standard deviation. of GSD could be used to improve the estimates in
When reporting the results, it is important to areas where the data are sparse.
provide also an estimation of the uncertainty of the
key parameters. A brief review of the interpretation
of survey results is available (Font, 2009). 6.4 Long-Term versus Short-Term
In population-weighted surveys, the final goal is Measurements
to estimate the exposure distribution of the popula-
Strong temporal and seasonal variation is one of
tion of a country or region and the percentage of
the characteristics of indoor radon activity concen-
houses exceeding the reference levels. Many biases
tration. Therefore, the length of measurement and
can distort the result of radon surveys. The use of
the season of measurement affect strongly the esti-
adequate statistical experience in the data analysis
mation of the annual average radon activity concen-
is therefore important (WHO, 2009).
tration. Seasonal variation is considered in Sections
7.3.5 and 7.3.6.
Both short-term and long-term measurements
6.3.6 Mapping the Proportion of Dwellings
have been used in radon measurements. The length
above Reference Level
of short-term measurement varies normally from 2
To prevent members of the public receiving high d to 1–3 months. The US EPA (1992) defines mea-
exposures to radon, and to reduce average expo- surements shorter than 90 d as short-term measure-
sures, it is necessary to identify the areas at risk. ments. The length of long-term measurements

104
 Strategies for Radon and Radon Progeny Measurements and Surveys

required or recommended by national authorities is 6.4.1 Integrating versus Time-Resolved

typically from 1 month to 1 year. In the USA, short- Measurements
term tests with a duration of 2 – 7 d have often been
Time-resolved measurements are more inform-
used in radon testing.
ative than integrating measurements, giving typic-
In a 2005 survey of radon measurement techni-
ally the hourly radon activity concentration for the
ques and protocols in European countries, the long-
whole measurement period, in addition to the
term integrating passive technique was the most
average of the total period. Due to the high price of
common method (Synnott and Fenton, 2005).
such continuous radon monitors, time-resolved mea-
Measurement periods of 1 month to 1 year were
surements have been mostly used in research. These
used. Some countries such as Sweden, Finland, and
kinds of measurements carried out with several
France recommend that measurements are carried
instruments monitoring simultaneously can give in-
out during the heating season (October to April) as
formation of activity concentration differences, radon
during this period higher radon activity concentra-
sources, and radon transport between different spa-
tions indoors would normally be expected. Other
ces. Strong diurnal variation in radon activity con-
countries, for example, the UK and Ireland, carry

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

centration, when there is a clear difference between
out radon measurements over any 3-month period
the temperatures of day and night, normally indi-
throughout the year and apply seasonal correction
cates an important role for convective flow of radon-
factors (Section 7.3.5). Baysson et al. (2003) ana-
bearing soil air (Section 7.2).
lyzed 11 000 measurements in France and deter-
Time-resolved measurements are needed at work-
mined seasonal correction factors for 2 and 6 month
places when there is need to explore the true radon
measurements to obtain an annual average.
activity concentration during working hours. Such
A 1 year measurement is the best choice for the
measurements are needed when an integrating
determination of the annual average radon activity
measurement has given a radon activity concentra-
concentration, but in practice has the disadvantage
tion clearly higher than the action level. Figure 6.1
of a reduced detector return rate and a long delay for
shows an example of a workplace where the average
the measurement results. In national surveys
radon activity concentration during working hours
carried out in many countries, a 1 year measuring
(130 Bq m23) was only 3% of the average during 1
period has been widely adopted either using only
week (4900 Bq m23) (Reisbacka, 2008). A reduction
one detector for the whole period or two or more sub-
in the mechanical air exchange rate and changes in
sequent measurements covering the whole 1 year
the indoor pressure conditions during night-time in-
period. The latter approach gives valuable informa-
creases the radon activity concentration remarkably.
tion on seasonal variations.

Figure 6.1. Variation of the radon activity concentration in a university seminar room resulting from variations in the adjustment of
mechanical ventilation during workdays and the weekend, beginning on 15 December 2006.

105
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

6.4.2 Predicting the Annual Average using Table 6.1. Variation of the ratio of short-term to annual average
Short-Term Measurements radon activity concentrations for different measurement periods

Two main factors affect the value of short-term Measurement period Coefficient of variation, COV, of period/
measurements as predictors of the annual average annual average ratioa
in a home. First, the short period of time, which
Minnesota UK Finland
gives only a poor indication of the annual average 75 housesb 91 housesc 326 housesd
radon activity concentration. Second, lower statistical
accuracy as well as inaccuracies in calibration and Two days: closed 76%
background effects play a more important role in Four days: closed 70%
short-term when compared with long-term measure- One week 63%e
Monthly: normal 40% 45%
ments. Unless the detector accurately records changes
Two months 29%
in radon activity concentrations, the resulting esti- Three months 25% 25% 22%
mate for the average activity concentration may not Semi annual 17% 18%
be accurate. Strong day-to-day temporal variation
a

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

which can result from weather or house operational COV ¼ 100  (GSD21), where GSD is the geometric standard
changes have been observed. On the other hand, the deviation of the period/annual average radon activity
concentration ratio.
detector’s response is usually calibrated under steady b
Steck (2005), corrected for instrumental variation. Closed-house
radon activity concentrations. Blind testing has conditions during the 2 and 4 d measurements means keeping all
shown that temporal fluctuations in radon activity windows closed, keeping doors closed except for normal entry and
concentration and increased humidity had a negative exit, and not operating fans or other machines which bring in air
influence on the precision (Sun et al., 2006; 2008). from outside.
c
Miles et al. (2012).
The annual average radon activity concentration d
Arvela et al. (2015).
in living spaces (AALS) is commonly used for radon e
Derived from other UK results.
risk assessment purposes. However, occupancy time
and breathing rates are needed for an accurate dose
assessment. In an attempt to get a quick, inexpen-
sive estimate of AALS, single short-term (2 –4 d)
measurements in the lowest lived-in level under
closed house (usually winter) conditions have been
used. This approach yields only poor estimates in
many cases. In three separate groups of Minnesota
houses, the geometric standard deviation (GSD) of
short-term measurements about the AALS was 1.5 –
1.8 (Steck, 1990; 2005). This means that the 95%
confidence interval (CI) of AALS predicted from
short-term measurements covers a factor of 10.
Wintertime short-term measurements generally
over-estimated the AALS by about 20%, while short-
term measurements taken in all seasons were scat-
tered symmetrically around the AALS (Section 7.3).
Figure 6.2. Short-term screening measurements versus annual
Table 6.1 gives the coefficient of variation of the average radon activity concentration in Minnesota (Steck, 2005).
various temporal measurement intervals based on
Midwest USA, UK, and Finnish studies. The annual
average radon activity concentration at the measure- Figure 6.3a exhibits a typical seasonal variation
ment site was used as the “gold standard”: the results pattern with high radon activity concentrations in
show a COV of greater than 25% for measurements of winter and lower activity concentrations in summer.
duration less than 3 months. Figure 6.2 shows the For comparison, Figure 6.3b shows an atypical vari-
linear regression and the large variation between ation pattern with highest activity concentrations in
short-term screening measurements (2–4 d in the summer including some results from monthly and
same house) and the annual average radon activity short-term measurements from the same 1-year
concentration in about 200 houses in the Minnesota period (Section 7.3.6).
study (Steck, 2005). (Note: the distribution of season- The introduction of seasonal correction factors
al/annual average ratios in 91 UK houses measured (SCF) increases the prediction accuracy, provided
in 3 months periods during 2 years is shown in the length of measurement is long enough (see
Figure 7.6.) Section 7.3.5). In a Canadian study, two consecutive

106
 Strategies for Radon and Radon Progeny Measurements and Surveys

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 6.3. (a) Typical seasonal variation for houses in Finland with high radon activity concentration in winter and lower activity
concentration in summer (Arvela et al., 1988). (b) Sample results from one measurement location with monthly and short-term
measurements (Steck, 2005). This house in Minnesota, USA, represents an atypical seasonal variation with a summer maximum.

6-month measurements in 4508 homes were used provide a fairly large database of geographically dis-
(Krewski et al., 2005b). Predicted annual average persed, short-term monitoring data, can be used to
radon activity concentrations, based on SCF, were in predict annual average living-area radon activity dis-
reasonable agreement with the observed average tributions for regions, individual states, and individual
value. Roughly 15– 30% of the predicted annual counties. This type of analysis illustrates the feasibility
average radon activity concentrations were within of using long-term concentration measurements to
10% of the observed values. The concordance “calibrate” short-term data, even if the long- and short-
between observed and predicted values falling below term measurements are from different homes.
or above 150 Bq m23 approached 90%. The results Seasonal variations and SCF for short-term mea-
for shorter measurement periods in the Canadian surements are reviewed in Section 7.3.
study were not equally promising (Section 7.3.5).
Although a single short-term measurement in a
6.4.3 Predicting the Past Thirty Years of
home is a poor predictor of the home’s annual average,
Radon Exposure from Annual Radon
collections of such measurements can be used to
Levels
characterize regional annual average radon activity
concentrations (Price and Nero, 1996). In the USA, Health studies on the harmful effects of radon
the State Residential Radon Surveys (SRRS), which face a considerable challenge to estimate the radon

107
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

activity concentration over long periods in the past result correctly identifying a house above the action
based on areal measurements (Steck, 2009). The threshold ( predictive value of a positive test) or of a
normal practice in current epidemiological studies house below the action threshold ( predictive value
has been to use one or more year-long contemporary of a negative test). Generally speaking, the farther
radon gas measurements for retrospective radon ex- the bulk of the results are from the threshold, the
posure reconstruction. As an example, in most more effective the test is. For example, in the USA
studies of the combined North American radon risk where the action level is 150 Bq m23 and 90% of the
analysis, an attempt was made to monitor all charcoal canister test results were below that level,
in-state homes occupied for a period of at least 1 the correct classification rate was 93% (White,
year within the exposure time window of interest, 1994). On the other hand, in a region where only
5 – 30 years prior to diagnosis of lung cancer 60% of the results were below the action level, the
(Krewski et al., 2005a, 2006). The radon measure- correct classification rate fell to 55%.
ments spanned on average an exposure period of A UK study has compared the predictive power of
19.2 years covering 77% of the 25 years period. 1 week, 1 and 3 month measurement periods during
If past activity concentrations are significantly a 1 year period (Groves-Kirkby et al., 2006).

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

different from present activity concentrations, then Table 6.2 indicates the threshold levels above/below
a systematic bias may be introduced in the risk as- which there can be 95% confidence that the indi-
sessment. If radon exhibits year-to-year variations, cated annual level is greater/less than the Action
then the variation will have a tendency to obscure Level of 200 Bq m23. These results are given for
the risk. However, the studies reviewed in Section track-etch detectors. The results and conclusions
7.3.7 show that in most cases, the annual variation are bound to the regional distribution of the radon
in homes ranges typically between 20 and 30% activity concentration and cannot be applied directly
(COV), without showing a persistent temporal to other areas. Calculated estimates are based on
trend. At individual sites, however, Harley et al. GSD values of Table 6.1. The levels can be estimated
(2011) and Steck (2009) observed clear temporal as 200/GSD2 and 200  GSD2 where the GSD is 1.45
trends, depending on climate and exposure to wind. for 1 month and 1.25 for 3 months. The levels are in
House modifications and installation of new heating, good agreement with the observed UK values. It
ventilation, and air conditioning (HVAC) systems should be noted, however, that this kind of estima-
may be the reason for a few observed dramatically tion can only be carried out if the distributions of the
large radon changes. radon activity concentrations are rather similar.
New retrospective measurement techniques have
been studied in order to improve the accuracy in the
determination of past exposures. These approaches, in 6.5 Homes and Workplaces
principle, make it possible to reconstruct the radon ex- The primary goal of both residential and work-
posure of a person using, for example, household place radon monitoring is to identify the homes and
objects. Section 5.4 reviews the current experience workplaces where reference levels are exceeded.
achieved using these techniques. As an example in the Strategies of residential and workplace radon sur-
residential epidemiological study in Missouri, long- veys and the principles of identifying radon affected
term exposure was estimated on the basis of glass areas have been considered in Section 6.3.
surface trap measurements (Alavanja et al., 1999). Information from radon maps can be used to
support decisions to carry out radon measurements
6.4.4 Using Short-Term Measurements
to Make Action Decisions
Table 6.2. Threshold levels above/below which there can be 95%
Short-term measurements are often used to make confidence that the indicated annual level is greater/less than the
Action Level of 200 Bq m23 (Groves-Kirkby et al., 2006). The
remedial action decisions. Often the decision proto-
levels in parentheses give the corresponding level derived using
col uses a well-defined radon threshold for that the GSD-values of Table 6.1
action. The actual effectiveness of the remediation
action decision depends on the actual radon distri- Confidence level Track-etch Track-etch Track-etch,
bution among investigated homes, spatial and tem- (95%) (Bq m23) HPA advicea
poral variation of the radon within a given home, One week One month Three months
and measurement result variation due to intrinsic
Lower 75 109 (95) 130 (128)
errors from calibration and analysis errors. Upper 518 478 (421) 300 (313)
The correct classification rate of a testing protocol
may be the best single indicator of its effectiveness. a
UK Health Protection Agency (HPA) advice: seasonal correction
Useful measures include the probability of a test carried out (now Public Health England).

108
 Strategies for Radon and Radon Progeny Measurements and Surveys

in homes and workplaces. For example, if an indoor compliance with regulatory requirements. These
workplace is in a radon-prone area, then radon mea- requirements include determination of individual ex-
surements are usually recommended. For radiation posure or dose, recording of exposures, compliance
protection purposes, the appropriate measurement with reference levels and dose limits, information
result should be compared with the relevant refer- and training of workers and health surveillance for
ence level. If mitigation is carried out to reduce workers.
radon exposure, then repeat measurements should In advance of a decision on the appropriate moni-
be made to confirm the effectiveness of the mitiga- toring strategy, the purpose of the monitoring and
tion system and records of the measurements should the particular measurement conditions must be
be kept (WHO, 2009). Remediated premises should taken into consideration. Therefore, it is necessary
be re-measured periodically to ensure that radon to consider whether the monitoring is carried out to
levels remain low. Measurements should also be determine the exposure of a person to radon or
repeated after any significant building work or radon progeny, or to assess the radon situation in
changes to an operational cycle affecting exposure dwellings or at workplaces.
conditions such as changes to the heating, ventila- The objectives and principles for selecting either

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

tion, and air-conditioning operation. areal or individual monitoring practices have been
High radon activity concentrations have been reviewed in Sections 4.4.3 and 6.1. Table 6.3 pre-
observed in underground mines and as a conse- sents general recommendations for selecting an ap-
quence, the control of radon in mines is regulated. propriate measurement strategy. The detector
Mining environments and atmospheric conditions choice and deployment have been considered in
set special requirements for radon and radon Section 6.3.2.5.
progeny monitoring. Underground workplaces form Area monitoring at workplaces should monitor
a similar work environment with a potential radon the working area of one or more persons working
risk. In industrial buildings, large work spaces, under similar exposure conditions in order:
special indoor air environments, building and foun-
dation structures, and ventilation strategies affect – to determine the annual average radon activity
the assessment of radon entry and indoor radon ac- concentration, or
tivity concentrations as well as monitoring. Diurnal – to determine the average potential alpha energy
and seasonal variations in radon activity concentra- concentration of short-lived radon progeny or any
tion affect the monitoring practices. At workplaces, other relevant quantity which is appropriate for
diurnal variations in ventilation practices may have a the calculation of the lung dose and the effective
very remarkable effect on monitoring results (see dose.
Figure 6.1). The large range of individual occupancy
times in workplaces affects monitoring requirements. The measurements should be taken with stationary
In workplaces where workers’ exposure to radon is devices during the common operation cycle which
considered as an occupational exposure, radon and should encompass all work procedures affecting the
radon progeny monitoring is required to demonstrate exposure to radon and radon progeny. Comprehensive

Table 6.3. Recommendations for selecting an appropriate measurement strategy

Personal measurement Area measurement

Radon In homes Commonly not used Long-term measurement in rooms with high
occupancy time (e.g., living room and sleeping room)
for determining the compliance with reference
values
At workplaces Persons with frequently changed location at In cases where the equilibrium factor does not vary
workplaces where the equilibrium factor does not significantly; for determining the exposure to radon
vary significantly the occupancy time and the physical activity of the
monitored person must be recorded separately
Radon In homes Commonly not used Only occasionally (e.g., when particular aerosol
progeny particle sources influence the equilibrium factor)
At workplaces In specific exposure situations for persons who In cases where spatial fluctuations of the aerosol
frequently change their location at workplaces particle concentration can be neglected; for
where considerable spatial and temporal determining the exposure to radon decay products
fluctuations of aerosol particle concentration occur the occupancy time and the physical activity of the
monitored person must be recorded separately

109
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

knowledge of the operation cycle is therefore neces- (e.g., the air flow through fans or pumps), to
sary. The measurement device must be installed at a ensure sustained effectiveness of the mitigation
position at which the activity concentration and the system (WHO, 2009).
activity size distribution of radon progeny are repre- (6) If any significant building work has taken place
sentative of those to which the workers are being or changes to the operational cycle affecting ex-
exposed. The sampling of radioactive substances posure conditions has occurred, such as changes
should ideally be taken at a height of 165 cm for to the heating, ventilation, and air-conditioning
standing work and 110 cm for sitting work in the operation, then measurements would need to be
immediate proximity to the worker (VDI, 1980). repeated.
Deviation from this rule can be allowed if evidence (7) If passive instruments are used, it must be
can be provided that the sampling is taken in the ensured that the instruments are stored in a low
working area with the highest exposure to radon radon environment at times at which no moni-
progeny. Area monitoring at workplaces will prefer- toring takes place.
ably be applied in cases where spatial fluctuations
can be neglected. In order to calculate individual Long-term and short-term measurements and the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

exposures, the individual occupancy times of the effect of diurnal variation in radon activity concen-
person at the workplace must be recorded. tration have already been considered in Section 6.4.
For the demonstration of compliance with refer-
ence levels or exposure limits, both passive and active
6.6 Individual Exposure Assessment:
radon detectors can be utilized in workplace monitor-
Time-Resolved Measurements
ing. The following aspects might be considered:
In epidemiological case –control studies, radon ac-
tivity concentrations in living rooms and bedrooms
(1) Long-term monitoring with an hourly recording of current and past dwellings of lung cancer cases
device. This strategy gives both the long-term ex- and controls were normally assessed retrospectively
posure and the exposure during working hours. by means of passive radon detectors (Darby et al.,
(2) Passive instruments for screening purposes. In 2005; 2006). Based on these measurements, individ-
normal cases when the working time radon ac- ual radon exposures to the inhabitants were then
tivity concentration is lower than the non- estimated for a period of several decades in the past.
working time activity concentration, the inte- The important factors of differences in breathing
grated long-term radon activity concentration rate and workplace exposure were not considered in
below the reference level indicates that it is these studies. The period 5–30 years prior to diag-
likely that the radon activity concentration in nosis or death from lung cancer is currently consid-
working hours is below the reference level. ered as the relevant time of radon exposure. Because
(3) If the measured radon activity concentration a long period of time in the past has to be analyzed
obtained with a passive device only slightly for radon exposures for each individual in the
exceeds the reference level, then further measure- cohort, the question arises as to whether a few
ments with a continuous monitoring device and locally measured indoor radon activity concentra-
hourly recording can be applied. These results tions can indeed be used as a measure for individual
can be utilized for the calculation of the average radon exposures.
radon activity concentration in working hours. Different strategies have been applied to quantify
(4) In cases where the reference level using a indoor radon activity concentrations. Details of the
passive device is significantly exceeded, e.g., by various procedures commonly used in epidemiologic-
a factor of 2 or more, the reference level for al studies can be found in Darby et al. (2006). For
the working time average is also likely to be example, radon measurements were performed
exceeded. during the time of the study in the last available
(5) If mitigation measures have been taken to home occupied for at least 2 years, or in all dwellings
reduce radon activity concentrations, then in the study area, or in all dwellings occupied for at
follow-up measurements are required to test least 1 year in the previous 25 or more years. For
their effectiveness. Long-term measurements periods where radon activity concentrations could
should be made at the same locations as the ori- not be determined experimentally, they were esti-
ginal measurements; however, short-term mea- mated indirectly based on mean radon activity con-
surements may also be started at the same time. centrations obtained from individuals in the control
If the levels are sufficiently reduced, long-term groups. Radon measurements were commonly per-
tests should be repeated periodically (e.g., every formed using passive detectors over various expos-
few years), in addition to regular physical checks ure periods ranging from several months to 1 year.

110
 Strategies for Radon and Radon Progeny Measurements and Surveys

In order to extrapolate the measured data to mean strategies such as personal and time-resolved moni-
annual indoor radon activity concentrations, season- toring should be considered.
al adjustment factors were applied if available. In An assessment of the individual radiation expos-
most cases, one detector was placed in the living ure can be achieved by personal sampling in which
room and another one in the bedroom as these are the sampler is worn by that monitored individual.
the two most frequently occupied rooms. Measured The sampling must be taken in the breathing zone
radon activity concentrations were then weighted by of the worker.
the time the investigated individuals spent in either The objective of areal measurements with station-
room in order to calculate a mean radon activity con- ary devices is the monitoring of the working area of
centration for a given dwelling. Sometimes, changes one or more persons working under similar exposure
in building and ventilation characteristics between conditions. The measurement device must be in-
the investigated individual and the current inhabi- stalled at a position at which the activity size distribu-
tants of a dwelling were also taken into account. tion of radon progeny and the activity concentration
Details of the various procedures commonly used in are representative of those to which the workers are
epidemiological studies can be found in Darby et al. exposed. This method will preferably be applied in

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(2006). cases where spatial fluctuations can be neglected. In
Even if assessments of indoor radon activity con- order to calculate individual exposures, the individual
centrations in epidemiological studies as described occupancy times of the person at the workplace must
above are carried out with the utmost care, they may be recorded.
not necessarily be a good proxy for the individual
radon exposure. For example, the time the investi-
6.6.2 Comparison of Integral and
gated individuals spent outside their homes, e.g.,
Time-Resolved Personal Measurements
outdoors, in the office or in other dwellings, etc., is
not known due to the retrospective nature of the Measurement methods and techniques for time-
measurements. Furthermore, lung doses are deter- resolved personal monitoring are listed in Section
mined not only by the radon activity concentrations in 5.5.1 for radon gas and in Section 5.5.2 for radon
air but also by individual breathing patterns, which progeny. Time-resolved measurements giving typic-
depend on the physical activity of that individual. ally the hourly radon exposure assessment for the
Ideally, estimates of individual exposure could be whole measuring period are more informative than
improved with the use of personal passive monitors integrating measurements. Because of the relatively
or with active personal monitors that permit the de- high price of such radon monitors, time-resolved
termination of indoor radon activity concentrations measurements have mainly been used in research
as a function of time. In principle, such time-resolved efforts.
measurements could be used with individual breath- Harley et al. (1991) used a radon monitor for per-
ing pattern data to obtain better estimates of “indi- sonal exposure measurements in 52 homes in the
vidual dose.” However, for residential epidemiology Chicago area with 84 occupants wearing the detec-
studies, personal monitoring is currently not a prac- tors. They found that the integrated personal expos-
tical option, especially for long-term measurements. ure was 70% of that predicted by the first floor radon
activity concentrations. Personal radon measure-
ments were not well correlated with measurements
6.6.1 Comparison of Areal and Personal
made in basements.
Exposure Assessment at Workplaces
A new electronic personal exposure meter
In workplaces where workers’ exposure to radon is (Karinda et al., 2008) was used to measure radon ac-
considered as occupational, the determination of the tivity concentrations in 12 tombs located in the
individual exposure or dose is required to demon- Valley of the Kings, Egypt (Gruber et al., 2011).
strate compliance with reference levels and dose Because the active exposure meters used are easy to
limits. A decision on an appropriate measurement handle, the guards agreed to wear them for 2–3 d,
strategy should be based on a detailed consideration and individual radon exposures could be quantified
of the exposure conditions. The exposure conditions for the first time with high time resolution for these
to be considered include the aerosol characteristics, individuals. The results obtained demonstrate that
the ventilation conditions, the occupancy time of the occupational dose from radon exposure inside
the worker at the workplace, and the type of the the tombs depends on location and period of stay
work activity, which determines the inhalation rate. inside the tomb.
Depending on the exposure conditions, areal as well In another application of the electronic personal
as personal monitoring may be applied. For example, exposure meter (Karinda et al., 2008), this detector
because conditions in mines are highly variable, was worn by 23 individuals over a period of about

111
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

1 week to investigate whether indoor radon activity radon exposure measured with the personal expos-
concentrations measured at home can indeed be ure meter was on average a factor of 2 higher than
used as a measure for individual radon exposure the estimated exposure at home based on measure-
(Gruber et al., 2015). For comparison purposes, ments in the living room and the bedroom. This dif-
areal measurements were performed over the same ference was caused primarily by the higher radon
period with passive detectors placed in the living activity concentrations in the offices when com-
room and the bedroom of the participants, and their pared with those measured in the home. However,
radon exposure in the home was estimated based on limitations of this study are that the measure-
these measurements weighted by their relative oc- ments were of short-term duration (1 week) and
cupancy of these rooms. Areal measurements were most of the participants worked for the same or-
also performed at the workplace of the participants. ganization, and therefore, the results are unlikely
The comparison showed that the total individual to be representative.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

112
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv009
Oxford University Press

7. Interpretation of Measurements

7.1 Variations of Areal and Local Radon concentrations varies greatly in EU countries, from 20
Activity Concentrations to 210 Bqm23, and the median from 20 to 180 Bq m23.
It should be noted, however, that the results are based
7.1.1 Worldwide Variation
mainly on non-representative volunteer data; nor has
National indoor radon surveys have been carried any population weighting been carried out. In many

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

out in a number of countries around the world using countries, the measurements have been focused on
different survey designs and exhibiting significant high radon areas.
areal variations. Spatial variation between the cells is shown in
The European indoor radon map (Dubois et al., Figure 7.2. The country-specific reported GSDs vary
2010) provides an extensive overview of radon activity typically from 1.7 to 2.2, the median for all countries
concentrations and areal variation in Europe (status being 1.91. This means that typically 95% of the
end of June 2009). The map is based on 10 km  measurements for a cell are in the range of 0.27 –3.7
10 km grid data. The data provided by national author- of the cell GM (GM/1.912 2 GM 1.912) (GM: geomet-
ities include results from 770 998 measurements in ric mean).
16 422 cells with the following descriptive statistics: Steck (1992) has analyzed spatial variations in the
Number of measurements per cell: upper Midwest of the USA in order to examine the
ability of standard radon measurement protocols to
† arithmetic mean (AM): 47 predict the long-term radon activity concentration in
† median (MED) with absolute deviation (MAD): 5+4 houses. In this study, 243 houses were monitored for
(MAD ¼ MED f|(AMi – MED (AMi)|g, AMi ¼ AM at least 1 year. Table 7.1 summarizes the pertinent dis-
in cell i) tribution parameters for spatial units ranging in size
† range: 1 – 2400 from a single floor of a house to the state of Minnesota.
Radon activity concentrations: House-to-house variations within a town-sized cluster
can be substantial. The table shows that the best esti-
† arithmetic mean (AM) of all cell means with coeffi- mate for the house-to-house variation about the town’s
cient of variation (COV) (COV ¼ SD/AM): 98 Bq mean radon activity concentration is 74%. For larger
m23 + 156% (Figure 7.1) geographical units, county, region, and state, the vari-
† median (MED) of all cell means with absolute ation is still greater. The house-to-house variation in
deviation (MAD) ¼ 62 + 22 Bq m23 (Figure 7.1) towns is in agreement with a median COV of 60%
† median of all cell medians (MED + MAD) ¼ 53 + observed in the 10 km  10 km squares of European
28 Bq m23 countries (Figure 7.2).
† percentage of all cell means .400 Bq m23: 2.28% The European and Midwest US results are in good
† coefficient of variation (COV) and geometric agreement with many regional and national radon
standard deviation (GSD) within cells (MED + surveys (Table 7.2). In the examples in the table, the
MAD) ¼ (60 + 27) %, 1.91 + 0.36% (Figure 7.2) GSD varies from 1.8 to 2.6. In the Irish study, the
overall GSD was 2.40. It should be noted that in the
individual 10 km  10 km squares (total 837) of the
The indoor radon map shows that there is no observ- European indoor radon map, the GSD ranged from
able geographic trend in radon activity concentrations. 1.23 to 5.86.
High radon activity concentrations can be observed in
all European countries, mainly due to varying bedrock
7.1.2 Spatial Variation within a House
and soil geology. Such high radon areas are associated,
for example, with granites and volcanic areas. On a re- Variation between two rooms in the ground floor is
gional scale, smaller “clusters” of cells with higher or caused by room-to-room variations in convective air
lower radon activity concentration can be observed. flows from beneath the floor slab, in air exchange,
Figure 7.1 shows that the mean of the radon activity and, on a smaller scale, also by the role of building

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

materials as a radon source. Room volume and the

coupling to other parts of house, such as the second
floor and the staircase, also affect the results.
Table 7.3 shows radon activity concentration ratios
between different levels of houses and between dif-
ferent rooms at the same level. For example, Chen
et al. (2008) determined the ratio of the geometric
means of 4238 bedrooms and 3669 basements mea-
sured in over 4400 houses. Together with other
Canadian results, the mean of the ratios in al-
together 5486 houses was 0.60 (Chen, 2003). Since
the radon activity concentrations on the first and
second floor were not significantly different from
each other, they were combined in the category “up-
Figure 7.1. Estimated spatial mean radon activity concentration stairs.” One study of 52 homes included the ratio of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

for European countries based on the arithmetic means of the
detectors worn by men and women to first and
10 km  10 km cells within the countries.
second floor activity concentrations (Harley et al.,
1991). Note that “ground” floor in the European
studies (Arvela et al., 2012; Wrixon et al., 1988) indi-
cate the same level in the house as the “first” floor in
the American studies (Fisher et al., 1998; Harley
et al., 1991; Marcinowski et al., 1994; Steck, 1992).
The original terminology was intentionally not
changed in order to avoid confusion when consulting
the corresponding papers.
The second floor to first floor or first floor to ground
floor radon activity concentration ratios of 0.66–0.9 are
indicative of air flows between the stories and dilution
of radon activity concentration when the first floor air
is mixing with the upstairs air volume. Due to vents,
flows of outdoor air, and variations in source strength,
the radon activity concentration may be also higher up-
stairs. Direct air flow to the second floor through pipe
Figure 7.2. Median COV and GSD within 10 km  10 km cells in penetrations and the intermediate floor may increase
European countries. second floor radon activity concentrations.

Table 7.1. Spatial variation of indoor radon activity concentration in the USA in units of different size (Steck, 1992)

Subunit Unit Ma Nb Distribution COVd Geom. Geometric Range Range

typec (%) mean standard deviation Min Max

Floor Room 19 5 Normalized 60 0.82 1.60 0.3 2.3

Floor House 416 208 Ratio 78 0.54 1.78 0.1 3.7
House Town 227 40 Deviation 74 1.74 1.45 1.1 3.9
House County 171 14 Deviation 98 1.98 1.30 1.4 3.5
House Region 343 7 Deviation 101 2.01 1.19 1.8 3.0
House State 243 1 Deviation 105 2.05 — — —
Town Region 39 7 Normalized 60 0.77 1.60 0.5 1.6
Town State 39 1 Normalized 46 0.72 1.46 0.3 1.6
Country State 14 1 Normalized 23 0.98 1.23 0.7 1.3
Region State 7 1 Normalized 11 1.02 1.11 0.9 1.3

a
Number of subunits.
b
Number of units.
c
All distributions are lognormal. Normalized-type distributions are constructed from ratios of each subunit’s value to the unit’s average.
Ratio ¼ first floor radon/ basement radon. Deviation distributions are distributions of subunit standard deviations and are not
normalized.
d
Coefficient of variation, COV ¼ 100 (GSD21) for normalized distributions. For deviation type distributions, COV ¼ 100 (GM21).

114
 Interpretation of Measurements

In houses with basements and semi-basements, Most of the measurements in Table 7.3 lasted for 1
the walls in contact with soil increase radon activity year. The high GSD values of the concentration ratio
concentrations in basements, especially when doors suggest that significant variation can exist in the
leading to the ground floor are closed. The geometric annual radon activity concentrations within a house.
mean (GM) of the first floor/basement ratio in The observed geometric standard deviations (GSDs)
Table 7.3 is typically 0.4–0.7, which is lower than the in the range of 1.3–1.9 indicate that a radon measure-
first floor/ground floor ratio in non-basement houses ment at only one location can incorrectly estimate the
(0.7–0.8). The air exchange between basement and average radon activity concentration by a factor of 2.
first floor varies in different national housing prac- The database for the variation of observations pre-
tices, affecting the radon activity concentration ratio. sented in Table 7.3 and the uncertainty estimation of
Due to additional room-specific radon sources, the the annual average exposure to radon is very limited.
variation expressed as GSD also increases to a level Heid et al. (2004) have analyzed the uncertainties of
of 1.85 from the lower level of 1.54 observed in houses the between-measurements variability in the German
with no basement. radon epidemiology study (Wichmann et al., 2005).
The mean coefficient of variation (COV) between Two simultaneous measurements carried out in the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

radon activity concentrations in rooms on the same bedrooms and living rooms of 4000 dwellings were
level of the house given in Table 7.3 is typically 9– analyzed. The result of a variance analysis yielded an
15%. In the Iowa study (Fisher et al., 1998), the overall SD of the log of error estimate of 0.3. This was esti-
mean COV (SD) for radon concentrations obtained mated to be a good approximation for the COV. In
from detectors placed in different locations on the summary, the COV of 30% has been utilized in the
same floor was 9.5% (10.7%) with a range of 0–119%. uncertainty analysis of Section 8.3 as a conservative
and “best estimate” of the error in different exposure
Table 7.2. Examples of geometric standard deviation (GSD) in estimates due to the variations of radon activity con-
regional and national radon surveys centrations between rooms. This approximation repre-
sents the variation between rooms both on the same
Country Survey No. of dwellings GM (Bq m23) GSD
floor and on different levels above the basement.
Japana National 899 12.7 1.78
Italyb National 5361 52 2.1
Irelandc National 11 319 57 2.40
Finlandd National 2866 70 2.45 7.2 Diurnal and Seasonal Variations
Canadae Winnipeg 2916 97.6 2.58 of Radon Activity Concentrations
a
Sanada et al. (1999), bBocchiccio et al. (2005), cFenell et al. The values of indoor radon activity concentrations
(2002), two detectors per dwelling, dMäkeläinen et al. (2009), are affected by many physical processes. For example,
e
Letourneau et al. (1992). radon entry is affected by radon emanation from

Table 7.3. Ratios and geometric standard deviations (GSD) between radon activity concentrations at different levels of houses. Activity
concentration ratios (means or geometric means, GM) or the mean coefficient of variation COV of radon activity concentration in different
rooms on the same level

Country House type Number Ratio Ratio statistic Ratio GSD COV (%)

Canadaa Basement 1260 Living room/basement Ratio of means 0.68

Canadab Basement 4238/3669 Bedroom/basement Ratio of GM 0.59
Canadac Basement 5486 Upstairs/ basement Mean of ratios 0.60
USAd Basement 208 First floor/ basement GM of ratios 0.54 1.78
USAe Basement 4350/2716 First floor /basement Ratio of means 0.40
USAf Basement þ 1 floor 471 First floor/basement Mean of ratios 0.61
USAf Basement þ 2 floors 417 First floor/basement Mean of ratios 0.53
Finlandg Semi-basement 249 First floor/basement GM of ratios 0.67 1.85 34.0%
USAe — 1561/4350 Second floor/first floor Ratio of means 0.90
UKh No basement ,2093 First floor / ground floor Ratio of means 0.66
Finlandg No basement 421 First floor/ground floor GM of ratios 0.84 1.54 19.8%
USAf No basement, 1 level 1.111 Room 1/room 2 — — — 9.5%
Finlandg No basement, 1 level 2241 Room 1 /room 2 GM of ratios 0.99 1.38 14.4%
USAi Basement 52 First floor/basement Ratio of means 0.57 6.5%
USAi Basement 52 Second floor/first floor Ratio of means 0.80 10.3

a
Letourneau et al. (1992), bChen et al. (2008), cChen (2003), dSteck (1992), eMarcinowski et al. (1994), fFisher et al. (1998), gArvela et al.
(2012), hWrixon et al. (1988), iHarley et al. (1991).

115
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

mineral grains into soil air, pressure difference-driven

air flow in porous soil media and through gaps of the
base floor, by infiltration in the building shell, exhal-
ation from building materials, and climate outdoors.
Radon entry processes increase the radon activity
concentration in indoor air, while air exchange of the
building decreases the radon activity concentration.
The same mechanisms that drive the air exchange
also drive the radon entry from the soil. A forced
mechanical ventilation further complicates the situ-
ation. Therefore, it is important to understand interac-
tions between ventilation and radon entry processes.
Different ventilation strategies aim at keeping the
air exchange, impurities, and moisture content of
indoor air at an acceptable level. In the case of radon

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

progeny, other effects such as plate-out to room sur-
faces affect their activity concentration. Different
radon entry mechanisms, such as emanation from Figure 7.3. Diurnal variation of the radon activity concentration
building materials and pressure-driven flow, may established in each level of a detached two-storey house with
contribute in different ways to variations in radon natural ventilation (measured from 27 February 2009 to 10
March 2009) (Beck, 2012).
activity concentrations. The radon activity concen-
tration in indoor air, CRn, can be expressed by the
following simplified equation 7.3 Physical Factors Affecting Indoor Radon
Activity Concentrations
CRn ¼ S=(NV) ð7:1Þ
Physical mechanisms affecting indoor radon activ-
21
where S is the radon entry rate (Bq h ); N the ity concentrations are: pressure difference and air
house air exchange rate (h21) ; and V the volume of exchange, radon entry from soil and building mate-
the house (m3). Variations in radon activity concen- rials, effect of wind on radon entry, stack effect, and
tration are created by variations in the entry and air exhaust ventilation, as well as meteorological para-
exchange rates. A house of greater height indicates meters causing seasonal and long-term variations.
a larger house volume and generally lower radon ac- This information is very import for reporting radon
tivity concentration. However, house height and exposures.
volume are connected also to entry rate and air ex-
change. Low air exchange rate increases the radon 7.3.1 Pressure Difference and Air Exchange
activity concentration.
Indoor–outdoor pressure difference and house air
Figure 7.3 shows a representative example of
exchange play an important role in the variation with
diurnal variation in radon activity concentration.
time in radon activity concentration. Natural ventila-
The results show the diurnal variations of the radon
tion is currently the most common ventilation strategy
activity concentration in the cellar, first, and second
in residential houses. Natural ventilation is based on
floor of a detached house with natural ventilation mea-
the pressure differentials and air flows in the envelopes
sured over 12 d. In houses with natural ventilation, a
of the buildings created by indoor–outdoor tempera-
night-time maximum is typical due to the cold outdoor
ture difference and wind. The physics of natural air ex-
temperature at night. The increased indoor–outdoor
change rate can be expressed, for example, using the
temperature difference and the resulting pressure dif-
following well-tested model (Sherman and Modera,
ference amplifies the flow of radon-bearing soil air into
1984):
living spaces and increases radon levels. Sections 7.3.2
and 7.3.3 deal with the processes affecting radon 2
Q ¼ ELA ½( fsr 2 DT 1=2 ) þ ( fwr v)2 1=2 ð7:2Þ
entry.
For comparison, the diurnal variation of the radon N ¼ QV 1 ð7:3Þ
activity concentration in a workplace is illustrated
in Figure 6.1 (Section 6.4.1). where Q (m3 h21) is the total infiltration rate; ELA the
A typical seasonal variation is characterized by a effective leakage area (m2); DT the indoor–outdoor
higher winter activity concentration compared with temperature difference; fsr the stack parameter, fwr the
summer activity concentration. Seasonal variations wind parameter; and v the wind speed. N is the house
are addressed in Sections 7.3.5 and 7.3.6. air exchange rate (h21); and V the house interior

116
 Interpretation of Measurements

volume (m3). The stack parameter is affected, for Opening windows increases air exchange. Opening
example, by the house height and distribution of lea- windows, especially in the lower parts of the house,
kages in the building shell. Envelope leakages and reduces or nearly brings to zero the pressure differ-
shielding against wind in the house surroundings ence at floor level and affects strongly the seasonal
affect the wind parameter. ELA describes the air- variation between spring and autumn seasons.
tightness of the building and ranges from less than
0.002 m2 in very air-tight houses to 0.2 m2 in very 7.3.2 Radon Entry from Soil and Building
leaky houses. A widely used recommendation for a Materials
qualified air exchange is 0.5 h21, which indicates one
change of the air volume in the house every 2 h. A Convective entry of soil gas is the dominant source
widely used indicator for air-tightness of the building of indoor radon in most houses with elevated concen-
is the air exchange rate at 50 Pa pressure difference trations (Nazaroff, 1992) (Section 5.2.1.3). The
(ACH50). Energy saving construction has altered venti- driving force for this entry is the small indoor–
lation practices by reducing air exchange rates. In outdoor pressure difference, typically 1–10 Pa. The
Nordic countries, most new dwellings are today pro- radon entry rate Qradon can be determined by the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

vided with balanced ventilation with a heat exchanger. pressure difference and resistance to flow of soil gas
Natural ventilation depends strongly on climatic at entry routes (Sherman, 1992).
factors, air-tightness of the building shell, and the use
Qradon ¼ Cdeep RLA vo ðDP=Po Þn ð7:4Þ
of ventilation windows or fresh air vents. Mechanical
or forced ventilation may be either by mechanical where Cdeep is the soil gas radon activity concentra-
exhaust ventilation or by supply and exhaust ventila- tion in deep soil; RLA the radon leakage area (m2)
tion. The latter is also called balanced ventilation. analogous to the effective leakage area of the building
Mechanical ventilation adds a forced component to shell [Equation (7.2)] and combines all information
natural ventilation. Therefore, air exchange is no about the flow resistance of foundation structures
longer solely dependent on climatic effects, as in the and building soil. DP represents the pressure differ-
case of natural ventilation [Equations (7.2) and (7.3)]. ence at floor level. The factor vo represents the refer-
Negative pressure differences from indoors to out- ence velocity related to the reference pressure Po. The
doors in houses are the key factor driving both air ex- leakage exponent n varies from 0.5 to 1 depending on
change and inflow of radon-bearing soil air into the role of gaps and soil in the total flow resistance.
buildings. Pressure difference due to the “stack effect” The flow resistance of the soil is determined by the
is proportional to indoor–outdoor temperature differ- air permeability of the soil. Soil types cover a very
ence and house height. The stack effect is caused by wide range of permeabilities, spanning more than 5
the difference in densities of the air columns of differ- decades. Clay represents a highly impermeable soil
ent temperatures indoors and outdoors. The colder air and coarse gravel a very high permeability. Figure 7.4
column outdoors with a higher density compared with shows the strong effect of soil permeability on radon
the less dense air column indoors causes a pressure
difference over the wall and floor structures. Typical
pressure differences in houses with natural venti-
lation at an outdoor temperature of 0oC are 1–3 Pa.
Mechanical exhaust ventilation increases the pressure
difference in a leaky house less than in a house with an
air-tight structure. The resulting typical pressure dif-
ference in moderately tight houses is 4–10 Pa. The use
of balanced ventilation with equal supply and exhaust
air flows affects only slightly the pressure difference.
However, increased air-tightness, e.g., in a modern
passive construction, may result in remarkable pres-
sure differences when the airflows are not balanced
(Arvela et al., 2015). In cold and cool climates, e.g., in
Nordic countries, balanced ventilation may need to be
adjusted for a minor negative pressure difference in
order to avoid long-term moisture problems. Typical Figure 7.4. Convective radon entry rates into a typical basement
through a 0.003 m slab-footer gap as a function of soil
pressure differences in Nordic houses with balanced
permeability. The permeability of the 0.15 m thick layer of gravel
ventilation are from 1 to 5 Pa (Jokisalo et al., 2008). beneath the basement slab is the variable parameter. The
All measures affecting pressure difference indoors basement is at 25 Pa pressure with respect to the atmosphere
affect radon entry and radon activity concentration. (Revzan and Fisk, 1992).

117
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

entry through a 0.003 m slab-footer gap. Note that a The role of building materials is highest in houses
0.15 m thick layer of gravel beneath the slab affects with masonry wall structures. Radon exhalation
the radon entry rate by a factor of 3–5. from concrete is higher than from burned clay brick
All gaps and openings in the floor construction or from furnace blast light-weight concrete block.
provide entry routes for radon-bearing soil air. In Globally, the typical range of 226Ra activity concen-
basements, radon-laden soil gas flows through tration in masonry building materials is quite
cracks in the floor slab and walls, block wall cavities, limited (10 –60 Bq m23), resulting also in a limited
plumbing connections, and sump wells. In most range of the radon exhalation rate. As a general
cases, the resistance of soil is much larger than the rule, the lower the radon activity concentration in
resistance of leakage gaps and openings in the foun- masonry houses, the more probable is a significant
dations. The flow through porous media, such as contribution from building materials to indoor radon
soil, is generally linear with respect to pressure. activity concentration.
Therefore, in a house with a floor slab in ground Soil moisture is an important factor affecting soil
contact, the flow rate of soil air can be assumed to be gas radon activity concentration and therefore also
proportional to the pressure difference at floor level. indoor radon activity concentrations. Partitioning of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

In this case, the exponent n in Equation (7.4) is radon gas between the water and air fractions of soil
equal to unity. In crawl space houses, the leakage pores is the main factor increasing soil air radon ac-
follows the same physical laws as air infiltration in tivity concentration with the increasing water satur-
the building shell; the leakage flow is proportional ation factor. Soil temperature is also an important
to pressure difference to the power of 0.5 –0.7. factor. Andersen (2001) provides a good review of
Radon entry from soil is highest during cold periods these effects. Rose et al. (1990) have studied season-
such as night-time and winter. Similarly, air ex- al variation in different temperature, soil type, and
change is elevated during these periods. However, soil moisture regimes in the USA, utilizing a theor-
the effect of radon entry is stronger and, normally, etical analysis of soil gas radon activity concentra-
winter activity concentrations are higher than tions. The study concludes that among the many
summer activity concentrations (Figure 7.7). effects of water on soil gas radon, the effect of the
The process of radon exhalation from all building varying proportion of pore space occupied by water
materials containing 226Ra is different from pres- appears to be among the largest and most universal.
sure difference-driven air flow. The emission is The study estimates that soil gas radon will be most
based on radon emanation from mineral grains into elevated by moisture effects in the eastern USA. In
the pore space of the material. Thereafter, diffusion some states, the summer activity concentration of soil
results in radon exhalation from building material radon was predicted to be higher than the winter ac-
surfaces into room air. Thoron is emitted from build- tivity concentration. A more recent review of the ex-
ing materials in a similar way. Radon emission from perimental and theoretical estimates, together with
building materials in a living environment is basic- soil moisture measurements over a period of 10 years,
ally a process with only a minor variation. The mois- indicates that variation in soil moisture is an import-
ture content of building material is a potential factor ant factor affecting the seasonal variation in indoor
increasing radon emanation from masonry materi- radon activity concentration (Arvela et al., 2015).
als (Sakoda et al., 2012). Radon emanation from con- Year-to-year variation in radon activity concentration
crete increases typically by a factor of 2 when may be markedly affected by long-term variation in
relative humidity in the atmosphere increases from soil moisture.
0 to 80% (Cozmuta et al., 2003). However, this effect
is not fully understood for normal living conditions.
7.3.3 Effect of Wind on Radon Entry
Therefore, the radon activity concentration due to
building materials is essentially controlled by the Wind has a significant effect on radon entry rate
air exchange in the building. In a house with and indoor radon activity concentrations (Riley et al.,
natural ventilation, radon activity concentration 1996; 1999). Wind first establishes depressurization
due to building materials has a maximum in sum- of the house, followed by a steady-state ground-
mertime when the air exchange is lowest and a surface pressure field which causes an inflow of soil
minimum in winter when air exchange is highest. gas and radon. Table 7.4 summarizes the effect when
Increased air exchange through open windows in only depressurization of the house has been consid-
summertime reduces strongly the summer activity ered. Depressurization of the house and the soil gas
concentration. Simple modeling estimates confirm pressure field reach a steady-state after perturbation
the difference between seasonal patterns with radon with a characteristic time of seconds to minutes. Soil
entry from building materials or pressure-driven gas concentration will reach a steady-state with a
soil air flow (Arvela et al., 1988). characteristic time that is determined by (1) soil gas

118
 Interpretation of Measurements

travel time from surface to the basement, from hours Including the effect of wind on ground-surface pres-
to months, depending on soil permeability, and (2) sures reduces the predicted radon entry rate relative
the time required to reach a radioactive steady-state, to a calm, low wind situation by as much as a factor of
typically several days. Figure 7.5 shows an example 3 at a permeability of 10211 m2 (fine sand) to 1000 at
of wind-induced variations. a permeability of 1028 m2 (medium coarse gravel).
In a test house in open terrain with no wind protec- The predicted indoor radon activity concentrations
tion and with an eaves-height of 3 m, the indoor differ by about the same factor. The pressure field
depressurizations caused by wind speed of 3.6 and created by wind reduces soil gas radon activity con-
8.3 m s21 were 2–11 Pa (Riley et al., 1996). The real centration by pushing and mixing radon-free air
pressurization depends on wind direction, house struc- beneath the house foundation.
ture, and air-tightness and shielding effect of the The observations indicate that the flushing effect
house surroundings. In the case of low indoor–outdoor of wind is an important factor affecting hourly,
temperature difference and an average wind speed of diurnal, and weekly variations in radon activity con-
3 m s21, the wind-induced pressure difference may be centrations. Seasonal variations in monthly average
comparable to, or greater than, the temperature- wind speeds are often rather limited, the average

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

induced pressure differences. wind speeds being generally 3–4 m s21. Therefore,
The effect of wind-induced ground surface pressures the effect on seasonal variation is not as high as on
on the radon entry rate may be very marked. In the shorter-term variation.
absence of wind-induced ground-surface pressures, the The variations of indoor radon activity concentra-
radon entry rate and hence indoor radon activity con- tions due to wind may differ greatly from house to
centration may increase by an order of magnitude as house. Permeability of building site, varying perme-
the soil permeability increases from 10211 to 1028 m2. ability in horizontal soil layers at different depths, use
of gravel as sub-slab filling material, ground surface
Table 7.4. Effect of different driving forces on radon activity profiles, and sensitivity to wind can cause large varia-
concentration normalized to soil activity concentration and total tions. The flushing effect will be emphasized in hilly
leakages (Sherman, 1992). In this presentation, the driving forces
areas where the forces of wind pressure pushing or
do not affect the soil gas radon activity concentration
sucking air from house subsoil are highest (see
Driving force Normalized radon activity Chapter 7.3.6). In houses with crawl space, the vari-
concentration ation depends on the interaction of wind speed, direc-
tion, and ventilation in crawl space (Miles, 2001).
Representative Range

Stack effect, basement 5 1–14

Stack effect, slab-on-ground 2 1–5 7.3.4 Comparison of Driving Forces
Wind 0.5 0–2
The effect of different driving forces, stack effect,
Exhaust ventilation 1 0.7–1.4
wind, and exhaust ventilation on radon activity

Figure 7.5. Indoor radon activity concentration and wind speed over a 3-week period (Riley et al., 1996).

119
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 7.6. Distribution of uncorrected seasonal/annual average ratios in 91 UK houses measured in three months periods over 2 years.

concentration is illustrated in Table 7.4 (Sherman, 7.3.5 Seasonal Correction Factors

1992). The effect of wind on soil gas radon concentra-
tion (Section 7.3.3) is not included. The radon activ- A seasonal correction factor (SCF) is a multiplying
ity concentration from each driving force scales in a factor applied to a measurement with a duration of
similar fashion with soil activity concentration and one or more months in order to derive a meaningful
total envelope and entry leakage, but differently annual average radon activity concentration. The cal-
with leakage distribution and pressure. Combining culations generally assume periodic annual, typically
these two effects allows a comparison of the induced sinusoidal, variation (Pinel et al., 1995). In the UK,
radon activity concentrations for conditions repre- SCFs were initially derived from two consecutive
sentative of normal housing. Table 7.4 displays 6-month measurements carried out in 2300 homes
the indoor radon activity concentration normalized and having start dates in all seasons. Figure 7.6 shows
to the soil activity concentration and total leakages. the distribution of uncorrected season/annual average
The results indicate that the highest activity con- ratios in 91 UK houses measured in 3 months periods
centrations occur during stack-dominated periods. during 2 years (Miles et al., 2012).
The stack effect (Section 7.3.1) causes the highest The modified UK correction factors based on these
variation in houses with basements due to differ- data are presented in Figure 7.7 together with the
ences in height. For example, in the case of houses original correction factors (Pinel et al., 1995). In the
with basements, the stack effect increases the nor- British approach, the correction factor is given for
malized reference radon activity concentration at all months. An alternative approach is to use only
the annual average outdoor temperature by a factor one average correction factor for heating season
of 5 and during cold winter weather by a factor of measurements. This has been found practical in the
14. The lowest activity concentrations occur during Nordic countries where only heating season mea-
wind-dominated periods, during warm periods, and surements are recommended.
windy weather. The effect of exhaust ventilation is For practical reasons, many SCFs are given in the
generally lower than the stack effect because the form of normalized radon activity concentration, i.e.,
increased air exchange is compensated by an ratio of the measured 1–3 months radon activity
increased pressure difference. concentration to the annual average. A seasonal cor-
In new very air-tight, low-energy construction prac- rection factor actually should be a multiplicative
tices, the role of pressure difference caused by mech- factor which gives the annual average on the basis of
anical ventilation can be higher, therefore increasing a short-term measurement.
radon activity concentration (Arvela et al., 2014). The dominant observation in the current SCFs is
High air-tightness of the base floor is required in a higher indoor radon activity concentration in
order to avoid elevated radon activity concentrations. wintertime when compared with summertime. The

120
 Interpretation of Measurements

current British (Pinel et al., 1995) and Irish (Burke seasonal correction factor for 3 months radon mea-
et al., 2010) correction factors are approximately 0.7 surements for five regions. These results show a nor-
for the coldest winter months and 1.3 for mid- malized radon activity concentration of 0.87–1.21 in
summer. Finnish studies in 3000 randomly chosen these regions in January, while the recommended na-
dwellings, with two subsequent 6 months measure- tionwide factor is 1.14. Figure 7.8 shows the seasonal
ments, show an average wintertime/annual correc- variation during 4 years in two dwellings (Denman
tion factor of 0.85 (Arvela, 1995). As a practical rule et al., 2007). Although SCFs illustrate the collective
based on many studies, the summer activity concen- variation of radon, the value of applying such correc-
tration is 50% of the winter activity concentration. tions to individual radon measurements is limited
Analysis of British and Irish indoor radon measure- because of the wide variation from the national
ments shows that there is a clear regional variation in average correction factors. Gillmore et al. (2005)
seasonality of radon levels (Burke and Murphy, 2011; emphasizes the effect of the complexity of underlying
Denman et al., 2007). The Irish analysis presents the geology and considerable variations in permeability of
underlying materials as a reason for a significant
number of occurrences where the application of a sea-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

sonal correction factor may give rise to over-estimated
or under-estimated radon levels. Variation in soil
moisture is a potential reason for the observed
marked year-to-year variations in indoor radon activ-
ity concentrations (see Section 7.3.2).
Increasing the length of measurement period when
calculating SCFs increases the accuracy. In the
Canadian study, two consecutive 6-month measure-
ments in 4508 homes were used (Krewski et al., 2005a;
2005b). Observed and predicted annual average radon
activity concentrations were in reasonable agreement.
Roughly 15–30% of the predicted annual average
radon activity concentrations were within 10% of the
observed values.
Based on the latest UK results (Miles et al., 2012),
presented in Table 7.5, spring and autumn measure-
ments give a better estimate of the annual average
radon activity concentrations than the best seasonal
Figure 7.7. Sinusoidal fit of monthly geometric mean radon
correction factors applied to all season measurements.
activity concentrations for the new UK study (solid line) (Miles
et al., 2012) and those presented by Pinel et al. (1995), normalized
to the same geometric mean radon activity concentration (dashed 7.3.6 Atypical Seasonal Variations
line). Note the values for Miles et al. (2012) refer to the middle of
each month, whereas the Pinel et al. (1995) data refer to the first Atypical seasonal and diurnal variations have been
day of each month. observed in areas of hilly permeable terrains. In the

Figure 7.8. Monthly average radon activity concentrations in two homes during 4 years, individual annual variation and average over the
whole investigation period with 95% confidence intervals (Denman et al., 2007).

121
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 7.5. Effects of different types of correction on the accuracy In a Norwegian study (Sundal et al., 2008), in-
of estimates of annual mean radon activity concentrations (Miles stantaneous changes in soil air flow directions and
et al., 2012)
in soil air radon activity concentration were recorded
Type of correction applied to Percentage of all corrected
when the outdoor temperature reached the annual
3-month results to obtain 3-month results that are within average outdoor temperature which is close to the
estimate of annual mean 30% of the 2-year mean activity deep soil temperature.
activity concentration concentration for the home The observations of the effect of wind in the anom-
alous areas described above can be applied to houses
No correction 74% in normal hillside areas, with no elevated subterra-
Seasonal correction based on 71%
Howarth and Miles (2008)
nean air flows, when the subsoil is permeable. Wind
Temperature correction based 76% direction is a potential reason for strong diurnal var-
on Miles (1998) iations and depending on the seasonal variation in
Seasonal correction based on 79% wind speed and direction also for seasonal variations.
Miles et al. (2012)
Use only spring or autumn 85%

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

results with no correction 7.3.7 Long-Term Variation in Annual Average
Radon Activity Concentrations
The annual average temporal radon behavior was
karst terrains of Huntsville, Alabama (Wilson et al., studied at 196 sites in 98 Minnesota houses (Steck,
1991) elevated radon levels were observed during sum- 2009). Seventeen hundred year-long indoor radon
mertime. Abrupt day-to-day changes were also measurements were made from 1983 to 2000 to deter-
observed. Similar observations have been made in hilly mine year-to-year radon fluctuations and long-term
eskers of coarse glacial permeable gravel in Finland temporal trends. Ten year-long measurements over a
(Arvela et al., 1994), in a volcanic region in Spain span of 13 years were made at a typical site. The
(Moreno et al., 2008), and in a permeable ice glacial median radon activity concentration was 120 Bq m23.
sediment in Norway (Sundal et al., 2008). Another The median radon activity concentration of the group
example of atypical seasonal variations is a house in of houses showed little year-to-year variation and no
Minnesota with maximum radon activity concentra- persistent temporal trends. However, at individual
tions during the summer months (Steck, 2005) as sites, year-to-year variations ranged from 3 to 110%.
shown in Figure 6.3b (Section 6.4.2). The median variation was 26%. Climate, exposure
In a Finnish esker study, the difference in tem- to wind, and radon activity concentration affected
perature between the soil air inside the esker and year-to-year variation, while house age, construction,
the outdoor air compels the subterranean soil air to or measurement floor did not. Variation in soil mois-
stream between the upper and lower esker areas. In ture is another potential reason for the observed
winter, the radon activity concentrations are ampli- marked year-to-year variations in indoor radon activ-
fied in the esker top area. In summer, activity con- ity concentrations (see Section 7.3.2). Some individ-
centrations are amplified in certain lower slope ual sites showed significantly larger radon changes
areas, while in winter, radon-free outdoor air is when modifications were made to the house structure
flowing into the esker and into subfoundation soil. and heating-ventilation systems.
Winter/summer concentration ratios were typically An annual average study in Iowa analyzed
in the range of 3 – 20 in areas with amplified winter year-to-year radon variation over spans up to 7 years
activity concentration, and 0.1 – 0.5 in areas with (Zhang et al., 2007). In the 61 houses with 3-yr-long
amplified summer activity concentrations. measurements over a 7-yr span, the COV had a mean
In a Spanish study, the average winter/summer of 24%, a median of 19%, and a range of 0–110%.
ratio was 1.7 on non-volcanic subsoils and 0.5 in vol- These statistics are similar to those of the Steck
canic areas. In comparison, as a rule of thumb in study (2009) reviewed above, which has a larger
normal terrains, the winter/summer ratio is close to sample size and longer study span.
2. Indeed, Harley and Terilli (1990) showed that over Hunter et al. (2005) examined 96 houses with
a 3-year measurement period, the seasonal change moderately elevated radon to track for 6 years with
in radon activity concentration varied by a factor of 3-month-long measurements in each year. After ex-
2 (summer to winter). Parallel with temperature tensive analysis that included building factors, they
variations, the abrupt day-to-day changes in these concluded that the year-to-year variation was of the
areas are caused by variations in wind speed and order of 40%. Only 15% of the variation was
direction. When hitting the slope of a hill, wind is explained by the building factors.
pushing outdoor air to the soil masses below the Measurements conducted over 17 years in a new
buildings. Jersey home showed averages and standard errors

122
 Interpretation of Measurements

of means for basement, first floor, and second floor of

26 + 18%, 13 + 11%, and 13 + 8% Bq m23, respect-
ively (Harley et al., 2011).
In conclusion, based on the studies above, the
reported values of year-to-year variation as expressed
as the COV are in the range of 25–40%.

7.4 Thoron Interference on Radon

Detection Systems
In most indoor environments, thoron (220Rn) is
present as well as radon (222Rn). In general, the
thoron activity concentration is negligible compared
with that of radon. This is not always the case as

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

recent studies have occasionally shown high thoron
activity concentrations in some areas (Nuccetelli
and Bochicchio, 1998; Wiegaud et al., 2000).
Although most radon detectors are designed to min- Figure 7.9. Relationship between the ratios of thoron to radon
imize the entry of thoron, there are a few reports on activity concentrations, TnC/RnC, and computed radon activity
the thoron contribution to the detector response concentrations RnC.
(Tokonami, 2010; Tokonami et al., 2001).
To take into account the thoron interference on
7.4.1 Time-Integrating Devices radon measurements, the observed activity concen-
7.4.1.1 Alpha-Track Detector. Passive radon tration (Cob) with a single passive radon detector can
(222Rn) detectors, in particular track detectors, are be expressed as follows:
commonly used for national radon surveys and epi-
Cob ¼ RnC þ (STn=Rn ) TnC ð7:5Þ
demiological studies. Therefore, several major alpha-
track detectors were examined with respect to their
Where RnC and TnC are the mean activity concen-
thoron interference. Tokonami (2010) summarized the
trations of radon and thoron during the exposure
thoron interference on radon measurements from the
period in Bq m23; and STn/Rn is the relative sensitiv-
viewpoint of relative thoron sensitivity of radon detec-
ity. The values of STn/Rn were estimated to be 0.025
tors. For the interference testing, detectors made in
for the short-term electrets and 0.033 for the long-
Canada, Germany, Italy, UK, and USA were selected.
term electrets.
The German detector had the highest sensitivity for
Figure 7.9 shows the relationship between the
thoron, followed by one from the USA. Special atten-
ratios of thoron to radon and “apparent” radon activity
tion must be paid to these results because these two
concentrations. For values of the ratio, TnC/RnC
detectors were used in major epidemiological surveys
above 10 on the abscissa in Figure 7.9, the computed
(Krewski et al., 2006; Wichmann et al., 2005). Since
radon activity concentrations for both electrets
the thoron sensitivities were 0.7–0.8 times of those for
increased rapidly. There is the possibility that thoron
radon, careful consideration will have to be taken for
activity concentrations are 10 times higher than
their practical use. The other investigated detectors
radon activity concentrations. Thus, there may be pro-
exhibited only very small thoron sensitivities, with
blems related to thoron interference on radon mea-
relative sensitivities of about 0.05. It should be noted,
surements with the electret systems when extremely
however, that the thoron interference may be much
high thoron activity concentrations are encountered.
larger if the detectors are placed near the walls, even
in the case of low thoron sensitivity.
7.4.2 Continuous Devices
7.4.1.2 Electret detector. It is known that 7.4.2.1 Ionization Chamber. Several types of
radon measurements with electret monitors may be PIC ( pulse ionization chamber) detectors are com-
affected by environmental parameters, e.g., tempera- mercially available. Although the most common PIC
ture, relative humidity (RH), the presence of ions in monitor operates as a current ionization chamber
the room, air drafts, gamma radiation, thoron in the for very high activity concentrations of radon, it
air, and external dust. Therefore, two types of elec- could be considered as a PIC detector for the mea-
trets were examined for their thoron interference on surements of environmental radon (Ishikawa, 2004).
radon measurements. It is widely used for measurements of environmental

123
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

radon (Franco-Marina et al., 2001; Huber et al., exposure condition of a mixture of radon and thoron,
2001; Ramola et al., 2000). The PIC detector has calculated standard deviations were relatively large,
some sensitivity to thoron and since this detector e.g., 7.3–21.7. In this case, the value is shown to be 7
type is a very common device for environmental (0–29) Bq m23 in Table 7.6. From the estimated sensi-
radon measurements, it is important to investigate tivities for thoron, it could be concluded that the rela-
the effect of thoron on the detector response. tive sensitivity to thoron of the PIC detector was about
Radon and thoron activity concentrations measured 10% on average. It indicated that the radon activity
with the radon/thoron discriminative detector (see concentration (Bq m23) measured in a mixture of
Figure 5.11) and the PIC detector in a reference radon and thoron was approximately the sum of the
chamber are shown in Table 7.6. The second column actual radon activity concentration (Bq m23) and 10%
in Table 7.6 indicates the time elapsed since starting of the thoron activity concentration (Bq m23). The
the comparison measurements of the two detectors for overestimation of radon (i.e., 10% of thoron activity
each exposure condition. The detectors were started concentration) due to the presence of thoron is negli-
well in advance of the comparison measurements so gible for general environments. However, care should
that they could have a stable response. be taken in thoron-enhanced areas. For example,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

The radon activity concentrations measured with Wiegand et al. (2000) reported that the median values
the radon/thoron discriminative monitor shown in of indoor radon and thoron activity concentrations
Table 7.6 exhibit a wide range. The radon/thoron dis- were 92 and 215 Bq m23, respectively, for cave dwell-
criminative monitor uses 218Po counts to estimate the ings in the region of Yan’an (China). If a PIC monitor
radon activity concentration. With a background level were used in this environment, radon activity concen-
of radon, 218Po counts are very small and the radon ac- trations would be overestimated by approximately
tivity concentration has a large uncertainty. Thus, it 20% on average.
seems that the radon activity concentrations indicated Such overestimation is also possible in other environ-
by the monitor can be lower than 0, for example, 2.7 + ments when the monitor is placed near thoron exhal-
31.8. In this case, the activity concentration is shown ation sources such as thorium-rich building materials.
as 3 (0–34) Bq m23 in Table 7.6. The relative sensitiv- A method to ascertain the presence of thoron using a
ity to thoron ranged from 9 to 14% (average: 12%) for PIC detector is to measure air sucked with a pump.
the exposure condition of thoron with background Operating this way, only a small fraction of thoron
radon. For a mixture of radon and thoron in the ap- would decay. Consequently, the thoron sensitivity can
proximate ratio of 1:1, the relative sensitivity ranged be increased compared with that in the diffusion mode.
from 0 to 12% (average: 7%). Standard deviations for If there is no significant difference in measured activity
the relative sensitivity were calculated using the usual concentrations between pumping and diffusion modes,
least squares error propagation equations. For the it indicates that the presence of thoron is negligible.

Table 7.6 Radon and thoron activity concentrations measured with the two types of detector

Exposure conditions Elapsed Radon/thoron monitor PIC monitor

time (h)
Radon activity Thoron activity Radon activity Relative sensitivity
concentration (Bq m23) concentration (Bq m23) concentration (Bq m23) for thoron (%)

Thoron with 0–1 3 (0 – 34) 634 + 132 91 + 18 14 + 6

background radon (1)
1–2 5 (0 – 34) 765 + 144 105 + 17 13 + 6
2–3 37 (0– 34) 711 + 140 104 + 15 9+7
Thoron with Average 15 (0– 34) 703 + 80 100 + 10 12 + 4
background radon (2)
0–1 3 (0 – 34) 710 + 139 84 + 16 11 + 5
1–2 40 (0– 34) 604 + 130 115 + 19 12 + 8
Mixture of radon Average 21 (0– 34) 657 + 95 100 + 12 12 + 5
and thoron
0 – 0.5 594 + 124 587 + 187 637 + 26 7 (0–34)
0.5– 1 600 + 125 747 + 208 669 + 28 9 (0–34)
1 – 0.5 632 + 128 627 + 194 627 + 29 0 (0–34)
1.5– 2 605 + 125 638 + 195 680 + 31 12 (0 –34)
2 – 2.5 621 + 129 548 + 183 656 + 30 6 (0–34)
Average 610 + 56 629 + 87 654 + 13 7 (0–34)

124
 Interpretation of Measurements

7.4.3 Mathematical Analysis of Radon/ grow-in) and therefore F is lower. The indoor ventila-
Thoron Atmospheres using Nuclear tion rate and hence the value of F depends on the
Track Detectors opening/shutting of windows, use of electric fans, air
conditioners, and dehumidifiers (Chen et al., 1998;
A computational method for the analysis of the
Iimoto, 2000; Iimoto et al., 2001; Iyogi et al., 2003). For
results of the exposure of nuclear track detectors to
example, measurements of F in hospitals in Taiwan
a mixture of radon and thoron atmospheres and the
showed that F was reduced from a value of about 0.7
determination of the decision threshold and the de-
to about 0.1 when a dehumidifier was in operation. In
tection limit are presented in Appendix B (D.
comparison, the hospital central air conditioner only
Schrammel, KIT, private communication, 2013).
reduced F by about 20% (Chen et al., 1998). Also mea-
surements of F made over 4–8 d at the end of each
month for a year in a typical Japanese apartment
7.5 Variation of Aerosol Parameter Values showed that the monthly values were lower in the
for Radon Progeny summer. In summer, when the windows were open or
In order to calculate doses from inhaled radon the air conditioner was in operation, F decreased to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

progeny, the activity size distribution of the radon about 0.2–0.3 compared with the winter months
progeny is required. As described in Section 4.6, the value of 0.6–0.7 (Iimoto, 2000). The filtration effect of
aerosol is created in two steps: After decay of the the air conditioner reduces the radon progeny activity
radon gas, the freshly formed radionuclides react concentration and therefore reduces F (Iimoto, 2000;
rapidly (,1 s) with trace gases and vapors and grow Tokonami et al., 1996). In dwellings and indoor work-
by cluster formation to form particles around 1– places, the mean values of F published in the litera-
3 nm in size. These unattached progeny may also ture vary between 0.2 and 0.7 (Tables 7.7 and 7.8).
attach to existing aerosol particles in the atmos- As described in Section 4.6.1, the fp value depends
phere. The attached progeny may have a tri-modal inversely on the ambient particle concentration. This
activity size distribution, which can be approxi- depends on ventilation rate and whether additional
mated by a combination of three lognormal distribu- aerosol sources are present, such as those due to tech-
tions (Porstendörfer, 2001). These consist of the nical processes, combustion, and human activities.
nucleation mode with activity median diameters The mean values of fp measured in dwellings range
(AMD) between 10 and 100 nm, the accumulation between 4% and 20% with some values greater than
mode with AMD values of 100 – 450 nm and a coarse 40% (Table 7.7) (Chen et al., 1998; El-Hussein, 2005;
mode with an AMD . 1 mm. Generally, the greatest Guo et al., 2012; Hopke et al., 1995; Huet et al.,
fraction of the potential alpha energy (PAE) is in the 2001a; Kojima and Abe, 1988; Kranrod et al., 2009;
accumulation mode. Mohamed, 2005; Reineking and Porstendörfer, 1990;
If radon (222Rn) gas measurements are made, then Tokonami et al., 1996b; Vargas et al., 2000; Yu et al.,
the equilibrium factor, F, is required for dose calcula- 1996). Similar values have also been measured in
tions to determine the potential alpha energy (PAE) indoor workplaces (Table 7.8).
exposure. ICRP uses representative F values of 0.4 In tourist caves with no additional ventilation and
for indoors (ICRP, 1993a) and 0.78 for outdoors high humidity, the particle concentration can be low
(ICRP, 1987). In its 2000 report, UNSCEAR assumes (, 4  103 cm23) with the result that fp is greater
similar F values of 0.4 for indoor exposures and 0.6 than about 10% (Table 7.9). For example, values of
for outdoor exposures (UNSCEAR, 2000). For a mea- fp measured in a natural tourist cave, Postojna,
sured radon activity concentration, the F value domi- Slovenia, by Butterweck et al. (1992) ranged from 6%
nates the contribution of radon progeny to lung doses to 16% with a mean of 10%. These values are similar
and thus has to be determined for an accurate dose to the ones measured in a limestone cave, Australia;
assessment. fp ¼ 11–18% (Solomon et al., 1992). However, further
The following sections discuss the variation of measurements carried out in the Postojna cave gave
aerosol parameter values due to different exposure higher values: the mean values of fp were about 60%
conditions. in the summer and about 12% in the winter
(Vaupotič, 2008b). These are comparable with those
measured in the Carlsbad Caverns, in Southern New
7.5.1 Equilibrium Factor, F, and Unattached
Mexico, in summer, which ranged from 25% to 60%
Fraction, fp, for 222Rn
with a mean of 44% (Cheng et al., 1997). In addition,
The value of F depends mainly on the ventilation particle concentration measurements carried out in
rate with F decreasing with increasing ventilation. As two tourist caves located in North Spain, indicated fp
the ventilation rate increases, there is less time for the values of 26% and 86% (Sainz et al., 2007). A much
radon gas to decay (i.e., for the radon progeny to lower value of fp has been measured in a Bozkov

125
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 7.7. Published values of unattached fraction, fp, and equilibrium factor, F, obtained from measurements in dwellings

Reference Place fp F

Keller and Folkers (1984) Germany 0.34 (0.07–0.90)

Kojima and Abe (1988) Japan. Detached two-story concrete house 0.043 (0.031–0.064) —
Wrixon et al. (1988) UK — 0.4
Reineking and Portendofer (1990) Aged aerosol, Germany 0.096 (0.016–0.25) 0.30 (0.15–0.49)
Closed rooms 0.062 (0.019–0.223) 0.24 (0.11–0.3)
Rooms with open windows
Subba Ramu et al. (1990) India — 0.39 (0.33–0.50)a
Farid (1993) Bangladesh 0.4 (0.33–0.5)a
Nsibande et al. (1994) Swaziland, Traditional 0.33
Modern 0.34
Apartment (ground floor) 0.34
Ramachanran and SubbaRamu (1994) Bombay, India 0.54 (0.15–0.97)
Hopke et al. (1995) USA and Canada 0.047 + 0.032 0.41 + 0.03

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Tokonami et al. (1996b) Japan. Second floor 0.08 0.34
Yu et al. (1996) Hong Kong, ACb 0.09 + 0.04 0.22 + 0.11
Natural ventilation 0.14 + 0.13 0.22 + 0.13
Electricfans 0.15 + 0.10 0.19 + 0.11
Chen et al. (1998) Urban dwellings of Kaohsiung, Taiwan. 0.055 (0.014–0.13) 0.49 (0.24–0.79)
Martinez et al. (1998) Mexico City, Mexico 0.41 + 0.17
Planinic et al. (1999) Osijek, East Croatia 0.44 (0.12–0.89)
Yu et al. (1999) Hong Kong, ACb 0.32 + 0.15
Natural ventilation 0.22 + 0.11
Electricians 0.17 + 0.17
Canoba and Lopez (2000) Aged aerosol, Argentina 0.09–0.29 0.11–0.33
Khan (2000) India, Lucknow 0.35 + 0.19
Kanpur 0.38 + 0.26
Aigarh 0.42 + 0.28
Iimoto (2000) Tokyo, Japan. Typical apartment, 3rd floor. — 0.43 (0.18–0.73)
Vargas et al. (2000) Spain 0.24 (0.12–0.4) 0.17 (0.12–0.22)
Detached house 0.43 (0.14–0.7) 0.06 (0.03–0.12)
Three-story farmhouse 0.14 (0.05–0.3) 0.39 (0.15–0.6)
Three-story house
Virk et al. (2000) India, Punjab 0.29
Yuan et al. (2000) China, Control area. 0.58 + 0.05
High background area 0.43 + 0.16
Huet et al. (2001a) France, aged aerosol 0.31 (0.08–0.67) 0.16 (0.04–0.45)
Clouvas et al. (2003) Thessaloniki, Greece 0.47 + 0.09 (0.2–0.7)
Lopez and Canoba (2003) Argentina 0.34 (0.1–0.8)
Misdaq (2003) Morocco 0.51 (0.40–0.56)
Abumurad and Al Tamimi (2005) Soum region, Jordan 0.4 (0.36–0.42)
El-Hussien (2005) El-Minia City, Egypt 0.09 (0.02–0.22) 0.31 (0.11–0.61)
Mohamed, A. (2005) El-Minia City, Egypt 0.11 (0.04–0.21) 0.35 (0.19-0.62)
Ramola (2005) Himalayas, India. Stone and mud houses 0.26 (0.02–0.9)
Sohrabi and Babapouran (2005) Ramsar, Iran 0.5 (0.39–0.73)
Clouvas et al. (2003) Thessaloniki, Greece 0.49 + 0.10 (0.2–0.7)
Kranrod et al. (2009)c Okinawa, Japan 0.19 (0.05–0.21) 0.14 + 0.01
Jilek et al. (2010) Czech Republic, Town Village 0.095 (0.04–0.25) 0.41 (0.26–0.63)
0.11 (0.05–0.23) 0.33 (0.19–0.55)
Chen and Marro (2011) Canada 0.54 (0.20–0.82)
Guo et al. (2012) China 0.093–0.17
Harley et al. (2012a) Ottawa, Canada, Basement 0.72 (0.59–0.86)

a
Range of mean values.
b
“AC” represents dwellings with air conditioning.
Measurements were also carried out with an air cleaner in operation, and the results were fp ¼ 0.52 (0.3120.71) and F ¼ 0.04 + 0.01.

dolomite cave, Czech Republic with values between that because of this negative correlation, radon gas
1% and 3% (Rovenská et al., 2008). measurements are a more robust indicator of dose
In conditions where the ventilation rate is not than the PAEC under a range of aerosol conditions
high, it has been shown that fp is negatively corre- normally encountered in dwellings and indoor work-
lated with F (Section 4.6.2). Again, it is emphasized places (Sections 4.5–4.7).

126
 Interpretation of Measurements

Table 7.8. Published values of unattached fraction, fp and equilibrium factor, F for 222Rn progeny obtained from measurements in indoor
workplaces

Place fp F Reference

Buildings with ACa; museums, universities and hotels. Taiwan 0.2–0.3 Iimoto et al. (2001)
Buildings with no additional aerosol sources. Germany 0.05 (0.02–0.14) Porstendörfer (2001)
Aomori Prefecture, Japanb 0.43 (0.35–0.52)c Iyogi et al. (2003)
Reinforced concrete buildings 0.39 (0.35–0.44)c
Wooden buildings with steel frame 0.26 (0.19–0.32)c
Reinforced concrete buildings with ACa
Café with smokers, Morocco 0.45 (0.40–0.53) Misdaq and Flata (2003)
Factory (marble), Morocco 0.55 Misdaq and Amghur (2005)
Hospitals with dehumidifier, Taiwan 0.06 Chen et al. (1998)
Nuclear power plants, Japan 0.065 (0.02–0.26) 0.3 + 0.1 Hattori and Ishida (1994)
Offices, Hong Kong 0.13 + 0.17 0.43 + 0.29 Yu et al. (1998)
Offices, Hong Kong 0.38 + 0.13 Yu et al. (2000)

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Offices, Tokyo, Japan 0.026 (0.017–0.035) 0.44 (0.3–0.5) Hattori et al. (1995)
Offices, Tokyo, Japan 0.06 (0.04–0.1) 0.39 (0.24–0.5) Tokonami et al. (1996a)
Offices, Tokyo, Japan 0.11 (0.08–0.16) 0.44 (0.36–0.6) Tokonami et al. (2003)
Offices, China Guo et al. (2012)
Pyramid, Egypt: entrance inside 0.36 (0.13–0.55) Hafez et al. (2003)
0.13 (0.01–0.29)
Kindergardens, Slovenia 0.15 (0.03–0.24) 0.42 (0.20–0.61) Vaupotič (2007)
Schools, Slovenia 0.13 (0.03–0.19) 0.49 (0.27–0.78) Vaupotič and Kobal (2006)
Schools, Kuwait 0.6 + 0.2 Maged (2006)
Schools, Tunisia 0.49 (0.4–0.55)c Labidi et al. (2010)
Spas, Lesvos Island, Greece 0.06–0.11c 0.21–0.44c Vogiannis et al. (2004a)
Spas, Lesvos Island, Greece 0.038–0.13c 0.19–0.31c Vogiannis et al. (2004b)
Spas, LoutraEdipsou, Greece 0.042–0.25c
Spas, Slovenia 0.14–0.57c Vaupotič and Kobal (2001)
Spa, Badgastein, Austria 0.21–0.45c Lettner et al. (1996)
Water supply facility, Germany 0.05 (0.03–0.09) Porstendörfer and Reineking (1999)

a
“AC” represents buildings with air conditioning.
b
Types of workplace include public office, hospital, school, manufacturing plants, and wholesale/retail buildings.
c
Range of mean values.

In mines, the particle concentration and the fp water vapor, trace gases, and the electrical charge
value depend on the use of diesel or electricpowered distribution of the radionuclides in the air.
equipment, the ventilation rate and the type of Porstendörfer (2001) and Reineking et al. (1994)
heating used during the winter months. If diesel measured the unattached size distribution with single
engines are used, the mine aerosol is dominated by screens and screen diffusion batteries. They found
diesel particles, resulting in a low unattached frac- that under “normal” conditions of humidity and radon
tion of about 1% or less (Butterweck et al., 1992; activity concentration, the activity size distribution of
Solomon et al., 1994). However, in a high-grade the unattached progeny can be approximated with
uranium ore mine in Canada, which used diesel- three lognormal distributions. The activity median
powered equipment, the ventilation rate was very thermodynamic diameter (AMTD) values measured
high (about one air change per 3 min). This resulted were 0.6, 0.85, and 1.3 nm with geometric standard
in a low value of F and a higher fp value than the deviations (GSD) of about 1.2. In places with high
expected value based on particle concentration. radon activity concentration, the fraction with the
Measurements carried out in the summer of 1996 greatest AMTD value (1.3 nm) was not observed.
showed that the average values of fp and F were The neutralization rate of the unattached clusters
about 6% and 0.08, respectively (Cavallo et al., 1999). increases with radon activity concentration and so it
is likely that modes below 1 nm are mainly associated
with neutral clusters, whereas modes above 1 nm
are charged clusters (Porstendörfer et al., 2005).
7.5.2 Particle Size Distributions
However, other workers have only measured a
7.5.2.1 Unattached 222Rn Progeny. The rela- uni-modal distribution with AMDs in the range 0.7–
tive activity size distribution of unattached radon 1.7 nm and with GSD values between 1.1 and 1.8
progeny clusters depends on the concentration of (Cheng et al., 1997; El-Hussein, 2005; El-Hussein

127
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 7.9 Published values of unattached fraction, fp, and equilibrium factor, F, for 222Rn progeny obtained from measurements in tourist
caves and underground wineries

Reference Place fp F

Tourist Caves
Butterweck et al. (1992) Postojna, Slovenia 0.1 (0.056–0.16) 0.36 (0.29–0.47)
Solomon et al. (1992) Royal Cave, Victoria, Australia 0.14 (0.11–0.18) 0.19–0.52
Cheng et al. (1997) Carlsbad Caverns, New Mexico 0.44 (0.25–0.59) 0.43 (0.36–0.48)
Misdaq and Ouguidi (2008) Karst Caves, Morocco 0.6 (0.57–0.63)
Rovenská et al. (2008) Bozkov Dolomite cave, CzechRepublic 0.01–0.03
Vaupotič (2008a) Postojna, Slovenia 0.14 + 0.08 0.58 + 0.13
Winter 1999a: 0.64 + 0.12 0.32 + 0.08
Summer 2001a: 0.17 + 0.06 0.58 + 0.13
Summer 2001b:
Underground wineries
Vaupotič (2008b; 2008c) Four wineries, Solovenia 0.08 + 0.02 0.63 + 0.16

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

0.09 + 0.02 0.48 + 0.06
0.12 + 0.04 0.49 + 0.14
0.20 + 0.06 0.25 + 0.08

a
Measurements made at the lowest point of the cave.
b
Measurements made at the railway station in the cave.

et al., 1998; Huet et al., 2001b; Mohammed, 1999). As nucleation and accumulation modes with the fraction
an example, Figure 7.10 gives the activity size distri- of the attached PAEC in the nucleation mode (fpn)
bution of unattached 214Pb measured with a granular being about 0.2 (Reineking et al., 1994).
bed diffusion battery in a dwelling situated in Measurements of the activity size distribution of the
Brittany, France (Huet et al., 2001b). The result for attached progeny in a dwelling in Okinawa, Japan,
218
Po was similar with an AMTD of 0.85 nm and a also showed a nucleation mode with an activity frac-
GSD of 1.25. tion of 0.14 (Kranrod et al., 2009). Porstendörfer
(2001) noted that in low ventilated rooms without
additional aerosol sources, the coarse mode is insig-
7.5.2.2 Attached 222Rn Progeny. The activity nificant because of the greater plate-out rate of large
size distribution of the attached radon progeny aerosol particles on room surfaces (Figure 7.11).
depends upon the exposure scenario and the type of Typically, in places with one dominant aerosol
aerosol sources. Activity size measurements have source, e.g., cigarette smoking, the activity size dis-
been carried out in underground mines, caves, dwell- tribution of the attached radon progeny can be
ings, and indoor workplaces. Results of experimental approximated by a single lognormal distribution.
studies show that the differences between the activity Porstendörfer (2001) measured an AMD of about
size distributions of the individual decay products 270 nm for attached radon progeny in room air con-
attached on aerosol particles are negligible (Huet taining a high particle concentration from cigarette
et al., 2001b; Porstendörfer, 1996). Therefore, for sim- smoke (Figure 7.12). The contributions of the PAEC
plicity and for dosimetry purposes, the aerosol distri- in the size ranges of nucleation particles and coarse
bution of each of the short-lived 222Rn progeny (i.e., of particles were negligible.
218
Po, 214Pb, and 214Bi) is assumed to be the same. Only a few activity size measurements have been
Some activity size distributions of attached 222Rn carried out in workplaces other than mines. Reichelt
progeny are given in Table 7.10 for dwellings and et al. (2000) carried out activity size measurements
workplaces other than mines. The results for of 222Rn progeny at several workplaces, including
dwellings without additional aerosols (i.e., for aged offices, workshops, factories, kitchens, agricultural
aerosols) show that the nucleation mode is not always facilities, and public buildings, such as schools,
observed but can be measured when additional aero- hospitals, and art galleries. Porstendörfer (2001)
sols are produced, for example, by cooking, candle summarized their results and suggested dividing
burning, tile stove heating, fumigating sticks, and gas workplaces into three categories regarding activity
combustion (Huet et al., 2001b; Marsh et al., 2002; size distribution and particle concentration:
NA/NRC, 1991). For an aged aerosol, Huet et al.
(2001b) found that the attached size distribution † Workplaces in rooms without coarse particles.
consisted only of the accumulation mode. However, † Workplaces with coarse particles generated
intercomparison measurements performed in a house by human activities and dispersion processes
in Germany, without additional aerosols, showed (Figure 7.13).

128
 Interpretation of Measurements

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 7.10. Typical example of activity size distribution (dA/AdlogD) of unattached 214Pb plotted as a function of particle diameter D
(mm). Taken from Huet et al. (2001b). EVE and Twomey are the algorithms that were used to derive the size distribution from the data.

Table 7.10. Measurement results of activity size distributionsa of attached 222Rn progeny inside dwellings without additional aerosols
(i.e., for aged aerosols) and inside workplaces other than mines. Mean measurement values are given and the extreme values are given in
parentheses

Reference Place Mode, i fpi AMD (nm) GSD

Dwellings (aged aerosol)

Tu et al. (1991) Rural a 248 (221–274)
Urban a 118 (86–150) 1.8 (1.3–2.2)
Tokonami et al. (1997) Tokyo a 100 3.4
El-Hussein et al. (1998) Egypt a 208 –222 2.4–2.5
Mohammed (1999) Egypt a 320 –340 2.7
Huet et al. (2001b) France a 190 (180–200) 1.6
Porstendörfer (2001) Germany na 0.3 (0–0.4) 0.7 (0.6–1.0) 20 –40 210 (120–350) 1.7–2.1 2.2 (1.6–3.0)
Kranrod et al. (2009) Japan nac 0.14 (0.09–0.21) 29 (23 –42) 1.6 (1.5–1.7)
0.81 (0.73–0.86) 267 (234–308) 1.7 (1.6–1.9)
0.05 (0.05–0.07) 1860–2520 1.4–1.6
Workplaces
Butterweck et al. (1992) Tourist cave a 228 (119 –289) 2.2 (1.3–6.0)
Solomon et al. (1992) Tourist cave a 170
Porstendörfer and Water supply facility na 0.16 50 1.5
Reineking (1999) 0.84 300 1.8
Porstendörfer (2001) Indoors: Without coarse modeb n a 0.3 (0.2–0.5) 15 –40 1.6–2.2
With coarse modec nac 0.7 (0.5–0.8) 300 (150–450) 2.2 (1.6–3.0)
0.3 15 –40 1.6–2.2
0.6 300 (150–450) 2.2 (1.6–3.0)
0.1 5000 (3000–8000) 1.8 (1.1–2.8)

a
Indices i ¼ n, a and c represent the nucleation, accumulation and coarse modes, respectively. fpi, fraction of attached potential alpha
energy concentration (PAEC) associated with mode i. GSDi, geometric standard deviation of mode i.
b
An fp value of 0.05 (0.02– 0.14) was assumed for indoor workplaces without coarse particles (Porstendörfer, 2001).
c
An fp value of 0.01 (0.007– 0.02) was assumed for indoor workplaces with coarse particles (Porstendörfer, 2001).

† Workplaces with one dominant aerosol source such when ventilation conditions vary considerably and
as combustion aerosols, resulting in a uni-modal when the ventilation rate is above 0.5 h21 (Reichelt,
distribution for the attached progeny (Figure 7.12). 2002).
For the first two categories, the nucleation mode repre- The relative size distribution of the aerosol attached
214
sents about 30% of the attached PAEC (Table 7.10). Pb activity concentration measured in a cabinet
Coarse particles produced by re-suspension may occur maker’s workshop is given in Figure 7.13 (Reichelt

129
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 7.11. Relative activity size distribution of the potential
Figure 7.12. Relative activity size distribution of the potential
alpha energy concentration (Cp) of the attached radon progeny in
alpha energy concentration (Cp) of the attached radon progeny
a moderately ventilated room (,0.5 h21), without additional
aerosol in air containing a high particle concentration of
aerosol sources. This was measured with a low-pressure cascade
combustion aerosol from diesel engines and cigarette smoke.
impactor in a house in Northern Bavaria, Germany (Reineking
Continuous line: Mine air (working þ diesel engine), AMDa ¼
et al., 1994). The measured size distribution (solid line) consists of
200 nm. Dashed line: Room air þ cigarette smoke, AMDa ¼
the nucleation mode (broken line) and the accumulation mode
270 nm. Taken from Porstendörfer (2001).
(dot-dash line). Adopted from Porstendörfer (2001).

Aerosol measurements in mines were mainly

et al., 2000). The measurements were carried out with carried out in the 1980s and 1990s. Butterweck et al.
a low-pressure cascade impactor. A tube diffusion (1992) carried out activity size measurements of
battery was connected to the front of the impactor to radon progeny in mines in Germany with a low-
remove the unattached activities. If this is not done, pressure cascade impactor and a high volume impact-
then the unattached progeny collected in the upper or. Their results showed that for a diesel-powered
stages of the impactor may be misinterpreted as the mine, the diesel aerosol dominates the mine aerosol
coarse mode (Gründel et al., 2005). resulting in a uni-modal distribution with an AMD of
To characterize the aerosol distribution in under- about 200 nm with a GSD of about 2.0 (Figure 7.12).
ground mines is difficult because of the highly vari- During non-working hours, the AMD increased to
able conditions and because of the different types of about 350 nm with a GSD of about 2.0.
mining conditions such as diesel or electric-powered Measurements have also been carried out in a
equipment, different ventilation rates, and the type uranium mine at the Olympic Dam, South Australia,
of heating used during the winter months (Cavallo, with a serial graded screen array and a diffusion
2000; Marsh et al., 2008). battery (Solomon et al., 1994). In areas of the mine
Aerosol particle size distribution measurements where there were large diesel-powered vehicles, the
were made in 27 areas in four uranium mines near AMD of the accumulation mode ranged from 200 to
Grants, NM (George et al., 1975). Mining activities 300 nm with a mean value of 250 nm and a GSD of
included drilling, blasting, slushing, ore hauling, about 2.5. In the areas of the mine where there were
and equipment maintenance. All mines were venti- no vehicles or the ventilation intakes were close by,
lated by downdraft main fans and smaller auxiliary the AMD values were smaller, in the range 90–
fans at stopes. Measurements were made with four 200 nm with a mean of 150 nm.
compact diffusion batteries and a jet cascade impactor. A few activity size measurements of 222Rn progeny
The AMD ranged from 90 to 300 nm with a mean have also been carried out in a diesel-powered
value of 170 nm (GSD ¼ 2.7). The unattached fraction, uranium mine in northern Saskatchewan, Canada
fp, ranged from 0.001 to 0.04 with a mean of 0.01. (Cavallo, 2000; Cavallo et al., 1999; Wu-Tu et al.,

130
 Interpretation of Measurements

Phipps, 2007). The activity size distribution of 212Bi

attached on aerosols is assumed to be the same as
that for 212Pb.
Published data on the activity size distributions of
the thoron decay product 212Pb are relatively sparse.
Measurement results for attached 212Pb are given in
Table 7.11 for dwellings and mines.
Activity size measurements performed by Kahn
et al. (1987) in mines showed that the aerosol size of
attached 212Pb was larger compared with that of
222
Rn progeny. The authors suggested that this may
Figure 7.13. Relative size distribution of aerosol-attached 214Pb be due to the longer radioactive half-life of 212Pb,
activity measurement in a cabinet maker’s workshop. The two which allows 212Pb atoms to spend more time in the
curves represent the distribution measured by the low-pressure vicinity of aerosols, leading to increased coagulation
impactor (step-like curve) and the fitted size distribution (smooth of aerosols and thus larger particle sizes. However,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

curve). Taken from Reichelt et al. (2000).
measurements performed by other workers have
shown that the AMD of the accumulation mode for
212
Pb and the 222Rn decay product, 214Pb are similar
1997). Because of the exceptionally high grade ore, at least for “typical” indoor air (Becker et al., 1984;
the mine ventilation rate was very high, about 3.6  Reineking et al., 1992a; 1992b).
104 m3 min21, which was estimated to be about one Butterweck et al. (1992) using impactors carried
air change per 3 min. Measurements were carried out activity size measurements of 212Pb and
out with an impactor during the winter of 1995 and 214
Pb/214Bi in mines. The size distributions of the
the summer of 1996. During the winter months, the attached progeny were uni-modal and the AMD
mine was heated to 58C by burning propane gas to values for 212Pb were similar to those of 222Rn
heat the ventilation air. As a result, the mine aerosol progeny 214Pb/214Bi. During mining activities, the
consisted of particles from the combustion of propane mean values of AMD for 212Pb ranged from 150 to
gas as well as diesel particles. Wintertime measure- 290 nm with a GSD of about 2–3, whereas during
ments carried out at a stope and a drilling area where non-working hours, the mean AMD values of
miners were working showed predominately a 300 nm and 400 nm were measured (Table 7.11).
bi-modal distribution for the attached progeny. The Measurements of the activity size distribution of un-
mean values of the fraction of the attached PAEC asso- attached 212Pb showed that for indoor and mining
ciated with the nucleation and accumulation modes environments, the unattached 212Pb were mostly
were about 65%, and 35%. The corresponding AMD neutral clusters with particle sizes of approximately
values were about 60 and 330 nm, respectively. The 1 nm (Chen et al., 1997). Further, measurements
summertime measurements showed that throughout carried out in a radon test chamber, as part of an inter-
the mine, the AMD values ranged from 50 to 120 nm comparison exercise, showed median diameters less
with a mean value of 85 nm and GSD of about 2.0. than 1 nm for unattached 212Pb (Cheng et al., 2000).
It is acknowledged that the exposure conditions in
mines today are significantly different from those 7.5.2.4 Particle Density. Particle density is
10– 20 years ago and that further measurements are required when assessing the aerodynamic median
required to characterize current mine aerosols. diameter from measurements of the thermodynamic
diameter. The first measurements of the density of
7.5.2.3 Thoron (220Rn) Progeny. Thoron airborne radioactivity were in settled dust on mine
( Rn) decays into the short-lived progeny of 216Po,
220
rafters in three mines in the Uravan region of the
212
Pb, and 212Bi, and it is the inhalation of these Colorado mining area. They were measured as having
progeny that gives rise to a lung dose. However, the a range of 2.4–2.7 g cm23. This indicated silica as the
PAE per unit activity of 212Pb is about 105 and 10 primary component (HASL, 1960). Aerosol particle
times as great as that of 216Po and 212Bi, respectively. numbers or mass concentrations have been reported
As a consequence, ICRP Publication 65 (ICRP, 1993a) in some locations, but aerosol density is rarely mea-
states that “For protection against thoron, it is sured (Kumara et al., 2012). Aerosol composition has
usually sufficient to control the intake of the decay been measured in New York State and is mainly road
product, lead-212, which has a half-life of 10.6 hours.” dust (51%), carbonaceous dust (21%), and compounds
However, doses per unit PAE exposure to thoron derived from fossil fuel emission such as sulphates.
progeny have been calculated by some authors by con- Measurements of mass in an industrial area in
sidering intakes of 212Pb and 212Bi (Kendall and Poland using X-ray photoelectron spectroscopy (XPS)

131
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table 7.11. Measurements of activity size distributionsa of attached 212

Pb inside dwellings and mines. Mean measurement values are
given and the extreme values are given in parentheses.

Reference Place Mode, i fpi AMD (nm) GSD

Dwellings
Becker et al. (1984) Germany, Rural a 190 (120–240 3.0 (2.1–5.3)
Urban a 230 (140–290) 2.6 (2.1–3.2)
Reineking et al. (1992a; 1992b), Germany na 0.14 (0.06–0.2) 30 –50 1.9–2.1
Porstendörfer (2001) 0.86 (0.8–0.94) 220 (175–270) 1.8 (1.5–2.1)
Zhang et al. (2010) China a 150 –160 1.7–2.2
City, Beijing a 110 (90–130) 2.5 (2.3–2.7)
Suburb, Beijing a 80 (50 –130) 2.9 (2.5–3.3)
Countryside a 50 (40 –60) 3.1 (2.1–3.6)
Brick houses
Cave dwellings
Mines

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Busigin et al. (1981) Canadian diesel-powered uranium mine. 88 2.3
Measurements at exhaust ventilation areab
Kahn et al. (1987) Two Canadian minesb,c: Diesel 100 —
Electrical 70 —
Butterweck et al. (1992) German mines: 190 3.1
Uraniummine, Gross-Schloppend 146 (113 –171) 2.0 (1.7–2.8)
Iron mine, Salzgitterd 290 (280–300) 2.2 (1.9–2.5)
Baritemine, Lauterbergd 400 1.6
With mining activityd 303 2.4
Without mining activity
Slate Mine, Fredeburg.
Without mining activity

a
Indices i ¼ n and a represent the nucleation and accumulation modes respectively. fpi, fraction of attached potential alpha energy
concentration (PAEC) associated with mode i. GSDi ¼ geometric standard deviation of mode i.
b
Measurement carried out with diffusion batteries but resolution was poor.
c
Measurements carried out during winter.
d
Measurements carried out during working hours with mining activity.

showed that for all particle sizes, including , 1 mm, 7.6 Estimation of Missing Exposure Data and
elemental carbon accounted for 80% of the mass and Uncertainties Involved
is the major surface element.
As a consequence of the spatial and temporal varia-
For a diesel-powered mine, it is generally assumed
tions of radon and radon progeny atmospheres, the a
that the aerosol is mainly dominated by the diesel
posteriori estimation of missing exposure data is only
aerosol. Several workers have calculated the effect-
possible with great uncertainties. This is particularly
ive density of diesel exhaust particles from measure-
important for epidemiological studies, where meas-
ments of the thermodynamic diameter (dth) and
urement errors and missing radon measurements
aerodynamic diameter (dae) of the exhaust particles
contribute to major uncertainties of individual expo-
(Olfert et al., 2007; Park et al., 2003). The effective
sures. Inclusion of measurement data in residential
density is the ratio of the particle density (r) and
epidemiological studies of radon and lung cancer
shape factor (x). Results indicate that the effective
usually requires detailed retrospective information
density decreases with increasing dth in the size
about residences over 15–30 years. In many epi-
range from 50 to 300 nm. This mainly occurs because
demiological studies, retrospective exposure mea-
particles become more highly agglomerated as size
surements were not possible in a fraction of the study
increases. The smaller particles are more compact
population. When radon measurements are not avail-
than the larger particles and therefore have a higher
able for a specific lung cancer case, indirect methods
effective density. Typically, the effective density
have been constructed based on measurements in the
varies from 1.2 to about 0.3 g cm23 depending on size
control population in the same study.
and fuel composition; higher effective densities are
For example, in one individual study, radon ex-
observed for high sulphur fuel. For dosimetry pur-
posure was imputed using models from prior area
poses, Marsh et al. (2011) assumed an effective
measurements (Raaschou-Nielson et al., 2008). In a
density of 0.6 g cm23 for radon progeny attached to
pooling study of 13 European studies, the missing
diesel exhaust particles in a mine.

132
 Interpretation of Measurements

value was estimated from either the arithmetic words, the “usual” activity concentration was the
mean of all controls or from area-specific control measured activity concentration corrected for dilu-
means (Darby et al., 2005). In the European collab- tion by random uncertainties in measuring radon
orative study, Darby et al. (2005; 2006) calculated a and year-to-year variability. The “usual” activity
mean “usual” activity concentration (or true long- concentration was estimated as one half of the mea-
term average value), taking account of the uncer- sured activity concentration for the high exposure
tainties in the radon measurements. In other group.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

133
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv017
Oxford University Press

8. Variabilities and Uncertainties of Radon and Radon Progeny

Exposure and Dosimetry

8.1 Introduction standard deviations, are evaluated from assumed

probability distributions based on experience or other
This section considers the uncertainties asso-
information.
ciated with calibration and field measurements of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

222 This definition gives the state-of-the-art in dealing
Rn and 220Rn exposures, radon progeny measure-
with all kind of measurements. Thus, all values that
ments, and derived quantities such as the equilib-
are derived from a measurement have to have an
rium factor. The sources of uncertainty associated
assigned uncertainty according to ISO (1995). The
with the determination of the long-term average
publication of incomplete uncertainties, like statistic-
radon activity concentration inferred from a meas-
al uncertainty only, or statistical variations is not
urement in a single year are described. Examples of
considered a sound basis for scientific work.
measurements and uncertainty evaluations are
The central task of a national metrology institute
given for illustrative purposes with particular atten-
(NMI), holding a primary standard, is to realize, to
tion to measurements carried out with etched-track
maintain, and to disseminate the legal units in com-
detectors (see also Appendix C). The sources of un-
pliance with the International System of Units (SI).
certainty associated with dosimetric calculations are
A calibration certificate issued by the NMI docu-
also described. Note that this section deals only with
ments a calibration traceable to national measure-
random uncertainties and not with systematic un-
ment standards, thus providing secondary standards.
certainties.
An example for a possible traceability chain is given
The underlying concept here is “uncertainty.” An
in Figure 8.1.
uncertainty is an intrinsic part of every measure-
ment, the quality of the measurement device will
only influence the value of the uncertainty not its 8.1.2 Variability of Long-term Average Radon
existence itself. Gas Exposures
As discussed above, the measurement uncertainty
8.1.1 The Meaning of Uncertainty in
includes any uncertainty of the calibration along
Metrology
the traceability chain as well as other sources of
The formal definition of the term “uncertainty of uncertainty associated with the field measurement.
measurement” is given by GUM (ISO, 1995) and in However, in the case of radon, the true activity con-
the VIM (JCGM, 2012) as follows: centration varies with time or exhibits a trend due
Uncertainty (of measurement): a parameter, asso- to other contributing factors that cannot always
ciated with the result of a measurement, that char- be monitored. In this case, the measurement of the
acterizes the dispersion of the values that could true activity concentration at a given time cannot be
reasonably be attributed to the measurand. repeated even if the same measurement conditions
This means that the parameter may be, for are put in place (e.g., the same instrument, the same
example, a standard deviation (or a given multiple place of measurement, and the same time of day).
of it), or the half-width of an interval having a stated For radiation protection purposes, an estimate of the
level of confidence. annual average radon activity concentration is often
Uncertainty of measurement comprises, in general, required.
many components. Some of these components can be The radon levels in a house exhibit temporal and
evaluated from the statistical distribution of the spatial variability. The temporal variability, is diurnal,
results of a series of measurements and can be charac- monthly, seasonal, and annual (for details, see Section
terized by experimental standard deviations. Other 7.2). The spatial variability is the variability inside
components, which also can be characterized by a particular house. Short-term measurements have

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 8.1. Traceability chain for the determination of an exposure to 222Rn activity concentration at the time of issuing this report. The
basic units (s, m, mol) are realized by an NMI (PTB). The derived units (Bq, m3) may be realized by an NMI but have to be at least
traceable to an NMI. The length of the traceability chain will influence the total uncertainty.

greater variability, which increases the uncertainty Figure 7.4. Rainfall may affect radon entry, for
in radon exposure assessment. Long-term measure- example, through variation in the soil air radon ac-
ments, which have smaller variability, are usually pre- tivity concentration caused by variations in radon
ferred to short-term measurements for estimating emanation from solid mineral grains.
the annual radon activity concentration. It is the esti- In most case –control residential studies, mea-
mated average annual radon activity concentration surements have been carried out in the living room
that is compared with reference levels for radiation and the bedroom and an average value weighted by
protection purposes (ICRP, 2007). the relative occupancy of the two rooms is calcu-
Meteorological factors can influence the amount lated, although ignoring differences in breathing
of radon-bearing soil air flowing from the soil into rates (Section 7.1.2). The exposure period of the
the house. For example, the analysis of Miles detectors is typically 3 months or a year for each
(2001) showed that four different houses had a very house in the study. If it is less than a year, then a
different response to outdoor temperature, wind seasonal correction factor is sometimes applied to
speed, and direction. Variation caused by these obtain the annual average radon activity concentra-
factors has been reviewed in Section 7.3. Outdoor tion (Section 7.3.5). To obtain a quantity that is a
temperature affects both the soil air entry rate and measure of the long-term average radon exposure
the air exchange within the house. Meteorological over 25–30 years for a given individual, a quantity
factors affect not only the transport of soil air into such as the time-weighted mean (TWM) activity
the house but also the radon activity concentration concentration is calculated (Darby et al., 2005; 2006;
in soil air. The effect of wind has been considered Heid et al., 2004). The TWM activity concentration
in Section 7.3.3. Permeability directly affects the is the mean radon activity concentration across all
flow rate of soil air into living spaces and the homes inhabited by an individual during the 25 or
typical soil type-dependent variation covers many 30 year period weighted by their relative residency
orders of magnitude. Section 7.3.2 gives examples time. Thus, for each house, the estimated annual
of subterranean air flows which strongly affect soil average radon concentration is assumed to be an in-
air radon activity concentration. Soil permeability dication of the long-term average during the resi-
has been considered in Section 7.3.2 and in dency time, which may be over many years.

136
 Variabilities and Uncertainties

8.1.3 Classification of Uncertainties in for missing data (Section 7.5.3). Berkson type errors
Exposure Assessment for also occur when “group-matched” correction factors
Epidemiological Studies for radon measurements are applied to all individuals
with certain characteristics in common, e.g., use of
For epidemiological studies, it is important to
seasonal correction factors (Heid et al., 2004).
identify and assess the sources of uncertainty asso-
Examples of classical type errors include measure-
ciated with the estimated long-term average radon
ment uncertainties and the uncertainty due to the
activity concentration, distinguishing between the
year-to-year variability associated with estimates of
two types of error models, namely the “classical” and
the long-term average radon gas activity concentra-
the “Berkson” models (Berkson, 1950; Heid et al.,
tion based on measurements made in a single year
2004).
(Darby et al., 2006; Heid et al., 2004). Heid et al.
Classical type errors arise when a quantity is mea-
(2004) describe the sources of uncertainties asso-
sured by some device and repeated measurements
ciated with estimates of the long-term average radon
vary around the true value. The additive classical
activity concentration and classify them into errors of
error model for a single source of uncertainty is:
classical or Berkson type.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Measured value ¼ true value þ measurement error The uncertainties associated with internal dosim-
etry can be considered as Berkson type errors
where the measurement error is a random variable
(Schafer and Gilbert, 2006). The individual peculiar-
with a mean of zero and is independent of the true
ities represent the inability of the dosimetric model
value.
to predict the individual’s true dose for a given
Berkson type errors are involved, for example, when
exposure.
the average value of a group is taken as the “measured
value” for an individual. The additive Berkson error
model for a single source of uncertainty is:
8.2 Uncertainty Evaluations: From
True value ¼ “measured value” þ individual peculiarity the Realization of the Unit to the Field
Measurement
where the individual peculiarity is a random vari-
able with a mean of zero and is independent of the Carrying out a residential radon epidemiological
“measured value.” study with radon activity concentration measure-
In residential radon case–control studies, Berkson ments obtained in dwellings (field measurements) is
type errors occur when estimates are made for a complex task. A calibrated device is required to
missing radon measurements based on indirect make field measurements (Figure 8.2).
methods. For example, Darby et al. (2006) used Such a device has assigned uncertainties from
average values in the control population as estimates its use in field measurements and from its former

Figure 8.2. Potential resources to rely on for a radon study: While a measurement and its result (quantity with assigned uncertainty) can
be expressed in a straightforward fashion according to GUM, the long-term average radon activity concentration, or dosimetric quantities
and their associated uncertainty requires further evaluation (STAR is an acronym for “Systems for Test Atmosphere with Radon”).

137
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

calibration (realization of the unit). In other words, device are required before starting a measurement
the measured average radon activity concentration campaign:
over the detector exposure period has a combined
uncertainty according to GUM covering both the - traceability,
calibration measurement and the field measure- - uncertainty,
ment. Further sources of uncertainty occur in the es- - detection limit, and
timation of the long-term average radon activity - range of application.
concentration and these are described in Section 8.3. Because the methodology of calibration differs for
radon and thoron, these are described separately in
the following text. However, for both radon and
8.2.1 Radon Gas Activity Concentration thoron, the calibration involves the determination of
the background activity concentration of the instru-
Before a field measurement is performed, the device ment, which can be the main source of uncertainty
to be used should have been calibrated. In principle, a when measuring low levels of radon or thoron.
calibration after a field measurement is also possible, Because, the background typically increases during

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

but the time between calibration and measurement the service life of a radon monitor, regular systematic
should not be too long and the calibration has to be checks are necessary.
carried out in the range of activity concentration that
is expected to occur. A calibration at 10 kBq m23 is not 8.2.1.1 Rn-220 Calibration. Two typical proce-
suitable for a measurement at 100 Bq m23. Although dures for the calibration of monitors for the activity
this seems to be obvious, it frequently occurs, in part concentration of 220Rn are used worldwide:
due to a lack of knowledge, but sometimes due to the
idea that a calibration at higher radon levels gives a (1) a primary calibration in a constant atmosphere
better statistical significance and thus a smaller un- based on a thorium emanation source, and
certainty for the calibration factor. In most cases, this (2) a secondary method based on calibration via a
is not a valid assumption since it ignores the question reference monitor enclosed in the same atmos-
of linearity. phere as the system under test.
Unfortunately, the devices for radon gas measure-
Both methods provide valid calibration factors. In
ment on the market are not necessarily reliable and
the case of the first approach, achievable relative
their determination of the value is rarely correct
combined uncertainties are 2–4% for k ¼ 2.1 In the
within the uncertainty stated by the manufacturer.
case of the second approach, the relative combined
A certification produced by a manufacturer is only
uncertainties can greatly vary. Realistic uncertain-
valid if the manufacturer is properly accredited for
ties are 4–8% for k ¼ 2.1
that. However, no manufacturer meets this criterion
at the moment.
8.2.1.2 Rn-222 Calibration. Three procedures
For example, tests performed at the Physikalisch-
for the calibration of monitors for the activity con-
Technische Bundesanstalt (PTB) on two measure-
centration of 222Rn are used:
ment devices of the same type (produced by the same
manufacturer) for the measurement of 220Rn showed (1) a primary method based on a reference activity
the following results: one differed by a factor of 1.9 concentration realized by a primary radon gas
and the other by a factor of 3.6 from the conventional- standard and a calibration volume (both values
ly true value. Deviations for other devices were as are traceable to national standards),
high as a factor of 4 (PTB, 2011). With a correct cali- (2) a secondary method based on calibration via a
bration factor, all measurements performed with reference monitor enclosed in the same atmos-
these devices would provide reliable data, without a phere as the system under test, and
correct calibration, a study based on these values will (3) a primary/secondary calibration in a constant
be of questionable value. atmosphere based on a radium emanation source.
To realize the unit, a calibration has to be per- This method is primary or secondary with respect
formed to establish traceability. The uncertainty of to the components used.
the calibration will be an intrinsic part of each meas-
1
urement later in the field. Therefore, the calibration The uncertainty stated is the expanded measurement uncertainty
procedures should be chosen with the same care as obtained by multiplying the standard measurement uncertainty
by the coverage factor k ¼ 2. It has been determined in accordance
the device itself. In Section 5.1 and Appendix A, the
with the “Guide to the Expression of Uncertainty in Measurement
general aspects are summarized as to how a field (GUM)”. The value of the measurand normally lies, with a
measurement should be prepared. For example, the probability of approximately 95%, within the attributed coverage
following physical characteristics of the measuring interval.

138
 Variabilities and Uncertainties

All facilities utilizing one or more of these methods k ¼ 2.5 + 0.1 in the corresponding activity concen-
have to be traceable to one of the facilities listed in tration level. The device is operated in the same
the Calibration and Measurement Capabilities of geometry as in calibration, thus no further correc-
the BIPM or in the BIPM key comparison database tions for the sampling are necessary.
(http://kcdb.bipm.org/). An ab initio calculation of the uncertainty corre-
All methods provide valid calibration coefficients. sponding to a single measurement based on the de-
In the case of approaches 1 and 3, achievable rela- tector parameters, such as response, measurement
tive combined uncertainties are 2 – 4% for k ¼ 2. In geometry, and counting statistics, is seldom possible
the case of the second approach, the relative com- because this information is not provided by the
bined uncertainties can vary widely, but realistic manufacturer. However, the manufacturer could
values are 4 – 6% for k ¼ 2. have provided data for error or uncertainty but of
A calibration factor for the radon monitor under unclear origin and an unclear mathematical base.
test is related to an activity concentration at a speci- Therefore, it is best to avoid problems caused by this
fied time C(t). This activity concentration is either situation by carrying out the following procedure.
given by a reference atmosphere or determined by The statistical fluctuation of the device in response

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

a reference radon monitor. For statistical reasons, to a constant atmosphere is used to determine the
a measuring interval of 24 h for the extrapolation type of uncertainty, the method of evaluation of uncer-
of C(t) from the reference activity concentration tainty by the statistical analysis of a series of observa-
Cr(t,l ), by taking account of radioactive decay, is tions, for the given conditions. This approach is easily
usually sufficient for activity concentration levels implemented in calibration, which can produce a cali-
above 1 kBq m23. For a sealed radon gas standard, bration coefficient including this uncertainty. Thus,
applying the radioactive decay constant l is always an instrument with low statistical power (for example,
valid. of low efficiency) will yield larger standard deviations
In the case of low level radon activity concentra- responding to a reference atmosphere than an instru-
tions (below 1 kBq m23), a calibration in a constant ment with a high statistical power. This is independ-
atmosphere is preferable to obtain small uncertain- ent of the quality of the reference atmosphere itself.
ties (approach 3). In field measurements, the situation is more
complex. Although the same effect of statistical vari-
8.2.1.3 Determination of an Average Activity ation caused by the instrument exists, the model of
Concentration in a Room. The following example analysis is much more unspecific. For example, is the
is a simple illustration of a measurement of the measurement a reading of a constant or changing
220
Rn activity concentration in a room for 10 h with atmosphere? To illustrate this, Figure 8.3 shows the
a direct reading instrument. That is with an active results of measurements of the thoron activity concen-
measurement device that provides continuous meas- tration in a room over 10 h. The reading of the instru-
urement values (quasi online) with a defined time ment is Cm(t) and by applying the calibration factor
resolution. The aim of this measurement is to deter- k and the correction of the background reading, Cm,bg
mine the mean activity concentration and to obtain the “true” activity concentration C(t) is calculated:
the corresponding uncertainty. C(t) ¼ (Cm2Cm,bg)k. The rather dramatic shift of the
At the start of the measurement, the device is absolute values here is typical for many thoron meas-
tested for its characteristic properties, such as the urement devices.
background. The reading of the device might include Under the assumption that the calculation of an
internal calibration factors, but for the determin- average activity concentration is reasonable, C m ¼
23
ation of a correct calibration factor, it will be taken (231 + 23) Bq m is determined. With the knowl-
as is. Thus, it is possible to obtain a “true” calibra- edge of the calibration equation for the mean activity
tion factor independent of the calibration provided concentration C  ¼ ðCm  C m;bg Þ k, a simple uncer-
by the manufacturer. This procedure is comparable tainty budget can be created (Table 8.1). Assuming
to the one explained in Appendix A, Equation (A1). all quantities are uncorrelated and the corresponding
A final advice to the user: Subtract the back- statistics is normal, then the combined uncertainty
ground from the reading of the device and multiply can easily be calculated using the law of propagation
this value with the calibration factor to obtain the of uncertainty (ISO, 1995).
true activity concentration. This calibration is only This is a simple but reasonable approach to gain
valid if the internal calibration factors are not information about the thoron situation in a room as
changed. well as to gain information about the quality of this
The detection limit of the device is 5 Bq m23, information. Since the dominating uncertainty in
its actual background reading was determined to this measurement is the fluctuation of the measured
be (10 + 1) Bq m23, and the calibration factor is activity concentration, the uncertainty associated

139
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Figure 8.3. A measured thoron activity concentration Cm(t) (lower curve) as a function of time t and the calculated thoron activity
 m ¼ (231 +
concentration C(t) (upper curve) obtained by applying the calibration factor and the background correction. The mean values C
 ¼ (553 + 62) Bq m23 with the standard variation are represented by the horizontal lines.
23) Bq m23 and C

 by a measurement of the activity concentration of

Table 8.1. Example for the determination of the mean activity concentration C 220
Rn
over 10 h

Quantity Value Standard uncertainty Indexa

m
C 231 Bq m23 23 Bq m23 87.0%
 m;bg
C 10 Bq m23 1 Bq m23 0.2%
k 2.5 0.1 12.8%
b
C 553 Bq m23 62 Bq m23

Result:

Quantity Value Expanded uncertainty Coverage factor Probability in the given coverage interval


C 0.55 kBq m23 0.12 kBq m23 2.00 95% (Normal distribution)

a
The index gives the amount of influence of a single uncertainty to the combined uncertainty.
b m  C  m;bg Þk, where C
 m and C
 m;bg are the mean measured activity concentration and the mean background activity concentration,
C ¼ ðC
respectively, and k is the calibration factor.

with the mean average activity concentration is rela- 8.2.2 Radon and Thoron Gas Exposures
 ¼ (0.55 + 0.12) kBq m23. As the un-
tively large: C
After a calibration of a thoron gas measuring device
certainty associated with the calibration factor is
has been performed, this system is available as a sec-
smaller, this system shows much less variation in re-
ondary standard for the measurement of activity
sponse to a constant reference atmosphere. Thus, it
concentration, and in combination with a time meas-
can be concluded that the room shows fluctuations
urement, also available for exposure determination
in the thoron activity concentration around a value
 with the assigned uncertainty. (see Section 8.2.1).
of C
Determining the exposure (with k ¼ 2) from the
This example shows that though the influence of
example given in Figure 8.3 would yield an exposure
the calibration uncertainty of the result is small, its
P of :
absolute value changes the entire result. Moreover,
the rather small uncertainty of the calibration will
 Dt ¼ ð5:5 + 1:2Þ kBq m3 h
P¼C ð8:1Þ
give better quality in field measurements.

140
 Variabilities and Uncertainties

Table 8.2. Example for the determination of the exposure P by a measurement of the activity concentration of 220Rn over 10 h

Quantity Value Standarduncertainty Distribution Indexa

m
C 231 Bq m23 23 Bq m23 Normal 95.4%
 m;bg
C 10 Bq m23 1 Bq m23 Normal 0.2%
K 2.50 0.05 Normal 3.5%
b
C 553 Bq m23 62 Bq m23
Dt 10.0 h 0.0981 h Rectangular 0.8%
Pc 5.525 kBq .h m23 0.589 kBq .h m23

a
The index gives the amount of influence of a single uncertainty to the combined uncertainty.
b m  C m;bg Þ k, where C
 m and C m;bg are the mean measured activity concentration and the mean background activity concentration,
C ¼ ðC
respectively, and k is the calibration factor.
c 
Exposure P ¼ CDt, where C is the calculated mean activity concentration and Dt is the exposure time (10 h).

with an exposure time, Dt of 10 h, and an assumed irradiations (220Rn and/or 222Rn) are applied over a

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

time resolution of 10 min. Thus, the uncertainty of period of several hours to days in a constant atmos-
the time measurement is less than 1% of the com- phere. The exposures are calculated from the mean
bined uncertainty (Table 8.2). values of the measured values of the reference
This approach to determine exposure is for short- standard and the irradiation time. They have to be
term measurements with direct reading instru- corrected for background effects.
ments and is therefore not used for epidemiological Appendix B gives an example of how a mixed ex-
studies, but is appropriate for the determination of posure (220Rn and 222Rn) calibration can be analyzed
workplace exposures. and applied to field measurements using matrix
The following sections describe the calibration of algebra. In this example, both types of detectors
non-direct reading devices for long-term exposure (radon and thoron) are exposed to a series of 220Rn
measurements of 220Rn and 222Rn. So-called non- exposures, and some of these detectors are addition-
direct reading devices are time-integrating sampling ally exposed to one 222Rn level.
devices (Section 5.2, especially SSNTD in Table 5.1
as well as thoron PADC given in Figure 5.11). The ab
8.2.2.2 Non-direct Reading Devices: Rn-222
initio calculation of measurement uncertainty is dif-
Exposure Calibration. There is one procedure
ficult and requires the identification of all sources of
available for the calibration of non-direct reading
uncertainty. Because etched-track detectors are
devices for 222Rn exposure. In analogy to the 220Rn
commonly used in radon studies, the sources of un-
exposure calibration discussed in the previous
certainty associated with measurements of 222Rn
section, it is a secondary method based on calibra-
with etched-track detectors were considered by
tion via a stable reference atmosphere.
many studies (e.g., Hanley et al., 2008; Hardcastle
There are quite a number of facilities, so-called radon
and Miles, 1996; Miles, 1994; Miles et al., 2004). The
chambers or STAR (System for Test Atmospheres
results provide a basis for an ab initio approach in
with Radon) available worldwide. These facilities
some special cases (see Section 8.2.2.2), but the im-
have to be traceable to one of the facilities listed in
plementation of this approach has to be done with
the Calibration and Measurement Capabilities of the
great care: missing a source of uncertainty, choosing
BIPM (http://kcdb.bipm.org/).
an improper probability function, or overestimating
As described earlier, there are different non-direct
the influence of a quantity can lead to incorrect
reading devices for the measurement of 222Rn avail-
results. So before relying on an ab initio approach,
able. Since etched-track detectors are very popular
the uncertainty model and its probability functions
in radon studies, it is worthwhile to take a closer
should be checked by exposure calibration. These
look at these systems. The following list provides in-
checks can be implemented in the quality assurance
formation on subjects which an ab initio approach
system in the same way as the direct reading devices.
for the calculation of the combined uncertainty in
calibration and field measurements must take into
8.2.2.1 Non-direct Reading Devices: Rn-220 account.
Exposure Calibration. There is one procedure
available for the calibration of non-direct reading † Variations of the etched-track material
devices for 220Rn exposure. It is a secondary method Track etch materials are commercially available
based on calibration via a stable reference atmos- with the quality of material varying from supplier
phere. The devices to be calibrated are enclosed in to supplier and from manufacturing batch to manu-
the same atmosphere as the reference devices. The facturing batch. Hanley et al. (2008) calculated a

141
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

relative standard uncertainty of 2.1% for the typical † Uncertainties due to chemical change of the etch
between-sheet variability for their detectors. The track material
processing laboratory should therefore implement Etch track material is subject to many changes
rigorous Quality Assurance controls in order to and variations with time. Hardcastle and Miles
identify the variation in the material. (1996) showed that the polymer sensitivity of
CR-39 to alpha radiation damage decreased over
† Uncertainties due to variation in the etching time due to aging and fading, which may lead to
process an underestimation of the radon activity concen-
Variations in etching conditions can alter track tration. Aging refers to the detector losing its sen-
sizes. Therefore, the etching parameters such as sitivity to record alpha tracks over time when
temperature, etch time, and chemical composition stored in air, possibly due to increases in cross-
and concentration of etchant, should be monitored linking of the polymer. Fading is the loss of alpha
and kept the same (Ibrahimi et al., 2009; Miles, tracks already recorded on a detector over time
1994; Miles et al., 2004). The uncertainties asso- when stored in air, possibly due to the partial
ciated with the etching process can be measured repair of damaged trails over time. As detectors

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

using a control sample of detectors each exposed are commonly placed for a minimum of 3 months
to the same amount of alpha radiation. An during a measurement, they are subject to both
alpha-emitting source such as 241Am can be used to aging and fading effects. The estimated relative
expose the detector. standard uncertainties due to aging and fading
given by Hanley et al. (2008) are 4.5 and 4.4%, re-
† Uncertainties of the automatic track counting spectively. The effect of aging (before exposure)
system and fading (after exposure) can be significantly
Inconsistent focusing and reading of the etch minimized if the plastic is stored either below 08C
track detectors can lead to misinterpretation of or in pure nitrogen gas at room temperature and
the etch track characteristics. Illumination may pressure.
vary, the focus of the counting system may drift,
the track recognition may change, and the † Response to other radiation sources, e.g., thoron
scratches or the surface defects may deteriorate and thoron progeny
the signal-to-background ratio, increasing the The set-up of the etch track material inside a
measurement error. The relative standard uncer- detector housing will influence its response to ra-
tainty of the automatic track counting system is diation. Since there is not only one source of radi-
estimated to be 5.6% (Hanley et al., 2008). ation, all potential sources of interference should
be checked and excluded. This is by nature not
† Uncertainty in the linearity of response possible with 220Rn. Therefore, the response to
220
Detectors may be exposed to a wide range of Rn has to be determined before using the detec-
radon activity concentrations. For example, in the tors in the field. A calculation of diffusion times or
UK, the mean annual radon activity concentration lengths is not enough to prove insensitivity to
220
in a home is 20 Bq m23, but activity concentra- Rn, since diffusion is not necessarily the fastest
tions above 10 000 Bq m23 have been measured. path of gas transportation inside the detector
At higher exposures, i.e., higher numbers of alpha housing. So if the transport of radon is not gov-
particles, the probability that a new track will erned by diffusion, the assumption of insensitivity
overlap a previous track is higher. When the to 220Rn is not valid.
tracks start to overlap, the calibration curve The measurement of the 222Rn exposure in the
becomes non-linear and a correction factor must field will in most cases be influenced by thoron.
be considered. The point at which the calibration This is not an uncertainty to be taken into account
line ceases to remain linear and the degree of lin- but an error to be corrected—but this correction
earity correction required both depend on the size has an uncertainty. Figure 8.4 shows the response
of the etched tracks (Ibrahimi et al., 2009). of a closed etched-track detector to a thoron activ-
Therefore, consistent etched-track size is required ity concentration in the PTB thoron chamber. This
for accurate determination across the expected ex- detector was used to measure 222Rn exposure only
posure range. At high radon exposures, where and its response to thoron needs to be considered.
track counting becomes unreliable, a calibration
in terms of total area of tracks can be considered A detailed example of the procedures to calculate
(Miles et al., 2004). Again a linear correction the combined standard uncertainty of a calibration
can be applied to allow for the areas of track coefficient for etched-track detectors is given in
overlapping. Annex C. This example does not try to calculate the

142
 Variabilities and Uncertainties

Section 7.1.2 deals with the spatial variation of

radon activity concentrations within houses. The ac-
tivity concentration differences are, typically, higher
between different floors compared with rooms on the
same floor. To summarize Section 7.1.2, the COV of
30% (Heid et al., 2004) is considered as the “best esti-
mate” and a conservative estimate of the uncer-
tainty in exposure estimates due to variations of
radon activity concentrations between rooms. This
estimate represents the variation between rooms
both on the same floor and on different levels above
the basement.
For the determination of the individual exposure
Figure 8.4. Number of tracks as a function of exposure to 220Rn
to radon, the additional exposure at the workplace,
for closed nuclear track detectors designed to measure the
exposure of 222Rn. with a daily occupancy factor of 0.333, equivalent to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

8 hours per working day (ICRP, 1994) and in outdoor
uncertainty ab initio, because the complete data for air, with a daily occupancy factor of 0.2 (UNSCEAR,
that were not available. The annex also describes 2000) must also be considered.
the procedure to calculate the uncertainty associated
with a measurement in the field together with the 8.3.2 Uncertainties in Extrapolating a
decision threshold and detection limit. The examples Short-term Measurement to an Annual
are consistent with ISO (1995) and are based on mea- Average
surements carried at the PTB, Germany. The reader Annual average radon activity concentration in
is also referred to the ISO 11665–4 (ISO, 2012c), living spaces has become the standard concept when
which describes procedures for the calculation of un- determining the exposure to radon. For practical
certainty associated with field measurements using reasons, normally shorter measurement periods
passive detectors. than a full year are used. Section 6.4.2 considers the
variation in the estimates of the annual average
radon activity concentration based on short-term
measurements. The coefficient of variation of the
8.3 Other Sources of Uncertainties in
1-month long radon measurements for the annual
Assessment of the Annual Average Radon
average radon level was typically 40% (Table 6.1).
Activity Concentration
The variation of half-year or 3 months averages
8.3.1 Uncertainties due to Spatial Variation (15–30%) is low enough to provide good substitutes
of Indoor Radon Activity Concentration for the annual average in most applications. The
in Dwellings coefficient of variation for a measurement with dur-
ation of 3 months was about 25%. Shorter-term (2 or
Variation in radon activity concentration between
3 d) measurements are very poor predictors of the
different rooms of a dwelling increases the overall un-
annual average.
certainty of the estimated radon exposure to residents
In some countries, the annual radon activity con-
when only one or two rooms have been measured.
centration was calculated from short-term measure-
Radon monitors are placed usually in two inhab-
ments by applying seasonal correction factors (Section
ited rooms, labeled as “living room” and “bedroom.”
7.3.5). The seasonal factors in one country are differ-
The average radon activity concentration for the
ent from the seasonal corrections factors in another
house is usually calculated as an average between
country. Therefore, every country should develop its
these two rooms weighted by their relative occupancy.
own correction factors to extrapolate the results
In the UK, for example, the relative occupancy of the
obtained for several months to the annual result if a
bedroom is assumed to be 0.55 based on results of
correction factor is to be used.
a national survey (Wrixon et al., 1988). The relative
occupancy of the “living room” is taken as 120.55 ¼
8.3.3 Uncertainties due to Long-term
0.45. The uncertainties arise from the fact that the
Variation in Annual Average Radon
radon activity concentrations in the rooms without
Activity Concentration
measurements may differ from the radon activity
concentration in the living room/bedroom, which is Section 7.3.7 deals with year-to-year radon varia-
used as a substitute for the activity concentrations in tions. Typical values of year-to-year variation
the other rooms. expressed as the coefficient of variation in the annual

143
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

average radon activity concentration in the studies 8.3.5 Uncertainties Associated with the
reviewed were in the range of 25–40%. The collabora- Estimate of Individual Exposure
tive radon risk analysis in EU countries found a Obtained with Areal Measurements
country-specific coefficient of variation of 17–62%.
For a given activity concentration of radon progeny
These estimates were based on repeated measure-
in air, the intake is directly proportional to the
ments of radon activity concentrations in the same
breathing rate. Therefore, lung doses are not only
dwelling in different years (Darby et al., 2006). In
determined by the radon progeny activity concentra-
some countries, the measurement period was for
tion but also by the individual breathing pattern,
1 year, while for others, it was for 2–3 months.
which depends upon the physical activity of the indi-
vidual. The breathing rate for an individual asleep in
8.3.4 Combined Uncertainty in the the bedroom is lower than that for an individual
Estimation of Long-term Average Radon awake in the living room. For example, for dosimetric
Activity Concentration modeling, the recommended values of daily breathing
The combined uncertainty of the long-term rates for an adult male at home are 0.45 m3 h21 for
sleep and 1.18 m3 h21 while awake (1/3 sitting and 2/

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

average radon activity concentration can roughly be
estimated on the basis of the reviews of the effect of 3 light exercise) (ICRP, 1994). An improved estimate
short-term radon measurements, spatial variation of the residential annual exposure to radon can be
within the dwelling, and the variation over years. made by taking account of both the occupancy and
Table 8.3 summarizes the uncertainties in the long- the average breathing rate in the monitored rooms.
term average radon activity concentration based on One possible approach is to calculate a weighted
measurements made in a single year with a meas- average radon activity concentration, CRn as follows:
urement period of 3 months (Table 6.1) (Miles et al., P
i CRni Oi Bi
2012; Steck, 2005). The uncertainties due to different CRn ¼ P ð8:2Þ
i Oi B i
sources have been combined using the law of propa-
gation of uncertainty (ISO, 1995) assuming these where CRni , Oi , and Bi are the average radon activity
quantities are multiplicative and independent. concentration, occupancy, and average breathing rate
The estimate of the combined uncertainty of 49% of the individual in room i, respectively. The occu-
can be compared with the detailed error analysis of pancy, Oi , represents the fraction of the time spent in
the German epidemiological analysis (Heid et al., room i as a fraction of the time spent at home.
2004). In the German study, year-long measure- The long-term radon gas measurements in dwellings
ments were used, in bedroom and living room. In with passive detectors are continuous with no time-
addition, the year-by-year variability was studied resolved information. Uncertainties, therefore arise if
using results of repeated 1 year measurements in a CRn is not representative of the radon activity concen-
given house. According to Heid et al. (2004), the tration in the room while it is occupied. For example,
resulting total error size estimate of 0.55 corre- the ventilation of the bedroom (due to opening or
sponds to a coefficient of variation of 60%. The un- closing windows) may be different at night while it is
certainties listed in Table 8.3 do not take into occupied compared with during the day when it is not
account of different breathing patterns of indivi- being used. The dwelling may only be occupied for part
duals in different rooms and that the 3-month mea- of the day or the week, but the passive detectors record
surements carried out in these rooms may not be continuously. Furthermore, the behavior of the inhabi-
representative of the individual exposure. tants may change during the monitoring period by in-
creasing or decreasing the ventilation.
Table 8.3. Example summary of sources of uncertainty, and the Measurements of the radon activity concentration
estimated combined uncertainty, in the long-term average radon in homes may be not be a good proxy for individual
activity concentration, in the same house without modifications. exposure because of the above reasons and because
A measurement period of 3 months has been used it does not take account of the individual’s exposure
spent outside their homes, e.g., outdoors, workplace,
Source of uncertainty Coefficient of
variation (%) or in other buildings (Section 6.6).

Three months versus 1 year long measurement 25%

Spatial variation within the house 30%
8.4 Uncertainties of Radon Progeny
Variation over years 30%
Measurements
Combined 49%
The general aspects of the uncertainty of the cali-
bration and the preparation of field measurements

144
 Variabilities and Uncertainties

are the same as described in Section 8.2.1 for radon content is controlled by an aerosol generator based on
gas activity concentrations. Carnauba wax and a High Efficiency-Particulate Air
Because the determination of the activity concen- (HEPA) filter system.
tration of progeny is much more dependent on the The activity concentrations of thoron and thoron
sampling process than on the measurement itself, progeny are adjusted by means of 10 open exhalation
this has to be reflected in the uncertainty budget. sources with 228Th, which are distributed inside
the chamber. For the addition of 222Rn, an external
8.4.1 Measurand and Derived Quantities source is available. Homogeneity of the environmen-
tal parameters and of the activity concentrations of
The measurement of a progeny activity concentra- thoron and its progeny in the chamber is ensured by a
tion, potential alpha energy concentration, an equi- special ventilation system. With this system, a com-
librium factor, or an unattached fraction requires a promise has been made between high circulation
number of simultaneous measurements (for example, rates due to the short half-life of thoron, on the one
an a-spectrum, the volume flow of the sample, times hand, and low flow velocities to avoid entrainment
for sampling and measurement). The measurands of aerosol particles from the chamber surfaces, on the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

are the number of counts in a special region of inter- other hand. Exposure parameters for calibration
est, the volume flow, and time. With these measur- purposes are (i) environmental parameters (tempera-
ands, the determination of an activity concentration ture, humidity, air pressure, aerosol size distribution),
as a derived quantity is possible. This is achieved by (ii) radon gas activity concentration (220Rn and/or
mathematical equations with nuclear data which 222
Rn), and (iii) radon progeny activity concentration
have uncertainties as well. (216Po,218Po, 212,214Pb, 212,214Bi, 212,214Po).
The easiest way to implement traceability to the The environmental parameters temperature and
derived value would be to calibrate the device in a ref- humidity are controlled by the climate control of the
erence atmosphere. This implies using the assigned air-conditioned chamber. The atmospheric para-
uncertainties in calibration and assessing the uncer- meters temperature, humidity, and pressure are
tainties in field measurements separately as described measured also by a sensor system distributed inside
in Section 8.2. the chamber. This system is traceable to national
Whether this approach can be used for a derived standards.
quantity depends on the quantity, the individual
device and its range of application. A calibration is
8.4.1.2 Reference fields for 222Rn progeny
only possible if for a quantity (measurand or derived
concentrations. There is one procedure available
quantity), a stable response in a range of parameters
for the calibration of 222Rn progeny devices. It is a
is achieved. Thus, different reference atmospheres
secondary method based on calibration via a stable
are necessary to assure that a calibration is feasible
reference atmosphere. The devices to be calibrated
for the scope of application.
are enclosed in the same atmosphere as the reference
devices. The irradiations are applied over a period of
8.4.1.1 Reference fields for 220Rn progeny several hours to days in a constant atmosphere.
concentrations. There is one procedure available There are quite a number of facilities (so-called
for the calibration of 220Rn progeny devices. It is a radon chambers or STAR) available worldwide. Some
secondary method based on calibration via a stable of these facilities can be operated for 222Rn progeny
reference atmosphere. The devices to be calibrated calibrations. These facilities have to be traceable to
are enclosed in the same atmosphere as the refer- one of the facilities listed in the Calibration and
ence devices and exposed over a period of several Measurement Capabilities of the BIPM (http://kcdb.
hours to days in a constant atmosphere. bipm.org/).
For the calibration of the activity concentration of
radon, thoron, and their progenies, a reference atmos-
8.4.2 An Example for the Determination of
phere has been established at PTB, Germany (BfS,
Derived Quantities
2006; 2007; BMU, 2009). This field consists of an air-
conditioned walk-in testing chamber of 6 m3 volume, Derived quantities are values calculated from one
in which the environmental parameters temperature, or more measurands (e.g., uncorrelated or correlated
air humidity, and aerosol content can be adjusted and count rates) in combination with some constants
controlled. The range of temperature is from 08C to (e.g., calibration factors, nuclear data and correction
708C, while the relative humidity can be controlled factors).
from 10% to 95%. A defined air movement is estab- The quantity radon activity concentration is a
lished by two fans on the roof of the chamber with an simple derived quantity with only one measurand
adjustable output from 30% to 100%. The aerosol (alpha counts), while the equilibrium equivalent

145
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

activity concentration Ceq includes three measur- conditions are variable so that F fluctuates with
ands (counts for 218Po, 214Pb,214Bi) or alternatively time. The calculations are based on the 24 individ-
two correlated measurement series (alpha counts of ual measurement results given in Figure 8.5.
218
Po and 214Po as a function of time). Assuming stable exposure conditions, the standard
The following simplified example shows a 24 h error of the mean is calculated for  so that its
pffiffiffiffiffiC
ffi
field measurement with a time resolution of 1 h to 
standard uncertainty u(C) ¼ sðCÞ= 24 (Table 8.4a).
obtain the equilibrium factor in a room. Two inde- However, if the exposure conditions are assumed to
pendent measuring devices were used to determine be variable then u(C) ¼ sðCÞ, where sðCÞ is the
the radon activity concentration CRn,m and the equi- standard deviation of C (Table 8.4b).
librium equivalent concentration Ceq,m. The symbols For stable exposure conditions, the determined
CRn,m and Ceq,m denote the readings from these mean equilibrium factor is F  ¼ (0.37 + 0.06) with
devices without applying corrections for calibration k ¼ 2 and the dominating uncertainty is the calibra-
or for the background readings. Figure 8.5 shows the tion of the two devices (Table 8.4a), whereas for un-
measurement results for both CRn,m and Ceq,m as well stable exposure conditions, the result is F  ¼ (0.37 +
as the calculated equilibrium factor, Fm ¼ Ceq,m/ 0.18) with k ¼ 2 (Table 8.4b). The mean value is the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

CRn,m. Although the functions CRn,m and Ceq,m seem same, but the difference in the uncertainty is large.
to run rather parallel as a function of time, their quo- In the given example, the assumption of a constant
tient shows large fluctuations. atmosphere is not sustainable within the uncertainty
To analyze the data further, the background read- of Table 8.4a. Nevertheless, the overall uncertainty
ings and the calibration factors of the devices are given in Table 8.4b can be considered as too large in
required together with their uncertainties. In this the case of quasi-stable situations in the room’s
example, the calibration factors are kRn ¼ (1.06 + atmosphere. If the data set is split at 14 h as shown
0.03) and keq ¼ (0.88 + 0.05) and their respective by the broken line in Figure 8.5, it means that the
background readings are CRn,bg ¼ (15 + 5) Bq m23 decrease in the radon activity concentration after
and Ceq,bg ¼ (5 + 2) Bq m23. The mean equilibrium
 is given by:
factor, F,
Table 8.4. The uncertainty budget for F assuming either (a) the
exposure conditions are stable so that the actual value of F is

 ¼ ð Ceq;m  Ceq;bg Þkeq
F ð8:3Þ
constant during the measurement period or (b) that the exposure
 Rn;m  CRn;bg ÞkRn conditions are variable so that F fluctuates with time
ðC
Quantity Value Standard uncertainty Indexa
where, C Rn;m and C
 eq;m are the mean radon activity
concentration and the mean equilibrium equivalent (a)
 eq;m
C 705 Bq m23 29 Bq m23b 26.2%
activity concentration.
Ceq,bg 5 Bq m23 2 Bq m23 0.1%
Table 8.4 gives the uncertainty budget for F as-
keq 0.88 0.05 50.8%
suming either (a) the exposure conditions are stable  Rn;m
C 1572 Bq m23 39 Bq m23b 10.0%
so that the actual value of F is constant during the CRn,bg 15 Bq m23 5 Bq m23 0.2%
measurement period or (b) that the exposure kRn 1.06 0.03 12.6%
c
F 0.37 0.03
(b)
 eq;m
C 705 Bq m23 140 Bq m23d 67.0%
Ceq,bg 5 Bq m23 2 Bq m23 0.0%
keq 0.88 0.05 5.4%
 Rn;m
C 1572 Bq m23 192 Bq m23d 26.3%
CRn,,bg 15 Bq m23 5 Bq m23 0.0%
kRn 1.06 0.03 1.3%

F 0.37 0.09

a
The index gives the amount of influence of a single uncertainty
to the combined uncertainty.
b
The standard uncertainty is set equal to the standard error
pffiffiffi
¼ sðCÞ= n, where s(C) is the standard deviation given in
Table 8.3b and n is the number of measurements, 24 in this
example.
Figure 8.5. Results of 24 h field measurements of the radon c
F ¼ ðC  Rn;m  CRn;bg ÞkRn , where, C
 eq;m  Ceq;bg Þkeq =ðC  Rn;m and
activity concentration CRn,m and the equilibrium equivalent 
Ceq;m are the mean radon activity concentration and the mean
activity concentration Ceq,m in a room. No corrections for equilibrium equivalent concentration. The respective background
calibration or for the background readings have been made. The readings and calibration factors are CRn,bg, Ceq,bg, kRn, and keq.
d
equilibrium factor Fm ¼ Ceq,m / CRn,m is also given at the y-axis to The standard uncertainty is set equal to the standard
the right. deviation, s(C).

146
 Variabilities and Uncertainties

14 h can be considered separately. Thus, results for F A comparison of dose-exposure conversion coeffi-
 h , t , 14 hÞ ¼ (0.39 + 0.09)
can be expressed as: Fð0 cients obtained by different lung dosimetry models

with k ¼ 2 and Fð14 h , t , 24 hÞ ¼ (0.35 + 0.15) for uranium mining exposure conditions, shown in
with k ¼ 2. These results overlap in the assigned Table 3.15, indicates a range of dose values from
uncertainties. 4.2 to 12.7 mSv WLM21, with an average value of
Splitting a data set has to be done with great care. 7.9 mSv WLM21. It should be noted, however, that
It should never be done to provide small uncertain- exposure conditions varied among the different
ties in results. If there is evidence of a change of model calculations and thus may overestimate the
exposure conditions, for example, a change in the differences between the various model predictions.
ventilation, then it may be reasonable. However, for Indeed, recent calculations for three different models,
radiation protection purposes, long-term measure- the compartmental HRTM model (ICRP, 1994), a de-
ments are generally required to estimate an annual terministic airway generation model (Winkler-Heil
average quantity. et al., 2002), and the stochastic airway generation
Summarizing the discussion of uncertainties in model IDEAL-DOSE (Hofmann et al., 2010) using
measurements, it is indispensable to provide clear the same mining exposure conditions produced rela-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

information about tively similar results, ranging from 8.3 to 11.8 mSv
WLM21 (2.3 to 3.3 mSv per mJ h m23). This suggests
(a) the basis of their measurement (traceability) that the application of current radon lung dosimetry
and models will produce dosimetric uncertainties of the
(b) the model of uncertainty analysis in which all order of about 30%.
sources of uncertainty are included in the
assigned uncertainty of the derived quantity.
8.5.2 Uncertainties of Model Parameter
Values used in Dose Calculations

8.5 Uncertainties of Dosimetric Results Currently, a variety of uncertainties of physical and

biological parameter values affect lung dosimetry.
The primary sources of the uncertainty of dosi- These include the relative contributions of sensitive
metric results are (1) the application of different basal and secretory cells to lung cancer risk (Section
dosimetric models, which vary with respect to lung 8.5.2.1), the size distributions of attached and un-
anatomy, deposition equations, clearance velocities, attached fractions (Section 8.5.2.2), the apportionment
location of target cells, and the application of differ- factors for bronchial, bronchiolar, and alveolar–inter-
ent mathematical modeling techniques, (2) still stitial regions for the calculation of average lung doses
existing uncertainties of relevant parameters, such (Section 8.5.2.3), and the correct value of the radiation
as the relative contribution of sensitive target cells, weighting factor for radon progeny alpha particles
apportionment factors, and the radiation weighting (Section 8.5.2.4).
factor for alpha particles, and (3) the inter-subject
variability of lung anatomy, particle deposition, 8.5.2.1 Sensitive Target Cells. At present,
particle clearance, and cellular dosimetry (Sections basal and secretory cells are considered to be the
3.4 – 3.8). primary target cells in bronchial epithelium (ICRP,
1994). Because of lack of more pertinent information,
it is further assumed that both cell types have the
8.5.1 Application of Different Dosimetric
same volumetric density of cell nuclei across the epi-
Models
thelium and the same radiosensitivity, i.e., the same
Over the last decades, a considerable number relative contribution to tumor induction. Hence bron-
of dosimetric models for inhaled radon progeny has chial doses are commonly expressed as the average of
been published in the open literature (see Section 50% basal and 50% secretory cell doses, except for the
3.9.1). These models utilize different anatomical IDEAL-DOSE model, where basal and secretory doses
models of the human lung, apply different depos- are weighted by their relative nuclear volumetric
ition equations, assume different mucociliary clear- densities (Winkler-Heil and Hofmann, 2005). Thus
ance velocities, and use different geometric models different assumptions about the relative contributions
of the bronchial epithelium. Furthermore, they are of basal and secretory cells will affect bronchial dose
based on different conceptual modeling philosophies estimates. For example, calculations for mining expos-
and computational methods, ranging from semi- ure conditions revealed that the choice of different
empirical compartment models to airway gene- target cells or any combination thereof leads to bron-
neration models and from analytical to stochastic chial doses ranging from 1.85 to 6.23 mGy WLM21
modeling techniques. (see Table 3.16), i.e., varying by a factor of about 3.

147
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

8.5.2.2 Radon Progeny Size Distributions. an average lung dose of 2.36 mGy WLM21. For com-
As shown in Tables 3.13 and 3.14, lung doses pre- parison, the application of non-uniform apportion-
dicted by different dosimetric lung models can vary ment factors (ABB:Abb:AAI ¼ 0.80:0.15:0.05), based on
by more than a factor of 2, depending mainly upon the relative cancer sensitivity of the three regions of
aerosol particle size, unattached fraction, and target the lung (Porstendörfer, 2002), results in an average
cell depth. Marsh and Birchall (2000) found that the lung dose of 4.07 mGy WLM21. Thus, calculated
unattached fraction and the activity median diam- average lung doses may vary by nearly a factor of
eter of the attached fraction are the aerosol para- 2 depending on the assumed apportionment factors.
meters that most affected the equivalent dose per
WLM. For example, it is well published that the 8.5.2.4 Radiation Weighting Factor. ICRP
attached fraction of aerosol particle size can change currently uses a radiation weighting factor of 20
from 100 to 400 nm, and the unattached fraction for alpha particles to convert the absorbed dose (in
from 0.05 to 6 nm. A change from 100 to 400 nm for gray) to an organ/tissue to an equivalent dose (in sie-
the attached fraction reduces the effective dose per verts). This value was chosen for radiation protection
WLM by about a factor 2 due to the less efficient dif- purposes only and is based on the observed relative

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

fusion deposition. On the other hand, the change of biological effectiveness (RBE) of alpha particles for
the size of the unattached fraction from 0.05 to 6 nm many types of radiation effects (ICRP, 2007). The
increases the effective dose per WLM by about a RBE values for alpha particles were derived primar-
factor 3 due to the less efficient filtration in the ily from animal experiments, but values obtained
nasal or oral region (see Figure 3.3). from human data were also considered (Harrison and
Muirhead, 2003). However, RBE values for in vitro
8.5.2.3 Apportionment Factors. In radiation oncogenic transformation in different immortalized
protection, lung cancer risk is commonly related to an cell lines irradiated by alpha particles (with varying
average lung dose. Depending on the weighting pro- LET and dose) range from 2.4 to 20, with an average
cedure applied to basal and secretory cells, doses to value of about 7 (Hofmann et al., 2004). Although the
the large bronchi (BB) are about a factor of 2 higher application of in vitro transformation data to in vivo
than those to the bronchiolar (bb) region (Table 3.16). carcinogenesis in bronchial tissue requires several
Moreover, doses to the alveolar–interstitial region extrapolations, these smaller values are consistent
are between one and two orders magnitude smaller with the reported range of RBE values from 4–8 for
than those to the bronchial and bronchiolar regions lung cancer incidence in rats exposed to radon
(Table 3.3). Thus, for the assessment of an average progeny (Cross and Monchaux, 1999). In conclusion,
lung dose, an additional weighting procedure has the actual radiation weighting factor for bronchial
been introduced through apportionment factors for carcinomas caused by radon progeny alpha particles
the BB, bb, and AI regions. In contrast to uniform may be smaller than 20 by about a factor of 2 or even
dose distributions in the case of external exposure to 3 (Hofmann et al., 2004).
gamma radiation or neutrons, the dose distribution
8.5.3 Inter-subject Variability
produced by inhaled radon progeny is distinctly non-
uniform. Assuming that the epithelium in each The primary biological parameters contributing
region is equally sensitive to cancer induction, this to inter-subject variability of bronchial doses are the
then raises the question whether each region of the variability of the extrathoracic airways, the bron-
human lung should contribute equally to lung cancer chial and alveolar airway structure and airway
risk in the case of radon progeny exposure or whether dimensions, random variations of breathing para-
the contribution of each region should be based meters, individual mucociliary clearance velocities,
on their relative dose. Moreover, the distribution and variation of the thickness of the bronchial
of lung cancers among bronchial and bronchiolar epithelium and related depths of target cells (see
airways observed in pathological examinations may Section 3.8.2). Calculations of inter-subject variabil-
be used as an additional weighting procedure. ity indicated that the asymmetry and variability of
ICRP (1994) currently uses a tissue weighting the airway geometry is the most important factor,
factor, wT, of 0.12 for the whole lung, consisting of followed by the filtering efficiency of the nasal pas-
BB, bb, and AI regions. The apportionment of the sages and by the diameter-related thickness of the
lung tissue weighting factor to each of the three bronchial epithelium (Hofmann et al., 2010).
regions of the lung (BB, bb, and AI) assumed by Several results of variability or uncertainty calcu-
ICRP (1994) is ABB:Abb:AAI ¼ 0.333:0.333:0.333. lations have been published in recent years, in
Based on the dose estimates listed in Table 3.16 and which the obtained dose distributions were repre-
assuming equal weighting of basal and secretory sented by lognormal distributions. For example, the
cells and equal weighting of regional doses leads to uncertainty analysis performed by NCRP (2009)

148
 Variabilities and Uncertainties

resulted in a geometric mean of 9 mSv WLM21, with of an exposure–effect relationship, where lung
a geometric standard deviation (GSD) of 1.6. Marsh cancer risk is related to measured average radon
et al. (2002) carried out a parameter uncertainty activity concentrations. However, this approach
analysis with the HRTM (ICRP, 1994) to calculate neglects the fact that lung cancer is caused by the
the probability distribution of the weighted equiva- inhalation of the short-lived radon progeny and not
lent dose to the lung per unit exposure to radon by the inhaled radon. Alternatively, lung cancer risk
progeny in the home. The resulting dose distribution can also be expressed in terms of a dose–effect
was approximated by a lognormal distribution with relationship, where lung doses are determined by the
a geometric mean of 14 mSv WLM21 and a GSD of application of dosimetric models, which consider the
1.5. Similar results were obtained by Birchall and effect of inhaled radon progeny. In both approaches,
James (1994) for the exposure to radon progeny in uncertainties of either radon exposure levels or lung
a mine. These results suggest that a GSD of about doses affect the estimation of lung cancer risk in epi-
1.5 is representative for the weighted equivalent demiological studies by assigning error bars on the
dose to the whole lung. The analysis of the effect x-axis. Note that the lung cancer risk plotted on the
of intra- as well as intersubject variability on bron- y-axis is also associated with error bars, which are

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

chial and bronchiolar doses indicated much wider particularly large at low radon levels.
lognormal dose distributions (Hofmann et al., 2010):
BB: median ¼ 3.2 mGy WLM21, GSD ¼ 2.3; bb: (1) The uncertainty of exposure levels in current epi-
median ¼ 2.3 mGy WLM21, GSD ¼ 4. Note that con- demiological studies is dominated by the retro-
sideration of intra-subject variability which is an in- spective reconstruction of past exposures from a
herent feature of the stochastic dosimetry model, limited number of measurement sites, a limited
produces wider dose distributions than determinis- number of passive detectors, and an exposure
tic dosimetry models, where each individual is char- time which represents only a small fraction of
acterized by a single dose value. time during which the initiation of lung cancer
could occur. Indeed, a recent study (H. Paretzke:
8.5.4 Summary of Uncertainties of Dose
personal communication) demonstrated that the
Calculations
individual exposures of 23 test persons, deter-
The above discussion of uncertainties of dosimet- mined by a personal exposure meter (Karinda
ric results suggests that the primary sources of un- et al., 2008), were on average a factor of 2 greater
certainties are the role of sensitive target cells, the than those estimated on the basis of indoor mea-
definition of the radon progeny size distributions, surements with passive devices.
particularly the unattached fraction, the choice of (2) Moreover, breathing rates depending on individ-
regional apportionment factors, and the appropriate ual physical activities, and hence bronchial dose,
value of the radiation weighting factor for radon were not known or were not recorded. For typical
progeny alpha particles in the bronchial region. indoor exposure, a breathing rate of 0.78 m3 h21
However, despite their significance, defined values is commonly assumed, consisting of 55% resting
of these factors are currently adopted in internation- (sleeping), 15% sitting awake, and 30% light ex-
al radiation protection regulations. ercise. Since the breathing rate for sleeping,
Uncertainty of parameter values as well as inter- sitting, and light exercise varies from 0.625 to
subject variations of anatomical and physiological 1.25 l for adult men, and from 0.444 to 0.992 l for
parameters lead to lognormal dose distributions adult women (ICRP, 1994), any other combin-
with GSDs of the order of 1.5, while the additional ation of activity patterns will lead to different
consideration of intra-subject variability yields GSDs breathing rates and thus radon inhalation rates.
of the order of 2.5 (Hofmann et al., 2010).
Although different dosimetry models have pro- Since radon gas measurements do not account for
duced significant differences in dosimetric results in the fact that doses to bronchial target cells are
the past, recent model predictions suggest that the caused by the short-lived radon progeny, an equilib-
uncertainties caused by the application of different rium factor has been applied to relate measured
radon lung dosimetry models will not introduce un- radon gas activity concentrations to the potential
certainties of dose estimates greater than 30 or 40%. alpha energy concentration (PAEC) of the progeny,
because this may be a better indicator of dose.
Measured equilibrium factors in different indoor
8.6 Effect of These Uncertainties on the
environments range 0.2 –0.9, with mean values
Analysis of Epidemiological Studies
between 0.3 and 0.4, depending primarily on venti-
Lung cancer risk due to radon exposure in epi- lation conditions (Porstendörfer, 1994). Thus, for in-
demiological studies is commonly reported in terms dividual exposure conditions, the corresponding

149
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

equilibrium factor can deviate from the average In conclusion, radon levels reported in epidemio-
value by a factor of 2. However, measurements have logical studies are associated with significant uncer-
shown that the equilibrium factor (F) is negatively tainty bars, which are largest at the lowest exposure
correlated with the unattached fraction (fp) for con- levels, thereby affecting the statistical uncertainty
ditions where the ventilation rate is relatively low. of derived lung cancer risk estimates at low radon
Taking account of this negative correlation between levels. However, it is in the low exposure region
F and fp, it has been shown that for indoor air, the where the most reliable estimates of lung cancer
radon gas activity concentration is a more robust in- risk are needed in order to derive statistically
dicator of dose than the PAEC under a range of significant radon exposure limits for homes and
aerosol conditions normally encountered. workplaces.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

150
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv015
Oxford University Press

9. Recommendations

This section provides recommendations regarding † Ancillary measurements for retrospective expos-
the optimum measurement strategies, i.e., best prac- ure estimates
tices, choice of measurement techniques, appropri- † Evaluation of population exposure or dose guidelines
ate recording of measurements, and reporting † Legal purposes

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

format of measurement results depending on the † Research
objectives. Target audiences are (i) authorities who
are planning radon surveys, (ii) those carrying out
measurements, (iii) those conducting epidemiologic- 9.1 Good Practice Recommendations
al studies, and (iv) those reviewing and evaluating
Each assessment of the radon and thoron situation
past (historical) studies.
in an area of interest is based on the measurements
Due to the importance of reliable measurements
of the respective activity concentrations. Since a
of radon activity concentrations, one of the past
measurement is the basis of all further evaluations, it
developments in international metrology addressed
is of fundamental importance to follow the rules of
two basic needs: (1) the harmonization of metrology
metrology to obtain reliable measurements. The un-
within the scope of the Mutual Recognition Arrange-
certainties in measurements drive the uncertainties
ment (MRA), an arrangement drawn up by the
in risk assessment. Strict adherence to the rules will
International Committee of Weights and Measures
broadly affect risk assessment capability.
for the mutual recognition of national standards and
Following these rules requires the correct usage of
of calibrations issued by national metrology institutes
an international vocabulary of metrology in general
and (2) the increased demands of the European
and the specific vocabulary of radon in particular
Atomic Energy Community (EURATOM) and the
(Glossary and Section 8). The usage of quantities
International Atomic Energy Agency (IAEA) direc-
and units according to international standards is
tives, transferred into national radiation protection
fundamental, as well as following the basic require-
regulations with regard to natural radioactivity and
ments of quality assurance.
its quality-assured measurements (Section 8).
In order to obtain accurate results, measurements
have to be performed with a traceable calibrated in-
strument and a measure of uncertainty has to be
Measurement objectives assigned to each measured value.
Radon gas or radon progeny are measured for Performing a measurement that is able to meet
diverse purposes, such as the determination of: scientific or legal criteria implies that the measure-
ment should be suited to the purpose with regard to:
† Individual exposure or remedial action decisions
† Post-remediation testing to test effectiveness of (1) its appropriateness (e. g., measured quantity,
mitigation systems sampling type, long- or short-term measure-
† Average population exposure in a region or ment), and
country (2) its properties (e. g., traceability, uncertainty, de-
† Distribution of exposure in a population tection limit, range of application).
† Demonstration of compliance with reference levels Good practice recommendations comprise good
and dose limits practice in recording and reporting, the use of app-
† Airborne progeny properties for lung dose ropriate nomenclature, e.g., SI units, and quality
calculations assurance for data validation.
or for other purposes such as: It must be emphasized that although performance
criteria are in place, it is up to individuals to carry
† Input for risk estimates from exposure distributions out the practices. Change in staff, location, budget,

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

etc. can impact on usual practices. Therefore, it is 9.2.1 Individual Dwellings

important that all aspects of the data acquisition
In dwellings, it is recommended that radon activ-
process be subject to ongoing evaluation.
ity concentrations be measured in at least two speci-
The combination of a reliable measurement and a
fied rooms. High occupancy rooms are more
well-designed and conducted survey is required for a
meaningful, such as the main living room and the
study to be scientifically valid. The above-listed
bedroom, than less inhabited rooms. As most of the
aspects of the measurements together with the fol-
indoor radon in a dwelling comes from the ground
lowing points for a survey should be considered.
subjacent to the building, the lowest inhabited level
† Specifying the objectives of the building is preferred as one of the measure-
† Target population, parameters to be estimated ment locations (Section 6.3).
(e.g., population densities, construction of In individual dwellings, long-term measurements
dwellings) are to be preferred and measurements over a period
† Inventory of resources, budget, staffing levels of 1 year are advisable. However, if for practical
required, detectors, data processing reasons this is not feasible, then a period as long as

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

† Choice of detector, installation, and collection of possible should be chosen, but not shorter than 3
detectors months. On the other hand, in some cases, e.g., for
† Requirements as to time schedule and accuracy radon screening in houses, short-term measure-
required ments represent a reasonable alternative.
† Sampling design, sample selection mechanism, There is a considerable seasonal variation of
and sample size determination radon activity concentrations. Seasonal correction
† Data collection method factors specific for a given climate have to be
† Information security and confidentiality (i.e., im- applied, i.e., specific seasonal correction factors have
plementation of guidelines on data protection) to be determined for a given region. National or re-
† Data processing methods, including imputation gional studies should therefore be undertaken to de-
and editing termine if there is an observable and reliable
† Specification of formulas for statistical quantities seasonal variation. Spring and autumn measure-
and measures of precision ments may give a better estimate of the annual
† Training of personnel and organization of field average radon activity concentration than the best
work seasonal correction factors applied to all measure-
† Allocation of resources to different survey ments over all seasons. An alternative approach is to
operations use only one average correction factor for heating
† Allocation of resources to control and evaluation season measurements. This has been found prac-
† Questionnaire design (if required) tical in Nordic countries where only heating season
measurements are recommended. Typical values of
year-to-year variation in the annual average radon
activity concentration in dwellings can be expressed
9.2 Recommendations Regarding by a coefficient of variation. Such coefficients of vari-
Measurement Strategies ation typically range from 25% to 40%. Ongoing
measurements should evaluate best estimates of the
Indoor radon measurements are required because
coefficient of variation as this has major impact on
it is not possible to predict with accuracy and preci-
uncertainties in risk assessment (Section 7.3.5).
sion the indoor radon levels in an individual build-
ing, including a dwelling or an indoor workplace.
9.2.2 Regional Indoor Radon Mapping
For radiation protection purposes, the appropriate
measurement result should be compared with the Radon surveys form an essential initial step in the
relevant reference level. If mitigation is carried out establishment of a national or regional radon
to reduce radon exposure, then repeat measure- program aiming to reduce the population risk. The
ments should be made to confirm the effectiveness of principal objective of a regional survey usually is to
the mitigation system and records of the measure- obtain the population distribution of annual average
ments should be kept. Remediated premises should radon activity concentrations. In the surveys of ex-
be re-measured periodically to ensure that radon posure of the population to indoor radon either
levels remain low. Measurements should also be random sample surveys or stratified sampling is
repeated after any significant building work or recommended. The distribution of indoor radon ex-
changes to an operational cycle affecting exposure posure of the population of a country or region indi-
conditions such as changes to the heating, ventila- cates the areas where action may be required to
tion, and air-conditioning operation. comply with national or regional reference levels

152
 Recommendations

(Section 6.3.2). In other words, information from If a dose assessment is required for radiation pro-
radon maps can be used to support decisions to tection purposes, then consideration of the actual
carry out radon measurements in homes and indoor parameters of the exposure situation should be
workplaces. For example, if an indoor workplace is taken into account. These may include, for example,
in a radon-prone area then radon measurements are occupancy times of the measured rooms (or in loca-
usually recommended. tions) and equilibrium factor (or radon progeny ac-
To obtain a reasonable population-weighted tivity concentrations). The assessment should also
average radon activity concentration for a given take into account of any regular (i.e., diurnal) varia-
area, the selection of houses should be based on tions in radon levels, in which case time-resolved
population densities. Other population-dependent measurements may be required. In such cases, mon-
criteria are the structure of the houses, such as itoring devices with 1 h time resolution are recom-
single family houses or multi-story town buildings, mended. However, care must be taken to assess the
and the construction material used, such as brick, full range of occupancy of the workplaces, taking
concrete, or wood. For the measurements in the into account of workers who might have non-typical
selected houses, the same recommendations apply occupancy.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

as discussed above for individual dwellings.
9.2.4 Residential Epidemiological Studies
9.2.3 Workplaces
Epidemiological studies of radon risk as a function
In indoor workplaces and in mixed-use buildings, of exposure rely on measurements or estimates of
which are used both by members of the public and radon or radon progeny exposure over long time
workers (e.g., schools, libraries, hospitals, residen- periods. Exposures from 2 to 3 month measurements
tial homes, shops, and cinemas), areal radon mea- are often imputed to even 30 year exposures with
surements with passive detectors are recommended amendments or assumptions, such as seasonal and
to demonstrate compliance with national reference annual correction factors. In some cases, long-term
levels. In general, measurements over a period of 1 measurements exist or are made retrospectively with
year are advisable. However, if for practical reasons, detectors that have accumulated radon progeny over
this is not feasible, then a period as long as possible long periods of time. Long-term measurements
should be chosen, but not shorter than 3 months. If provide the best uncertainty estimates for risk as-
appropriate, seasonal correction factors can be sessment compared with short-term measurements
applied to obtain the annual average radon activity (Section 6.4). For reliable dose assessments, how-
concentration as discussed above. ever, individual time-resolved monitoring is needed.
Regulatory requirements apply in workplaces But for residential epidemiology studies, personal
where the exposure to radon is considered as occupa- monitoring is currently not a practical option, espe-
tional. In such workplaces, individual exposure or cially for long-term measurements (Section 6.6).
dose assessment are required to demonstrate compli- Although the radon progeny deliver the relevant
ance with reference levels and dose limits. Depending lung dose, i.e., to bronchial epithelium, their use for
upon exposure conditions, individual as well as area residential exposure assessment is limited by the
monitoring may be applied (Sections 6.5.2 and 6.6.1). short time duration of measurement and the re-
If the spatial and temporal conditions are very vari- quirement of air sampling equipment. Exposure as-
able or if the individual frequently changes exposure sessment, especially for purposes of risk estimation,
sites with different exposure conditions, then individ- requires long-term data including evaluation of
ual monitoring is generally recommended, if appropri- retrospective exposure. Passive alpha track detec-
ate. For example, personal monitors are used in many tors that monitor residential radon gas atmospheres
underground workplaces, such as mines, where the for up to 1 year have generally replaced short-term
exposure conditions are variable. On the other hand, measurements. Emerging research indicates that
if spatial and temporal variations can be neglected alpha track detectors can use progeny surface depos-
and occupancy times are known, individual exposures ition to measure individual, time-resolved radon
may reasonably be approximated by areal measure- progeny exposures (Section 6.6.2).
ments. In such cases, individual occupancy times of A major uncertainty affecting the accuracy of ex-
the person at the workplace (or at the measured loca- posure estimates is the diurnal, weekly, monthly,
tions) should be recorded. seasonal, and annual variability of radon levels in
Radon progeny measurements are recommended residences. If, for practical reasons, this may not be
at workplaces where the equilibrium factor varies feasible, then a period as long as possible should be
significantly because of variation in the ventilation chosen, but at least a period of 3 months. The
or fluctuations in aerosol particle concentration. annual variability of the estimates of average radon

153
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

activity concentrations should also be taken into 9.2.5 Miner Epidemiological Studies
account when estimating risks in residential epi-
Since mines are generally classified as occupation-
demiology studies. Using seasonal correction factors,
al exposures, legally prescribed radiation protection
the radon activity concentrations determined for
practices have to be performed. Radon progeny mea-
measurement periods between 3 months and 1 year
surements were the historical choice in most under-
can be converted into annual values. If a national
ground uranium mines. The reported units were in
survey is carried out in a phased fashion (say four
Working Level Months (WLM), a unit that described
sequential 3 month measurement periods or three 4
the total alpha energy concentration of progeny per
month periods), then, in principle, its results could
liter of air multiplied by the working months spent
be used to generate site-specific seasonal correction
at various locations.
factors (Section 7.3).
Occupational exposure assessment in under-
In principle, the same measurement strategies
ground mines still relies upon real-time equipment
apply as for individual dwellings. In addition, the
to measure progeny mainly because regulations are
following aspects must be considered:
based on dose. Due to the dependence of lung doses

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

on individual physical activities, and because expos-
† Occupancy patterns (daily, yearly)
ure conditions are very variable, personal measure-
† Radon data for previous homes, workplaces, and
ments should be carried out, although they may be
outdoors
supplemented by areal measurements. In the case of
† Physical activities
uranium miners, airborne radioactivity other than
† Smoking status, also spousal smoking
radon progeny, i.e., long-lived radionuclides in the
† Gender and age
uranium ore dust, and external gamma radiation
† Occupant work-related information
also contribute to their exposure.
Since the actual lung carcinogens are the radon
progeny and risk-based studies would require dose 9.2.6 Retrospective Measurements
estimates that build upon exposure data, the add- For the estimation of the retrospective exposure to
itional measurement parameters related to radon radon in residential epidemiological studies, radon
progeny exposure are: activity concentrations can be derived from the
measurement of the long-lived radon progeny with
† Equilibrium factor surface or volume trap detectors (Section 5.4).
† Unattached fraction of the PAEC Examples of suitable surface traps are:
† Activity size distributions of the attached and un-
attached radon progeny † Glass in a picture or photograph frame
† A wall mirror in a bedroom or living room
The residential radon activity concentration is † The outer vertical surfaces of glass in a display
mainly measured by passive radon detectors. cabinet or vitrine
Although passive radon detectors are usually † Glass in a door between rooms
designed to detect radon efficiently and exclusively, † The flat glass surface of a clock
several types (actually used in major epidemiologic-
Examples of suitable volume traps are:
al studies) can detect thoron together with radon. In
this case, these detector readings may include both † Filling material of cushions
radon and thoron signals and lung cancer risk will † Mattresses
be given as a biased estimate when epidemiological
studies are carried out (Section 5.2.3). Uncertainty estimates for retrospective measure-
Information on doses derived from thoron progeny ments are not well established. This is an area for
inhalation is much sparser than that from radon worthwhile research.
progeny. Indoor levels of thoron are generally much
9.2.7 Outdoor Radon Mapping
lower than radon levels unless the materials of the
internal surfaces of the building have a high content Methods available for outdoor radon mapping are
of thorium. Considering the spatial distribution of aerial gamma surveys, soil gas measurements, and
the thoron activity concentration indoors, it is ques- measurements of ambient radon activity concentra-
tionable to apply the equilibrium factor of thoron tions. The population density distribution and the
for the determination of thoron progeny activity variation in geology are valuable input parameters
concentrations. Therefore, direct measurements of when designing radon maps and identifying high
thoron progeny ( particularly 212Pb) are recom- radon areas. Spatial and temporal variations are im-
mended (Section 4.6). portant factors to be considered. Spatial variations

154
 Recommendations

depend on the geology of the study area and hence Table 9.1. Recommended devices for radon (222Rn) gas areal
determine the grid size of the measurement sites. measurements for indoor workplaces, mixed-use buildings, or
dwellings
Temporal variations are caused by variable weather
conditions, such as snowfall, inversion, etc. For Purpose Measurement Device Cost
example, variation in outdoor radon activity concen- type
trations observed in 10 km  10 km squares or in
towns is typically described by a median coefficient Assessment of Long-term Alpha track Low
of variation (COV) of 60% and by a geometric stand- exposure sampling detectors
ard deviation (GSD) of 1.9, with a typical variation ( 3 months) Electret Medium
ionization
in the range of 1.7 – 2.2. In national radon surveys, chambers
the GSD is typically in the range of 1.8–2.7 Determination of Short-term Continuous High
(Sections 4.4 and 7.1). ratio of average sampling radon monitorsb
working time to (1 week)
one week activity
concentrationa

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

9.3 Recommendations Regarding Post remediation Short-term Continuous High
Measurement Techniques test to test sampling radon monitors Low/
effectivenessc (1 week) Passive Medium
At present, a wide variety of measurement detectors (Alpha
methods and detectors are available for radon and track, electret,
activated
radon progeny measurements. Depending on the charcoal)
objectives of the planned study, the first decision to
be made is whether to measure radon or radon a
Annual average activity concentration during working hours
progeny. Although it is the radon progeny which can be estimated my multiplying the annual average activity
cause lung cancer, the measurement of radon activ- concentration by the ratio of the average activity concentration
during the working hours of a week to the average activity
ity concentrations is a much simpler and cheaper al-
concentration during the whole week.
ternative for exposure assessment purposes. The b
These are active monitors that have the ability to produce time-
purpose of the measurements will usually prescribe resolved measurements.
c
the techniques. Table 9.1 recommends devices for As well as long-term testing, short-term measurements (1 week)
indoor radon (222Rn) gas areal measurements. may be started at the same time.
In the case of radon monitoring, either short-term
exposure assessment is limited by the short duration
or long-term, or active or passive, or areal or person-
of the measurement and the requirement of real-
al measurement methods can be applied. In add-
time air sampling equipment (Section 5.3).
ition, the selection of an appropriate measurement
To explore the full potential of radon progeny mea-
device depends on the required sensitivity and ac-
surements, determination of radon progeny activity
curacy. Another consideration is the related cost/
concentrations should be combined with measure-
benefit relationship. For example, passive, time-
ments of the size distributions of the attached
integrating devices are recommended for long-term
(cascade impactors, diffusion batteries) and the un-
measurements of the radon activity concentration,
attached fraction (screen samplers, diffusion batter-
either for areal or personal monitoring. To avoid the
ies) (Section 5.3.3).
potential contribution of thoron to the measured
radon signal, the sensitivity of the radon monitor to
thoron must be known. In the case of suspected high 9.4 Recommendations Regarding Recording
thoron activities, it is recommended to use a separ- and Reporting of Measurements
ate thoron detector (Section 5.2).
In the case of individual monitoring, the required Key features in recording and reporting of the
time-resolution should be correlated with physical results of radon measurements depend on the specif-
activity, i.e., personal breathing rates, depending on ic purpose of the measurement.
gender and age, and diurnal occupancy with respect In the case of radon measurements aimed at the
to exposure conditions. The variation of the breath- assessment of radon levels in homes or indoor work-
ing rate is a source of uncertainty in estimates of places rather than the assessment of exposures to
“individual dose” or “individual exposure” in epi- specific individuals, such as national radon surveys,
demiological studies (Sections 6.6 and 8.3.5). the following key parameters should be recorded
In contrast to time-integrating radon monitoring, and reported.
radon progeny measurements are usually short- Measurements:
term measurements with filter devices combined
with alpha detectors. Their use for residential † Date, time, and duration of measurement

155
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

† Persons carrying out the measurement characterizing personal exposure conditions and con-
† Location of measurement and placement of comitant exposures which may contribute to lung
detectors cancer risk, such as cigarette consumption, should be
† Sampling method considered as well.
† Instrumentation used (methods, detectors)
† Uncertainty associated with the measurement, Personal exposure conditions:
which should be taken into account when setting
† Occupancy parameters
reference values to comply with the limits recom-
† Physical activities
mended by international directives or national
legislation Confounding factors:
† Detection limit and decision level
† Quality assurance for data validation † Smoking status (active, passive)
† Meteorological conditions, e.g., temperature, hu- † Additional carcinogenic factors
midity, rain, or snowfall † Occupational exposure

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Results: In the case of radon progeny measurements, the fol-
lowing key parameters should be recorded and
† Persons analyzing the data reported:
† Local and temporal distribution of results
† Application of seasonal or annual correction † Equilibrium factor
factors, taking into account their high variability † Size distribution of attached and unattached
† Statistical parameters: arithmetic mean and radon progeny (activity median diameter, geomet-
standard deviation (or coefficient of variation) ric standard deviation)
and/or geometric mean and geometric standard † Unattached fraction of the PAEC
deviation
† Percentage exceeding reference levels and identi- In summary, the above recommendations regarding
fication of houses above action thresholds measurement strategies, measurement techniques,
and recording and reporting of measurements
In the case of individual exposure assessments, such provide a framework of practical advice to those who
as radon measurements associated with epidemio- are planning, conducting and reporting radon mea-
logical lung cancer studies, additional key parameters surements for a variety of purposes.

156
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv012
Oxford University Press

Appendix A Radon and Radon Progeny Metrology and Quality

Assurance of Measurements

A.1 Metrological Traceability For the first time, the calibration of commercial
radon devices is possible in constant reference atmo-
Metrological traceability is a property of a meas-
spheres and therefore small uncertainties are
urement result whereby the result can be related to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

achieved. The relative standard uncertainty for the
a reference through a documented unbroken chain
radon activity concentration at approximately 360
of calibrations, each contributing to the measure-
Bq m23 is u(C) ¼ 1.1%.
ment uncertainty.
In the case of 220Rn, a reference atmosphere has
Traceability of the unit, for radon and its progeny
been available since 2009. This primary standard is
typically Bq m23 or J m23, means that the calibra-
based on a certified activity standard of 228Th, a cer-
tion of the device used in the measurement is linked
tified known volume, and measurement of the 220Rn
to a chain of calibrations originating at a national or
emanation factor for the activity standard, which is
international metrology laboratory. At the origin,
determined online and continuously over the entire
e.g., a national metrology institute, the unit is rea-
period of the calibration (Röttger et al., 2010). The
lized in the form of a primary method. For example, 220
Rn gas emanating from the 228Th source is deter-
a radon activity gas standard (Bq) is transferred
mined by online measurement g-ray spectrometry
without loss to a known volume (m3), thus creating a
through the disequilibrium of the 228Th activity
homogeneous reference atmosphere (Bq m23). This
(measured via 224Ra) and the 212Pb activity.
reference atmosphere is a primary standard.
A primary standard has been available for 222Rn
progeny since 1999 (Paul et al., 1999) and for 220Rn
Primary standards progeny since 2009 (Röttger et al., 2009).
In the case of 222Rn, the creation of a reference at-
mosphere based on the development of a radon gas Secondary standards
standard (Dersch and Schotzig, 1998; Picolo, 1996) A device (e.g., system under test) that is exposed
became a standard procedure worldwide (Paul et al., to a reference atmosphere can be calibrated, thus be-
2000). The use of this procedure for the calibration coming a secondary standard. If this device is used
of commercial devices is limited to activity concen- for further calibrations of other devices, for example,
trations above 1 kBq m23, because below this level, in other parts of the world, it is referred to as a
the measurement uncertainty of the commercial transfer standard.
devices is too large due to the limited statistics of
counts in their active volumes.
The CIPM MRA
For reference atmospheres below 1000 Bq m23, a
time constant has been generated to perform long- National metrology institutes work independently
term calibrations (t  5 d) of commercial devices. from each other but they cooperate as signatories of
With the development of the low-level radon refer- the Mutual Recognition Arrangement of the
ence chamber at PTB (Linzmaier, 2012), constant Committee International des Poids et Mesures
activity concentrations from 1900 Bq m23 down to (CIPM MRA), an “Arrangement on the mutual rec-
150 Bq m23 are available. For this purpose, several ognition of the equivalence of national standards
emanation sources have been manufactured. These and of calibration certificates issued by national me-
emanation sources generate a constant reference at- trology institutes.” This assures under certain condi-
mosphere in the low-level radon reference chamber. tions that a calibration certificate from one national
The emanation coefficients of the sources are mea- metrology institute is accepted to be valid by the
sured and analyzed according to the count rate ap- others.
proach of a gas-tight closed 222Rn emanating 226Ra One central point of the arrangement is a regula-
source. tion for worldwide comparison measurements in key

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

areas, so-called key comparisons, to obtain informa- Measurements for a scientific or legal purpose have
tion about the degree of equivalence of national to be in agreement with ISO/IEC 17025 (ISO, 2005).
measurement standards and calibration procedures. This standard covers all aspects of quality assurance.
Under the MRA, the International Bureau of Weights
and Measures (Bureau International des Poids et ISO 9001 and ISO/IEC 17025
Mesures, BIPM) publishes the results of the key
ISO 9001 (ISO, 2008) is a generic management
comparisons in a “key comparison database” which
standard that can be applied to any business enter-
is accessible via the internet. The database contains
prise, public administration, or government depart-
the results of comparisons between national (and
ment. Growth in the use of management systems
international) standards, expressed in terms of
generally has increased the need to ensure that la-
degrees of equivalence with respect to an agreed
boratories can operate in a quality management
reference value, as well as statements about the
system that is seen as compliant with ISO 9001
measuring capabilities of the metrology institutes and
(ISO, 2008) as well as demonstrating technical com-
the associated uncertainty budgets for calibrations.
petence. Therefore, ISO 17025 (ISO, 2005) was
These capabilities are also supported by supplemen-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

written to incorporate all the ISO 9001 (ISO, 2008)
tary comparisons, which extend the reach of the key
requirements that are relevant to the scope of testing
comparisons. This structure of key and supplemen-
and calibration services as well as specifying the
tary comparisons is organized through the CIPM and
technical requirements for technical competence.
its laboratory, the BIPM, as well as through a series of
Testing and calibration laboratories that comply
regional metrology organizations (ROMs). The CIPM
with ISO 17025 (ISO, 2005) will also operate in ac-
MRA also requires that all national metrology insti-
cordance with ISO 9001 (ISO, 2008).
tutes provide evidence of a suitable quality system,
The ISO 17025 (ISO, 2005) standard itself com-
either by self-declaration and peer review or by certifi-
prises five elements: scope, normative references,
cation by a third party (ISO/IEC 17025, 2005).
terms and definitions, management requirements,
Entries for “radon” can be found at http://kcdb.
and technical requirements. The last two elements
bipm.org/.
contain the actual accreditation requirements.
Accreditation requirements are divided into manage-
ment requirements (quality system, document control,
A.2 Quality Assurance
review of requests, tenders and contracts, subcontract-
Definition and Purpose ing of tests and calibrations, purchasing services and
supplies, service to client, complaints, control of non-
Quality assurance (QA) includes all planned and
conforming testing and/or calibration work, corrective
systematic actions that are necessary to provide ad-
action, preventive action, control of records, internal
equate confidence in the accuracy of measurements.
audits, management reviews), and technical require-
A quality assurance program encompasses require-
ments (general, personnel, accommodation and envir-
ments on organization and management of the
onmental conditions, test and calibration methods
service and requirements on technical equipment
and method validation, equipment, measurement
and provisions for carrying out measurements.
traceability, sampling, handling of test and calibra-
Quality assurance should include validation of
tion items, assuring the quality of test and calibra-
methods and verification of results which in turn
tion results, reporting the results).
involves all the actions by which the adequacy of
equipment, instruments, and procedures are assessed
Quality Control
against specified requirements. It should ensure that
equipment and instruments function correctly, proce- The radon service should maintain an appropriate
dures are correctly established and followed, quantifi- monitoring system to prove that specified require-
able errors are within acceptable limits, and records ments are met and processes are under control. The
are correctly and promptly maintained. The quality modality of recording relevant process parameters
assurance program and the regular checks made for shall ensure the repeatability of measurements.
quality control shall be fully documented. From an analysis of the measurement process, the
General requirements for the competence of la- relevant parameters should be deduced and moni-
boratories are set out by the international standard tored by control charts to reveal early trends and to
ISO/IEC 17025 (ISO, 2005) which should be adopted prevent malfunctions of procedures and equipment. A
by radon services carrying out radon measurements. QA Plan shall specify all activities, measurements,
The competence of a radon service can be formally and documentations to be carried out. The activities
recognized by an authoritative body working under can encompass regular calibrations, cross-checks of
a national accreditation scheme. measurement results (interlaboratory comparison

158
 Appendix

exercises), duplicates or collocated measurements, are used to calculate the calibration factor kt for the
laboratory and field background measurements. reading of the item being calibrated.
The provision of objective evidences that specified
requirements have been fulfilled, is an essential pre- C
kt ¼ ðA:1Þ
requisite to verify processes and to accomplish confi- Ct;tc  Ct;bg
dence in results.
An example of a resulting uncertainty budget is
given in Table A.1.
Validation of Methods
The results for the ratio of the activity concentra-
Validation is the confirmation that requirements tions in the reference volume and the mean values
for a specific intended use have been fulfilled. of the readings of the calibration item is kt ¼
Standard validation procedures are type tests of (0.96 + 0.4) for coverage factor k ¼ 2 (see Section
instruments, comparisons, and calibrations, and 8.2.1 for the definition of the coverage factor). In the
should be as extensive as necessary to meet the respective range of the activity concentration, the
needs of a given application. reading of the calibration item must be multiplied

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

by the calibration factor kt after subtraction of the
Type Test of Radon Instruments background. The uncertainty stated is the expanded
measurement uncertainty obtained by multiplying
By a type test, one or more radon instruments rep- the standard measurement uncertainty by the
resentative of the manufacturing production are coverage factor k ¼ 2, which has been determined in
checked for compliance with requirements specific accordance with the “Guide to the Expression of
to the intended use. It covers mechanical, electrical, Uncertainty in Measurement (ISO, 1995).” The value
and radiological examinations which can be carried of the measurand then normally lies, with a probabil-
out by the manufacturer or by an approved testing ity of 95%, within the attributed coverage interval.
laboratory. Specific requirements for radon measur-
ing instruments are laid down in the standard series
IEC 61577(IEC, 2000; 2006; 2009; 2011).
A.4 Example of the Analysis of Uncertainties
in an Interlaboratory Comparison

A.3 Example of the Analysis of Uncertainties Passive radon devices consisting of a diffusion
in a Calibration by a Primary Radon Activity chamber with a solid-state detector inside are com-
Standard monly used for long-term measurements to deter-
mine the average 222Rn activity concentration. In
For the measurements, the calibration item order to estimate the measurement uncertainty of
(system under test) is placed into the radon refer- this method, an analysis of the results of the annual
ence chamber of known volume Vr. Then the calibra- intercomparisons conducted by the German Federal
tion item is exposed to radon inside the closed Office for Radiation Protection (BfS) has been
chamber. The radon activity concentration C in the undertaken (Beck et al., 2005; 2007; 2009). About 20
chamber results from the activity A0 of the radon national and international radon services have par-
gas standard at a reference date given in the ticipated in the intercomparison each year since
certificate, the volume of the chamber Vr and the dis- 2003. Figure A.1 shows a summary of the results.
placed volume of the measurement item Vt to The box plots illustrate the distribution of the single
C ¼ A0/(Vr 2Vt). The readings of the calibration item
are recorded over a period of 1 – 3 d, at given inter-
vals. During this time, the radon activity concentra- Table A.1. A reduced uncertainty budget according to the “Guide
tions in the chamber decreases as a result of to the Expression of Uncertainty in Measurement” (ISO, 1995) for
radioactive decay i.e. C ¼ A0 exp (2ltc). a calibration of a 222Rn measuring device in a reference
The instrument background Ct,bg of the calibra- atmosphere (primary standard)
tion item caused by contamination of the measure-
Quantity Value Standard uncertainty Index
ment volume, in particular with 210Pb and its decay
products, are determined over a period of 24 h by C 959.8 Bq m23 16.7 Bq m23
means of a measurement in radon-free air, and A0 60.00 Bq 1.00 Bq 68.9%
taken into account for the calibration. Vr 2Vt 0.048600 m3 243.1026 m3 6.2%
The readings of the calibration Ct,tc item are aver- Ct,tc 1000.0 Bq m23 10.0 Bq m23 24.8%
Ct,bg 10.00 Bq m23 2.00 Bq m23 0.0%
aged over the respective interval. The mean value is
kt 0.960 0.0193
corrected by the respective background. These data

159
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

quality level have been involved in the analysis.

Measurement services whose average results deviate
by more than 10% from the reference level were
excluded.
From Figure A.1, the nominal measurement un-
certainty of a single 222Rn measurement using
radon diffusion chambers with solid-state detectors
can be deduced. This nominal measurement uncer-
tainty represents the characteristic uncertainty of
the measurement method irrespective of the individ-
ual uncertainty of a particular radon service. From
this figure, it is estimated that within a range from
500 to 1000 kBq h m23, the expanded uncertainty
of the method is approximately 30% (expanded
Figure A.1. Summary of an interlaboratory comparison for uncertainty: value of the measurand that normally

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

instruments with solid state nuclear track detectors at BfS since lies within this interval with a probability of 95%).
2003, box plots indicate the distribution of the single For exposures to 222Rn above 2000 kBq h m23, the
measurement results of the participating radon service for a expanded uncertainty is somewhat lower (about 20%)
given reference exposure to radon (boundary of the boxes
and below 300 kBq h m23 higher. All uncertainties
indicates the 25th and the 75th percentile, a line within the boxes
marks the median, whiskers indicate the 10th and the 90th relate to the measurement uncertainty (IEC, 2015) of
percentiles, and points indicate the 5th and 95th percentile). a single 222Rn measurement with a statistical safety
of 95%. The standard uncertainty in the exposure
range from 500 to 1000 kBq h m23, which corresponds
results around the corresponding reference value. to an annual average 222Rn activity concentration of
Only radon services which could prove an acceptable about 100 Bq m23, is thus approximately 15%.

160
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv013
Oxford University Press

Appendix B Analysis of Results of the Nuclear Track Detector

Exposure at PTB in View of Cross Sensitivity to Radon/Thoron and
the Determination of Decision Threshold and Detection Limit

D. Schrammel, KIT ( private communication) If these thoron detectors are exposed to 1000 kBq
m23 h of thoron, and 2000 kBq m23 h of radon,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

This appendix describes a computational method
the number of tracks will be: wT0 1000 þ
for the analysis of the results of the exposure of
wT2  2000 þ 71:28 ¼ 6797:
nuclear track detectors at PTB to a mixture of radon
and thoron activity concentrations. Combining the detectors (that is using always a
Both types of detectors (radon and thoron) were radon and a thoron detector together), the individ-
exposed to a series of 220Rn exposures (102, 500, 1500, ual exposure to 222Rn and 220Rn can be determined.
3000 kBq m23 h), some of these detectors were add- To cover this, an uncertainty for single track density
itionally exposed to one 222Rn level (1250 kBq m23 h). is calculated leading to
An additional background is assumed for all detec-
tors: nR0 (for the radon detectors) and nT0 (for the 32 2 3 2 3
mR mR0 1:10E2
thoron detectors). 6 7 6 7 6 7
½uðni Þ ¼ A 4 mT 5 with 4 mR2 5 ¼ 4 3:27E2 5
By using the formula
b bR 8:34
Ay¼x ðB:1Þ 2 3 2 3
mT0 2:75E1
6 7 6 7
with A being the exposure matrix, x the vector of or 4 mT2 5 ¼ 4 4:28E2 5 ðB:4Þ
result (mean value of the number of tracks of the re- bT 5:42
spective group of nuclear track detectors), the result
vector y can be calculated (Tables B.1 and B.2).
while the uncertainty of a single number of tracks is
The values in the vector y represent the result of
supposed to be covered by Equation (B.5) in which
solving Equation (B.1):
nÞ is the uncertainty of a group of m nuclear track
uð
T 1 1 detectors:
y ¼ Uy AT U1
x x with Uy ¼ ðA Ux AÞ ðB:2Þ
pffiffiffiffiffi
Since the mathematical operations are time- uðni Þ ¼ m uðn  Þ: ðB:5Þ
consuming, the use of a qualified software tool is
advisable: The overall uncertainty of a measurement is given
2 3 2 3 2 3 2 3 by the matrix Un covering the respective sum of
wR0 0:100 uðwR0 Þ 0:0034 uncertainties: background, results according to
6 7 6 7 6 7 6 7
yR ¼ 4wR25 ¼ 4 3:23 5 with 4uðwR2 Þ5 ¼ 4 0:017 5ðB:3Þ Equations (B.2) and (B.3), and the single track
nR0 52:85 uðnR0 Þ 2:14 density according to Equation (B.5).
2 3 2 3 2 3 2 3 Using these results, an unknown exposure
wT0 0:135 uðwT0 Þ 0:0046 E220
can be determined by solving Equation (B.6):
yT ¼ 4 wT2 5 ¼ 4 2:69 5 with 4 uðwT2 Þ 5 ¼ 4 0:042 5: E222
nT0 71:28 uðnT0 Þ 5:08
" # " # " # " #
nR wR0 wR2 E220 nR0
Example 1 ¼ þ ;
nT wT0 wT2 E222 nT0
If these radon detectors are exposed to 1000 kBq
" # " #" # " #
m23 h of thoron, and 2000 kBq m23 h of radon, nR 0:10 3:23 E220 52:85
the number of tracks will be: wR0 1000 þ ¼ þ ; ðB:6Þ
wR2 2000 þ 52:85 ¼ 6617: nT 1; 35 2:69 E222 71:28

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table B.1. Exposure matrix and the respective matrix of variances according to Equation (B1). The thoron detectors at the exposure level
of 1500 kBq m23 h (thoron) and at 1250 kBq m23 h (radon) did not pass the linearity check and were therefore dismissed

Exposure matrix Matrix of variances

A
u2 ð
ni Þ 0
E(Rn-220) E(Rn-222) Ebg Ux ¼
0 ...

Radon detectors

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Thoron detectors

This yields: Table B.2. Analysis of tracks according to the results of

calibration in the PTB’s reference chambers. The corresponding
2 31 matrices were calculated according to Equations (B.2 –B.8)

E220 6w 7 Quantity Value (kBq m23 h) Standard uncertainty of

6 R0 wR2 7 nR nR0
the value (kBq m23 h)
¼6 7  ;
E222 4 wT0 wT2 5 nT nT0
|fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflffl} 220
Q Rn exposure 571.8 145.4
222
Rn exposure 430.1 10.7

0:696 0:789
Q¼ ; ðB:7Þ
0:330 0:0243
It is obvious from this result that the cross-
correlation of the response of nuclear track detectors
The uncertainty of the exposure is the square root of toward 222Rn or 220Rn, respectively, has to be consid-
the diagonal elements of the matrix UE given by ered with great care.
For practical use of the combined system, the de-
termination of the decision threshold and detection
U E ¼ Q U n QT : ðB:8Þ
limit is needed. According to ISO 11929 (ISO, 2010),
this can be done using the following equation:
The determination of U n has to be performed care-
fully (sum of all uncertainties, see statements ~ ¼ 0Þ
~ ðE
decision threshold: E ¼ k1a u ðB:9Þ
above). #
detection limit: E ¼ E þ k1b ~ ¼ E# Þ ðB:10Þ
~ ðE
u
Example 2
A set of detectors is exposed and the tracks ana- The decision threshold and detection limit are
lyzed to be: defined by the respective uncertainties for special
exposure conditions. Calculating the limit for one
number density in the thoron monitor: 2000 cm22 exposure (radon/thoron), the corresponding other
number density in the radon monitor: 1500 cm22 exposure value (thoron/radon) is kept constant. The

162
 Appendix B

Table B.3. Decision threshold and detection limit determined ½ERn-220 ; ẼRn-222 ¼ 0; 1 and for the detection limit
according to ISO 11929 for a pair of nuclear track detectors with ½ERn-220 ; ẼRn-222 ¼ E# Rn-222 ; 1.
Solving Equations (B.9) and (B.10) can be tedious,
Quantity Decision threshold Detection limit
(kBq m23 h) (kBq m23 h) so an iterative solution might be chosen instead.
With the probability of the error of first and
Exposure to 220Rn 47.4 133.6 second order a ¼ b ¼ 5%, the results are given in
Exposure to 222Rn 10.4 21.1 Table B.3.
This example shows the importance of well-
defined calibration and exposure conditions.
Although the analysis of data is time-consuming,
uncertainties are calculated in the same way as for the final result drastically improves the quality of
actual exposures. results obtained in all subsequent measurements.
The exposure to 220Rn is represented for the decision All users of radon monitors, passive or active,
threshold with the exposure vector ½ẼRn-220 ¼ 0; should be aware of the problem of cross-sensitivity
ERn-222 ; 1 and for the detection limit with to 220Rn or 222Rn, respectively. Since this effect does

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

½ẼRn-220 ¼ E#
Rn-220 ; ERn-222 ; 1. influence the uncertainty of the measurement, it is
The exposure to 222Rn is represented for the fundamental for the determination of the decision
decision threshold with the exposure vector threshold and detection limit as well.

163
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv014
Oxford University Press

Appendix C Measurement Method using Solid-State Nuclear Track

Detectors and the Expression of Results

In the following example, a number of soild-state where Cr is the measured activity concentration and
nuclear track detectors were calibrated in the radon Crbg is the specific background value obtained with
reference chamber of the Physikalisch-Technische the reference instrument and kr is the calibration

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Bundesanstalt (PTB), Germany. For the reference ex- factor for the instrument. The uncertainty budget
posure, each nuclear track detector and its housing for exposure P is given in Table C.1 and has been cal-
positioned in a specific geometry were brought into culated according to ISO (1995).
the radon reference chamber, together with the active During this exposure, a number m of nuclear
radon monitor calibrated as reference standard, and track detectors were exposed. The variation in the
exposed to 222Rn. The irradiations extended over a track density was used to determine the uncertainty
period of several hours to days. The exposures are cal- nÞ of the mean track density n
uð . This is a simplified
culated from the mean values of the measured values example, for a special batch of track-etch detectors
of the reference standard and the irradiation time. only.
They were corrected for background effects. The radon P ni
activity concentration was recorded in the radon refer- u2 ðni Þ
¼P
n ; ðC:2Þ
ence chamber at intervals of 10 min. A number of the 1
u2 ðni Þ
detectors were not exposed and were used to deter-
mine the background effects for each device. where i is the index of a given track-etch detector.
Tables C.1–C.3 give the results of the measure- A statistical consistency check is carried out as
ments, calculated quantities, and their associated un- follows:
certainties for the determination of a calibration
coefficient starting with a traceable reference atmos- X ðni  n
Þ
M¼ ; if M . x20:95;m1 then
phere. Table C.4 gives an example of how this calibra- u2 ðni Þ
tion information can be used in field measurements. rffiffiffiffiffiffiffiffiffiffiffiffiffi ðC:3Þ
M
f ¼ ; else f ¼ 1
m1
C.1 Realization of the Unit: Determination where f takes account of variation other than
of a Calibration Coefficient for Nuclear counting statistics. uncertainty, uð nÞ, is then calcu-
Track Detectors lated as:
A simple example to illustrate the general principle sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
for using the ISO (1995) procedures for the calibration 1
nÞ ¼ f P 1 :
uð ðC:4Þ
of nuclear track detectors is presented in the following u2 ðni Þ
uncertainty budget. The calibration conditions are well
known in terms of the exposure period, Dt ¼ (t2  t1 ) To correlate the track density to the exposure,
and radon activity concentration CRn-222 : This calibra- several exposures have to be performed and ana-
tion is performed in the radon reference chamber start- lyzed. However, for simplicity, only one exposure is
ing at time t1 and ending at t2 . considered in this example (Table C.2). Also for il-
The exposure is determined with a secondary stand- lustrative purposes, f ¼ 1 for the unexposed detec-
ard (i.e., with reference instrument) that has been cali- tors, while for the exposed detectors f . 1,
brated using a primary standard (Sections 5.1 and A.1). indicating that the variation in track densities is
The exposure P is given by larger than expected.
If a group of m nuclear detectors is exposed to dif-
P ¼ CRn-222 ðt2  t1 Þ with ferent radon levels, a linearization can be performed
  ðC:1Þ
CRn-222 ¼ kr Cr  Crbg and if the linear model passes the consistency check,

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENT AND REPORTING OF RADON EXPOSURES

Table C.1. Uncertainty budget for the exposure P according to Table C.3. Example of the analysis of an exposure of the nuclear
(ISO, 1995) based on Equation (C.1) track detectors at 1500 kBq m23 h and 12 detectors without
exposure (background detectors) to obtain a calibration coefficient
Quantity Value Standard uncertainty Indexa k, according to Equation (C.5)

CRn-222 30.7 kBq m23 0.5 kBq m23 Quantity Value Standard uncertainty Indexa
kr 1.031 0.014 89.7%
Cr 29.8 kBq m23 0.13 Bq m23 9.9% P 1500 kBq m23 h 22 kBq m23 h 3.2%
Crbg 59 Bq m23 3 Bq m23 0.0% n
 2030 cm22 160 cm22 96.5%
t2 49.00 h 0.04 h 0.2% n
 bg 69 cm22 6 cm22 0.3%
t1 0.0 h 0.04 h 0.2% k 0.77 kBq m23 h cm2 0.06 kBq m23 h cm2
P 1500 kBq h m23 22 kBq h m23
a
The index gives the amount of influence of a single uncertainty
a
The index gives the amount of influence of a single uncertainty to the combined uncertainty.
to the combined uncertainty.

Table C.4. Example of the analysis of nuclear track detectors to

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Table C.2. Example of the analysis of the track density for an obtain an exposure according to Equation (B.6)
exposure of 10 nuclear track detectors at 1500 kBq m23 h and the
track density for 12 detectors without exposure (background Quantity Value Standard uncertainty Indexa
detectors)
k 0.77 kBq m23 h cm2 0.06 kBq m23 h cm2 56.8%
P (kBq m23 h) m n
 (cm22) M x20:95;m1 f nÞ (cm22)
uð n
 3290 cm22 220 cm22 43.2%
n
 bg 50 cm22 5 cm22 0.0%
0 12 68.94 8.46 19.68 1 5.57 P 2500 kBq m23 h 260 kBq m23 h
1500 10 2026.78 477.17 16.92 7.28 155.77 Dt 2000 h 14 h 0.4%

C 1.25 kBq m23 0.13 kBq m23

a
The index gives the amount of influence ofa single uncertainty to
the calibration coefficient, k, can be calculated as the combined uncertainty.
follows:

P  ¼ (1.25 + 0.26) kBq m23 (with a coverage

C
k¼ : ðC:5Þ factor k ¼ 2).
ð
nn bg Þ
These results show the expanded measurement un-
The calculated value of k and its associated uncer- certainty obtained by multiplying the standard meas-
tainty is given in Table C.3 for this example. urement uncertainty by the coverage factor k ¼ 2. It
is determined in accordance with the “Guide to the
Expression of Uncertainty in Measurement (GUM).”
C.2 Assigning Uncertainty to a Measurand The value of the measurand then normally lies, with
in a Field Measurement a probability of 95%, within the attributed coverage
interval.
Following from Equation (C.5), an unknown ex- This method is strictly only valid, if the condi-
posure to a radon activity concentration (mean tions and constraints set up by all equations for
 can be obtained by:
value C) calibration are valid for the field measurements as
P ¼ ð
nn
 bg Þ k: ðC:6Þ well. A good example, where this may not be the
case, is the presence of thoron (220Rn) in field mea-
surements. The thoron activity concentration for
 ¼ ð
C
nn
 bg Þ k
¼
P
ðC:7Þ
Dt Dt the above example is assumed to be negligible in
the calibration procedure and in the field measure-
Applying the law of propagation of uncertainty (ISO, ment, or the nuclear track detectors are assumed to

1995) to Equation (C.7), the uncertainty budget for C be insensitive to thoron. However, this cannot
can be determined (Table C.4). always be assumed. For example, all systematic
The determined exposure (of a field measurement) investigations at the PTB thoron chamber showed a
is expressed in the form significant effect of the presence of 220Rn activity
P ¼ (2.5 + 0.6)  103 kBq m23 h (with a coverage concentration on 222Rn measuring devices particu-
factor k ¼ 2) larly for open detectors. For closed detectors, where
and the mean value for the radon activity concen- the nuclear track detector is mounted in a closed
tration during the time of measurement (2000 h) is container, the sensitivity to thoron (220Rn) is less
given by (Section 7.4).

166
 Appendix C

The decision threshold and detection limit can be  #:

and the detection limit C
calculated according to ISO 11929 (ISO, 2010). The
decision threshold and detection limit are defined by # ¼ C
C  þ k1b u e  #Þ
 ¼C
~ ðC ðC:9Þ
the respective uncertainties for special exposure con-
ditions. With the probability of the error of first- and In the given example C  ¼ 6 Bq m23 and C  # ¼ 15
second-order, typically a ¼ b ¼ 5%, the results can be Bq m23 is determined. Detailed and simple exam-
expressed by the decision threshold C : ples for the determination of the uncertainties for
different detector types, their decision threshold,
and detection limit can be found in ISO 11665-4
 ¼ k1a u
C e
 ¼ 0Þ
~ ðC ðC:8Þ (ISO, 2012c).

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

167
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Journal of the ICRU Vol 12 No 2 (2012) Report 88 doi:10.1093/jicru/ndv016
Oxford University Press

References

Abu-Jarad, F., and Fremlin, J. H. (1982). “The activity of Andersen, C. E., Ulbak, K., Damkjaer, A., Kirkegaard, P.,
radon daughters in high-rise buildings,” Health Phys. and Gravesen, P. (2001). “Mapping indoor radon-222 in
43, 75–80. Denmark: design and test of the statistical model used
Abu-Jarad, F., Wilson, C. K., and Fremlin, J. H. (1981). in the second nationwide survey,” Sci. Total Environ.
“The registration of the alpha-particles from polonium 272, 231– 241.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

isotopes plated-out on the surface of the plastic detec- Andreae, M. O. (2013). “The aerosol nucleation puzzle,”
tors LR-115 and CR-39,” Nucl. Tracks 5, 285 –290. Science 339, 911 –912.
Abummurad, K. M., and Al-Tamimi, M. H. (2005) Anjos, R. M., Umisedo, N., da Silva, A. A. R., Estellita, L.,
“Natural radioactivity due to radon in Soum region, Rizzotto, M., Yoshimura, E. M., Velasco, H., and Santos,
Jordan,” Radiat. Meas. 29, 77–80. A. M. A. (2010). “Occupational exposure to radon and
Alavanja, M. C. R., Lubin, J. H., Mahaffey, J. A., and natural gamma radiation in the La Carolina, a former
Brownson, R. C. (1999). “Residential radon exposure gold mine in San Luis Province, Argentina,” J. Environ.
and the risk of lung cancer in Missouri,” Am. J. Public Radioact. 101, 153 –158.
Health 89, 1042–1048. Annanmäki, M. K., Oksanen, E., and Markkanen, M.
Al-Azmi, D., Al-Abed, T., Alnasari, M. S., Borham, E. E., (1996). “Radon at workplaces other than mines and
Chekir, Z., Khalifa, M. S., and Shweikani, R. (2012). underground excavations,” Environ. Int. 22(Suppl. 1),
“Coordinated indoor radon surveys in some Arab coun- 769– 772.
tries,” Radioprotection 47, 205 –217. Appleton, J. D., Miles, J. C. H., Green, B. M. R., and
Alberigi, S., Pecequilo, B. R. S., Lobo, H. A. S., and Larmour, R. (2008). “Pilot study of the application of
Campos, M. P. (2011). “Assessment of effective doses Tellus airborne radiometric and soil geochemical data for
from radon levels for tour guides at several galleries of radon mapping,” J. Environ. Radioact. 99, 1687–1697.
Santana Cave, Southern Brazil, with CR-39 detectors: Arvela, H. (1995). “Seasonal variation in radon concentra-
preliminary results,” Radiat. Prot. Dosim. 145, tion of 3000 dwellings with model comparisons,”
252– 255. Radiat. Prot. Dosim. 59, 33–42.
Albert, R. E., and Lippmann, M. (1971). “Bronchial clear- Arvela, H., Voutilainen, A., Mäkeläinen, I., Castrén, O.,
ance abnormalities in man produced by cigarette smoke and Winqvist, K. (1988). “Comparison of predicted and
and lung disease,” Chest 59(5 Suppl.), 3S. measured variations of indoor radon concentration,”
Albert, R. E., Lippmann, M., Peterson, H.T., Jr., Berger, Radiat. Prot. Dosim. 24, 231– 235.
J., Sanborn, K., and Bohning, D. (1973). “Bronchial de- Arvela, H., Voutilainen, A., Honkamaa, T., and Rosenberg,
position and clearance of aerosols,” Arch. Intern. Med. A. (1994). “High indoor radon variations and the
131, 115–127. thermal behavior of eskers,” Health Phys. 67, 254– 260.
Allen, J. E., Henshaw, D. L., Keitch, P. A., Fews, A. P., and Arvela, H., Holmgren, O., and Reisbacka, H. (2012).
Eatough, J. P. (1995). “Fat cells in red bone marrow of “Radon prevention in new construction in Finland: a na-
human rib: their size and spatial distribution with tionwide sample survey in 2009,” Radiat. Prot. Dosim.
respect to radon-derived dose to haemopietic tissue,” 148, 465– 474.
Int. J. Radiat. Biol. 68, 669 –678. Arvela, H., Holmgren, O., Reisbacka, H., and Vinha, J.
Al-Mustafa, H., Al-Jarallah, M. I., Fazal-ur-Rehman, and (2014). “Review of low-energy construction, air-tightness,
Abu-Jarad, F. (2005). “Radon concentration measure- ventilation strategies and indoor radon: results from
ments in the desert caves of Saudi Arabia,” Radiat. Finnish houses and apartments,” Radiat. Prot. Dosim.
Meas. 40, 707 –710. 162, 351–363.
Alter, W. H., and Oswald, R. A. (1983). “Results of indoor Arvela, H., Holmgren, O., and Hänninen, P. (2015). “Effect
radon measurements using the Track Etchw method,” of soil moisture on seasonal variations in indoor radon
Health Phys. 45, 425 –428. concentration: modelling and measurements in 326
Alter, W. H., and Oswald, R. A. (1987). “Nationwide distri- Finnish houses,” Radiat. Prot. Dosim. (doi:10.1093/rpd/
bution of indoor radon measurements: a preliminary ncv182).
data base,” J. Air Pollut. Control Assoc. 37, 227 –231. Asgharian, B., Hofmann, W., and Miller, F. J. (2001).
Andersen, C. E. (2001). “Numerical modelling of “Mucociliary clearance of insoluble particles from the
radon-222 entry into houses: an outline of techniques tracheobronchial airways of the human lung,” J. Aerosol
and results,” Sci. Total Environ. 272, 33–42. Sci. 32, 817–832.

# Crown copyright 2015. Reproduced with the permission of the Controller of Her Majesty’s Stationery Office/Queen’s Printer for
Scotland and Public Health England.
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Aurand, K., Jacobi, W., and Schraub, A. (1956). Beck, T. R. (2012). Personal communication. Federal
Untersuchungen über die Folgeprodukte des Radons im Office for Radiation Protection, BfS Germany.
Gasteiner Thermalwasser (Springer, Komm). Beck, T. R., Schwedt, J., and Hamel, P. (2005). “Quality as-
Aytekin, H., Baldık, R., Çelebi, N., Ataksor, B., Taşdelen, surance of individual radon measurements,” pp. 731–739
M., and Kopuz, G. (2006). “Radon measurements in the in Natural Radiation Enviroment VII, McLaughlin, J. P.,
caves of Zonguldak (Turkey),” Radiat. Prot. Dosim. 118, Simopoulos, S. E., and Steinhäusler, F., Eds. (Elsevier
117– 121. Science, Oxford).
Azimi-Garakani, D., Flores, B., Piermattei, S., Susanna, Beck, T. R., Buchröder, H., Foerster, E., and Schmidt, V.
A. F., Seidel, J. L., Tommasino, L., and Torri, G. (1988). (2007). “Interlaboratory comparison for passive radon
“Radon gas sampler for indoor and soil measurements measuring devices at BfS,” Radiat. Prot. Dosim. 125,
and its applications,” Radiat. Prot. Dosim. 24, 269 –272. 572– 575.
Baias, P. F., Hofmann, W., Winkler-Heil, R., Cosma, C., and Beck, T. R., Buchröder, H., and Schmidt, V. (2009).
Duliu, O. G. (2010). “Lung dosimetry for inhaled radon “Performance tests for instruments measuring radon
progeny in smokers,” Radiat. Prot. Dosim. 138, 111–118. activity concentration,” Appl. Radiat. Isot. 67, 876– 880.
Bailey, M., Ansoborlo, E., Etherington, G., Gregoratto, D., Becker, H. K., Reineking, A., Scheibel, H. G., and
Guilmette, R., Marsh, J., Paquet, F., and Smith, J. Porstendörfer, J. (1984). “Radon daughter activity size

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(2009). “Proposed updating of the ICRP Human distributions,” Radiat. Prot. Dosim. 7, 147 –150.
Respiratory Tract Model,” in Proceedings of the 12th Berkson, J. (1950). “Are there two regressions?,” J. Am.
International Congress of the International Radiation Stat. Assoc. 45, 164 –180.
Protection Association (IRPA12). BfS (2006). Bundesamt für Strahlenschutz Bericht
Baixeras, C., Climent, H., Font, L. L., Bacmeister, G. U., über das vom Bundesamt für Strahlenschutz fachlich
Albarracin, D., and Monnin, M. M. (1997). “Using und administrativ begleitete Forschungsprogramm
SSNTDs in soil and indoors in two mediterranian loca- Strahlenschutz des Bundesministeriums für Umwelt,
tions for radon concentration measurements,” Radiat. Naturschutz und Reaktorsicherheit. BfS-SCHR-40
Meas. 28, 713 –716. Strahlenschutzforschung—Programmreport 2005.
Baldwin, F. (1994). “Basal cells in human bronchial epithe- BfS (2007). Bundesamt für Strahlenschutz. Bericht über das
lium,” Anat. Rec. 238, 360– 367. vom Bundesamt für Strahlenschutz fachlich und adminis-
Baldwin, F., Hovey, A., McEwen, T., O’Connor, R., Unruh, H., trativ begleitete Forschungsprogramm Strahlenschutz des
and Bowden, D. H. (1991). “Surface to nuclear distances Bundesministeriums für Umwelt, Naturschutz und
in human bronchial epithelium: relationships to penetra- Reaktorsicherheit. BfS-SCHR-44 Strahlenschutzforschung
tion by Rn daughters,” Health Phys. 60, 155–162. – Programmreport 2006. ISBN 978-3-86509-766-8.
Bale, W. F. (1980). “Memorandum to the files, March 14, Bi, L., Li, W. B., Tschiersch, J., and Li, J. L. (2010). “Age
1951: hazards associated with radon and thoron,” and sex dependent inhalation doses to members of the
Health Phys. 38, 1061– 1066. public from indoor thoron progeny,” J. Radiol. Prot. 30,
Ball, T. K., and Miles, J. C. H. (1993). “Geological and geo- 639– 658.
chemical factors affecting the radon concentration in Billon, S., Morin, A., Caer, S., Baysson, H., Gambard, J. P.,
homes in Cornwall and Devon, UK,” Environ. Geochem. Backe, J. C., Rannou, A., Tirmarche, M., and Laurier, D.
Health 15, 27–36. (2005). “French population exposure to radon, terrestrial
Barnet, I., Mikšová, J., and Fojtı́ková, I. (2005). “Indoor— gamma and cosmic rays,” Radiat. Prot. Dosim. 113, 314–320.
soil gas relationship in the Central Bohemian Plutonic Be, M.-M., Chiste, V., Dulieu, C., Browne, E., Chechev, V.,
Complex,” Ann. Geophys. 48, 93–99. Kuzmenko, N., Helmer, R., Nichols, A., Schonfeld, E., and
Barros, N. G., Steck, D. J., and Field, R. W. (2014). “A com- Dersch, R. (2004). Table of Radionuclides, Monographie
parison of winter short-term and annual average radon BIPM-5, http://www.bipm.org/utils/common/pdf/monographie
measurements in basements of a radon-prone region RI/Monographie_BIPM-5_Tables_Vol1.pdf and http://www.
and evaluation of further radon testing indicators,” bipm.org/utils/common/pdf/monographieRI/Monographie_
Health Phys. 106, 535 –544. BIPM-5_Tables_Vol2.pdf (accessed) (Bureau International
Bartlett, D. T., Gilvin, P. J., Dixon, D. W., Solanki, H. L., des Poids et Mesures, Sevres, France).
and Miles, J. C. H. (1986a). “The performance of the Birchall, A., and James, A. C. (1994). “Uncertainty ana-
NRPB Radon Personal Dosemeter,” Radiat. Prot. lysis of the effective dose per unit exposure to radon
Dosim. 17, 139 –142. progeny and implications for ICRP risk-weighting
Bartlett, D. T., Browne, C. W., Gilvin, P. J., Dixon, D. W., factors,” Radiat. Prot. Dosim. 53, 133 –140.
and Miles, J. C. H. (1986b). “A radon personal dose- BMU (2009). Bundesministerium für Umwelt, Naturschutz
meter for miners,” Int. J. Rad. Appl. Instrum. D Nucl. und Reaktorsicherheit. Generation and Characterization
Tracks Radiat. Meas. 12, 721– 724. of Reference Atmospheres of Thoron Decay Products for
Bateman, H. (1910). Cambridge Philosophical Society the Calibration of Measuring Devices for Thoron Decay
Proceedings 15. Products, Final BMU-Report-4453.
Baysson, H., Billon, S., Laurier, D., Rogel, A., and Bochicchio, F., Campos-Venuti, G., Nuccetelli, C.,
Tirmarche, M. (2003). “Seasonal correction factors for Piermattei, S., Risica, S., Tommasino, L., and Torri, G.
estimating radon exposure in dwellings in France,” (1996). “Results of the representative Italian national
Radiat. Prot. Dosim. 104, 245– 252. survey on radon indoors,” Health Phys. 71, 741 –748.

170
 References

Bochicchio, F., Campos-Venuti, G., Piermattei, S., progeny from respiratory tract to blood,” Radiat. Prot.
Nuccetelli, C., Risica, S., Tommasino, L., Torri, G., Dosim. 102, 343–348.
Magnoni, M., Agnesod, G., Sgorbati, G., Bonomi, M., Buzinny, M., Sakhno, V., and Romanchenko, M. (2009).
Minach, L., Trotti, F., Malisan, M. R., Maggiolo, S., “LSC-based approach for radon in soil gas measure-
Gaidolfi, L., Giannardi, C., Rongoni, A., Lombardi, M., ment,” pp. 7– 11 in LSC 2008, Proceedings of an
Cherubini, G., D’Ostilio, S., Cristofaro, C., Pugliese, M., International Conference on Advances in Liquid
Martucci, V., Crispino, A., Cuzzocrea, P., Sansone Scintillation Spectrometry, Eikenberg, J., Jagi, M., and
Santamaria, A., and Cappai, M. (2005). “Annual Beer, H., Eds. (Radiocarbon, Tucson).
average and seasonal variations of residential radon Canoba, A. C., and Lopez, F. O. (2000). “Measurement of
concentration for all the Italian Regions,” Radiat. Meas. the unattached fraction of 222Rn progeny using wire
40, 686– 694. screens,” J. Radioanal. Nucl. Chem. 245, 539 –544.
Booker, D. V., Chamberlain, A. C., Newton, D., and Stott, Canoba, A. C., López, F. O., and Oliveira, A. A. (1999).
A. N. B. (1969). “Uptake of radioactive lead following in- “Radon determination by activated charcoal adsorption
halation and injection,” Br. J. Radiol. 42, 457 –466. and liquid scintillation measurement,” J. Radioanal.
Bossew, P. (2010). “Radon: exploring the log-normal Nucl. Chem. 240, 237 –241.
mystery,” J. Environ. Radioact. 101, 826– 834. Cavallo, A. J. (2000). “Understanding mine aerosols for

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Boulaud, D., and Chouard, J. C. (1992). “Submicron-sized radon risk assessment,” J. Environ. Radioact. 51,
aerosol and radon progeny measurements in a uranium 99–119.
mine,” Radiat. Prot. Dosim. 45, 91–94. Cavallo, A., Hutter, A., and Shebell, P. (1999). “Radon
Brenner, D. J., Little, J. B., and Sachs, R. K. (2001). “The progeny unattached fraction in an atmosphere far from
bystander effect in radiation carcinogenesis. II: a quan- radioactive equilibrium,” Health Phys. 76, 532 –536.
titative model,” Radiat. Res. 155, 402 –408. CCSI (2002). U.K. Childhood Cancer Study Investigators.
Brown, W. L., and Hess, C. T. (1992). “Measurement of the “The United Kingdom Childhood Cancer Study of ex-
biotransfer and time constant of radon from ingested posure to domestic sources of ionising radiation: 1:
water by human breath analysis,” Health Phys. 62, Radon gas,” Br. J. Cancer 86, 1721–1726.
162– 170. Chambers, D. B., and Zielinski, J. M. (2011). “Residential
Brunke, E.-G., Labuschagne, C., Parker, B., van der Spuy, radon levels around the world,” pp. 828 –839 in
D., and Whittlestone, S. (2002). “Cape Point GAW Encyclopedia of Environmental Health, Nriagu, J., Eds.
Station 222Rn detector: factors affecting sensitivity and (Elsevier, New York).
accuracy,” Atmos. Environ. 36, 2257– 2262. Chameaud, J., Perraud, R., Masse, R., and Lafuma, J.
Bucci, S., Pratesi, G., Viti, M. L., Pantani, M., Bochicchio, (1981). “Contribution of animal experimentation to the
F., and Venosos, G. (2011). “Radon in workplaces: first interpretation of human epidemiological data,” pp.
results of an extensive survey and comparison with 222– 227 in International Conference Radiation
radon in homes,” Radiat. Prot. Dosim. 145, 202 –205. Hazards in Mining: Control, Measurement and Medical
Burke, O., and Murphy, P. (2011). “The use of volunteer Aspects. Gomez, M., Eds. (Society of Mining Engineers
radon measurements for radon mapping purposes: an of the American Institute of Mining, Metallurgical, and
examination of sampling bias issues,” J. Radiol. Prot. Petroleum Engineers, New York).
31, 319– 328. Chameaud, J., Masse, R., and Lafuma, J. (1984).
Burke, O., Long, S., Murphy, P., Organo, C., Fenton, D., “Influence of radon daughter exposure at low doses on
and Colgan, P. A. (2010). “Estimation of seasonal correc- occurrence of lung cancer in rats,” Radiat. Prot. Dosim.
tion factors through Fourier decomposition analysis—a 7, 385– 388.
new model for indoor radon levels in Irish homes,” Chao, C. Y. H., Tung, T. C. W., Chan, D. W. T., and Burnett,
J. Radiol. Prot. 30, 433 –443. J. (1997). “Determination of radon emanation and back
Busigin, A., van der Vooren, A., and Phillips, C. R. (1978). diffusion characteristics of building materials in small
Attached and Unattached Radon Daughters: Measurements chamber tests,” Build. Environ. 32, 355 –362.
and Measurement Techniques in Uranium Mines and in the Charles, M. W. (2004). “The skin in radiological protec-
Laboratory. AECB Report 1169 (Atomic Energy Control tion—recent advances and residual unresolved issues,”
Board, Ottawa, Canada). Radiat. Prot. Dosim. 109, 323– 330.
Busigin, A., van der Vooren, A., and Phillips, C. R. (1981). Charles, M. W., Mill, A. J., and Darley, P. J. (2003).
“Measurements of the total and radioactive aerosol size “Carcinogenic risk of hot-particle exposures,” J. Radiol.
distribution in a Canadian uranium mine,” Am. Ind. Prot. 23, 5– 28.
Hyg. Assoc. J. 42, 310– 314. Chavez, A., Monnin, M., Segovia, N., Seidel, L. J., Peńa, P.,
Butterweck, G., Porstendörfer, J., Reineking, A., and Moreno, A., and Balcazar, M. (1997). “Automatic SSNTD
Kesten, J. (1992). “Unattached fraction and the aerosol exchanger for soil radon determinations,” Radiat. Meas.
size distribution of the radon progeny in a natural cave 27, 587–591.
and mine atmospheres,” Radiat. Prot. Dosim. 45, Chen, J. (2003). “Estimate of annual average radon con-
167– 170. centration in the normal living area from short-term
Butterweck, G., Schuler, Ch., Vezzu, G., Muller, R., Marsh, tests,” Health Phys. 85, 740 –744.
J. W., Thrift, S., and Birchall, A. (2002). “Experimental Chen, J., and Marro, L. (2011). “Assessment of radon equi-
determination of the absorption rate of unattached radon librium factor from distribution parameters of

171
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

simultaneous radon and radon progeny measurrments,” Cornelis, J., Vanmarcke, H., Landsheere, C., and Poffijn,
Radiat. Environ. Biophys. 50, 597 –601. A. (1993). “Modeling radon progeny absorbed in glass,”
Chen, T.-R., Cheng, Y.-S., Hopke, P. K., Tung, C.-J., and Health Phys. 65, 414 –417.
Pourprix, M. (1997). “Electrical mobility and size distri- Cosma, C., Moldovan, M., Dicu, T., and Kovacs, T. (2008).
bution of aged 212Pb nanometer carriers in nitrogen “Radon in water from Transylvania (Romania),” Radiat.
gas,” J. Aerosol Sci. 28, 1465–1477. Meas. 43, 1423– 1428.
Chen, C.-J., Liu, C.-C., and Lin, Y.-M. (1998). “Measurement Cozmuta, I., van der Graaf, E. R., and de Meijer, R. J.
of equilibrium factor and unattached fraction of radon (2003). “Moisture dependence of radon transport in con-
progeny in Kaohsiung, Taiwan,” Appl. Radiat. Isot. 49, crete: measurements and modeling,” Health Phys. 85,
1613–1618. 438– 456.
Chen, J., Tracy, B. L., Zielinski, J. M., and Moir, D. (2008). Crawford-Brown, D. J. (1989). “The biokinetics and
“Determining the sample size required for a community dosimetry of radon-222 in the human body following
radon survey,” Health Phys. 94, 362– 365. ingestion of groundwater,” Environ. Geochem. Health
Cheng, Y. S., and Yeh, H. C. (1980). “Theory of screen type 11, 10– 17.
diffusion battery,” J. Aerosol Sci. 11, 313 –319. Crookes, W. (1903). “The emanation of radium,” Roy. Soc.
Cheng, K.-H., Cheng, Y.-S., Yeh, H.-C., Guilmette, R. A., Lond. 71, 405– 408.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Simpson, S. Q., Yang, Y.-H., and Swift, D. L. (1996). “In Cross, F. T. (1988a). Radon Inhalation Studies in Animals,
vivo measurements of nasal airway dimensions and DOE/ER-0396 (National Technical Information Service,
ultrafine aerosol deposition in the human nasal and Springfield, VA).
oral airways,” J. Aerosol Sci. 27, 785– 801. Cross, F. T. (1988b). “Radon inhalation studies in
Cheng, Y.-S., Chen, T.-R., Wasiolek, P. T., and Van Engen, animals,” Radiat. Prot. Dosim. 24, 463 –466.
A. (1997). “Radon and radon progeny in the Carlsbad Cross, F. T., and Monchaux, G. (1999). “Risk assessment of
Caverns,” Aerosol Sci. Technol. 26, 74–92. radon health effects from experimental animal studies.
Cheng, Y. S., Chen, T. R., Yeh, H. C., Bigu, J, Holub, R., A joint review of PNNL (USA) and CEA-COGEMA
Tu, K., Knutsond, E. O., and Falk, R. (2000). (France) data,” pp. 85– 105 in Indoor Radon Exposure
“Intercomparison of activity size distribution of thoron and Its Health Consequences. Quest for the True Story of
progeny and a mixture of radon and thoron progeny,” Environmental Radon and Lung Cancer, Inaba, J.,
J. Environ. Radioact. 51, 59–78. Yonehara, H. and Doim. M., Eds. (Kodansha Scientific
Cherouati, D. E., and Djeffal, S. (1988). “Measurements of Limited, Tokyo).
radon and radon daughters in dwellings in Algiers,” Cross, F. T., Palmer, R. F., Filipy, R. E., Dagle, G. E., and
Radiat. Prot. Dosim. 25, 137– 139. Stuart, B. O. (1982). “Carcinogenic effects of radon
Chevillard, A., Ciais, P., Karstens, U., Heimann, M., daughters, uranium ore dust and cigarette smoke in
Schmidt, M., Levin, I., Jacob, D., Podzun, R., Kazan, V., beagle dogs,” Health Phys. 42, 33–52.
Sartorius, H., and Weingartner, E. (2002). “Transport of Cross, F. T., Palmer, R. F., Dagle, G. E., Busch, R. H., and
222
Rn using the regional model REMO: a detailed com- Buschbom, R. L. (1984). “Influence of radon daughter
parison with measurements over Europe,” Tellus B 54, exposure rate, unattachment fraction, and disequilib-
850– 871. rium on occurrence of lung tumours,” Radiat. Prot.
Çile, S., Altınsoy, N., and Çelebi, N. (2010). Radon concen- Dosim. 7, 381– 384.
trations in three underground lignite mines in Turkey,” Damon, P. E., and Hyde, H. I. (1952). “Scintillation tube
Radiat. Prot. Dosim. 138, 78–82. for the measurement of radioactive gases,” Rev. Sci.
Clouvas, A., Xanthos, S., and Antonopoulos-Domis, M. Instrum. 23, 766.
(2003). “Long term measurements of radon equilibrium Darby, S. C., Whitley, E., Howe, G. R., Hutchings, S. J.,
factor in Greek dwellings,” Radiat. Prot. Dosim. 103, Kusiak, R. A., Lubin, J. H., Morrison, H. I., Tirmarche,
269– 271. M., Tomasek, L., Radford, E. P., Roscoe, R. J., Samet, J.
Cohen, B. L. (1986). “A national survey of 222Rn in U.S. M., and Yao, S. X. (1995). “Radon and cancers other
homes and correlating factors,” Health Phys. 51, 175–183. than lung cancer in underground miners: a collabora-
Collier, C. G., Strong, J. C., Baker, S. T., Eldred, T., tive analysis of 11 studies,” J. Natl. Cancer Inst. 87,
Humphreys, J. A. H., and Cobb, L. M. (1999). “Effects of 378– 384.
continuous inhalation exposure of rats to radon and its Darby, S., Hill, D., Auvinen, A., Barros-Dios, J. M.,
progeny at various levels of dose and dose rate: interim Baysson, H., Bochicchio, F., Deo, H., Falk, R.,
results,” Radiat. Res. 152, S141 –S144. Forastiere, F., Hakama, M., Heid, I., Kreienbrock, L.,
Collier, C. G., Strong, J. C., Humphreys, J. A., Timpson, N., Kreuzer, M., Lagarde, F., Mäkeläinen, I., Muirhead, C.,
Baker, S. T., Eldred, T., Cobb, L., Papworth, D., and Oberaigner, W., Pershagen, G., Ruano-Ravina, A.,
Haylock, R. (2005). “Carcinogenicity of radon/radon Ruosteenoja, E., Schaffrath Rosario, A., Tirmarche, M.,
decay product inhalation in rats—effect of dose, dose rate Tomásek, L., Whitley, E., Wichmann, H. E., and Doll, R.
and unattached fraction,” Int. J. Radiat. Biol. 81, (2005). “Radon in homes and risk of lung cancer: collab-
631–647. orative analysis of individual data from 13 European
Cornelis, J., Landsheere, C., van Trier, A., Vanmarcke, H., case– control studies,” Br. Med. J. 330, 223 –228.
and Poffijn, A. (1992). “Experiments on glass-absorbed Darby, S., Hill, D., Deo, H., Auvinen, A., Barros-Dios, J.
polonium-210,” Int. J. Radiat. Appl. Instr. Part A. Appl. M., Baysson, H., Bochicchio, F., Falk, R., Farchi, S.,
Radiat. Isot. 43, 127 –138. Figueiras, A., Hakama, M., Heid, I., Hunter, N.,

172
 References

Kreienbrock, L., Kreuzer, M., Lagarde, F., Mäkeläinen, Radiation Protection, Earth Sciences and the Environment
I., Muirhead, C., Oberaigner, W., Pershagen, G., (World Scientific Publishing, Singapore).
Ruosleenoja, E., Schaffrath Rosario, A., Tirmarche, M., Durridge (2000). Durridge Company, Inc. Rad7 User
Tomasek, L., Whitley, E., Wichmann, H.-E., and Doll, R. Manual (Durridge Company Inc., Billerica, MA).
(2006). “Residential radon and lung cancer—detailed Eappen, K. P., and Mayya, Y. S. (2004). “Calibration
results of a collaborative analysis of individual data on factors for LR-115 (type-II) based radon thoron discrim-
7148 persons with lung cancer and 14,208 persons inating dosimeter,” Radiat. Meas. 38, 5– 17.
without lung cancer from 13 epidemiological studies in Eappen, K. P., Sapra, B. K., and Mayya, Y. S. (2007). “A
Europe,” Scand. J. Work Environ. Health 32 (Suppl. 1), novel methodology for online measurement of thoron
1– 84. using Lucas scintillation cell,” Nucl. Instr. Methods
Deetjen, P., Falkenbach, A., Harder, D., Jöckel, H., Kaul, Phys. Res. A 572, 922 –925.
A., and Philipsborn, H. (2005). Radon als Heilmittel Eappen, K. P., Nair, R. N., and Mayya, Y. S. (2008).
(Kovac Verlag, Hamburg) (in German). “Simultaneous measurement of radon and thoron using
Denman, A. R., Crockett, R. G. M., Groves-Kirkby, C. J., Lucas scintillation cell,” Radiat. Meas. 43, 91– 97.
Phillips, P. S., Gillmore, G. K., and Woolridge, A. C. Eatough, J. P., and Henshaw, D. L. (1992). “Radon and
(2007). “The value of seasonal correction factors in asses- thoron associated dose to the basal layer of the skin,”

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

sing the health risk from domestic radon—a case study Phys. Med. Biol. 37, 955 –967.
in Northamptonshire, UK,” Environ. Int. 33, 34–44. Eatough, J. P., Worley, A., and Moss, G. R. (1999). “Personal
Dersch, R., and Schotzig, U. (1998). “Production and monitoring of 218Po and 214Po radionuclide deposition
measurement of 222Rn standards,” Appl. Radiat. Isot. onto individuals under normal environmental exposure
49, 1171 –1174. conditions,” Phys. Med. Biol. 44, 2227–2239.
Desrosiers, A., Kennedy, A., and Little, J. B. (1978). “222Rn Ekman, J., Helgesson, J., Karlsson, L. B., Mohsenpour,
daughter dosimetry in the Syrian golden hamster M., Riihimäki, I., Touboltsev, V., Jalkanan, P., Virtanen,
lung,” Health Phys. 35, 607 –623. A., Kettunen, H., Huikari, J., Nieminen, A., Moore, I.,
Detao, X., and Fuqi, Z. (1997). “The pilot application of Penttilä, H., Arstila, K., Äystö, J., Räisänen, J., and
radon dosimeter for miners,” Chin. J. Radiat. Mediat. Whitlow, H. J. (2006). “Retention of Pb isotopes in glass
Prot. 17, 123– 126 (in Chinese). surfaces for retrospective assessment of radon expos-
Dimitrova, I., Pressyanov, D., Georgiev, S., and Yankov, P. ure,” Nucl. Instr. Meth. Phys. Res. B 249, 544 –547.
(2011). “Logistic of surveys of retrospective radon con- El-Hussein, A. (1996). “Unattached fractions, attachment
centrations by home-stored CDs/DVDs,” Radiat. Prot. and deposition rates of radon progeny in indoor air,”
Dosim. 145, 300 –304. Appl. Radiat. Isot. 47, 515 –523.
Doi, M., Fujimoto, K., Kobayashi, S., and Yonehara, H. El-Hussein, A. (2005). “A study on natural radiation expos-
(1994). “Spatial distribution of thoron and radon con- ure in different realistic living rooms,” J. Environ.
centrations in the indoor air of a tradiational Japanese Radioact. 79, 355 –367.
wooden house,” Health Phys. 66, 43– 49. El-Hussein, A., Ahmed, A. A., and Mohammed, A. (1998).
Dua, S. K., and Hopke, P. K. (1996). “Hygroscopic growth “Radiation dose to the human respiratory tract from in-
of assorted indoor aerosols,” Aerosol Sci. Technol. 24, halation of radon-222 and its progeny,” Appl. Radiat.
151– 160. Isot. 49, 783– 790.
Dua, S. K., Kotrappa, P., and Gupta, P. C. (1983). El-Hussein, A., Mohammed, A., and Ahmed, A. A. (1999).
“Influence of relative humidity on the charged fraction “Radon exhalation and ultrafine fraction of radon progeny
of decay products of radon and thoron,” Health Phys. 45, in closed room air,” Atmos. Environ. 33, 183–190.
152– 157. Ellett, W. H., and Nelson, N. S. (1985). “Epidemiology and
Dua, S. K., Hopke, P. K., and Raunemaa, T. (1999). risk assessment: testing models for radon-induced lung
“Hygroscopicity of diesel aerosols,” Water Air Soil Poll. cancer risk,” pp. 70– 107 in Indoor Air and Health,
112, 247 –257. Gammage, R. B., and Kaye, S. V., Eds. (Lewis
Dubois, G. (2005). An Overview of Radon Surveys in Europe, Publishers, Chelsea, MI).
EUR 21892 EN, http://radonmapping.jrc.ec.europa. Elster, J., and Geitel, H. (1902). “On the radioactive eman-
eu/it (accessed). (European Commission, Luxembourg). ation in atmospheric air,” Physikal. Zeitschr. 4, 522.
Dubois, G., Bossew, P., Tollefsen, T., and De Cort, M. EPA (1992). U.S. Environmental Protection Agency.
(2010). “First steps towards a European atlas of natural National Residential Radon Survey: Summary Report,
radiation: Status of the European indoor radon map,” EPA 402-R-92-011 (US Environmental Protection
J. Environ. Radioact. 101, 786– 798. Agency, Washington, DC).
Duport, P., Madelaine, G., Zettwoog, P., and Pineau, J. F. EPA (1993). U.S. Environmental Protection Agency. Map
(1980). “Enregistrement des rayonnements alpha dans of Radon Zones, EPA 402-F-93-013 (US Environmental
le dosimètre individuel et le dosimètre de site du Protection Agency, Washington, DC).
Commissariat à l’Énergie Atomique,” pp. 609– 615 in EPA (2003). U.S. Environmental Protection Agency. EPA
Solid State Nuclear Detectors: Proceedings of the 10th Assessment of Risks from Radon in Homes, EPA
International Conference, François, H., Kurtz, N., and 402-R-03-003 (U.S. Envionmental Protection Agency,
Massue, J.-P., Eds. (Elsevier, New York). Washington, DC).
Durrani, S. A., and Ilic, R., Eds. (1997). Radon EU (2011). European Union. Proposal for a Council
Measurements by Etch Track Detectors: Applications in Directive Laying Down Basic Safety Standards for

173
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Protection Against the Dangers Arising from Exposure decay products in some Iranian non-uranium under-
to Ionising Radiation. Draft presented under Article 31 ground mines,” Radiat. Prot. Dosim. 118, 111 –116.
Euratom Treaty for the opinion of the European Fenell, S. G., Mackin, G. M., McGarry, A. T., and Pollard,
Economic and Social Committee, COM(2011) 593 final, D. (2002). “Radon exposure in Ireland,” Int. Congr. Ser.
2011/0254 (NLE) (European Commission, Brussels). 1225, 71– 77.
EU (2014). European Union. “Council Directive 2013/59/ Ferron, G. A., Kreyling, W. G., and Haider, B. (1988).
EURATOM of 5 Decmber 2013 laying down basic safety “Inhalation of salt aerosol particles—II. Growth and de-
standards for protection against the dangers arising position in the human respiratory tract,” J. Aerosol Sci.
from exposure to ionising radiation, and repealing 19, 611–631.
Directives 89/618/Euratom, 90/641/Euratom, 96/29/ Fews, A. P., Henshaw, D. L., Keitch, P. A., Close, J. J., and
Euratom, 97/43/Euratom and 2003/122/Euratom,” Wilding, R. J. (1999). “Increased exposure to pollutant
J. Eur. Union, L 13, 1– 73. aerosols under high voltage power lines,” Int. J. Radiat.
Evans, R. D., and Goodman, C. (1940). “Determination of Biol. 75, 1505–1521.
thoron content of air and its bearing on lung cancer Field, R. W., Steck, D. J., Parkhurst, M. A., Mahaffey, J.
hazards in industry,” J. Industr. Hyg. Tox. 22, 89–91 A., and Alavanja, M. C. (1999). “Intercomparison of
[reprinted in Health Phys. (1980) 38, 919 –928]. retrospective radon detectors,” Environ. Health

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Fakir, H., Hofmann, W., and Aubineau-Laniece, I. (2005). Perspect. 107, 905 –910.
“Microdosimetry of radon progeny alpha particles in Firestone, R. B., and Shirley, V. S. (1999). Table of Isotopes
bronchial airway bifurcations,” Radiat. Prot. Dosim. (Wiley & Sons, New York).
117, 382 –394. Fisenne, I. M., George, A. C., and Keller, H. W. (1990). “A
Fakir, H., Hofmann, W., and Caswell, R. S. (2008). “Radon calibration and quality assurance program for environ-
progeny microdosimetry in human and rat bronchial mental radon measurements,” J. Res. Natl Inst. Stand.
airways: the effect of crossfire from the alveolar region,” Technol. 95, 127 –134.
Radiat. Prot. Dosim. 130, 149– 161. Fisher, E. L., Field, R. W., Smith, B. J., Lynch, C. F., Steck,
Fakir, H., Hofmann, W., Tan, W. Y., and Sachs, R. K. (2009). D. J., and Neuberger, J. S. (1998). “Spatial variation of
“Triggering-response model for radiation-induced by- residential radon concentrations: the Iowa Radon Lung
stander effects,” Radiat. Res. 171, 320–331. Cancer Study,” Health Phys. 75, 506 –513.
Falk, R., More, H., and Nyblom, L. (1992). “Measuring Fitzgerald, B., and McLaughlin, J. P. (1996). “The meas-
techniques for environmental levels of radon-220 in air urement and assessment of 210Po surface activity on
using flow-through Lucas cell and multiple time ana- glasses in dwellings,” Environ. Int. 22 (Suppl. 1),
lysis of recorded pulse events,” Appl. Radiat. Isot. 43, 845– 855.
111 –118. Fleischer, R. L., Meyer, N. R., Hadley, S. A., MacDonald, J.,
Falk, R., Mellander, H., Nyblom, L., and Ostergren, I. and Cavallo, A. (2001). “Personal radon dosimetry from
(1996). “Retrospective assessment of radon exposure by eyeglass lenses,” Radiat. Prot. Dosim. 97, 251–258.
measurements of 210Po implanted in surfaces using an Font, L. I. (2009). “On radon surveys: design and inter-
alpha track detector technique,” Environ. Int. 22(Suppl. pretation,” Radiat. Meas. 44, 964– 968.
1), 857 –861. Ford, J. R., and Terzaghi-Howe, M. (1992). “Basal cells are
Falk, R., Philipson, K., Svartengren, M., Jarvis, N., Bailey, the progenitors of primary tracheal epithelial cell cul-
M., and Camner, P. (1997). “Clearance of particles tures,” Exp. Cell Res. 198, 69–77.
from small ciliated airways,” Exp. Lung Res. 23, Franco-Marina, F., Segovia, N., Ruiz, W., Godinez, L.,
495– 515. Tavera, L., Lopez, A., Chavez, A., Pena, P., and Ponciano,
Falk, R., Philipson, K., Svartengren, M., Bergmann, R., G. (2001). “Short and long term indoor radon survey in
Hofmann, W., Jarvis, N., Bailey, M., and Camner, P. Mexico City,” Radiat. Meas. 34, 545–548.
(1999). “Assessment of long-term bronchiolar clearance Frank, A. L., and Benton, E. V. (1973). “A diffusion
of particles from measurements of lung retention and chamber radon dosimeter for use in mine environment,”
theoretical estimates of regional deposition,” Exp. Lung Nucl. Instr. Meth. 109, 537 –539.
Res. 25, 495 –516. Friedmann, H. (2005). “Final results of the Austrian
Falk, R., Almren, K., and Ostergren, I. (2001). “Experience Radon Project,” Health Phys. 89, 339– 348.
from retrospective radon exposure estimations for indi- Froňka, A., Moučka, L., and Jeřábek, M. (2008).
viduals in a radon epidemiological study using solid- “Detection properties of a measuring system for a con-
state nucleat track detectors,” Sci. Total Environ. 272, tinuous soil radon concentrations monitoring,” Radiat.
61–66. Prot. Dosim. 130, 56– 59.
Farid, S. M. (1993). “Measurement of the equilibrium Fujimoto, K., Kobayashi, S., Uchiyama, M., Doi, M., and
factor for 222Rn progeny in Bangladeshi dwellings,” Nakamura, Y. (1997). “Nationwide indoor radon survey
Health Phys. 65, 493 –496. in Japan,” Jpn. J. Health Phys. 32, 41–51 (in Japanese).
Farkas, A., and Szöke, I. (2013). “Simulation of bronchial Garland, L. H., Beier, R. L., Coulson, W., Heald, J. H., and
mucociliary clearance of insoluble particles by computa- Stein, R. L. (1962). “The apparent sites of origin of car-
tional fluid and particle dynamics methods,” Inhal. cinomas in the lung,” Radiology 78, 1– 11.
Toxicol. 25, 593 –605. Gastineau, R. M., Walsh, P. J., and Underwood, N. (1972).
Fathabadi, N., Ghiassi-Nejad, M., Haddadi, B., and “Thickness of bronchial epithelium with relation to ex-
Moradi, M. (2006). “Miners’ exposure to radon and its posure to radon,” Health Phys. 23, 857 –860.

174
 References

Genrich, V. (1995). AlphaGUARD PQ-2000/MC50 concentrations in the United Kingdom,” Health Phys.
Multiparameter Radon Monitor, Genitron Instruments 64, 2– 12.
GmbH. Gundersen, L. C. S., and Schumann, R. R. (1996).
George, A. C. (1984). “Passive, integrated measurement of “Mapping the radon potential of the United States:
indoor radon using activated carbon,” Health Phys. 46, Examples from the Appalachians,” Environ. Int. 22
867– 872. (Suppl. 1), 829 –837.
George, A. C. (1996). “State-of-the-art instruments for Guo, Q., Zhang, L., and Guo, L. (2012) Assessment of the
measuring radon/thoron and their progeny in dwell- unattached fraction of indoor radon progeny and its con-
ings—a review,” Health Phys. 70, 451– 463. tribution to dose: a pilot study in China. J. Radiol. Prot.
George, A. C., Hinchliffe, L., and Sladowski, R. (1975). 32, 447– 454.
“Size distribution of radon daughter particles in Hadler, N. J. C., Iunes, P. J., Osorio, A. M., and Paulo, S.
uranium mine atmospheres,” Am. Ind. Hyg. Assoc. J. R. (1991). “Relationship between track size and energy
36, 484– 490. for alpha particles in CR-39,” Nucl. Tracks Radiat.
Geranios, A., Nikolopoulos, D., Louizi, A., and Karatzi, A. Meas. 19, 313 –318.
(2004). Multiple radon survey in spa of Loutra Edipsou Haefeli-Bleuer, B., and Weibel, E. R. (1988). “Morphometry
(Greece),” Radiat. Prot. Dosim. 112, 251 –258. of the human acinus,” Anat. Rec. 220, 401–414.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Gilbert, E. S., Cross, F. T., and Dagle, G. E. (1996). Hafez, A. F., Bishara, A. A., Kotb, M. A., and Hussein, A.
“Analysis of lung tumor risk in rats exposed to radon,” S. (2003). “Regular radon activity concentration and ef-
Radiat. Res. 145, 350 –360. fective dose measurements inside the great pyramid
Gillmore, G. K., Phillips, P. S., Denman, A. R., and with passive nuclear track detectors,” Health Phys. 85,
Gilbertson, D. D. (2002). “Radon in the Creswell Crags 210– 215.
Permian limestone caves,” J. Environ. Radioact. 62, Hamori, K., Toth, E., Losonci, A., and Minda, M. (2006).
165– 179. “Some remarks on the indoor radon distribution in a
Gillmore, G. K., Phillips, P. S., and Denman, A. R. (2005). country,” Appl. Radiat. Isot. 64, 859 –863.
“The effects of geology and the impact of seasonal cor- Hanley, O., Gutierrez-Villanueva, J. L., Currivan, L., and
rection factors on indoor radon levels: a case study ap- Pollard, D. (2008). “Assessment of the uncertainties in
proach,” J. Environ. Radioact. 84, 469– 479. the Radiological Portection Institute of Ireland (RPII)
Goble, R., and Socolow, R. (1990). “High radon houses: radon measurements service,” J. Environ. Radioact. 99,
questions about log normal distribution and implica- 1578–1582.
tions for epidemiology and risk assessment,” pp. 52–59 Haque, A. K. M. M., and Collinson, A. J. L. (1967).
in Proceedings of the 1990 International Symposium on “Radiation dose to the respiratory system due to
Radon and Radon Reduction Technology, EPA-600/9-91 radon and its daughter products,” Health Phys. 13,
026a, vol. 1. (US Environmental Protection Agency, 431– 443.
Washington, DC). Hardcastle, G. D., and Miles, J. C. H. (1996). “Ageing and
Gomez, J. C., Oliveira, A. A., Arnaud, M. I., and Ciallella, fading of alpha particle tracks in CR-39 exposed to air,”
H. E. (1993). “Radon in dwellings in Argentina,” pp. Radiat. Prot. Dosim. 67, 295– 298.
391– 397 in Proceedings of the International Conference Harley, J. H. (1980). “Sampling and measurement of air-
on High Levels of Natural Radiation (International borne daughter products of radon,” Health Phys. 38,
Atomic Energy Agency, Vienna). 1067–1074.
Gosink, T. A., Baskaran, M., and Holleman, D. F. (1990). Harley, N. H. (1984). “Comparing radon daughter dose: en-
“Radon in the human body from drinking water,” vironmental versus underground exposure,” Radiat.
Health Phys. 59, 919 –924. Prot. Dosim. 7, 371– 375.
Groves-Kirkby, C. J., Denman, A. R., Crockett, R. G. M., Harley, N. H. (1988). “Radon daughter dosimetry in the
Phillips, P. S., Woolridge, A. C., and Gillmore, G. K. rat tracheobronchial tree,” Radiat. Prot. Dosim. 24,
(2006). “Time-integrating radon gas measurements in 457– 461.
domestic premises: comparison of short-, medium- and Harley, J. H., and Fresco, J. M. (1951). “Retention of radon
long-term exposures,” J. Environ. Radioact. 86, daughter products in the respiratory system”, p. 29 in
92–109. HASL Report No. 22 (Atomic Energy Commission, New
Gruber, E., Salama, E., and Rühm, W. (2011). “Real-time York).
measurement of individual occupational radon exposures Harley, N. H., and Pasternack, B. S. (1972). “Alpha absorp-
in tombs of the Valley of the Kings, Egypt,” Radiat. Prot. tion measurements applied to lung dose from radon
Dosim. 144, 620–626. daughters,” Health Phys. 23, 771 –782.
Gründel, M., Reineking, A., and Porstendörfer, J. (2005). Harley, N. H., and Pasternack, B. S. (1973). “Experimental
“Activity size distribution in outdoor air: Short-lived absorption applied to lung dose from thoron daughters,”
(214Po, 214Bi/214Po) and long-lived (210Pb, 210Po) radon Health Phys. 24, 379 –386.
and thoron (212Pb, 212Po) decay products and 7Be,” pp. Harley, N. H., and Pasternack, B. S. (1982). “Environmental
454– 458 in Natural Radiation Enviroment VII, radon daughter alpha dose factors in a five-lobed human
McLaughlin, J. P., Simopoulos, S. E., and Steinhäusler, lung,” Health Phys. 42, 789–799.
F., Eds. (Elsevier Science, Oxford). Harley, N. H., and Robbins, E. S. (1992). “222Rn alpha dose
Gunby, J. A., Darby, S. C., Miles, J. C. H., Green, B. M. R., to organs other than lung,” Radiat. Prot. Dosim. 45,
and Cox, D. R. (1993). “Factors affecting indoor radon 619– 622.

175
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Harley, N. H., and Robbins, E. S. (1994). “A biokinetic Hering, S. V., Friedlander, S. K., Coolins, J. J., and Richards,
model for 222Rn gas distribution and alpha dose in L. W. (1979). “Design and evaluation of a low pressure im-
humans following ingestion,” Environ. Int. 20, 605–610. pactor. 2,” Environ. Sci. Technol. 13, 184–188.
Harley, N. H., and Terilli, T. B. (1990). “Predicting annual Hinds, W. C. (1982). Aerosol Technology: Properties,
average indoor 222Rn concentration,” Health Phys. 59, Behavior, and Measurement of Airborne Particles (John
205– 209. Wiley and Sons, New York).
Harley, N. H., Chittaporn, P., Roman, M. H., and Hofmann, W. (1982a). “Dose calculations for the respira-
Sylvester, J. (1991). “Personal and home 222Rn and tory tract from inhaled natural radioactive nuclides as a
gamma-ray exposure measured in 52 dwellings,” Health funtion of age—II. Basal cell dose distributions and
Phys. 61, 737 –744. associated lung cancer risk,” Health Phys. 43, 31–44.
Harley, N. H., Meyers, O. A., and Robbins, E. S. (1992). Hofmann, W. (1982b). “Mathematical model for the post-
“222Rn dosimetry in the dog lung,” Radiat. Prot. Dosim. natal growth of the human lung,” Respir. Physiol. 49,
45, 611–617. 115– 129.
Harley, J. H., Jetter, E. S., and Nelson, N. (1994). “Elimination Hofmann, W. (1982c). “Cellular lung dosimetry for inhaled
of radon from the body,” Environ. Int. 20, 573–584. radon decay products as a base for radiation-induced
Harley, N. H., Cohen, B. S., and Robbins, E. S. (1996). lung cancer risk assessment. I. Calculation of mean cel-

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

“The variability in radon decay product bronchial dose,” lular doses,” Radiat. Environ. Biophys. 20, 95– 112.
Environ. Int. 22(Suppl. 1), 959– 964. Hofmann, W., and Sturm, R. (2004). “Stochastic model of
Harley, N., Chittaporn, P., Medora, R., and Merrill, R. particle clearance in human bronchial airways,”
(2010). “Measurement of the indoor and outdoor 220Rn J. Aerosol Med. 17, 73 –89.
(thoron) equilibrium factor: application to lung dose,” Hofmann, W., and Winkler-Heil, R. (2011). “Radon
Radiat. Prot. Dosim. 141, 357– 362. lung dosimetry models,” Radiat. Prot. Dosim. 145,
Harley, N. H., Chittaporn, P., and Marsciano, A. (2011). 206– 212.
“Residential radon remediation: performance over 17 Hofmann, W., Steinhäusler, F., and Pohl, E. (1979). “Dose
years,” Radiat. Prot. Dosim. 145, 194 –197. calculations for the respiratory tract from inhaled
Harley, N. H., Chen, J., Chittaporn, P., Sorimachi, A., and natural radioactive nuclides as a funtion of age—I.
Tokonami, S. (2012a). “Long term measurements of Compartmental deposition, retention and resulting
indoor radon equilibrium factor,” Health Phys. 102, dose,” Health Phys. 37, 517– 532.
459– 462. Hofmann, W., Martonen, T. B., and Graham, R. C. (1989a).
Harley, N. H., Fisenne, I. M., and Robbins, E. S. (2012b). “Predicted deposition of nonhygroscopic aerosols in the
“Attempted validation of ICRP 30 and ICRP 66 respira- human lung as a function of subject age,” J. Aerosol.
tory models,” Radiat. Prot. Dosim. 152, 14– 17. Med. 2, 49–68.
Harley, N. H., Chittaporn, P., Cook, G. B., and Fisenne, I. Hofmann, W., Martonen, T. B., and Ménache, M. G.
M. (2014). “Radon water to air transfer measured in a (1989b). “Age-dependent lung dosimetry of radon
bathroom in an energy-efficient homewith a private progeny,” pp. 317 –344 in Extrapolation of Dosimetric
well,” Radiat. Prot. Dosim. 160, 231 –234. Relationships for Inhaled Particles and Gases, Crapo,
Harrison, J. D., and Marsh, J. W. (2012). “Effective dose J.D., Smolko, E., Miller, F., Graham J. and Hayes A.,
from inhaled radon and its progeny,” Ann. ICRP 41, Eds. (National Technical Information Service,
378– 388. Springfield, VA).
Harrison, J. D., and Muirhead, C. R. (2003). “Quantitative Hofmann, W., Martonen, T. B., and Menache, M. G. (1990).
comparisons of canmcer induction in humans by intern- “A dosimetric model for localised radon progeny accu-
ally deposited radionuclides and external radiation,” mulations at tracheobronchial bifurcations,” Radiat.
Int. J. Radiat. Biol. 79, 1 –13. Prot. Dosim. 30, 245– 259.
HASL (1960). Health and Safety Laboratory. Hofmann, W., Ménache, M. G., and Graham, R. C. (1993).
Experimental Environmental Study of AEC Leased “Radon progeny dosimetry in the rat lung,” Health
Uranium Mines, HASL-91 (US Atomic Energy Agency, Phys. 64, 279 –290.
New York). Hofmann, W., Koblinger, L., and Mohamed, A. (1996).
Hattori, T., and Ishida, K. (1994). “Equilibrium factor and “Incorporation of biological variability into lung dosim-
unattached fraction of radon progeny in nuclear power etry by stochastic modeling techniques,” Environ. Int.
plants,” Radiat. Prot. Dosim. 55, 191 –197. 22(Suppl. 1), 995 –1003.
Hattori, T., Ichiji, T., and Ishida, K. (1995). “Behaviour of Hofmann, W., Bergmann, R., and Koblinger, L. (1999).
radon and its progeny in a Japanese office,” Radiat. “Characterization of local particle deposition patterns
Prot. Dosim. 62, 151– 155. in human and rat lungs by different morphometric
Heid, I. M., Kuchenhoff, H., Miles, J., Kreienbrock, L., and parameters,” J. Aerosol Sci. 30, 651 –667.
Wichmann, H. E. (2004). “Two dimensions of measure- Hofmann, W., Asgharian, B., and Winkler-Heil, R. (2002).
ment error: classical and Berkson error in residential “Modeling intersubject variability in particle deposition
radon exposure assessment,” J. Exp. Anal. Environ. in human lungs,” J. Aerosol Sci. 33, 219 –235.
Epidemiol. 14, 365– 377. Hofmann, W., Fakir, H., Aubineau-Laniece, I., and Pihet,
Hering, S. V., Flagan, R. C., and Friedlander, S. K. (1978). P. (2004). “Interaction of alpha particles at the cellular
“Design and evaluation of new low pressure impactor. level—implications for the radiation weighting factor,”
1,” Environ. Sci. Technol. 12, 667 – 673. Radiat. Prot. Dosim. 112, 493 – 500.

176
 References

Hofmann, W., Winkler-Heil, R., and Hussain, M. (2010). radon decay products under typical indoor domestic
“Modeling intersubject variability of bronchial doses for conditions,” Sci. Total Environ. 272, 97–103.
inhaled radon progeny,” Health Phys. 99, 523 –531. Hunter, N., Howarth, C. B., Miles, J. C. H., and Muirhead,
Hofmann, W., Winkler-Heil, R., Truta, L. A., and C. R. (2005). “Year-to-year variations in radon levels in
Tschiersch, J. (2014). “Application of a Monte Carlo lung a sample of UK houses with the same occupants,” pp.
dosimetry code to the inhalation of thoron progeny,” 438– 447 in Radioactivity in the Environment. Volume
Radiat. Prot. Dosim. 160, 96–99. 7: The Natural Radiation Environment (NRE-VII)
Holub, R. F., and Knutson, E. O. (1987). “Measuring (Elsevier, New York).
polonium-218 diffusion-coefficient spectra using mul- Hunter, N., Muirhead, C. R., Tomasek, L., Kreuzer, M.,
tiple wire screens,” pp. 340– 356 in Radon and Its Decay Laurier, D., Leuraud, K., Schnelzer, M., Grosche, B.,
Products: Occurrence, Properties and Health Effects. Placek, V., Heribanova, A., and Tirmarche, M. (2013).
Hopke, P. K., Eds. (American Chemical Society, “Joint analysis of three European nested case– control
Washington, DC). studies of lung cancer among radon exposed miners: ex-
Holub, R. F., Knutson, E. O., and Solomon, S. B. (1988). posure restricted to below 300 WLM,” Health Phys. 104,
“Tests of the graded wire screen technique for measur- 282– 292.
ing the amount and size distribution of unattached Hursh, J. B., and Mercer, T. T. (1970). “Measurement of

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

212
radon progeny,” Radiat. Prot. Dosim. 24, 265 –268. Pb loss rate from human lungs,” J. Appl. Physiol. 28,
Honig, A., Paul, A., Röttger, S., and Keyser, U. (1998). 268– 274.
“Environmental control of the German radon reference Hursh, J. B., Morken, D. A., Davis, T. P., and Lovaas, A.
chamber,” Nucl. Instr. Meth. Phys. Res. A 416, 525– 530. (1965). “The fate of radon ingested by man,” Health
Hopke, P. K., Wasiolek, P., Montassier, N., Cavallo, A., Phys. 11, 465– 476.
Gadsby, K., and Socolow, R. (1992). “Measurement of Hursh, J. B., Schraub, A., Sattler, E. L., and Hofmann,
activity-weighted size distributions of radon decay pro- H. P. (1969). “Fate of 212Pb inhaled by human subjects,”
ducts in a normally occupied home,” Radiat. Prot. Health Phys. 16, 257 –267.
Dosim. 45, 329 –331. IAEA (2014). International Atomic Energy Agency.
Hopke, P. K., Jenson, B., Li, C. S., Montassier, N, Radiation Protection and Safety of Radiation Sources:
Wasiolek, P., Cavallo, A. J., Gatsby, K., Socolow, R. H., International Basic Safety Standards. General Safety
and James, A. C. (1995). “Assessment of the exposure to Requirements Part 3 (International Atomic Energy
and dose from radon decay products in normally occu- Agency, Vienna).
pied homes,” Environ. Sci. Technol. 29, 1359– 1364. Ibrahimi, Z.-F., Howarth, C. B., and Miles, J. C. H. (2009).
Hosoda, M., Sorimachi, A., Yasuoka, Y., Ishikawa, T., “Sources of error in etched-track radon measurements
Sahoo, S. K., Furukawa, M., Hassan, N. M., Tokonami, and a review of passive detectors using results from a
S., and Uchida, S. (2009). “Simultaneous measurements series of radon intercomparisons,” Radiat. Meas. 44,
of radon and thoron exhalation rates and comparison 750– 754.
with values calculated by UNSCEAR equation,” ICRP (1979). International Commission on Radiological
J. Radiat. Res. 50, 333 –343. Protection. Limits for Intakes of Radionuclides
Hosoda, M., Ishikawa, T., Sorimachi, A., Tokonami, S., and by Workers, ICRP Publication 30, Part 1, Ann. ICRP
Uchida, S. (2011). “Development and application of a 2(3– 4).
continuous measurement system for radon exhalation ICRP (1980). International Commission on Radiological
rate,” Rev. Sci. Instr. 82, doi:http://dx.doi.org/10.1063/1. Protection. Limits for Intakes of Radionuclides
3527065. by Workers, ICRP Publication 30, Part 2. Ann. ICRP
Howarth, C. B., and Miles, J. C. H. (2008). Validation 4(3– 4).
Scheme for Organisations Making Measurements of ICRP (1987). International Commission on Radiological
Radon in Dwellings: 2008 Revision. HPA-RPD-047 Protection. Lung Cancer Risk from Indoor Exposures to
(Public Health England, London). Radon Daughters, ICRP Publication 50, Ann. ICRP
Huang, S. H., Lin, T. H., Su, C. S., and Weng, P. S. (1986). 17(1).
“Analysis of sparking technique for counting and mag- ICRP (1993a). International Commission on Radiological
nification of etched holes in cellulose nitrate films,” Protection. Protection Against Radon-222 at Home and
Nucl. Tracks Radiat. Meas. 11, 115–121. at Work, ICRP Publication 65, Ann. ICRP 23(2).
Huber, J., Ennemoser, O., and Schneider, P. (2001). ICRP (1993b). International Commission on Radiological
“Quality control of mitigation methods for unusually Protection. Age-Dependent Doses to Members of the
high indoor radon concentrations,” Health Phys. 81, Public from Intake of Radionuclides: Part 2. Ingestion
156– 162. Dose Coefficients, ICRP Publication 67, Ann. ICRP
Huet, C., Tymen, G., and Boulaud, D. (2001a). “Long-term 23(3–4).
measurements of equilibrium factor and unattached ICRP (1994). International Commission on Radiological
fraction of short-lived radon decay products in a dwell- Protection. Human Respiratory Tract Model for
ing—comparison with Praddo Model,” Aerosol Sci. Radiological Protection, ICRP Publication 66, Ann.
Technol. 35, 553 –563. ICRP 24(1–3).
Huet, C., Tymen, G., and Boulaud, D. (2001b). “Size distri- ICRP (2007). International Commission on Radiological
bution, equilibrium ratio and unattached fraction of Protection. The 2007 Recommendations of the International

177
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Commission on Radiological Protection, ICRP Publication Environment VIII, Paschoa, A. S., and Steinhäusler, F.,
103, Ann. ICRP 37(2–4). Eds. (American Institute of Physics, Melville, NY).
ICRP (2010). International Commission on Radiological Iimoto, T. (2000). “Time variation of the radon equilibrium
Protection. Lung Cancer Risk from Radon and Progeny factor in a reinforced concrete dwelling,” Radiat. Prot.
and Statement on Radon, ICRP Publication 115, Ann. Dosim. 92, 319 –321.
ICRP 40(1). Iimoto, T., Shirakata, Y., Tokonami, S., Furukawa, M., and
ICRP (2014). International Commission on Radiological Kurosawa, R. (1998). “Continuous 220Rn concentration
Protection. Radiological Protection Against Radon monitor using an electrostatic collection method,”
Exposure, ICRP Publication 126, Ann. ICRP 43(3). Radiat. Prot. Dosim. 77, 185– 189.
ICRU (2006). International Commission on Radiation Iimoto, T., Kosako, T., and Sugiura, N. (2001).
Units and Measurements. Sampling of Radionuclides “Measurements of summer radon and its progeny con-
in the Environment, ICRU Report 75, J. ICRU 6(1). centrations along with environmental gamma dose
IEC (2000). International Electrotechnical Commission. rates in Taiwan,” J. Environ. Radioact. 57, 57– 66.
Radiation Protection Instrumentation—Radon and Radon Iimoto, T., Tokonami, S., Morishita, Y., and Kosako, T.
Decay Product Measuring Instruments—Part 2: Specific (2004). “Application of activated charcoal radon collec-
Requirements gor Radon Measuring Instruments, IEC tors in high humidity environments,” J. Environ.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

61577-2 ed.1.0 (International Electrotechnical Radioact. 78, 69– 76.
Commission, Geneva). Ingersoll, J. G., Stitt, B. O., and Zapalac, G. H. (1983). “A
IEC (2006). International Electrotechnical Commission. fast and accurate method for measuring radon exhal-
Radiation Protection Instrumentation—Radon and ation rates from building materials,” Health Phys. 45,
Radon Decay Product Measuring Instruments—Part 1: 550– 552.
General Principles, IEC 61577-1 ed2.0 (International Ishikawa, T. (2004). “Effects of thoron on a radon detector
Electrotechnical Commission, Geneva). of pulse-ionization chamber type,” Radiat. Prot. Dosim.
IEC (2009). International Electrotechnical Commission. 108, 327– 330.
Radiation Protection Instrumentation—Radon and Ishikawa, T., Tokonami, S., and Nemeth, C. (2007).
Radon Decay Product Measuring Instruments—Part 4: “Calculation of dose conversion factors for thoron decay
Equipment for the Production of Reference Atmospheres products,” J. Radiol. Prot. 27, 447– 456.
Containing Radon Isotopes and Their Decay Products Ishimori, Y., and Maruo, Y. (2005). “Radon exhalation rate
(STAR), IEC 61577-4 ed.1.0 (International monitoring in/around the closed uranium mine sites in
Electrotechnical Commission, Geneva). Japan,” Int. Congr. Ser. 1276, 291– 292.
IEC (2011). International Electrotechnical Commission. Ishimori, Y., Ishikawa, T., and Tokonami, S. (2004).
Radiation Protection Instrumentation—Radon and “Radon intercomparison experiment at PTB in
Radon Decay Product Measuring Instruments—Part 3: Germany,” Jpn. J. Health Phys. 39, 263– 267.
Specific Requirements for Radon Decay Product ISO (1995). International Organization of Standardization.
Measuring Instruments, IEC 61577-3 ed.2.0 IEC (2015) Guide to the Expression of Uncertainty in Measurement
(International Electrotechnical Commission, Geneva). (International Organization of Standardization,
IEC (2015). International Electrotechnical Commission. Geneva).
Radiation Protection Instrumentation – Determination ISO (2005). International Organization of Standardization.
of Uncertainty of Measurement, IEC TR 62461:2015 General Requirements for the Competence of Testing and
(International Electrotechnical Commission, Geneva). Calibration Laboratories, ISO/IEC 17025 (International
Ielsch, G., Cushing, M. E., Combes, Ph., and Cuney, M. J. Organization of Standardization, Geneva).
(2010). “Mapping of the geogenic radon potential in ISO (2008). International Organization of Standardization.
France to improve radon risk management: method- Quality Management Systems—Requirements, ISO 9001
ology and first application to region Bourgogne,” (International Organization of Standardization,
J. Environ. Radioact. 101, 813– 820. Geneva).
Igarashi, G., Saeki, S., Takahata, N., Sumikawa, K., ISO (2010). International Organization of Standardization.
Tasaka, S., Sasaki, Y., Takahashi, M., and Sano, Y. Determination of the Characteristic Limits (Decision
(1995). “Ground-water radon anomaly before the Kobe Threshold, Detection Limit and Limits of the Confidence
earthquake in Japan,” Science 269, 60–61. Interval) for Measurements of Ionizing Radiation—
Iida, T., Ikebe, Y., and Tojo, K. (1991). “An electrostatic Fundamentals and Application, ISO 11929 (International
radon monitor for measurements of environmental Organization of Standardization, Geneva).
radon,” Res. Lett. Atmos. Electr. 11, 55– 59. ISO (2012a). International Organization of Standardization.
Iida, T., Ikebe, Y., Suzuki, K., Ueno, K., Wang, Z., and Jin, Measurement of Radioactivity in the Environment—Air:
Y. (1996). “Continuous measurements of outdoor radon Radon 222: Part 2: Integrated Measurement Method for
concentrations at various locations in East Asia,” Determining Average Potential Alpha Energy
Environ. Int. 22(Suppl. 1), 139– 147. Concentration of Its Short-Lived Decay Products, ISO
Iida, T., Rahman, N. M., Matsui, A., Yamazawa, H., and 11665-2 (International Organization of Standardization,
Moriizumi, J. (2008). “The measurement of size distribu- Geneva).
tion of indoor natural radioactive aerosols by imaging ISO (2012b). International Organization of Standardization.
plate technique,” pp. 137–140 in The Natural Radiation Measurement of Radioactivity in the Environment—Air:

178
 References

Radon 222: Part 3: Spot Measurement Method of the 88–98 in Biology, Toxicology, and Carcinogenesis of
Potential Alpha Energy Concentration of Its Short-Lived Respiratory Epithelium, Thomassen, D. G., and Nettesheim,
Decay Products, ISO 11665-3 (International Organization P., Eds. (Hemisphere Publishing Corporation, New York).
of Standardization, Geneva). Jokisalo, J., Kalamees, T., Kurnitski, J., Eskola, L.,
ISO (2012c). International Organization of Standardization. Jokiranta, K., and Vinha, J. (2008). “A comparison of mea-
Measurement of Radioactivity in the Environment—Air: sured and simulated air pressure conditions of a detached
Radon 222: Part 4: Integrated Measurement Method for house in a cold climate,” J. Build. Phys. 32, 67–89.
Determining Average Activity Concentration Using Kahn, A., Bandi, F., Gleasure, R. W., Phillips, C. R., and
Passive Sampling and Delayed Analysis, ISO 11665-4 Duport, P. (1987). “Underground measurements of
(International Organization of Standardization, Geneva). aerosol and activity distributions of radon and thoron
Iyogi, T., Ueda, S., Hisamatsu, S., Kondo, K., Sakurai, N., progeny,” pp. 219 –238 in Radon and Its Decay
and Inaba, J. (2003). “Radon concentration in indoor oc- Products: Occurrence, Properties, and Health Effects,
cupational environments in Aomori Prefecture, Japan,” Hopke, P. K., Eds. (American Chemical Society,
J. Environ. Radioact. 67, 91–108. Washington, DC).
Jacobi, W. (1964). “The dose to the human respiratory Kappel, R. J. A., Keller, G., Kreienbrock, L., and Nickels,
tract by inhalation of short-lived 222Rn and 220Rn decay R. M. (1993). “An epidemiological study using passive

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

products,” Health Phys. 10, 1163 –1175. radon measurement by liquid scintillation counting,”
Jacobi, W. (1972). “Activity and potential alpha-energy of pp. 319– 323 in Liquid Scintillation Spectrometry 1992,
222
Rn- and 220Rn-daughters in different air atmo- Noakes, J. E., Schonhofer, F., and Polach, H. A., Eds.
spheres,” Health Phys. 22, 441– 450. (Radiocarbon, Tucson, AZ).
Jacobi, W. (1984). “Possible lung cancer risk from indoor Karinda, F. L., Haider, B., and Rühm, W. (2008). “A new
exposure to radon daughters,” Radiat. Prot. Dosim. 7, electronic personal exposure meter for radon gas,”
395– 401. Radiat. Meas. 43, 1170– 1174.
Jacobi, W., and Eisfeld, K. (1980). Dose to Tissues and Kávási, N., Somlai, J., Vigh, T., Tokonami, S., Ishikawa, T.,
Effective Dose Equivalent by Inhalation of Radon-222, Sorimachi, A., and Kovács, T. (2009). “Difficulties in the
Radon-220 and Their Short-Lived Daughters, GSF dose estimate of workers originated from radon and
Report S-626 (Gesellschaft für Strahlen- und radon progeny in a manganese mine,” Radiat. Meas. 44,
Umweltforschung, München). 300– 305.
Jakupi, B., Krstić, G., Tonić, M., and Ilić, R. (1997). Kávási, N., Kobayashi, Y., Kovács, T., Somlai, J., Jobbágy,
“Radon in mines and dwellings in Kosovo and Metohia,” V., Nagy, K., Deák, E., Berhés, I., Bender, T., Ishikawa,
Radiat. Meas. 28, 691– 694. T., Tokonami, S., Vaupotič, J., Yoshinaga, S., and
James, A. C. (1988). “Lung dosimetry,” pp. 259 –309 in Yonehara, H. (2011). “Effect of radon measurement
Radon and Its Decay Products in Indoor Air, Nazaroff, methods on dose estimation,” Radiat. Prot. Dosim. 145,
W. W., and Nero, A. V., Eds. (Wiley Interscience, 224– 232.
New York). Keller, G., and Folkerts, K. H. (1984). “Radon-222 concen-
James, A. C., Greenhalgh, J. R., and Smith, H. (1977). trations and decay-product equilibrium in dwellings
“Clearance of lead-212 ions from rabbit bronchial epi- and in the open air,” Health Phys. 47, 385 –398.
thelium to blood,” Phys. Med. Biol. 22, 932– 948. Keller, G., Hoffmann, B., and Feigenspan, T. (2001).
James, A. C., Strong, J. C., Cliff, K. D., and Stranden, E. “Radon permeability and radon exhalation of building
(1988). “The significance of equilibrium and attachment materials,” Sci. Total Environ. 272, 85– 89.
in radon daughter dosimetry,” Radiat. Prot. Dosim. 24, Kemski, J., Klingel, R., Siehl, A., and Stegemann, R.
451– 455. (2005). “Radon transfer from ground to houses and pre-
James, A. C., Birchall, A., and Akabani, G. (2004). diction of indoor radon in Germany based on geological
“Comparative dosimetry of BEIR VI revisited,” Radiat. information,” pp. 820 –832 in The Natural Radiation
Prot. Dosim. 108, 3 –26. Environment VII, McLaughlin, J. P., Simopoulos, S. E.,
JCGM (2012). Joint Committee for Guides in Metrology. and Steinhäusler, F., Eds. (Elsevier Science, New York).
International Vocabulary of Metrology—Basic and Kenawy, M. A., and Morsy, A. A. (1991). “Measurements of
General Concepts and Associated Terms (VIM), 3rd ed., environmental radon-222 concentration in indoor and
JDGM 200 (International Organization of outdoors in Egypt,” Nucl. Tracks Radiat. Meas. 19,
Standardization, Geneva). 343– 346.
Jilek, K., Thomas, J., and Tomasek, L. (2010). “First Kendall, G. M., and Phipps, A. W. (2007). “Effective and
results of measurements of equilibrium factors, F and organ doses from thoron decay products at different
unattached fraction, fp of radon progeny in Czech dwell- ages,” J. Radiol. Prot. 27, 427 –435.
ings,” Third European IRPA Congress 14– 18 June Kendall, G. M., and Smith, T. J. (2002). “Doses to organs
2010, Helsinki, Finland (www.irpa2010europe.com). and tissues from radon and its decay products,”
Johansson, G. I., Samuelsson, C., and Pettersson, H. J. Radiol. Prot. 22, 389 –406.
(1984). “Characterisation of the aerosol and the activity Kendall, G. M., and Smith, T. J. (2005). “Doses from radon
size distribution of radon daughter in indoor air,” and its decay products to children,” J. Radiol. Prot. 25,
Radiat. Prot. Dosim. 7, 133 –137. 241– 256.
Johnson, N. F., Hubbs, A. F., and Thomassen, D. G. (1990). Kesten, J., Butterweck, G., Porstendörfer, J., Reineking,
“Epithelial progenitor cells in the rat respiratory tract,” pp. A., and Heymel, H. J. (1993). “An online a-impactor for

179
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

short-lived radon daughters,” Aerosol Sci. Technol. 18, Kranrod, C., Tokonami, S., Ishikawa, T., Sorimachi, A.,
156– 164. Janik, M., Shingaki, R., Furukawa, M., Chanyotha, S.,
Khan, A. J. (2000). “A study of indoor radon levels in and Chankow, N. (2009). “Mitgation of the effective dose
Indian dwellings, influencing factors and lung cancer of radon decay products through the use of an air
risks,” Radiat. Meas. 32, 87–92. cleaner in a dwelling in Okinawa, Japan,” Appl. Radiat.
Khan, A. J., Varshney, A. K., Prasad, R., Tyagi, R. K., and Isot. 67, 1127 –1132.
Ramachandran, T. V. (1990). “Calibration of a CR-39 Kreuzer, M., Walsh, L., Schnelzer, M., Tschense, A., and
plastic track detector for the measurement of radon and Grosche, B. (2008). “Radon and risk of extrapulmonary
its daughters in dwellings,” Nucl. Tracks Radiat. Meas. cancers: results of the German uranium miners’ cohort
17, 497– 502. study, 1960– 2003,” Br. J. Cancer 99, 1946–1953.
Khursheed, A. (2000). “Doses to systemic tissues from Krewski, D., Lubin, J. H., Zielinski, J. M., Alavanja, M.,
radon gas,” Radiat. Prot. Dosim. 88, 171– 181. Catalan, V. S., Field, R. W., Klotz, J. B., Letourneau, E.
Kies, A., Biell, A., Rowlinson, L., and Feider, M. (1996). G., Lynch, C. F., Lyon, J. I., Sandler, D. P., Schoenberg,
“Radon survey in the Grand-Duchy of Luxembourg— J. B., Steck, D. J., Stolwijk, J. A., Weinberg, C., and
indoor measurements related to house features, soil, Wilcox, H. B. (2005a). “Residential radon and risk of
geology and environment,” Environ. Int. 22(Suppl. 1), lung cancer: a combined analysis of 7 North American

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

805– 808. case– control studies,” Epidemiology 16, 137 –145.
Kim, C.-K., Lee, S.-C., Lee, D.-M., Chang, B.-U., Rho, Krewski, D., Mallick, R., Zielinski, J. M., and Letourneau,
B.-H., and Kang, H.-D. (2003). “Nationwide survey of E. G. (2005b). “Modeling seasonal variation in indoor
radon levels in Korea,” Health Phys. 84, 354– 360. radon concentrations,” J. Exp. Anal. Environ.
Knutson, E. O. (1999). “History of diffusion batteries in Epidemiol. 15, 234– 243.
aerosol measurements,” Aerosol Sci. Technol. 31, 83–128. Krewski, D., Lubin, J. H., Zielinski, J. M., Alavanja, M.,
Knutson, E. O., George, A. C., Knuth, R. H., and Koh, B. Catalan, V. S., Field, R. W., Klotz, J. B., Letourneau, E.
R. (1984). “Measurements of radon daughter particle G., Lynch, C. F., Lyon, J. L., Sandler, D. P., Schoenberg,
size,” Radiat. Prot. Dosim. 7, 121– 125. J. B., Steck, D. J., Stolwijk, J. A., Weinberg, C., and
Knutson, E. O., Tu, K. W., Solomon, S. B., and Strong, J. Wilcox, H. B. (2006). “A combined analysis of North
(1988). “Intercomparison of three diffusion batteries for American case– control studies of residential radon
the measurement of radon decay product particle size and lung cancer,” J. Toxicol. Environ. Health A 69,
distributions,” Radiat. Prot. Dosim. 24, 261– 264. 533– 597.
Knutson, E. O., George, A. C., Shebell, P., and Gogolak, C. Kulmala, M., Kontkanen, J., Junninen, H., Lehtipalo, K.,
V. (1994). “EML thoron gas measurements,” Radiat. Manninen, H. E., Nieminen, T., Petaja, T., Sipila, M.,
Prot. Dosim. 56, 263– 266. Schobesberger, S., Rantala, P., Franchin, A., Jokinen,
Kobal, I., Vaupotič, J., Udovč, H., Burger, J., and Stropnik, T., Jarvinen, E., Aijala, M., Kangasluoma, J., Hakala,
B. (1990). “Radon concentrations in the air of Slovene J., Aalto, P. P., Paasonen, P., Mikkila, J., Vanhanen, J.,
(Yugoslavia) underground mines,” Environ. Int. 16, Aalto, J., Hakola, H., Makkonen, U., Ruuskanen, T.,
171– 173. Mauldin, R. L., Duplissy, J., Vehkamaki, H., Back, J.,
Koblinger, L., and Hofmann, W. (1985). “Analysis of Kortelainen, A., Riipinen, I., Kurten, T., Johnston, M.
human lung morphometric data for stochastic aerosol V., Smith, J. N., Ehn, M., Mentel, T. F., Lehtinen, K. E.
deposition calculations,” Phys. Med. Biol. 30, 541– 556. J., Laaksonen, A., Kerminen, V-M., and Worsnop, D. R.
Koblinger, L., and Hofmann, W. (1988). “Stochastic mor- (2013). “Direct observation of atmospheric aerosol nu-
phological model of the rat lung,” Anat. Rec. 221, cleation,” Science 339, 943 –946.
533– 539. Kumara, P., Hopke, P. K., Raja, S., Casuccio, G., Lersch, T.
Koblinger, L., and Hofmann, W. (1990). “Monte Carlo mod- L., and West, R. R. (2012). “Characterizatiion and het-
elling of aerosol deposition in human lungs. Part I: erogeneity of coarse particles across an urban area,”
simulation of particle transport in a stochastic lung Atmos. Environ. 46, 449– 459.
structure,” J. Aerosol Sci. 21, 661 –674. Kusnetz, H. L. (1956). “Radon daughters in mine atmo-
Koblinger, L., Hofmann, W., Graham, R. C., and Mercer, R. spheres: a field method for determining concentrations,”
R. (1995). “Aerosol inhalation in the rat lung. Part I: ana- Am. Ind. Hyg. Assoc. Q. 17, 85–88.
lysis of the rat acinus morphometry and construction of a Labidi, S., Al-Azmi, D., Mahjoubi, H., and Ben Salah, R.
stochastic rat lung model,” J. Aerosol Med. 8, 7–19. (2010). “Radon in elementary schools in Tunisia,”
Kojima, H., and Abe, S. (1988). “Measurements of the total Radioprotection 45, 209 –217.
and unattached radon daughters in a house,” Radiat. Lagarde, F., Falk, R., Almren, K., Nyberg, F., Svensson,
Prot. Dosim. 24, 241– 244. H., and Pershagen, G. (2002). “Glass-based radon-
Kotrappa, P., Dempsey, J. C., Hickey, J. R., and Stieff, L. R. exposure assessment and lung cancer risk,” J. Exp.
(1988). “An electret passive environmental 222Rn Anal. Environ. Epidemiol. 12, 344– 354.
monitor based on ionization measurement,” Health Langroo, M. K., Wise, K. N., Duggleby, J. G., and Kotler,
Phys. 54, 47– 56. L. H. (1991). “A nationwide survey of 222Rn and gamma
Kotrappa, P., Dempsey, J. C., Ramsey, R. W., and Stieff, L. radiation levels in Australian homes,” Health Phys. 61,
R. (1990). “A practical E-PERMw (electret passive envir- 753– 761.
onmental radon monitor) system for indoor 222Rn meas- L’Annunziata, M. F., Ed. (2003). Handbook of Radioactivity
urement,” Health Phys. 58, 461 –467. Analysis, 2nd ed. (Academic Press, London).

180
 References

Lario, J., Sánchez-Moral, S., Cañaveras, J. C., Cuezva, S., Lubin, J. H., Wang, Z. Y., Boice, J.D., Jr., Xu, Z. Y., Blot, W.
and Soler, V. (2005). “Radon continuous monitoring in J., Wang, L. D., and Kleinerman, R. A. (2004). “Risk of
Altamira Cave (northern Spain) to assess user’s annual lung cancer and residential radon in China: pooled
effective dose,” J. Environ. Radioact. 80, 161– 174. results of two studies,” Int. J. Cancer 109, 132– 137.
Laurer, G. R., Estrada, J. J., and Cohen, N. (1999). “Lung Lucas, H. F. (1957). “Improved low-level alpha-scintillation
exposure from inhalation of radon progeny: calculated counter for radon,” Rev. Sci. Instrum. 28, 680–683.
from in vivo measurements of 210Pb in the skull,” Lundin, F. E., Wagoner, J. K., and Archer, V. E. (1971).
Health Phys. 76, 380 –387. Radon Daughter Exposure and Respiratory Cancer,
Lawrence, C. E., Akber, R. A., Bollhöfer, A., and Martin, P. Quantitative and Temporal Aspects: Report from the
(2009). “Radon-222 exhalation from open ground on and Epidemiological Study of United States Uranim
around a uranium mine in the wet-dry tropics,” Miners, Joint Monograph No. 1 (National Technical
J. Environ. Radioact. 100, 1–8. Information Service, Springfield, VA).
Letourneau, E. G., Zielinski, J. M., Krewski, D., and Maged, A. F. (2006). “Radon concentrations in elementary
McGregor, R. G. (1992). “Levels of radon gas in Winnipeg schools in Kuwait,” Health Phys. 90, 258 –262.
homes,” Radiat. Prot. Dosim. 45, 531–534. Mahaffy, J. A., Parkhurst, M. A., James, A. C., Cross, F. T.,
Lettner, H., Hubmer, A. K., Rolle, R., and Steinhäusler, F. Alavanja, M. C. R., Boice, J. D., Ezrine, S., Henderson,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

(1996). “Occupational exposure to radon in treatment P., and Brownson, R. C. (1993). “Estimating past expos-
facilities of the radon-spa Badgastein, Austria,” ure to indoor radon from household glass,” Health Phys.
Environ. Int. 22(Suppl. 1), 399– 407. 64, 381– 391.
Leuraud, K., Schmelzer, M., Tomasek, L., Hunter, N., Maher, E. F., and Laird, N. M. (1985). “EM algorithm re-
Tirmarche, M., Grosche, B., Kreuzer, M., and Laurier, construction of particle size distributions from diffusion
D. (2011). “Radon, smoking and lung cancer risk: battery data,” J. Aerosol Sci. 16, 557 –570.
Results of a joint analysis of three European case- Mäkeläinen, I. (1986). “Experiences with track etch detec-
control studies among uranium miners,” Radiat. Res. tors for radon measurements,” Nucl. Tracks Radiat.
176, 375– 387. Meas. 12, 717 –720.
Li, W., and Hopke, P. K. (1993). “Initial size distributions Mäkeläinen, I., Arvela, H., and Voutilainen, A. (2001).
and hygroscopicity of indoor combustion aerosol parti- “Correlations between radon concentration and indoor
cles,” Aerosol Sci. Technol. 19, 305 –316. gamma dose rate, soil permeability and dwelling substruc-
Li, W. B., Tschiersch, J., Oeh, U., and Hoeschen, Ch. (2010). ture and ventilation,” Sci. Total Environ. 272, 283–289.
“Lung dosimetry of inhaled thoron decay products,” in Mäkeläinen, I., Moisio, S., Reisbacka, H., and Turtiainen,
Proceedings of the 12th Congress of the International T. (2005). “Indoor occupancy and radon exposure in
Radiation Protection Associaton (IRPA-12) (International Finland,” pp. 687– 693 in Proceedings of the Seventh
Atomic Energy Agency, Vienna). International Symposium on the Natural Radiation
Linzmaier, D. (2012). Entwicklung einer Low-Level- Environment. Radioactivity in the Environment, Vol. 7
Radon-Referenzkammer, PhD thesis (Gottfried Wilhelm (Elsevier, NY).
Leibniz Universität, Hannover, Germany). Mäkeläinen, I., Kinnunen, T., Reisbacka, H., Valmari, T.,
Lively, R. S., and Ney, E. P. (1987). “Surface radioactivity and Arvela, H. (2009). Radon in Finnish Dwellings—
resulting from the deposition of 222Rn daughter pro- Sample Survey 2006, STUK-A242 (Radiation and Nuclear
ducts,” Health Phys. 52, 411–415. Safety Authority, Helsinki) (abstract in English).
Lively, R. S., and Steck, D. J. (1993). “Long-term radon Malanca, A., and Gaidolfi, L. (1997). “Environmental
concentrations estimated from 210Po embedded in radon in some Brazilian towns and mines,” Radiat.
glass,” Health Phys. 64, 485 –490. Prot. Dosim. 69, 211 –216.
Long, S., and Fenton, D. (2011). “An overview of Ireland’s Marcinowski, F., Lucas, R. M., and Yeager, W. M. (1994).
National Radon Policy,” Radiat. Prot. Dosim. 145, “National and regional distributions of airborne radon
96–100. concentrations in U.S. homes,” Health Phys. 66,
López, F. O., and Canoba, A. C. (2003). “Passive method 699– 706.
for the determination of the equilibrium factor between Markovic, V. M., Stevanovic, N., and Nikezic, D. (2011).
222
Rn gas and its short period progeny,” J. Radioanal. “Doses from beta radiation in sensitive layers of human
Nucl. Chem. 258, 269 –274. lung and dose conversion factors due to 222Rn/220Rn
Lubin, J. H., Tomasek, L., Edling, C., Hornung, R. W., progeny,” Radiat. Environ. Biophys. 50, 431 –440.
Howe, G., Kunz, E., Kusiak, R. A., Morrison, H. I., Marques, A. L., dos Santos, W., and Geraldo, L. P. (2004).
Radford, E. P., Samet, J. M., Tirmarche, M., Woodward, “Direct measurement of radon activity in water from
A., and Yao, S. X. (1997). “Estimating lung cancer mor- various natural sources using nuclear track detectors,”
tality from residential radon using data for low expo- Appl. Radiat. Isot. 60, 801 –804.
sures of miners,” Radiat. Res. 147, 126 –134. Marsh, J. W., and Bailey, M. R. (2013). “A review of
Lubin, J. H., Linet, M. S., Boice, J.D., Jr., Buckley, J., lung-to-blood absorption rates for radon progeny,”
Conrath, S. M., Hatch, E. E., Kleinerman, R. A., Tarone, Radiat. Prot. Dosim. 157, 499– 514.
R. E., Wacholder, S., and Robison, L. L. (1998). “Case – Marsh, J. W., and Birchall, A. (1998). Sensitivity Analysis
control study of childhood acute lymphoblastic leukemia of the Weighted Equivalent Lung Dose per Unit
and residential radon exposure,” J. Natl. Cancer Inst. Exposure from Radon Progeny, NRPB-M929 (Health
90, 294– 300. Protection Agency, Chilton, Didcot, UK).

181
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Marsh, J. W., and Birchall, A. (1999a). “Letter to the Ménache, M. G., Hofmann, W., Asgharian, B., and Miller,
editor—The thoron issue: monitoring activities, meas- F. J. (2008). “Airway geometry models of children’s
uring techniques and dose conversion factors,” Radiat. lungs for use in dosimetry modeling,” Inhal. Toxicol. 20,
Prot. Dosim. 81, 311 –312. 101– 126.
Marsh, J. W., and Birchall, A. (1999b). “Determination of Menzler, S., Piller, G., Gruson, M., Rosario, A. S.,
lung-to-blood absorption rates for lead and bismuth Wichmann, H.-E., and Kreienbrock, L. (2008).
which are appropriate for radon progeny,” Radiat. Prot. “Population attributable fraction for lung cancer due to
Dosim. 83, 331 –337. residential radon in Switzerland and Germany,” Health
Marsh, J. W., and Birchall, A. (2000). “Sensitivity Phys. 95, 179–189.
analysis of the weighted equivalent lung dose per unit Mercer, T. T. (1976). “The effect of particle size on the
exposure from radon progeny,” Radiat. Prot. Dosim. 87, escape of the recoiling RaB atoms from particulate sur-
167– 178. faces,” Health Phys. 31, 173– 175.
Marsh, J. W., Birchall, A., Butterweck, G., Dorrian, M.-D., Mercer, R. R., and Crapo, J. D. (1989). “Anatomical model-
Huet, C., Ortega, X., Reineking, A., Tymen, G., Schuler, ing of microdosimetry of inhaled particles and gases,”
Ch., Vargas, A., Vezzu, G., and Wendt, J. (2002). pp. 69–78 in Extrapolation of Dosimetric Relationships
“Uncertainty analysis of the weighted equivalent lung for Inhaled Particles and Gases, Crapo, J., Smolko, E.,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

dose per unit exposure to radon progeny in the home,” Miller, F., Graham, J., and Hayes, A., Eds. (Academic
Radiat. Prot. Dosim. 102, 229– 248. Press, San Diego, CA).
Marsh, J. W., Birchall, A., and Davis, K. (2005). Mercer, R. R., Russell, M. L., and Crapo, J. D. (1991). “Radon
“Comparative dosimetry in homes and mines: estimation dosimetry based on the depth distribution of nuclei in
of K-factors,” pp. 290–298 in The Natural Radiation human and rat lungs,” Health Phys. 61, 117–130.
Environment VII, McLaughlin, J.P., Simopoulos, S.E., Michielsen, N., and Tymen, G (2007). “Semi-continuous
Steinhäusler F., Eds. (Elsevier Science, NY). measurements of the unattached radon decay products
Marsh, J. W., Bessa, Y., Birchall, A., Blanchardon, E., size distributions from 0.5 to 5 nm by an array of
Hofmann, W., Nosske, D., and Tomasek, L. (2008). annular diffusion channels,” J. Aerosol Sci. 38,
“Dosimetric models used in the Alpha-Risk Project to 1129 –1139.
quantify exposure of uranium miners to radon gas and Michielsen, N., Voisin, V., and Tymen, G. (2005).
its progeny,” Radiat. Prot. Dosim. 130, 101– 106. “Measurement of the unattached radon decay products
Marsh, J. W., Blanchardon, E., Gregoratto, D., Hofmann, W., with an annular diffusion channel battery,” pp.
Karcher, K., Nosske, D., and Tomasek, L. (2012). 339– 345 in The Natural Radiation Environment VII,
“Dosimetric calculations for uranium miners for epidemio- McLaughlin, J. P., Simopoulos, S. E., and Steinhäusler,
logical studies,” Radiat. Prot. Dosim. 149, 371–383. F., Eds. (Elsevier Science, New York).
Martinez, T. A., Lartigue, J. A., Navarrete, M. A., Cabrera, Miles, J. C. H. (1994). “Mapping the proportion of the
L. A., Gonzalez, P. B., Ramirez, A. A., and Elizarraras, housing stock exceeding a radon reference level,”
V. A. (1998). “Long term and equilibrium factor indoor Radiat. Prot. Dosim. 56, 207– 210.
radon measurements,” J. Radioanal. Nucl. Chem. 236, Miles, J. (1998). “Mapping radon-prone areas by log-
231– 237. normal modeling of house radon data,” Health Phys. 74,
Martonen, T. B., Graham, R. C., and Hofmann, W. (1989). 370– 378.
“Human subject age and activity level: factors Miles, J. (2001). “Temporal variation of radon levels in
addressed in a biomathematical deposition program for houses and implications for radon measurement strat-
extrapolation modeling,” Health Phys. 57(Suppl 1), egies,” Radiat. Prot. Dosim. 93, 369– 375.
49–59. Miles, J. C. H. (2002). “Use of a model data set to test
Mazur, D., Janik, M., Łoskiewicz, J., Olko, P., and methods for mapping radon potential.” Radiat. Prot.
Swakoń, J. (1999). “Measurements of radon concentra- Dosim. 98, 211 –218.
tion in soil gas by CR-39 detectors,” Radiat. Meas. 31, Miles, J. C. H., and Appleton, J. D. (2005). “Mapping vari-
295– 300. ation in radon potential both between and within geo-
McDowell, E. M., Newkirk, C., and Coleman, B. (1985). logical units,” J. Radiol. Prot. 25, 257– 276.
“Development of hamster tracheal epithelium. II. Cell Miles, J. C. H., Kendall, G. M., Ibrahimi, Z.-F., and
proliferation in the fetus,” Anat. Rec. 213, 448 –456. Howarth, C. B. (2004). “Practical procedures for a radon
McLaughlin, J. P. (1972). “The attachment of radon daugh- etched track dosimetry service,” J. Radiol. Prot. 24,
ter products to condensation nuclei,” Proc. Roy. Irish 165– 171.
Acad. 72, 51–70. Miles, J. C. H., Howarth, C. B., and Hunter, N. (2012).
McLaughlin, J. P. (1998). “The application of techniques to “Seasonal variation of radon concentrations in UK
assess radon exposure retrospectively,” Radiat. Prot. homes,” J. Radiol. Prot. 32, 275– 287.
Dosim. 78, 1 –6. Misdaq, M. A. (2003). “Radon, thoron and their progenies
McLaughlin, J. (2010). “An overview of thoron and its measured in different dwelling rooms and reference
progeny in the indoor environment,” Radiat. Prot. atmospheres by using CR-39 and LR-115 SSNTD,”
Dosim. 141, 316 –321. Appl. Radiat. Isot. 59, 273 –280.
Mellander, H., and Enflo, A. (1992). “The alpha track Misdaq, M. A., and Amghar, A. (2005). “Radon and thoron
method used in the Swedish Radon Epidemiological emanation from various marble materials: Impact on
Study,” Radiat. Prot. Dosim. 45, 65–71. the workers,” Radiat. Meas. 39, 421 –430.

182
 References

Misdaq, M. A., and Flata, K. (2003). “The influence of the NA/NRC (1991). National Academies/National Research
cigarette smoke pollution and ventilation rate on Council. Comparative Dosimetry of Radon in Mines and
alpha-activities per unit volume due to radon and its Homes (National Academies Press, Washington, DC).
progeny,” J. Environ. Radioact. 67, 207– 218. NA/NRC (1999a). National Academies/National Research
Misdaq, M. A., and Ouguidi, J. (2008). “Radon, thoron and Council. Health Effects of Exposure to Radon, BEIR VI
their decay products in natural caves with nuclear track (National Academies Press, Washington, DC).
detectors,” Radiat. Prot. Dosim. 130, 110– 114. NA/NRC (1999b). National Academies/National Research
Mishra, U. C., and Subba Ramu, M. C. (1988). “Natural Council. Risk Assessment of Radon in Drinking Water
radioactivity in houses and mine atmospheres in India,” (National Academies Press, Washington, DC).
Radiat. Prot. Dosim. 24, 25–28. NCRP (1984). National Council on Radiation Protection
Mohammed, A. (1999). “Activity size distributions of short and Measurement. Evaluation of Occupational and
lived radon progeny in indoor air,” Radiat. Prot. Dosim. Environmental Exposures to Radon and Radon
86, 139– 145. Daughters in the United States, NCRP Report No. 78
Mohammed, A. (2005). “Study on radon and radon (National Council on Radiation Protection and
progeny in some living rooms,” Radiat. Prot. Dosim. Measurement, Bethesda, MD).
117, 402 –407. NCRP (1988). National Council on Radiation Protection

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Monchaux, G. (2004). “Risk of fatal versus incidental lung and Measurement. Measurement of Radon and Radon
cancer in radon-exposed rats: a reanalysis of French Daughters in Air, NCRP Report No. 97 (National
data,” Arch. Oncol. 12, 7 –12. Council on Radiation Protection and Measurement,
Monchaux, G. (2005). “Contribution of animal experimen- Bethesda, MD).
tal data for risk assessment of exposure to radon decay NCRP (2009). National Council on Radiation Protection
products,” pp. 66–76 in The Natural Radiation and Measurement. Uncertainties in Internal Radiation
Environment VII, McLaughlin, J. P., Simopoulos, S. E., Dose Assessment, NCRP Report No. 164 (National
and Steinhäusler, F., Eds. (Elsevier Science, New York). Council on Radiation Protection and Measurement,
Monchaux, G., and Morlier, J. P. (2002). “Influence of ex- Bethesda, MD).
posure rate on radon-induced lung cancer in rats,” Nazaroff, W. W. (1992). “Radon transport from soil to air,”
J. Radiol. Prot. 22, A81– A87. Rev. Geophys. 30, 137 –160.
Monchaux, G., Morlier, J. P., Morin, M., Chameaud, J., Nelder, J. A., and Mead, R. (1965). “A Simplex method for
Lafuma, J., and Masse, R. (1994). “Carcinogenic and function minimisation,” Comput. J. 7, 308 –313.
cocarcinogenic effects of radon and radon daughters in Nero, A. V., Schwehr, M. B., Nazaroff, W. W., and Revzan,
rats,” Environ. Health Perspect. 102, 64– 73. K. L. (1986). “Distribution of airborne radon-222 con-
Monchaux, G., Morlier, J. P., Altmeyer, S., Debroche, M., centrations in US homes,” Science 234, 992– 997.
and Morin, M. (1999). “Influence of exposure rate on Nero, A. V., Leiden, S. M., Nolan, D. A., Price, P. N., Rein,
lung cancer induction in rats exposed to radon progeny,” S., Revzan, K. L., Woolenberg, H. R., and Gadgil, A. J.
Radiat. Res. 152, S137 –S140. (1994). “Statistically based methodologies for mapping
Morawska, L. (1989). “Two ways of determining the 222Rn of radon ‘actual’ concentrations: the case of Minnesota,”
emanation coefficient,” Health Phys. 57, 481 –483. Radiat. Prot. Dosim. 56, 215– 219.
Moreno, V., Baixeras, C., Font, L. I., and Bach, J. (2008). Neznal, M., Matolı́n, M., Just, G., and Turek, K. (2004a).
“Indoor radon levels and their dynamics in relation “Short-term temporal variations of soil gas radon con-
with the geological characteristics of La Garrotxa, centration and comparison of measurement techni-
Spain,” Radiat. Meas. 43, 1532 –1540. ques,” Radiat. Prot. Dosim. 108, 55– 63.
Morken, D. A. (1973). “Biological effects of radon on the Neznal, M., Neznal, M., Matolı́n, M., Barnet, I., and
lung,” pp. 501 in Noble Gases, Stanley, R. E., and Mikšová, J. (2004b). The New Method for Assessing the
Moghissi, A. A., Eds., CONF-730915 (National Radon Risk of Building Sites, CGS Special Papers 16
Technical Information Service, Springfield, VA). (Czech Geiological Survey, Prague).
Morken, D. A., and Scott, J. K. (1966). Effects on Mice of Nikezic, D., and Yu, K. N. (2004). “Formation and growth
Continual Exposure to Radon and Its Decay Products of tracks in nuclear track material,” Mater. Sci. Eng. R
on Dust, UR-669 (National Technical Information 46, 51–123.
Service, Springfield, VA). Nikiforov, A. I., and Schlesinger, R. B. (1985).
Mortensen, J. D., Schaap, R. N., Bagley, B., Stout, L., “Morphometric variability of the human upper bron-
Young, J. D., Stout, A., Burkart, J. A., and Baker, C. D. chial tree,” Respir. Physiol. 59, 289– 299.
(1983). A Study of Age Specific Human Respiratory NIOSH (1971). National Institute for Occupational Safety
Morphometry: Final Report, TR 01525-010 (University and Health. Radon Daughter Exposure and Respiratory
of Utah Research Institute, Salt Lake City, UT). Cancer Quantitative and Temporal Aspects, Lundin,
Mortensen, J. D., Stout, L., Bagley, B., Burkart, J. A., and F.F., Jr., Wagoner, J. K., and Archer, V. E., Eds.
Schaap, R. N. (1989). “Age related morphometric ana- (National Technical Information Service, Springfield,
lysis of human lung casts,” pp. 50– 68 in Extrapolation VA).
of Dosimetric Relationships for Inhaled Particles and Nsibande, M. C., Mahlobo, M., and Farid, S. M. (1994).
Gases, Crapo, J., Smolko, E., Miller, F., Graham, J., and “Radon levels inside residences in Swaziland,” Sci. Tot.
Hayes, A., Eds. (Academic Press, San Diego, CA). Environ. 151, 181 –185.

183
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Nuccetelli, C., and Bochicchio, F. (1998). “The thoron progenies at the German radon reference chamber,” Int.
issue: monitoring activities, measuring techniques and Congr. Ser. 1225, 161 –167.
dose conversion factors,” Radiat. Prot. Dosim. 78, Peterman, B. F., and Perkins, C. J. (1988). “Dynamics of
59–64. radioactive chemically inert gases in the human body,”
Nussbaum, E., and Hursh, J. B. (1957). “Radon solubility Radiat. Prot. Dosim. 22, 5– 12.
in rat tissue,” Science 125, 552– 553. Phalen, R. F., and Oldham, M. J. (2001). “Methods for
Nussbaum, E., and Hursh, J. B. (1958). “Radon solubility in modeling particle deposition as a function of age,”
fatty acids and triglycerides,” J. Phys. Chem. 62, 81–84. Respir. Physiol. 128, 119–130.
Oberstedt, S., and Vanmarcke, H. (1996). “Volume traps— Phalen, R. F., Oldham, M. J., Beaucage, C. B., Crocker, T.
a new retrospective radon monitor,” Health Phys. 70, T., and Mortensen, J. D. (1985). “Postnatal enlargement
222– 226. of human tracheobronchial airways and implications for
OECD (1985). Organisation for Economic Co-operation particle deposition,” Anat. Rec. 212, 368 –380.
and Development. Metrology and Monitoring of Radon, Phillips, J. D., Duval, J. S., and Ambroziak, R. A. (1993).
Thoron and Their Daughter Products (OECD National Geophysical Data Grids: Gamma-Ray,
Publications and Information Center, Paris). Gravity, Magnetic, and Topographic Data for the
Oikawa, S., Kanno, N., Sanada, T., Abukawa, J., and Conterminous United States (US Geological Survey,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Higuchi, H. (2006). “A survey of indoor workplace radon Denver, CO).
concentration in Japan,” J. Environ. Radioact. 87, Picolo, J. L. (1996). “Absolute measurement of radon 222
239– 245. activity,” Nucl. Instr. Meth. Phys. Res. A 369, 452– 457.
Olfert, J. S., Symonds, J. P. R., and Collings, N. (2007). Pinel, J., Fearn, T., Darby, S. C., and Miles, J. C. H. (1995).
“The effective density and fractal dimension of particles “Seasonal correction factors for indoor radon measure-
emitted from a light-duty vehicle with a diesel oxidation ments in the United Kingdom,” Radiat. Prot. Dosim. 58,
catalyst,” J. Aerosol Sci. 38, 69– 82. 127– 132.
Page, D., and Smith, D. M. (1992). “The distribution of Planinic, J., Faj, Z., Radolic, V., Smit, G., and Faj, D.
radon and its decay products in some UK coal mines,” (1999). “Indoor radon dose assessment for Osijek,”
Radiat. Prot. Dosim. 45, 163– 166. J. Environ. Radioact. 44, 97–106.
Pahapill, L., Rulkov, A., and Swedjemark, G. A. (1996). Pohl, E., and Pohl-Rüling, J. (1976). “Determination of en-
Radon in Estonian Buildings. Establishment of a vironmental or occupational 222Rn in air and water and
226
Measurement System and Obtained Results, SSI-rapport Ra in water with feasible and rapid methods of sam-
96:13 (Swedish Radiation Protection Institute, pling and measurement,” Health Phys. 31, 343– 348.
Stockholm). Pohl, E., and Pohl-Rüling, J. (1977). “Dose calculations
Papastefanou, C. (2007). “Measuring radon in soil gas and due to inhalation of 222Rn, 220Rn and their daughters,”
groundwaters: a review,” Ann. Geophys. 50, 569 –578. Health Phys. 32, 552 –555.
Papastefanou, C., Manolopoulou, M., Stoulos, S., Porstendörfer, J. (1994). “Properties and behaviour or
Ioannidou, A., and Gerasopoulos, E. (2005). “Elevated radon and thoron and their decay products in the air,”
radon concentrations in a Pleistocenic cave operating as J. Aerosol Sci. 25, 219 –263.
a show cave,” Int. Congr. Ser. 1276, 204 –205. Porstendörfer, J. (1996). Radon: measurements related to
Paridaens, J., and Vanmarcke, H. (1999). “The usability of dose,” Environ. Int. 22(Suppl. 1), 563 –583.
wood as a volume trap for the purpose of retrospective Porstendörfer, J. (2001). “Physical parameters and dose
radon exposure assessment,” Health Phys. 76, 657– 663. factors of the radon and thoron decay products,” Radiat.
Park, K., Cao, F., Kittelson, D. B., and McMurray, P. H. Prot. Dosim. 94, 365– 373.
(2003). “Relationship between particle mass and mobil- Porstendörfer, J. (2002). “Influence of physical parameters
ity for diesel exhaust particles,” Environ. Sci. Technol. on doses from radon exposures,” Int. Congr. Ser. 1225,
37, 577– 583. 149– 160.
Passo, C.J., Jr., and Floeckher, J. M. (1991). “The LSC ap- Porstendörfer, J., and Reineking, A. (1999). “Radon: char-
proach to radon counting in air and water,” pp. 375– 384 acteristics in air and dose conversion factors,” Health
in Liquid Scintillation Counting and Organic Phys. 76, 300 –305.
Scintillators, Ross, H., Noakes, J. E., and Spaulding, J. Porstendörfer, J., Pagelkopf, P., and Gründel, M. (2005).
D., Eds. (CRC Press, Boca Raton, FL). “Fraction of the positive 218Po and 214Pb clusters in
Paul, A., Röttger, S., Honig, A., Sulima, T., Buchholz, A., indoor air,” Radiat. Prot. Dosim. 113, 342 –351.
and Keyser, U. (1999). “Measurement of short-lived Pressyanov, D. S. (2012). “Retrospective measurements of
radon progenies by simultaneous ag-spectrometry at thoron and radon by CDs/DVDs: a model approach,”
the German radon reference chamber,” Nucl. Instr. Radiat. Prot. Dosim. 149, 464– 468.
Meth. A 434, 303– 312. Pressyanov, D., Buysse, J., Van Deynse, A., Poffijn, A., and
Paul, A., Honig, A., Röttger, S., and Keyser, U. (2000). Meesen, G. (2001). “Indoor radon detected by compact
“Measurement of radon and radon progenies at the discs,” Nucl. Instr. Meth. A 457, 665 –666.
German radon reference chamber,” Appl. Radiat. Isot. Preston, D. L., Ron, E., Tokuoka, S., Funamoto, S., Nishi,
52, 369– 375. N., Soda, M., Mabuchi, K., and Kodama, K. (2007).
Paul, A., Honig, A., Röttger, S., and Keyser, U. (2002). “Solid cancer incoidence in atomic bomb survivors:
“Metrology of the activity concentration of radon and its 1958–1998,” Radiat. Res. 168, 1 –64.

184
 References

Price, P. N., and Nero, A. V. (1996). “Joint analysis of long- Reineking, A., and Porstendörfer, J. (1986). “High-volume
and short-term radon monitoring data from the nor- screen diffusion batteries and a-spectrometry for meas-
thern U.S.,” Environ. Int. 22(Suppl. 1), 699 –714. urement of the radon daughter activity size distribu-
Prichard, H. M., Venso, E. A., and Dodson, C. L. (1992). tions in the environment,” J. Aerosol Sci. 17, 873 –879.
“Liquid scintillation analysis of 222Rn in water by alpha- Reineking, A., and Porstendörfer, J. (1990). “‘Unattached’
beta discrimination,” Radioact. Radiochem. 3, 28–36. fraction of short-lived Rn decay products in indoor and
PTB (2011). Physikalisch-Technische Bundesanstalt. PTB outdoor environments: an improved single-screen
Jahresbericht 2011 (PTB, Braunschweig, Germany). method and results,” Health Phys. 58, 715– 727.
Qureshi, A. A., Kakar, D. M., Akram, M., Khattak, N. U., Reineking, A., Scheibel, H. G., Hussain, A., Becker, K. H.,
Tufail, M., Mehmood, K., Jamil, K., and Khan, H. A. and Porstendörfer, J. (1984). “Measurements of stage ef-
(2000). “Radon concentrations in coal mines of ficiency functions including interstage losses for a
Baluchistan, Pakistan,” J. Environ. Radioact. 48, Sierra and a Berner impactor and evaluation of data by
203– 209. a modified Simplex method,” J. Aerosol Sci. 15,
Raabe, O. G. (1969). “Concerning the interactions that 376– 380.
occur between radon decay products and aerosols,” Reineking, A., Becker, K. H., and Porstendörfer, J. (1988).
Health Phys. 17, 177 –185. “Measurements of activity size distributions of the

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Raabe, O. G., Yeh, H. C., Schum, G. M., and Phalen, R. F. short-lived radon daughters in the indoor and outdoor
(1976). Tracheobronchial Geometry: Human, Dog, Rat, environment,” Radiat. Prot. Dosim. 24, 245 –250.
Hamster, LF-53 (US Government Printing Office, Reineking, A., Butterweck, G., Kesten, J., and
Washington, DC). Porstendörfer, J. (1992a). “Thoron gas concentration and
Raaschou-Nielsen, O., Andersen, C. E., Andersen, H. P., aerosol characteristics of thoron decay products,” Radiat.
Gravesen, P., Lind, M., Schüz, J., and Ulbak, K. (2008). Prot. Dosim. 45, 353–356.
“Domestic radon and childhood cancer in Denmark,” Reineking, A., Butterweck, G., Porstendörfer, J., Strong,
Epidemiology 19, 536– 543. J. C., Vanmarcke, H., and Van Dingenen, R. (1992b).
Rad Elec (1999). Rad Elec, Inc. E-PERM System Manual “Intercomparison of methods for investigating the phys-
(Rad Elec, Inc., Frederick, MD). ical characteristics of radon decay products in the
Radolić, V., Vuković, B., Šmit, G., Stanić, D., and Planinić, indoor environment,” Radiat. Prot. Dosim. 45, 41– 46.
J. (2005). “Radon in the spas of Croatia,” J. Environ. Reineking, A., Knutson, E. A., George, A. C., Solomon, S.
Radioact. 83, 191 –198. B., Kesten, J., Butterweck, G., and Porstendörfer, J.
Ramachandran, T. V., and Sathish, L. A. (2011). (1994). “Size distribution of unattached and aerosol-
“Nationwide indoor 222Rn and 220Rn map for India: a attached short-lived radon decay products: some results
review,” J. Environ. Radioact. 102, 975– 986. of intercomparison measurements,” Radiat. Prot. Dosim.
Ramachandran, T. V., and Subba Ramu, M. C. (1994). 56, 113–118.
“Variation of equilibrium factor F between radon and its Reisbacka, H. (2008). “Radon measurments in convention-
short-lived decay products in an indoor atmosphere,” al Finnish workplaces during measurement seasons
Nucl. Geophys. 8, 499 –503. 2005–2007,” pp. 130 –134 in Nordic Society for
Ramamurthi, M., and Hopke, P. K. (1991). “An automated Radiation Protection, Proceedings of the NSFS XV
semicontinuous system for measuring indoor radon Conference, Stralevern Rapport 2008:13 (Norvegian
progeny activity-weighted size distributions, dp: 0.5 – Radiation Protection Authority, Osteras).
500 nm,” Aerosol Sci. Technol. 14, 82–92. Rericha, V., Kulich, M., Rericha, R., Shore, D. L., and
Ramola, R. C. (2005). “Levels of indoor radon, thoron, and Sandler, D. P. (2006). “Incidence of leukemia, lymph-
their progeny in Himalaya,” Int. Congr. Ser. 1276, oma, and multiple myeloma in Czech uranium miners:
215– 216. a case-cohort study,” Environ. Health Perspect. 114,
Ramola, R. C., Kandari, M. S., Negi, M. S., and Choubey, 818– 822.
V. M. (2000). “A study of diurnal variation of indoor Revzan, K. L., and Fisk, W. J. (1992). “Modeling radon
radon concentrations,” Hoken Butsuri 35, 211– 216. entry into houses with basements: the influence of
Reddy, K. V. K., Reddy, M. S., Reddy, C. G., Reddy, P. Y., structural factors,” Indoor Air 2, 40–48.
and Reddy, K. R. (2012). “Spatial and vertical distribu- Richardson, R. B., Eatough, J. P., and Henshaw, D. L.
tion of radon and thoron in a typical Indian dwelling,” (1991). “Dose to red bone marrow from natural radon
J.Radioanal. Nucl. Chem. 292, 1089– 1092. and thoron exposure,” Br. J. Radiol. 64, 608 –624.
Reichelt, A. (2002). Untersuchungen zur Strahlensituation Riley, W. J., Gadgil, A. J., Bonnefous, Y. C., and Nazaroff, W.
durch Radon und Radonfolgeprodukte an Arbeitsplätzen W. (1996). “The effect of steady winds on radon-222 entry
in Gebäuden. Abschlussbericht über das from soil into houses,” Atmos. Environ. 30, 1167–1176.
Forschungsvorhaben StSch4086 im Auftrag des Riley, W. J., Robinson, A. L., Gadgil, A. J., and Nazaroff, W.
Bundesministers für Umwelt, Naturschutz und W. (1999). “Effects of variable wind speed and direction
Reaktorsicherheit, TÜV Süddeutschland. on radon transport from soil into buildings: model devel-
Reichelt, A., Reineking, A., Lehmann, K. H., opment and exploratory results,” Atmos. Environ. 33,
Porstendörfer, J., Schwedt, J., and Streil, T. (2000). 2157–2168.
“Radon at workplaces,” P-1b-51, pp. 1– 9 in Proceedings Robbins, E. S., Meyers, O. A., and Harley, N. H. (1990).
of the 10th International Congress of the International “Quantification of the nuclei of human bronchial epithe-
Radiation Protection Association, IRPA-10. lial cells from electron micrographs for radon risk

185
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

analysis,” in Proceedings of the XIIth International case of the tourist caves,” J. Hazard. Mat. 145,
Congress for Electron Microscopy (San Francisco Press, 368– 371.
San Francisco, CA). Sajó-Bohus, L., Greaves, E. D., Pálfalvi, J., Urbani, F., and
Roos, B., and Samuelsson, C. (2005). “The behaviour of Merlo, G. (1997). “Radon concentration measurements
Rn-222 decay products at the air-glass interface and its in Venezuelan caves using SSNTDS,” Radiat. Meas. 28,
implications for retrospective radon exposure esti- 725– 728.
mates,” pp. 813– 819 in Natural Radiation Enviroment Sakashita, T., Doi, M., Nakamura, Y., and Iida, T. (2004).
VII, McLaughlin, J. P., Simopoulos, S. E., and “A case study of radon-222 transport from continental
Steinhäusler, F., Eds. (Elsevier Science, Oxford). North-East Asia to the Japanese islands in winter by
Roos, B., and Whitlow, H. J. (2003). “Computer simulation numerical analysis,” J. Environ. Radioact. 72,
and experimental studies of implanted 210Po in glass 245– 257.
resulting from radon exposure,” Health Phys. 84, Sakoda, A., Ishimori, Y., and Yamaoka, K. (2011). “A com-
72–81. prehensive review of radon emanation measurements
Rose, A. R., Ciokosz, E. J., and Washington, J. W. (1990). for mineral, rock, soil, mill tailing and ash,” Appl.
“Effects of regional and seasonal variations in soil mois- Radiat. Isot. 69, 1422– 1435.
ture and temperature on soil gas transport,” paper no. Sakoda, A., Ishimori, Y., Fukao, K., Yamaoka, K.,

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

C-VI-5 in International Symposium on Radon and Kataoka, T., and Mitsunobu, F. (2012). “Lung dosimetry
Radon Reduction Technology, Vol. III. of inhaled radon progeny in mice,” Radiat. Environ.
Röttger, A., Honig, A., Schmidt, V., Buchröder, H., Rox, A., Biophys. 51, 425 –442.
Butterweck, G., Schuler, Ch., Maringer, F. J., Sakoda, A., Ishimori, Y., Yamaoka, K., Kataoka, T., and
Michielsen, N., Howarth, C., Miles, J. C. H., Vargas, A., Mitsunobu, F. (2013). “Absorbed doses of lungs retained
Ortega, X., Burian, I., Turtiainen, T., Jachs, P., in airway lumens of mice and rats,” Radiat. Environ.
Edelmaier, R., and Hagberg, N. (2005). “Comparison of Biophys. 52, 389 –395.
calibration facilities for radon activity concentration: Salmon, P. L., Berkovsky, V. I., and Henshaw, D. L. (1998).
Euromet Project 657,” Metrologia 42, doi:10.1088/ “Relative importance of inhalation and ingestion as
0026-1394/42/1A/06003. sources of uptake of 210Pb from the environment,”
Röttger, A., Honig, A., Schmidt, V., Buchröder, H., Rox, A., Radiat. Prot. Dosim. 78, 279– 293.
Butterweck, G., Schuler, Ch., Maringer, F. J., Jachs, P., Salonen, L. (2010). “Comparison of two direct LS methods
Edelmaier, R., Michielsen, N., Howarth, C. B., Miles, J. for measuring 222Rn in drinking water using a/b liquid
C. H., Vargas, A., Ortega, X., Burian, I., Turtiainen, T., scintillation spectrometry,” Appl. Radiat. Isot. 68,
and Hagberg, N. (2006). “Radon activity concentra- 1970–1979.
tion—a Euromet and BIPM supplementary compari- Samuelsson, C. (1988). “Retrospective determination of
son,” Appl. Radiat. Isot. 64, 1102 –1107. radon in houses,” Nature 334, 338 –340.
Röttger, A., Honig, A., and Arnold, D. (2009). “The Samuelsson, C., and Johansson, L. (1994). “Long-lived
German thoron progeny chamber—concept and applica- radon decay products as a long-term radon exposure in-
tion,” Appl. Radiat. Isot. 67, 839 –842. dicator,” Radiat. Prot. Dosim. 56, 123– 126.
Röttger, A., Honig, A., Dersch, R., Ott, O., and Arnold, D. Sanada, T., Fujimoto, K., Miyano, K., Doi, M., Tokonami, S.,
(2010). “A primary standard for activity concentration Uesugi, M., and Takata, Y. (1999). “Measurement of na-
of 220Rn (thoron) in air,” Appl. Radiat. Isot. 68, tionwide indoor Rn concentration in Japan,” J. Environ.
1292–1296. Radioact. 45, 129–137.
Rovenská, K., Thinová, L., and Ždı́mal, V. (2008). Sanchis Aldas, J., Dolovich, M., Chalmers, R., and
“Assessment of the dose from radon and its decay pro- Newhouse, M. T. (1971). “Regional aerosol clearance in
ducts in the Bozkov dolomite cave,” Radiat. Prot. smokers and nonsmokers,” Chest 59, p2S.
Dosim. 130, 34– 37. Särndal, C. E., Swensson, B., and Wretman, J. (1992).
Saccomanno, G., Auerbach, O., Kuschner, M., Harley, N. Model Assisted Survey Sampling (Springer-Verlag,
H., Michels, R. Y., Anderson, M. W., and Bechtel, J. J. New York).
(1996). “A comparison between the localization of lung Schafer, D. W., and Gilbert, E. S. (2006). “Some statistical
tumors in uranium miners and in nonminers from 1974 implications of dose uncertainty in radiation dose–re-
to 1991,” Cancer 77, 1278– 1283. sponse analyses,” Radiat. Res. 166, 303– 312.
Sägusa, J., Yamasaki, K., Tsujimoto, T., Yamauchi, T., and Scheib, C., Appleton, D., Jones, D., and Hodgkinson, E.
Shimo, M. (1996). “Development of an apparatus for (2006). “Airborne uranium data in support of radon po-
measuring ground exhalation rates of 222Rn and 220Rn,” tential mapping in Derbyshire, Central England,” pp.
Environ. Int. 22, 483– 490. 210– 219 in Radon Investigations in the Czech Republic
Sahoo, B. K., Mayya, Y. S., Sapra, B. K., Gaware, J. J., XI and the 8th International Workshop on the
Banerjee, K. S., and Kushwaha, H. S. (2010). “Radon ex- Geological Aspects of Radon Risk Mapping, Barnet, I.,
halation studies in an Indian uranium tailings pile,” Neznal, M., and Pacherova, P., Eds. (Czech Geological
Radiat. Meas. 45, 237– 241. Survey, Prague).
Sainz, C., Quindós, L. S., Fuente, I., Nicolás, J., and Schmitz, J., and Nickels, R. M. (2001). “Exposure to radon/
Quindós, L. (2007). “Analysis of the main factors affect- radon decay products in water works,” Sci. Total
ing the evaluation of the radon dose in workplaces: the Environ. 272, 347 –348.

186
 References

Schubauer-Berigan, M. K., Daniels, R. D., and Pinkerton, Somlai, J., Gorjánácz, Z., Várhegyi, A., and Kovács, T.
L. E. (2009). “Radon exposure and mortality among (2006). “Radon concentration in houses over a closed
white and American Indian uranium miners: An update Hungarian uranium mine,” Sci. Total Environ. 367,
of the Colorado Plateau cohort,” Am. J. Epidemiol. 169, 653– 665.
718–730. Somlai, K., Tokonami, S., Ishikawa, T., Vancsura, P.,
Scott, A. G. (1981). “A field method for measurement of Gáspár, M., Jobbágy, V., Somlai, J., and Kovács, T.
radon daughters in air,” Health Phys. 41, 403 –405. (2007a). “222Rn concentrations of water in the Balaton
Sevcova, M., Sevc, J., and Thomas, J. (1978). “Alpha ir- Highland and in the southern part of Hungary, and the
radiation of the skin and the possibility of late effects,” assessment of the resulting dose,” Radiat. Meas. 42,
Health Phys. 35, 803 –806. 491– 495.
Sharma, N., Hess, C. T., and Thrall, K. D. (1997). “A com- Somlai, J., Kávási, N., Szabó, T., Várhegyi, A., and Kovács,
partmental model of water radon contamination in the T. (2007b). “The function of radon in curing respiratory
human body,” Health Phys. 72, 261 –268. diseases in the therapeutic cave of Tapolca,”
Sherman, M. (1992). “Simplified modeling for infiltration J. Radioanal. Nucl. Chem. 273, 363– 370.
and radon entry,” in Proceedings of the ASHRAE/ Soto, J., and Gómez, J. (1999). “Occupational doses from
DOE/BTECC Conference, Thermal Performance of the radon in Spanish spas,” Health Phys. 76, 398– 401.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

Exterior Envelopes of Buildings Conference V (adden- Steck, D. J. (1990). “A comparison of EPA screening mea-
dum), LBL-31305 (National Technical Information surements and annual 222Rn concentrations in state-
Service, Springfield, VA). wide surveys,” Health Phys. 58, 523– 530.
Sherman, M. H., and Modera, M. P. (1984). Comparison of Steck, D. J. (1992). “Spatial and temporal indoor radon
Measured and Predicted Infiltration Using the variations,” Health Phys. 62, 351 –355.
LBL-Infiltration Model, LBL-17001 (Lawrence Berkeley Steck, D. J. (2005). “Residential radon risk assessment:
Laboratory). how well is it working in a high radon region?” pp. 1 –13
Shimo, M., Saegusa, J., Tsujimoto, T., Yamasaki, K., and in 15th International Radon Symposium (American
Yamauchi, T. (1994). “Development of in-situ type Association of Radon Scientists and Technologists,
radon/thoron exhalation rate measuring system,” Bull. Fletcher, NC).
Gifu Coll. Med. Tech. 10, 19– 26 (in Japanese). Steck, D. J. (2009). “Annual average indoor radon varia-
Shweikani, R., and Hushari, M. (2005). “The correlations tions over two decades,” Health Phys. 96, 37– 47.
between radon in soil gas and its exhalation and concen- Steck, D. J., and Field, R. W. (1999). “The use of track
tration in air in the southern part of Syria,” Radiat. registration detectors to reconstruct contemporary and
Meas. 40, 699 –703. historical airborne radon (222Rn) and radon progeny
Sill, C. W. (1977). “Integrating air sampler for determin- concentrations for a radon-lung cancer study,” Radiat.
ation of Rn-222,” pp. 97– 104 in Program Report on the Meas. 31, 401 –406.
Workshop on Methods for Measuring Radiation in and Steinbuch, M., Weinberg, C. R., Buckley, J. D., Robinson,
Around Uranium Mills, Weaver, C. L., Ed. (Atomic L. L., and Sandler, D. P. (1999). “Indoor residential
Industrial Forum, Inc., Washington, DC). radon exposur and risk of childhood acute myeloid leu-
Sinclair, D., Countess, R. J., and Hoopes, G. S (1974). kaemia,” Br. J. Cancer 81, 900– 906.
“Effect of relative humidity on the size of atmospheric Stoop, P., Glastra, P., Hiemstra, Y., de Vries, L., and
aerosols particles,” Atmos. Environ. 8, 1111–1117. Lembrechts, J. (1997). Results of the Second Dutch
Skliarov, V., Röttger, A., Honig, A., Korostin, S., Kuznetsov, National Survey on Radon in Dwellings,
S., Lapenas, A., Milevsky, V., Ivaniukovich, A., RIVM-Report No. 610058006 (National Institute of
Kharitonov, I., and Sepman, S. (2009). “COOMET.RI(II)- Public Health and the Environment, Bilthoven, The
S1.Rn-222 (169/UA/98): Rn-222 volume activity compari- Netherlands).
son,” Metrologia 46, doi:10.1088/0026-1394/46/1A/06006. Streil, T., Oeser, V., and Feige, S. (2002). “An electronic
Sohrabi, M., and Babapouran, M. (2005). “New public dose radon dosimeter as a new multipurpose device—a
assessment from internal and external exposures in bridge between dosimetry and monitoring,” Geofı́s. Int.
low- and elevated-level natural areas of Ramsar, Iran,” 41, 285– 288.
Int. Congr. Ser. 1276, 169– 174. Su, Y., Feng, Y., Gao, H., Xing, J., Song, C., Pang, X.,
Solomon, S. B., and Ren, T. (1992). “Counting efficiencies Wang, X., and Wei, H. (2007). “Analysis of the reasons of
for alpha particles emitted from wire screens,” Aerosol recently some radioactive source accidents and sugges-
Sci. Technol. 17, 69– 83. tions for management countermeasures,” Radiat. Prot.
Solomon, S. B., Cooper, M. B., O’Brien, R. S., and 27, 193– 198 (in Chinese).
Wilkinson, L. (1992). “Radon exposure in a limestone Subba Ramu, M. C., Muraleedharan, T. S., and
cave,” Radiat. Prot. Dosim. 45, 171 –174. Ramachandran, T. . (1988). “Calibration of a solid state
Solomon, S. B., O’Brien, R. S., Wilks, M., and James, A. C. nuclear track detector for the measurement of indoor
(1994). “Application of the ICRP’s new respiratory tract levels of radon and its daughters,” Sci. Total Environ.
model to an underground uranium mine,” Radiat. Prot. 73, 245– 255.
Dosim. 53, 119–125. Subba Ramu, M. C., Shaikh, A. N., Muraleedharan, T. S.,
Solomon, S. B., Langroo, R., Lyons, R. G., and James, J. and Ramachandran, T. V. (1990). “Measurements of the
M. (1996). “Radon exposure to tour guides in Australian equilibrium factor for 222Rn daughters in dwellings in
show caves,” Environ. Int. 22, 409 –413. India,” Sci. Tot. Environ. 99, 49– 52.

187
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

Sumesh, C. G., Vinod Kumar, A., Nair, R. N., Tripathi, R. and the unattached fraction fp of radon progeny in the
M., and Puranik, V. D. (2012). “Estimation of thoron environment,” Environ. Int. 22 (Suppl. 1), S611– S616.
concentration using scintillation cell,” Radiat. Prot. Tokonami, S., Iimoto, T., Ichiji, T., Fujitaka, K., and
Dosim. 150, 536 –540. Kurosawa, R. (1996c). “Continuous radon monitor using
Sun, K., Majdan, M., Field, D. W., and Field, R. W. (2006). a two-filter method,” Radiat. Prot. Dosim. 63, 123– 126.
“Field comparison of commercially available short-term Tokonami, S., Iimoto, T., Ichiji, T., Tanaka, N., and
radon detectors,” Health Phys. 91, 221– 226. Kurosawa, R. (1996d). “Integrated measurement of
Sun, K., Budd, G., McLemore, S., and Field, R. W. (2008). equilibrium equivalent radon and thoron concentra-
“Blind testing of commercially available short-term tions using cellulose nitrate film,” Radiat. Meas. 26,
radon detectors,” Health Phys. 94, 548– 557. 689– 699.
Sundal, A. V., Valen, V., Soldal, O., and Strand, T. (2008). Tokonami, S., Takahashi, F., Iimoto, T., and Kurosawa, R.
“The influence of meteoerological parameters on soil (1997). “A new device to measure the activity size distri-
radon levels in permable glacial sediments,” Sci. Tot. bution of radon progeny in a low level environment,”
Environ. 389, 418 –428. Health Phys. 73, 494 –497.
Suomela, M., and Kahlos, H. (1972). “Studies on the elim- Tokonami, S., Yang, M., and Sanada, T. (2001).
ination rate and the radiation exposure following inges- “Contribution from thoron on the response of passive

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

tion of 222Rn rich water,” Health Phys. 23, 641 –652. radon detectors,” Health Phys. 80, 612– 615.
Surbeck, H. (1996). “A radon-in-water monitor based on Tokonami, S., Matsuzawa, T., Ishikawa, T., Iimoto, T.,
fast gas transfer membranes”, pp. 177 –191 in Proc. Int. Yonehara, H., and Yamada, Y. (2003). “Changes of
Conf. Technologically Enhanced Natural Radiation indoor aerosol characteristics and their associated vari-
(TENR) Caused by Non-uranium Mining, IAEA-INIS ation on the dose conversion factor due to radon
Vol. 28, Issue 21 (Glowny Instytut Gornictwa, progeny inhalation,” Radioisotopes 52, 285 –292.
Katowice). Tokonami, S., Sun, Q., Akiba, S., Zhou, W., Furukawa, M.,
Suzuki, G., Yamaguchi, I., Ogata, H., Sugiyama, H., Ishikawa, T., Hou, C., Zhang, S., Narazaki, Y., Ohji, B.,
Yonehara, H., Kasagi, F., Fujiwara, S., Tatsukawa, Y., Yonehara, H., and Yamada, Y. (2004). “Radon and
Mori, I., and Kimura, S. (2010). “A nation-wide survey thoron exposures for cave residents in Shanxi and
on indoor radon from 2007 to 2010 in Japan,” J. Radiat. Shaanxi Provinces,” Radiat. Res. 162, 390 –396.
Res. 51, 683 –689. Tokonami, S., Takahashi, H., Kobayashi, Y., Zhuo, W., and
Synnott, H., and Fenton, D. (2005). An Evaluation of Hulber, E. (2005). “Up-to-date radon –thoron discrim-
Radon Reference Levels and Radon Measurement inative detector for a large scale survey,” Rev. Sci.
Techniques and Protocols in European Countries Instrum. 76, 113505.
(Radiological Protection Institute of Ireland, Wexford). Tomášek, L. (2013). “Lung cancer risk from occuptational
Tanner, A. B. (1991). “Error in measuring radon in soil gas and environmental radon and role of smoking in two
by means of passive detectors,” Nucl. Geophys. 5, 25–30. Czech nested case-control studies.” Int. J. Environ. Res.
Tempfer, H., Hofmann, W., Schober, A., Lettner, H., and Pub. Health 10, 963– 979.
Dinu, A. L. (2010). “Deposition of radon progeny on skin Tomášek, L., Kunz, E., Müller, T., Hulka, J., Heribanova,
surfaces and resulting radiation doses in radon A., Matzner, J., Placek, V., Burian, I., and Holecek, J.
therapy,” Radiat. Environ. Biophys. 49, 249– 259. (2001). “Radon exposure and lung cancer risk—Czech
Thinová, L., and Burian, I. (2008). “Effective dose assess- cohort study on residential radon,” Sci. Tot. Environ.
ment for workers in caves in the Czech Republic: experi- 272, 43–51.
ments with passive radon detectors,” Radiat. Prot. Tomášek, L., Rogel, A., Tirmarche, M., Mitton, N., and
Dosim. 130, 48– 51. Laurier, D. (2008a). “Lung cancer in French and Czech
Thomas, J. W. (1972). “Measurement of radon daughters uranium miners: radon-associated risk at low exposure
in air,” Health Phys. 23, 783– 789. rates and modifying effects of time since exposure and
Thomas, J. W., and LeClare, P. C. (1970). “A study of the age at exposure,” Radiat. Res. 169, 125 –137.
two-filter method for radon-222,” Health Phys. 18, Tomášek, L., Rogel, A., Laurier, D., and Tirmarche, M.
113– 122. (2008b). “Dose conversion of radon exposure according
Todorovic, N., Nikolov, J., Forkapic, S., Bikit, I., Mrdja, D., to new epidemiological findings,” Radiat. Prot. Dosim.
Krmar, M., and Veskovic, M. (2012). “Public exposure to 130, 98–100.
radon in drinking water in Serbia,” Appl. Radiat. Isot. Tommasino, L., Cherouati, D. E., Seidel, J. L., and Monnin,
70, 543– 549. M. (1986). “A plastic-bag sampler for passive radon moni-
Tokonami, S. (2010). “Why is 220Rn (thoron) measurement toring,” Nucl. Tracks Radiat. Meas. 12, 681–684.
important?” Radiat. Prot. Dosim. 141, 335 –339. Tommasino, L., Cardellini, F., Kozak, K., and Mazur, J.
Tokonami, S., Matsumoto, M., Furukawa, M., Fujimoto, (2012). “New measurement-strategies for radon in
K., Fujitaka, K., and Kurosawa, R. (1996a). “Behavior of water and soil,” in 11th International Workshop on the
radon and its progeny at working place,” pp. 151 –153 in Geological Aspects of Radon Risk Mapping (Czech
Proceedings of the 9th International Congress of the Geological Survey, Prague).
International Radiation Protection Association Tóth, E., Hámori, K., and Minda, M. (2006). “Indoor radon
(IRPA-9). in Hungary (lognormal mysticism),” in Radon
Tokonami, S., Iimoto, T., and Kurosawa, R. (1996b). Investigations in the Czech Republic XI and the 8th
“Continuous measurement of the equilibrium factor F International Workshop on the Geological Aspects of

188
 References

Radon Risk Mapping, Barnet, I., Neznal, M., and Van Dilla, M. A., and Taysum, D. H. (1955). “Scintillation
Pacherová, P., Eds. (Czech Geological Survey, Prague). counter for assay of radon gas,” Nucleonics 13, 68–69.
Trautmannsheimer, M., Schindlmeier, W., and Börner, K. Vanmarcke, H., Berkvens, P., and Poffijn, A. (1989). “Radon
(2003). “Radon concentration measurements and per- versus Rn daughters,” Health Phys. 56, 229–231.
sonnel exposure levels in Bavarian water supply facil- Vargas, A., Ortega, X., and Porta, M. (2000). “Dose conver-
ities,” Health Phys. 84, 100– 110. sion factor for radon concentration in indoor environ-
Trotti, F., Mozzo, P., Temporin, A., Lanciai, M., Predicatori, ments using a new equation for the F 2 fp correlation,”
F., Righetti, F., and Tacconi, A. (1996). “CR-39 track Health Phys. 78, 80– 85.
detectors applied to measurements of 210Po embedded in Vargas, A., Michielsen, N., Le Moing, C., Rio, M., Tymen,
household glass,” Environ Int. 22(Suppl. 1), 863–869. G., and Ortega, X. (2005). “Determination of 218Po nano-
Tschiersch, J., Li, W. B., and Meisenberge, O. (2007). meter size distribution in a controlled environment by
“Increased indoor thoron concentrations and implica- two new systems,” pp. 361– 370 in Natural Radiation
tion to inhalation dosimetry,” Radiat. Prot. Dosim. 127, Enviroment VII, McLaughlin, J. P., Simopoulos, S. E.,
73–78. and Steinhäusler, F., Eds. (Elsevier Science, Oxford).
Tsivoglou, E. C., Ayer, H. E., and Holiday, D. A. (1953). Vásárhelyi, A., Csige, I., Hakl, J., and Hunyadi, I. (1997).
“Occurrence of non-equilibrium atmospheric mixtures “Spatial distribution of radon content of soil-gas and

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

of radon and its daughters,” Nucleonics 11, 40– 45. well-waters measured with etched track radon moni-
Tu, K. W., and Knutson, E. O. (1988). “Indoor radon tors,” Radiat. Meas. 28, 685 –690.
progeny particle size distribution measurements made Vaupotič, J. (2007). “Nano-size radon short-lived progeny
with two different methods,” Radiat. Prot. Dosim. 24, aerosols in Slovenian kindergartens in wintertime,”
251– 255. Chemosphere 69, 856– 863.
Tu, K. W., Knutson, E. O., and George, A. C. (1991). “Indoor Vaupotič, J. (2008a). “Comparison of various methods of
radon progeny aerosol size measurements in urban, surban, estimating radon dose at underground workplaces in
and rural regions,” Aerosol Sci. Technol. 15, 170–178. wineries,” Radiat. Environ. Biophys. 47, 527– 534.
Tuccimei, P., Moroni, M., and Norcia, D. (2006). Vaupotič, J. (2008b). “Nanosize radon short-lived decay
“Simultaneous determination of 222Rn and 220Rn exhal- products in air of the Postojna Cave,” Sci. Tot. Environ.
ation rates from building materials used in Central 393, 27–38.
Italy with accumulation chambers and a continuous Vaupotic, J. (2008c) Levels of nanosize radon decay pro-
solid state alpha detector: influence of particle size, hu- ducts in inddor air: a comparison of different environ-
midity and precursors concentration,” Appl. Radiat. ments. Coll. Antropol. 32, (Suppl. 2), 99– 104.
Isot. 64, 254– 263. Vaupotič, J., and Kobal, I. (2001). “Radon exposure in
Twomey, S. (1975). “Comparison of constrained linear in- Slovenia spas,” Radiat. Prot. Dosim. 97, 265– 270.
version and an iterative nonlinear algorithm applied to Vaupotič, J., and Kobal, I. (2006). “Effective doses in
the indirect estimation of particle size distributions,” schools based on nanosize radon progeny aerosols,”
J. Comput. Phys. 18, 188– 200. Atmos. Environ. 40, 7494–7507.
Tymen, G., Robe, M. C., and Rannou, A. (1992). VDI (1980). Association of German Engineers.
“Measurements of aerosol and radon daughters in five Determination of the Dust Concentration in the Workplace
radon houses,” Radiat. Prot. Dosim. 45, 319– 322. for Industrial Hygiene Purpose, VDI 2265 (Verein
Ui, H., Tasaka, S., Hayashi, M., Osada, K., and Iwasaka, Deutscher Ingenieure e.V., Düsseldorf) (in German).
Y. (1998). “Preliminary results from radon observation Virk, H. S., Sharma, N., Singh, R. P., and Ramachandran,
at Syowa Station, Antarctica during 1996,” Polar T. V. (2000). “Measurement of indoor radon and thoron
Meteorol. Glaciol. 12, 112–123. levels in some districts of Punjab,” Ind. J. Environ.
UNSCEAR (2000). United Nations Scientific Committee Prot. 20, 268– 271.
on the Effects of Atomic Radiation. Sources and Effects Vogiannis, E., Nikolopoulos, D., Louizi, A., and
of Ionizing Radiation, UNSCEAR 2000 Report to the Halvadakis, C. P. (2004a). “Radon exposure in the
General Assembly, with Scientific Annexes (United thermal spas of Lesvos Island—Greece,” Radiat. Prot.
Nations Publications, New York). Dosim. 111, 121– 127.
UNSCEAR (2008). United Nations Scientific Committee Vogiannis, E., Nikolopoulos, D., Louizi, A., and
on the Effects of Atomic Radiation. Effects of Ionizing Halvadakis, C. P. (2004b). “Radon variations during
Radiation, UNSCEAR 2006 Report, Volume II: treatment in thermal spas of Lesvos Island (Greece), ”
Scientific Annexes C, D, and E Annex E: Sources-to- J. Environ. Radioact. 76, 283– 294.
effects assessment for radon in workplaces and homes von Doebeln, W., and Lindell, B. (1965). “Some aspects of
(United Nations Publications, New York). radon contamination following ingestion,” Arkiv Physik
Vacquier, B., Caer, S., Rogel, A., Feurprier, M., Tirmarche, 27, 531– 572.
M., Lucioni, C., Quesne, B., Acker, A., and Laurier, D. Walsh, C., and McLaughlin, J. P. (2001). “Correlation of
210
(2008). “Mortality risk in the French cohort of uranium Po implanted in glass with radon gas exposure:
miners: extended follow-up 1946–1999,” Occup. sensitivity analysis of critical parameters using a
Environ. Med. 65, 597 –604. Monte-Carlo approach,” Sci. Total Environ. 272,
Valmari, T., Mäkeläinen, I., Reisbacka, H., and Arvela, H. 195 – 202.
(2011). “Finnish radon situation analysed using national Walsh, L., Tschense, A., Schnelzer, M., Dufey, F., Grosche,
measurement database,” Radiat. Prot. Dosim. 145, 101–106. B., and Kreuzer, M. (2010). “The influence of radon

189
 MEASUREMENTS AND REPORTING OF RADON EXPOSURES

exposures on lung cancer mortality in German uranium Winkler-Heil, R., Hofmann, W., Marsh, J., and Birchall, A.
miners, 1946–2003,” Radiat. Res. 173, 79–90. (2007). “Comparison of radon lung dosimetry models for
Weibel, E. R. (1963). Morphometry of the Human Lung the estimation of dose uncertainties,” Radiat. Prot.
(Academic Press, New York). Dosim. 127, 27– 30.
Weingartner, E., Burtscher, H., and Baltensperger, U. Winkler-Heil, R., Hofmann, W., and Hussain, M. (2014).
(1997). “Hygroscopic properties of carbon and diesel “Stochastic dosimetry model for radon progeny in the
soot particles,” Atmos. Environ. 31, 2311 –2327. rat lung,” Radiat. Prot. Dosim. 160, 104 –107.
Wendt, J., Rehs, B., and Porstendörfer, J. (1988). Wrixon, A. D., Green, B. M. R., Lomas, P. R., Miles, J. C.
Bestimmung der dosisrelevanten Parameter (F_Faktor, H., Cliff, K. D., Francis, E. A., Driscoll, C. M. H., James,
fp-Faktor, Aktivitätsgrößenverteilung) im Hinblick auf A. C., and O’Riordan, M. C. (1988). Natural Radiation
Vergleich der outdoor- indoor Radonkonzentration. Exposure in UK Dwellings, NRPB-R-190 (Health
Abschlussbericht, StSch. 4084 (Federal Ministry for the Protection Agency, Chilton, Didcot).
Environment, Nature Conservation, Building and Wu-Tu, K., Fisenne, I. M., and Hutter, A. R. (1997). Short-
Nuclear Safety, Berlin). and Long-Lived Radionuclide Particle Size Measurements
Whicker, J. J., and McNaughton, M. W. (2009). “Work to in a Uranium Mine, Report EML-588 (National Technical
save dose: contrasting effective dose rates from radon Information Service, Springfield, VA).

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

exposure in workplaces and residences against the Yasuoka, Y., and Shinogi, M. (1997). “Anomaly in atmos-
backdrop of public and occupational regulatory limits,” pheric radon concentration: a possible precursor of the
Health Phys. 97, 248 –256. 1995 Kobe, Japan, earthquake,” Health Phys. 72,
White, S. B. (1994). “Making mitigation decisions based on 759– 761.
short-term tests of 222Rn,” Health Phys. 67, 180 –182. Yasuoka, Y., Shinogi, M., Ishii, T., Kataoka, Y., Kubo, T.,
White, S. B., Bergsten, J. W., Alexander, B. V., Rodman, N. Suda, H., Tokonami, S., and Ishikawa, T. (2004).
F., and Phillips, J. L. (1992). “Indoor 222Rn concentra- “Determination of radon concentration in water using
tions in a probability sample of 43,000 houses across 30 liquid scintillation counter,” Radioisotopes 53, 123–131.
states,” Health Phys. 62, 41– 50. Yeates, D. B., Aspin, N., Levinson, H., Jones, M. T., and
Whittlestone, S., and Zahorowski, W. (1998). “Baseline radon Bryan, A. C. (1975). “Mucociliary tracheal transport
detectors for shipboard use: development and deployment rates in man,” J. Appl. Physiol. 39, 487– 495.
in the First Aerosol Characterization Experiment (ACE Yeh, H. C., and Schum, G. M. (1980). “Models of human
1),” J. Geophys. Res. 103, 16743–16751. lung airways and their application to inhaled particle
WHO (1986). World Health Organization. Indoor Air deposition,” Bull. Math. Biol. 42, 461 –480.
Quality Research: Report on a WHO Meeting (WHO Yeh, H. C., Schum, G. M., and Duggan, M. T. (1979).
Press, Geneva). “Anatomic models of the tracheobronchial and pulmon-
WHO (2009). World Health Organization. WHO Handbook ary regions of the rat,” Anat. Rec. 195, 483– 492.
on Indoor Radon: A Public Health Perspective (WHO Yokoyama, S., Mori, N., Shimo, M., Fukushi, M., and
Press, Geneva). Ohnuma, S. (2011). “Measurement of radon concentra-
Wichmann, H. E., Rosario, A. S., Heid, I. M., Kreuzer, M., tion in water using the portable radon survey meter,”
Heinrich, J., and Kreienbrock, L. (2005). “Increased lung Radiat. Prot. Dosim. 146, 23–26.
cancer risk due to residential radon in a pooled and extended Yu, K. N., Young, E. C. M., and Li, K. C. (1996). “A survey
analysis of studies in Germany,” Health Phys. 88, 71–79. of radon properties for dwellings for Hong Kong,”
Wiegand, J., Feige, S., Quingling, X., Schreiber, U., Radiat. Prot. Dosim. 63, 55–62.
Wieditz, K., Wittmann, C., and Xiarong, L. (2000). Yu, K. N., Young, E. C., Stokes, M. J., and Tang, K. K.
“Radon and thoron in cave dwellings (Yan’an, China),” (1998). “Radon properties in offices,” Health Phys. 75,
Health Phys. 78, 438 –444. 159– 164.
Wilkening, M. (1990). Radon in the Environment (Elsevier, Yu, K. N., Cheung, T., Guan, Z. J., Young, E. C. M., Mui, B.
Amsterdam). W. N., and Wong, Y. Y. (1999). Concentration of 222Rn,
220
Willeke, K. (1976). “Temperature dependence of particle Rn and their progeny in residences in Hong Kong,”
slip in a gaseous medium,” J. Aerosol Sci. 7, 381 –387. J. Environ. Radioact. 45, 291– 308.
Wilson, D. L., Gammage, R. B., Dudney, C. S., and Saultz, Yu, K. N., Cheung, T., Guan, Z. J., Mui, B. W. N., and Ng,
R. J. (1991). “Summertime elevation of 222Rn levels in Y. T. (2000). “222Rn, 220Rn and their progeny concentra-
Huntsville, Alabama,” Health Phys. 60, 189– 197. tions in offices in Hong Kong,” J. Environ. Radioact. 48,
Winkler-Heil, R., and Hofmann, W. (2002). “Comparison of 211– 221.
modeling concepts for radon progeny lung dosimetry,” Yuan, Y., Morishima, H., Shen, T., Koga, T., Wei, L., and
pp. 169 –177 in High Levels of Natural Radiation and Sugahara, T. (2000). “Measurements of Rn-222, Rn-220
Radon Areas: Radiation Dose and Health Effects, and their decay products in the environmental air of the
Burkart, W., Eds. (Elsevier, Amsterdam). high background radiation areas in Yangjiang, China,”
Winkler-Heil, R., and Hofmann, W. (2005). “Stochastic J. Radiat. Res. 41 (Suppl.), 25–30.
radon lung dosimetry—modeling variability of bron- Zhang, Z., Smith, B., Steck, D. J., Quo, Q., and and Field,
chial cellular doses,” pp. 624– 631 in Natural Radiation R. W. (2007). “Variation in yearly residential radon con-
Enviroment VII, McLaughlin, J. P., Simopoulos, S. E., centration in the upper Midwest,” Health Phys. 93,
and Steinhäusler, F., Eds. (Elsevier Science, Oxford). 288– 297.

190
 References

Zhang, L., Guo, Q., and Zhuo, W. (2010). “Measurement of inhaled short-lived radon decay products on human
the 212Pb particle size distribution indoors,” Radiat. lung dose,” Radiat. Prot. Dosim. 63, 197 –206.
Prot. Dosim. 141, 371 –373. Žunić, Z. S., McLaughlin, J. P., Walsh, C., and
Zhuo, W., Tokonami, S., Yonehara, H., and Yamada, Y. Benderac, R. (1999). “The use of SSNTDs in the
(2002). “A simple passive monitor for integrating mea- retrospective assessment of radon exposures in high
surements of indoor thoron concentrations,” Rev. Sci. radon rural communities in Yugoslavia,” Radiat.
Instrum. 73, 2877–2881. Meas. 31, 343 –346.
Zmazek, B., Italiano, F., Zivcic, M., Vaupotic, J., Kobal, I., Žunić, Z. S., Yarmoschenko, I. V., Birovljev, A., Bochicchio,
and Martinelli, G. (2002). “Geochemical monitoring of F., Quarto, M., Obryk, B., Paszkowski, M., Celikovic, I.,
thermal waters in Slovenia: relationship to seismic ac- Demajo, A., Ujic, P., Budzanowski, M., Oiko, P.,
tivity,” Appl. Radiat. Isot. 57, 919 –930. McLaughlin, J. P., and Waligorski, M. P. R. (2007).
Zock, C., Porstendörfer, J., and Reineking, A. (1996). “The “Radon survey in the high natural radiation region of
influence of the biological and aerosol parameters of Niska Banja, Serbia,” J. Environ. Radioact. 92, 165–174.

Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016

191
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016
Downloaded from http://jicru.oxfordjournals.org/ at Hirosaki University on April 5, 2016