Scintillation Detectors

Slide 1
Scintillation Detectors. A module developed for the International Atomic Energy
Agency as part of a training course for the maintenance of Nuclear Electronic
Systems.

Slide 2
A scintillation detector starts with a scintillator, a material that will produce a flash
of light when struck by nuclear radiation. The scintillator is usually attached to a
photomultiplier tube. The photomultiplier tube is a vacuum tube, flat on one end.
The inside of the flat portion of the tube is coated with a photocathode material.
This is a material with a low work function, such that when light from the
scintillator strikes the photocathode, electrons are emitted. The electrons are
then collected into a charge multiplier region in the photomultiplier tube. Here the
charge is multiplied or amplified producing output pulse of charge that is
proportional to the number of electrons being emitted from the photocathode.
This pulse is also then proportional to the amount of light produced in the
scintillator, which is proportional to the amount of energy deposited by the
radiation.

Slide 3
Scintillators are frequently connected to the photomultiplier tube using a device
called a light pipe.

Slide 4
Light pipes have three general functions. The first is to match geometries as
shown in the next slide.

Slide 5
In this case, the scintillator is a thin slab of material and is used to detect the
presence of a beam of charged particles, in this case a beam of beta particles.
The particles pass through the detector depositing only a small portion of their
energy. The light from the detector exits through the edge and passes through
the light pipe into the photomultiplier tube.

Slide 6
A second function of light pipes is to separate the photomultiplier tube from the
radiation environment. For instance, the photomultiplier tube is very, very
sensitive to magnetic fields. If it is required that the scintillation detector be
placed in a magnetic field, then a light pipe is frequently used to guide the light
away from the magnetic field to the photomultiplier tube. A third function is to
improve resolution.

1
Slide 7
The sensitivity of the photocathode surface varies greatly across the face of the
photomultiplier tube as shown in this slide,. The portion near the center of the
photocathode may be considered 100% efficient while regions in the outer areas
of photomultiplier tube may be as low 20% efficient. This means that some
regions in the outer edges of the photomultiplier tube would only produce one-
fifth of the charge for the same amount of light striking the photomultiplier tube.

Slide 8
If the scintillation crystal is attached directly to the photomultiplier tube, the light
from radiation interacting close to the interface would interact only with a small
region of the photocathode. Depending on whether the light interacted with the
outer region or central portion of the photocathode, one could get a large
variation in the amount of charge for a given amount of energy deposited in the
scintillation crystal. On the other hand, if the scintillation crystal is separated
from the photomultiplier tube, then if the radiation strikes near the end of the
scintillation crystal, the light is averaged over the whole area of the photocathode
surface giving a uniform or consistent amount of charge emitted by the
photocathode.

Slide 9
An example of the improvement in resolution that can be achieved by using a
light pipe to connect a crystal to the phototube are shown on this slide. Here the
relative resolution of detector is shown as a function of the thickness of the light
pipe. In this case, a 10 cm diameter scintillation crystal is connected to a 12 ½
cm diameter photomultiplier tube. One can see that for this detector system, the
resolution for Co-60 is greatly improved with a light pipe that has a thickness of
about 2.5 centimeters.
Slide 10
Scintillation Detector Systems. Common systems for using a scintillation
detector are shown on this slide. First, in order to operate, the scintillation
detector requires a high voltage. The voltage depends on the type of
photomultiplier tube that is used in the system and may vary from about 1,000
volts to about 3,000 volts. The scintillation detector actually requires a number of
different voltages. Therefore, the high voltage is usually fed through a voltage
divider into the scintillation detector with voltages for each of the dynodes, the
focusing grid, and the anode taken from different points of the voltage divider.

The signal from the scintillation detector is then fed into a preamplifier. Although
these two units may be separated, they are frequently combined into one
module. The signal from the preamplifier is then fed to a main amplifier where it
is amplified and finally sent to a pulse analyzing system.

2
Slide 11
Properties of Scintillators.
The first function of the scintillator is to convert energy from the radiation to the
light. The maximum sensitivity of most photomultiplier tubes occurs when the
wavelength of the light spectrum peaks around 440 nm. This is in the range of
blue to ultraviolet light in the visible light spectrum. The second property of
scintillators is that they must be transparent to light.

Slide 12
Scintillators can be divided into four main classes.
The first are the inorganic scintillators exemplified by thallium activated sodium
iodide. That is sodium iodide with a small amount of thallium added. Other
examples of inorganic scintillators might be thallium activated cesium iodide or
lithium iodide. Another type of inorganic scintillator is BGO or bismuth germinate
and a fifth might be zinc sulfide activated with silver. There are other inorganic
scintillators besides these five. The first three listed here sodium iodide, cesium
iodide, and lithium iodide are all single crystals, which limits they size of the
crystals and makes them expensive. They also are all hygroscopic. This means
that they will readily absorb moisture from the air. Therefore, they must be
hermetically sealed in order to protect the crystals. If they absorb moisture their
scintillation properties change and they become ineffective as scintillators. This
condition can be observed because the crystals will change color.

A second class of scintillators would be the organics. Two examples of organic

scintillators are anthracene and stilbene. These are low Z materials which
means they would be good for detecting charged particles. The problem with
these materials is that they are not single crystals, but are actually made up of a
multitude of smaller crystals and the light reflects at the planes between the
crystals. Therefore, they are limited in their transparency properties and the
ability for light to escape the crystal. The maximum useful thickness of these
detectors is about one or two centimeters. In addition, anthracene is a
carcinogenetic material.

A third class of scintillator are the plastics. This type of scintillator might be
exemplified by the use of polystyrene with some type of material added to
activate it. The additive might be something like tetraphenylbutadene. Different
combinations of materials will give the scintillator different properties. Sizes of
plastic scintillators range from a few cubic centimeters to meters.

A fourth class of scintillator would be the liquid scintillator. Historically, the liquid
has been toluene plus some material such as PBO or PBD or POPOP added as
an activator. However because of the toxic nature of toluene it has been
supplemented in more recent years with mineral oil based scintillators. Other
materials, including water have been used as scintillators in large detectors used

3
in high energy physics experiments. In the laboratory, the size of the liquid
scintillators are usually in the range of 10 or 20 ml to a few liters.

Slide 13
This slide shows some of the thousands of photomultiplier tubes used in to detect
light from 50,000 tons of water used as a liquid scintillator in system to detect
neutrinos from the sun.

Slide 14
An example of how the scintillation process works in an activated inorganic
scintillator such as sodium iodide is shown here. In a single crystal, the electrons
are all located in very well defined energy levels as shown here in what is known
as a Ferme energy level diagram where the energy of the electrons is shown as
the vertical scale. The last electrons in the atoms are all in the valence shell.
Then there is a forbidden gap where there are no levels. And finally there will be
a conduction band. If the electrons moved into the conduction band, they are
free to move around the crystal. Radiation entering the crystal will excite an
electron and raise it from the valence band into the conduction band. Nature,
however, does not like to be in the excited level, so the electron moves around
the conduction band until it reaches the bottom of the level and then de-excites
by dropping back down into the valence band. When the electron drops back to
valence band, it will emit a light photon. However, for most crystals this light
would be too energetic. That is, the wavelength would be too short.

Slide 15
Therefore, in order to increase wavelength of the photon, a material is added that
will introduce extra energy levels in the forbidden gap. In the case of sodium
iodide the material would be a small amount of thallium. Now when the electron
de-excites, it drops to the energy level in the forbidden gap and then down to the
valence band. This means that the photons are lower in energy or the
wavelengths are longer moving them into the desirable range around 440 nm.

Slide 16
The importance of having a proper spectral response from the scintillator is
shown in this slide. Here, the response of two typical photomultiplier tubes, with
a bialkali photocathode and an S-11 photocathode are shown. It should be
pointed out that the responses here have been normalized to 10 units. They are
not typical of the actual sensitivity of the photomultiplier tubes. The light output of
typical thallium activated sodium iodide and sodium activated cesium iodide
scintillators are also plotted on the diagram and one can see that they peak at
about 430-440 nm, which matches well with the response of the photomultiplier
tubes. On the other hand, if one takes a look at thallium activated cesium iodide,
one finds that it peaks at around 550 nm, well outside the range of maximum
sensitivity of the photomultiplier tubes.

4
Slide 17
Other properties to consider when selecting a scintillator are first the conversion
efficiency, that is how much energy must be deposited in the detector to produce
one usable light photon. For thallium activated sodium iodide, this is about 13%.
In other words, for every 26 eV of energy deposited in the scintillator, one usable
light photon will be produced. Anthracene is about 5 % efficient, plastic is about
2%, and liquid scintillators that can be simply considered as a liquid plastic are
also about 2%. BGO is about 3% efficient.

Another factor to consider is the density of the scintillator. This becomes

important because one wants a high density for stopping and detecting gamma
rays or high energy photons. On the other hand, if one wants to detect electrons,
one wants a low density scintillator. High density scintillators mean that they
have high z nuclei. Charged particles interacting with high z nuclei may back
scatter out of the detector. One wants the electrons to scatter in a forward
direction. Therefore one wants a low z material.

The final property to consider when selecting a scintillator is the mean lifetime of
the conversion process. Remember that the radiation is exciting atoms. The
electrons decay back down to the low energy state with a mean lifetime much
like the radioactive decay of nuclei. If the mean lifetime is too long, the light is
emitted over a long period of time and this will result in a very slow rise time
pulse. Therefore one wants the mean lifetime to be on the order of
microseconds or less. Usually the shorter the lifetime the better.

Slide 18
Properties of a number of scintillators are given here. First consider the total
number of photons per MeV of energy deposited in the scintillator. This is the
conversion efficiency. For thallium activated sodium iodide, one sees that about
38,000 photons are produced for each MeV of energy deposited in the detector.
Thallium activated cesium iodide is even higher; almost twice as high. BGO
produces only 8,200 photons per MeV.

The second property to consider is the wavelength of the light. That is whether it
will be a spectral match with the photocathode. Here one sees that sodium
iodide has a peak at 415 nm, whereas cesium iodide has a peak at 540 nm.

The third factor one wants to consider is the specific gravity or the density of the
material. One finds that sodium iodide has a specific gravity of 3.67. BGO is
much better 7.13, which means that it would be much better for stopping
photons.

A fourth factor is the decay constant or the mean lifetime in microseconds. One
sees that thallium activated sodium iodide has a decay constant of 0.23 µs. This
will produce a pulse with about a quarter of a microsecond rise time. On the
other hand, if one takes a look at cesium iodide one finds that the decay constant

5
is about 1 microsecond resulting in about a 4 microsecond rise time in the output
pulses.

Finally, if one looks at the relative pulse heights with the bialkali photomultiplier
tube, comparing to the pulse heights to thallium sodium iodide normalized to one,
one finds that thallium activated cesium iodide is 0.49 primarily because of the
mismatch between the spectrum of the light output and sensitivity of the
photocathode.

If one looks at cesium iodide again, one finds that it has many good properties
such as high specific density and high conversion efficiency, however the
spectral match causes it to produce a relatively low pulse height with a typical
photomultiplier tubes.

Slide 19
Photomultiplier Tubes.
Photomultiplier tubes consist of a cylindrical vacuum tube with a photocathode on
the inside. Light entering through the end hits the photocathode producing a
pulse of charge. This process is shown on the next slide.

Slide 20
One end of the photomultiplier tube is flat to provide a better surface for coupling
to the scintillator or light pipe. On the inside of the flat end is the photocathode.
When light hits the photocathode, electrons are ejected.

Slide 21
The photocathode consists of material with low work functions, such as
antimony, potassium and cesium, or sodium and potassium. Low work functions
mean that the outer electrons of the atom are bound with very low energies so it
is very easy for low energy light photons to knock an electron out of the atom.
The sensitivity of these materials peaks at about 440 +/- 40 nm. A typical S-11
response is about 10% efficient. This means that for every 10 light photons
hitting the photocathode, about 1 electron will be ejected. The newer bi-alkali
photomultiplier tubes have about 16% efficiency or 1 photoelectron for about
every 6 photons hitting the photocathode.

Slide 22
When the electron is ejected from the photocathode it first sees a focusing grid.
The focusing grid is at a potential of about 100 volts positive with respect to the
photocathode surface. The electrons are accelerated toward the focusing grid
and pass through it to a special electrode called a dynode, which is about 100
volts positive with respect with the focusing grid. Thus the electrons that strike
the dynode have gained an energy of about 200 electron volts of energy.
When the electrons strike the dynode, which has a special low work function
surface, a number of electrons will be ejected usually between 3 and 4. The

6
secondary electrons are then focused onto a second dynode, which sits at about
100 volts positive with respect to the first dynode. When these electrons which
have gained 100 eV of energy strike the second dynode the charge multiplication
process is repeated and one has even more electrons, which are then attracted
to a third dynode. This process is repeated for ten or more dynodes.

Slide 23
To summarize, we have the focusing electrode, which sits at about 100 volts
positive. The electrons are accelerated to about 200 electron volts and hit the
first dynode. The dynodes are special electrodes that are decoded with a special
type of material, such as cesium antimony or silver manganese. For each
electron striking the dynode, 3 or 4 electrons are released, multiplying the
charge. The secondary electrons are then focused and accelerated toward
another dynode, repeating the process. Most photomultiplier tubes will have 10
dynodes so the process is repeated 10 times or the total charge multiplication will
be a factor of 410, which is equal to 106 or 1 million. Therefore the charge of one
electron will be amplified by a factor of 1 million.

Slide 24
When the electrons leave the last dynode, the charge is then collected onto an
anode producing an output pulse.

Slide 25
The total multiplication of the system depends upon the high voltage that has
been applied and typically will vary as about the seventh power of the applied
high voltage. This means that if the detector is to be stable, the high voltage
power supply must be very well regulated.

Slide 26
The dynode structures of a number of typical photomultiplier tubes are shown
here and on the next slide. One of the oldest of the dynode structures is the box
and grid dynode structure.

A much newer design is what is known as a venetian blind dynode structure.

Although the use of this dynode structure is limited because the large
capacitance between the dynodes results in relatively slow response times, it
does have one significant advantage. When one has high counting rates the
demand for charge coming from the last dynodes often exceeds the ability of the
dynode to produce the charge resulting in a non-linearity in the amplification
process as the count rate is increased. The venetian blind dynode structure
provides significantly better stability as a function of count rate than the other
photomultiplier designs.

7
Slide 27
Some other dynode structures include the focused linear grid dynode structure
and the circular grid dynode structure, which is probably the most common
design. The circular grid design is commonly called a squirrel cage.

Slide 28
This is a photograph of a typical 10 dynode circular grid photomultiplier tube.
Notice the many pins in the tube base. Each dynode requires a separate
voltage.

Slide 29
This is another view of the circular grid design with the glass cut away and the
screen taken off of the circular dynode structure.

Slide 30
This is a picture of the end of the photomultiplier tube showing the focusing grid
structure.

Slide 31
In order to increase the light collection efficiency some photomultiplier tubes are
made with an enlarged end region. In this case the diameter of the photocathode
is 7 ½ centimeters. Photomultiplier tube are made as large as 12 ½ centimeters.
Note the reddish color on the end. This is the photocathode. If the phototube
has developed a leak and has lost its vacuum, the air will turn this reddish color
into a transparent color.

Slide 32
Each dynode, the focusing electrode, and the anode require a different voltage,
therefore, rather than having 11 or 12 different power supplies, one generally
uses a voltage divider to supply the different voltages. The voltage divider is
simply a resister chain used to divide the high voltage. The resistor chain is
usually built into the tube socket.

Slide 33
A diagram for a typical voltage divider for a photomultiplier tube with ten dynodes
is shown here. As one can see, the total resistance of the system would be
approximately 6 MΩ. If one applies a voltage of 1,000 volts, this means that the
power supply must be capable of supplying approximately 0.5 milliamp of
current.

Slide 34
The voltage divider is frequently built into the base of the tube socket as shown
here. This particular photomultiplier tube setup is one that we built in our
laboratory.

8
Slide 35
This slide shows a commercially made tube base, again with the voltage divider
built into the base of the photomultiplier tube.

Slide 36
The output pulse from the photomultiplier tube is a pulse of charge. The rise time
of the charge pulse is determined by the activation processes in scintillation
crystal and is typically measured on the order of nanoseconds. The decay of the
pulse charge is determined by the mean lifetime of the excited levels in the
scintillator. These life times range from microseconds for some of the inorganic
materials to as low as a few nanoseconds for the plastic and liquid scintillators.
The pulse of charge is usually converted to a voltage pulse in the preamplifier
and then the decay time of the charge pulse becomes the rise time of the voltage
pulse.

Slide 37
Resolution
The resolution of any system is defined as the full width of the peak at one half of
the maximum height of the peak or FWHM in energy divided by the energy of the
peak times 100%.

Slide 38
An example is shown in this slide. Here a Cs-137 spectrum was recorded with a
sodium iodide detector. The photo peak is at 661.7 keV. In this case, the FWHM
is 43.3 keV and the resolution of the photo peak is 6.5%. For sodium iodide, Cs-
137 is the source typically used to determine the resolution of the system.

Slide 39
For scintillation detectors, the resolution will be proportional to the square root of
the total number of electrons reaching the first dynode.

Slide 40
Factors that affect the resolution in a scintillation detector are first the energy of
the radiation. Obviously, the more energy in the radiation striking the detector
and deposited in the detector, the more photoelectrons will be produced.

Second is the conversion efficiency. How efficient is the detector at converting

the energy deposited to usable light?

Third, is the collection efficiency of the system. How efficient is the detector at
collecting the light and getting it to the photocathode surface? One way of
improving the efficiency is to coat the outside of the scintillator with a white
material such that when light strikes the surface, it is re-emitted back into the
detector.

9
Fourth, what is the transparency of the scintillator? If the light is absorbed before
it gets out of the detector it does not do any good.

Fifth, what is the sensitivity of the photocathode? In other words, how many light
photons must hit the cathode to produce a photoelectron?

And finally, what fraction of the photoelectrons emitted by the photocathode

actually reach the first dynode? This is a function of the adjustment of the
focusing grid.

Slide 41
Factors such as the energy of the radiation and the properties of the scintillator
usually can not be adjusted, However, two factors that can be varied to obtain
the optimum resolution in the scintillation detector are first the focusing grid and
the high voltage. The easiest way to optimize the focusing grid is to observe the
output pulse of the photomultiplier tube with an oscilloscope and adjust the
voltage on the grid to obtain the maximum pulse height. If an oscilloscope is not
available, one can look at the photo peaks in a spectrum. The higher the energy
or channel number of the photo peak, the more electrons from the photocathode
are striking the first dynode.

The second factor that one can adjust is the high voltage itself. However,
changing the high voltage also changes the amplification of the phototube and to
obtain optimum resolution one must do an energy calibration at each setting of
the high voltage. This becomes very tedious and time consuming.

Slide 42
Rather than recalibrating the system for each setting of the high voltage, one can
optimize the high voltage by looking at a spectrum such as Co-60. Cobalt-60 has
two photo peaks that are close enough together that they cannot be completely
resolved with a scintillation detector. If one takes a look at the peak of the 1.17
MeV photo peak and compares that to the valley between the two photo peaks
and takes a ratio the number of counts at the peak height to the number of
counts in the valley, one can obtain a measure of the resolution of the system.
The greater the peak to valley ratio, the better the resolution.

Slide 43
This table shows an example of the peak to valley ratio as a function of the high
voltage for a given NaI detector. Here, five voltages were tried, varying from
1130 volts to 1440 volts. For each voltage setting, the amplifier gain had to be
readjusted to keep the spectrum and the photo peaks in the same relative
location on the multi-channel analyzer. One can observe the peak to valley ratio
varying from 8.5 up to a maximum of 10.1. Increasing the high voltage above
1340 volts caused the peak to valley ratio to drop off very rapidly to 5.0

10
Slide 44
Applications of scintillation detectors.
One of the most common applications of scintillation detectors is to detect
gamma rays and measure their energy. For this purpose the most common
detector is the sodium iodide.

Slide 45
The most commonly encountered NaI detectors are the integral units where the
scintillation crystal is permanently mounted to the photomultiplier tube and sealed
into one integral package. Many sizes are available. Some of the more
common sizes are shown here in this diagram. Among the more common sizes
are a five centimeter diameter by five centimeter high crystal mounted to a 5
centimeter ten stage photomultiplier tube.

Another common type often encountered is the 7 ½ centimeter diameter by 7 ½

centimeter high scintillation detector mounted to a 7 ½ centimeter diameter
phototube. Most of these integral units will incorporate a magnetic shield in the
covering. However, some of the older units, especially some of those made by
the Harshaw Company did not incorporate the magnetic shields. Without the
magnet shield, the phototubes are sensitive to the earth’s magnetic field and the
gain of the system will vary as the direction of the tube is changed.

Slide 46
When measuring the activities of sources a variety of geometries can be used.
One of the more common geometries that is frequently shown in the literature is
displayed here. In this case, one is using a point source located 25 centimeters
above the top of the detector. This is the standard geometry used by
manufacturers when measuring the resolution of the detector. Because the
source is located so far away from the detector all the photons enter the detector
perpendicular to the end. This results in the best resolution. However, because
the source is located so far away from the detector, there is only a very small
probability that a gamma ray will hit the detector. In other words, this is a low
geometry system. A more common practice would be to place the source
directly on top of the detector.

Most sources used in actual practice are not point sources, but rather are discs
as shown here. The source is commonly placed directly on top of the detector.
However a good practice would be to place a thin sheet of plastic or some other
low Z material between the source and detector to serve as a beta shield. In this
case, the beta particles would be stopped in the plastic producing a minimum of
bremsstrahlung radiation that would complicate the gamma ray spectrum.

Slide 47
Another common geometry is the one used to measure liquid samples. In this
case sample might be a one liter bottle placed directly on top of the detector.

11
Slide 48
Another geometry that is used for counting samples is the Marinelli beaker. This
geometry is commonly used for large samples and increases the detection
efficiency by surrounding the detector.

Slide 49
A diagram of a Marinelli beaker is shown here. It is a large beaker, either 1, 3, or
4 liters in size, and has a large hole or cavity in the center. The beaker is then
placed directly over the detector and completely surrounds the detector.

Slide 50
Another geometry frequently used to increase the detection efficiency for small
samples is the well type crystal.

Slide 51
Here a hole is drilled into the detector and the sample is placed into the well.
This increases the solid angle for detection. However it does have some
disadvantages. One disadvantage is that high energy photons may pass
completely through the walls of the detector without interacting. The second
disadvantage is that if one two low energy photons that are in coincidence, that is
they are emitted from the source within a microsecond of each other and both
interact in the detector, the energy from the two gamma rays can be summed
together making the event look like one high energy photon.

Slide 52
This is shown here where the source is emitting two photons with similar
energies in a cascade. In this case, the half life of the gamma ray is 1.49 ns
following the decay Iodine-125 by electron capture. When the Iodine-125 decays
by electron capture, a 27.5 keV X-ray is emitted. If one has source outside of the
detector one gets a peak due to the gamma rays and X-rays and a very small
sum peak, which is the sum of the energies of the two low energy photons.
However, if the source is in the well, one gets a much higher probability of
detection of both photons as shown in the peak, which is much larger, but at the
same time one also gets a large sum peak, which looks like a second gamma ray
and sometimes this can complicate the analysis of the spectrum.

Slide 53
Other Applications of Scintillation Detectors.
Among some of the other applications for scintillation detectors are gamma
cameras. These are devices that not only detect the gamma rays, but also will
give some information about the origin of the gamma ray.

Slide 54
One type of gamma camera is shown here. In this example, a large slab of
sodium iodide is monitored by a number of different photomultiplier tubes. It is
placed behind a lead collimator. The purpose of the collimator is to limit the

12
gamma rays detected to those gamma rays that are entering perpendicular to the
sodium iodide crystal. By comparing the signals from the different photomultiplier
tubes, one can determine where in the sodium iodide crystal the particular
gamma ray interacted.

Slide 55
The results are shown here, where one is looking at a skeleton of a person that
has been injected with the radioisotope technetium-99m.

Slide 56
Another example of gamma cameras is positron emission tomography or PET.

Slide 57
In this case a large ring of sodium iodide detectors are arranged in a circle.

Slide 58
In this case the patient is injected with a chemical compound that contains a
relatively short lived positron emitter. When the positron interacts in the
surrounding tissue it annihilates producing two half MeV photons that are emitted
at 180 degrees to each other. The sodium iodide detectors are operating in
coincidence so that when two detectors both detect a 0.511 MeV photon at the
same time one knows that the positron emitter is on a straight line between the
two detectors as shown here. By analyzing a large number of events, one can
locate where in the body the compound is located.

Slide 59
A picture of the PET device is shown here. The patient is placed on the table
and slowly passed through the ring of detectors. Since the positron emitters are
concentrated in the organ or tissue of interest, one can pinpoint their location.

Slide 60
A similar device often used in medical diagnostics is the CAT scan device. Here
an electron accelerator focuses a beam of electrons onto a target producing x-
rays. The x-rays are then collimated and focused through the body to a detector.
By rotating a device around the body, one can pinpoint variations in the density
of the body such as bones or tumors.

Slide 61
Another application of scintillation detectors in the laboratory is the use of liquid
scintillation detector systems. These systems are usually used to detect low
energy beta emitters such as tritium or carbon-14 or higher energy beta emitters
such as potassium-40.

Slide 62
These systems use small amounts of liquid scintillator placed in vials such as
shown here. This vial contains 14 ml of a liquid scintillator.

13
Slide 63
The sample, in this case a wipe used to check for lose surface contamination, is
placed directly in the scintillator as shown here. This has the advantage in that it
has a very high detection efficiency.

Slide 64
The scintillation vials are then placed in an automatic counting system such as
the Packard Tri-carb system shown here.

Slide 65
Most automatic liquid scintillation counters have three separate systems that
must be maintained. First the scintillation detector electronics, and data
acquisition systems; second, the refrigeration system; and third the sample
changer.

Slide 66
A diagram of a typical data acquisition system found in many older systems is
shown here. The system uses two photomultiplier tubes, one on each side of the
liquid scintillator vial. This not only increases the detection efficiency of the light,
but also serves to reduce the background. When counting samples for tritium or
C-14, the beta particles have very low energies. This combined with the low
conversion efficiency means that very few light photons are produced. This fact
coupled with the small solid angle that the vials present to the photomultiplier
tube and the efficiency of the photocathodes, means that there may be only one
or two electrons emitted from the photocathode per event. However, the
photocathode with its low work function is continually emitting electrons, so the
background from the system could be very high. These electrons are called the
dark current. To reduce the background, the pulses from the photomultiplier tube
are fed to a coincidence circuit and only those events that have an interaction in
both photomultiplier tubes at the same time are then counted. The pulses from
the photomultiplier tube are also fed to a summation circuit where they are added
together and then fed to a pulse height analyzing system. The pulse height
analyzing system shown here which is typical of older units consists of simply
amplifiers, pulse height analyzers, and scalers.

Slide 67
A similar system using a computer for analyzing the beta spectra is shown here
where the pulses are fed to an ADC or analog to digital converter and then
analyzed by a computer system. This type of system will be found in most of the
newer liquid scintillation systems.

Slide 68
Two beta spectra are shown here. The unquenched spectrum is from a clean
sample, that is a sample that does not affect the performance of the liquid

14
scintillator. However, many materials that are placed in the liquid scintillator will
either affect the conversion efficiency of the scintillator or absorb some of the
light photons before they can escape from the scintillator thus reducing the pulse
amplitude as shown in the quenched spectra. Some of the newer counting
systems will use an external source to check for quenching and then adjust the
amplifier gain to correct the counting rate.

Slide 69
A second system that must be maintained in most liquid scintillation counting
systems is the refrigeration system. Cooling the photocathode reduces the dark
current and lowers the background. The refrigeration system is usually a
mechanical system that also requires regular maintenance.

The third system is the sample changer. Since one wants be able to count many
samples automatically, one needs a mechanical sample changer.

Slide 70
Such a sample changing system is shown here is a Packard Tri-carb system. In
the counter, approximately 200 samples can be placed in the detector and
counted automatically in sequence.

Slide 71
Still another application for scintillation detectors is for the detection of alpha
particles.

Slide 72
In this case, most detectors use a very thin crystal or powder of silver activate
zinc sulfide or ZnS(Ag) coated on the face of the photomultiplier tube. Silver
activated ZnS has a high conversion efficiency, but is not transparent. Therefore,
these scintillators must be very thin. If the scintillator is thin then beta particles or
electrons from photon interactions cannot deposit enough energy to be detected.
The background in the system is due primarily to cosmic radiation and is very
low. These detectors must be placed in a light tight enclosure or have a very
thin, light tight window or covering. The window must be thin enough to allow the
alpha particles to pass through and deposit enough energy in the detector to be
detected.

Slide 73
Such an alpha counting system is shown here. The sample is placed in the tray.

Slide 74
The tray is then closed and the system forms a light tight chamber and the high
voltage turned on.

Slide 75
Scintillation detectors are also used as the probes for survey meters.

15
Slide 76
The major disadvantage of using scintillation detectors as a probe is that they are
more fragile than the typical G-M Detector and also are much more expensive.
The major advantage, however, is that they are much more sensitive to gamma
rays.

Slide 77
One variation of the convectional scintillation detector is to replace the
photomultiplier tube with the photodiode to convert light to charge. A
disadvantage of this system is that one no longer has the amplification or charge
multiplication from the photomultiplier tube, but must rely completely upon
electronic amplification in the pre-amplifier and amplifier. One advantage,
however, is that one gets a much broader range of spectral sensitivity than one
can get with either S11 or bi-alkali photocathodes that one has with
photomultiplier tube. This means that the system will work well with crystals such
as thallium activated cesium iodide. Another advantage is the small size. One
no longer has the bulky, large size of the photomultiplier tube, but rather has a
very small photodiode. This, of course, can also be a disadvantage when one
wants to use large crystals.

Slide 78
The spectral sensitivity of a typical diode is shown here. One can see that
although they will work with wavelengths as low as 440 nm, they work much
better if the wavelength is longer around the range of 500-700 nm. Thallium
activated cesium iodide has a peak around 560 nm, which means that it would
work well with this diode. Other scintillators such as BGO also work quite well
although the light output is much lower.

Slide 79
Some typical shapes offered by the Bicron company are shown here. One can
see the rectangular size with photodiodes on either the end or along one edge of
the scintillator.

Slide 80
Typical sizes are shown here. The largest is a 18 mm x 18 mm x 6 cm long
crystal.

The resolution of such a system is 8.6 keV with the Cesium 137 source. This is
not as good as typical sodium iodide crystals using photomultiplier tubes, but is
very acceptable.

Slide 81
Another geometry is shown here, which is a detector offered by Scionix
Company. Here one has a 5 cm diameter crystal attached to a photodiode.

16
Slide 82
Performance of such a detector is shown in this slide. One can see the 122 keV
peak, the 511 keV peak from positron annihilation and the 1.238 keV peak from a
gamma ray.

Slide 83
Another variation of scintillation detectors are TLD or Thermal Luminescent
Dosimeters.

Slide 84
Going back to the Fermi energy level diagram as shown here, one finds that
when radiation interacts with a TLD material, electrons are raised from the
valence band up to the conduction band. They then decay back down to an
impurity energy level that is located in the forbidden gap. But unlike ordinary
scintillation detectors, where the electron can continue to decay down to the
valence band, with TLD dosimeters, they are trapped at this location. The more
radiation that bombards the TLD Dosimeter, the more electrons will be trapped in
this forbidden gap. Upon heating the TLD, the electrons gain enough energy to
move back into the conduction band and then from the conduction band most of
them will drop back to the valence band and in the process they will emit light
that could be detected by a photomultiplier tube.

Slide 85
TLD Dosimeters come in a variety of sizes and shapes. Some small ones shown
here are approximately 3 mm x 3 mm x 1 mm thick. TLD material may also be in
the form of a powder or in small rods.

Slide 86
A typical TLD detector readout unit is shown here.

Slide 87
The TLDs are placed on a small heater and then moved into a light tight region.
The heater is turned on, heating the TLD, producing small flashes of light that are
read out by the photomultiplier tube.

Slide 88
A newer version of a TLD reader is shown in this system. The latest systems for
TLD readers, use lasers to raise the electrons from the forbidden area up to the
conduction band as opposed to using a heater.

Slide 89
A short summary of scintillation detectors would be as follows. First they operate
at room temperature as opposed to some radiation detectors that require liquid
nitrogen temperatures.

17
Second, they may be a very large size, as large as 50 cm x 50 cm in terms of a
solid inorganic detector or size of large rooms in case of a liquid scintillator.

They are very versatile instruments. They can be used in the laboratory for
measuring radiation or they may be used as hand held survey instruments.

They may have a relatively high density such as sodium iodide or BGO
detectors, which means that they are then good for detecting gamma rays. This
ends the module on scintillation detectors.

18