NIST Measurement Services:

Alpha-Particle NIST
Special
Calibrations (2004) Publication
250–5a

J. M. R. Hutchinson

National Institute of Standards and Technology

Technology Administration, U.S. Department of Commerce
NIST Special Publication 250–5a

NIST MEASUREMENT SERVICES:

Alpha-Particle
Calibrations (2004)

J. M. R. Hutchinson

Ionizing Radiation Division

Physics Laboratory
National Institute of Standards and Technology
100 Bureau Drive
Gaithersburg, MD 20899-0001

(Supersedes NIST Special Publication 250-5, July 1987)

January 2004

E N T OF C O M
TM M
AR

ER
D EP

CE
ICA
UN
IT

ER

D
E

ST AM
ATES OF

U.S. Department of Commerce

Donald L. Evans, Secretary
Technology Administration
Phillip J. Bond, Under Secretary of Commerce for Technology
National Institute of Standards and Technology
Arden L. Bement, Jr., Director
 Certain commercial entities, equipment, or materials may be identified in this document in order to
describe an experimental procedure or concept adequately. Such identification is not intended to imply
recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended
to imply that the entities, materials, or equipment are necessarily the best available for the purpose.

National Institute of Standards and Technology Special Publication 250–5a

Natl. Inst. Stand. Technol. Spec. Publ. 250–5a, 51 pages (Jan. 2004)
CODEN: NSPUE2

U.S. GOVERNMENT PRINTING OFFICE

WASHINGTON: 2004

For sale by the Superintendent of Documents, U.S. Government Printing Office

Internet: bookstore.gpo.gov—Phone: (202) 512-1800—Fax: (202) 512-2250
Mail: Stop SSOP, Washington, DC 20402-0001
 Table of Contents

Abstract................................................................................................................ iv

List of Figures ....................................................................................................... v

List of Tables ........................................................................................................ v

1. Introduction ....................................................................................................... 1

2. Description of Service ....................................................................................... 1

3. Design Philosophy and Theory ......................................................................... 3

4. Description of Systems ................................................................................... 15

5. General Operational Procedures .................................................................... 16

6. Step-by-Step Procedures................................................................................ 18

7. Determination of Uncertainties........................................................................ 24

8. Quality Control and Records ........................................................................... 26

9. References .....................................................................................................27

10. Appendices
A. Table of counting yields ....................................................................... 29
B. Percentage reduction of yields with source depth................................ 35
C. Figure of counting yields with z and energy......................................... 38
D. Figure of counting yields for layered sources ...................................... 39
E. Certificates
i. small 2πα detector ...................................................................... 41
ii. large-area 2πα detector ............................................................. 43
iii. 0.8πα detector........................................................................... 45

iii
 ABSTRACT

This document describes the alpha-particle calibration services offered by the

Radioactivity Group of the National Institute of Standards and Technology (NIST)
(Scheduled Calibrations; 43030C, 43040C, and 43050C). The fundamental
measurement quantities are defined, the measurement approach is described (or
reviewed), and the operating procedures are described from the point of view of
the calibration technician or metrologist. The measurement uncertainties, and
how they are estimated, are described. Methods for maintaining and assessing
quality control, e.g., international comparisons, MQA programs, etc. are also
briefly reviewed. This edition updates and supersedes NIST Special Publication
250-5, July 1987.

iv
 List of Figures

1. Schematic of the small 2πα PC

2. Electrical hook-up for all counters

3. Photo of the small 2πα

4. Relative count rate as a function of counter voltage for the small 2πα PC

5. Relative count rate as a function of source position for the small 2πα PC

6. Schematic of the large 2πα PC

7. Photo of the the large 2πα PC

8. Schematic of the 0.8πα SC

9. Photo of the 0.8πα SC

10. Pulse height spectrum for the 0.8πα SC

11. Count rate versus angle of emission for the 0.8πα SC

12. Counting yield, Y(0), as a function of alpha-particle energy, for a plane source
at the surface (depth x=0) of a backing of atomic number z.

13. Reduction ratio Yav(x0)/Y(0) for uniform layer sources extending to depth x0
below the surface of the backing.
a. For backings with z=4 and 13
b. For backings with z=26 and 78

List of Tables

1. Specifications for calibrations using the 2π proportional counters and the 0.8π
defined solid angle counter
2. Approximate relative scintillation efficiencies of various commercially available
materials at room temperature

3.Listing of count rate correction factors and typical 1σ uncertainties for 2π and
0.8π measurements
4. Counting yield Y(0) of alpha-particle sources at the surface of a backing of
atomic number z, as a function of the source energy E
5. Relation between the counting yield of a plane alpha-particle source at the
surface of a backing, Y(0), and the yield of a uniform layer source extending to a
depth x0 below the surface

v
1. Introduction

Alpha particles emitted from radioactive sources are measured at the

National Institute for Standards and Technology (NIST) in three systems, the 2π
steradian, α-proportional counting system (denoted small 2πα PC), the large area
2π steradian, α-proportional counting system (denoted the large 2πα PC), and
the 0.8π steradian α-CsI(Tl) scintillation system (denoted 0.8πα SC). The full
energy of the alpha particles is absorbed in the counting gas in the first two
systems and in the scintillator in the second system. A ("frequency") spectrum of
alpha-particle counts in a counting period versus alpha-particle energy deposited
in either the gas or scintillator is collected and displayed in a multichannel
analyzer within a computer. Overall counting rates are determined from sums of
all or parts of these spectra as discussed below.

2. Description of Service [1]*: 43030C, 43040C, and 43050C

Alpha-particle sources are submitted by outside laboratories to NIST for

calibration of either their 2πα steradian emission rate or of their total activity.
Emission rate is measured in alpha-particles per second detected in 2π geometry
- in this case by one or both of the 2π proportional counters. Because the
resulting pulse-height distribution is continuous down to zero energy and alpha
particles cannot always be separated from pulses produced by recoiling
daughters and instrument noise in the proportional counter the spectrum must be
extrapolated from above 100 keV down to zero energy. (This is discussed further
in the General Operational Procedures section.) In the case of measurements
with the 0.8πα SC and sometimes the proportional counters as we shall see,
activity, No, is measured in becquerels (Bq), the SI unit which is also
dimensionally "per second," and represents the total number of disintegrations
per second produced by the alpha-particle emitting source. When the 2πα PC’s
are used for activity measurement, conversion factors that when multiplied into
the emission rate for a given source, convert the result to activity for that source.

__________________
*Numbers in brackets indicate literature references at the end of this document.

1
 Calibrations are checked by simultaneously calibrating a previously-
calibrated standard source.

Typical calibration reports for 2πα-emission rate measurement and

0.8πα-particle emission rate are given in Appendices Ei, Eii, and Eiii.

Sources submitted by the customer to NIST for calibration must satisfy

certain requirements. Packaging for all sources must be in compliance with
Department of Transportation (DOT) and Nuclear Regulatory Commission (NRC)
regulations. Copies of regulations may be obtained from Operations Division,
Office of Hazardous Materials, Department of Transportation, Washington, D.C.
20590. Postal regulations prohibit mailing radioactive materials, which require a
caution label under DOT regulations. Alpha-particle solid sources must be
supplied in special source holders such that the active area is not touched by any
material. For sources measured in the small 2πα PC counter (Calibrations
43030C and 43050C) the diameter of the source must be less than 10 cm and
that of the active surface less than 9 cm. For the large 2πα PC (Calibrations
43030 and 43050C) the maximum dimensions on a rectangular source must be
18cm by 30cm. For the 0.8π α SC counter (Calibration 43040C), the maximum
diameter is only 1.6 cm.

Further specifications for these calibration services are given in the

following table. (Table 1)

Table 1. Specifications for calibrations using the proportional counters and the
0.8π defined solid angle counter.

Calibrations 42030C
and 43050C Calibration 43040C
Counting system NIST proportional NIST 0.8 π defined-solid-
counters angle counter
Sources calibrated for Alpha-particle emission Total activity
rate into 2π steradians
Nominal uncertainty 2σ 1.5 percent 1.5 percent
Activity range (0.4 - 104 ) s-1 (40 - 104) s-1
Maximum diameter Small area: 10 cm
1.6 cm
Maximum size Large area: 18 cm by
30cm

2
3. Design Philosophy and Theory *

The proportional counters [2] are used because they record alpha particles with
little interference from emitted beta particles or gamma rays. In proportional
counters, all alpha particles leaving the source into the forward hemisphere
ionize atoms in the counting gas, with the number of ions produced proportional
to the α-particle energy deposited. The electrons released in the ionization are
accelerated in an electric field toward a collecting electrode, multiplying their
number by impact ionization. The resulting electrical pulse is electronically
amplified, and sorted as to pulse height. The fact that the source is "internal"
(inside the counter) allows all emitted particles to be recorded without absorption
by possible counter casings. (For further discussion of proportional counters see
NCRP 58 [3].)

The advantage of these detection systems is that sources up to relatively

large areas can be measured; however, because of the high detection
efficiencies and count rate limitations in the electronics, the measurements are
limited to sources with activities of less than 20,000 Bq. At the lower end, count
rates should be greater than 10 times background, which is on the order of 0.06
s-1.

3.1. The small 2πα Proportional Counter (small 2πα PC)

The calibration of alpha-particle sources started at the National Institute for

Standards and Technology in the early 1950's [4]. The alpha-particle detector
used consisted of a 12-cm-diameter hemispherical gas-flow proportional counter
as shown in Figs. 1, 2, and 3.

* Excerpted in part from Lucas [2]

3
12 CENTIMETER DIAMETER PROPORTIONAL COUNTER CHAMBER
(small 2πα PC)

Fig. 1. Schematic diagram of the small area 2πα PC

4
 Fig. 2. Hook-up of electronic components of all three counters

This counter has been in use constantly since that time and is still much
used today. Description of the characteristics and usage of such detectors
follows.

Fig.3. Photograph of the small 2π α PC

5
 Extrapolation to Zero Pulse Height - Regardless of the type of detection
system used, an accurate determination of the total alpha-particle count rate
requires that the alpha-particle pulse-height spectrum be recorded and extrap-
olated to zero pulse height, as shown in Fig. 4. The high-count rate at low
channel numbers is primarily due to system noise. At NIST we generally fit to a
second-order polynomial, although a simple first-order polynomial (straight-line
extrapolation) is often adequate. When the pulse-height spectrum is extrapolated
to zero pulse-height in this way, the total count rate remains almost constant
despite a large change in the system gain, whereas the count rate above a fixed
discrimination level varies significantly with the same change in the system gain.
The effect of using an extrapolation-to-zero pulse height is demonstrated in Fig. 4
in terms of a plot of relative count rate versus counter voltage and in Fig. 5 in
terms of a plot of relative count rate versus point-source position.

Fig.4. Relative counting rate versus counter voltage for the small 2πα PC
(see Fig. 1). An 241Am point source was located in the center of the
chamber. The counting gas was P-10 at atmospheric pressure. The
solid line is the count rate data using an extrapolation to zero pulse
height. The dashed line represents data from count rates above a fixed
discrimination level.

6
Fig.5. Relative count rate versus position for an 241Am point source in the small
2πα PC (see Fig. 1). The counter voltage was 1400V. The counting gas
was P-10 at atmospheric pressure. The solid line is the count rate
extrapolated to zero pulse height. The dashed line is the count rate
above a fixed discrimination level.

Maximum Source Size - For uniformly high detection efficiency (1.000 ±

0.005) from the surface of an extended source, the source diameter should not
be larger than about one half of the counter diameter. The maximum permissible
source size using this criterion depends upon the electric field distribution within
the counter (especially the shape and position of the anode wire loop) and upon
the composition of the counting gas (for example, it is larger for P-10 than for
methane [5]) and can be determined for any given counter by plotting the count
rate versus the position of a point source. Fig. 5 shows such data for the small
2πα PC at NIST under normal operating conditions (1 atm. of P-10 gas, 1400 V).
Note the difference when the pulse-height spectrum is extrapolated to zero pulse
height.

Counting Gas - In a proportional counter the counting gas serves two

purposes:

1. It provides a medium in which an alpha particle can dissipate its

kinetic energy to form free electrons and ionized atoms or molecules.

7
 2. It provides a medium in which the free electrons so formed can be
accelerated in the presence of an electric field to produce additional ionization
(that is, electron multiplication and hence amplification or "gas gain").

Methane and P-10 (90 percent argon, 10 percent methane) are the gases
most often used for gas-flow proportional counters. The gas of choice in any
given system or for any given application depends upon a number of factors,
including cost, availability, purity, and background count rate. There are also
other considerations. A much higher electric field ( that is, a much higher anode
voltage) is required in methane than in P-10 for the same average electron
multiplication, and the electron multiplication in methane is less affected by the
presence of small concentrations of electronegative gas impurities, such as O2
and H20 [5]. However, if the pulse-height spectrum is extrapolated to zero pulse
height, the effect of the change in system gain due to the presence of impurities
in the counting gas will be minimal, even with P-10.

Scattering - For reasons of stability and convenience, an alpha-particle

source is commonly mounted on a flat metal backing. When such a source is
measured in a detector having 2π geometry, the 2π counting rate, C2π, is
generally not one half of the activity, No, because some of the alpha particles
initially emitted downward are backscattered into the sensitive volume of the
detector, while some of the alpha particles initially emitted upward are scattered
or absorbed in the source or both. As a result, the measured C2π,/No ratio can
vary significantly from 0.50. This is shown in Appendices A, B, C, and D, which
are from the latest and most comprehensive [6] of a number of papers dealing
with this subject. Note that, even for a weightless source, the correction can be
very significant.

3.2 The NIST large area 2π large area α (large 2πα PC) proportional counter.
Figs. 6 and 7 show this counter.

Fig.6. Schematic diagram of the NIST large 2πα PC (not to scale)

8
 Fig.7. Photograph of the large area counter.

The functioning of the large 2πα PC is virtually identical with that of the small 2πα
PC. All above descriptions apply.

3.3 0.8πα Scintillation Counter

The 0.8πα SC, the so-called Robinson counter [7], after the designer of the
instrument, is a CsI(Tl) scintillation counter, with defined solid angle, and a three
dimensional central baffle (Figs. 8 and 9). Alpha particles striking the scintillator
produce light flashes proportional to the energy imparted. These flashes are
transformed to electrical pulses and amplified for pulse-height analysis. The
baffle is designed such that the detection efficiency is insensitive to positional
changes in the source with respect to the detector. As shown in [7], a 10 mm
vertical or horizontal displacement of the source results in a count rate change on

9
the order of 0.1 percent. This detector should not be used with sources with
diameters greater than 1.6 cm. However, since the efficiency is approximately 40
percent of that of the 2π counters, maximum source activities can be
correspondingly greater. Unfortunately, the background is significantly higher
than that in the 2π counters and, therefore, the minimum activity sources are
significantly higher - on the order of 40 Bq. A pulse height spectrum is shown in
Fig. 10.

Fig. 8. Schematic of 0.8πα SC.

10
Fig. 9. Photograph of the 0.8πα SC.

11
 Fig. 10. The pulse-height spectrum from a sample of 239Pu in the 0.8πα SC
showing the extrapolation to zero pulse.

In about 1960 an experimental investigation of the energy and angular

distributions of the emitted and scattered alpha particles, was carried out at
NIST by Walker [8]. Her results for all emitted alpha particles at various angles
are shown in Fig. 11 Backscattered alpha particles are all emitted within about
200 of the plane of the source (70° to 90° in Fig. 11). Therefore if the solid angle
subtended by the detector is such as to exclude all alpha particles emitted within
20° or more of the plane of the source, then the scattering correction, and the
uncertainty associated with it, is eliminated. The slight slope in the central
portion of the curve was apparently due to the center of rotation of the detector
not being coincident with the source. Figure 11 also shows the angles
subtended by several of the defined-geometry counters for the standardization
of alpha-particle sources.

12
Fig. 11. Count rate versus angle of emission for alpha particles emitted by 210Po
that had been deposited on the surfaces of platinum and monel disks.
(0 is perpendicular to the surface of the disk.) See Ref. 8 for complete
details. Also shown are the angles subtended by the detectors in
several defined-geometry counters.

Several defined-geometry counters have been described in the literature

[7,9,10,11]. One problem that arises in the use of such counters, especially
those with large solid angles, is that the geometry is very dependent upon the
position of the source in the counter. The design described by Robinson (9)
almost eliminates such position dependence over a wide range of horizontal and
vertical displacements of the source.

To obtain the highest accuracy from a defined-geometry counter, it is

desirable to minimize the energy loss and the scattering in the region between
the alpha-particle source and the collimator(s) and detector. As a result most

13
[9,10], but not all [7,11], counters of this type are operated with the chamber
under vacuum and with a solid scintillator as the alpha-particle detector. A
question that always arises is, "What is the intrinsic detection efficiency of the
solid scintillator?" Our results at NIST indicate that the alpha-particle-detection
efficiencies of our commercially prepared inorganic scintillation crystals [cesium
iodide (thallium) and calcium fluoride (europlum)] and organic scintillation
plastics are at least 0.999. The alpha-particle-detection efficiencies of zinc
sulfide (silver) layers are considerably more variable because this material is
available only in the form of microcrystalline powder that must be deposited as a
thin layer (often with a binder) to form a suitable detector. Nonetheless, for
carefully prepared deposits, efficiencies of 0.999 appear to be reproducibly
attainable [9]. There are, however, important differences between the various
solid scintillators, especially in relative light output for alpha particles compared
with beta particles and gamma rays. Table 2 lists a number of commonly used
scintillators, their relative light outputs, and their α/β response ratios.

Table 2. Approximate relative scintillation efficiencies of various commercially

available materials at room temperature. The relative light output of
sodium iodide (thallium) per unit energy deposited by beta particles is
taken as 1.00.

Solid scintillator α β α/β

ZnS(Ag) 2.0 1 2.0

CsI(Tl) 0.4 0.5 0.8

NaI(Tl) 0.5 1.00 0.5

CsBr(Tl) 0.05 0.15 0.3

KBr(Tl) 0.01 0.04 0.25

CaF2(Eu) 0.1 0.4 0.25

NaCl(Ag) 0.01 0.04 0.2

Organic 0.02 0.25 0.08

Ionization
detectors 1

14
 If one is counting a source that emits only alpha particles, the a/β response
ratio is not of significance. Mechanical and chemical properties would be the
basis on which to select a scintillator. But if the source also emits beta particles
or gamma rays or both, or if these radiations constitute an important part of the
background, then the a/β response ratio may be the most important parameter in
the choice of a scintillator. For example, suppose that one is counting 5.5-MeV
alpha particles in the presence of 500-keV conversion electrons and that both
types of particles deposit their full energies in the scintillator.* If the scintillator is
zinc sulfide (silver), the light output due to the alpha particles will be more than
20 times that due to the conversion electrons. Hence the conversion electrons
may be readily discriminated against on the basis of pulse height. If the
scintillator is an organic plastic, the light outputs due to the two types of particles
will be the same. In this case it will not be possible to discriminate against the
conversion electrons on the basis of pulse height.

4. Description of Systems

4.1. Small and large 2πα PC’s: The 2πα proportional counter systems are
pictured in Figures 1, 2, 3, 6, and 7. The source is centered inside each
counter and sits on the aluminum base plate. Alpha particles emitted into the
sensitive volume ionize counting gas (P-10) atoms, which are attracted and
cascade to the positively charged anode loop where they are collected. The
pulse of charge is presented to a charge sensitive preamplifier after which the
pulse is amplified to overload. The resulting spectrum is collected in a multi-
channel analyzer within a computer system.

By amplifying to overload, the condition is nearly satisfied that all alpha

particles, regardless of energy, will be detected in the data collection system.
Consequently under these conditions, extrapolation to zero pulse height is more
easily accomplished because the low-energy region can be examined in detail.

4.2. 0.8πα SC: The 0.8πα CsI(Tl) scintillation "Robinson" counter is shown in
Figs. 8 and 9. The source is mounted on the "source mount" and the enclosed
volume between it and the scintillator is flushed with hydrogen gas to permit
the alpha particles to move with minimum loss of energy. The outside baffle
and the Al stop are accurately machined with tolerances on the order of 0.005
cm so that the detection geometry is known to approximately 0.1 percent. Light
pulses are collected in the light pipe and passed into the phototube. The
resulting pulses are amplified to overload as with the 2π systems and the
spectrum is collected in a computer.

__________________
* In actual practice, one generally tries to minimize the energy deposition of the
beta particles and gamma rays by using a solid scintillator that is just thick
enough to stop the alpha particles.

15
 Without the inside baffle, the source geometrical factor would be sensitive
to vertical positioning of the source. As explained by Robinson [9], the three
dimensional central baffle is so designed that changes in the geometrical factor
defined by the outside baffle due to changes in source position, are offset by
equal and opposite changes in the screening of the alpha-particle beam by the
central baffle. Hutchinson et al. [7] have shown that for a vertical displacement of
0.4 cm or a horizontal displacement of 1 cm of the source from the central
position, the change in the count rate varies by less than 0.3 percent.

A big advantage of this counter relates to the fact that multiply-scattered

alpha particles, essentially all of which are emitted at grazing angles to the
source, are not detected. The difficult problem of estimating the back-scattering
contribution is thus avoided.

The total efficiency ε0.8π , for this geometry, approximately 0.8π steradians,
of the counter was deduced from measurements on thin sources with the small
2π and 0.8π counters for different values of atomic number, Z, of the backing
material [12].

The 2π count rate, N2π divided by the 0.8π count rate, N 0.8π was
plotted as a function of Z, and extrapolated to Z = 0 where, theoretically, ε0.8π
= N0.8π/2N2π = N0.8π/N0. A schematic diagram of the spectrum and extrapolation
to zero pulse height for the 0.8π counter is shown in Fig. 10. The multi-
channel analyzer is set to integrate counts above a level just above noise. The
number of true counts in the noise region are estimated by means of the
extrapolations.

5. General Operational Procedures

5.1. Small and large 2πα PC’s: After the source has been introduced into the
counter and the sensitive volume flushed until the pulses observed on the pulse-
height analyzer have reached their maximum height, the counting procedure is
initiated. Typically, the counting proceeds in the order: background, standard
reference source, submitted source, background. The counting times are
adjusted so that 106 counts from the source are collected (corresponding, of
course, to 0.1 percent statistical uncertainty). These total counts are collected
usually over five repetitions, or approximately 2 x 105 counts per measurement.

The functioning of the instrument is checked by comparing the

measurement results for the standard, corrected for decay, and the background
with previous results.

16
 Five collection windows are set covering the lower end of the spectrum for
the purpose of obtaining an extrapolation in the case where the total number of
counts into 2π are required. The extrapolation is based on the assumption that
the "true" tail of the spectrum (e.g. as in Fig. 10) can be represented as a flat
horizontal line with a height corresponding to the spectrum minimum.

So-called type B errors are estimated at the lσ level and are added to type
A errors in quadrature. The overall uncertainty quoted in the certificates is two
times the calculated combined uncertainty [13] as shown in Appendices Ei, Eii,
and Eiii.

5.2. 0.8πα SC: Generally the same procedure for data collection and reduction is
employed with the 0.8πα SC as with the 2π counters. Only activity measurements
are made with the 0.8πα SC, therefore all measurements involve extrapolated
values to which a calibration factor of 1/ε0.8π = 5.0638 is applied (see reference
6). A sample certificate is shown in Appendix Eiii. The one significant operational
difference from the 2π counter is that the correct source height must be
determined by varying the vertical position of the source through screwing the
mount in or out as required. The count rates will go through a maximum with
changing vertical adjustment representing the correct vertical setting.
Source positioning is achieved in the horizontal and vertical directions as follows:

Horizontal - The available spacers fit most source diameters that we

encounter. For those sources of a different diameter, a new spacer can be
made, or the source may be centered on the block and secured with a
minimum of double-stick tape. The tape method is not recommended for
thin source mounts such as thin platinum.

Vertical - The optimum vertical position for counting a source will vary with
the thickness of the source mount, and the type of deposit (electroplated,
vacuum evaporation, or dried deposit). This position is determined by
counting the source (or one of a batch of similar sources) at the "0" turns or
stop position, which is the point at which the block is against the bottom of
the counter. Then, count at each 1/4 or 1/2 turn down, to the point where the
count obviously decreases. These measurements can be plotted as count-
rate-vs-breechblock position, and the operating point selected. The position
which is thus determined can then be used for sources of the same
configuration.

17
6. Step-by-Procedures

6.1 Preliminary

• Customer contact - give specifications for physical dimensions and activity

limits, emphasize that source container must protect active area from
contact with anything within the container, and send calibration booklet to
the customer.

• Inspection for damage - if damage has occurred, the customer will be

notified before proceeding with the calibration.

• Test check measurements - perform measurements of background and a

calibrated test source to make sure the system is operating correctly.

• NIST paperwork and acceptance procedure- submit completed NIST 364

for approval before arrival of materials, notify the Chief of Health Physics
of the arrival and departure of all plutonium and uranium, complete other
necessary forms (NIST 64,77,796A).

Test folders - request test folder from OPMS on receipt of material to be

calibrated. Note dates of material received and returned.

6.2 Counting Procedure

The measurements are taken in the following order: background, standard

reference source, submitted source, background. The counting times are
adjusted so that 106 counts from the source are collected. The functioning of the
instrument is checked by comparing the measurement results for the standard,
corrected for decay and background, with previous results. The data is reduced
as described in the following sections and certificates are prepared and
submitted to the customer.

Procedures for calibrations using three instruments are described: (1) small
2πα PC used for thin alpha-particle sources up to a diameter of 10 cm, (2) large
2πα PC for large area sources up to a rectangular shape of 10 cm by
15 cm, and (3) the 0.8πα SC used for the determination of source activity for thin
sources up to a diameter of 2.5 cm.

18
6.3 Set-up Procedure for All Counters

• Introduce source into the counter.

• Turn on the High Voltage and other electronics.
• Turn on the air pressure valve Flush P-10 gas through the system until the
pulses recorded on an oscilloscope reach a maximum value.

6.4 2π Alpha- Particle Step-by-Step Procedure

To Start

• Turn on the electronics.

• Turn on the air pressure valve.
• Raise top of counter by pulling handle up slowly.
• Place source in the center of the counter, handling the source with gloves to
prevent possible contamination.
• Lower top of counter by pulling handle down slowly.
• Turn on P-10 gas flow-pressure at 7 psi.
• Open black knob on regulator and flush rapidly for 10 minutes until pulses on
the oscilloscope have reached a maximum.
• Turn on high voltage (1450 volts) at the start of the flushing.
• Reduce flow to approximately 60 cc per minute.
• Use the following settings for the Tenelec TC 247 Dual Amplifier –coarse gain
of 200 and fine gain of 10.0.

Data Collection

• Feed the pulses from the amplifier into the computer which uses a MAESTRO
program services.
• Initiate job control.
• Use file named: prop1.job.
• Edit file to establish set-up parameters including:
- number of sets of measurements(spectra)
- elapsed live times
- sample description
-file name
-counter description.
• Collect spectral data (hit OPEN).
• Hand record history of measurements.

19
Data Reduction

When the data collection is complete, bring up file called SPECTINC, which sets
6 regions to be used in an extrapolation procedure which gives the integrated
total count-rate.
Typically the regions would be channel integrals from 0-20, 21-40,41-60,61-
80,81-100,101-1000.
Then, on EDIT file, type number of files to be reduced.
Run program called NLEVCON.EXE, which executes the data reduction into
counts per second for each integral region specified above.
As a check on the results arrived at above, run the data called NLEVEL.EXE
which performs the extrapolation automatically.

To Change Source

• Cut off High Voltage.

• Turn off gas flow at black knob.
• Raise top of counter by pulling handle up slowly.
• Remove source carefully.
• Place next source in center of counter.
• Lower counter top slowly.
• Turn on gas with black knob.
• Continue as above.

To Shut Down

• Turn off electronics.

• Turn of gas (close black knob, main valve, and release the diaphragm control)
• Turn off air-pressure valve.
• Take smears of counter and immediate area.

6.5 Large Area 2πα Proportional Counter

To Start

Turn on the electronics

• Turn on the air pressure valve to 40 psi.
• Release 6 stops on the base-plate movement-switch in down position.
• Lower base-plate handle down slowly.
• Place source in the center of the counter, handling the source with gloves to
prevent possible contamination.
• Raise baseplate by switching on in up position.
• Turn on P-10 gas flow.
• Open black knob on regulator and flush rapidly at 3.0 cubic feet per hour for
10 minutes.

20
• Turn on high voltage (1150 volts) at the start of the flushing.
• Reduce flow to approximately 1.5 cubic feet per hour.
• Use the following settings for theTenelec TC 247 Dual Amplifier –coarse gain
of 200 and fine gain of 10.0.

Data Collection

• Feed the pulses by datalink from the amplifier into the computer which uses a
MAESTRO program services.
• Initiate job control.
• Use file named: prop1.job.
• Edit file to establish set-up parameters including:
- number of sets of measurements(spectra)
- elapsed live times
- sample description
-file name
-counter description.
• Collect spectral data (hit OPEN).
• Hand record history of measurements.

Data Reduction

When the data collection is complete, bring up file called SPECTINC, which sets
6 regions to be used in an extrapolation procedure, which gives the integrated
total count-rate.
Typically the regions would be channel integrals from 0-20, 21-40,41-60,61-
80,81-100,101-1000.
Then, on EDIT file, type number of files to be reduced.
Run program called NLEVCON.EXE, which executes the data reduction into
counts per second for each integral region specified above.
As a check on the results arrived at above, run the data called NLEVEL.EXE
which performs the extrapolation automatically.

To Change Source

• Cut off High Voltage.

• Turn off gas flow at black knob.
• Lower base-plate.
• Remove source carefully.
• Place next source in center of counter.
• Raise base-plate slowly.
• Turn on gas with black knob.
• Continue as above.

21
To Shut Down

• Turn off electronics ( High Voltage and NIM BIN).

• Turn of gas (close black knob, main valve, and release the diaphragm control)
• Turn off air-pressure valve.
• Take smears of counter and immediate area.

6.6 0.8π Defined Solid-Angle-Alpha-Particle Counter

To Start

• Turn on electronics (NIM BIN) and high voltage interlock box.

• High voltage power supply must be in the off position.
• Remove electrical connection from interlock to breechblock.
• Carefully unscrew breechblock which is also the source mount.
• Be ready to support the weight of the breechblock (approximately 2000 grams
when it comes free).
• Place source in center of breechblock (spacers for sources of various
diameters in drawer under counter).
• Replace block and turn to stop position carefully. (Care should be taken not to
advance source to the point of contact with the central baffle).
• Replace electrical connection from the interlock.
• Turn on gas (hydrogen) at main valve and adjust diaphragm control to about
10 psi. (The gas is exhausted directly to the outside.)
• Open black knob between regulator and flow meter and flush for about 10
minutes, at 3 L min-1 as indicated on the flow meter.
• Turn on high voltage, (+) 1500, at start of flushing.
• Reduce flow rate to about 0.5 L min-1; observe flow rate on flow meter.
• Connect amplifier output to analyzer at attenuator input; coarse gain + 4, fine
gain = 3.20.
• Counting may begin.

Data Collection

• Feed the pulses by datalink from the amplifier into the computer which uses a
MAESTRO program services.
• Initiate job contro.l
• Use file named: prop1.job.

22
Edit file to establish set-up parameters including:
- number of sets of measurements(spectra)
- elapsed live times
- sample description
-file name
-counter description.
• Collect spectral data (hit OPEN).
• Hand record history of measurements.

Data Reduction

When the data collection is complete, bring up file called SPECTINC, which sets
6 regions to be used in an extrapolation procedure, which gives the integrated
total count-rate.
Typically the regions would be channel integrals from 0-20, 21-40,41-60,61-
80,81-100,101-1000.
Then, on EDIT file, type number of files to be reduced.
Run program called NLEVCON.EXE, which executes the data reduction into
counts per second for each integral region specified above.
As a check on the results arrived at above, run the data called NLEVEL.EXE
which performs the extrapolation automatically.

To Change Source

• Turn off high voltage.

• Turn off gas flow by closing black knob.
• Remove interlock lead.
• Remove source carefully.
• Take smears of block and counter area and count at Health Physics.
• Replace block and turn to stop position.
• Turn off interlock control.

To Shut Down

• Turn off electronics (high voltage and NIM bin).

• Turn off gas (close black knob, main valve, and release diaphragm).
• Remove interlock lead.
• Remove source carefully.
• Take smears of block and counter area and count at Health Physicslace block
and turn to stop position.
• Replace block and turn to stop position.
• Turn off interlock control.

23
7. Determination of Uncertainties

The corrections and uncertainty components considered are given

in Table 3.

Table 3. Listing of count rate correction factors and typical 1σ uncertainties for 2π and
0.8π measurements.
_________________________________________________________________________________
2π Counters 0.8π Counter

Typical Typical Typical Typical

Correction
Uncertainty Correction Uncertainty
Factor (%) Factor (%)
Correction 1σ 1σ
______________________________________________________________________________________

Counter geometry 2.000 0.05 40.000 0.05

Extension and non-uniformity of source 1.000 0.05 1.005 0.05

Self-absorption and scattering from 1.100 0.5 1.000 0.05

Source and support

Scattering off counter wall 1.000 0.02

Transmission through collimator edge 1.000 0.02

Scattering in/on detector 1.000 0.01 1.000 0.01

Transmission through detector 1.000 0.01 1.000 0.01

(no count)

Background 0.999 0.01 0.990 0.01

Deadtime 1.010 0.01 1.010 0.01

Extrapolation to zero pulse height 1.050 0.5 1.010 0.5

Counting Statistics 0.1 1.000 0.1

______________________________________________________________________________________

24
 Although this table lists typical uncertainties, one could, by using these
numbers, derive typical total uncertainties for the various measurements by
combining the relevant uncertainty components in quadrature and multiplying by
an arbitrary factor of 2. The measurements and these "typical" uncertainties in
percent are follows:
In Table 3 typical corrections and uncertainties associated with the deter-
mination of alpha-particle activity are given. A source with a thickness equal to
one fourth of the alpha-particle range., and an activity of 2000s-1 is assumed.
These uncertainties will vary in individual cases depending on such factors as
source size, source uniformity, alpha-particle energy, impurities, interfering
radiations and count rate to background ratio. Appendices E give examples of
uncertainty computation in three specific cases. Uncertainties for other sample
measurements will, in general, be different from these two examples but are
nearly always within 0.5 percent in the quoted overall uncertainty. A number of
the entries in Table 3 have been considered and evaluated but are normally
negligible and are not included in the examples given in Appendices E.

The uncertainty estimates are given at roughly the lσ level. For the most
part the table is self-explanatory. Symbols used in this section are identified
in Table 3.where the symbols are defined in Table 3. Now follow some notes
on the sources of error:

7.1. Counter Geometry: The uncertainty in the detector solid angle Ω(0.8π) is
estimated from possible differences between the calculated values, previously
described, based on alternate extrapolations to Z = 0.

7.2. Extension and Non-Uniformity of Source: If the source is large enough, it will
impinge on the region of point-source position for which the response of the
detector is significantly (>0.1 percent) different from that at the center. These
effects have been previously described in the Design Philosophy and Theory
Section.

7.3. Self-Absorption and Scattering from Source and Support: This effect is very
much larger for 2π sources because, as previously pointed out, back and
forward scattered particles involve those alphas which are coming off the source
at grazing angles. For even moderately thick sources, this correction could be
quite significant for 2π sources yet be completely insignificant in 0.8π geometry
because the 0.8π counter measures those forward directed, relatively scatter-
free, particles. As discussed on page 11 under Scattering, the activity of an
alpha-particle-emitting source can be obtained from the 2π extrapolated 2π
counting rate using a theoretical 2π/No counting rate [8].

25
7.4. Scattering off Counter Wall: This effect does not exist in the 2πα PC's
counter because alpha particles do not reach the wall. The dimensions of the
0.8π counter are such that very rare large-angle scattering would be required to
produce such events.

7.5. Transmission through Collimator Edge: This applies only to the 0.8πα SC
and is negligible based on the range of alpha particles in the collimator material.

7.6. Scattering in/on Detector: This is a conceivable, but entirely negligible effect
for these counters.

7.7. Background: This is not observed as being variable with time and could
supposedly be driven down through collection of massive statistics. Generally,
enough counts are accumulated to reduce the background uncertainty
contribution to less than 0.1 percent.

7.8. Dead-time: The expression for dead-time correction is given in the mea-
surement equations for uncompensated systems. However, the multichannel
analyzer, which is the dominant source of dead-time losses, has a highly precise
dead-time compensator so that for a one percent dead time correction, for
example, the uncertainty due to this effect is entirely negligible.

7.9. Extrapolation to Zero Pulse Height: The shapes of the spectra in the low
energy region have significant curvature. Consequently, an extrapolation based
on the minimum value has significant uncertainty.

7.10 Counting Statistics: The standard deviation of the mean of repeated counts
is taken as the basic measure of statistical uncertainty.

The overall uncertainty is estimated by summing in quadrature, the

standard errors of all calculated or estimated uncertainties and multiplying by a
factor of three which then gives approximately a 99 percent confidence level.

8. Quality Control and Records

The day-to-day control of the instruments uses a known standard, which is

measured at the time of a calibration.

Cross-checks between the counters and other specialized counters such as

the 0.1π counter [4] are also performed. For example, a given calibration will be
performed and a quick cross-check of this source on another counter will be
made. Such checks provide a feeling for the quality of a source, a check on the
functioning of the calibration counting system, and cross calibration in the
(unlikely) event that questions are raised at a later date.

26
 Records maintained are a customer log sheet including customer name,
date received, kind of source, source number and identification including
Radioactive Source (RS) number and test number, date calibrated, date returned
to customer. In addition an alpha calibration record book includes hard copy of
the data, data reduction procedures, calibration results, and copy of the
certificate. Computer records include information contained in the customer log
sheet and the alpha calibration record book.

The only large international intercomparison of alpha-particle-emitting

sources was performed under the sponsorship of the B1PM in 1963 on a solution
of 241Am. The agreement of NIST (then NBS) with the international average was
in the 0.1 percent range. It was concluded at that time that the calibration of
alpha-particle emitters was well understood and no further international
comparisons have been made on alpha-particle emitters.

9. REFERENCES

1. NBS Special Publication 250, "NBS Calibration Services Users Guide 1986-
88," G. A. Uriano, et al., Editors, U.S. Department of Commerce, National Bureau
of Standards, U.S. Government Printing Office, Washinton, D.C. July 1986.

2. The Standardization of Alpha-Particle Sources. L.L. Lucas. Proceedings of

the ASTM Conference of Effluent and Environmental Radiation Surveillance, July
9-14, 1978, Johnson, Vermont, in ASTM Spec. Tech. Publ. 698. American
Society for Testing and Materials (ASTM), pp. 342-354 (1980).

3. NCRP Report 58, A Handbook of Radioactivity Measurements Procedures,

Section 3.7 "Alpha Particle Counting", Mann, W.B. (ed) National Council on
Radiation Protection and Measurements, Washington (1978).

4. Temmer, G.M. and Wycoff, J.M., "Alpha-Ray Measurements Program"

National Bureau of Standards, Report No. 2598, 1953.

5. Hawkings, R.C., Merritt, W.F., and Craven, J.H. in Recent Developments and
Techniques in the Maintenance of Standards, Her Mayesty's Stationary Office,
London, 1952, p. 35.

6. Berger, M.J. “Counting Yields for Beta and Alpha Particle Sources”, NIST
Internal Report, NISTIR 6464(2000).

7. Hutchinson, J.M.R., Lucas, L.L., Mullen, P.A. "Study of the Scattering

Correction for Thick Uranium-Oxide and Other a-Particle Soures-II:
Experimental", Int. J. Appl. Radiat. Isotopes, 27, 43 (1976).

27
8. Walker, D.H. "an Experimental Study of the Backscattering of 5.3-MeV Alpha
Particles from Platinum and Monel Metal", Int. J. Appl. Radiat. Isotopes, 16, 183
(1965).

9. Robinson, H.P. in Metrology of Radionuclides, International Atomic Energy

Agency, Vienna, Austria, 1960, p. 147.

10. Jaffey, A.H., Flynn, K.F., Bentley, W.C., and Karttunen, J.O., Phys. Rev. C
9, 1974, p. 1991.

11. Bambynek, W.B. in Standardization of Radionuclides, International Atomic

Energy Agency, Vienna, Austria, 1960, p. 147.

12. Hutchinson, J.M.R., Lucas, L.L., Mullen, P.A. "Study of the Scattering
Correction for Thick Uranium-Oxide and Other a-Particle Sources-II:
Experimental", Int. J. Appl. Radiat. Isotopes, 27, 43 (1976).

13. Giacomo, P. "News from BIPM, " Metrologia 17, 73 (1981).

14. Taylor, B.N., and Kuyatt, C.E. “Guidelines for Evaluating and Expressing
the Uncertainty of NIST Measurement Results”, NIST Technical Note
1297(1994).

28
 10. APPENDICES

Appendix—A

29
30
31
32
33
34
Appendix—B

35
36
37
Appendix—C

Fig. 12. Counting yields, Y(o), as a function of alpha-particle energy, for a plane
source at the surface (depth x=o) of a backing of atomic number Z.

38
Appendix—D

Fig. 13. Redaction ratio Yav (x0)/Y(o) for uniform layer sources extending to
depth x0 below the surface of the backing
a. for backings with Z=4 and 13
b. for backings with Z=26 and 78.

39
Appendix—D (continued)

Fig. 13b

40
 APPENDIX--Ei

41
 NOTES

(1)
The total number of alpha particles counted per second
emitted into
a 2π-steradian geometry (including those scattered).

(2)
The overall uncertainty is two times the value found from
combining quadratically the standard deviations of the
mean, or approximations thereof, of the following:

a) one standard deviation of the mean

of five measurements 0.12
percent

b) pulse-height extrapolation 0.70

percent
Percent difference in the certified value
between the estimate based on horizontal
extrapolation from the minimum point on
the spectrum to that from the same spectral
point to zero count rate at zero energy.

c) system live time 0.1

percent
Using the two-source method, the live time
was verified to within 0.1 percent.

For further information, contact Michael P. Unterweger at (301)

975-5536, or Pamela A. Hodge
at (301) 975-5544.

Source identification 7404

NIST Test No. 268916-03

42
 APPENDIX--Eii

43
 NOTES

(1) The total number of alpha particles counted per second emitted into a 2π-
steradian geometry (including those scattered).

(2) The overall uncertainty is two times the value found from combining
quadratically the standard deviations of the mean, or approximations thereof,
of the following:

a) one standard deviation of the mean

of five measurements 0.03 percent

b) pulse-height extrapolation 0.70 percent

Percent difference in the certified value

between the estimate based on horizontal
extrapolation from the minimum point on
the spectrum to that from the same spec-
tral point to zero count rate at zero energy.

c) system live time 0.1 percent

Using the two-source method, the live time

was verified to within 0.1 percent.

For further information, contact Dr. M.P. Unterweger at (301) 975-5536, or Pamela
A. Hodge at (301) 975-5544.

Source BV691
NIST test No. 268749-03

44
 APPENDIX--Eiii

45
 NOTES

(1)
The overall uncertainty is two times the value found from combining quadratically
the standard deviations of the mean, or approximations thereof, for the following:

a) one standard deviation of the mean of

five measurements 0.07 percent

b) pulse-height extrapolation to zero energy 0.70 percent

Percent difference in the certified value

between the estimate based on horizontal
extrapolation from the minimum point on
the spectrum to that from the same spec-
tral point to zero count rate at zero energy.

c) system live time 0.1 percent

Determined with a gated oscillator (∀ 0.1

percent) plus 1 percent multiplied by the
live time divided by the dead time.

(2)
Defined-solid-angle counter with scintillation detector.

For further information, please contact M.P. Unterweger at (301) 975-5536 or Pamela A.
Hodge at (301) 975-5544.

Source identification 4904SG-34

46