.
4
.
-
1
e.
.
-
bo
.
i
I OF I
ORNL
P 2105
-
EX
every
.
-
1
.:
1
50
I56
3.6
14.0
E:
IN
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963
.
o row -p-2/05

CFSTI PRICES:
1966
H.C. $ locus; MN, 550
MAY
PHYSICAL PROBLEMS OF DOSIMETRY OF NEUTRONS AND PROTONS
Cos-660472-1
J. A. Auxier
Health Physics Division
Oak Ridge National Laboratory
Oak Ridge, Tennessee
REHEASEO POR ANNOUNCEVENT
IN NUCLEAR SCIENCE ABSTRACT:S
The problems associated with the physical dosimetry of neutrons and
protons are functions of several parameters including the coexistence of
gamma or other radiation fields, the range of dose rates or particle
fluences that may be encountered, the range of energies of the particles
and associated electromagnetic radiation that may be encountered, and
various geometrical fact:urs such as the shielding configurations, shield-
ing parameters, and size of the biological target. For particle energies
of less than 20 Mev, exposures of personnel to direct proton beams are not
a problem if reasonable precautions are taken to protect the worker. For
particle energies in the range of 20 to N250 Mev, the dosimetry for protons
is much simpler than for neutrons except for the sharp variation in dose
as a function of position in a large target such as man. The distribution
of dose rises slowly over most of the track length, rises more markedly
near the end of the particle tracks, and then falls off abruptly.
Consequently, a measure of the proton fluence and energy range is the
most important problem. For neutrons in this energy region, there is an
initial buildup which may be quite sharp and then a very gradual decrease
in both particle fluence and dose with increasing depth in the target.
*Research sponsored by the U. S. Atomic Energy Commission under contract...
with the Union Carbide Corporation.
LEGAL NOTICE
.
.
This report was prepared as an account of Govorament sponsored work. Neither the United
statos, por the Commission, aor any person acung on behalf of the Commission:
A, Makos any warranty or representation, expressed or implied, with respect to the accu-
racy, completeness, or unofuldous of the Information contained in this report, or that the use
i of any information, apparatus, motbod, or procons diacload in this roport may not Infringe
printely omsed rights; or
B. Asnumor any liabiliuos with respect to the un of, or for damages rosulung from the
un of any information, apparatus, method, or procesu dlacioned in the report.
As used in the abovo, "person acting on beball of the Commission" includes any em-
ploym or contructor of the Commission, or employee of such contractor, to the extent that
vuod omployee or contractor of the Commission, or employee of such contractor prepares,
dlouminates, or provides accons to, lay information pursuant to klo employment or contract
with the Commission, or kilo employment with such contractor.
In this energy region, detectors for both neutrons and protons may be of
the nuclear reaction type, e.g., the production of "lc by the 12c(n,2n)lic
reaction. As the energy increases still further into the hundreds of Mev
and greater, associated fluences of other particles and quanta greatly
complicate the problems of dosimetry. For proton energies of about 300 Mev,
pion production thresholds are reached and the types of radiation increases.
In general, for protons up to 400 Mev the dose is about 6 to 10 x 10-8
rad/proton/cm2,'1' In this region, direct measurements of energy deposition,
e.g., using anthropomorphous phantoms and tissue-equivalent ionization
chambers, are the most widely used and widely accepted method of dosimetry,
especially when used in conjunction with particle fluence measurements or
other means from which the quality factor may be determined. As human
exposure to radiations of greater than 20 Mev are, in general, limited to.
the people working directly with accelerators, this paper will not discuss
at length the solution to the problems of dosimetry in this energy region.
For extensive and comprehensive reviews of the problems of high-energy
dosimetry, there are excellent papers on this subject by Baarli and
Sullivan, (2) and by Turner, et al. (1) In addition, a range of problems
was discussed in November 1965 at the Symposium on Accelerator Radiation
Dosimetry and Experience at the Brookhaven National Laboratory. Likewise,
since most of the personnel exposures to neutrons and protons in the
nuclear industry and in nuclear research are due to neutrons originating
in fission and in low-energy accelerators, and most radiobiological research
utilizes radiations in the energy region below 20 Mev, the bulk of this
paper will deal with neutrons in the energy region from thermai to about
20 Mev.
For neutron dosimetry in the energy interval below 20 Mev, it is
possible to use the Bragg-Gray relationship and assume secondary and primary
radiation equilibrium in order to expedite the measurement of absorbed dose.
Also in this energy region, certain nuclear reactions, e.g., the widely used
threshold detector system, may be used to measure the fluence as a function
of energy in broad energy increments such that the dose can be computed.
In this case, the typical relation for computing dose shown in Figure 1
actually expresses the energy deposited by the incident particles (kerma)
rather than the energy absorbed so that the results of a computation of this
type can be compared with an absorbed dose measurement, e.g., with a propor-
tional counter, only if secondary equilibrium can be assumed. In general,
the measurement of particle energy spectra is more desirable than an absorbed
dose measurement because the results may be used in the calculation of other
arwin
. .w
fundamental parameters, e.g., LET distributions as a function of depth in
.
the target material. The more detail in which such spectra can be determined,
. . .
the more accurately the dose distribution functions can be calculated. For
. .
-
this reason, at several laboratories the emphasis in physical dosimetry is
- - -
on the measurement of the energy spectra of the incident particles. However,
this parameter is still at least one step short of the ideal, i.e., the measure-
ment or calculation of the microscopic as well as the macroscopic distribution
of energy loss or absorption throughout the target of interest. Figure 2
shows three neutron spectra: (1) that for the HPRR, (2) that for the Y-12
accident, and (3) that for the Yugoslav accident. Figure 3 shows typical
energy fluence histograms measured with a threshold detector system based
largely on fission-foil detectors. Figure 4 shows the LET distributions
calculated for the center of a 30-cm-diam cylinder of tissue for these
spectra. The comparable LET distrib tions at 1.5 cm and 28.5 cm from the
front of the phantom are shown in Figures 5 and 6, respectively. The
marked differences in these energy and LET distributions emphasize the
importance of measurements other than dose alone.
For personnel exposures to massive, acute radiation fields such as in
the case of nuclear accidents, the definitive meaşurement of the field is
.
made more difficult because the detectors must be passive devices kept for
long periods of time without attention or maintenance. Therefore, the
threshold detector system has proved particularly suitable. However, the
use of threshold detectors based on fission foils has been hampered over
mon.
.
.
- -
-
the years by several factors: (1) the high initial cost of a good set of
detector foils and shield, (2) the large (gram) quantities of fissile .
material, particularly plutonium and neptunium, and (3) the expensive and
extensive electronic apparatus necessary for analysis. Other lesser
problems, nontheless real, include: (1) the difficulty or impossibility
of analysis if early recovery is not accomplished, (2) the difficulty in
calibrating the electronics system, and (3) the high level of dose required
to get an accurate measurement, i.e., several tens of rads, even for cases
where early recovery of the system is feasible. For this reason recent
developments in solid-state particle-track detectors have had an important
influence on this field. The basic TDU system is modified to the extent
that the gram quantities of fissile material are reduced to milligram
quantities plated on thin backing plates and placed immediately adjacent
to glass or plastic detectors. In this case the fission fragments from
the nuclear fission incident upon the glass or plastic detectors produce
a track which may be rendered visible under the optical microscope by a
simple etching technique. This system is more sensitive, i.e., doses of
the order of 1 rad of fission neutrons may be measured with an accuracy
of 10%, and early recovery is unnecessary. In addition the need for
extensive electronic apparatus is obviated. Figuze 7 shows a completed
detector of this type. Unlike the standard fission foils, the sulfur or
other high-energy detector which must be used with this system can be
analyzed with simple counters and electronic systems. Figure 8 shows a
glass plate which has been irradiated by fission fragments and etched in
a solution of sodium hydroxide to render the tracks visible.
Another important neutron detector for acute exposure conditions is
the sodium in the bloo. and hair on the body. Both have proved to be
valuable in the case of nuclear accidents, e.g., the Los Alamos Scientific
Laboratory accident of 1958 3) and the accident at the United Nuclear
Corporation's Wood River Junction plant in 1964.(4) However, there are
many parameters which must be evaluated in order to interpret blood sodium
and hair activation data in terms of dose. These include orientation of
the victim relative to the source, his size, the spectrum to which he is
exposed, etc. These factors have been studied by several experimenters
and at least one calculational study has been completed. Figure 9 shows
some of the results and indicates why additional studies are needed. One
of the most significant findings is that epithermal or "resonance" energy
neutrons cannot be neglected.
For the dosimetry of personnel expose: to low-level, chronic radiation
fields, such systems as the threshold detectors are not suitable in general,
eam
6
- we amerikan man transitional
and must be replaced by detectors having a longer integrating cycle. For
example if the spectrum does not change markedly with time, a measurement
may be made with laboratory-type instrumentation which may be used to call.
brate a thermal detector-moderator system which can be used to integrate
flux. for a long period of time, especially if the detector element has a
sufficiently long half-life, e.g., the 5.3-year half-life of 60co. A fissilie
foil, e.g., depleted uranium or thorium, could be used in conjunction with a
glass plate placed in a personnel meter which could then serve as a long-term
integrating detector. It must be emphasized that these single detector
systems are valid only when the spectrum is relatively invarient in the work
space and time.
Where exposures to neutrons in the dose range up to several times
tolerance values are likely but unpredictable, the nuclear emulsion is still
the most widely used detector. However, the lowest neutron energy for which
emulsions may be used, i.e., 2500 kev, is so high as to be a serious limita-
tion for heavily moderated fission spectra.
There are other systems which have been proposed for personnel monitor-
ing, including: (1) diode detectors for which carrier life times can be
correlated with neutron dose for certain applications, and (2) use of photo-
,
luminescent or thermoluminescent systems with two detectors, one of which.
-
-
.
-
has a greater sensitivity to neutrons than its mate. However, the uncertain-
-
ties encountered with these systems are great at present, and their limitations
should be explored fully before they are used for routine work. The diode
detector, which shows a change of carrier lifetime as a function of "dede,"
is a convenient and sensitive system but has a low-energy cutoff of about
250 kev, and the response is more nearly proportional to energy fluence
than to absorbed dose in tissue. The luminescence systems are handicapped
at present because the neutron response is represented, in general, by a
small difference between two large numbers.
Most available neutron dosime try systems have been evaluated at the
DOSAR Facility at Oak Ridge National Laboratory (ORNL) utilizing the Health
Physics Research Reactor (ho kR). Figure 10 shows a typical arrangement form
exposure. Figure 11 lists the relative response of several systems used former
nuclear accident dosimetry. Many of the observations given above were based
on data from this and similar experiments.
Summary
For convenience, the problems associated with the physical dosimetry
. of neutrons and protons can be divided roughly into two groups--those for
neutrons and protons of less than 20 Mev and those for neutrons and protons
of 20 Mer and greater. The problems of dosimetry above 20 Mev are, in
general, limited to those associated with accelerator facilities; thus,
the number of people involved is much less than for the facilities for which
neutrons and protons of less than 20 Mev are frequently encountered. Because
elastic processes are normally less important than inelastic processes at the
higher energies, it is important to recognize the difference in methodology
and techniques for attacking these problems. For absorbed dose measurements,
the assumption of equilibrium of incident and secondary particles cannot be
applied to the higher energy region, and, in general, measurements of flux
and at least a crude measurement of particle spectra must be made. Fortunately,
there are several convenient detectors by which flux may be measured, e.g., the
120(n, 2n) llc reaction and the 12C(p,np)11c reaction. Similar reactions are:
useful for the energy region below 20 Mev, e.g., the widely used threshold
detector systems, and in addition the energy region lends itself to absorbed
energy measurements based on the Bragg-Gray principle, i.e., assuming that
equilibrium of primary and secondary particles is possible. In general,
protons of less than 20 rev are not an important practical problem due to
their low penetrating power. Neutrons from fission sources and low-energy
accelerators are by far the most important sources of personnel exposures
in the region below 20 Mev, and they are also the energies most commonly
used in radiobiological research. For both energy regions, it is now
considered most important to determine sufficiently the parameters associated
with the radiation field such that the most fundamental interaction phenomena
may be computed. Particularly important are the linear energy transfer (LET)
distribution and the macroscopic distribution of dose throughout the organism,
including the various organs in the larger organisms such as man. The emphasis
of research at the present time is upon physical measurements at a sufficiently
definitive level that comparisons can be made with comprehensive calculations
using digital computers, so that once a computation technique is proven it can
be used to compute the various parameters in more detail than can be accomplished
practically with experimental methods.
24
10
References
1. J. E. Turner, et al., Health Phys. 10 (11), 783 (1964).
2. J. Baarli and A. H. Sullivan, Health Phys. 11 (5), 353 (1965).
3. D. F. Peterson; V. E. Mitchell, and W. H. Langham, Health Phys. ©,
1 (1961).
4. J. . Auxier, Nuclear Safety 6 (3), 298 (1965).
UNCLASSIFIED
ORNL-LR-DWG. 43850
TISSUE DOSE EQUATION FOR NEUTRONS OF ENERGY E
D(Mev/g) = EX Oilili
· where
0; = cross section of atoms of type i
1 - 2 mM(m + M)2
di = number of atoms of type i
figures
...
.
.
y
.
-
-
-
.
..
-
-
-
...
-
-
.
..
.
.
.
..
-
----
--
--
--TH
&
N(E)
ORNL-DWG. 66-3010

NEUTRON LEAKAGE SPECTRA
HPRR
Y-12
0.001 E
YUGOSLAV
0.0001
TTTT
0.00001
படடடடடடடடடப்பா
0.1
1
10
E (MeV) -
figure 2.
ORNL-DWG. 66-2673
HPRR
---- Y-12 ACCIDENT
--- YUGOSLAV ACCIDENT
PER CENT IN INTERVAL
0,75
18.0

1.50
2.5
NEUTRON ENERGY SPECTRA
..
.
figure 3

ORNL- DWG. 66-3009
PERCENT ABSORBED DOSE AS A FUNCTION OF LET
15 cm
PENETRATION DEPTH
HPRR
---- Y-12
--- YUGOSLAV
% ABSORBED DOSE
•
2
4
6
.
20
5
8 10
9
200
7
30 40 60 80 100
50 70 90
LET (kE WIN

figure 5.
ORNL-OWG. 66-3008
PERCENT ABSORBED DOSE AS A FUNCTION OF LET
1.5 cm
PENETRATION DEPTH
-- HPRR
----- Y-12
-.- YUGOSLAV
% ABSORBED DOSE
- 2
4
!
6,8 10
20
30 40 60 80 100
50 70 90
200
-
LET (KEVINE

-
LET ( ke ۷/۴)
figure 6
.DWG -اORN 66
-
3007
PERCENT ABSORBED DOSE AS A FUNCTION OF LET
---
--- ------
PENETRATION DEPTH 28.5 cm
HPRR
--
YUGOSLAV
% ABSORBED DOSE
- للللللللللل
4 |
20 .oاوا6 | 2
5
و 7
oo OO ا60 ا 40
30
200
50 70 90 .
-
LET
(
kev
ا ( م /

E6w.cru.........
Firm
OAK MOGE NATIONAL LABORATORY
O 0000

000



.
.
1
.
Ligure 8
figure 9.

ORNL-DWG. 65-6267
.
ERROR vs CYLINDER SIZE FOR MONTE CARLO
CODE COMPARED WITH HPRR EXPERIMENTS
0.90
NEUTRONS ESCAPING FROM
PHANTOM
.
0.80
0.70
NEUTRONS ESCAPING ABOVE
THERMAL
0.60
NUMBER TRACED FOR STATISTICAL
ACCURACY IN MONTE CARLO CODE
0.50
THERMALS PRODUCED
0.40
0.30
ERROR ( MONTE CARLO vs
HPRR EXPERIMENT )
0.20
.
0.10
ESCAPING THERMALS
.
1
اللللللللللللللللللاه
20
10 12
RADIUS (cm)
can

OC.
4
........
Il respiro
-
.
****.....mesine
,
OD
bat
VIVID
.
n
on
todo momenti
t
:
16
figure 10
.
...!!
.
1


figurelle
ORNL Dwg. 66-387
Intercomparison Data
Burst B158D, October 19, 1965
Fast
Thermal
neutrons neutrons
Date
(rads) (rads).
Burst B159D, October 20, 1965
Fast
Thermal
Gamma
neutrons neutrons radiation
(rads) (rads)
Gamma
radiation
Study
group
(r)
Date
A
10-22
403
.
60
10-22
37
23.5
22.3
B
10-22
65
10-22
41
22
Revised
417
384
362
389
0.13
Revised
36
69
89
0.09
30
18
37
363
0.22
66
25
10-22
Revised
35
10-22
Revised
0.14
51
22
10-22
-............
<0.1
0.26
61
10-22
23
25
316
364
19
Revised
0.17
10-22
55
10-22
23
404
352
364
352
Revised
24.3
23.9
23.9
25.5
0.18
Revised
0.11
0.13
......
--
-
-
-
.
.
F
10-22
353
10-22
Revised
13.3
20.7
-.-.-.•
10-22
360
0.21
52
10-22
26
30
10-22
370
10-22
28.8
XOS
SW-
1
END
DATE FILMED
6 / 14 /66







antakan
D
.
L:............
............,...
-!:
aris.
.....
.
.
...........
...---