.
.
.
,
.
4
e
.
i
1
:
.
-
i
i
TA
I OFI
ORNL P
1736

'
L:
.
..
.....
14
1
.-
.
-
-
---
+
-
1
.
.
:
4
엘
​NU 3 6
Min
.
40
...
-
.
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS - 1963
ORNr-p-1786
Corf-651026-2
The SNAP Sielding Program at ORNL*
NOV 1 8 866
ML,'
C. E. Ciifora ana ū. Levin
Oak Riage Vationai Laboratory
Oak Rage, Tennessee
The SNAP shielding program at Oak Ridge National Laboratory is only one
phase or a program of theoretical analysis, calculation, and experimentation
............ war....'
related to the radiation shielding problem. Obviously, the requirements of the

ncisi
program have become more stringent and exacting as nuclear energy has been ap-
ve
piiea to mobile systems. As part of the departe, but not forrotten AVP program,
simbo
wie Tower Shielding Facility waij constructed at Oak Ridge to make possible in-air
ürge-scale experiments in support of the shielding program. Despite the demise
l
RELEASED FOR ANNOUNCUEINT
IN NUCLEAR SCIENCE ABSTRACTS
ut the ANZ prograni, a great deal of valuable technology, including shieiding
i
w
cata, was generated.
The Tower Shielding Facility has since then been used in
caecking calculations for vehicle shielding and in establishing the scieiding
effectiveness of various structures, and will in the near future be used :) check
ne shielding of a SNAP-2/10A reactor. The aims of the program remain tíu...
iti in
N
ievelopment of advanced techniques for accurate calculation of the dose de-
I
.
W
livered through and aroună a shield to any point on a given dose plane, ano 9
7
47.
!
permit optimization of shield shape and weight.
.
.!
!
.
!
!
.
!
!.
!!
The Tower Shielding Facility at ORIVL is shown in slike i (Photo 53816).
Control room and .counting rooms are shielded by concrete and earth fill. The
towers are about 300 feet high and can elevate a load of 55 tons in the plane of
the two nearer towers to a height of about 125 feet. The reactor shown here and
in the next slide (Photo. 55847) is the second reactor to be usea here and, as the
"
"
LEGAL NOTICE
This report was prepired as an account of Government sponsored work. Neither the United
States, aor the Commission, nor any person scung on behalf of the Commission:
A. Makes any warranty or representation, expressed or implied, with respect to the accu-
racy, completeness, or usefulness of the information contained in this report, or that the use
of any information, apparatus, method, w process disclosed in this report may not laſringe
B. Assunos any liabilides with respect to the use o!, or for damages resulting from the
use of any information, apparatue, method, or process disclosed in this report.
As used in the above, “person acting on behalf of the Commission” Includes any em-
ployce or contractor of the Commission, or employee of such contractor, to the extent that
such employee or contractor of the Commission, or employee of such contractor prepares,
disseminates, or provides access to, any information pursuant to h.8 employment or contract
with the Commiosion, or hio employment with sucha contractor.
privately ownod rights; or
-
-
1
first reactor, has a water-mode:cated and cooled core. The third slide (Photo
50366-C) shows the flexibility of experimental arrangements achievable with this
*Research sponsored by the ü. 3. Atomic Energy Commission under contrüct with
the Union Carbide Corporation.
installation. The reactor ic now enclosed in a collimator 60 that a beam of
radiation may be directed over a range of directions. The nearer sphere is a
10-1't detector collimator that, when filled with lead and water, attenuates neu-
tron dose by a factor of 104 and gamma dose a factor of 108 thus permitting
discrimination of spacial distribution of incoming radiation. This facility then
we plan to use for some work with & SNAP reactor, which will be described later.
A natural question in connection with shielding research 16. "Why is it
necessary?" After all, shields have been designed and built for many years now
with considerable success. Back in 1957 a shielded reactor was flown in an air-..
plane and measured neutron dose rates were found to be within a factor of 10 of
calculated predictions in most cases. The simplest answer, of course, is that
factors of 10 are not good enough in space applications. However, in addition
there is the problem of uncertainty, for some calculations, especially when
departure is made from the simplest geometries, often differ widely from measure-
ments.
The dose rate at any given point within a shielded volume consists of
several components each of which is contributed by a distinct set of physical
processes.
These component parts of the total dose rate can vary sharply with
change of geometry, materials, and radiation source. As a matter of fact, know-
ledge of the physical processes themselves is constantly being supplemented; that
is, interaction cross sections are still under investigation.
In the SNAP systems that have been seriously proposed, the principal concern,
60 far as shielding is concerned, has been the protection of instruments from
radiation damage. The damage threshold in instrument payloads is invariably
reached earlier in the course of accumulation of neutron dose than of garama-ray
dose. For this reason, most attention has been focused on the neutron component
of radiation emanating from the reactor core..
..
motie jesü säiendiño materian Tür neutrons in space systems is jitni müü
yuride. Ii is rejatively idõiiü, aving a fabrication density oë 0.74 to 0.77
cī icon
empezaüure aria a macroscopic cross secüion ci about 0.05 per cmu waen the
Tiiciüesses considerea are not ove:: 2 1. A few years ago the vaiue of the vari-
Cür microscopic cross sections inau make up the average value oi macroscopic
مت
Cross seccions were very mica in coubt. Some cross sections aad just been de-
cassizieâ, aná to veriày calculations in wrica geometry piayeå & minimü role,
Como experiments on the transmission of neutrons through slads of lithiw gyáride
un conducted at the Lower Shielding Facility. The 12-ft aiameter sperical
Cummimaüor was used to obtain beans of neutrons 6 in. in diameter anā i5 ira.
da VS
Waüdroit
ململبد
Lucha sia
Wiis
OK
om diameter that were directed though siads of lithium nyáride or various thick-
C
VW
coses. Cancuations were cone üring the Monte Carlo tecinique as incorporateä
ine 052? code developed at Jais liage. The results are shown in the next sidde
while
was were interpretea as coa
ühe relative accuracy or cross sections anā
I
-
we adequacy or the nuüber ci nisuories seiected for calculation.
La
UU
Iniö wüs one on the mirst tests oz the Monte Cario methoá cođe mowa as
Om at väis Ridge Tavionai Taboratury. Monte Cario is the most accurate methoá
presentny available for cancuiation of radiation transmission through materiais
inanü it tãe only eüäca ürac is aot limitec by geometry of the source-sciela
cuvination. ñowever, wäere are still problems in using this metod in that gooà
cccuracy or cose prediction requires a suificient number of histories to be rü
vaca üsefüs biasing techniques Encong Wita a evaluation of the staüisticãi error
merenu in the calculation (aü : east 5,000). Also consistent errors in prograxi-
was wre noi äitzicuit to make arà very difficult to discover, once made. The
prolet on transmission ünzouga tihe slads, for example, required many hours of
machine time on a CW-2604 computer to evaluate diasing techniques aná to ceauce
the statistical deviatica cor 25,000 neutron historie...
+
Our6
hu
There are, of course, other less meticulous methods of calculation. The
next slide lists scme of them. The dose predictions for the SNAPSHOT experi-
ments -- the orbited SNAP-IOA reactor of Atomice International -- were made by the
point-kemel method as developed by Goldstein at United Nuclear Corp. However,
the dose at the payload plane was not dominated by the neutrons that penetrated
the shield but by the neutrons scettered around the shield by structural com-
ponents of the system. Even 50, the predictions obtained were quite good for a
small unmanned system, varying from 10% below the measured dose to 300% above. In
the past, calculations of dose rates for the ANP program using kernel methods
yielded results that were from 50% too high to a factor of 2.7 too low.
The scattering of neutrons by structure was Illustrated at ORNL in a series
of experiments utilizing the TSR-II reactor. The next.slide shows the scheme of
the experiments (ORNL 2-01-056-35--1300). The truncated cone below the TSR-II
core is a water tank, representing a SNAP system shield. A beryllium block,
simulating a SNAP reactor control drum is mounted adjacent to the reuctor core
as shown in the next slide (Photo 58047). In addition to the beryllium, a small
cylinder of nickel was mounted above. The beryllium could be swunes away from the
reactor vessel 60 as to come out farther from the shield shadow cone, as shown in.
the next slide (ORNL 2-01-056-35-1381).
The detector collimatca traversed horizontally from the center line of the
shield to about 8 in. to the right of the edge of the shield. In other words,
it moved from the shadow of the shield out to where it looked direccly at the
beryllium. 'In the next slide are some of the results (ORNL 2-01-056-35-1383).
Positions 0, 1, 2, 3, are positions of the beryllium progressively farther out
from the reactor vesse), Ihe rise in dose rate clearly indicated that the direc-
tions of neutrons are changed in collision with beryllium atoms and increase the ...
aose received even well within the shadow of the shield. The next slide (ORNL
2-01-056-35-1384) shows the same results but with tighter collimation, 3-in.
collimator as compared to an 8-in. collimator in the previous results. The total
dose rate per watt of reactor power is down as would be expected, but the change
in dose rate with beryllium position is greater. The next slide shows (ORML
2-01-056-35-1385) the increments of dose for each position of the beryllium blocks.
The background has been subtracted so that these dose rates represent increments
ciue to scattering only. The effect of the very small, l-in. diameter by l-in.
hign nickel cylinder is clearly visible above background. These experiments gave
an indication of the total effect to be expected from structural scattering and in
the SNAPSHOT flight the scattering effect was the dominant dose contributor over
most of the dose plane.
over the next couple of years the program at ORNL 16 oriented toward develop-
ment of advanced mathematical techniques for calculating accurately neutron
transport through shields and for calculating by iterative processes a minimum
weight design for a given mission profile. The program includes improving the
05R code, including amplification of the cross-section library and coupling of this
ccke with the Sn code to achieve shorter machine time for a lerger number of
neutron histories. The program a].30 envisages development of a two-dimensional
Sn-Pn code and coupling of 05R and OGRE codes and Sn-CGRE: for rapid calculation of
secondary gamma-ray generation anä the resultant dose contribution.
First problems to be undertaken by these mathematical programs will be
related to a SNAP 2/10A system ani to provide experimental verification of the
calculations. A SNAP reactor will be installed at the Tower Shielding Facility.
The experimental program will be divided into three major parts. First will be
znapping of the leakage flux from the core as shown in the next slide (ORNL Dwg.
65-7228).
-6-
The reactor will be suspended on a lead screw at the end of a boom fixed to
A rotatable column. In the position shown here, with the reactor at not over one
watt power, a collimator for a neutron spectrometer will rotate about the core
while maintaining its focus on one point in the reactor. This procedure with
.
yield a source spectrum for use in subsequent calculations..
The second part of the progran is represented by the next slide (ORNL Dwg.
65-7229). Here the dose and spectrum transmitted through a lithium hydride
shield will be mapped. The detector collimator will be placed in a dry chamber
that will be shielded from air-scattered and structure-scattered radiation by
concrete cover slabs. The shield itself will be surrounded by a polyethylene and
iron collar to exclude air-scattered neutrons. The design of this collar and
the validity of this experimental configuration is at present the subject of both
theoretical and experimental study. Calculations indicate that the collar will
de effective in preventing distortion of shield performance in space by the earth-
bound environment. However, an experiment utilizing the TSR-II will be conducted
shortly to verify this.
The configuration is shown in the next slide.
The third phase of the program will be concerned with structural scattering
contributions as shown in the next slide (ORNL Dwg. 65-7231). A view of the overall
Facility for the reactor installation is shown in the next slide (ORNL Dwg.
65-7616). It is hoped that the installati.on will be flexible enough to allow
some extension of the program into thicker shields and eventually to allow veri-
Tication of shield shaping calculstions that may in the future contribute to weight
Gaving in manned as well as unmann.ed space systeús. ' .
12/ 7 / 651
DATE FILMED
END
.
.....
..
...
.
...
.
ram.....
...
........
....
.
.