..
$
.
.
.
V
-IOFI
- ORNLP
2599

+
*
-
:
;
.
.
e
.
.
1
rin
I
C
.
4
.
i
69
'T
.
A
:
*
EEEFEFEE
-
.
top
| 1.25 1.1.4 1.6
MICROCOPY RESOLUTION TEST CHART
NATIONAL BUREAU OF STANDARDS -1963
ORNO D 25.99
MASTAR
1.09. .50 2tP
1.C..
NOV 29 :ják
-
CONF-661019.5
RATIO OF CAPTURE TO FISSION IN 238 Pu AT keV NEUTRON ENERGIES*
A. Lottin, ** L. W. Weston, G. de Saussure,
and J. H. Todd

RELEASED FOR ANNOUNCEMENT
Oak Ridge National Laboratory
Oak Ridge, Tennessee
IN NUCLEAR SCIENCE ABSTRACTS
ABSTRACT
The neutron capture to fission ratio, a, was measured for 239Pu for neu-
tron incident energies from 20 to 600 keV. A pulsed beam of neutrons was
collimated on a sample of plutonium placed in the center of a large hydrog-
eneous gamma-ray scintillator poisoned with gadolinim. A capture event in
the sample was characterized by a single pulse of the scintillator due to the
cascade of capture gamma rays, while a fission event was characterized by a
pulse due to the prompt fission gamma rays followed, a few microseconds later,
by auditional pulses due to the gamma rays produced when the thermalized fission
S
neutrons were captured in the gadolinium of the scintillator. Below 100 keV
the neutron energies were measured by the time-of-flight technique with a
resolution of 7 nsec/m. Above 100 keV approximately monoenergetic neutrons
were used. Similar measurements were also performed on 2351) and the results
are included for completeness. The uncertainties in the values of a obtained
are approximately 10%. The data are consistent with measurements performed
by other laboratories.
*Research sponsored by the U. S. Atomic Energy Commission under contract
with Union Carbide Corporation.
-
WARE
* :
**On assignment from Centre d'Etudes Nucléaire, Saclay.
!
..
"
-2-
I.
INTRODUCTION
The neutron capture to fission ratio, a, 18 of perticular importance in
the design of breeder reactors, and reactor designers have requested this
parameter for 239Pu in the energy range from 10 keV to 1 MeV to an accuracy
of 5%.2 Hopkins and Divens have measured ü for 239 Pu with approximately mono-
energetic neutrons at nine neutron energies between 30 keV and 1 MeV, including
two measurements below 150 keV, at 30 + 20 keV and at 65 + 15 kev. The values
of alpha obtained at these two energies differ by more than a factor of two.
!
The technique used in the present measurements was very similar to that used
by Hopkins and Diven, the main difference being that below 100 keV the incident ::
neutron energies were obtained by the time-of-flight technique, making it pos-
sible to determine a for many different energies, and with a relatively good
energy resolution, in the range from 20 keV to 100 keV.
II. EXPERIMENTAL TECHNIQUE
:
Since the experimental technique is described in detail elsewhere, 4 only
the principle of the measurement will be given here. A pulsed beam of neutrons
was collimated into a sample of the isotope under investigation (238Pu or 235U)
placed in the center of a 1.25-m-diam hydrogeneous gamma-ray scintillator
poisoned with gadolinim. The tank containing the liquid scintillator is
described in ref. 5. This tank was filled with approximately 250 gallons of
NE-224, a commercial liquid scintillator,* in which gadolinium 2-ethylhexoate
had been dissolved at a concentration of 16 g/liter. This scintillator was
viewed by eight 58 AVP photomultipliers. A capture event in the sample was
characterized by a single pulse of the scintillator due to the cascade of
capture gamma rays, while a fission event was characterized by a pulse due to
,
*NE-224 produced by Tuclear Enterprises, Ltd., Winnipeg, Canada.
the prompt fissioz gamma rays followed, a few microseconds later, by addi-
tional pulses due to the gamma rays produced when the thermalized fission
neutrons were captured in the gadolinium of the scintillator.
Two microseconds after a "prompt" pulse was detected, a delayed coinci-
dence circuit was activated to inspect, during a time interval of 32 us, for
additional pulses due to the capture of fission neutrons in the gadoliniu.
The probability, €, of detecting a pulse in a 32-us interval just following &
fission was approximately 88%, whereas the probability, P, of detecting a
pulse during a "random" 32-us interval was approximately 7%. This technique for
identifying fission and capture events was also used in the measurements
2.
reported in refs. 3, 4, and 6, where more detailed discussions may be found.
The electronics used was similar to that used in ref. 4 except that a
"fast-slow" coincidence arrangement was used to gather pulse-height and time
information. The pulse-height discriminator of the slow system (cross-over pick-
off) determined if an event was to be analyzed. The fast pulses from the eight
photomultipliers were summed in a passive network and triggered a tunnel diode
discriminator with a bias of approximately 1/4 that of the slow system. The
output of the tunnel diode discriminator was used for time information if it
was in coincidence with the slow discriminator output.
WELLNES
S
"The time distributions and pulse-height distributions of the pulses from
the scintillator were stored simultaneously in two modified 400-channel analyzers.
The time analyzer was divided into two sections to store separately the dis-
tributions of pulses followed by delayed coincidences and not followed by
delayed coincidences. Figure 1 shows an example of time-of-flight spectra.
The pulse-height analyzer was divided in four sections to store separately, the
LEGAL NOTICE
.
. .
.
.
.
.
This report was prepared as an account of Government sponsored work. Nelther the United
States, acr the Commission, nor any person acting on beball of the Commission:
A. Makes any varranty or representation, expressed or implied, with raspoct to the accu-
racy, completenens, or usefulness of the information contained in this report, or that the we
of any information, apparatus, method, or process disclosed in this report may ant indringe
privately owned righto; or
B. Assumes any liabilities with respect to the use of, or for damages rosulung from the
use of any information, appuratus, molhod, or process Usclosed in this report,
As used in the above, "person acting on behalf of the Commission" Includes any om-
ployee or contractor of the Commission, or onployee of such contrtotor, to the extent that
such employee or contractor of the Commission, or employee of such contractor preparos,
dissominates, or provides ACCOsa to, any Information pursuant to his employment or contract
with the Commission, or his employmont with such contractor.
GO YULE
chi.
...
.
.
..
h
.:
+5
bara
.
.
th
.
35,000
ORNL-DWG 65-9688
EVENTS NOT FOLLOWED
BY DELAYED NEUTRONS
L (CAPTURE AND _.
BACKGROUND)
EVENTS FOLLOWED BY
DELAYED NEUTRONS
_(FISSION).
25,000
SAMPLE
EVENTS
COUNTS
"
..
.
TARGET
GAMMA RAYS
To
.
L
:
10,0
oo
BACKGROUND SAMPLE
EVENTS, EVENTS
PULSE-HEIGHT PULSE-HEIGHT
ANALYZED ANALYZED
5,000
com assessm
e
ns
100
200
30C
CHANNEL NUMBER (~ 2.3. ns/CHANNEL)
.
400

"
HE
*
*
*
.
-5-
evente 'time-gated" over two time-of-flight intervals, followed and not
followed by delayed coincidences. The position of these two time intervals
is shown in Fig. 1. The prompt interval corresponds to the time during which
neutrons are arriving at the sample, whereas the background interval is of
equal duration but at a later time, when the neutron burst has passed the
sample. Figure 2 shows an example of pulse-beight distributions after background
subtraction of the coincidence and anticoincidence events in 238Pu.
The ORNL 3-M pulsed Van de Graaff was used to accelerate protons onto
a Lithium target; neutrons were produced via the Phi(p,n) reaction. The proton
burstó were 2-ns wide FWAM and repeated every 32 us. Neutrons of the required
energies were obtained by changing the high voltage of the Van de Graaff, and
hence the energy of the proton impinging upon the lithium target. Near the
threshold for the "Li(p,n) reaction (E. = 1.88 MeV) the 30-kev neutrons produced
were kinematically collimated in the forward direction by the motion of the
center-of-mass of the reaction.
The experimental arrangement for measurements
at 30-kev neutron energies is sketched in Fig. 3. At higher proton energies
the neutrons were emitted more isotropically ari a neutron collinator was
necessary. The arrangement used in this case is sketched in Fig. 4.
The time-of-flight distributions of the various events were obtained with
a time-to-pulse-height converter. Figure 1 shows the tine-of-flight spectra
í'or 30-keV incident neutrons on 239Pu. The right-hand side of the figure
corresponds to events followed by delayed coincidences (fissioni events and
accidental delayed coincidences) whereas the left-hand side corresponds to
events not followed by delayed coincidences (capture events, background and
fissions not followed by delayed coinaidences). The narrow peak to the right
.......
.
......
...
..
..
.
.
...
.
.
ORNL-DWG 66-9687
8000

7000
O EVENTS FOLLOWED BY
DELAYED COINCIDENCES
(FISSION)
• EVENTS NOT FOLLOWED
BY DELAYED COINCIDENCES
(CAPTURE PLUS ~ 8%
OF FISSION)
6000
5000
COUNTS
3000
Olutaanotoo cool
poooo000000
CHANNEL NUMBER
UM
ORNL-LR-DWG 49319R2

.
.
0
1
::.
yar TV
ww
w
L
+
6299
www
...
..
.
.
w
HEX
SCINTILLATOR
- LEAD SHIELD
- TARGET
--PHOTOMULTIPLIER (8)
M
- SAMPLE HOLDER
Aramelo
1.1..
4-in.-THICK
LEAD SHIELD ..
.
FL"
.
.
..
.
.
Lid
.
ORNL-LR-DWG 49320

.
4-in-THICK LEAD SHIELD
- SAMPLE
< PHOTOMULTIPLIER (8)
ntrenamiento
n met een maniera convenientemente con cordones e
.
de
restaurants
111
111111111
levereiro series me
wanneer
november
O
THIUM LOADED
PARAFFIN
j
L
ESE
Cum
.
ii:
B
1:':1 LE
1
HT
2
E
9
.
.
VM
-
-
-
CONTO
::1
1
D
D
III
UUTI
til
.
UNI
II
IIIII
IT
TA
UUNI CUTII
I
III
1
III
IE
11
.
.
2
2
NO
doo
.
.
VN
0
.
.
0
*
SCINTILLATOR
I IIMIIIIII
*
////////////IIIIIIIIIIIIIIIIII
TARGET
of the time spectra is causeż by the gama rays produced when the proton burst
arrives on the lithium target. A lead shield was placed around the target,
as showa on Fig. 3, to reduce the intensity of this peak. The time-of-flight
was measured relative to the target gamma-ray peak.
The width of this peak is
a measure of the time resolution of the system. The height of the target gamma-
ray peak in the delayed coincidence spectrum is a measure of the accidental
delayed coincidence probability. The width of the "sample events" peak in
Fig. 1 is determined by the energy spectrum of the neutron beam, hence by the
H target thickness, angle of collimation, etc. The time resolution of the
system during operation was 7 ns. The measurements at 1-n flight path yielded
time-of-flight and pulse-height spectra similar to those in Figs. 1 and 2 but
with somewhat poorer signal to background.
The samples used were 100 g of 285U and 20 and 40 g of 239Pu. The 2350
sample consisted of ten 2-in.-diam discs that were each sandwiched between
0.005-in.-thick Al discs. The Ai discs were spaced in a 5-in.-OD aluminum can
so that the spacing between the 235U plates was 0.10 in. The Al can had 0.020-
in.-thick end vindows. The 20-8 289 Pu sample was one disc 3 7. in diameter
and similarly canred. The 40-g 238Pu samle consisted of four 2-in, discs,
also similarly canned. The 40-g 238Pu sample was used for measurements at im
and the 20-8 sample was used for the short flight path measurement.
.
The isotopic composition of the 235ų sample was 97.5% 235U, 1.64 284U,
0.06% 2360, and 0.73% 2384. The 239Pu. samples contuined 0.07% 240pu and
0.01% 241Pu. The 288Pu was reduced to metal, rolled, and canned by Los Alamos
Scientific Laboratory.
4
+
*
.
:
T
ri
.
9
7."
-20-
III. ANALYSIS OF THE DATA
It was shown in ref. 4 that a can be obtained from the ratio, y, of the
sample events not followed by delayed coincidence to the sample events followed
by delayed coincidence by the relation:
y[1 - C. (1-P)] - C.11-P)
a
=
(1)
1 - P - y
Where C. = 1 - € is the probability that no delayed coincidence would be detec-
ted following a fission, P is the random delayed coincidence probability
defined previously, and €ple, is the ratio of the efficiency of the scintillator
for detecting a fission event to the efficiency for detecting a capture event.
In ref. 4 it was also show:n how the values of C, and eple, can be obtained from
by delayed coincidences. The sample events from which y is obtained must be
corrected for two types of backgrounds; a neutron-beam-independent background
background of the detector under operating conditions (cosmic rays, detector
activation), and a neutron-dependent background due to the capture gamma rays
emitted when neutrons were absorbed in the 25 mils of aluminum in which the
samples were canned and when neutrons scattered by the sample were eventually
captured in the solution or in some structural material.
The neutron-independent background was measured by placing a quartz window
in front of the Van de Graaff lithium target, thereby preventing the proton
beam from reaching that target and producing neutrons. The neutron-associated
background was obtained by replacing the plutonium sample with a lead sample
(also canned in aluminum) whose thickness was computed to "mock up" the
11-
scattering of the plutonium.
Pulse-height spectra and time-of-flight spectra
of the two kinds of l'ackgrounds were measured. For the measurements with the
two uranium isotopes, an additional small background correction was necessary
to correct for 238U and 294U isotopic impurities in the sample. These cor-
rections were computed from reported cross sections of those isotopes. 8
For neutrori energies above 100 keV and at 30 keV (near threshold of the
Li(p,n) reaction) the values of y were obtained as the ratios of the areas
corresponding to sample events (Fig. 2). The neutron energies were computed
from the measured proton energies, and the energy resolution was determined
by the thickness of the lithium target.
In the energy range from 20 to 100 keV, the ratio y was computed channel
by channel from time spectra similar to that shown in Fig. 1.
The neutron
energy corresponding to each channel was obtained by the time-of-flight over
im with a time resolution of 7 ns/m.
The major uncertainties in the values of a were a 6 to 10% w certainty
associated with the uncertainty in the relative efficiencies of the scintillator
to detect fission and capture events (€p/€, in Eq. 1) and an imcertainty in
the value of a of 0.007 associated with the uncertainty in the probability e
to detect a pulse in a 32-s time interval after a fission (c. in Eq. 1).
The exact values of these uncertainties were functions of the neutron energy
and off the isotope considered.
IV. RESULTS AND COMPARISON WITH OTHER MEASUREMENTS
Table I lists the values of a obtained for 23872 and 285U for 40 neutron
energies. The values for 238Pu are compared in Fig. 5 with values obtaineit by
12.
Table I. a of 238Pu and 2350
-
· E
(kev)
239 Pu
:
335U
239 Pu
236U
(kev)
i
17.7
38.5
40.5
42.3
18.3
18.8
19.4
20.2 + 0.6
21.0
21.7
44.5
46.7
48.5 :
51.0 + 2.5
54.5
57.5
22.4
0.253 + 0.017 0.340 + 0.016
0.226 + 0.016 0.360 + 0.016
0.246 + 0.016 0.365 + 0.016
0.244 +0.017 0.365 + 0.015
0.286 + 0.017 0.361 + 0.015
:0.199 + 0.027 0.335 + 0.015
0.198 + 0.028 0.339 + 0.016
0.195 + 0.030 0.322 1 0.013
0.178 + 0.032 0.329 + 0.012
0.176 + 0.025 0.300 + 0.011
0.174 + 0.022 0.355 + 0.012
0.169 + 0.021 0.315 + 0.012
0.165 + 0.020 · 0.329 + 0.013
0.160 + 0.021 0.350 + 0.016
.0.172. + 0.034 0.349 + 0.022
23.1
23.9
60.7
0.395 + 0.108 0.325 + 0.0:49
0.490 + 0.109 0.274 + 0.044
0.443 + 0.097 0.337 + 0.043
0.442 + 0.089 0.342 + 0.041
0.350 + 0.075 0.445 + 0.043
0.353 + 0.071 0.406 + 0.038
0.406 + 0.071 0.340 + 0.031
0.409 + 0.048 0.360 + 0.031
0.371 + 0.040 0.391 + 0.031
0.353 + 0.036 · 0.373 + 0.027
0.350 + 0.034 0.382 +0.025
0.355 + 0.030 0.368 + 0.024
0.327 + 0.027 0.372 + 0.022
0.289 + 0.025 0.349 + 0.020
0.281 + 0.023 0.372 + 0.020
0.329 + 0.033* 0.384 + 0.033*
0.297 + 0.020. 0.370 + 0.018
08303 + 0.019 : 0.367 + 0.018
0.288 + 0.019 .0.376 + 0.018
0.299 + 0.019 0.374 + 0.017
0.228 + 0.017 0.374 + 0.017
24.8
25.7
64.0
68.0
72.0
26.8
77.0
82.0 + 5.0
27.9.
29.0
30.1 + 1.2
31.0
32.3
33.8
200.0 + 7.0
300.0 +6.0
400.0 + 6.0
500.0 16.0
600.0 + 5.0
0.127 + 0.008 0.254 + 0.010
0.116 + 0.011 0.215 + 0.010
0.078 + 0.011 0.164 + 0.008
0.065 + 0.005 0.154 + 0.005
0.032 + 0.002 0.129 + 0.004
35.3
37.0
-
...
*The total error on a is given at 30.1 keV; values at other energies were normalized to
this value, and only the statistical error is given for the other energies.
n
.
.
.
.
t
ZsDuo T
740%
20Orde.
u
,
:
.
-.
'
*'
*
$
R
ORNL-DWG 66-9475
• PRESENT EXPERIMENT
A KANNE et al. (1955)
• HOPKINS AND DIVEN (1962)
A SPIVAK et al. (1957)
a OF 239 Pu..
1
-
101
-
TOIT
100
1000

NEUTRON ENERGY (keV)
-14-
F. "
PT ":11 TATA: 1!-...-AS
*
++
S
A
NEW M
AY
2WPFYTTETEA
TA ..
. TETE
.
..
.....
.
.
-
.
..
--.
-.-
.-
-.
....
--
-
A
.
ne
Hopkins and Diven by a very similar technique, and with values derived from
measurements of n by Spivak et al. The figure also shows the result of a mea-
surement by Kanne et al.20 made by a foil irradiation in a reactor spectrum.
· The agreement between the various data is considered satisfactory, especially
considering the different resolutions with which the measurements were per-
V
.
!
formed.
........
Figure 6 is a similar comparison between values of a for 235U. Again
the agreement between various sets of data is fair. Unfortunately, the most
accurate data are all taken by very similar techniques so that a possibility
remains that some systematic error may be associated with all these measurements.
In addition, the uncertainties are generally larger than the 5% requested by
reactor designers.
.
. .
. .
.
ORNL-DWG 66-9474
• PRESENT EXPERIMENT
A KANNE et al. (1955)
• ORNL-RPI (1966)
O VAN SHI-DI et al. (1965)
O WESTON et al. (1964)
o HOPKINS AND DIVEN (1962)
A SPIVAK et al. (1957)
Q OF 235J..
TY
100
UN
UT
14 4 III
H
.
-100
04
TOT
10
100
NEUTRON ENERGY (keV)
1000


....
.
.
'L
'
A
.
.
.
int
...
A V
'
A
N
--
..
-16-
REFERENCES
1. H. A. Bethe, Proc. Intern. Conf. Peaceful Uses At. Energy, Geneva, 1955
4, 321.
2. Computation of EANDC Requests, EANDC 55U (1966).
UN VI
3. J. C. Hopkins and B. C. Diven, Nucl. Sci. Eng. 12, 169 (1962).
4. L. W. Weston, G. de Saussure, and R. Gwin, Nucl. Sci. Eng. 20, 80 (1964). ".
5. J. H. Gibbons, R. L. Macklin, P. D. Miller, and J. I. Neiler, Phys. Rev. 122,
182 (196.1).
Van Shi-Di et al., "Interaction of Neutrons with U-235 Nuclei at an Energy
0.0002-30 kev," P-2024, Preprint, Joint Institute for Nuclear Research,
Dubna, 1965.
A
.
7. J. B. Marion and J. L. Fowler, Fast Neutron Physics, Part 1, pg. 133,
Interscience Publishers, New York, 1960.
.
&
Neutron Cross Sections, Vol. III, Z = 88 to 98, BNL-325, 2nd ed., Supp. 2. ...
9. P. E. Spival, B. G. Erozolimsky, G. A. Dorofeev, V. N. Levrenchik, I. E.
Kutikov, and Y. P. Dobrynin, At. Energ. I, No. 3, 21 (1956). Translation:
J. Nucl. Energy 4, Part II, 79 (1957).
10. W. R. Kanne, H. B. Stewart, and F. A. White, Proc. · Inter. Conf. Peaceful
Uses At. Energy, Geneva, 1955, 4, p. 315; see also s. Oleksa, J. Nucl. Energy
5, 16 (1957).
'..
.
.!
1
-17-
FIGURE CAPTIONS
Fig. 1. Time-of-Flight Spectrum for ~ 30-kev Neutrons of 238Pu.
Fig. 2. Pulse-Height Spectra for 238Pu.
Fig. 3. Experimental Arrangement for ~ 30-ket Neutron Energy Measurement.
Fig. 4. Experimental Arrangement for Time-of-Flight Measurements of a.
Fig. 5. Comparison of Various Values of a of 238Pu in the Range from
10 to 1000 keV.
Fig. 6. Comparison of Various Values of a of 2850 in the Range from
1 to 1000 keV.
.
i
'
-'
*
:
..
"
.
i
END



DATE FILMED
12/ 21 / 66




.
2.
99
IL
3
-
2
.
.
I
!
2
.1
...