operated by los alamos national security, llc for the u.s. … · 2017. 2. 28. · pu series (0.126...
Post on 05-Sep-2020
0 Views
Preview:
TRANSCRIPT
Operated by Los Alamos National Security, LLC for the U.S. Department of Energy's NNSA
Los Alamos National Laboratory
Subcritical Copper-Reflected α-phase Plutonium (SCRαP) Measurements and
Simulations
NCSP TPR March 2017
Operated by Los Alamos National Security, LLC for the U.S. Department of Energy's NNSA
Los Alamos National Laboratory *Institut de Radioprotection et de Sûreté Nucléaire (IRSN)
J. Hutchinson, R. Bahran, T. Cutler, W. Monange*, J. Arthur, M. Smith-Nelson, E. Dumonteil*
Los Alamos National Laboratory
March 2017 | 3
Overview • Introduction • Experiment Design • Experiment Overview • Preliminary Results • Future work
Los Alamos National Laboratory
March 2017 | 4
Introduction
Los Alamos National Laboratory
March 2017 | 5
Design/Conduct/Analyze Subcritical Validation Experiments
o Nuclear Data and Transport Codes • Fill integral experiment database deficiencies + • Find differential nuclear data library deficiencies For different….
o Energy Ranges (Thermal, Intermediate, Fast) o Multiplication Ranges (Low, Medium, High) o Materials (Fissile, Moderator, Reflector) o Neutron Reactions
o Uncertainty Quantification
Los Alamos National Laboratory
March 2017 | 6
Recent Advances in Subcritical Experiments • We have come a long way since the first subcritical
measurements at CP-1 in 1942. • Many organizations (LANL, LLNL, SNL, IAEA, IRSN, CEA,
universities, and others) have pursued subcritical experiments and/or simulations in recent years.
• The BeRP ball reflected by nickel benchmark evaluation was published in the 2014 edition of the ICSBEP handbook.
• This benchmark was the first: o Published benchmark evaluation of measurements performed at DAF.
o Benchmark evaluation using new MCNP capabilities for subcritical systems (the MCNP list-mode patch and MCNP6 list-mode capabilities).
o Benchmark using the Feynman Variance-to-Mean method.
o LANL-led subcritical experiment in the ICSBEP handbook.
• This benchmark was the culmination of several years of subcritical experiment research.
• BeRP-tungsten published in 2016 edition of ICSBEP handbook.
Los Alamos National Laboratory
March 2017 | 7
Experiment Design (CED-1 and CED-2)
Los Alamos National Laboratory
March 2017 | 8
Subcritical Copper-Reflected α-phase Plutonium (SCRαP) Integral Experiment
• SCRαP Preliminary Design (w/ MCNP®6) o BeRP (Beryllium-Reflected Plutonium).
• 4.5-kg WG α-phase stainless-steel clad plutonium sphere.
o High-purity nested copper shells • C101 Cu alloy (99.99 wt.% Cu).
Los Alamos National Laboratory
March 2017 | 9
Subcritical Copper-Reflected α-phase Plutonium (SCRαP) Integral Experiment
• SCRαP Preliminary Design (w/ MCNP®6) o High-density interleaved polyethylene
shells • Wide range of achievable subcritical
multiplication values will help: • Identify deficiencies and quantify
uncertainties in nuclear data • Validate computational methods related to
neutron multiplication inference.
There are two purposes for the configurations with polyethylene: • They allow for higher multiplication factor than with copper alone • They allow for a different neutron spectra (and resulting sensitivity)
for the same multiplication factor.
Los Alamos National Laboratory
March 2017 | 10
Subcritical Copper-Reflected α-phase Plutonium (SCRαP) Integral Experiment
• MC-15 (Multiplicity Counter 15) was used to estimate three benchmark parameters: o Detector singles count rate (R1) i.e. the
count rate in the detector system o Doubles count rate (R2) i.e. the rate in
the detector system in which two neutrons from the same fission chain are detected
o Leakage multiplication (ML) i.e. the number of neutrons escaping a system per starter neutron.
Photograph and MCNP® model of the MC-15 detector system. 15 He-3 tubes inside polyethylene.
Records list-mode data (a time list of every recorded neutron event to a resolution of 128 nsec).
Los Alamos National Laboratory
March 2017 | 11
Subcritical Copper-Reflected α-phase Plutonium (SCRαP) Integral Experiment
Photograph and MCNP® model of the MC-15 detector system. 15 He-3 tubes inside polyethylene.
Records list-mode data (a time list of every recorded neutron event to a resolution of 128 nsec).
For the SCRαP experiment, two MC-15 systems were present and collected data in the same time list.
Los Alamos National Laboratory
March 2017 | 12
Sensitivity Results with MCNP®6
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.00.82
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.89
1.83
1.8
1.77
1.751.73
1.721.7
1.69 1.68 1.68 Pure Cu-Reflected BeRP Ball
kef
f
Subcritical Copper Shell Thickness [in]
avg. neutron energy causing fission [MeV]
• Cu-only base configurations shown.
• Decided to only measure configurations up to 4.0” thickness of total Cu: o Cu total neutron cross
section sensitivity (and average neutron energy causing fission) level out at the 5”-thick mark.
o Beyond 4.0” there are additional issues (weight, cost, criticality safety).
Los Alamos National Laboratory
March 2017 | 13
Sensitivity Results with MCNP®6
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.00.82
0.84
0.86
0.88
0.90
0.92
0.94
0.96
0.98
1.89
1.83
1.8
1.77
1.75
1.731.72
1.71.69
1.68 1.68 Pure Cu-Reflected BeRP Ball
kef
f
Spherical Copper Shell Thickness [in]
avg. neutron energy causing fission [MeV]
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
k eff a
bsol
ute
sens
itivi
ty
63Cu (n,total)
65Cu (n,total)
• Cu-only base configurations shown.
• Decided to only measure configurations up to 4.0” thickness of total Cu: o Cu total neutron cross
section sensitivity (and average neutron energy causing fission) level out at the 5”-thick mark.
o Beyond 4.0” there are additional issues (weight, cost, criticality safety).
Los Alamos National Laboratory
March 2017 | 14
Interleaving Poly Provides Additional Configurations
0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.891.83
1.8 1.771.75
1.731.72
1.7 1.691.68
239Pu (n,total) - Pure Cu-Reflected BeRP Ball avg. neutron energy causing fission [MeV] 63,65Cu (n,total) - Pure Cu-Reflected BeRP Ball
kef
f abs
olut
e se
nsiti
vity
keff
Los Alamos National Laboratory
March 2017 | 15
Interleaving Poly Provides Additional Configurations
0.82 0.84 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.000.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.891.83
1.8 1.771.75
1.731.72
1.7 1.691.68
1.741.66
1.55 1.53 1.52
239Pu (n,total) - Pure Cu-Reflected BeRP Ball 239Pu (n,total) - Pure Cu and Poly-Reflected BeRP Ball avg. neutron energy causing fission [MeV] 63,65Cu (n,total) - Pure Cu-Reflected BeRP Ball 63,65Cu (n,total) - Pure Cu and Poly-Reflected BeRP Ball
kef
f abs
olut
e se
nsiti
vity
keff
14% epithermal4.4% thermal
3.6% epithermal0% thermal
Los Alamos National Laboratory
March 2017 | 16
Sensitivity Results with MCNP®6 • These results were compared to critical
configurations in the ICSBEP handbook. o Note that there are a limited number of
copper-reflected critical experiments (8 series with U fuel and 2 series with Pu fuel).
• Total Cu-63 (all energies): o For SCRaP experiment, the maximum
sensitivity (0.143) was greater than the two Pu series (0.126 max) but less than for some of the U series (0.200 max).
• Total Cu-63 (intermediate energy regime): o For SCRaP experiment, the maximum
sensitivity for the 16 configurations (0.018) is greater than the two Pu experimental series by nearly an order of magnitude (0.002), but similarly less than that for some of the U experimental series (0.051).
0.85 0.90 0.95 1.00-0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14 Thermal Intermediate Fast
k eff a
bsol
ute
sens
itivi
ty
keff
Cu+Poly configurations
Cu only configurations
Cu-63 (n,total)
Configurations with HDPE have increased sensitivity in the intermediate energy range.
Los Alamos National Laboratory
March 2017 | 17
0.85 0.90 0.95 1.00-0.010-0.0050.0000.0050.0100.0150.0200.0250.0300.0350.0400.0450.0500.0550.0600.0650.070
Thermal Intermediate Fast
k eff a
bsol
ute
sens
itivi
ty
keff
Sensitivity Results with MCNP®6 • These results were compared to critical
configurations in the ICSBEP handbook. o Note that there are a limited number of
copper-reflected critical experiments (8 series with U fuel and 2 series with Pu fuel).
• Cu-65 sensitivity results yielded same conclusions as Cu-63.
Cu-65 (n,total)
Configurations with HDPE have increased sensitivity in the intermediate energy range.
Los Alamos National Laboratory
March 2017 | 18
Uncertainty estimates • Experimental uncertainties for 4 experimental
parameters were calculated. • Used criticality eigenvalue calculations for these
estimates as described in a previous work [J. HUTCHINSON, T. CUTLER “Use of Criticality Eigenvalue Simulations for Subcritical Benchmark Evaluations” Transactions of the ANS Winter Meeting, Las Vegas NV (2016)].
Parameter Experimental Uncertainty
Uncertainty
ML
Pu radius ± 2 mils 0.18 Pu isotopics ± 0.5% 0.19
Cu thickness ± 0.3 cm 0.03 Cu mass ± 0.5% 0.00006
R1
Pu radius ± 2 mils 1024 Pu isotopics ± 0.5% 1045
Cu thickness ± 0.3 cm 141 Cu mass ± 0.5% 0.34
R2
Pu radius ± 2 mils 37450 Pu isotopics ± 0.5% 41336
Cu thickness ± 0.3 cm 5252 Cu mass ± 0.5% 13.1
Estimate of experimental uncertainties for Configuration 15 (0.5 inch-thick HDPE surrounded by 3.5 inch-thick copper).
Cu mass was expected to be a minor uncertainty, which the table confirms.
Many lessons-learned from the previous Ni and W benchmarks were used to minimize experimental uncertainties.
Los Alamos National Laboratory
March 2017 | 19
Experiment Overview (CED-3B)
Los Alamos National Laboratory
March 2017 | 20
Experiment configurations • 17 total configurations: o 1 Bare o 8 Cu-only configurations o 7 Cu+HDPE configurations o 1 HPDE-only configuration
• In order to determine the detector efficiency, Cf-252 source replacement measurements were performed. o The source strength of the Cf-252
source at the time of the measurements was 7.59e5 fissions/sec +/- 1.0%.
Los Alamos National Laboratory
March 2017 | 21
Experiment configurations
• 17 total configurations: o 1 Bare o 8 Cu-only configurations o 8 Cu+HDPE configurations
• In order to determine the detector efficiency, Cf-252 source replacement measurements were performed. o The source strength of the Cf-252
source at the time of the measurements was 7.59e5 fissions/sec +/- 1.0%.
Thickness (inches) Simulated keff
HDPE Cu HDPE+Cu LANL IRSN
0.0 0.0 0.0 0.774 0.777
0.0 0.5 0.5 0.837 0.829
0.0 1.0 1.0 0.871 0.862
0.0 1.5 1.5 0.894 0.884
0.0 2.0 2.0 0.911 0.900
2.0 1.0 3.0 0.917 0.907
0.0 2.5 2.5 0.924 0.914
2.0 2.0 4.0 0.929 0.921
2.0 2.0 4.0 0.935 0.919
0.0 3.0 3.0 0.935 0.923
0.0 3.5 3.5 0.944 0.933
0.0 4.0 4.0 0.951 0.939
1.5 2.0 3.5 0.951 0.939
1.0 2.5 3.5 0.957 0.943
0.5 3.0 3.5 0.958 0.942
0.5 3.5 4.0 0.965 0.948
4.0 0.0 4.0 - -
Los Alamos National Laboratory
March 2017 | 22
Experiment configurations
Configurations 0-7
Los Alamos National Laboratory
March 2017 | 23
Experiment configurations
Configurations 8-16
Los Alamos National Laboratory
March 2017 | 24
Experiment configurations
Configurations 0-7
Los Alamos National Laboratory
March 2017 | 25
Experiment configurations
Configurations 8-16
Los Alamos National Laboratory
March 2017 | 26
Preliminary Results (CED-3B + CED-4A)
Los Alamos National Laboratory
March 2017 | 27
Analysis method
• Neutron noise analysis o Rossi-alpha o Time interval analysis o Feynman variance to mean
• Hansen Dowdy • Hage-Cifarelli
o Others… • Analysis method used here
is documented in detail in the BeRP/Ni and BeRP/W ICSBEP evaluations.
Data are separated into gates (of time-width τ)
y-axis is the number of gates that contained exactly n events (Cn)
x-axis is the number of neutrons recorded in the gate (n)
∑= nCTime τ
τ
Los Alamos National Laboratory
March 2017 | 28
Feynman histogram results
• Deviation from Poisson (solid lines) increases as system multiplication increases.
• Mean of histogram is proportional to the detector count rate.
• Width of histogram is proportional to the doubles count rate.
Configurations 0-5 (1024 micro-sec gate-width)
0 10 20 30 40 50 60 70 80 90 100102
103
104
105
106
107
Cn
n
bare
Los Alamos National Laboratory
March 2017 | 29
Feynman histogram results
• Deviation from Poisson (solid lines) increases as system multiplication increases.
• Mean of histogram is proportional to the detector count rate.
• Width of histogram is proportional to the doubles count rate.
0 25 50 75 100 125 150 175 200 225 250102
103
104
105
Cn
n
Configurations 6-11 (1024 micro-sec gate-width)
Los Alamos National Laboratory
March 2017 | 30
Feynman histogram results
• Deviation from Poisson (solid lines) increases as system multiplication increases.
• Mean of histogram is proportional to the detector count rate.
• Width of histogram is proportional to the doubles count rate.
0 25 50 75 100 125 150 175 200 225 250102
103
104
105
106
Cn
n
Configurations 12-16 (1024 micro-sec gate-width)
Los Alamos National Laboratory
March 2017 | 31
Singles count rate (R1)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 161000015000200002500030000350004000045000500005500060000650007000075000800008500090000
R1 (c
ts/s
ec)
Configuration
SCRaP Singles Rate (R1)( )( )
!
)1()1(0
r
prnnnm n
n
r
∑∞
=
+−⋅⋅⋅−=
ττ
( ) ( )
( )∑∞
=
=
0nn
nn
C
Cp
τ
ττ
( ) ( )τ
ττ 1
1mR =
Reduced factorial moment:
normalized fraction of gates that recorded n events:
Singles count rate:
Gate-width (τ)
Los Alamos National Laboratory
March 2017 | 32
Singles count rate (R1) Detector efficiency (from Cf-252 measurements):
This is the count rate from the Cf-252 measurements.
FS is the reported spontaneous fission emission rate of the Cf-252 source. is the average number of neutrons emitted per Cf-252 fission.
( ))1(
1
SSFR
ντ
ε =
)1(Sν
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1x104
2x104
3x104
4x104
5x104
6x104
7x104
8x104
9x104
SCRaP Singles Rate (R1)
R1 (c
ts/s
ec)
Configuration
0.000
0.005
0.010
0.015
0.020
0.025
0.030
SCRaP efficiency (ε)
effic
ienc
y
Plotted together to show that the reason that the count rate goes down significantly for the configurations with HDPE is due to the decrease in efficiency (which is caused by neutron absorption primarily in the hydrogen).
Los Alamos National Laboratory
March 2017 | 33
Excess variance The excess variance (deviation of a Feynman histogram from a Poisson distribution) is proportional to Y2, given by:
( )[ ]
τ
τττ
212
2
)(21)( mm
Y−
=
0 250 500 750 1000 1250 1500 1750 2000
0.0
2.0x105
4.0x105
6.0x105
Y2
Time Width (µsec)
The amount of excess variance increases with Cu thickness (due to an increase in the system multiplication).
Configurations with Cu only
Los Alamos National Laboratory
March 2017 | 34
Excess variance The excess variance (deviation of a Feynman histogram from a Poisson distribution) is proportional to Y2, given by:
( )[ ]
τ
τττ
212
2
)(21)( mm
Y−
=
The amount of excess variance increases with the system multiplication.
Configurations with Cu+HDPE
0 250 500 750 1000 1250 1500 1750 2000
0.0
2.0x105
4.0x105
6.0x105
Y2
Time Width (µsec)
HDPE can increase or decrease Y2 (due to a competition between multiplication and detector efficiency (due to absorption in H).
Los Alamos National Laboratory
March 2017 | 35
Neutron lifetime/slowing-down time A fit can be performed on the Y2 curves to calculate the neutron lifetime/slowing-down time (1/λ):
λττλω
λτ−−−=
e11),(2
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16-50
0
50
100
150
200 SCRaP Lifetime (1/λ)
Neu
tron
lifet
ime/
slow
ing-
dow
n tim
e (m
icro
-sec
)
Configuration
0.000
0.005
0.010
0.015
0.020
0.025
0.030
SCRaP efficiency (ε)
effic
ienc
y
The MC-15 detector system has a slowing-down time of around 35 micro-seconds. For the configurations with Cu only, the result is approximately 35 micro-seconds as expected, but it is significantly larger for the configurations that include HDPE hemishells.
This can also be determined via Rossi-alpha analysis.
Los Alamos National Laboratory
March 2017 | 36
Doubles count rate (R2) Doubles count rate:
),()()(
2
22 τλω
ττ
YR =
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
1x104
2x104
3x104
4x104
5x104
6x104
7x104
8x104
9x104
SCRaP Singles Rate (R1)
R1 (c
ts/s
ec)
Configuration
103
104
105
106
SCRaP Doubles Rate (R2)
R2 (c
ts/s
ec)
Los Alamos National Laboratory
March 2017 | 37
Leakage multiplication (ML)
( )( )
31224
1
)1(23
)1(
)2()1()2(2
)1(
)2()1(1
1
42
4
1
1
2
CCCC
RRC
C
C
withC
CCM
S
I
ISS
I
IS
L
−=
−=
−−=
−=
+−=
ετντ
νννν
ννν
)1(Sν
)1(Sν 1st factorial moment of Pu-240 Pν 2nd factorial moment of Pu-240 Pν 1st factorial moment of Pu-239 Pν 2nd factorial moment of Pu-239 Pν
)2(Sν
)1(Iν
)2(Iν
Assumes that there are no emissions from (α,n) neutrons.
Leakage multiplication is related to the multiplication factor (keff).
Los Alamos National Laboratory
March 2017 | 38
Leakage multiplication (ML)
( )( )
31224
1
)1(23
)1(
)2()1()2(2
)1(
)2()1(1
1
42
4
1
1
2
CCCC
RRC
C
C
withC
CCM
S
I
ISS
I
IS
L
−=
−=
−−=
−=
+−=
ετντ
νννν
ννν
)1(Sν
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 160
2
4
6
8
10
12
14
16
ML
Configuration
SCRaP Leakage Multiplication (ML)
Los Alamos National Laboratory
March 2017 | 39
Leakage multiplication (ML) )1(Sν
As previously discussed, preliminary simulations provided multiplication factor (keff) results. These results were used to approximate leakage multiplication.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 150
2
4
6
8
10
12
14
16
18
20 LANL simplified model IRSN detailed model Measured
Leak
age
Mul
tiplic
atio
n (M
L)
Configuration
The LANL MCNP models were simplified models (perfect spherical reflectors with no materials present outside the reflectors) but the IRSN MORET models had additional details (MC-15 detectors, detailed reflector hemishells, etc.).
Los Alamos National Laboratory
March 2017 | 40
Leakage multiplication (ML) )1(Sν
As previously discussed, preliminary simulations provided multiplication factor (keff) results. These results were used to approximate leakage multiplication.
The LANL MCNP models were simplified models (perfect spherical reflectors with no materials present outside the reflectors) but the IRSN MORET models had additional details (MC-15 detectors, detailed reflector hemishells, etc.).
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16-4
-2
0
2
Leak
age
Mul
tiplic
atio
n (C
-E)/E
(%)
Configuration
(C-E)/E (%) for IRSN MORET detailed models
Los Alamos National Laboratory
March 2017 | 41
Future work (CED-4A + CED-4B)
Los Alamos National Laboratory
March 2017 | 42
What’s next? • This experiment will be evaluated and
documented in an upcoming version of the ICSBEP handbook.
• Results will hopefully be used to improve cross-section libraries.
• Data set will also be used to validate subcritical analysis methods.
Los Alamos National Laboratory
March 2017 | 43
Thank you for your attention.
This work was supported by the DOE Nuclear Criticality Safety Program, funded and managed by the National Nuclear Security Administration for the Department of Energy.
top related