Access to this full-text is provided by EDP Sciences.
Content available from The European Physical Journal Conferences
This content is subject to copyright. Terms and conditions apply.
06008
Radiation Characterization Summary for the
WSMR Fast Burst Reactor Environment at the 6-
Inch Location
Danielle Redhouse1,*, Edward S. Lum2, and Johnathon Koglin3
with Edward J. Parma1, Curtis D. Peters1, Mikhail Finko3, Jesse M. Roebuck1, David W.
Vehar1, Frank Sage, Andrew M. Tonigan1, Ryan Mulcahy1, Thomas A. Ball1, Elliott
Pelfrey1, Melissa Moreno1, Karissa Currie1, and Patrick J. Griffin1
1Sandia National Laboratories, Albuquerque, NM, USA
2Los Alamos National Laboratory, Los Alamos, NM, USA
3Lawrence Livermore National Laboratory, Livermore, CA, USA
Abstract. The characterization of the neutron, prompt gamma-ray, and
delayed gamma-ray radiation fields for the White Sands Missile Range
(WSMR) Fast Burst Reactor, also known as molybdenum-alloy Godiva
(Molly-G) has been assessed at the 6-inch irradiation location. The neutron
energy spectra, uncertainties, and common radiation metrics are presented.
Code-dependent recommended constants are given to facilitate the
conversion of various dosimetry readings into radiation metrics desired by
experimenters. The Molly-G core was designed and configured similarly to
Godiva II, as an unreflected, unmoderated, cylindrical annulus of uranium-
molybdenum-alloy fuel with molybdenum loading of 10%. At the 6-inch
position, the axial fluence maximum is about 2.4×1013 n/cm2 per MJ of
reactor energy; about 0.1% of the neutron fluence is below 1 keV and 96%
is above 100 keV. The 1-MeV Damage-Equivalent Silicon (DES) fluence
is estimated at 2.2×1013 n/cm2 per MJ of reactor energy. The prompt
gamma-ray dose is roughly 2.5E+03 rad(Si) per MJ and the delayed
gamma-ray dose is about 1.3E+03 rad(Si) per MJ.
* Corresponding author: drredho@sandia.gov
Radiation characterization summary for the
WSMR fast burst reactor environment at the
6-inch location
© The Authors, published by EDP Sciences. This is an open access article distributed under the terms of the Creative Commons
Attribution License 4.0 (https://creativecommons.org/licenses/by/4.0/).
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
1 Introduction
In the late 1950’s, Los Alamos National Laboratory (LANL) designed and developed
the Moly-Godiva reactor during the Rover program period at the Pajarito Site. The core was
designed as an unreflected, unmoderated, cylindrical annulus of uranium-molybdenum
alloy. The assembly was moved to the U.S. Army’s Survivability, Vulnerability, and
Testing Directorate (SVAD) at White Sands Missile Range (WSMR) in June of 1964 [7].
The Moly-Godiva reactor was renamed the WSMR Fast Burst Reactor (FBR), nicknamed
Molly-G. Molly-G was designed and configured similarly to Godiva-II but had a
molybdenum loading of 10%. Molly-G was developed to conduct neutronic studies by
providing an intense neutron burst over a narrow pulse width, with high dose capabilities.
Initial criticality of the Molly-G at WSMR was achieved on July 7th, 1964 [18].
Characterization of the neutron and gamma-ray environments in WSMR FBR include
the 6-inch (152.4 mm) and 24-inch (609.6 mm) leakage free-field environment external to
the FBR core radially spaced on y-plate at the axial fluence maximum and includes both
accurate neutronic modeling and experimental efforts [1]. The neutronics model for the
WSMR FBR was redundantly modeled in MCNP by both LANL and Sandia National
Laboratories (SNL) and modeled by Lawrence Livermore National Laboratory (LLNL) in
their proprietary neutronic code MERCURY. Due to the proprietary nature of MERCURY,
these results will not be enclosed in this report. Experimental observations, including
passive and active dosimetry, are important to determine the accuracy of the model.
Presented are the energy dependent neutron fluence, some of the conversion constants for
radiation metrics, the axial and radial neutron fluence profiles, and the time-dependent
responses for different pulse sizes.
2 Environment Description
WSMR FBR was designed by Dr. Thomas F. Whimett and Dr. Gordon E. Hansen to
perform neutron studies in a burst environment [7]. WSMR FBR can operate in a steady-
state mode with the operating power level limited to 8kW. In pulse mode, a maximum pulse
size of 2.0 MJ (∆T of 250°C) with a full-width half-maximum (FWHM) of 50 s can be
attained [16].
The WSMR FBR is fueled by six uranium-molybdenum solid metal ring elements. The
fuel uses uranium enriched to 93.0 weight percent U-235 and 10.0 weight percent
molybdenum. The fuel elements are clad in 6061-aluminum at 0.5 cm thick. Each fuel rings
dimensions are an internal diameter of 2 inches (5.08 cm) and an external diameter 4 inches
(10.18 cm), with an individual height of 1.28 inches (3.25 cm). WSMR FBR is controlled
by two fuel control rods, one fuel burst rod, and a fueled safety block. The control rods are
positioned axially in 5 fuel elements, while the burst rod is positioned axially in 4 fuel
elements in the normal core configuration [7][16][17].
Historically, the WSMR FBR has been used for a wide variety of experiment campaigns
including those for space applications and radiation effects sciences. The main irradiation
location for the WSMR FBR is within the reactor room on an elevated table. The table
features a y-plate with standard reproduceable test positions in 1/5-inch increments along
three separate ‘legs’. Figure 1 shows the WSMR FBR along with its irradiation table.
Figure 2 shows the y-plate with three legs that extend out to the edges of the test table;
these include the ‘A-Leg’, ‘B-Leg’, and ‘C-Leg’. By convention, ‘A-Leg’ or ‘Long Leg’
extends from the core west, ‘B-Leg’ extends N-NE (~60 angle), and ‘C-Leg’ extends S-SE
(~60 angle). Experiments are classically placed at the axial fluence maximum outside of
the core, roughly 3 inches from the bottom of the core. The WSMR FBR is situated in an
open-air cell approximately 48 ft feet (30.48 cm) by 48 feet (30.48 cm), with a 23 ft height
to the ceiling. WSMR FBR features a new state-of-the-art cooling system allowing
continued safe pulsing operations. Cooling is done in both steady-state mode and after
pulses using chilled/dried compressed air.
Figure 1. WSMR “Molly-G” Fast Burst Reactor and test table.
2
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
1 Introduction
In the late 1950’s, Los Alamos National Laboratory (LANL) designed and developed
the Moly-Godiva reactor during the Rover program period at the Pajarito Site. The core was
designed as an unreflected, unmoderated, cylindrical annulus of uranium-molybdenum
alloy. The assembly was moved to the U.S. Army’s Survivability, Vulnerability, and
Testing Directorate (SVAD) at White Sands Missile Range (WSMR) in June of 1964 [7].
The Moly-Godiva reactor was renamed the WSMR Fast Burst Reactor (FBR), nicknamed
Molly-G. Molly-G was designed and configured similarly to Godiva-II but had a
molybdenum loading of 10%. Molly-G was developed to conduct neutronic studies by
providing an intense neutron burst over a narrow pulse width, with high dose capabilities.
Initial criticality of the Molly-G at WSMR was achieved on July 7th, 1964 [18].
Characterization of the neutron and gamma-ray environments in WSMR FBR include
the 6-inch (152.4 mm) and 24-inch (609.6 mm) leakage free-field environment external to
the FBR core radially spaced on y-plate at the axial fluence maximum and includes both
accurate neutronic modeling and experimental efforts [1]. The neutronics model for the
WSMR FBR was redundantly modeled in MCNP by both LANL and Sandia National
Laboratories (SNL) and modeled by Lawrence Livermore National Laboratory (LLNL) in
their proprietary neutronic code MERCURY. Due to the proprietary nature of MERCURY,
these results will not be enclosed in this report. Experimental observations, including
passive and active dosimetry, are important to determine the accuracy of the model.
Presented are the energy dependent neutron fluence, some of the conversion constants for
radiation metrics, the axial and radial neutron fluence profiles, and the time-dependent
responses for different pulse sizes.
2 Environment Description
WSMR FBR was designed by Dr. Thomas F. Whimett and Dr. Gordon E. Hansen to
perform neutron studies in a burst environment [7]. WSMR FBR can operate in a steady-
state mode with the operating power level limited to 8kW. In pulse mode, a maximum pulse
size of 2.0 MJ (∆T of 250°C) with a full-width half-maximum (FWHM) of 50 s can be
attained [16].
The WSMR FBR is fueled by six uranium-molybdenum solid metal ring elements. The
fuel uses uranium enriched to 93.0 weight percent U-235 and 10.0 weight percent
molybdenum. The fuel elements are clad in 6061-aluminum at 0.5 cm thick. Each fuel rings
dimensions are an internal diameter of 2 inches (5.08 cm) and an external diameter 4 inches
(10.18 cm), with an individual height of 1.28 inches (3.25 cm). WSMR FBR is controlled
by two fuel control rods, one fuel burst rod, and a fueled safety block. The control rods are
positioned axially in 5 fuel elements, while the burst rod is positioned axially in 4 fuel
elements in the normal core configuration [7][16][17].
Historically, the WSMR FBR has been used for a wide variety of experiment campaigns
including those for space applications and radiation effects sciences. The main irradiation
location for the WSMR FBR is within the reactor room on an elevated table. The table
features a y-plate with standard reproduceable test positions in 1/5-inch increments along
three separate ‘legs’. Figure 1 shows the WSMR FBR along with its irradiation table.
Figure 2 shows the y-plate with three legs that extend out to the edges of the test table;
these include the ‘A-Leg’, ‘B-Leg’, and ‘C-Leg’. By convention, ‘A-Leg’ or ‘Long Leg’
extends from the core west, ‘B-Leg’ extends N-NE (~60 angle), and ‘C-Leg’ extends S-SE
(~60 angle). Experiments are classically placed at the axial fluence maximum outside of
the core, roughly 3 inches from the bottom of the core. The WSMR FBR is situated in an
open-air cell approximately 48 ft feet (30.48 cm) by 48 feet (30.48 cm), with a 23 ft height
to the ceiling. WSMR FBR features a new state-of-the-art cooling system allowing
continued safe pulsing operations. Cooling is done in both steady-state mode and after
pulses using chilled/dried compressed air.
Figure 1. WSMR “Molly-G” Fast Burst Reactor and test table.
3
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 2. WSMR test table with the y-plate installed. Curved dosimetry plates are shown at the 6-inch
position on the ‘A-Leg’, B-Leg, and ‘C-Leg’ of the y-plate.
3 Neutronic Models
The MCNP model of the WSMR FBR is shown in XY and XZ plane views in Figure 3.
The MCNP model includes the uranium-molybdenum fuel elements, safety block, control
rods, and burst rod. The control rods and burst rods can be vertically adjusted in the model
to any position desired. The model is typically run at standard temperature and pressure,
using room-temperature cross sections (293.6 K) from ENDF/B-VIII.0. Various cross
section temperatures can be modeled if desired [11].
Neutron energy spectra and fluence per fissions were calculated using a 6-cm diameter
tally sphere located in-core and ex-core on the long leg (i.e., A Leg) level with the fluence
maximum centerline, and in mirror positions on the opposite side of the reactor from the
long leg. Calculations were performed using the MCNP k-code mode for the neutron
fluence and metrics. For the SNL MCNP neutronic model of the WSMR FBR, the system
keff was noted as 1.00019 0.00001. This value coincides with a system configuration of
the burst rod at full insertion, with control rod 1 inserted at 2838 mils and control rod 2
inserted at 2800 mils.
Figure 3. SNL MCNP Model views showing the above and side profile of the WSMR Fast Burst
Reactor.
The neutron energy spectrum was calculated for 640 -group and 89-group energy
structures using the MCNP model presented in Section 2 for the 6-cm diameter tally sphere
[4]. MCNP6.2 version was used with the ENDF/B-VIII.0 cross sections. Room temperature
cross sections were used for the calculations [3][11][15]. The model was run in the k-code
mode using both neutrons and photons. This arrangement allowed for both the neutron
energy spectra and the prompt gamma-ray energy spectrum to be generated in a single run.
The gamma-ray spectrum is not presented here. The results generated in the tally sphere
were in units of neutron fluence per source neutron. These results were then converted to
fluence per fission. The neutron fluence results are then converted from fluence per fission
to fluence per MJ of reactor power using 192.4 MeV per fission. This value represents the
fission fragment, neutron, prompt gamma-ray, capture gamma-ray, and delayed gamma-ray
energy deposition in the reactor core per fission event. These energy deposition values were
calculated using MCNP. The 89-group neutron energy spectrum result was used as the trial
function for LSL-M3, a least-squares unfolding code [9][12][14].
Figure 5 and Figure 6 show the MCNP-generated 640-group (black) and 89-group
(grey) neutron energy fluence at the 6-inch position on a linear and logarithmic y-axis,
respectively. The units on the y-axis are in lethargy fluence or energy fluence, equal to
Edϕ/dE (MeV/MeV-cm2-MJ). With the energy fluence represented linearly on the y-axis
and the neutron energy on the x-axis represented logarithmically, the area under the curve
represents the total neutron fluence. This representation allows for the best visual depiction
of the fluence over the complete neutron energy range.
The energy fluence is found to peak near 1 MeV. The thermal peak occurs at ~0.07 eV
(7.0E-8 MeV). The results for the 640-group and 89-group neutron energy fluence are in
very close agreement, as expected. The 640-group energy fluence calculation shows higher
fidelity in the spectrum, which is more notably observed in the energy range of 1E-4 to 1
MeV. This structure is considered to be real and not an artifact of the code or cross section
set [10][13]. It is caused by resonances in the cross sections, especially from the oxygen
elastic scattering cross section. The calculated standard deviation for each energy bin in the
640-group energy fluence in the energy range from 0.005 eV (5.0E-9 MeV) to 6 MeV is
4
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 2. WSMR test table with the y-plate installed. Curved dosimetry plates are shown at the 6-inch
position on the ‘A-Leg’, B-Leg, and ‘C-Leg’ of the y-plate.
3 Neutronic Models
The MCNP model of the WSMR FBR is shown in XY and XZ plane views in Figure 3.
The MCNP model includes the uranium-molybdenum fuel elements, safety block, control
rods, and burst rod. The control rods and burst rods can be vertically adjusted in the model
to any position desired. The model is typically run at standard temperature and pressure,
using room-temperature cross sections (293.6 K) from ENDF/B-VIII.0. Various cross
section temperatures can be modeled if desired [11].
Neutron energy spectra and fluence per fissions were calculated using a 6-cm diameter
tally sphere located in-core and ex-core on the long leg (i.e., A Leg) level with the fluence
maximum centerline, and in mirror positions on the opposite side of the reactor from the
long leg. Calculations were performed using the MCNP k-code mode for the neutron
fluence and metrics. For the SNL MCNP neutronic model of the WSMR FBR, the system
keff was noted as 1.00019 0.00001. This value coincides with a system configuration of
the burst rod at full insertion, with control rod 1 inserted at 2838 mils and control rod 2
inserted at 2800 mils.
Figure 3. SNL MCNP Model views showing the above and side profile of the WSMR Fast Burst
Reactor.
The neutron energy spectrum was calculated for 640 -group and 89-group energy
structures using the MCNP model presented in Section 2 for the 6-cm diameter tally sphere
[4]. MCNP6.2 version was used with the ENDF/B-VIII.0 cross sections. Room temperature
cross sections were used for the calculations [3][11][15]. The model was run in the k-code
mode using both neutrons and photons. This arrangement allowed for both the neutron
energy spectra and the prompt gamma-ray energy spectrum to be generated in a single run.
The gamma-ray spectrum is not presented here. The results generated in the tally sphere
were in units of neutron fluence per source neutron. These results were then converted to
fluence per fission. The neutron fluence results are then converted from fluence per fission
to fluence per MJ of reactor power using 192.4 MeV per fission. This value represents the
fission fragment, neutron, prompt gamma-ray, capture gamma-ray, and delayed gamma-ray
energy deposition in the reactor core per fission event. These energy deposition values were
calculated using MCNP. The 89-group neutron energy spectrum result was used as the trial
function for LSL-M3, a least-squares unfolding code [9][12][14].
Figure 5 and Figure 6 show the MCNP-generated 640-group (black) and 89-group
(grey) neutron energy fluence at the 6-inch position on a linear and logarithmic y-axis,
respectively. The units on the y-axis are in lethargy fluence or energy fluence, equal to
Edϕ/dE (MeV/MeV-cm2-MJ). With the energy fluence represented linearly on the y-axis
and the neutron energy on the x-axis represented logarithmically, the area under the curve
represents the total neutron fluence. This representation allows for the best visual depiction
of the fluence over the complete neutron energy range.
The energy fluence is found to peak near 1 MeV. The thermal peak occurs at ~0.07 eV
(7.0E-8 MeV). The results for the 640-group and 89-group neutron energy fluence are in
very close agreement, as expected. The 640-group energy fluence calculation shows higher
fidelity in the spectrum, which is more notably observed in the energy range of 1E-4 to 1
MeV. This structure is considered to be real and not an artifact of the code or cross section
set [10][13]. It is caused by resonances in the cross sections, especially from the oxygen
elastic scattering cross section. The calculated standard deviation for each energy bin in the
640-group energy fluence in the energy range from 0.005 eV (5.0E-9 MeV) to 6 MeV is
5
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
less than 1%. A similar but less resolved structure is also seen in the 89-group energy
fluence. The 89-group energy structure is the energy grouping used in the NuGET code [5].
4 Experiments
The results of the MCNP and MERCURY calculations of the WSMR environment
represent a good initial representation for the neutron spectrum characterization. However,
a considerable uncertainty in the results can exist due to model representation uncertainty
(geometry, density, and composition) and uncertainty in the transport cross sections. An
improved spectrum can be obtained by combining this initial trial spectrum with measured
integral values that correspond to the reactions rates from high-fidelity passive dosimetry
reactions. For this work, the LSL-M3 code was used to generate an 89-group neutron
energy spectrum that is used in the NuGET code [5].
The selection of passive dosimetry foils and activation reactions has been studied and
evaluated over many years. A summary of the work can be found in noted references of this
manuscript [1][6][10][11][13]. No complete and perfect set of activation reactions exists
that allows the neutron fluence energy spectrum to be calculated by dosimetry alone.
However, there are enough reactions to cover the relevant energy range with high -fidelity
dosimetry cross sections to allow for adjusted neutron fluence results to be generated with a
quantified accuracy. The passive dosimetry foils and the associated neutron activation
reactions used to perform the neutron fluence characterization for the WSMR 6-inch (152.4
mm) and 24-inch (609.6 mm) environments are shown in Table 4 and Table 5.
The foils and activation reactions chosen for the analysis represent expert judgment and
references to previous work [1][10][13]. A complete set of dosimetry foils and reaction data
will vary for a given neutron environment being characterized [2]. Typically, neutron
activation resulting in the emission of protons (n,p), neutrons (n,2n), (n,n’), or alpha
particles (n,) represent high-neutron-energy reactions of 1 MeV or greater. Neutron
activation resulting in prompt gamma-ray emission from radiative capture (n,) or fission
reactions determine the shape of the thermal and epithermal region of the neutron spectrum.
Covering foils with cadmium (Cd) and/or boron (B) can allow for resonances above the
associated cutoff energies to become more prominent, allowing for additional integral
quantities to be included in the analysis. Typically, only thermal activation or fission foils
are covered in Cd or B.
A total of 23 different foil types, resulting in 35 different reactions, were irradiated [2].
Three fission foils (U-235, U-238, and Np-237) were irradiated individually in a bare or
cadmium cup configuration in a single 24 MJ steady-state operation. An historic Pu-239
foil activity was used from an older dosimetry set characterized and run with a nickel foil
for tracking purposes. This historic activity was nickel corrected to match the 24 MJ steady-
state operation conducted in 2022 [2].
The free-field environment was used for all the foil irradiations noted previously. A
high variation in the flux is known to exist radially and axially within the reactor cell.
However, a test region was identified where the neutron fluence varies 5%, this region is
the highlighted (yellow) section on the curved plates shown in Figure 4. Additional
uncertainty in the analysis was included to account for possible variation due to geometry
effects.
Figure 4. Typical Dosimetry Foils – Characterization Test Setup.
5. Spectrum Characterization Results
The neutron energy spectrum is first calculated using MCNP and then a least-squares
spectrum or genetic algorithm adjustment is performed using passive neutron activation
dosimetry measurements to produce a “characterized” neutron spectrum. The resulting
spectrum can then be considered “characterized” in that it quantifies the “true” neutron-
fluence energy spectrum with a stated accuracy, including energy dependent uncertainties
and a covariance matrix. The process for determining a characterized neutron spectrum is to
first generate an a priori 640-group and 89-group neutron energy trial spectra from the
MCNP models presented in Section 2 at the 6-cm diameter tally sphere location. If the
least-squares spectrum is found to have inconsistent inputs when comparing the MCNP a
priori trial spectrum and the measured dosimetry results, as determined by a measure of the
chi-squared per degree of freedom (2/dof) in the spectrum adjustment, the dosimetry
measurements would be reexamined and/or the MCNP model would be reexamined and
modified or modeled with greater fidelity. The results of the MCNP calculations of the
WSMR environment represent a good initial representation for the neutron spectrum
characterization.
Also required in the analysis was an initial uncertainty estimate in the neutron spectrum
as a function of energy, an initial energy-dependent correlation matrix, the energy
dependent self-shielding factors, and the dosimetry cross section library that also included
uncertainties and covariance matrices [15]. In addition to the counting uncertainty, an
additional 2% uncertainty was included for the foils to address uncertainty contributions
due to positioning and possible geometrical effects in the central region of the core. The
output was in the 89-group NuGET format described earlier.
The resulting values for 2/dof are 0.74 for the 6-inch testing position, which represents
a highly acceptable value. Figure 5 and Figure 6 for the 6-inch location show the same
representation of the neutron energy spectra described in the Experimentation section, with
the additional 89-group adjusted spectrum result from the LSL analysis (blue). The adjusted
neutron fluence in Figure 6 shows very close agreement to that calculated using the MCNP
model.
6
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
less than 1%. A similar but less resolved structure is also seen in the 89-group energy
fluence. The 89-group energy structure is the energy grouping used in the NuGET code [5].
4 Experiments
The results of the MCNP and MERCURY calculations of the WSMR environment
represent a good initial representation for the neutron spectrum characterization. However,
a considerable uncertainty in the results can exist due to model representation uncertainty
(geometry, density, and composition) and uncertainty in the transport cross sections. An
improved spectrum can be obtained by combining this initial trial spectrum with measured
integral values that correspond to the reactions rates from high-fidelity passive dosimetry
reactions. For this work, the LSL-M3 code was used to generate an 89-group neutron
energy spectrum that is used in the NuGET code [5].
The selection of passive dosimetry foils and activation reactions has been studied and
evaluated over many years. A summary of the work can be found in noted references of this
manuscript [1][6][10][11][13]. No complete and perfect set of activation reactions exists
that allows the neutron fluence energy spectrum to be calculated by dosimetry alone.
However, there are enough reactions to cover the relevant energy range with high -fidelity
dosimetry cross sections to allow for adjusted neutron fluence results to be generated with a
quantified accuracy. The passive dosimetry foils and the associated neutron activation
reactions used to perform the neutron fluence characterization for the WSMR 6-inch (152.4
mm) and 24-inch (609.6 mm) environments are shown in Table 4 and Table 5.
The foils and activation reactions chosen for the analysis represent expert judgment and
references to previous work [1][10][13]. A complete set of dosimetry foils and reaction data
will vary for a given neutron environment being characterized [2]. Typically, neutron
activation resulting in the emission of protons (n,p), neutrons (n,2n), (n,n’), or alpha
particles (n,) represent high-neutron-energy reactions of 1 MeV or greater. Neutron
activation resulting in prompt gamma-ray emission from radiative capture (n,) or fission
reactions determine the shape of the thermal and epithermal region of the neutron spectrum.
Covering foils with cadmium (Cd) and/or boron (B) can allow for resonances above the
associated cutoff energies to become more prominent, allowing for additional integral
quantities to be included in the analysis. Typically, only thermal activation or fission foils
are covered in Cd or B.
A total of 23 different foil types, resulting in 35 different reactions, were irradiated [2].
Three fission foils (U-235, U-238, and Np-237) were irradiated individually in a bare or
cadmium cup configuration in a single 24 MJ steady-state operation. An historic Pu-239
foil activity was used from an older dosimetry set characterized and run with a nickel foil
for tracking purposes. This historic activity was nickel corrected to match the 24 MJ steady-
state operation conducted in 2022 [2].
The free-field environment was used for all the foil irradiations noted previously. A
high variation in the flux is known to exist radially and axially within the reactor cell.
However, a test region was identified where the neutron fluence varies 5%, this region is
the highlighted (yellow) section on the curved plates shown in Figure 4. Additional
uncertainty in the analysis was included to account for possible variation due to geometry
effects.
Figure 4. Typical Dosimetry Foils – Characterization Test Setup.
5. Spectrum Characterization Results
The neutron energy spectrum is first calculated using MCNP and then a least-squares
spectrum or genetic algorithm adjustment is performed using passive neutron activation
dosimetry measurements to produce a “characterized” neutron spectrum. The resulting
spectrum can then be considered “characterized” in that it quantifies the “true” neutron-
fluence energy spectrum with a stated accuracy, including energy dependent uncertainties
and a covariance matrix. The process for determining a characterized neutron spectrum is to
first generate an a priori 640-group and 89-group neutron energy trial spectra from the
MCNP models presented in Section 2 at the 6-cm diameter tally sphere location. If the
least-squares spectrum is found to have inconsistent inputs when comparing the MCNP a
priori trial spectrum and the measured dosimetry results, as determined by a measure of the
chi-squared per degree of freedom (2/dof) in the spectrum adjustment, the dosimetry
measurements would be reexamined and/or the MCNP model would be reexamined and
modified or modeled with greater fidelity. The results of the MCNP calculations of the
WSMR environment represent a good initial representation for the neutron spectrum
characterization.
Also required in the analysis was an initial uncertainty estimate in the neutron spectrum
as a function of energy, an initial energy-dependent correlation matrix, the energy
dependent self-shielding factors, and the dosimetry cross section library that also included
uncertainties and covariance matrices [15]. In addition to the counting uncertainty, an
additional 2% uncertainty was included for the foils to address uncertainty contributions
due to positioning and possible geometrical effects in the central region of the core. The
output was in the 89-group NuGET format described earlier.
The resulting values for 2/dof are 0.74 for the 6-inch testing position, which represents
a highly acceptable value. Figure 5 and Figure 6 for the 6-inch location show the same
representation of the neutron energy spectra described in the Experimentation section, with
the additional 89-group adjusted spectrum result from the LSL analysis (blue). The adjusted
neutron fluence in Figure 6 shows very close agreement to that calculated using the MCNP
model.
7
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 5. WSMR 6-inch – SNL LSL Adjusted 89-Group Neutron Lethargy Fluence Energy Spectrum
Compared to the MCNP Calculated Results (linear–log).
Figure 6. WSMR 6-inch – SNL LSL Adjusted 89-Group Neutron Lethargy Fluence Energy Spectrum
Compared to the MCNP Calculated Results (log–log).
The results for some integral metrics and conversion factors are shown in Table 1. The
total neutron fluence is often normalized to 1.00 and the other values for fluence are in
reference to this value. These values were calculated as part of the LSL analysis.
Conversion values to translate to n/cm2 are given for fissions in the reactor, MJ of reactor
energy, and 58Ni(n,p)58Co activity at the characterized location on the Y-plate. It should be
noted that portions of the energy spectrum are highly correlated, so the uncertainty in the
integral metric can be much less than the average uncertainty [9][14].
Table 1. WSMR 6-inch Location – SNL Integral Neutron Spectrum Metrics and Associated
Uncertainties.
Metric
Integral
Response
Standard
Deviation
(%)
Total Neutron Fluence
Average Neutron Energy = 1.445 MeV
1.00
---
Fluence > 3 MeV
0.137
5.2
Fluence > 1 MeV
0.513
3.6
Fluence > 100 keV
0.968
0.58
Fluence > 10 keV
0.997
0.46
Fluence < 1 keV
0.001
10.7
Fluence < 1 eV
0.0002
31.6
Fluence 1-MeV(Si) Eqv. E722-19
(Ref. 1-MeV value = 95 MeV-mb)
83.8 MeV-mb
0.897
1.1
Total Fluence Conversion ([n/cm2]/fission)
6.791E-04
0.3
Total Fluence Conversion ([n/cm2]/MJ)
2.403E+13
0.5
Total Neutron Silicon Dose
(rad[Si]/MJ)
1.677E+03
---
Ionizing Si Dose (rad[Si]/[n/cm2])
9.951E+02
---
Percent Neutron Si Dose Ionizing (%)
59.3
---
Total Prompt Gamma-Ray Ionizing Silicon Dose
(rad[Si]/MJ)
9.476E+02
Total Delayed Gamma-Ray Ionizing Silicon Dose
(rad[Si]/MJ)
2.162E+03
Total Ionizing Dose (rad[Si]/MJ)
4.105E+03
Figure 7 shows the results for the sulfur radial neutron fluence profile for the free -field
environment at an axial position of 3 inches from the bottom of the test table for the sulfur
pellets arranged perpendicular from the core. The nickel foils and sulfur tablets were
arranged on an extended piece of tape that spanned from the 6-inch position to end of the
test table on all three legs of the y-plate. The solid line represents the average value for all
of the pellets measured on the ‘B-Leg’ at standard 1-inch increments. The counting
uncertainties are within each data point symbol. The results show that for the nickel and
sulfur activation reactions, there is no significant variation in the fast neutron fluence
between the legs on the y-plate.
8
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 5. WSMR 6-inch – SNL LSL Adjusted 89-Group Neutron Lethargy Fluence Energy Spectrum
Compared to the MCNP Calculated Results (linear–log).
Figure 6. WSMR 6-inch – SNL LSL Adjusted 89-Group Neutron Lethargy Fluence Energy Spectrum
Compared to the MCNP Calculated Results (log–log).
The results for some integral metrics and conversion factors are shown in Table 1. The
total neutron fluence is often normalized to 1.00 and the other values for fluence are in
reference to this value. These values were calculated as part of the LSL analysis.
Conversion values to translate to n/cm2 are given for fissions in the reactor, MJ of reactor
energy, and 58Ni(n,p)58Co activity at the characterized location on the Y-plate. It should be
noted that portions of the energy spectrum are highly correlated, so the uncertainty in the
integral metric can be much less than the average uncertainty [9][14].
Table 1. WSMR 6-inch Location – SNL Integral Neutron Spectrum Metrics and Associated
Uncertainties.
Metric
Integral
Response
Standard
Deviation
(%)
Total Neutron Fluence
Average Neutron Energy = 1.445 MeV 1.00 ---
Fluence > 3 MeV
0.137
5.2
Fluence > 1 MeV
0.513
3.6
Fluence > 100 keV
0.968
0.58
Fluence > 10 keV
0.997
0.46
Fluence < 1 keV
0.001
10.7
Fluence < 1 eV 0.0002 31.6
Fluence 1-MeV(Si) Eqv. E722-19
(Ref. 1-MeV value = 95 MeV-mb)
83.8 MeV-mb
0.897 1.1
Total Fluence Conversion ([n/cm2]/fission) 6.791E-04 0.3
Total Fluence Conversion ([n/cm2]/MJ) 2.403E+13 0.5
Total Neutron Silicon Dose
(rad[Si]/MJ) 1.677E+03 ---
Ionizing Si Dose (rad[Si]/[n/cm2]) 9.951E+02 ---
Percent Neutron Si Dose Ionizing (%)
59.3
---
Total Prompt Gamma-Ray Ionizing Silicon Dose
(rad[Si]/MJ) 9.476E+02
Total Delayed Gamma-Ray Ionizing Silicon Dose
(rad[Si]/MJ) 2.162E+03
Total Ionizing Dose (rad[Si]/MJ) 4.105E+03
Figure 7 shows the results for the sulfur radial neutron fluence profile for the free -field
environment at an axial position of 3 inches from the bottom of the test table for the sulfur
pellets arranged perpendicular from the core. The nickel foils and sulfur tablets were
arranged on an extended piece of tape that spanned from the 6-inch position to end of the
test table on all three legs of the y-plate. The solid line represents the average value for all
of the pellets measured on the ‘B-Leg’ at standard 1-inch increments. The counting
uncertainties are within each data point symbol. The results show that for the nickel and
sulfur activation reactions, there is no significant variation in the fast neutron fluence
between the legs on the y-plate.
9
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 7. A-Leg (Red Dotted) and B-Leg (Black Solid) Nickel and Sulfur Normalized Radial Neutron
Fluence Profiles, referenced from the center of the FBR.
7 Conclusions
This report presents the characterized neutron radiation environments for the free-field
environment for the WSMR FBR, called Molly-G [1]. The characterized 6-inch location of
the y-plate on the test table is presented. 640-group and 89-group neutron energy spectra
were calculated using MCNP, utilizing a newly created high-fidelity model of WSMR FBR
by SNL. Each neutron spectrum was adjusted to align more closely with neutron-activation
dosimetry. The adjustment was performed using the least-squares code LSL-M3, a
modified version of LSL-M2 [9][14]. Neutron conversion factors are presented for various
dosimetry readings into radiation metrics desired by experimenters.
Acknowledgement: This article has been authored by an employee of National Technology &
Engineering Solutions of Sandia, LLC under Contract No. DE-NA0003525 with the U.S. Department
of Energy (DOE). The employee owns all right, title and interest in and to the article and is solely
responsible for its contents. The United States Government retains and the publisher, by accepting the
article for publication, acknowledges that the United States Government retains a non-exclusive, paid-
up, irrevocable, world-wide license to publish or reproduce the published form of this article or allow
others to do so, for United States Government purposes. The DOE will provide public access to these
results of federally sponsored research in accordance with the DOE Public Access Plan.
References
[1] D.R. Redhouse (2022), E.S. Lum, J. Koglin, “Radiation Characterization Summary: WSMR-
Molly G at the 6-Inch and 24-Inch Irradiation Locations,” SAND2022-13797, Sandia
National Laboratories, Albuquerque, NM, October 2022
[2] ASTM E720-16 – “Standard Guide for Selection and Use of Neutron Sensors for Determining
Neutron Spectra Employed in Radiation-Hardness Testing of Electronics,” ASTM
Standard, Published 2016
[3] IRDFF-II (2014), v1.03, March 3, 2014, https://www-nds.iaea.org/IRDFF/
[4] MCNP – “A General Monte Carlo N-Particle Transport Code, Version 6.1,” LA-UR-03-1987,
Los Alamos, NM, April 2003
[5] K. R. DePriest (2004) and P. J. Griffin, “NuGET User’s Guide: Revision 0,” SAND2004-1567,
Sandia National Laboratories, Albuquerque, NM, April 2004
[6] K. R. DePriest (2006), P. J. Cooper, E. J. Parma, “MCNP/MCNPX Model of the Annular Core
Research Reactor,” SAND Report SAND2006-3067, Sandia National Laboratories,
Albuquerque, NM, May 2006
[7] L. B. Engle (1962), P. C. Fisher, Los Alamos Report LAMS-2642, 1962
[8] P. C. Fisher (1964), L. B. Engle, “Delayed Gammas from Fast-Neutron Fission of Th232, U233,
U235, U238, and Pu239,” Physical Review, Vol. 134, Number 4B, pp. B796 – B816, 25
May 1964
[9] P. J. Griffin (1994a), J. G. Kelly, and J.W. VanDenburg, “User's Manual for SNL-SAND-II
Code,” SAND93-3957, April 1994
[10] P. J. Griffin (1994b), J. G. Kelly, and D. W. Vehar, “Updated Neutron Spectrum
Characterization of SNL Baseline Reactor Environments,” SAND93-2554, April 1994
[11] P. J. Griffin (2011a), C. D. Peters, and D. W. Vehar, “Recommended Neutron Dosimetry Cross
Sections for the Characterization of Neutron Fields,” RADECS 2011 Proceedings, 2011
[12] W.N. McElroy (1967), S. Berg, T. Crockett, R. G. Hawkins, “A Computer-Automated Iterative
Method for Neutron Flux Spectra Determination by Foil Activation, Vol. II: SAND-II
(Spectrum Analysis by Neutron Detectors II) and Associated Codes,” AFWL-TR-67-41,
September 1967
[13] E. J. Parma (2014), T. J. Quirk, L. L. Lippert, P. J. Griffin, G. E. Naranjo, and S. M. Luker,
“Radiation Characterization Summary: ACRR 44-Inch Lead-Boron Bucket Located in the
Central Cavity onf the 32-Inch Pedestal at the Core Centerline (ACRR-LB44-CC-32-cl),”
SAND13-3406, April 2013
[14] F. W. Stallmann (1985), “LSL-M2: A Computer Program for Least-Squares Logarithmic
Adjustment of Neutron Spectra,” NUREG/CR-4349, ORNL/TM-9933, March 1985
[15] A. Trkov, P.J. Griffin, S.P. Simakov, et. al, “IRDFF-II: A New Neutron Metrology Library,”
Nuclear Data Sheets, Vol. 163, pg. 1-108, January 2020
[16] D. Loaiza, D. Gehman (2006), “End of an Era for the Los Alaomos CCritical Experiments
Facility: History of Critical Assemblies and Experiments (1946-2004)”, Annuals of Nuclear
Energy, Vol. 33, pg. 1339-1359, September 2006
[17] H.C. Paxton (1983), “A History of Critical Experiments at Pajarito Site”, Los Alamos National
Laboratory, LA-9685-H, March 1983
10
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Figure 7. A-Leg (Red Dotted) and B-Leg (Black Solid) Nickel and Sulfur Normalized Radial Neutron
Fluence Profiles, referenced from the center of the FBR.
7 Conclusions
This report presents the characterized neutron radiation environments for the free-field
environment for the WSMR FBR, called Molly-G [1]. The characterized 6-inch location of
the y-plate on the test table is presented. 640-group and 89-group neutron energy spectra
were calculated using MCNP, utilizing a newly created high-fidelity model of WSMR FBR
by SNL. Each neutron spectrum was adjusted to align more closely with neutron-activation
dosimetry. The adjustment was performed using the least-squares code LSL-M3, a
modified version of LSL-M2 [9][14]. Neutron conversion factors are presented for various
dosimetry readings into radiation metrics desired by experimenters.
Acknowledgement: This article has been authored by an employee of National Technology &
Engineering Solutions of Sandia, LLC under Contract No. DE-NA0003525 with the U.S. Department
of Energy (DOE). The employee owns all right, title and interest in and to the article and is solely
responsible for its contents. The United States Government retains and the publisher, by accepting the
article for publication, acknowledges that the United States Government retains a non-exclusive, paid-
up, irrevocable, world-wide license to publish or reproduce the published form of this article or allow
others to do so, for United States Government purposes. The DOE will provide public access to these
results of federally sponsored research in accordance with the DOE Public Access Plan.
References
[1] D.R. Redhouse (2022), E.S. Lum, J. Koglin, “Radiation Characterization Summary: WSMR-
Molly G at the 6-Inch and 24-Inch Irradiation Locations,” SAND2022-13797, Sandia
National Laboratories, Albuquerque, NM, October 2022
[2] ASTM E720-16 – “Standard Guide for Selection and Use of Neutron Sensors for Determining
Neutron Spectra Employed in Radiation-Hardness Testing of Electronics,” ASTM
Standard, Published 2016
[3] IRDFF-II (2014), v1.03, March 3, 2014, https://www-nds.iaea.org/IRDFF/
[4] MCNP – “A General Monte Carlo N-Particle Transport Code, Version 6.1,” LA-UR-03-1987,
Los Alamos, NM, April 2003
[5] K. R. DePriest (2004) and P. J. Griffin, “NuGET User’s Guide: Revision 0,” SAND2004-1567,
Sandia National Laboratories, Albuquerque, NM, April 2004
[6] K. R. DePriest (2006), P. J. Cooper, E. J. Parma, “MCNP/MCNPX Model of the Annular Core
Research Reactor,” SAND Report SAND2006-3067, Sandia National Laboratories,
Albuquerque, NM, May 2006
[7] L. B. Engle (1962), P. C. Fisher, Los Alamos Report LAMS-2642, 1962
[8] P. C. Fisher (1964), L. B. Engle, “Delayed Gammas from Fast-Neutron Fission of Th232, U233,
U235, U238, and Pu239,” Physical Review, Vol. 134, Number 4B, pp. B796 – B816, 25
May 1964
[9] P. J. Griffin (1994a), J. G. Kelly, and J.W. VanDenburg, “User's Manual for SNL-SAND-II
Code,” SAND93-3957, April 1994
[10] P. J. Griffin (1994b), J. G. Kelly, and D. W. Vehar, “Updated Neutron Spectrum
Characterization of SNL Baseline Reactor Environments,” SAND93-2554, April 1994
[11] P. J. Griffin (2011a), C. D. Peters, and D. W. Vehar, “Recommended Neutron Dosimetry Cross
Sections for the Characterization of Neutron Fields,” RADECS 2011 Proceedings, 2011
[12] W.N. McElroy (1967), S. Berg, T. Crockett, R. G. Hawkins, “A Computer-Automated Iterative
Method for Neutron Flux Spectra Determination by Foil Activation, Vol. II: SAND-II
(Spectrum Analysis by Neutron Detectors II) and Associated Codes,” AFWL-TR-67-41,
September 1967
[13] E. J. Parma (2014), T. J. Quirk, L. L. Lippert, P. J. Griffin, G. E. Naranjo, and S. M. Luker,
“Radiation Characterization Summary: ACRR 44-Inch Lead-Boron Bucket Located in the
Central Cavity onf the 32-Inch Pedestal at the Core Centerline (ACRR-LB44-CC-32-cl),”
SAND13-3406, April 2013
[14] F. W. Stallmann (1985), “LSL-M2: A Computer Program for Least-Squares Logarithmic
Adjustment of Neutron Spectra,” NUREG/CR-4349, ORNL/TM-9933, March 1985
[15] A. Trkov, P.J. Griffin, S.P. Simakov, et. al, “IRDFF-II: A New Neutron Metrology Library,”
Nuclear Data Sheets, Vol. 163, pg. 1-108, January 2020
[16] D. Loaiza, D. Gehman (2006), “End of an Era for the Los Alaomos CCritical Experiments
Facility: History of Critical Assemblies and Experiments (1946-2004)”, Annuals of Nuclear
Energy, Vol. 33, pg. 1339-1359, September 2006
[17] H.C. Paxton (1983), “A History of Critical Experiments at Pajarito Site”, Los Alamos National
Laboratory, LA-9685-H, March 1983
11
EPJ Web of Conferences 308, 06008 (2024) https://doi.org/10.1051/epjconf/202430806008
ISRD 17
Available via license: CC BY 4.0
Content may be subject to copyright.
