Progress of Thermal Scattering Law Development and ......Progress of Thermal Scattering Law...
Transcript of Progress of Thermal Scattering Law Development and ......Progress of Thermal Scattering Law...
-
Progress of Thermal Scattering Law Development and Evaluations
at North Carolina State University
Cole Manring, Colby Sorrell, Ben LarameeAyman I. Hawari
Nuclear Reactor Program
Department of Nuclear Engineering
North Carolina State University
Raleigh, North Carolina, USA
Technical Program Review
Nuclear Criticality Safety ProgramMarch 26 – 27, 2019 • Amarillo, TX, USA
-
Acknowledgement
The many graduate students, postdocs, andresearch staff at North Carolina StateUniversity
Collaboration with LLNL and Bettis labs◼ David Heinrichs, Michael Zerkle, Jesse Holmes
Funding◼ US NNSA Nuclear Criticality Safety program
◼ US Naval Nuclear Propulsion Program
-
FY 2018/2019 4 new TSL evaluations
◼ 3 first-of-a-kind evaluations
Modern predictive methods for thermal neutron crosssection calculations based on the use of atomisticsimulations
◼ Ab initio lattice dynamics
◼ Molecular dynamics (ab initio and classical) New materials
All states of matter (solid, liquid, gas)
Imperfect structure
Expanding the FLASSH thermal scattering analysis platform◼ Removed approximations such as incoherent approximation,
cubic approximation, atom site approximation and SCT approximation
Initiated the integration of the Generalized Nuclear Data Structure (GNDS) format into FLASSH
-
Objectives
-
Objectives
Completed light water (H2O) Finalizing molten salt FLiBe
-
( ) ( )2
coh incoh s
d σ 1 ES κ,ω S κ,ω
d dE 4π E
= +
)()()( ω,κSω,κSω,κS ds
+=
is composed of two parts)( ω,κS
The scattering law
Using first Born approximation combined with Fermi pseudopotential, it can
be shown that the double differential scattering cross section has the form
Van Hove’s space-time formulation
( ) ( ) ( )1
, ,2
i r tS G r t e drdt
−
− −
=
where G( Ԧ𝑟,t) is the dynamic pair correlation function and can be expressed in terms of time dependent atomic positions.
Neutron Thermalization
-
2
( , )2
s
Binelastic
d ES
d dE k T E
=
Tk
EE
B
−= Energy transfer
Tk
)cosEE2EE(
B
−+= Momentum transfer
−
−−= dteeS ttis)(
2
1),(
deet ti 2/ 1
)2/sinh(
)(
2)(
−
−−=
() – density of states (e.g., phonon frequency distribution)
The scattering law (TSL) is the Fourier transform of a Gaussian correlation
function
( ) ( )s B sS , k T S κ,ω =
Since 1960s
GASKET
NJOY/LEAPR
INCOHERENT
APPROXIMATION
-
Thermal Scattering Law Analysis
Key development in the last 20 years isthe use of atomistic simulations methodsto support the evaluation process
◼ Produce data necessary to calculate the TSL including
DOS for evaluation of TSL
Direct access to TSL using correlation analysis
-
ENDF/B-VIII TSL EvaluationsMaterial ENDF Library Name Evaluation
BasisInstitution
Beryllium metal tsl-Be-metal.endf DFT/LD NCSUBeryllium oxide (beryllium) tsl-BeinBeO.endf DFT/LD NCSUBeryllium oxide (oxygen) tsl-OinBeO.endf DFT/LD NCSULight water (hydrogen) tsl-HinH2O.endf MD CABLight water ice (hydrogen) tsl-HinIceIh.endf DFT/LD BAPLLight water ice (oxygen) tsl-OinIceIh.endf DFT/LD BAPLHeavy water (deuterium) tsl-DinD2O.endf MD CABHeavy water (oxygen) tsl-OinD2O.endf MD CAB
Polymethyl Methacrylate (Lucite)
tsl-HinC5O2H8.endfMD NCSU
Polyethylene tsl-HinCH2.endf MD NCSUCrystalline graphite tsl-graphite.endf MD NCSU
Reactor graphite(10% porosity)
tsl-reactor-graphite-10P.endf
MD NCSU
Reactor graphite(30% porosity)
tsl-reactor-graphite-30P.endf
MD NCSU
Silicon carbide (silicon) tsl-CinSiC.endf DFT/LD NCSUSilicon carbide (carbon) tsl-SiinSiC.endf DFT/LD NCSUSilicon dioxide (alpha phase) tsl-SiO2-alpha.endf DFT/LD NCSUSilicon dioxide (beta phase) tsl-SiO2-beta.endf DFT/LD NCSUYttrium hydride (hydrogen) tsl-HinYH2.endf DFT/LD BAPLYttrium hydride (yttrium) tsl-YinYH2.endf DFT/LD BAPLUranium dioxide (oxygen) tsl-OinUO2.endf DFT/LD NCSUUranium dioxide (uranium) tsl-UinUO2.endf DFT/LD NCSUUranium nitride (nitrogen) tsl-NinUN.endf DFT/LD NCSUUranium nitride (uranium) tsl-UinUN.endf DFT/LD NCSU
-
ENDF/B-VIII TSL EvaluationsMaterial ENDF Library Name Evaluation
BasisInstitution
Beryllium metal tsl-Be-metal.endf DFT/LD NCSUBeryllium oxide (beryllium) tsl-BeinBeO.endf DFT/LD NCSUBeryllium oxide (oxygen) tsl-OinBeO.endf DFT/LD NCSULight water (hydrogen) tsl-HinH2O.endf MD CABLight water ice (hydrogen) tsl-HinIceIh.endf DFT/LD BAPLLight water ice (oxygen) tsl-OinIceIh.endf DFT/LD BAPLHeavy water (deuterium) tsl-DinD2O.endf MD CABHeavy water (oxygen) tsl-OinD2O.endf MD CAB
Polymethyl Methacrylate (Lucite)
tsl-HinC5O2H8.endfMD NCSU
Polyethylene tsl-HinCH2.endf MD NCSUCrystalline graphite tsl-graphite.endf MD NCSU
Reactor graphite(10% porosity)
tsl-reactor-graphite-10P.endf
MD NCSU
Reactor graphite(30% porosity)
tsl-reactor-graphite-30P.endf
MD NCSU
Silicon carbide (silicon) tsl-CinSiC.endf DFT/LD NCSUSilicon carbide (carbon) tsl-SiinSiC.endf DFT/LD NCSUSilicon dioxide (alpha phase) tsl-SiO2-alpha.endf DFT/LD NCSUSilicon dioxide (beta phase) tsl-SiO2-beta.endf DFT/LD NCSUYttrium hydride (hydrogen) tsl-HinYH2.endf DFT/LD BAPLYttrium hydride (yttrium) tsl-YinYH2.endf DFT/LD BAPLUranium dioxide (oxygen) tsl-OinUO2.endf DFT/LD NCSUUranium dioxide (uranium) tsl-UinUO2.endf DFT/LD NCSUUranium nitride (nitrogen) tsl-NinUN.endf DFT/LD NCSUUranium nitride (uranium) tsl-UinUN.endf DFT/LD NCSU
-
Thermal Scattering Cross-Sections
Evaluation DFT/LD
PHONONVASP
Optimize the
system structure
Generate
displacements
Evaluate
Hellmann–
Feynman Forces
Evaluate the
Dynamical Matrix
Sample the
Brillouin Zone
DFT
Pseudo-
potential
Quantum
Mechanical
model
LD
Phonon
model
Harmonic
potential
Density of StatesThermal Scattering
Law S(α, β)
FLASSHThermal Scattering Cross Section
-
Thermal Scattering Cross-Sections
Evaluation MD/QM
Equilibrate the
systemGenerate particle
trajectories
Velocity Auto-
Correlation
Function (VACF)
Evaluate classical
Gcl(r, t) and Icl(k, t)
LAMMPS MD
Pair potential
Classical system
Density of States
Thermal Scattering
Law S(α, β)
Quantum Correction
Thermal Scattering
Cross SectionFLASSH
-
Computational Capabilities Hybrid mini cluster - 17 nodes
▪ 324 CPU cores▪ 22 Nvidia GPUs▪ Expanding…..
Parallel computations▪ Atomistic simulations▪ TSL analysis▪ Neutronic simulations▪ System design
VASP, PHONON, LAMMPS PREPRO, NJOY, FUDGE, SAMMY, MCNP, Serpent, GEANT4, McStas, PARET, RELAP, COMSOL
-
Heavy Paraffinic Oil
0.0 0.1 0.2 0.3 0.4 0.5 0.60
5
10
15
20
25
30
DO
S (
eV-1
)
Energy (eV)
0.000 0.001 0.0020.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
DO
S (
1/e
V)
Energy (eV)
0 500 1000 1500 2000-1.0-0.8-0.6-0.4-0.20.00.20.40.60.81.0
VA
CF
(ar
b)
Time (fs)
-
Light Water
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.70
1
2
3
4
5
6
7
8
9
10
DO
S (
eV
-1)
Energy (eV)
283K 93 bond
293K 93 bond
300K 93 bond
323K 94 bond
350K 94 bond
373K 94 bond
400K 94 bond
423K 94 bond
450K 94 bond
473K 94 bond
500K 94 bond
523K 94 bond
550K 94 bond
573K 94 bond
600K 94 bond
623K 94 bond
650K 94 bond
12 6
2inter intra 214 1 exp ( ) ( )
2i j i j
i jtot
r OH eq eq
i j i jO O O O ij
q qU U U D r r K
r r r
= + = − + + − − − + −
Flexible TIP4P/2005 potential
-
Light Water
0.02 0.04 0.06 0.08 0.100.96
0.98
1.00
1.02
1.04
1.06
Ratio
Energy (eV)
NCSU 473K/293K
ENDF8 473K/293K
ENDF7 450K/293K
Dritsa [EXFOR #22613] (473 K / 293 K)
10-6
10-5
10-4
10-3
10-2
10-1
100
101
102
10-40
10-38
10-36
10-34
10-32
10-30
10-28
10-26
10-24
10-22
10-20
10-18
10-16
10-14
10-12
10-10
10-8
10-6
10-4
10-2
100
102
104
106
S(a
lpha,b
eta
)
Beta
8.19721E-8
1.31155E-6
3.27888E-5
1.31155E-4
0.00118
0.01312
0.11804
1.06236
10.28258
100.4158
819.72081
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102 103 10410-4010-3810-3610-3410-3210-3010-2810-2610-2410-2210-2010-1810-1610-1410-1210-1010-810-610-410-2100102104106108
1010
S(a
lpha
,be
ta)
Alpha
0
1E-6
1.02345E-5
1.04745E-4
0.001
0.01027
0.10507
1.0062
10.29797
105.39471
398.10717
1E-5 1E-4 0.001 0.01 0.1 1 1010
100
1000
Tota
l X
S (
ba
rns)
Energy (eV)
283K
293K
300K
323K
350K
373K
400K
423K
450K
473K
500K
523K
550K
573K
600K, 80 atm
623K, 100 atm
650K, 150 atm
File 7
-
Liquid FLiBe – FY 2019 Eutectic with a mixture of
2:1 ratio of LiF and BeF2
Melting Point: 732K Boiling Point:1703K
DFT and MD analysis (with QM corrections)
TSL evaluation between 750K and 1500K
0.00 0.05 0.10 0.15
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
De
nsity o
f S
tate
s (
1/e
V)
Energy (eV)
Li
Born-Mayer potential
-
Hydrofluoric Acid
Strong hydrogen bonding
Unique molecular structure dynamics
Example CMD Potential:
LJ + 3 pt. chg.Perturbed Morse Osc.
𝑉 𝑟 = 𝐶 𝑒𝑥𝑝 −𝐷 𝑟 − 𝑟0 − 12 + 𝐸𝑟𝑉 𝑟 = 𝐴
𝐵
𝑟
12
−𝐵
𝑟
6
+⋯
-
FLASSH Code
-
FLASSH Code FeaturesNJOY FLASSH
Coherent Inelastic
Remove Incoherent Approximation
Remove Short Collision Time (SCT) Approximation
Integral against alpha differential cross section
NumericalDefault: Analytical
Optional: Numerical
a, B Gridding User inputDefault: Automatic grid
Optional: User input
Parallel Computing Using OpenMP
Graphite User Interface
Syntax and Error Checking
Coherent Elastic Calculation Supported Structure Hexagonal, FCC, BCC Any crystal structure
Supported MaterialsGraphite, Beryllium, Beryllium Oxide, Aluminum, Lead, Iron
Any material
Compound Materials 2 elements with ratio 1:1Any number of elements
with any ratio
Remove Cubic Approximation
Remove Atom Site Approximation
Coherent Elastic Scattering Cross Section
Over Ewald SphereOn every reciprocal lattice
point
Need to modify source code if calculating other materials
Yes No
-
Liquid Physics in FLASSH
Separation of the diffusive DOS from the continuous (solid) DOS in LEAPR
Convolution of the solid and liquid TSL components
𝑆𝑡𝑜𝑡𝑎𝑙 𝛼, 𝛽 = 𝑆𝑑𝑖𝑓𝑓. 𝛼, 𝛽 ∗ 𝑆𝑐𝑜𝑛𝑡. 𝛼, 𝛽 𝛽
-
Liquid Physics in FLASSHConstruct fine beta grid (for convolution):
Call convolve_grid subroutine to determine appropriate resolution and lower/upper beta limits
Build liquid TSL model over the new beta grid:
Call liquid model function (e.g., lang for Langevin)
◼ Call besk1 (Bessel) function if necessary
Interpolate solid TSL onto new beta grid:
Call interp_grid subroutine to interpolate values for every convolution ‘window’
Convolve the liquid and solid components:
Call convolve subroutine to perform the convolution
Construct total TSL:
Add in extra DW term
Output results:
Write TSL components to various files
-
Free Gas◦ Assumes a Maxwellian velocity distribution
Crystal Lattice◦ Compound nucleus effects separated from lattice effects
◦ Transition probability
◦ Self Scattering Law
Identical to that used in thermal scattering
Describes the energy-momentum phase space of a material
Doppler Broadening
23
( )( )
2
20
'1 '' ( ', ) ( ') , ( ', ) expFG FG FG
E EEE dE S E E E S E E
E
− −= =
( )( ) ( )
2
0
2 2
0
( , )
4 / 2
s
B
SE d
E E k T
−
=
− − +
-
Ab initio lattice dynamics◦ Predictive density of states (DOS)
◦ Current DOS implemented in the ENDF/B-VIII.0 cross section library for U in UO2
Doppler Broadening
24
◦ Fluorite Structure
◦ 2x2x2 supercell
◦ GGA-PBE+U
-
Ab initio lattice dynamics◦ Predictive density of states (DOS)
◦ Current DOS implemented in the ENDF/B-VIII.0 cross section library for U in UO2
Doppler Broadening
25
◦ Fluorite Structure
◦ 2x2x2 supercell
◦ GGA-PBE+U
-
FLASSH Generalized TSL
Full Equation Cubic Approximation( )2 21
3sse k =
2
0
(2 th
2 2
)co
m s
B
W dM k T
k
=
Function of the density of states ρ(ω) Function of the polarization vector and dispersion relations
2
2 coth2 2
s
s
s
s B
eW
MN k T
=
( )2 0 20
2( e)
!
1,
2
U U
s
i
n
t
n
V
n
We e dtW
S Gn
+
−−
=
= =
( )( )
2
/2
1
1 1
(0) 2 sinh / 2
s
sB
e eG e
N k T
−= ( )( )
/ 2
1
1 ( )
2 sinh / 2G e
−=
-
5.49E-3 0 0
0 5.49E-3 0
0 0 5.49E-3
FLASSH Generalized TSL
27
Beryllium Metal
o HCP (P63/mmc)
o Ab initio lattice dynamics
o 4x4x3 Supercell Exact Debye-Waller Matrix (Å2)
-
5.76E-3 -1.58E-5 1.01E-6
-1.58E-5 5.74E-3 -5.18E-7
1.01E-6 -5.18E-7 5.02E-3
FLASSH Generalized TSL
28
Beryllium Metal
o HCP (P63/mmc)
o Ab initio lattice dynamics
o 4x4x3 Supercell Exact Debye-Waller Matrix (Å2)
-
FUDGE/GNDS Ongoing
collaboration with LLNL
-
Summary FY 2018
◼ 10 new TSL evaluations contributed to ENDF/B-VIII.0
◼ 5 first-of-a-kind evaluations
FY 2019◼ 4 new TSL evaluations
◼ 3 first-of-a-kind evaluations
Modern predictive methods for thermal neutron cross sectioncalculations based on the use of atomistic simulations
◼ Ab initio lattice dynamics
◼ Molecular dynamics (ab initio and classical) New materials
All states of matter (solid, liquid, gas)
Imperfect structure
FLASSH is a new thermal scattering analysis platform that uses a generalized theoretical approach for TSL calculations◼ Removed approximations such as incoherent approximation, cubic
approximation, atom site approximation and SCT approximation
Progress on FY 2019 tasks