PCA benchmark analysis using the ADVANTG3.0.1/MCNP6 ...indico.ictp.it/event/7994/session/7/contribution/34/...PCA benchmark “12/13” configuration – PCA core with components mock

Joint ICTP-IAEA Workshop on the Evaluation of Nuclear Reaction Data for Applications

2-13 Oct 2017, ICTP, Trieste, Italy

Joint ICTP-IAEA Workshop on the Evaluation of Nuclear Reaction Data for Applications

2-13 Oct 2017, ICTP, Trieste, Italy

PCA benchmark analysis using the ADVANTG3.0.1/MCNP6.1.1b codes

PCA benchmark analysis using the ADVANTG3.0.1/MCNP6.1.1b codes

Mario Matijević

University of Zagreb, Faculty of Electrical Engineering and Computing, Department of Applied Physics

Unska

3, 10000 Zagreb, Croatia

Mario Matijević

University of Zagreb, Faculty of Electrical Engineering and Computing, Department of Applied Physics

Unska

3, 10000 Zagreb, Croatia

[email protected]

mailto:[email protected]

Contents:Contents:

1. PCA benchmark definition2. PCA benchmark description3. ADVANTG3.0.1/MCNP6.1.1b codes4. PCA response functions5. PCA results using ADVANTG3.0.1/MCNP6.1.1b6. Summary and conclusions

2

Pool Critical Assembly Pressure Vessel (PCA) benchmarkPool Critical Assembly Pressure Vessel (PCA) benchmark––

A well known benchmark from the SINBAD databaseA well known benchmark from the SINBAD database––

Based on the PCA facility experiments at the ORNLBased on the PCA facility experiments at the ORNL––

A small poolA small pool--type highly enriched experimental reactortype highly enriched experimental reactor––

Measured and calculated (DORT) equivalent fission fluxesMeasured and calculated (DORT) equivalent fission fluxes––

Validation of transport theory codes and XS librariesValidation of transport theory codes and XS libraries––

Prediction of the inPrediction of the in--vessel neutron flux gradientsvessel neutron flux gradients

Scope of the PCA benchmarkScope of the PCA benchmark––

Validation of the methodology to predict the reaction rates in eValidation of the methodology to predict the reaction rates in exx--core regioncore region––

Qualification of the pressure vessel fluence calculation methodoQualification of the pressure vessel fluence calculation methodologylogy––

Simulation of the neutron flux gradient inside carbon steel (RPVSimulation of the neutron flux gradient inside carbon steel (RPV))––

RPV surveillance programs of existing USA RPV surveillance programs of existing USA NPPsNPPs

(RPV damage)(RPV damage)––

Measured RR inside RPV (A4, A5, A6) and water gap inMeasured RR inside RPV (A4, A5, A6) and water gap in--front (A3, A2)front (A3, A2)––

Well defined neutron source, materials, and simple geometryWell defined neutron source, materials, and simple geometry––

DORT libraries in the PCA benchmark: BUGLEDORT libraries in the PCA benchmark: BUGLE--93, SAILOR93, SAILOR--95, BUGLE95, BUGLE--9696––

Computational requirements by the U.S. NRC Regulatory Guide 1.19Computational requirements by the U.S. NRC Regulatory Guide 1.1900

1. PCA benchmark definition1. PCA benchmark definition

3

2. PCA benchmark description2. PCA benchmark description

PCA pressure vessel wall benchmark facility MC model of PCA benchmark facility (water removed)

PCA benchmark “12/13” configuration––

PCA core with components mock up the corePCA core with components mock up the core--toto--cavity region in cavity region in PWRsPWRs––

Al plate, Thermal shield (TS), pressure vessel simulator (PVS), Al plate, Thermal shield (TS), pressure vessel simulator (PVS), void box (VB)void box (VB)––

Water gap between Al plate and TS: 12 cmWater gap between Al plate and TS: 12 cm––

Water gap between TS and RPVS: 13 cmWater gap between TS and RPVS: 13 cm––

PCA facility is immersed in a large pool of water (coolant and mPCA facility is immersed in a large pool of water (coolant and moderator)oderator)––

PCA corePCA core

hashas

25 material test reactor (MTR) plate25 material test reactor (MTR) plate--type elements (e=93%)type elements (e=93%)

4


Cross sectional view through the control element

PCA standard MTR fuel element (water included)

PCA control rod (water included)

5

PCA benchmark “12/13” configuration

Lead

(11.34 g/cc)

Boron carbide

(1.6 g/cc)

Aluminum

Al-U alloy (fuel plate)


6



7



reaction rateseq

i

φσ

=

Equivalent fission fluxes at detector locations:

Experimental measurements: A1 – A7

( ) ( )

( ) ( )

( )

iE

eqi

E

E

E E dE

E E dE

E dE

σ φφ

σ ϕ

ϕ

=∫

∫∫

Maxwell

1/E slowing

fission

8


9

Reactions of interest (Reactions of interest (shielding calculationsshielding calculations))

––

237237Np(n,f)Np(n,f)137137Cs, Cs, 238238U(n,f)U(n,f)137137Cs, Cs, 103103Rh(n,n')Rh(n,n')130m130mRh, Rh, 115115In(n,n')In(n,n')115m115mIn, In, 5858Ni(n,p)Ni(n,p)5858Co, Co, 2727Al(n,Al(n,αα))2424NaNa

––

Results are given per unit PCA core neutron source (normalized)Results are given per unit PCA core neutron source (normalized)––

CalculatedCalculated--toto--measured (C/M) ratios of equivalent measured (C/M) ratios of equivalent 235235U fission fluxesU fission fluxes––

MeasurementsMeasurements

inin

core midplane (z=0 and y=0) core midplane (z=0 and y=0) atat

locations A1 to A7locations A1 to A7––

Experimental access tubes: steel in PV and Plexiglas in water loExperimental access tubes: steel in PV and Plexiglas in water locationscations––

Minimization of the perturbations of the neutron fieldMinimization of the perturbations of the neutron field––

DORT reaction rates (/atom/s) are given for dosimeters A1DORT reaction rates (/atom/s) are given for dosimeters A1--A8 A8

Critical configuration (Critical configuration (eigenvalue calculationseigenvalue calculations))––

Fully inserted control rod reaches bottom of the fuelFully inserted control rod reaches bottom of the fuel––

Withdrawn length is measured from the bottom of the fuelWithdrawn length is measured from the bottom of the fuel––

Safety rods (S1, S2 and S3) critical positions: 48.26 cmSafety rods (S1, S2 and S3) critical positions: 48.26 cm––

Regulating rod (RR) position: 38.43 cmRegulating rod (RR) position: 38.43 cm––

Total critical mass of Total critical mass of 235235U: 3336.01 gU: 3336.01 g

3. ADVANTG3.0.1/MCNP6.1.1B codes3. ADVANTG3.0.1/MCNP6.1.1B codesADVANTG3.0.1 code (ORNL)ADVANTG3.0.1 code (ORNL)

––

Automated meshAutomated mesh--based tool for generating VR parameters for MCNP codebased tool for generating VR parameters for MCNP code––

Approximate 3Approximate 3--D multigroup SN forward/adjoint transport solutionsD multigroup SN forward/adjoint transport solutions––

Denovo SN solver developed at ORNL with CADIS/FWDenovo SN solver developed at ORNL with CADIS/FW--CADIS formalismCADIS formalism––

VR parameters: spaceVR parameters: space--energy weightenergy weight--windows (WW) and biased source distributions windows (WW) and biased source distributions (SB) cards for (SB) cards for thethe

MCNP inputMCNP input

MCNP6.1.1b code (LLNL)MCNP6.1.1b code (LLNL)––

generalgeneral--purpose Monte Carlo Npurpose Monte Carlo N--Particle code with arbitrary 3D geometryParticle code with arbitrary 3D geometry––

neutron, photon, electron, or coupled neutron, photon, electron, or coupled n/p/en/p/e

transporttransport––

XS libraries: continuous, discrete, multigroup, XS libraries: continuous, discrete, multigroup, S(a,bS(a,b) law, dosimetry,) law, dosimetry,……––

powerful general source, rich collection of VR techniques, flexipowerful general source, rich collection of VR techniques, flexible talliesble tallies––

Pointwise XS data with MAKXSF for XS libraries with Doppler broaPointwise XS data with MAKXSF for XS libraries with Doppler broadeningdening

PCA model with MCNP in z=0 cm plane PCA model with MCNP in y=30.84 cm plane

10

11MCNP model of the ¼ cask ADVANTG SN mesh for VR parameters Adjoint function (1.8-2.4) MeV

point detector = adjoint source

forward flux

adjoint flux

detector

source

source

detector

forward transport:

adjoint transport:

source

detector

†1ˆ( , ) ( , ) ( , )q r E q r E r ER

φ=

†( , )( , )Rw r Er Eφ

=

0 †

( , )( , )ˆ( , ) ( , )q r E Rw r Eq r E r Eφ

= =

† ( , )( , )( , ) ( , )

d

d

r Eq r Er E r Eσ

σ φ=∫

Target weight

Initial weight

Weighted adjoint source

Biased source

CA

DIS

FW-

CA

DIS

Hybrid shielding

†( , ) ( , )dq r E r Eσ= Adjoint source

3. ADVANTG3.0.1/MCNP6.1.1B codes3. ADVANTG3.0.1/MCNP6.1.1B codes

4. PCA response functions4. PCA response functionsCross sectionCross sectionss from the IAEA NDS servicefrom the IAEA NDS service

––

ENDF/B retrieval service (ENDF/B retrieval service (wwwwww--nds.iaea.orgnds.iaea.org))––

IRDFIRDF--2002 and IRDFF2002 and IRDFF--2014 dosimetry libraries2014 dosimetry libraries

Selection of XS library and reaction type

XS reaction explorer XS reaction data in the ENDF-6 format

XS reaction plotting and extracting for MCNP

Cross sections with reconstructed resonances and applied Doppler broadening at the temperature 293

K

IRDFF-1.05 library

Al-27(n,a)Na-24

12

4. PCA response functions4. PCA response functionsCross sectionCross sectionss from the IAEA NDS servicefrom the IAEA NDS service

––

IRDFFIRDFF--2014 and IRDF2014 and IRDF--2002 dosimetry library2002 dosimetry library––

Neutron excitation (Neutron excitation (n,nn,n’’) of the first isomeric state ) of the first isomeric state RhRh

(56.12 min ) and In (4.486 h) (56.12 min ) and In (4.486 h) ––

MetastableMetastable

isomers isomers 103m103mRh and Rh and 115m115mIn with reactions MF/MT = 3/51In with reactions MF/MT = 3/51

[1] Joel A. Kulesza

& Roger L. Martz (2017) Evaluation of the Pool Critical

Assembly Benchmark with Explicitly Modeled Geometry Using MCNP6, Nuclear Technology, 197:3,

284-295, DOI: 10.1080/00295450.2016.1273711

[1]

13

DE/DF cards define the response

spectrum for point detector tallies

5. PCA results using ADVANTG3.0.1/MCNP6.1.1b5. PCA results using ADVANTG3.0.1/MCNP6.1.1b

14

MCNP6.1.1b eigenvalue results:

keff = (0.99924 ±

0.00100)

c c ------------------

criticality source criticality source ----------------------kcodekcode

2000 1.0 50 350 1e42000 1.0 50 350 1e4ksrcksrc

20.22 31.46 0 $ starting point20.22 31.46 0 $ starting pointc c --------------------------------------------------------------------------------------------

keff cycle number

keff cycle number

Entropyofsource

Averagekeff

keff neutron flux in z=0 cm plane keff neutron flux RE in z=0 cm plane


15

ADVANTG3.0.1 parameters:ADVANTG3.0.1 parameters:––

FWFW--CADIS method for one reaction on all detectors (A1CADIS method for one reaction on all detectors (A1--A8)A8)––

S4/P1, SC spatial differencing, S4/P1, SC spatial differencing, epseps=1e=1e--3, 1.6e6 mesh cells3, 1.6e6 mesh cells––

ANISNANISN--format coupled format coupled nn--gg

multigroup library BPLUS (updated BUGLEmultigroup library BPLUS (updated BUGLE--96)96)––

BUGLEBUGLE--96 library: LWR shielding, PV dosimetry, VITAMIN96 library: LWR shielding, PV dosimetry, VITAMIN--B6 collapsingB6 collapsing––

5 weighting spectra of 15 weighting spectra of 1--D model of a reactor cavity and D model of a reactor cavity and bioshieldbioshield––

BPLUS library: 47n/20g, 393 isotopes, ENDF/BBPLUS library: 47n/20g, 393 isotopes, ENDF/B--VII.0 evaluationVII.0 evaluation––

Mixed 11 pure materials into 26857 macro materialsMixed 11 pure materials into 26857 macro materials––

Memory requirements (GB): 6Memory requirements (GB): 6--9 FW Denovo, 109 FW Denovo, 10--14 ADJ Denovo14 ADJ Denovo

MCNP6.1.1b parameters:MCNP6.1.1b parameters:––

Continuous XS for neutron transport with Doppler correction and Continuous XS for neutron transport with Doppler correction and S(a,bS(a,b) law) law––

MCNP6.1.1b with distributed (PVM) calculation on Quad Core CPUMCNP6.1.1b with distributed (PVM) calculation on Quad Core CPU––

Using point detectors inside small void spheres at centers of A1Using point detectors inside small void spheres at centers of A1--A8A8––

Mesh tally for capturing the global MC neutron transportMesh tally for capturing the global MC neutron transport––

card card ““ctmectme

720720””

$ cumulative time in min $ cumulative time in min ––

About 2e6 neutron histories per MC simulationAbout 2e6 neutron histories per MC simulation––

Point detectors have Point detectors have RERE on average 1on average 1--2 % 2 %

Integrated forward flux in z=0 cm

Integrated adjoint flux in z=0 cm

WW first group in z=0 cm

5. PCA results using ADVANTG3.0.1/MCNP6.1.1b5. PCA results using ADVANTG3.0.1/MCNP6.1.1bADVANTG3.0.1ADVANTG3.0.1 FWFW--CADIS CADIS resultsresults for for 2727Al(n,a)Al(n,a)2424NiNi

Integrated adjoint flux with contours

(A8)

(A1)

(A8)

(A1)

(core)

(TS)

(RPV)

(VB)

ψ† w=R/ψ†

ψ ψ†

16


MC neutron flux solution for Al-27(n,a)Ni-24 in z=0 cm plane

MC neutron flux RE on 1 sigma level for Al-27(n,a)Ni-24 in z=0 cm plane

Location 237Np(n,f) 238U(n,f) 27Al(n,a) 58Ni(n,p) 115In(n,n') 103Rh(n,n') MCNPAvg

± sigDORT

Avg

± sig

A1 0.862 - 0.801 0.887 0.902 0.945 0.88 ± 0.02 0.91 ± 0.02

A2 - - 0.852 0.938 0.986 - 0.93 ± 0.04 0.92 ± 0.01

A3 0.908 - 0.765 0.835 0.877 - 0.85 ± 0.03 0.96 ± 0.02

A4 0.869 0.882 0.874 0.898 0.935 0.905 0.90 ± 0.01 0.94 ± 0.03

A5 0.914 0.880 0.929 0.919 0.935 0.877 0.91 ± 0.01 0.92 ± 0.03

A6 0.878 0.884 0.968 0.957 0.969 0.881 0.92 ± 0.02 0.91 ± 0.04

A7 0.939 - - - - - 0.94 ± 0.00 0.89 ± 0.00

A8 - - - - - - - -

MCNP6.1.1b results for all reactionsMCNP6.1.1b results for all reactions

Equivalent fission fluxes C/M ratios

(“-” experimental result not provided in the PCA benchmark)

RPV

17[Lower quality results since ADVANTG was used with default eps=1e-3]

5. PCA results using ADVANTG3.0.1/MCNP6.1.1b5. PCA results using ADVANTG3.0.1/MCNP6.1.1bIsotopes of Fe and selfIsotopes of Fe and self--shieldingshielding

––

Neutron spectrum Neutron spectrum ““softeningsoftening””

inside the RPV simulatorinside the RPV simulator––

Neutrons > 1 MeV are shifted to resonance region of Fe => selfNeutrons > 1 MeV are shifted to resonance region of Fe => self--shielding!shielding!––

Important reaction becomes inelastic neutron scattering (Important reaction becomes inelastic neutron scattering (n,nn,n’’))

Total cross section (MT=1) of iron isotopes Inelastic scattering (MT=3) of iron isotopes18

( )2

2 2 2, 02 2

0

( ) 2 sin 2( )sin 2 4 (2 1)sin( ) 1/ 4

nt l l l l

N gE E E N lE E

πσ δ δ π δΓ= Γ − Γ + − + +

− + ΓD

D

Hard-sphere nuclear model

( / )tan( / )

ll

l

J RN R

δ =D

D

l=0 (s-wave) neutrons

l=1 (p-wave) neutrons

0 /Rδ = D

11 0 tan ( / )Rδ δ −= − D

(LWR)

(FBR)

20 0

,0 0 2 20

4( )( ) 14( )t pot

E E E REE E E

σ σ σ−Γ ⎡ ⎤= + +⎢ ⎥− + Γ Γ⎣ ⎦D

1/v interpolation

Breit-

Wigner

Prob. tables


0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

10 20 30 40 50 60 70 80 90 100

Detector position (cm)

FOM

fact

or

237Np(n,f) 238U(n,f) 27Al(n,a) 58Ni(n,p) 115In(n,n') 103Rh(n,n')

MCNP FOM factors for all reactionsMCNP FOM factors for all reactionsComputational model metric addressing memory and CPU timeFigure-of-merit (FOM) factor: CPU time necessary to reach a given level REFOM = 1/(RE2xT), where RE is MC rel.error and T is CPU time in min

FOM factors for different detector locations

19

5. PCA results using ADVANTG3.0.1/MCNP6.1.1b5. PCA results using ADVANTG3.0.1/MCNP6.1.1bADVANTG3.0.1ADVANTG3.0.1 CADIS CADIS optimizationoptimization ofof A3A3 detectordetector

MC neutron flux solution for Al-27(n,a)Ni-24 in z=0 cm plane

MC neutron flux RE on 1 sigma level for Al-27(n,a)Ni-24 in z=0 cm plane

Integrated adjoint flux in

z=0 cmWW first group

in z=0 cm

ψ† w=R/ψ†

20

(A3)

(A3)

5. PCA results using ADVANTG3.0.1/MCNP6.1.1b5. PCA results using ADVANTG3.0.1/MCNP6.1.1bADVANTGADVANTG3.0.13.0.1 FWFW--CADIS global flux solutionCADIS global flux solution

Integrated contributon flux

in z=0 cm

Integrated forward and adjoint flux in

z=0 cm

MC neutron flux global solution in z=0 cm plane MC neutron flux RE in z=0 cm plane

(RPV) (TS)ψ

ψ†

more histories needed!

average RE ≈

30%

21

local SN m

esh

refinement

?

ψc

=ψψ†

Commercial hybrid shielding codes today:Commercial hybrid shielding codes today:––

SCALE6.2 (CADISSCALE6.2 (CADIS//FWFW--CADIS in MAVRIC)CADIS in MAVRIC)––

ADVANTG3.0.1 + MCNP5v1.6 (or MCNP6.1.1b)ADVANTG3.0.1 + MCNP5v1.6 (or MCNP6.1.1b)––

AA33MCNP (TORT + MCNP5v1.6)MCNP (TORT + MCNP5v1.6)––

ATTILA + MCNP6.1.1BATTILA + MCNP6.1.1B

6. Summary and conclusions6. Summary and conclusionsHybrid shielding:Hybrid shielding:

––

FWFW--CADIS for distributed detectors and global flux solutionCADIS for distributed detectors and global flux solution––

CADIS for local answers, would require 6 reactions x 8 locationsCADIS for local answers, would require 6 reactions x 8 locations

= 48 runs!= 48 runs!––

SN based VR technique removes burden of manual VR preparationSN based VR technique removes burden of manual VR preparation––

Computational tradeComputational trade--off between SN and MC simulationsoff between SN and MC simulationsPCA benchmark:PCA benchmark:

––

Good agreement between calculated (C) and measured (M) resultsGood agreement between calculated (C) and measured (M) results––

Obtained results in accordance with the Regulatory Guide 1.190Obtained results in accordance with the Regulatory Guide 1.190––

Sensitivity of BPLUS shielding library on the iron XS (selfSensitivity of BPLUS shielding library on the iron XS (self--shielding)shielding)––

Differences due to XS libraries, weighting spectrum, core modeliDifferences due to XS libraries, weighting spectrum, core modeling,ng,……––

More neutron groups are preferred due to spectral effectsMore neutron groups are preferred due to spectral effects

22

References:References:

1.1.

““Shielding Integral Benchmark Archive and DatabaseShielding Integral Benchmark Archive and Database,,””

SINBAD SINBAD ReactiorReactior

Shielding Shielding Benchmark Experiments, OECDBenchmark Experiments, OECD--NEA, 2011.NEA, 2011.

2.2.

I. I. RemecRemec

and F.B. and F.B. KamKam, , ““Pool Critical Assembly Pressure Vessel Facility BenchmarkPool Critical Assembly Pressure Vessel Facility Benchmark,,”” NUREG/CRNUREG/CR--6454 (ORNL/TM6454 (ORNL/TM--13205), August 1997.13205), August 1997.

3.3.

International Atomic Energy Agency, Vienna, Austria, InternationInternational Atomic Energy Agency, Vienna, Austria, International Reactor Dosimetry al Reactor Dosimetry File 2002 (IRDFFile 2002 (IRDF--2002), 2006.2002), 2006.

4.4.

U.S.NRC, U.S.NRC, Calculational and Dosimetry Methods for Determining Pressure VesCalculational and Dosimetry Methods for Determining Pressure Vessel sel Neutron FluenceNeutron Fluence, Regulatory Guide 1.190, 2001., Regulatory Guide 1.190, 2001.

5.5.

J.C. Wagner and A. J.C. Wagner and A. HaghighatHaghighat, , ““Automated variance reduction of Monte Carlo Automated variance reduction of Monte Carlo shielding calculations using the discrete ordinates adjoint funcshielding calculations using the discrete ordinates adjoint function,tion,””

Nuclear Science Nuclear Science and Engineeringand Engineering, vol. 128, no. 2, pp. 186, vol. 128, no. 2, pp. 186--208, 1998.208, 1998.

6.6.

G.I. Bell and S. G.I. Bell and S. GlasstoneGlasstone, , Nuclear Reactor TheoryNuclear Reactor Theory, Van , Van NostrandNostrand

Reinhold Company, Reinhold Company, Litton Educational Publishing, 1970.Litton Educational Publishing, 1970.

7.7.

M.B Chadwick, P. M.B Chadwick, P. ObloObložžinskyinsky, M. Herman et al., , M. Herman et al., ““ENDF/BENDF/B--VII.0: next generation VII.0: next generation evaluated nuclear data library for nuclear science and technologevaluated nuclear data library for nuclear science and technology,y,””

Nuclear Data Nuclear Data SheetsSheets, vol. 107, no. 12, pp. 2931, vol. 107, no. 12, pp. 2931--3060, 2006.3060, 2006.

8.8.

A. A. TrkovTrkov, M. Herman, and D.A. Brown, , M. Herman, and D.A. Brown, ““ENDFENDF--6 Formats Manual6 Formats Manual””, BNL, 2011., BNL, 2011.9.9.

T. T. GoorleyGoorley, "MCNP6.1.1, "MCNP6.1.1--Beta Release Notes", LABeta Release Notes", LA--URUR--1414--24680 (2014).24680 (2014).10.10.

S.W. Mosher, S.R. Johnson, A.M. S.W. Mosher, S.R. Johnson, A.M. BevilBevil

et al., et al., ““ADVANTG ADVANTG ––

An Automated Variance An Automated Variance Reduction Parameter GeneratorReduction Parameter Generator””, ORNL/TM, ORNL/TM--2013/416, Rev. 1, August 2015. 2013/416, Rev. 1, August 2015.

23

Documents

PCA benchmark analysis using the ADVANTG3.0.1/MCNP6 ...indico.ictp.it/event/7994/session/7/contribution/34/...PCA benchmark “12/13” configuration – PCA core with components mock