nucifer project: full simulation scheme · 2010. 8. 5. · reactor simulation neutrino spectra...

Reactor simulationNeutrino spectra simulation

Detector simulationExpected sensitivity and conclusions

Nucifer Project: Full simulation schemeFrom reactor to detector response

Jonathan Gaffiot on behalf of the Nucifer collaboration

CEA/DSM/Irfu: SPhN, SPP, SEDI, SIS, SENAC

CEA/DAM/DIF/DPTA/SPN

CNRS/IN2P3: Subatech

August 3rd, 2010

Nucifer

Jonathan Gaffiot on behalf of the Nucifer collaboration Nucifer Project: Full simulation scheme 1 / 27



From reactor to detector response

Goal: time prediction of detected neutrino spectra

⇒ cf. T. Lasserre and R. Granelli presentations




MCNP Utility for Reactor EvolutionA ’N4’ french PWR simulation: InputsA ’N4’ french PWR simulation: OutputsValidation and non proliferation studies

Outlook

1 Reactor simulationMCNP Utility for Reactor EvolutionA ’N4’ french PWR simulation: InputsA ’N4’ french PWR simulation: OutputsValidation and non proliferation studies

2 Neutrino spectra simulation

3 Detector simulation

4 Expected sensitivity and conclusions





MCNP Utility for Reactor Evolution

Principle

Monte Carlo: given static geometry and compositions, simulates neutron flux

Evolution code: given a static neutron flux, simulates composition evolution

MURE iterates these 2 simulations to get a depletion code

MURE: a recent open source library for reactor simulation [1]

C++ code coupled with MCNP for Monte Carlo simulation

Developed and supported by CNRS/IN2P3: IPNO, LPSC and Subatech

Available @ NEA data bank since 2009 (http://www.nea.fr/abs/html/nea-1845.html)

Adapted to non proliferation needs: C++ interface for inputs description,graphical interface based on ROOT for outputs analysis, coupling with fissionproducts β decay database, off equilibrium effect evaluation. . .





A ’N4’ french PWR simulation: Inputs

Simulation of Double Chooz reactors: french PWR type ’N4’, 4.27 GWth

Geometry, initial materials, nuclear databases, power history and time steps

317 pellet/rod and264 rods

205 assemblies

Core with differentenrichment zones





A ’N4’ french PWR simulation: Outputs

Fuel inventory, reaction rates, neutron flux, keff. . . at each time step





Validation and non proliferation studies

Validation

NEA benchmark:4 Westinghouse assemblies

NEA benchmark:Quarter core Westinghouse

(N. Capellan PhD thesis)

Independent NEA benchmarks andsentivity studies [1]

Comparison with deterministic codesAPOLLO2 and DRAGON

Non proliferation scenarii studies [2, 3]

Osiris Candu VHTR(V. M. Bui PhD thesis) (S. Cormon PhD thesis)




The difficulty: convert electron spectra to neutrino spectraThe microscopical approachBESTIOLE: a new code to simulate neutrino spectraBESTIOLE’s results

Outlook

1 Reactor simulation

2 Neutrino spectra simulationThe difficulty: convert electron spectra to neutrino spectraThe microscopical approachBESTIOLE: a new code to simulate neutrino spectraBESTIOLE’s results







The difficulty: convert electron spectra to neutrino spectra

β decay: measurable electron energy, need to deduce neutrino energy

Single β branch: direct relation between neutrino and electron energy

Spectrum of a nucleus: superposition of many β branches

Spectrum of a reactor: superposition of hundreds of fission products, i.e.thousands of β branches, and still unknown nuclei and branches

Kinetic energy (MeV)0 1 2 3 4 5

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

branchβsingle

branchνsingle

0E / 20E

single β branch

Kinetic energy (MeV)0 0.5 1 1.5 2 2.5 3

Inte

grat

ed y

ield

per

50

keV

bin

s

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

Mn56 spectrum of ν

Mn56 spectrum of β

nucleus: many branchesKinetic energy (MeV)

0 1 2 3 4 5 6 7 8 9

arbi

trar

y un

its

-410

-310

-210

-110

1

spectrumβmeasured

spectrumνconverted

235U: hundreds of nuclei





Historical approach for spectrum conversion

e−: accurate measurement of total spectra for each isotopes of interest

Fit of these spectra with some tens of effective branches

ν: converted virtual spectra from these effective branches

Reference for all neutrino experiment: ILL e− measurement @ 3%(1980s), andspectra conversion, see [4, 5, 6]

kinetic energy (MeV)β2 3 4 5 6 7 8 9

Inte

grat

ed y

ield

per

50

keV

bin

s

-210

-110

1 spectrumβMeasured

1 branch

2 branches

3 branches

4 branches





The microscopical approach

Today in nucleardatabases (JEFF,JENDL, ENDF)

1 More than 700 fissionproducts

2 More than 10000 βbranches

3 Effective models forunknown nuclei

No free parameter

Not precise enough innorm: 5-10% with e−

data Kinetic energy (MeV)1 2 3 4 5 6 7 8

part

icle

/ fis

sion

-310

-210

-110

1

spectrumβsimulated

spectrumβmeasured

spectrumνsimulated

spectrumνconverted

U235





BESTIOLE: a new code to simulate neutrino spectra

Mixing the two approaches

1 ∼ 90% of physicalbranches from nucleardatabases

2 Match residues witheffective branches

3 Reverse real and effectivebranches to obtainneutrino spectra

kinetic energy (MeV)β2 3 4 5 6 7 8

pred

ictio

n / d

ata

0

0.2

0.4

0.6

0.8

1

Built

Fitted

BESTIOLE: a new C++ code (Th. A. Mueller Ph.D. thesis)

Inputs: standard nuclear databases (ENSDF format) and fission rates

Simulate neutrino spectra for each fission product and each fissile isotopes

Intrinsic error propagation with calculation of covariance matrix





BESTIOLE’s results

Systematic 3% shift between historical and BESTIOLE spectra

Kinetic energy (MeV)2 3 4 5 6 7 8

( si

mul

atio

n -

data

) /

data

-0.02

0

0.02

0.04

0.06

0.08

0.1

electron residues

neutrino residues

U after fitting procedure235Residues for

+3% systematicnormalization shift

fit of electron data

To be submitted to Phys. Rev. C.




GEANT4 simulation of the detectorCalibration: simulation and detectorFighting correlated backgroundsThe results of the experiment

Outlook



3 Detector simulationGEANT4 simulation of the detectorCalibration: simulation and detectorFighting correlated backgroundsThe results of the experiment






GEANT4 simulation of the detector

High fidelity description [7]

Simulation of scintillation with dedicated model and fine liquid properties

Multiple measurements of optical properties of detector components

Intrinsic digitization of collected photons on simulated photomultipliers

Output: ROOT files with the same format than experimental data

Benefits from Double Chooz developments





Calibration: simulation and detector

Shallow depth laboratory + No shielding → Background rate of several kBq

(pe)totQ0 200 400 600 800 1000 1200

0

1000

2000

3000

4000

5000 Simulated 60Co

Measured 60Co

Co source60Simulated and measured

E (MeV)0 1 2 3 4 5

(p

e)to

tQ

0

200

400

600

800

1000

1200

1400

1600

1800

/ ndf = 1.036 / 32χProb 0.7925p0 0± 0 p1 7.143± 323.5

/ ndf = 1.036 / 32χProb 0.7925p0 0± 0 p1 7.143± 323.5

/ ndf = 0.7671 / 32χProb 0.8573p0 0± 0 p1 7.15± 330.5

/ ndf = 0.7671 / 32χProb 0.8573p0 0± 0 p1 7.15± 330.5

Cs137

Na22

Co60

4.4 MeVγAmBe: + recoil proton

5.5 MeV≈Simulation

Measurement

Simulated and measured calibration curves

1 24195Am → 4

2α +23793 Np

2 42α +9

4 Be → 136C

∗

3 136C

∗ → 126C

∗ + n (some MeV)4 12

6C∗ →12

6 C + γ (4.4 MeV)

(pe)totQ1000 1500 2000 2500 3000

0

1000

2000

3000

4000

5000

6000

7000

8000Simulated AmBe

Measured AmBe)DγH(n,2.2 MeV

4.4 MeVγ+ recoil proton

5.5 MeV≡

Simulated and measured AmBe source





Fighting correlated backgrounds

Pulse Shape Discrimination

Due to mass difference, α, p and e− have different ionising density

Light emission is then a bit quicker with e− than ion

Pulse Shape Discrimination can be used to discriminate particles

The experiment: introduction of 222Rn in Nucifer

Among 222Rn daughters, 214Bi decays β− on 214Po214Po is α emitter with half-life 164 µs

Time correlation + 2 different particles: Bi/Po decays mimic a ν signal





The results of the experiment

Comparison simulation/experiment

(pe)totQ0 200 400 600 800 10001

10

210

310

410

Rn decay chain with background222Simulation of

Bi)214(βBackground

Sum of all contributions

Po)214(α

Po)218(αRn) + 222(α

Bi)218(β

(pe)totQ200 400 600 800 1000 1200 1400

1

10

210

310

Rn decay chain222Measured

Bi)214(β

Po)214(α

Po)218(αRn) + 222(α

A clear indication of Pulse Shape Discrimination for α

(pe)totQ0 200 400 600 800 1000

tot

/Qta

ilQ

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4Rn chain222Simulated Pulse Shape Discrimination for

Po)218(αRn) + 222(α

Po)214(α

γ + β

(pe)totQ0 100 200 300 400 500 600 700 800 900 1000

to

t /

Qta

ilQ

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35 Background

Rn 222Background +

Rn experiment222Pulse Shape Discrimination for

180 300




Expected sensitivity and count rateConclusions and perspectives

Outlook




4 Expected sensitivity and conclusionsExpected sensitivity and count rateConclusions and perspectives





Expected sensitivity and count rate

At power plant: 25 m from reactor core of constant power 3.3 GWth

Detector: 0.8 m3 of liquid with 50% efficiency

A point with 1% statistical error each 3 days

Sensitivity to ∼ 55 kg of plutonium

l acrylic [cm]

# gamma [%]

Weeks0 10 20 30 40 50 60 70

Neu

trin

o /

Wee

k

15000

16000

17000

18000

19000

20000

Reactor: ON OFF ON

Refuelling 1/3of the core





Conclusions and perspectives

A full simulation scheme from reactor to detector

Inputs: Standard databases, geometrical properties and history of thermal power

Outputs: Direct comparison with experimental data

Many new developements: evolutionnary Monte-Carlo for reactor simulation,improvement of ν spectra, detector simulated from scintillation to digitization

Non proliferation scenario studies with Nucifer on power and research reactors

Perspectives

Comparison of Osiris ν spectra simulation with data

Fine tuning of GEANT4 simulation on final detector

Accurate non proliferation studies with final Nucifer performances




Back-up slides




References

O. Meplan et al., MCNP Utility for Reactor Evolution - Description of themethods, first applications and results, Proc. ENC, 2005

M. Fallot et al. “Nuclear reactor simulations for unveiling diversion scenarios:capabilities of the antineutrino probe,” Proc. GLOBAL, 2009

F. Yermia et al. “The Nucifer experiment: antineutrino detection for reactormonitoring,” Proc. GLOBAL, 2009

F. Von Feilitzsch, A. A. Hahn and K. Schreckenbach, “Experimental BetaSpectra From Pu-239 And U-235 Thermal Neutron Fission Products And TheirCorrelated Anti-Neutrinos Spectra,” Phys. Lett. B 118 (1982) 162.

K. Schreckenbach et al., “Determination Of The Anti-Neutrino Spectrum FromU-235 Thermal Neutron Fission Products Up To 9.5-Mev,” Phys. Lett. B 160(1985) 325.

A. A. Hahn et al., “Anti-Neutrino Spectra From Pu-241 And Pu-239 ThermalNeutron Fission Products,” Phys. Lett. B 218 (1989) 365.

A. Porta. “Reactor neutrino detection for non proliferation with the Nuciferexperiment,” Proc. TAUP, 2009, and J. Phys. Conf. Ser., 203, 2010




See also

MURE: http://lpsc.in2p3.fr/gpr/MURE/html/MURE/MURE.html

MURE @ NEA: http://www.nea.fr/abs/html/nea-1845.html

BESTIOLE: Th. A. Mueller Ph.D. thesis

GEANT4: http://geant4.cern.ch/

ROOT: http://root.cern.ch/drupal/

D. Lhuillier et al. The Nucifer experiment: reactor monitoring with antineutrinosfor non proliferation purpose. Proc. GLOBAL, 2009

A. Porta et al. Reactor neutrino detection for non proliferation with the NUCIFERexperiment. Proc. ANIMMA, 2009 IEEE 10.1109/ANIMMA.2009.5503653

L. Giot et al. Proc. PHYSOR, 2008

M. Fallot et al. Proc. Nuclear Data, B., 2007

B. Guillon et al. Proc. GLOBAL, 2007




Conversion procedure

1 Fit of the measured spectrum’s tail with an effective branch

2 Substraction of the branch

3 Iteration ∼ 30 times

Kinetic energy (MeV)2 3 4 5 6 7 8 9

a.u

.

-510

-410

-310

-210

-110

1

spectrumβmeasured

branchβ st1


a.u

.

-510

-410

-310

-210

-110

1

stepst spectrum after the 1β

branchβ nd2

Kinetic energy(MeV)2 3 4 5 6 7 8 9

a.u

.

-510

-410

-310

-210

-110

1

stepnd spectrum after the 2β

branchβ rd3

Individual conversion ofeach branch with energyconservation

Sum of all of them to getthe total spectrum


a.u

.

-510

-410

-310

-210

-110

1

spectrumβmeasured

spectrumνconverted




Main corrections brought by MURE

Out equilibriumspectra: long-livedfission products

Neutron capture onfission products

Shape of neutronflux (axial offset,pilot rods. . . )

kinetic energy (MeV)ν1.5 2 2.5 3 3.5 4 4.5 5 5.5 6R

elat

ive

diffe

renc

e to

12

hour

s irr

adia

tion

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

100 days irradiation



U235




Noise simulations

Noise measurement with HPGe as input to simulation


nucifer project: full simulation scheme · 2010. 8. 5. · reactor simulation neutrino spectra...

Documents