Element by R2S Method
Klemen Ambroˇziˇc
Joˇzef Stefan Institute Jamova cesta 39 1000 Ljubljana, Slovenia
klemen.ambrozic@ijs.si Luka Snoj
Joˇzef Stefan Institute Jamova cesta 39 1000 Ljubljana, Slovenia
luka.snoj@ijs.si ABSTRACT
The JSI TRIGA reactor has several sample irradiation facilities with well characterized neutron fields by both simulations and measurements data. Because of this, JSI TRIGA has become a ref- erence center for neutron irradiation of detector for ATLAS experiment (CERN). However, γ-ray characterization of irradiation facilities is yet to be performed. Monte Carloγ-ray characterizations up to date have accounted for promptγ-rays, neglecting delayedγ contribution. Measurements sug- gest that delayedγ-rays may contribute to up to30 %of totalγ flux, and is the only contribution of γ-rays after reactor shutdown. Initial steps in accounting for delayedγ-rays using R2S method have been performed and described in this paper.
A Rigorous two-step method (R2S) described here utilizes Monte Carlo particle transport code MCNP 6.1 for particle transport part of the problem, and FISPACT-II for neutron activation and delayedγ-ray generation, and custom Python scripts, joining the two codes.
An example on the use of code is presented, in terms of evaluation on utilization of activated nuclear fuel as a viableγ-ray source in sample irradiation. For this example, fresh nuclear fuel is considered and a silicon pipe sample modeled in the computational model. Two nuclear fuel activation time regimes were simulated, short (several hours) and long term (in terms of 1 MWdand10 MWd fuel burn-up).
1 INTRODUCTION
The Joˇzef Stefan Institute (IJS) TRIGA reactor is a250 kWpool type reactor featuring numerous irradiation facilities with different characteristics in terms of size and neutron field properties [1].
Due to good neutron characterization, both computationally [1] and with measurements [2], the reac- tor has become a reference facility for neutron irradiation of ATLAS experiment (CERN) detectors [3].
Currently γ characterization of the reactor is also being performed. Prompt γ characterization is being performed using the Monte Carlo MCNP 6.1 code [4] with ENDF VII.0 [5] nuclear data li- braries [6]. However these results are valid only for short irradiations on moderate to high power, as the measurements with ionization chamber indicate, that delayedγ-rays due to neutron activation contribute up to30 %of totalγ-ray flux in steady state operation [7, 8, 9], and is the only source of γ rays after reactor shutdown. Thoroughγ-ray characterization in terms of computational analysis and measurements in an operational reactor and after reactor shutdown, for both prompt and delayed
γ-rays is planned in the near future. Initial steps in modeling the delayed γ-ray contributions from activated nuclear fuel, due to the fact that it can be a viableγ-ray source and is a main contributor of delayedγ-rays, have been taken, and methods and results described in this paper.
A rigorous two-step method (R2S) using Monte Carlo particle transport code MCNP 6.1 [4] for par- ticle transport part of the problem, and FISPACT-II [10] for neutron activation and delayed γ-ray generation part with custom in-house developed Python scripts, combining the two codes.
2 RIGOROUS TWO-STEP METHOD (R2S)
An R2S method [11] in reactor physics is a method of coupling particle transport capabilities with neutron activation analysis and delayed γ-ray generation. The basic idea is to divide the problem geometry into small voxels and calculate neutron spectra and total neutron flux values in each of them, or even subdivide the voxel to conform with geometry (cell-under-voxel approach) [12] by the means of a Monte Carlo neutron transport. Voxel neutron flux and spectrum, along with voxel material composition and mass is used in neutron activation analysis calculation, generating time dependent isotope inventory and delayedγray spectra and intensities respectively. The latter results are used in a Monte Carlo photon transport as time dependentγ sources of delayedγ in a respective voxel for delayed photon field, spectrum, kerma and dose particle transport calculations.
3 COMPUTATIONAL SETUP
Concrete
Graphite
Alu. housing
366 cm
370 cm
655 cm 274 cm
~490 cm
198 cm 91.4 cm
264 cm Heavy
concrete Core
Lead
Boral Polyethylene Heavy
concrete door Door plug
244 cm
274 cm Thermalizing column Tangential irr.
port Radial piercing irr.
port
Thermal column door Plug
Lead
Radial beam irr. port
35 ° 33.5 °
Polyethylene
Boral
Steel shield Reactor tank 106 cm
Thermal irr.port
TOP SIDE
Figure 1: Schematic of the JSI TRIGA Mark II reactor, side (left) and top (right) view.
JSI TRIGA is a pool type reactor, with maximum steady state power of 250 kW (Fig. 1). The fuel elements are arranged in an annular configuration into rings A-F (Fig. 2). A detailed JSI TRIGA reactor MCNP model, based on the criticality benchmark model [13], which has been thoroughly validated by multiple experiments [14, 15, 16], was used for total neutron flux and spectra distribution inside the fissionable material part of the fuel element (U-ZrH fuel meat) of several standard JSI TRIGA fuel elements (12 wt%uranium, enrichment 20 %). Neutron activation for fresh fuel meat (U-ZrH) only was calculated, as it is the main contributor of delayedγ-rays.
Kerma in sample was calculated using voxel volume averaged energy deposition estimators, and the
compositions, as well as major isotopes, contributing to the dose on the sample were also calculated.
Each of the core ring fuel element representatives fuel meat was divided in10axial ×5radial ×
A1 B1 B2 B3
B4 B5 B6
C2 C1 C3 C4
C5 C6 C7
C8 C9
C10 C11 C12 D2 D1 D3 D4 D5 D6 D7
D9 D10 D11 D12
D13 D14 D15 D16 D17 D18 E2 E1 E3 E4 E5 E6 E7
E8 E9
E12 E13 E14 E15
E16 E17
E18 E19 E20 E21 E22 E23 E24 F2 F1
F3 F4 F5 F6 F7 F8 F9 F10
F11 F12
F13
F14F15 F16 F17 F18
F19 F20
F21 F22
F23 F24 F25 F26 F27 F28 F30F29 TriC
20% LEU fuel element
Empty/irradiation in-core position Representative fuel element
Neutron source
Compensating control rod Shim control rod
Regulating control rod Pulse control rod
Figure 2: JSI TRIGA core with typical fuel element arrangement and denoted represen- tative fuel elements from each core ring.
72.5 cm
Upper pin
Graphite reflector
Zr rod Fuel meat
38.1 cm
Lower pin Lower grate 3.76 cm Molybdenum disc
Upper grate
Fuel element Fuel meat segmentation along element
length
Fuel meat segment division
over radius and angle
3.81 cm
Figure 3: JSI TRIGA fuel element and fuel meat divi- sion for R2S calculation.
5azimuthal voxels, and a 100-group neutron spectra and total flux values for steady state operation were calculated in each of them. An irradiation plan in terms of reactor power and duration for the whole core, along with voxel calculated neutron spectra, fluxes and material compositions was used to construct neutron activation code FISPACT-II input files for each voxel respectively. After the calculations for all the voxels were performed, a Python script was used, to merge the output calculations data accordingly and construct MCNPγ sources for each stage of neutron irradiation or shutdown. A flow-chart of the procedure is depicted in Fig. 4. Two irradiation time regimes were simulated: short-term irradiation, where reactor operated for a few hours, and long-term irradiation, where a typical 40 h/woperation at full reactor power was simulated until a burn-up of 1 MWdor 10 MWdwas achieved.
As the gamma irradiation in the JSI TRIGA reactor is performed mostly to study radiation hardness of Si based semiconductors, the main motivation of this work is to use the R2S method to calculate kerma and dose on Si. For this purpose a cylindrical pipe silicon sample was modeled around the activated fuel element, with kerma and dose distributions calculated in it (Fig. 5).
Neutron transport
Voxel data division
Neutron spectrum energy group conversion, tot. flux
Time dependent activity, isotopic composition and γ spectrum.
Input file for neutron energy group conversion Batch procedure for multiple input evaluation ź Tot. neutron flux
ź Neutron group flux in predetermined group structure
Preparation of inputs files for activation eval.
Batch procedure for serial input calculations
Activities, isotopic compositions, γ spectra isotopic contributions to activity and dose, uncertainties
Construction of bulk γ-ray source from activation data
γ-ray
transport Doserate range
evaluaiton Voxel data merging, activities,
uncertainties, isotopic contributions to dose Merging time
dependent results in order
Time dependent activities, dose-rates, uncertainties, isotopic contributions to dose Neutron
transport input file
Neutron flux and spectrum in voxels
Doserate sample value and error
START
END
KEY: Text files Monte Carlo calc.
Use of script FISPACT-II calc.
Tot.
flux
Group neutron flux
Neutron flux and spectra preparation
Activation time coursr
Doserate calculations, merging voxel into bulk source
Corect temporal sequence, result preparation
Figure 4: Flow chart of the R2S calculation.
600 118
40 Silicon sample
TRIGA fuel
Figure 5: Irradiation of the silicon sample with irradiated fuel element.
4 RESULTS
In this section, an overview on the results for both short-term and long-term irradiations using the above mentioned R2S approach on will be presented. Activities of fuel elements, dose and kerma rates on the silicon sample are given in this section. Fuel elements inside the JSI TRIGA reactor are arranged in concentric rings. A sigle element from each ring was selected for calculations (Fig. 2).
4.1 Short activation
A short fuel activation of several hours is considered in this case. This corresponds to day to day operations, where reactor is operational for several hours at full reactor power. Fresh nuclear fuel prior to irradiation is considered. Kerma-rate and dose-rate results are given in terms of kerma and dose-rate range on the sample, due to their distribution in the silicon sample, with maximal values being in the most inner part at center hight of the sample, and lowest values at the sample outer edges.
All of the results have a computational uncertainty below1 %.
106 107 108 109 1010 1011 1012 1013 1014
101 102 103 104 105 106 107 108 109 1010
A [Bq]
t [s]
B2 1h B2 2h B2 4h
B2 8h B2 10h B2 20h
(a) B2 fuel element activity.
106 107 108 109 1010 1011 1012 1013 1014
101 102 103 104 105 106 107 108 109 1010
A [Bq]
t [s]
E5 1h E5 2h E5 4h
E5 8h E5 10h E5 20h
(b) E5 fuel element activity.
10-6 10-5 10-4 10-3 10-2 10-1 100 101 102 103
101 102 103 104 105 106 107 108 109 1010
Dose-rate H*10 [Sv/h]
t [s]
B2 1h B2 2h B2 4h
B2 8h B2 10h B2 20h
(c) B2 fuel element dose-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 1010
Dose-rate H*10 [Sv/h]
t [s]
E5 1h E5 2h E5 4h
E5 8h E5 10h E5 20h
(d) E5 fuel element dose-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 1010
Kerma-rate [Gy/h]
t [s]
B2 1h B2 2h B2 4h
B2 8h B2 10h B2 20h
(e) B2 fuel element kerma-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 1010
Kerma-rate [Gy/h]
t [s]
E5 1h E5 2h E5 4h
E5 8h E5 10h E5 20h
(f) E5 fuel element kerma-rate to sample.
Figure 6: Two representative fuel elements (B2 and E5), with different activation times. X-axis denotes time after the irradiation.
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14
10-2 10-1 100 101
P(E)
E [MeV]
B2 1 h B2 20 h
(a) B2 fuel elementγ-ray spectra after1 hand20 h of operation at full reactor power.
10-4 10-3 10-2 10-1
101 102 103 104 105
Rel. contribution to contact dose [%]
tcol [s]
134I 138Cs 93Sr 142La 91Rb 89Rb 144La 90Rb 135I 92Sr 101Mo 89Kr 140Cs 95Y
(b) Isotope contributions to the contact dose in B2 element, after20 h of operation at full reactor power.
Figure 7
Spectra ofγ rays for each voxel element are calculated during the neutron activation calculation procedure, to enable us to produce sources of delayedγ-rays respectively. An integral spectrum can also be calculated (Fig. 7a). In addition, isotope contribution to contact dose is also calculated, which gives insight on radiation hazards due to neutron activation of components in a nuclear facility (Fig.
7b).
4.2 Long activation
For long term activations, a40 h/woperation at full power is considered, until a burnup of1 MWd or10 MWd is achieved, starting with fresh fuel. Fuel element activity, kerma-rate and dose-rate to the silicon sample after reactor shutdown are calculated.
γ-ray spectra emitted by activated voxels have again been summed in terms of fuel element γ-ray spectra (Fig. 9a) as a whole, and isotopic contribution to contact dose have been calculated (Fig. 9b).
107 108 109 1010 1011 1012 1013 1014
101 102 103 104 105 106 107 108 109 10101011
A[Bg]
tcol[s]
B2, 1 MWd C3, 1 MWd D4, 1 MWd E5, 1 MWd
(a)1 MWdfuel elements activity.
108 109 1010 1011 1012 1013 1014
101 102 103 104 105 106 107 108 109 10101011
A[Bg]
tcol[s]
B2, 10 MWd C3, 10 MWd D4, 10 MWd E5, 10 MWd
(b)10 MWdfuel elements activity.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 10101011
Dose-rate H*10 [Sv/h]
tcol[s]
B2, 1 MWd C3, 1 MWd D4, 1 MWd E5, 1 MWd
(c)1 MWdfuel elements dose-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 10101011
Dose-rate H*10 [Sv/h]
tcol[s]
B2, 10 MWd C3, 10 MWd D4, 10 MWd E5, 10 MWd
(d)10 MWdfuel elements dose-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 10101011
Kerma rate[Gy/h]
tcol [s]
B2, 1 MWd C3, 1 MWd D4, 1 MWd E5, 1 MWd
(e)1 MWdfuel elements kerma-rate to sample.
10-7 10-6 10-5 10-4 10-3 10-2 10-1 100 101 102
101 102 103 104 105 106 107 108 109 10101011
Kerma rate[Gy/h]
tcol [s]
B2, 10 MWd C3, 10 MWd D4, 10 MWd E5, 10 MWd
(f)10 MWdfuel elements kerma-rate to sample.
Figure 8: Results for representative fuel elements with burn-up of1 MWdand10 MWd.
0 0.05 0.1 0.15 0.2 0.25
10-3 10-2 10-1 100
P(E)
E [MeV]
1 MWd 10 MWd
(a) Emitted γ-ray spectra from B2 element after 1 MWdand10 MWdburn-up.
10-7 10-6 10-5 10-4 10-3 10-2 10-1
102 103 104 105 106 107 108
Rel. contribution to contact dose [%]
tcol
La140 Np239I135 Nb97I133 Nb97mI133 Nb95Zr95 Ce143Sr91 Xe135
(b) Isotopic contribution to contact dose on B2 el- ement after10 MWdburn-up.
Figure 9 4.3 Results comment
γ-decaying isotopes in the nuclear fuel are mainly produced in nuclear fission, and their dis- tribution is proportional to fission rates and therefore to power distribution. The differences in the γ-activities of fuel elements irradiated for same periods of time are therefore attributed to power dis- tribution which has been calculated per fuel element [15]. The ranges of kerma-rates and dose-rates can be attributed to the axial neutron flux distributions, which has also been evaluated both compu- tationally and by measurements [2].
For longer irradiation times, or until a desired burn-up is achieved, we can clearly observe the effect of increased neutron flux distribution in inner elements. Theγ-decaying isotope production rate is again proportional to neutron flux, and with shorter irradiation time, fewer of the so produced isotopes have decayed up to the irradiation end. This affects the isotopes, which reach saturation in several hours, which can are visible in the Fig. 8 with differences between fuel elements for same burn-up.
With the increased burn-up, long-lived isotopes are produced linearly with the increased irradiation time and burn-up, therefore the10×increase in activities and sample dose-rates and kerma-rates at 10 MWdburn-up, compared to the1 MWd.
5 SUMMARY
The R2S method presented here aims to provide preliminary analysis on contribution of delayed γ-rays to photon kerma and dose-rates in the reactor irradiation facilities. Initial measurement of delayedγ-ray dose in the vicinity of several fuel elements show agreement with the calculations.
The codes and scripts developed for this problem specific calculation procedure give valuable insight on developing a generalized method for R2S calculations, with possible ”geometry under voxel” ap- proach implemented [12], while current method performs division, conforming to model geometry.
The model segmentation, neutron flux and spectrum calculations, as well as neutron activation and delayed γ-ray source generation would be performed automatically. The code would be used for delayedγ-field characterization of the JSI TRIGA reactor, which would increase utilization forγ-ray radiation hardness studies.
R2S methods are applied almost exclusively to fusion devices, where little to no utilization on fis- sion reactors. In the following months, the method will be applied to the whole JSI TRIGA reactor core, along with the development of the above mentioned generalized R2S code and with γ field measurements in the JSI TRIGA reactor during reactor operation and after reactor shutdown. The
measurements will serve to establish measurement procedures ofγmeasurements in mixedγ-neutron fields and to validate the computational results.
The current and a generalized method will be used in the future to evaluate component activation in- side nuclear power plants, as well as for accurate classification of nuclear waste after nuclear facility’s decommissioning, therefore increasing long-term storage facility utilization.
REFERENCES
[1] Luka Snoj, Gaˇsper ˇZerovnik, and Andrej Trkov. Computational analysis of irradiation facilities at the JSI TRIGA reactor. Applied Radiation and Isotopes, 70(3):483 – 488, 2012.
[2] Gaˇsper ˇZerovnik, Tanja Kaiba, Vladimir Radulovic, Anˇze Jazbec, Sebastjan Rupnik, Loc Bar- bot, Damien Fourmentel, and Luka Snoj. Validation of the neutron and gamma fields in the JSI TRIGA reactor using in-core fission and ionization chambers. Applied Radiation and Isotopes, 96:27 – 35, 2015.
[3] Luka Snoj and Borut Smodiˇs. 45 years of TRIGA Mark II in Slovenia. InProceedings of the International Conference Nuclear Energy for New Europe, September 2011.
[4] T Goorley, M James, T Booth, F Brown, J Bull, LJ Cox, J Durkee, J Elson, M Fensin, RA Forster, et al. Initial MCNP6 release overview. Nuclear Technology, 180(3):298–315, 2012.
[5] M.B. Chadwick, P. Obloinsk, M. Herman, et al. ENDF/B-VII.0: Next Generation Evaluated Nuclear Data Library for Nuclear Science and Technology.Nuclear Data Sheets, 107(12):2931 – 3060, 2006. Evaluated Nuclear Data File ENDF/B-VII.0.
[6] Klemen Ambroˇziˇc, Luka Snoj, and Gaˇsper ˇZerovnik. Computational analysis of the dose rates at JSI TRIGA reactor irradiation facilities. To be published this year, 8 2016.
[7] Vladimir Radulovi´c, Damien Fourmentel, Lo¨ıc Barbot, Jean-Franois Villard, Tanja Kaiba, Zerovnik Gaˇsper, and Luka Snoj. Measurements of miniature ionization chamber currents inˇ the JSI TRIGA mark II reactor demonstrate the importance of the delayed contribution to the photon field in nuclear reactors. Nuclear Instruments and Methods in Physics Research Section A:Accelerators, Spectrometers, Detectors and Associated Equipment, 804:149 – 154, 2015.
[8] Damien Fourmentel, Vladimir Radulovi´c, Lo¨ıc Barbot, Jean Francis Villard, Gaˇsper ˇZerovnik, Luka Snoj, Mikolaj Tarchalski, Krysztof Pytel, and Fadhel Malouch. Delayed gamma mea- surements in different nuclear research reactors bringing out the importance of the delayed contribution in gamma flux calculations. In Advancements in Nuclear Instrumentation Mea- surement Methods and their Applications, ANIMMA, Lisbon, April 2015. Instituto de Plasmas e Fus˜ao Nuclear.
[9] Vladimir Radulovi, Damien Fourmentel, Loc Barbot, Jean-Franois Villard, Tanja Kaiba, erovnik Gaper, and Luka Snoj. Measurements of miniature ionization chamber currents in the{JSI} {TRIGA}mark{II}reactor demonstrate the importance of the delayed contribution to the photon field in nuclear reactors. Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 804:149 – 154, 2015.
[10] Jean-Christophe Sublet C, James W. Eastwood, J. Guy Morgan, Michael Fleming, and Mark R.
Gilbert. The FISPACT-II User Manual. UK Atomic Energy Authority, Culham Science Centre, Abingdon, Oxfordshire, OX14 3DBr, issue 7 edition, 9 2015.
[11] A. Davis. Radiation shielding of fusion systems. PhD thesis, University of Birmingham, April 2010.
[12] P. Sauvan, J. P. Cataln, F. Ogando, R. Jurez, and J. Sanz. Development of the R2SUNED Code System for Shutdown Dose Rate Calculations. IEEE Transactions on Nuclear Science, 63(1):375–384, Feb 2016.
[13] R. Jeraj, M. Ravnik, and Mark II TRIGA. Reactor U (20)–Zirconium Hydride Fuel Rods In Wa- ter With Graphite Reflector, International Handbook of Evaluated Criticality Safety Benchmark experiments. Technical report, NEA/NSC/DOC/(95) 03/III.
[14] M. Ravnik and R. Jeraj. Research reactor benchmarks. Nuclear Science and Engineering, 145(1):145–152, 2003.
[15] L. Snoj and M. Ravnik. Power peakings in mixed TRIGA cores. Nuclear Engineering and Design, 238(9):2473–2479, 2008.
[16] L. Snoj, A. Kavˇciˇc, G. ˇZerovnik, and M. Ravnik. Calculation of kinetic parameters for mixed TRIGA cores with Monte Carlo. Annals of Nuclear Energy, 37(2):223–229, 2010.
[17] John Hale. Data for protection against ionizing radiation from external sources: Supplement to ICRP publication 15data for protection against ionizing radiation from external sources: Sup- plement to ICRP publication 15. ICRP publication 21. by the international commission on ra- diological protection. paper, $7.50. pp.100, with numerous figures and tables elmsford, n.y., pergamon press, 1973. Radiology, 111(3):716–716, June 1974.
