Neutronics Calculation of RTP Core
Mohamad Hairie B. Rabir, Muhammad Rawi B. Mohamed Zin, Julia Bt. Abdul Karim, Abi Muttaqin B. Jalal Bayar, Mark Dennis Anak Usang, Muhammad KhairulAriff B. Mustafa, Na’imSyauqi B. Hamzah, Norfarizan Bt. Mohd Said, Muhammad Husamuddin B. Abdul jalil 1
Nuclear & Reactor Physics Section, Reactor Technology Centre, Technical Support Division, Malaysian Nuclear Agency, Bangi, 43000 Kajang, Selangor, Malaysia.
[email protected],+603-89112000 ext: 2327 / 6133
ABSTRACT Reactor calculation and simulation are significantly important to ensure safety and better utilization of a research reactor. The Malaysian’s PUSPATI TRIGA Reactor (RTP) achieved initial criticality on June 28, 1982. The reactor is designed to effectively implement the various fields of basic nuclear research, manpower training, and production of radioisotopes. Since early 90s, neutronics modelling were used as part of its routine in-core fuel management activities. The are several computer codes have been used in RTP since then, based on 1D neutron diffusion, 2D neutron diffusion and 3D Monte Carlo neutron transport method. This paper describes current progress and overview on neutronics modelling development in RTP. Several important parameters were analysed such as keff, reactivity, neutron flux, power distribution and fission product build-up for the latest core configuration. The developed core neutronics model was validated by means of comparison with experimental and measurement data. Along with the RTP core model, the calculation procedure also developed to establish better prediction capability of RTP’s behaviour. Keywords: TRIGLAV, MCNPX, RTP, Neutronics calculation Area of research: Nuclear Engineering, Reactor Physics, Neutronics Analysis
INTRODUCTION The Malaysian 1MW PUSPATI TRIGA Reactor (RTP) was designed to effectively implement the various fields of basic nuclear research and education. It incorporates facilities for advanced neutron and gamma radiation studies as well as for isotope production, sample activation, and student training. RTP has reached its first criticality on 28 Jun 1982 with excess reactivity of $ 0.15. It uses standard TRIGA UZrH1.6 fuel of 8.5, 12 and 20 wt. % of U with 19.9 % of U-235 enrichment. It has an annular core surrounded by graphite reflector and cooled by natural convection. The top and bottom grid plate is made of Al-6061. RTP has 4 control rods which are made up of boron carbide, see Table 1. Three of them are from fuel follower type and the other is air follower. The fuel follower control rods (FFCR) made up of 8.5 wt. % UZrH1.6 and B4C absorber on top of the fuel section. The air follower control rod (AFCR) only contain B4C absorber and an air section. RTP used mainly for beam experiments, samples analyses, education and trainings. The reactor utilizes hydride fuel which is a homogeneous mixture of uranium and zirconium hydride (UZrH). The ZrH is used as main moderator. The crosssectional view of Core-15, the latest core configuration is shown in Figure 1. Elements are arranged in seven circular rings and the spaces between the fuel rods are filled with water that acts as coolant and moderator.
Table 1, The details specification of TRIGA fuel elements and control rods Fuel Element
Fuel Follower Control Rod
0.3175 1.765 0.05 0.05
0.3175 1.665 0.05 0.05
Geometrical data Outer Radius of Zr rod (cm) Outer Radius of fuel (cm) Air gap thickness (cm) Cladding thickness(cm) Fuel composition Uranium (wt. %) Enrichment (wt. %) H:Zr ratio Absorber
8.5
12 19.9 1.6
20
8.5 19.7 1.6 B4C
Figure 1. RTP Core-15 configuration with 7 annular rings start with the central thimble (CT) or Aring, B-ring, C-ring, D-ring, E-ring, F-ring and the outermost G-ring. RTP core neutronics analysis have been done using deterministic and Monte Carlo code. Monte Carlo methods have gained interest due to the ability to more accurately model complex 3-dimesional geometries. TRIGLAV is a deterministic code, based on diffusion approximation of transport equation, which uses WIMSD program to calculate a unit cell averaged cross section data. TRIGLAV program package is developed for reactor calculations of mixed cores in TRIGA Mark II research reactor. It can be applied for fuel element burnup calculations, for power and flux distributions calculations and for critically predictions. In previous work, in reference 2 and 3, RTP MCNP model with fresh fuel was developed and validated against experimental data. However, a core model with burned fuel is needed in order to obtain accurate analysis for current and future core parameters. The main aim of this work is to develop MCNP model of the PUSPATI TRIGA Reactor considering the fuel burnup. The calculation procedure also developed to establish better prediction capability of RTP’s behaviour.
METHOD Detailed analysis method can be found in reference 1, 2 and 3. However, the calculation procedure have never been described before for RTP analysis. Figure 2 shows the calculation procedure developed and applied in developing the new core MCNP model where fuel burnup is taken into account.
Figure 2, Calculation procedure which involves the use of TRIGLAV and MCNP code along with validation via measurement or reactor operational data. Burnup measurements are for future work As shown in the calculation procedure, determination of fuel burnup will be based on TRIGLAV calculation. Figure 3 shows the results of the core burnup where the excess reactivity decreased at the end of cycle (EOC) and increased every time the core reshuffled at the beginning of cycle (BOC) of a new core configuration. Based on the core burnup, the burnup value for each individual fuel rod can be determined. TRIGLAV output produced individual fuel burnup in MWD as shown in figure 4. Figure 4 is actual Core-15 configuration with fuel burnup value at BOC. Actinides and fission product buildup calculation will be done using MCNPX. Each fuel with specific burnup value will be matched to its actinides and fission products inventory, and used in material input deck of the MCNP core model. Figure 5 shows the results for actinides build-up against fuel burnup. 8.00 Measured excess reactivity at zero power
6.00 5.00 4.00
TRIGLAV excess reactivity at zero power
3.00 2.00 1.00
TRIGLAV excess reactivity at full power
400
360
320
280
240
200
160
120
80
40
0.00
0
Excess reactivity $
7.00
Acc. Burnup MWD
Figure 3, Excess reactivity at BOC and EOC of each new core
Figure 4, Fuel burnup (MWD) at BOC for Core-15
mass (g)
1.00E+03 1.00E+02
U-238
1.00E+01
U-235
1.00E+00
U-236 Pu-239
1.00E-01
Pu-240
1.00E-02
Np-239
1.00E-03
Np237
1.00E-04
Pu241
1.00E-05
U-237
1.00E-06 0
2
4
6 Fuel burnup (MWD)
8
10
Figure 5, Major actinides in RTP fuel calculated using MCNPX
12
RESULTS Figure 6 shows the MCNP model of Core-15. The first validation for the developed burned core model is the criticality or the excess reactivity. Table 2 shows the results for excess reactivity calculation and comparison with measured value at BOC of Core-15. This results shows the ratio of calculated over experimental value (C/E) and it is 1.004, a relatively small discrepancy.
Figure 6, Core-15 MCNP model Table 2, Calculation and measured value for excess reactivity ($) of Core-15 at BOL
$
MCNPX 5.20
Measured 5.18
C/E 1.004
Criticality validation as shown above however, only represent the global behaviour of the core, not the local behaviour which represent the heterogeneity of Core-15. Validating this local behaviour will be based on the thermal neutron flux measurement inside the core. Therefore, an in-core thermal neutron flux measurement was performed using self-powered neutron detector (SPND) in several location in the core. The SPND size is relatively small and can be fitted in any flux holes in the core. The measurement was done online with Vanadium based SPND. Detail measurement work of the SPND however will not be discussed here and can be found in reference 1. Figure 7 shows the SPND setup during the measurement. Figure 8 shows the comparison between calculated and measured thermal neutron flux. This results shows the discrepancy between both fluxes values are within 10%. The measurement was done at 500kW, thus the MCNP results are also scaled to this power level. The description for MCNP tally scaling method used in this work can be found in reference 2.
Figure 7, Core-15 in-core online thermal neutron flux measurement setup using SPND
Figure 8, Core-15 in-core online thermal neutron flux measurement at 500kW comparison with MCNP calculation. The C/E value correspond to each data point in the graph.
The control rod worth of Core-15 was also used as validation data for the MCNP model. Two control rod curve data used for comparison is TRANSIENT rod, an air follower control rod type and REGULATING rod, a fuel follower control rods. The discrepancy between the measurement and calculation results are relatively big (maximum different around 30%). This may be due to the lack of understanding of how the performance of the control rods element especially boron carbide after relatively long irradiation time in the core.
Figure 9, Integral and differential curve of control rods comparisons The difference between the fresh core and burned core model were also tested via power peaking factor calculation. As shown in figure 10, it is clearly will lead to wrong determination of power distribution if burned fuel model is not used. The power peaking calculated using fresh fuel model seems to be lower than the burned fuel model especially in B-ring and E-ring. TRIGLAV prediction is conservatively higher than MCNPX simulation.
Figure 10, Power peaking factor comparisons for fresh and burned core model
CONCLUSION The results show this work has successfully developed neutronics calculation model for RTP with burned fuel. Calculation procedure was also established to ensure the quality of the results and better neutronics analysis to support in-core fuel management strategy. Good agreement (relatively) between calculation and measurement: MCNP model over estimation within 10% for flux and keff. The control rod element model lack in material changes behaviour after long time irradiation in the core. Future work should be done on the matter and validation via burnup measurement.
REFERENCE 1. Mohamad Hairie Rabir, Julia Abdul Karim, Norfarizan Mohd Said, Muhammad KhairulAriff Mustafa, Abi Muttaqin Jalal Bayar, Muhd Husamuddin Abdul Khalil, Dr Muhammad Rawi Md Zin, In-Core Neutron Flux Determination, Nuclear Bulletin of Malaysia 2016 Vol. 18, ISSN:13945610,Jun 2016, Pages 26-28; http://www.nuklearmalaysia.org/nuklearmalaysia_org/media/File/BNM/ebnm2016vol18/ebnm201 6vol18.html 2. Mohamad HairieRabir, Muhammad Rawi Md Zin, Mark Dennis Usang, Abi Muttaqin Jalal Bayar, Na'im Syauqi Bin Hamzah, Neutron Flux and Power in RTP Core-15, AIP Conference Proceedings 1704, 050018 (2016); http://dx.doi.org/10.1063/1.4940114 3. Mohamad Hairie Rabir, Mark Dennis Usang, Na’im Syauqi Hamzah, Julia Abdul Karim, Mohd Amin Sharifuldin Salleh, DETERMINATION OF NEUTRONIC PARAMETERS FOR RTP USING MCNP CODE,JurnalSainsNuklear Malaysia, 2013, Vol. 25(1):53-67I SSN: 2232-0946
