APPLICATION OF KARATE TO SUPERCRITICAL WATER REACTORS

Gy. Hegyi, Cs. Maráczy, Á. Brolly, P. Vértes, G. Hordósy

KFKI Atomic Energy Research Institute, H-1525 Budapest 114, POB 49, Hungary
[email protected]

ABSTRACT

The overall objective of the High Performance Light Water Reactor (HPLWR) project
was to assess the feasibility of a high efficiency light water reactor operating at
thermodynamically supercritical region. In a once-through concept, the water enters the
reactor as water and exits as high pressure steam without change of phase. With low
investment costs an efficiency of about 44% is expected. The evaluation and improvement of
the Japanese concept [1] was carried out by eight institutions within the 5th Framework
Programme of the EC [2,3]. The present paper summarizes the activity of KFKI Atomic
Energy Research Institute in the field of neutronics and outlines the future in the 6th FP.

INTRODUCTION

The HPLWR belongs to the six reactor types currently being investigated within the
framework of the Generation IV International Forum. These reactors are expected to exceed
the current nuclear reactors in reliability, safety, electricity generation costs and proliferation
resistance. The most visible advantages of the HPLWR are the low construction costs because
of size reduction of components and buildings compared to current LWR and the low
electricity production costs due to high efficiency.

RESEARCH RESULTS OF KFKI-AEKI IN THE FIELD OF CORE DESIGN

The evaluation and improvement of the Japanese concept concentrated on the

following areas: basic design requirements, plant architecture, core design, neutronics and
thermal-hydraulics analyses, materials and corrosion, safety, economics. The work was
carried out by eight institutions within the 5th Framework Programme of the EC.

The role of our institute was to perform neutronic transport and core diffusion
calculations. The water density at the top of the HPLWR assemblies is very low, so to reach
sufficient moderation in the tight-lattice hexagonal assemblies the application of water rods
extending to seven hexagons is necessary ( Figure 1). Preliminary calculations have shown
that the maximum keff deviation of our deterministic transport code from Monte Carlo results
does not exceed 0.9% even in case of 0.096 g/cm3 water density.
 Figure 1
30o sector of the reference HPLWR assembly (5th FP).

As the coolant density along the axial direction shows remarkable change, coupled neutronic-
thermohydraulic calculations are essential which take into account the heating of moderator in
the special water rods of the hexagonal assemblies. We have prepared a parametrized
diffusion cross section library of the HPLWR assembly at low burnup. The cross sections
depend on the following parameters: burnup, coolant density, water rod density, fuel
temperature, 135Xe, 149Sm, 235U, 238U and 239Pu concentration. The parameter range covers the
cold zero power and hot full power states. The parametrized cross section library has been
built into the GLOBUS nodal code of the KARATE [4] program system. The feedback
parameters were calculated with the SPROD code of the Tokyo University, which was
coupled with the KARATE core calculation module. Preliminary core calculations have been
done to demonstrate the applicability of the code. The axial density distribution of the coolant,
the water rod and the water insulator between them can be seen in Fig. 2.
800

700

600
Density [kg/m3]

500

400 C oolant
Insulator
300 W ater rod

200

100

0
5.38E -02 5.92E -01 1.13E + 00 1.67E +00 2.21E +00 2.75E +00 3.28E + 00 3.82E +00

H eig h t [m ]

Figure 2
Axial water density distributions of the reference HPLWR assembly (5th FP)
under nominal conditions.
 During the HPLWR core calculations the two energy group structure was used in the
coupled KARATE-SPROD code system. Because of the coolant axial density change the
need for checking the applicability of the two-group structure emerged. Cross section
calculations were performed with the MULTICELL deterministic transport code for the
reference fuel assembly without Gd absorbers at zero burnup. The technological parameter
range (coolant density, water rod density, 135Xe concentration, fuel temperature) corresponded
to the nominal conditions. The multigroup cross sections were collapsed to 2, 4 and 6 group
diffusion type cross sections using the criticality spectrum (B1 equations) for each parameter
combination. The cross sections were used to calculate one representative assembly. The
temperature and equilibrium Xe distributions were calculated with the KARATE-SPROD
nodal code using 2 energy groups. The SNAP finite difference (FD) code was used to study
the effect of the 2, 4 and 6 group scheme, where the cross sections were based on the frozen
distributions. The results showed that the maximum deviation between the multiplication
factors was about 0.2 %, so the use of the limited number of groups in this calculational
scheme does not lead to considerable error. Figure 3 shows the axial linear heat rate
distributions using the different options.

H P L WR av e rag e fu e l ro d
S N AP FD N G = 2 S N AP FD N G = 4
S N AP FD N G = 6 K AR ATE N o d a l N G = 2

25000
Linear heat rate

20000
[W/m]

15000
10000
5000
0
0.0 1.0 2.0 3.0 4.0 5.0
H e ig h t [m ]

Figure 3
Axial linear heat rate distributions of an average HPLWR fuel rod

The main results of our activity during FP 5:

• Reactivity coefficients of fresh core are similar to PWR EOC conditions (zero boron).
• The applicability of zirconium hydride solid moderator rods instead of using water
moderator tubes was investigated from the point of view of possible cycle length. The
application of ZrH showed no clear advantage.
• Possibility of reactivity control via adjusting of flow rate of water rods is very limited.

In 2005 the NUKENERG Project became a Hungarian sponsored contribution to the

Supercritical Water Reactor research: The Hungarian National Office for Research and
Technology launched a call to support research institutes which participate in international
research and development projects. "Development of Technological Elements for New
Nuclear Energy Production Methods" (NUKENERG) includes both fusion and fission related
issues. The fission part covers various fields of the Supercritical Water Reactor research: core
design, safety, thermal-hydraulics, materials. There are two participants in the fission part: the
KFKI Atomic Energy Research Institute and the Nuclear Technology Institute (NTI) of the
Budapest University of Technology and Economics.

In September 2006 the "HPLWR Phase 2" FP 6 project had its Kick-Off Meeting in
Budapest. Seven research institutes, two universities and an industrial partner works on the
concept, including 3 AER member institutes: VTT, ÚJV and KFKI-AEKI.

As in the "HPLWR Phase 2" FP 6 project there is limited acknowledgement of code

development, we have used our national programme to upgrade and test our code system.
Benchmarking against other codes including high precision Monte Carlo calculations are
necessary because of the change from hexagonal to square geometry [5].

The MULTICELL transport code was validated solving the TAKAHAMA spent fuel
benchmark [6], which is a test for predicting the isotopic composition of fuel on the basis of
measured data. The benchmark can also be used to test the multiplication factor and power
distribution provided by MULTICELL. The reference data were calculated with MCNP.
Figure 4 shows the map of the 17x17 fuel assembly with the indication of the 45˚ symmetry
sector. On the boundary of the assembly reflective boundary condition is imposed.

Figure 4
Map of the 17x17 TAKAHAMA fuel assembly.

The deviation of the multiplication factors is 0.3%. Figure 5 shows the power peaking
factors calculated by MCNP and MULTICELL. The numbering of fuel pins corresponds to
the reading sequence in the 1/8 symmetry sector.
 MCNP MULTICELL

1.2

Relative Power
1
0.8
0.6
0.4
0.2
0
0 10 20 30 40 50
Pin No.

Figure 5.
Power peaking factors of the TAKAHAMA assembly

For testing the burnup process the coupling of the MCNP Monte Carlo code to TIBSO
burnup code with more than one thousand isotopes was realized. The spectrum and shielded
resonance cross sections of some isotopes calculated by MCNP were transferred to the
subsequent TIBSO burnup calculation. The coupled code can be used for the calculation of
SCWR fuel composition. In figure 6 the burnup of a hypothetical SCWR fuel pin is
demonstrated.

MULTICELL MCNP-TIBSO

1.30

1.20

1.10
Kinf

1.00

0.90

0.80

0.70
0 20000 40000 60000 80000
Burnup [MWd/tU]

Figure 6.
Burnup of a hypothetical SCWR fuel pin.

In the newly developed square GLOBUS code the response matrices of the nodes
calculated by KIKO3D were applied. The response matrices of the square nodes in flux-
current representation were converted to partial current representation and were applied in the
GLOBUS partial current iteration scheme. The results of the code for the Langenbuch 3-D
test case are compared to that of the CUBBOX code in the following table, which shows
sufficient agreement.

Local Power Density [W/cm]

Node: 1 2 3 4
CUBBOX 279.86 186.17 258.26 231.45
GLOBUS SQR 283.83 184.09 259.17 232.63
Difference[%] +1.4 -1.1 +0.3 +0.5
Table 1
Results of the Langenbuch 3-D Test Case
 FUTURE PROSPECTIVES

The more heterogeneous fuel assembly structure than usual, the smaller node size than
usual, the steep axial water density variation and last but not least the lack of experimental
results require a thorough verification with the help of a Monte Carlo code. The step by step
verification procedure still being carried out includes the test of few group cross section
generation, the test of axial streaming and axial reflector description method, the test of
interaction of assemblies and radial reflector description method. The last test is the realistic
full core calculation with reflector. The preliminary results [7] show the applicability of the
KARATE code system to supercritical water reactor research. Our future goal is the analysis
of HPLWR assemblies and core to improve core performance and to calculate basic safety
features with the KARATE code system.

REFERENCES:

1. Dobashi, K., Oka, Y., Koshizuka, S. "Conceptual Design of a High Temperature

Power Reactor Cooled and Moderated by Supercritical Light Water", 6th International
Conference on Nuclear Engineering, ICONE6, ASME, NY, NY(1998)

2. D. Squarer, Y. Oka, D. Bittermann, N. Aksan, C. Maraczy, R. Kyrki-Rajamaki, A.

Souyri, P. Dumaz: High Performance Light Water Reactor (HPLWR) FISA-2001 EU
Research in Reactor Safety, European Commission, Luxembourg, 12-14 November
2001, ISBN 92-894-3455-4, p. 620-630

3. D. Squarer, T. Schulenberg, D. Struwe, Y. Oka, D. Bittermann, N. Aksan, C.

Maraczy, R. Kyrki-Rajamäki, A. Souyri and P. Dumaz: High Performance Light
Water Reactor, Nuclear Engineering and Design, Volume 221, Issue 1-3, April 2003,
p. 167-180

4. Cs. Hegedűs, Gy. Hegyi, G. Hordósy, A. Keresztúri, M. Makai, Cs. Maráczy, F.

Telbisz, E. Temesvári, P. Vértes: The KARATE Program System, PHYSOR 2002,
Seoul, Korea, October 7-10, 2002

5. J. Hofmeister, T. Schulenberg, J. Starflinger: "Optimization of a fuel assembly for a

HPLWR" International Congress on Advances in Nuclear Power Plants, Seoul,
KOREA, May 15-19, 2005, Proceedings of ICAPP ’05, Paper 5077

6. http://www.nea.fr/html/science/wpncs/sfcompo.

7. S. Fehér: Personal communication: Nuclear Technology Institute (NTI) of the

Budapest University of Technology and Economics. 2006, Budapest