Neutronic Evaluation of MSBR System Using MCNP Code

The concept of Molten Salt Reactor use Th to breed fissile U, where an initial source of fissile material needs to be provided. However, there is no available U and so; the fissile fuel supply is one of the unresolved problems. Thus, it is necessary to use existing fissile materials such as U or Pu to produce U. Current studies analyze the fuel transition from U/Th or Pu/Th to U/Th and, in this context, the present work evaluates the criticality and the neutron flux of MSBR (Molten Salt Breeder Reactor) considering the fuel: (i) mix of Th and enriched U; (ii) the combination of Th and reprocessed Pu; and (iii) matrix of reprocessed Pu/minor actinides (MAs) and Th. The goal is to verify which of these fuels can be used as initial fissile supply. The MSBR core was simulated by MCNPX 2.6.0 code and the criticality model presents similar behavior of previous studies. The results show that reprocessed fuels could have a potential to be used as initial fissile supply, but these fuels present a neutron flux profile less flattens than traditional U/Th. It is possible that a new distribution of fuel elements may improve this profile and future simulations will be performed to evaluate this behavior. The uranium, must has high enrichment value to be used as initial seed. Other studies need be performed to evaluates the uranium enrichment and the U/Th ratio that produces similar core criticality to traditional fuel.


INTRODUCTION
The energy demands are expected to grow at different rates around the world. Several studies have shown that nuclear power plants may represent very attractive options to the energy generation especially due to greenhouse gas emissions reductions. However, the economics, proliferation, safety, and waste production are major barriers to the expansion of nuclear power. In this context, design, though it was never experienced in practice [1]. In a MSR system, the fuel is a mixture of salts, generally molten fluoride containing fissile materials, which circulate between the reactor core and the heat exchanger. Thorium fueled MSR concept is one of the most promising nuclear reactor designs currently being studied, because to the high availability of naturally occurring this fuel.
Several works have been studied this nuclear system (e.g. [2][3][4][5][6][7][8][9][10]) where the main interest is to using 232 Th to breed 233 U. However, considering that is no available 233 U in the nature, it is crucial to analyze which initial fissile material that could give a flexible transition in fuel cycle to MSR system. The traditional concept proposes to use 235 U as initial fissile fuel supply, but Pu and minor actinides (MAs) could be used instead of U for initial fissile loading. The present work studies the neutronic behavior of MSBR system using the following initial fissile source: (a) uranium, (b) reprocessed Pu and (c) a matrix of reprocessed Pu/MAs. The goal is evaluates the use these nuclides as initial seed of the system in the fission/transmutation process. The MCNPX 2.6.0 code was used to calculate the criticality and the neutron flux profile at steady state of MSBR [11].

Evaluated Fuels and Material Compositions
The standard fuel salt of MSBR system is a mixture of LiF-BeF2-ThF4-UF4. In the ORNL report there are two mole percentage of UF4, 0.3 and 0.232% which represent a conceptual and an optimized MSBR design respectively [12]. These values correspond to mass percentage (mp) of fissile material ( 233 U) of 2.26% and 1.76%. In order to evaluate other nuclides as initial fissile loading in MSBR system, the current study simulates four fuels types considering the two values of fissile content reported by ORNL. The following fuel types were simulated:  Table 1 present the salt composition of the evaluated fuels. The isotopic calculation was based on mole percentage and on mp of fissile material. In the heavy metal mass, the highest mp of fissile isotopes implies in the lowest thorium concentration. Thorium, uranium, plutonium and minor actinides all form suitable fluoride salts that readily dissolve in the LiF-BeF2-ThF4 mixture. The fissile and fertile isotopes can be easily separated from one another in fluoride form. The isotopic composition of the Pu and Pu/MAs was calculated by previous studies using ORIGEN code [13].
These studies simulate a typical burnup of PWR fuel with 4.5% of initial enrichment at a cycle of 33 GWd/MTU, where the spent fuel remained in the cooling pool for five years. Table 2  In the MSBR, the graphite is the principal material other than salt. The core contains graphite for neutron moderation and reflection. In the simulations the moderator elements and the reflector blocks are made of natural graphite. Also, the reactor vessel is composed of Hastelloy N that is an alloy developed specially for use in molten fluoride systems. Among the major constituents, chromium is the least resistant to attack by fluorides. The chromium content of Hastelloy N is low enough for the alloy to have excellent corrosion resistance toward the salts. Table 3 present the composition in mass percentage of moderator, reflector and reactor vessel used in the simulations.

Computational Model of MSBR
The MSBR configurations use the data from a conceptual design developed by ORNL [12]. Figure 1 illustrates the geometry of the simulated system and Table 4 present, the main dimensions of the simulated system. Eight initial neutron sources were uniformly distributed in active core of the reactor. The estimated standard deviation is around 310 -4 .
The nuclear data were downloaded from ENDF/BVII-1 (Evaluated Nuclear Data File) website [14] and processed with NJOY99 (Nuclear Data Processing System) [15] at the operational temperature of MSBR, about 900 K. Thus, these data was added to the MCNPX 2.6.0 library.

Evaluated Parameters
The MCNPX estimates the effective multiplication factor (keff) with the respective standard deviation (σST) of the simulated model. This work evaluates the criticality and the neutron flux of the MSBR at steady state condition to the EU, RU, Pu and Pu/AMs fuels.
To evaluate the neutron flux distribution in the core, the average neutron flux was calculated by MCNPX to each reactor cell. The TMESH card was used which allows the user to tally particles on a mesh independent of the problem geometry [11]. normalization was performed using the following equation [11]: where N  is the normalized flux; MCNPX  is the flux estimated by MCNPX; P is the reactor power level;  is the average number of fission neutrons and Q is the recoverable energy per fission event.
The values of  , Q and keff are calculated by the MCNPX and the power level is designed by MSBR project at P = 2250 MWt [11]. Table 5 present the effective multiplication factor (keff) of MSBR system among several works.

RESULTS AND DISCUSSION
The keff was calculated at beginning of cycle (BOC) to traditional RU considering 1.76% of fissile content. The keff calculated by de MCNPX 2.6.0, in the present study, agree with others works. The biggest difference is related to SERPENT code. According to reference [8], this discrepancy may be due the simplifications in Zone II performed in SERPENT model.  Table 6 presents the keff of the MSBR core for the evaluated fuels. The RU presents the highest keff values and the RA has the smallest one. The same mass percentage (mp) of fissile isotopes does not produce similar keff values, because the RP and RA fuels have nuclides which present high neutrons cross section for radiative capture. Furthermore, the atomic masses of the fissile isotopes are different. For the same mass and the same mp, the fuel that contains fissile isotopes with lightest atomic mass has more fissile atoms. In this way, among the evaluated fuels, the fissions number is the highest for the RU, which produces the highest keff value. Because these factors, the concentration of fissile isotopes in EU, RP and RA, must be higher than 2.26% in fuel mass.
However, in the uranium mass, the EU fuel has 91% of 235 U (Table 1). This fact is negative for the non-proliferation issue because this high enrichment value is forbidden by international treatises.
Thus, the present study forsakes the use 235 U and focuses on the analysis of RP and RA as initial fissile seed. The respective fissile content of Pu and Pu/AMS is 62 and 54% (Table 2). Although it is a high fissile content value, the radiotoxicity of the spent fuel hinders its proliferation. Moreover, the Pu and Pu/AMs are diluted in thorium mass to form the RP and RA fuels, where the percentage of fissile isotopes in fuel matrix is less than 54%. Table 7 present the keff to RP and RA fuels as function of fissile content in the fuel matrix and as function of mass percentage of heavy metal. As expected, the gradual increase of fissile isotopes concentration produces higher keff values (Table 7).
When RP and RA have 16% and 25% of fissile isotopes the keff is similar to RU fuel with 1.76% (see Table 6). Note that the fissile content 233 U/U in RU is 92% (Table 1)

CONCLUSION
The criticality of the simulated MSBR, using RU fuel at BOC, is similar to previous studies.
Among the evaluated fuels RP and RA seems to be promisor to initial fissile supply. Although the reprocessed Pu and Pu/MAs have about of 62 and 54% of fissile isotopes, the radiotoxicity of this spent fuel makes it difficult for proliferation. Diluting these isotopes in thorium mass, 16 and 25% of the fissile isotope in fuel matrix of RP and RA, produce similar criticality of traditional RU. On the other hand, the presence of neutron absorbers in RP and RA reduce the radial neutron flux and produce a central peak in its profile. Studies to evaluate a new distribution of fuel elements are important to flatten the neutron flux.
Regarding to EU fuel, the used uranium needs to be highly enriched. The enrichment value estimated by this work is impractical due the limits of enrichment values established by the nonproliferation treaty. New studies need be performed to evaluate a uranium enrichment values that does not exceeds 20%. In these studies, it is necessary calculates the U/Th ratio that produces similar core criticality to traditional RU.
The use of plutonium and minor actinides as initial fissile source in MSBR, could contribute to reduction to waste production. However, it is essential evaluate the core behavior during the burnup, the spent fuel composition and the breeding ratio value among the evaluated fuels. Future works will simulate MSBR system to study these technical features where other nuclear codes can be used to compare the results.