The existing parallel computing schemes for Monte Carlo criticality calculations suffer from a low efficiency when applied on many processors. We suggest a new fission matrix based scheme for efficient parallel computing. The results are derived from the fission matrix that is combined from all parallel simulations. The scheme allows for a practically ideal parallel scaling as no communication among the parallel simulations is required, and inactive cycles are not needed. (C) 2009 Elsevier Ltd. All rights reserved.
We have described a fission matrix based method that allows to cancel the inactive cycles in Monte Carlo criticality calculations. The fission matrix must be sampled in the course of the Monte Carlo calculation using a space mesh with sufficiently small zones as it causes the fission matrix be insensitive to errors in the initial fission source. The k(eff) and other quantities can be derived by means of the final fission matrix. The confidence interval for the k(eff) estimate can be conservatively determined via the variance in the fission matrix.
The Monte Carlo fission matrix acceleration methods aim at accelerating the convergence of the fission source in inactive cycles of Monte Carlo criticality calculations. In practice, however, these methods may corrupt the fission source, or slow down its convergence. These phenomena have not been completely understood so far. We demonstrate the convergence problems, and explain their reasons.
A new adaptive stochastic approximation method for an efficient Monte Carlo calculation of steady-state conditions in thermal reactor cores is described The core conditions that we consider are spatial distributions of power, neutron flux, coolant density, and strongly absorbing fission products like Xe-135. These distributions relate to each other; thus, the steady-state conditions are described by a system of nonlinear equations. When a Monte Carlo method is used to evaluate the power or neutron flux, then the task turns to a nonlinear stochastic root-finding problem that is usually solved in the iterative manner by stochastic optimization methods. One of those methods is stochastic approximation where efficiency depends on a sequence of stepsize and sample size parameters. The stepsize generation is often based on the well-known Robbins-Monro algorithm; however, the efficient generation of the sample size (number of neutrons simulated at each iteration step) was not published yet. The proposed method controls both the stepsize and the sample size in an efficient way; according to the results, the method reaches the highest possible convergence rate.
A comparative safety study has been performed on sodium vs. lead/bismuth as coolant for accelerator-driven systems. Transient studies are performed for a beam overpower event. We examine a fuel type of recent interest in the research on minor actinide burners, i.e. uranium-free oxide fuel. A strong positive void coefficient is calculated for both sodium and lead/bismuth. This is attributed to the high fraction of americium in the fuel. It is shown that the lead/bismuth-cooled reactor features twice the grace time with respect to fuel or cladding damage compared to the sodium-cooled reactor of comparable core size and power rating. This accounts to the difference in void reactivity contribution and to the low boiling point of sodium. For improved safety features the general objective is to reduce the coolant void reactivity effect. An important safety issue is the high void worth that could possibly drive the system to prompt criticality.
Support vector machines (SVMs), a relatively new paradigm in statistical learning theory, are studied for their potential to recognize transient behavior of detector signals corresponding to various accident events at nuclear power plants (NPPs). Transient classification is a major task for any computer-aided system for recognition of various malfunctions. The ability to identify the state of operation or events occurring at an NPP is crucial so that personnel can select adequate response actions. The Modular Accident Analysis Program, version 4 (MAAP4) is a program that can be used to model various normal and abnormal events in an NPP. This study uses MAAP signals describing various loss-of-coolant accidents in boiling water reactors. The simulated sensor readings corresponding to these events have been used to train and test SVM classifiers. SVM calculations have demonstrated that they can produce classifiers with good generalization ability for our data. This in, turn indicates that SVMs show promise as classifiers for the learning problem of identifying transients.
This paper describes principles of Accelerator-Driven Transmutation of Nuclear Wastes (ATW) and gives some flavour of the most important topics which are today under investigations in many countries. Pin assessment of the potential impact of ATW on a future of nuclear energy is also given. Nuclear reactors based on self-sustained fission reactions - after spectacular development in fifties and sixties, that resulted in deployment of over 400 power reactors - are wrestling today more with public acceptance than with irresolvable technological problems. In a whole spectrum of reasons which resulted in today's opposition against nuclear power few of them are very relevant for the nuclear physics community and they arose from the fact that development of nuclear power had been handed over to the nuclear engineers and technicians with some generically unresolved problems, which should have been solved properly by nuclear scientists. In a certain degree of simplification one can say, that most of the problems originate from very specific features of a fission phenomenon: self-sustained chain reaction in fissile materials and very strong radioactivity of fission products and very long half-life of some of the fission and activation products. And just this enormous concentration of radioactive fission products in the reactor core is the main problem of managing nuclear reactors: it requires unconditional guarantee for the reactor core integrity in order to avoid radioactive contamination of the environment; it creates problems to handle decay heat in the reactor core and finally it makes handling and/or disposal of spent fuel almost a philosophical issue, due to unimaginable long time scales of radioactive decay of some isotopes. A lot can be done to improve the design of conventional nuclear reactors (like Light Water Reactors); new, better reactors can be designed but it seems today very improbable to expect any radical change in the public perception of conventional nuclear power. In this context a lot of hopes and expectations have been expressed for novel systems called Accelerator-Driven Systems, Accelerator-Driven Transmutation of Waste or just Hybrid Reactors. All these names are used for description of the same nuclear system combining a powerful particle accelerator with a subcritical reactor. A careful analysis of possible environmental impact of ATW together with limitation of this technology is presented also in this paper.
The IABAT project - Impact of Accelerator Based Technologies on Nuclear Fission Safety - started in 1996 in the frame of 4(th) Framework Programme of the European Union, R&D specific programme Nuclear fission safety 1994-1998, area A.2 Exploring innovative approaches/Fuel cycle concepts, as one of the first common European activities in ADS. The project was completed October 31, 1999. The overall objective of the IABAT project has been a preliminary assessment of the potential of Accelerator-Driven Systems (ADS) for transmutation of nuclear waste and for nuclear energy production with minimum waste generation. Moreover, more specific topics related to nuclear data and code development for ADS have been studied in more detail. Four ADSs have been studied for different fuel/coolant combinations: liquid metal coolant and solid fuel, liquid metal coolant and dispersed fuel, and fast and thermal molten salt systems. Target studies comprised multiple target solutions and radiation damage problems in a target environment. In a tool development part of the project a methodology of subcriticality monitoring has been developed based on Feynman-alpha and Rossi-alpha methods. Moreover, a new Monte-Carlo burnup code taking full advantage of continuous neutron cross-section data has been developed and benchmarked. Impact on the risk from high-level waste repositories fi om radiotoxicity reduction using ADS has been assessed giving no crystal-clear benefits of ADS for repository radiotoxicity reduction but concluding some important prerequisites for effective transmutation. In proliferation studies important differences between critical reactors and ADS have been underlined and non-proliferation measures have been proposed. In assessment of accelerator technology costing models have been created that allow the circular and linear accelerator options to be compared and the effect of parameter variations examined. The calculations reported show that cyclotron systems would be more economical, due mainly to the advantage of the cost of RF power supplies. However, the accelerator community regards with skepticism the possibility of transporting and extracting more than a 10mA beam current from a 1GeV cyclotron and therefore technical factors may limit the application of cyclotrons. Finally, this review summarizes development of nuclear data in the energy region between 20 Mev and 150 MeV. Neutron and proton transport data files for Fe, Ni, Pb, Th, U-238 and Pu-239 have been created. The high-energy part of the data files consists completely of results from model calculations, which are benchmarked against the available experimental data. Although there is obviously future work left regarding fine-tuning of several parts of the data files, the representation of nuclear reaction information up to 150 MeV is already better than can be attained with intranuclear cascade codes.
Subcritical Assembly in Dubna (SAD), a project funded by the International Science and Technology Centre, driven in collaboration with many European partners, may become the first Accelerator Driven Subcritical experiment coupling an existing proton accelerator of 660 MeV with a compact MQX-fuelled subcritical core. The main objective of the SAD experiment is to study physics of Accelerator Driven System ranging from a very deep subcriticality up to k(eff) of 0.98. All experiences with subcriticality monitoring from previous subcritical experiments like MUSE, Yalina and IBR-30 booster mode will be verified in order to select the most reliable subcriticality monitoring technique. Particular attention will be given to validation of the core power-beam current relation. Moreover, some studies have been done to assess possibility of power upgrade for SAD.
Investigations aimed at the development of neutron and proton cross-section evaluations for Th-232 at intermediate energies in the range of 0 to 200 MeV are described The coupled-channel optical model has been used to calculate the neutron total, elastic, and reaction cross sections and the elastic scattering angular distributions. Evaluations of the neutron and charged particle emission cross sections and of the fission cross sections have been obtained on the basis of the statistical description that includes direct, preequilibrium, and equilibrium mechanisms of nuclear reactions. The Kalbach parameterization of angular distributions has been used to describe the double-differential cross sections of emitted neutrons and charged particles in ENDF/B-VI format.
Investigations aimed at the development of neutron cross-section evaluations for U-238 at intermediate energies are briefly described. The coupled-channels optical model is used to calculate the neutron total, the elastic and reaction cross sections, and the elastic-scattering angular distributions. Evaluations of the neutron and charged particle emission cross sections and of the fission cross sections are obtained on the basis of the statistical description that includes direct, preequilibrium, and equilibrium mechanisms of nuclear reactions. The Kalbach parameterization of angular distributions is used to describe the double-differential cross sections of emitted neutrons and charged particles in ENDF/B-VI format.
Subcritical experiments, devoted to investigation of characteristics of accelerator-driven systems, have been constructed at the Joint Institute for Power and Nuclear Research - Sosny in Minsk, Belarus. Different methods for reactivity determination have previously been investigated in the thermal ADS experiment "Yalina", and recently, a coupled fast-thermal facility "Yalina Booster" was launched. This study presents the neutron kinetic characteristics of the Yalina and the new Yalina Booster setups, and points out some important differences. For the Yalina setup, neutron kinetic parameters, such as keff, α, βeff and Λ have been determined by Monte Carlo simulations and they have previously been verified experimentally. For Yalina Booster, these parameters have been estimated by Monte Carlo simulations in a preliminary study, and they will be verified in upcoming experiments.
Different reactivity determination methods have been investigated, based on experiments performed at the subcritical assembly Yalina in Minsk, Belarus. The development of techniques for on-line monitoring of the reactivity level in a future accelerator-driven system (ADS) is of major importance for safe operation. Since an ADS is operating in a subcritical mode, the safety margin to criticality must be sufficiently large. The investigated methods are the Slope Fit Method, the Sjostrand Method and the Source Jerk Method. The results are compared with Monte Carlo simulations performed with different nuclear data libraries. The results of the Slope Fit Method are in good agreement with the Monte Carlo simulation results, whereas the Sjostrand Method appears to underestimate the criticality somewhat. The Source Jerk Method is subject to inadequate statistical accuracy.
Two subcritical configurations of the zero-power coupled subcritical core YALINA-Booster have been identified through pulsed neutron source measurements. The area ratio and the slope fitting reactivity estimation methods have been utilized as well as the pulsed Rossi-a noise method. The measurements showed that despite the inhomogeneous two-zone core composition a clear single exponential prompt neutron decay was obtained. Spatial spread of the results and converegence issues related to the area ratio method are addressed.
A subcritical zero-power source-driven coupled core, the YALINA-Booster. has been constructed for experimental investigations of neutron kinetics of source-driven systems. In this study, the reactivity of two subcritical configurations has been determined by the area ratio method. The prompt neutron decay constants have been evaluated through slope fitting of the prompt neutron decay as well as through the pulsed Rossi-alpha method. It is shown that the slope fitting method and the pulsed Rossi-alpha method give stable results whereas the area ratio method results show spatial dependence. The reasons for the spatial spread are addressed.
The performance of the fusion–fission hybrid system based on the molten salt (flibe) blanket, driven by a plasma based fusion device, was analyzed by comparing transmutation scenarios of actinides extracted from the LWR (Sweden) and RBMK (Lithuania) spent nuclear fuel in the scope of the EURATOM project BRILLIANT. The IAEA nuclear fuel cycle simulation system (NFCSS) has been applied for the estimation of the approximate amount of heavy metals of the spent nuclear fuel in Sweden reactors and the SCALE 6 code package has been used for the determination of the RBMK-1500 spent nuclear fuel composition. The total amount of trans-uranium elements has been estimated in both countries by 2015. Major parameters of the hybrid system performance (e.g., kscr, keff, Φn(E), equilibrium conditions, etc.) have been investigated for LWR and RBMK trans-uranium transmutation cases. Detailed burn-up calculations with continuous feeding to replenish the incinerated trans-uranium material and partial treatment of fission products were done using the Monteburns (MCNP + ORIGEN) code system. About 1.1 tons of spent fuel trans-uranium elements could be burned annually with an output of the 3 GWth fission power, but the equilibrium stage is reached differently depending on the initial trans-uranium composition. The radiotoxicity of the remaining LWR and RBMK transmuted waste after the hybrid system operation time has been estimated.
The paper present results of Monte Carlo modeling of an Experimental Accelerator Driven System (ADS), which employs a subcritical assembly and a 660 MeV proton accelerator operating at the Laboratory of Nuclear Problems of the Joint Institute for Nuclear Research in Dubna. The mix of oxides PuO2 + UO2 MOX fuel designed for the reactor will be adopted for the core of the assembly. The design of the experimental subcritical assembly in Dubna (SAD) is based on the core with a nominal unit capacity of 30 kW (thermal). This corresponds to the multiplication coefficient K-eff = 0.95 and the accelerator beam power of I kW. A subcritical assembly has been modeled in order to increase power of this experimental set up. Different options for the target and fuel elements have been considered.
The radiation fields outside the planned experimental Sub-critical Assembly in Dubna (SAD) have been studied in order to provide a basis for the design of the concrete shielding that cover the reactor core. The effective doses around the reactor, induced by leakage of neutrons and photons through the shielding, have been determined for a shielding thickness varying from 100 to 200 cm. It was shown that the neutron flux and the effective dose is higher above the shielding than at the side of it, owing to the higher fraction of high-energy spallation neutrons emitted in the direction of the incident beam protons. At the top, the effective dose was found to be similar to 150 mu Sv s(-1) for a concrete thickness of 100 cm, while similar to 2.5 mu Sv s(-1) for a concrete thickness of 200 cm. It was also shown that the high-energy neutrons (> 10 MeV), which are created in the proton-induced spallation interactions in the target, contribute for the major part of the effective doses outside the reactor.
The radiation fields and the effective dose at the Sub-critical Assembly in Dubna (SAD) have been studied with the Monte Carlo code MCNPX. The effective dose above the shielding, i.e. in the direction of the incident proton beam of 3.0 mu A, was found to be about 190 mu Sv h(-1). This value meets the dose limits according to Russian radiation protection regulations, provided that access to the rooms in this area is not allowed for working personnel during operation.By separating the radiation fields into a spallation- and a fission-induced part, it was shown that the neutrons with energy higher than 10MeV, originating exclusively from the proton-induced spallation reactions in the target, contribute for the entire part of the radiation fields and the effective dose at the top of the shielding. Consequently, the effective dose above the SAD reactor system is merely dependent on the proton beam properties and not on the reactivity of the core.
The proton source efficiency (ψ*) was studied for homogeneous and heterogeneous distributions of minor actinides in a nitride-fuelled and lead-bismuth-cooled accelerator-driven system. The findings from the MCNPX simulations indicate that, compared to a homogeneous configuration, a gain in ψ* by up to 16% can be obtained by distributing the minor actinides heterogeneously, Cm being placed in the inner zone of the active core and Am in the outer zone. The reason for this is the higher fission probability for neutrons for Cm than for Am in the energy range below 1.0 MeV. Moreover, a comparative study of two different physics packages available in MCNPX, the Bertini and the CEM models, has been performed, focusing on the production of neutrons in the spallation target and on the proton source efficiency. The Bertini model was found to produce a higher number of neutrons in the low-energy range (below ∼15 MeV) than the CEM model. Consequently, the Bertini model also over-estimates ψ* by about 10%, compared to the CEM model.
In order to study the beam power amplification of an accelerator-driven system (ADS), a new parameter, the proton source efficiency psi* is introduced. psi* represents the average importance of the external proton source, relative to the average importance of the eigenmode production, and is closely related to the neutron source efficiency rho*, which is frequently used in the ADS field. rho* is commonly used in the physics of subcritical systems driven by any external source (spallation source, (d,d), (d, t), Cf-252 spontaneous fissions, etc.). On the contrary, psi* has been defined in this paper exclusively for ADS studies where the system is driven by a spallation source. The main advantage with using psi* instead of rho* for ADS is that the way of defining the external source is unique and that it is proportional to the core power divided by the proton beam power, independent of the neutron source distribution.
Numerical simulations have been performed with the Monte Carlo code MCNPX in order to study psi* as a function of different design parameters. It was found that, in order to maximize psi* and therefore minimize the proton current needs, a target radius as small as possible should be chosen. For target radii smaller than similar to30 cm, lead-bismuth is a better choice of coolant material than sodium, regarding the proton source efficiency, while for larger target radii the two materials are equally good. The optimal axial proton beam impact was found to be located similar to 20 cm above the core center. Varying the proton energy, psi*/E-p was found to have a maximum for proton energies between 1200 and 1400 MeV Increasing the americium content in the fuel decreases psi* considerably, in particular when the target radius is large.
A demonstration facility for Accelerator Driven Systems has been proposed to be constructed at the Joint Institute of Nuclear Research in Dubna. The Subcritical Assembly in Dubna project proposes to couple an existing proton accelerator of 660 MeV and 1 mu A current with a specially designed U-Pu MOX subcritical core. Project objectives, technical description and current status of the project are presented in this paper.
In 1966, Philadelphia Electric has put into operation the Peach Bottom I nuclear reactor, it was the first high temperature gas reactor (HTGR); the pioneering of the helium-cooled and graphite-moderated power reactors continued with the Fort St. Vrain and THTR reactors, which operated until 1989. The experience on HTGRs lead General Atomics to design the gas turbine - modular helium reactor (GT-MHR), which adapts the previous HTGRs to the generation IV of nuclear reactors. One of the major benefits of the GT-MHR is the ability to work on the most different types of fuels: light water reactors waste, military plutonium, MOX and thorium. In this work, we focused on the last type of fuel and we propose a mixture of 40% thorium and 60% uranium. In a uranium-thorium fuel, three fissile isotopes mainly sustain the criticality of the reactor: U-235, which represents the 20% of the fresh uranium, U-233, which is produced by the transmutation of fertile Th-212, and Pu-239, which is produced by the transmutation of fertile U-238. In order to compensate the depletion of U-235 with the breeding of U-233 and Pu-239, the quantity of fertile nuclides must be much larger than that one of 235 U because of the small capture cross-section of the fertile nuclides, in the thermal neutron energy range, compared to that one of U-235. At the same time, the amount of U-235 must be large enough to set the criticality condition of the reactor. The simultaneous satisfaction of the two above constrains induces the necessity to load the reactor with a huge mass of fuel; that is accomplished by equipping the fuel pins with the JAERI TRISO particles. We start the operation of the reactor with loading fresh fuel into all the three rings of the GT-MHR and after 810 days we initiate a refueling and shuffling schedule that, in 9 irradiation periods, approaches the equilibrium of the fuel composition. The analysis of the k(eff) and mass evolution, reaction rates, neutron flux and spectrum at the equilibrium of the fuel composition, highlights the features of a deep burn in-core fuel management strategy for a uranium-thorium fuel.
One of the major benefits of the Gas Turbine-Modular Helium Reactor is the capability to operate with several different types of fuel; either Light Water Reactors waste, military plutonium or thorium represent valid candidates as possible types of fuel. In the present studies, we performed a comparison of various nuclear data libraries by the Monte Carlo Continuous Energy Burnup Code MCB applied to the Gas Turbine-Modular Helium Reactor operating on a thorium fuel. A thorium fuel offers valuable attractive advantages: low fuel cost, high reduction of actinides production and the possibility to enable the reactor to act as a breeder of fuel by the neutron capture of fertile Th-232. We evaluated the possibility to mix thorium with small quantities, about 3% in atomic composition, of Pu-239, U-233 and U-235. The mass of thorium must be much larger than that one of plutonium or uranium because of the low capture cross section of thorium compared to the fission one of the fissile nuclides; at the same time, the quantity of the fissile isotopes must grant the criticality condition. These two simultaneous constraints force to load a huge mass of fuel in the reactor; consequently, we propose to allocate the fuel in TRISO particles with a large radius of the kernel. For each of the three different fuels we calculated the evolution of the fuel composition by the MCB code equipped with five different nuclear data libraries: JENDL-3.3, JENDL-3.2, JEFF-3, JEF-2.2 and ENDF/B.
In the present study we investigate the influence of the fuel axial shuffling and the operational control rod maneuvering on the performances of the one-pass (no reprocessing) deep-burn incineration of light water reactor waste in the gas turbine-modular helium reactor. After an irradiation period, the fuel axial shuffling schedule has to take into account the fuel depletion profile generated by the adjustments of the position of the operational control rods, because the insertion of the rods strongly alters the neutron flux shape. We aimed at implementing a numerical simulation as close as possible to a real scenario and therefore took advantage of the powerful geometrical modeling capability of the MCB code to describe the reactor in a detailed three-dimensional geometry model in which we simulated over 120 different burnable materials, each of them undergoing a different neutron flux intensity. We adjusted the position of the control rods every 90 effective full-power days of irradiation to maintain the core as close as possible to the critical condition; thereafter, we recalculated the neutron flux and cross sections by a new MCNP/ MCB run. At the present time, this sophisticated approach can be realized only by a computer cluster of ten 64-bit processors working in parallel mode. The fuel axial shuffling adds from 3 to 5% to the transmutation rates of 239Pu, plutonium, and all actinides, which range from 80 to 86, 50 to 53, and 46 to 48%, respectively; the present results are 5 to 14% less compared to the case of a two-pass (reprocessing) deep burn. The efficiency of transmuting minor actinides has been estimated by comparing the long-term radio-toxicity of the fresh and irradiated americium and curium fuel; this comparison revealed that it is not worthwhile to transmute americium and curium in the current design of the gas turbine-modular helium reactor by a one-pass deep burn.
Preliminary studies have been performed on operation of the gas turbine-modular helium reactor (GT-MHR) with a thorium based fuel. The major options for a thorium fuel are a mixture with light water reactors spent fuel, mixture with military plutonium or with with fissile isotopes of uranium. Consequently, we assumed three models of the fuel containing a mixture of thorium with 239Pu, 233U or 235U in TRISO particles with a different kernel radius keeping constant the packing fraction at the level of 37.5%, which corresponds to the current compacting process limit. In order to allow thorium to act as a breeder of fissile uranium and ensure conditions for a self-sustaining fission chain, the fresh fuel must contain a certain quantity of fissile isotope at beginning of life; we refer to the initial fissile nuclide as triggering isotope. The small capture cross-section of 232Th in the thermal neutron energy range, compared to the fission one of the common fissile isotopes (239Pu, 233U and 235U), requires a quantity of thorium 25-30 times greater than that one of the triggering isotope in order to equilibrate the reaction rates. At the same time, the amount of the triggering isotope must be enough to set the criticality condition of the reactor. These two conditions must be simultaneously satisfied. The necessity of a large mass of fuel forces to utilize TRISO particles with a large radius of the kernel, 300 μm. Moreover, in order to improve the neutron economics, a fuel cycle based on thorium requires a low capture to fission ratio of the triggering isotope. Amid the common fissile isotopes, 233U, 235U and 239Pu, we have found that only the uranium nuclides have shown to have the suitable neutronic features to enable the GT-MHR to work on a fuel based on thorium.
The deep burn fuel cycle for the incineration of military plutonium in the GT-MHR is studied using the Monte-Carlo burnup code. The irradiation is DF is so rich in fissile isotopes that the TF cannot guarantee a negative reactive feedback, and the presence of erbium as burnable poison is absolutely necessary for the reactivity safety reasons. At beginning of life (BOL) the fuel composed of DF, consisting of fresh military plutonium, after an irradiation period of three years the fuel is reprocessed into post driver fuel (PDF). The mass flow of the GT-MHR fuelled by military plutonium at the equilibrium of the fuel composition shows that 66% of 239Pu is burned in three years and 92% in six years.
In the future development of nuclear energy, the graphite-moderated helium-cooled reactors may play an important role because of their valuable technical advantages: passive safety, low cost, flexibility in the choice of fuel, high conversion energy efficiency, high burnup, more resistant fuel cladding, and low power density. General Atomics possesses a long experience with this type of reactor, and it has recently developed the gas turbine-modular helium reactor (GT-MHR), a design where the nuclear power plant is structured into four reactor modules of 600 MW(thermal). Amid its benefits, the GT-MHR offers a rather large flexibility in the choice of fuel type; Th, U, and Pu may be used in the manufacture of fuel with some degrees of freedom. As a consequence, the fuel management may be designed for different objectives aside from energy production, e.g., the reduction of actinide waste production through a fuel based on thorium. In our previous studies we analyzed the behavior of the GT-MHR with a plutonium fuel based on light water reactor (LWR) waste; in the present study we focused on the incineration of military Pu. This choice of fuel requires a detailed numerical modeling of the reactor since a high value of keff at the beginning of the reactor operation requires the modeling both of control rods and of burnable poison; by contrast, when the GT-MHR is fueled with LWR waste, at the equilibrium of the fuel composition, the reactivity swing is small.
We performed a numerical comparative analysis of the burnup capability of the Gas Turbine-Modular Helium Reactor (GT-MHR) by the Monte Carlo Continuous Energy Burnup Code (MCB). The MCB code is an extension of MCNP that includes the burnup implementation; it adopts continuous energy cross sections and it evaluates the transmutation trajectories for over 2,400 decaying nuclides. We equipped the MCB code with three different nuclear data libraries: JENDL-3.2, JEF-2.2 and ENDF/B-6.8 processed for temperatures from 300 to 1,800 K.
The GT-MHR model studied in this paper is fueled by actinides coming from the Light Water Reactors waste, converted into two different types of fuel: Driver Fuel and Transmutation Fuel. The Driver Fuel supplies the fissile nuclides needed to maintain the criticality of the reactor, whereas the Transmutation Fuel depletes non-fissile isotopes and controls reactivity excess. We set the refueling and shuffling period to one year and the in-core fuel residency time to three years.
The comparative analysis of the MCB code consists of accuracy and precision studies. In the accuracy studies, we performed the burnup calculation with different nuclear data libraries during the year at which the refueling and shuffling schedule set the equilibrium of the fuel composition. In the precision studies, we repeated the same simulations 20 times with a different pseudorandom number stride and the same nuclear data library.
We investigated some important neutronic features of the deep burn modular helium reactor (DB-MHR) using the MCNP/MCB codes. Our attention was focused on the neutron flux and its spectrum, capture to fission ratio of Pu-239 and the temperature coefficient of fuel and moderator. The DB-MHR is a graphite-moderated helium-cooled reactor proposed by General Atomic to address the need for a fast and efficient incineration of plutonium for nonproliferation purposes as well as the management of light water reactors (LWRs) waste. In fact, recent studies have shown that the use of the DB-MHR coupled to ordinary LWRs would keep constant the world inventory of plutonium for a reactor fleet producing 400 TWe/y. In the present studies, the DB-MHR is loaded with Np-Pu driver fuel (DF) with an isotopic composition corresponding to LWRs spent fuel waste. DF uses fissile isotopes (e.g. Pu-239 and Pu-241), previously generated in the LWRs, and maintains criticality conditions in the DB-MHR. After an irradiation of three years, the spent DF is reprocessed and its remaining actinides are manufactured into fresh transmutation fuel (TF). TF mainly contains non-fissile actinides which undergo neutron capture and transmutation during the subsequent three-year irradiation in the DB-MHR. At the same time, TF provides control and negative reactivity feedback to the reactor. After extraction of the spent TF, irradiated for three years, over 94% of Pu-239 and 53% of all actinides coming from LWRs waste will have been destroyed in the DB-MHR. In this paper we look at the operation conditions at equilibrium for the DB-MHR and evaluate fluxes and reactivity responses using state of the art 3-D Monte Carlo simulations.
One of the major benefits of the Gas Turbine-Modular Helium Reactor is the capability to operate with several different types of fuel; either Light Water Reactors waste, military plutonium or thorium represent valid candidates as possible types of fuel. In the present studies, we performed a comparison of various nuclear data libraries by the Monte Carlo Continuous Energy Burnup Code MCB applied to the Gas Turbine-Modular Helium Reactor operating on a thorium fuel. A thorium fuel offers valuable attractive advantages: low fuel cost, high reduction of actinides production and the possibility to enable the reactor to act as a breeder of fuel by the neutron capture of fertile 232Th. We evaluated the possibility to mix thorium with small quantities, about 3% in atomic composition, of 239Pu, 233U and 235U. The mass of thorium must be much larger than that one of plutonium or uranium because of the low capture cross section of thorium compared to the fission one of the fissile nuclides; at the same time, the quantity of the fissile isotopes must grant the criticality condition. These two simultaneous constraints force to load a huge mass of fuel in the reactor; consequently, we propose to allocate the fuel in TRISO particles with a large radius of the kernel. For each of the three different fuels we calculated the evolution of the fuel composition by the MCB code equipped with five different nuclear data libraries: JENDL-3.3, JENDL-3.2, JEFF-3, JEF-2.2 and ENDF/B.
Numerical applications implemented on the Monte Carlo method have developed in line with the increase of computer power; nowadays, in the field of nuclear reactor physics, it is possible to perform burnup simulations in a detailed 3D geometry and a continuous energy description by the Monte Carlo method; moreover, the required computing time can be abundantly reduced by taking advantage of a computer cluster. In this paper we focused on comparing the results of the two major Monte Carlo burnup codes, MONTEBURNS and MCB, when they share the same MCNP geometry, nuclear data library, core thermal power, and they apply the same refueling and shuffling schedule. While simulating a total operation time of the Gas Turbine-Modular Helium Reactor of 2100 effective full power days and a one-pass deep burn in-core fuel management schedule, we have found that the two Monte Carlo codes produce very similar results both on the criticality value of the core and the transmutation of the key actinides.
Reaction rates were measured by the foil activation technique to obtain neutron spectrum information in a subcritical core driven by an external neutron source. The experimental results are compared with Monte Carlo calculations in order to examine the capability of the Monte Carlo code MCNP together with ENDFB-6.8. JEFF-3.1.1 and CENDL-3.1 neutron cross section libraries to predict the neutron spectrum dependent reaction rates correctly in a subcritical core. The focus lies on fast neutrons. A discrepancy is found in the calculated-to-experimental values of the reaction rates and an inaccurate cross section is identified in CENDL-3.1.
The Radiation Test Facility (RTF) is under construction at the Institute for Theoretical and Experimental Physics to control the electronics under irradiation of particles that imitate cosmic rays (protons, carbon, aluminum, iron, tin, bismuth, and uranium). For the norms of radiation safety of personnel and users of the RTF to be observed, a local shielding and beam dump must be designed. Simulations of the dose rates around the designed shielding and beam dump are carried out in the present work.
This work presents results of activation-aided determination of threshold reaction rates (RRs) in 92 samples of 209Bi, natPb, 197Au, 181Ta, 169Tm, natIn, 93Nb, ,64Zn, 65Cu, 63Cu, 59Co, 19F, and 12C and in 121 samples of 27Al. All the samples are aligned with the proton beam axis inside and outside the demountable 92-cm-thick Pb target of 15-cm diameter assembled of 23 4-cm-thick discs. The samples are placed on 12 target disks to reproduce the long axis distribution of protons and neutrons. The target was exposed to an 800-MeV proton beam. The total number of protons onto the target was (6.0 ± 0.5) X1015. The RRs were determined by the direct gamma spectrometry techniques. In total, 1196 gamma spectra have been measured, and about 1500 RRs have been determined. The measured RRs were simulated by the MCNPX and SHIELD codes. A generally acceptable agreement of simulations with experimental data has been found.
Nuclide production cross sections measured at the Institute for Theoretical and Experimental Physics (ITEP) for the targets of (nat)Cr, (56)Fe, (nat)Ni, (93)Nb, (181)Ta, (nat)W, (nat)Pb, and (209)Bi irradiated by protons with energies from 40 to 2600 MeV were used to estimate the predictive accuracy of several popular high-energy transport codes. A general agreement of the ITEP data with the data obtained by other groups, including the numerous GSI data measured by the inverse kinematics method was found. Simulations of the measured data were performed with the MCNPX (BERTINI and ISABEL options), CEM03.02, INCL4.2 + ABLA, INCL4.5 + ABLA07, PHITS, and CASCADE.07 codes. Deviation factors between the calculated and experimental cross sections have been estimated for each target and for the whole energy range covered by our measurements. Two-dimensional diagrams of deviation factor values were produced for estimating the predictive power of every code for intermediate, not measured masses of nuclei targets and bombarding energies of protons. Further improvements of all tested here codes are recommended. In addition, new measurements at ITEP of nuclide yields from the (208)Pb target irradiated by 500-MeV protons are presented. A good agreement between these new data and the GSI measurements obtained by the inverse kinematics method was found.
This work presents the cross sections for radioactive nuclide production in Fe-56( p, x) reactions determined in six experiments using 300, 500, 750, 1000, 1500, and 2600 MeV protons of the external beam from the ITEP U-10 proton accelerator. In total, 221 independent and cumulative yields of radioactive residuals of half-lives from 6.6 min to 312 d have been obtained. The radioactive product nuclide yields were determined by direct gamma-spectrometry. The measured data have been compared with the experimental data obtained elsewhere by the direct and inverse kinematics methods and with calculation results of 15 different codes that simulated hadron-nucleus interactions: MCNPX (INCL, CEM2K, BERTINI, ISABEL), LAHET (BERTINI, ISABEL), CEM03 (.01,. G1,. S1), LAQGSM03 (.01,. G1,. S1), CASCADE-2004, LAHETO, and BRIEFF. Most of the data obtained here are in a good agreement with the inverse kinematics results and disprove the results of some earlier activation measurements that were quite different from the inverse kinematics measurements. The most significant calculation-to-experiment differences are observed in the yields of the A < 30 light nuclei, indicating that further improvements in nuclear reaction models are needed, and pointing out as well to a necessity of more complete experimental measurements of such reaction products.
The cross sections for nuclide production in thin Fe-56 and Cr-nat targets irradiated by 0.04-2.6-GeV protons are measured by direct gamma spectrometry using two gamma spectrometers with the resolutions of 1.8 and 1.7 keV for the Co-60 1332-keV gamma line. As a result, 649 yields of radioactive residual product nuclei have been obtained. The Al-27(p, x)Na-22 reaction has been used as a monitor reaction. The experimental data are compared with the MCNPX (BERTINI, ISABEL), CEM03.02, INCL4.2, INCL4.5, PHITS, and CASCADE07 calculations.
The cross sections for nuclide production in thin Nb-93 and Ni-nat targets irradiated by 0.04- to 2.6-GeV protons have been measured by direct gamma spectrometry using two gamma spectrometers with the resolutions of 1.8 and 1.7 keV in the Co-60 1332-keV gamma line. As a result, 1112 yields of radioactive residual nuclei have been obtained. The Al-27(p, x)Na-22 reaction has been used as a monitor reaction. The experimental data have been compared with the MCNPX (BERTINI, ISABEL), CEM03.02, INCL4.2, INCL4.5, PHITS, and CASCADE07 calculations.