Design of asynchronous algorithms for solving the neutron transport equation on massively parallel and heterogeneous architectures
This PhD thesis work aims at designing an efficient solver for the solution to the neutron transport equation in Cartesian and hexagonal geometries for heterogeneous and massively parallel architectures. This goal can be achieved with the design of optimal algorithms with parallel and asynchronous programming models.
The industrial framework for this work is in solving the Boltzmann equation associated to the transportof neutrons in a nuclear reactor core. At present, more and more modern simulation codes employ an upwind discontinuous Galerkin finite element scheme for Cartesian and hexagonal meshes of the required domain.This work extends previous research which have been carried out recently to explore the solving step ondistributed computing architectures which we have not yet tackled in our context. It will require the cou-pling of algorithmic and numerical strategies along with programming model which allows an asynchronousparallelism framework to solve the transport equation efficiently.
This research work will be part of the numerical simulation of nuclear reactors. These multiphysics computations are very expensive as they require time-dependent neutron transport calculations for the severe power excursions for instance. The strategy proposed in this research endeavour will decrease thecomputational burden and time for a given accuracy, and coupled to a massively parallel and asynchronousmodel, may define an efficient neutronic solver for multiphysics applications.
Through this PhD research work, the candidate will be able to apply for research vacancies in highperformance numerical simulation for complex physical problems.
One-sided communication mechanisms for data decomposition in Monte Carlo particle transport applications
In the context of a Monte Carlo calculation for the evolution of a PWR (pressurized water reactor) core, it is necessary to compute a very large number of neutron-nucleus reaction rates, involving a data volume that can exceed the memory capacity of a compute node on current supercomputers. Within the Tripoli-5 framework, distributed memory architectures have been identified as targets for high-performance computing deployment. To leverage such architectures, data decomposition approaches must be used, particularly for reaction rates. However, with a classical parallelization method, processes have no particular affinity for the rates they host locally; on the contrary, each rate receives contributions uniformly from all processes. Access to decomposed data can be costly when it requires intensive use of communications. Nevertheless, one-sided communication mechanisms, such as MPI RMA (Message Passing Interface, Remote Memory Access), make these accesses easier both in terms of expression and performance.
The objective of this thesis is to propose a method for partial data decomposition relying on one-sided communication mechanisms to access remotely stored data, such as reaction rates. Such an approach will significantly reduce the volume of data stored in memory on each compute node without causing a significant degradation in performance.
HPC two-phase simulations with lattice Boltzmann methods and adaptative mesh refinement
CEA/STMF develops computational fluid dynamics (CFD) codes in thermohydraulics that aim to quantify mass and energy transfers in nuclear cycle systems such as reactors and management devices of radioactive wastes. This thesis focuses on Lattice Boltzmann Methods (LBM) adapted to Adaptive Mesh Refinement (AMR) inside a generic computing environment based on Kokkos and executable on multi-GPU supercomputers. The proposed work consists in developing LB methods in the Kalypsso-lbm code to simulate coupled partial differential equations (PDEs) modelling incompressible two-phase and multi-component flows such as those encountered in downstream cycle devices. Once the developments have been completed, they will be validated with reference solutions. They will allow a comparison of various interpolation methods between blocks of different sizes in the AMR mesh. A discussion will be held on the refinement and de-refinement criteria that will be generalized for these new PDEs. Finally, benchamrks of performance will quantify the contribution of AMR for 3D simulations when the reference simulation is performed on a static and uniform mesh. This work will use supercomputers which are already operational (e.g., Topaze-A100 from CEA-CCRT), as well as the future exascale supercomputer Alice Recoque depending on the progress of its installation.
Development of extracting systems for the isotopic enrichment of chlorine
Chlorine (Cl) is naturally composed of 76% 35Cl, which through neutron capture forms 36Cl, a long-lived gamma emitter (t1/2 = 301 000 years), and sulfur 36S, which accelerates corrosion phenomena, and 24% 37Cl with a drastically lower neutron capture section. A supply of 37Cl is therefore necessary in order to operate these reactors. Techniques currently exist that enable the enrichment of chlorine, such as ultracentrifugation, liquid-phase thermal diffusion, or laser isotope separation. The enrichment of chlorine by liquid-liquid extraction technics has been recently developed within CEA. The objective of the thesis is to identify and implement chemical systems allowing the 37Cl enrichment by a separative chemistry process. The thesis subject aims to identify on the basis of literature data initially, the families of ligands and, within these families, the best candidates for the 37Cl enrichment. Next, the synthesis and purification of the selected molecules will be carried out in the laboratory. Finally, the enrichment properties of the successfully synthesised ligands will be evaluated by separative chemistry, by quantification of chlorine isotopes using Inductively coupled plasma mass spectrometry (ICP-MS).
The thesis will be carried out at the recycling and energy recovery processes laboratory (LRVE) at the CEA in Marcoule.
The ideal candidate will be a Master's student in their second or third year of engineering school, studying chemistry, organic chemistry or analytical chemistry. The multidisciplinary nature of the skills acquired and the rigour developed by the student during the experiments undertaken will be valuable assets for the future PhD student.
Modeling the impact of defects in Steel–Concrete Structures. Identification of critical defects through metamodeling and optimization algorithms
To meet growing constructability challenges, steel–concrete (SC) structures are emerging as a promising alternative to conventional reinforced concrete structures. These elements are composed of infill concrete, two external steel plates, and steel shear studs that ensure composite action. While such structures present a clear interest due to their overall mechanical behavior, the presence of the steel plates prevents visual inspection of the concrete casting quality. It is therefore essential to characterize the impact of possible defects. This is the context of the proposed PhD research. Building upon recent results obtained in the laboratory, the goal is to develop a numerical framework to account for defects in steel–concrete structures. The thesis will be structured in several stages: validation of a modeling strategy for the mechanical behavior of defect-free SC structures, introduction of defects in the simulations and assessment of the applicability of the numerical approach, development of a metamodel and sensitivity analysis, and identification of critical defect configurations through optimization algorithms. One of the operational objectives of this doctoral work is to provide a tool capable of identifying critical defect configurations (size, position, and number) with respect to a given target quantity of interest (such as loss of strength or reduction in average stiffness). The research will therefore rely on the use and further development of state-of-the-art numerical tools in the fields of finite element modeling, optimization techniques, sensitivity analysis, and metamodeling. The thesis will be carried out within a rich collaborative environment, notably in partnership with EDF.
Effects of structural heterogeneities on air flow through reinforced concrete walls
The containment building represents the third barrier to confinement in nuclear power plants. Its role is to protect the environment in the event of a hypothetical accident by limiting releases to the outside. Its function is therefore closely linked to its tightness. Traditionally, the estimation of the leakage rate is based on a sound knowledge of transfer properties (such as permeability), combined with a chained (thermo-)hydro-mechanical simulation approach. While the mechanical behavior of the structure is now broadly well understood, progress is still needed in the comprehension and quantification of fluid flow. This is particularly true in the presence of heterogeneities (cracks, honeycombs, construction joints, reinforcements, cables, etc.), which represent situations that can locally disturb permeability. This is the context of the present PhD topic.
The work will consist, through a methodology combining experimental testing and numerical simulation, in improving the representation of fluid flow by explicitly accounting for the impact of heterogeneities. An initial analysis will define an experimental plan, which will then be carried out. The results will be analyzed in order to empirically characterize the influence of each type of heterogeneity tested on transfer properties. A simulation approach, exploiting the experimental findings, will then be developed using finite element and discrete methods. Finally, the applicability of the methodology to a real-scale structure will be assessed, while explicitly accounting for uncertainties regarding the presence and impact of such heterogeneities (probabilistic approach).The PhD will therefore rely on state-of-the-art experimental and numerical tools and methods, and will be conducted in a rich collaborative context (CEA, ASNR, EDF).
Study of plutonium oxalate formation mechanisms – Application to molten salt reactors
Molten salt reactors (MSRs) offer a promising alternative for sustainable nuclear energy production, thanks to their intrinsic safety and their ability to close the nuclear fuel cycle, notably through the use of a fast neutron spectrum. This type of reactor can use liquid chloride salts containing plutonium and other actinides as fuel. As part of the development of this nuclear pathway, the CEA supports the development of a PuCl3 production process. The synthesis of this chloride has already been carried out at small scale at the CEA and elsewhere in the world. Several starting materials can be used for the synthesis of the trichloride, notably plutonium metal, plutonium oxide and plutonium oxalate. The most industrially promising synthesis route is the oxalate route, because it can be transferred to the equipment already present at the La Hague site. This process consists of converting the oxalate into plutonium chloride via a gas–solid reaction with a chlorinating agent, such as HCl for example. However, the reaction mechanism and the decomposition of the oxalate in a chlorinated environment are still poorly understood. A detailed understanding of this transformation would make it possible to optimize operating conditions and facilitate the scale-up of this synthesis. The topic will initially focus on determining the reaction mechanism of Ce oxalate (a surrogate for Pu) to the chloride. Small-scale studies will be performed to identify the various reaction intermediates using analytical techniques such as X-ray diffraction (XRD), thermogravimetric analysis / differential thermal analysis (TGA/DTA) and analysis of the gases produced during the reaction. The kinetics as well as the enthalpy changes will also be studied in order to obtain key data for modelling a large-scale process. Subsequently, an optimization of the PuCl3 synthesis at the scale of a few tens of grams will be carried out. These studies will first be conducted under non-radioactive conditions on a surrogate to validate the experimental approach, before being transposed to radioactive conditions.
Effects of alpha decay on the alteration of nuclear glasses: simulation, understanding, and consideration in geochemical models
This Ph-D at the CEA on the alteration of nuclear glass is central to the challenges of sustainable radioactive waste management. The doctoral student will acquire expertise in materials and modeling, paving the way for exciting careers in research, engineering, or the nuclear industry. In deep geological storage, contact with groundwater can cause glass alteration, which is the main source of radionuclide release. The CEA is developing a multi-scale model that needs to be adapted to take into account the effects of glass self-irradiation. The aim of the thesis is to identify the mechanisms modified by irradiation and to parameterize the model. The doctoral student will conduct controlled irradiation experiments on non-radioactive glasses and compare them to ²44Cm-doped active glass. The structural and physicochemical changes induced will be characterized using various techniques (Raman, IR, NMR, SEM, TEM, DSC, etc.). Targeted alteration tests will be used to observe the impact of the level of damage on the kinetics of alteration. The results will be used to adjust and validate the predictive model under conditions representative of geological storage. The work will be carried out both in an active environment (shielded cells) and in an inactive laboratory. An M2 internship is available on the same subject. Profile: M2 or materials engineer, physical chemistry.
Optimization of the conditions for the electrolytic synthesis of metallic uranium
Reprocessed uranium (RepU), derived from the reprocessing of spent nuclear fuel, represents a material whose reuse in power plants would allow for the sustainable management of energy resources. Accordingly, the CEA is supporting the nuclear industry to evaluate the feasibility of enriching this RU via the laser route. This technology requires, as a process input, uranium in the form of a metallic alloy. Consequently, an upstream process for the synthesis of metallic uranium must be developed and optimized to build a sovereign RepU sector.
One of the routes under study for synthesizing metallic uranium is the electrolysis of uranium oxide, previously dissolved in high-temperature molten fluoride salt media. This synthesis, which was previously implemented in the United States using the aluminum synthesis process, now requires a re-appropriation and optimization of experimental conditions.
In a first phase, the PhD student will conduct a systematic study of the electrolyte, in order to understand the influence of key parameters—salt composition, temperature range, redox environment, material compatibility, and oxide solubility—on the behavior of the electrolysis bath. For each parameter, targeted tests will be conducted: thermochemical characterization of the salt (melting point, volatility, purification, etc.), evaluations of the kinetics and the solubility limit of uranium oxide in the bath (a crucial point of the process), electrochemical tests aimed at identifying redox systems of interest, as well as studies on the resistance of materials when in contact with the molten salt and liquid metal. All of these investigations will make it possible to define the optimal experimental conditions for the controlled implementation of metal synthesis by oxide electrolysis.
In a second phase, once these conditions are established, the work will focus on the formation of the metal at the electrode, its recovery, and its characterization. The quantity and quality of the metal produced after electrolysis will be the major criteria for validating the selected experimental parameters.
All acquired data will be utilized for the design of pilot and industrial scale electrolyzers, and will feed into future digital models that will be developed. The results obtained may be the subject of presentations at international conferences and publications.
These studies will be carried out at the laboratory scale using active material, with work phases on simulants to grasp the implementation of the process and scaling up. The host laboratory, which operates in both these environments, specializes in the implementation of thermal processes and pyrochemical tests.
The candidate should ideally have a Master 2 or engineering school degree in chemistry or physics.
At the end of this thesis work, the PhD student will have acquired expertise in experimental techniques related to metallic synthesis by electrolysis, from the design of electrochemical devices to the multi-scale characterization of materials. Furthermore, their involvement in a sovereign project focused on strategic metals will open up numerous employment prospects in academic research or industrial R&D, both in the nuclear sector and in other fields of chemistry and materials.
Impact of fission products and microstructure on the thermophysical properties of LWR (U,Pu)O2-x fuel
In France, mixed oxide fuel (MOX, (U,Pu)O2) is currently deployed in several pressurized water reactors (PWRs) operated by EDF. To ensure continued low-carbon electricity production, a broader use of MOX fuel across the French nuclear fleet is expected to become essential in the near future. During reactor operation, U1??Pu?O2?? fuels undergo significant changes in their physical properties and microstructure, primarily due to the accumulation of dozens of lighter elements generated by plutonium’s fission, commonly referred to as fission products (FPs). Because of the high radiotoxicity of irradiated fuel, surrogate materials known as SIMMOX have been developed. In a previous PhD project, we established a synthesis route enabling the production of SIMMOX doped with up to twelve fission products, successfully reproducing the microstructure of irradiated PWR MOX fuel.
To maintain an adequate margin to fuel melting during irradiation, it is crucial to understand how the thermophysical and thermodynamic properties of MOX fuel evolve under these conditions. This PhD project aims to measure these properties on a representative MOX composition currently used in EDF reactors. The key properties of interest include thermal conductivity, heat capacity, and melting temperature. These measurements will be carried out at the JRC-Karlsruhe (Germany) during a research stay of approximately 12 months. Subsequently, the samples will be returned to CEA-Marcoule, where the impact of high-temperature exposure on actinide and fission product speciation, as well as on the microstructural evolution of the MOX fuel, will be investigated. In parallel, the experimental work will be complemented by thermodynamic modeling using the CALPHAD approach, in order to identify the mechanisms and phase equilibria governing high-temperature behavior during property measurements.