Electrolyte motion in Li-ion cell: experimental and modelling approach
Li-ion batteries are an essential source of energy storage for decarbonising our energy use. The gradual introduction of silicon-based compounds into the negative electrode – traditionally made of graphite – appears to be the preferred route to increasing the technology’s energy density to between 300 and 350 Wh/kg. Silicon has a specific capacity of 3,580 mAh/g, which is almost ten times that of graphite. However, this enormous capacity is associated with a 280% volume expansion of the material, compared with just 10% for graphite during intercalation.
At the Li-ion cell level, this ‘breathing’ phenomenon occurs and must be managed. In particular, when a rigid cell is charged, the ‘breathing’ of the active materials expels the electrolyte from the electrochemical core. A lithium salt gradient may develop in the electrode plane in the electrolyte.
This thesis aims to gain a better understanding of this phenomenon by reproducing it experimentally in a specific cell, to establish measurements of lithium salt concentration, and to simulate the phenomenon in a P4D model based on an existing P2D model.
During his PhD, the candidate will develop a wide range of skills and know-how, in a unique combination.
Experimental and theoretical approach to select deep eutectic solvents for the delamination of photovoltaic panels
Multimaterial assemblies recycling such as photovoltaic (PV) panels is necessary to guarantee the circularity of materials. Today, PV panels recycling is mostly managed through pyrometallurgy which is limited in terms of materials recovery (polymers are incinerated and recovered silicium and silver yields are low) and emits noxious and greenhouse gases, such as HF generated by fluorinated polymer incineration.
Chemical delamination processes consist in the immersion of panels into a hot solvent to swell the polymeric tie-layers and separate the materials. These processes have the potential to largely improve recovery yields of all materials, but they are environmentally impacting due to the operating conditions and solvent use.
The solvent choice defines the impact of the process. In the current literature, this choice is mostly based on trial and error approaches that are tedious. Moreover, efficient solvents for PV delamination are generally toxic and flammable, which limits the environmental relevance of the processes.
Deep eutectic solvents (DES), which are the association of a hydrogen bond donor and acceptor (HBD and HBA), present interesting solvent properties for such processes. The use of organic hydrophobic molecules as HBA or HBD can impart the solvents with a dual hydrophobic and hydrophilic character that brings a capacity to interact with various molecules to the solvent, including polymers. As HBA/HBD combinations are numerous, it is important to develop a systematic methodology for their choice.
The present topic proposes a methodology based on a combination of theoretical modelling and experimental validation measurements to better understand the contributions of HBD/HBA structures on DES properties. The study will focus on the identification of the interactions between DES and EVA (ethylene vinyl acetate), a widely used polymer in PV panels, through the study of phenomenon of solubilization and solvolysis (covalent bond breakage) to identify the contributions to the delamination phenomenon. At the end of the thesis, the DES selection will be validated through the demonstration of the delamination of a PV panel and the analysis of the separated layers.
Design and implementation of innovative alloys for high-performance NdFeB magnets via the grain-boundary diffusion process
Permanent magnets based on rare earth elements (RE), particularly neodymium-iron-boron (Nd-Fe-B) magnets, are essential for high-performance motors and generators for use in electric vehicles and wind turbines. To increase their resistance to demagnetization, heavy rare earth elements (HRE) such as dysprosium (Dy) or terbium (Tb) are partially substituted for neodymium. However, these HREs, although crucial for magnetic properties, are critical materials with supply risks. Hence, the challenge is to reduce their content while maintaining performance, in order to meet the growing demand for high-performance magnets. In this context, the grain boundary diffusion (GBD) process of HREs emerges as the most effective solution. It enables the diffusion of HREs from the surface of sintered magnets toward grain boundaries and then into the magnetic phase grains. This localization of HREs at the grain peripheries improves resistance to demagnetization while preserving high remanence, thereby reducing the required amount of HREs.
The objective of this Ph.D is to provide a comprehensive understanding of the diffusion and localization mechanisms of HREs while limiting parasitic reactions that unnecessarily consume them. The doctoral work will combine a comprehensive experimental approach, including the development of HRE-RE-M (where M is a transition metal) alloys for GBD, the study of diffusion, the characterization of equilibrium phases, and detailed magnetic and thermodynamic analyses, particularly through an in-depth study of phase diagrams. This approach aims to define the optimal compositions and diffusion conditions to achieve the targeted magnetic performances.
Development of a new class of positive electrode materials for Na-ion batteries
Disordered Rock-Salts materials derived from Li2MnO2F have been extensively studied over the past decade as potential high-energy-density positive electrode materials for Li-ion batteries. At the CEA, our experience with these materials, which are highly sensitive to usage conditions, has enabled us to surpass state-of-the-art performance in full cells incorporating them. Surprisingly, very few studies have been reported on their sodium equivalents and their use in Na-ion batteries, despite their composition being low in critical and expensive materials, making them particularly suitable for this technology. This is likely due to their exacerbated instability compared to the lithiated phases. In a previous internship, we were able to develop synthesis and electrode implementation protocols that allowed us to achieve specific capacities exceeding 220 mAh·g?¹ with a material of this type, which could make it competitive with a LiFePO4-based Li-ion technology. The thesis will therefore focus on optimizing the material's composition and understanding the influence of the synthesis protocol in terms of structure (X-ray diffraction analysis) and material surface state (spectroscopy).
Selective precipitation of zirconium: an innovative step for future nuclear reprocessing processes
The recycling of spent nuclear fuel is a strategic mission for the CEA aimed at improving the sustainability of the fuel cycle and reducing the impact of radioactive waste. In this context, zirconium has been identified as one of the most problematic fission products in the PUREX process due to its complex chemistry in nitric medium. It is involved in co-extraction phenomena, the formation of interfacial cruds, the degradation of contactor hydrodynamic performance, and the reduction of decontamination factors.
The objective of this PhD thesis is to develop a selective zirconium precipitation step directly within the dissolution liquor prior to liquid-liquid extraction. The central scientific challenge is to understand how to control the selectivity of this precipitation in a highly nitric and complex medium.
The proposed approach is based on the use of a precipitating agent adhering to the CHON principle. The work will first focus on single-metal systems and subsequently on mixtures representative of dissolved fuel, in order to identify competitive phenomena between elements.
The expected results pertain to the understanding of precipitation mechanisms, the identification of parameters governing process selectivity, and the demonstration of quantitative zirconium removal without significant actinide loss. This approach could lead to a more robust PUREX process, better suited to high-burnup fuels. The PhD candidate will develop expertise at the interface of coordination chemistry, separation chemistry, and nuclear process engineering, within a research environment of high scientific and industrial impact featuring an extensive array of characterization techniques.
Development of Monte Carlo methods for reactor kinetics analysis: application to the ZOE pile
The Monte Carlo code TRIPOLI-5 is under development with the mid-term aim of simulating nuclear reactor behavior under both steady-state and transient conditions. This is achieved by leveraging high-performance computing (HPC) architectures and natively integrating physical feedback effects. Methods for addressing non-stationary transport problems are categorized into three main approaches: the point kinetic approximation, the determination of the asymptotic behaviour, and solving the exact kinetic equations which allows for a comprehensive characterization of core behavior. In this PhD thesis, we propose to evaluate the performance of these three approaches within the HPC context of TRIPOLI-5 and subsequently implement them by leveraging and generalizing the strategies derived from previous work. In particular, an additional challenge in this thesis will involve addressing photonuclear reactions, which induce a coupling between neutron and photon tranport equations. The developments achieved during this thesis will then be applied to the ZOE pile, France's first nuclear reactor, built by the CEA at the Fontenay-aux-Roses site, which achieved criticality on December 15, 1948. A set of experimental measurements available in historical reports will provide a validation basis for the reactor kinetics analysis methods developed in TRIPOLI-5. Since the ZOE pile was heavy-water moderated, photonuclear reactions on deuterium must be explicitly accounted for to accurately determine the evolution of reactor kinetics.
Monte Carlo Methods for Generating Diffusion Data for Light Water Reactor Core Calculations
Neutronic calculations for light water reactors (LWRs) require determining the power distribution in the core, a computationally expensive task. To address this challenge, deterministic simulations are traditionally performed in two steps. First, isolated 2D calculations are carried out: fuel assembly simulations using transport theory with reflective boundary conditions. Then, full 3D simulations are performed: these use homogenized few-group cross sections generated in 2D to model the entire core via diffusion theory.
However, the deterministic codes used for the first step rely on approximations (e.g., energy dependence of cross sections, treatment of anisotropic scattering), which limit the accuracy of the results. In contrast, Monte Carlo methods offer an approximation-free alternative capable of producing reference-level results.
This thesis aims to develop and qualify Monte Carlo methods for generating diffusion coefficients and multi-group cross sections in the TRIPOLI-5 code (co-developed by the CEA and the ASNR). These data will then be used as input for deterministic codes to perform full-core neutronic simulations.
Few-group diffusion coefficients, which are essential for diffusion solvers, are not available in nuclear data libraries. Current deterministic methods rely on approximations, introducing uncertainties. This thesis will propose solutions to compute these coefficients with high fidelity in a continuous-energy Monte Carlo context.
The thesis will explore various approaches, including methods for calculating diffusion coefficients (e.g., the Cumulative Migration Area method) and leakage models. For each approach, the accuracy of the results will be evaluated with the goal of industrial implementation in TRIPOLI-5.
The implementation of these methods in TRIPOLI-5 will improve the accuracy of diffusion data for downstream solvers and overcome the limitations of current deterministic toolchains. Ultimately, this thesis will contribute to establishing TRIPOLI-5 as a reference platform for generating high-fidelity data for light water reactor core calculations, while paving the way for industrial integration of these methods.
Modeling of flow regime transitions in horizontal and inclined pipes: dynamic aspects and scalability
Understanding fluid flows in the piping systems of industrial facilities is of crucial importance for their safety and operation. In accidental scenarios studied for the primary circuit of a pressurized water reactor (PWR), two-phase flows within pipes may occur, exhibiting a wide variety of regimes depending on the encountered conditions. The transitions between these different regimes (stratified, annular, slug, bubbly, etc.) under such conditions are the subject of extensive experimental and numerical research. In particular, system-scale thermohydraulic simulation codes, such as CATHARE, must accurately compute these regime transitions.
This thesis aims to improve the prediction at the system scale of two-phase flow regime transitions in horizontal or inclined pipes. The project is structured around three main objectives. The first objective seeks to better account for scaling and fluid transposition issues by investigating models that influence the stratified/intermittent and stratified/annular transitions, including interfacial friction. The second objective focuses on predicting the transition between bubbly and intermittent regimes by revisiting the modeling of bubble segregation and mixing, and by incorporating new experimental data. The third objective, directly related to the previous one, addresses the transient effects of regime transitions, particularly by exploring the use of interfacial area transport equations.
The software tools used will include the CATHARE code, while experimental data will be provided from CEA facilities and the literature.
Effects of turbulence and mixing on the interactions between a hydrogen-air flame and a spray
This project studies the interactions between a hydrogen-air flame and a water spray in the context of a severe accident (e.g., hydrogen release in a nuclear containment). The goal is to analyze how sprays influence flame propagation and generated overpressures, which can either inhibit combustion or accelerate it through turbulence and obstruction mechanisms.
The experimental program is divided into:
1/ Characterization of spray-induced turbulence (1 or 2 nozzles, in open air or confined vessel) using PIV (Particle Image Velocimetry).
2/ Study of combustion impact: comparison of flame dynamics (H2/air) with and without spray, measuring propagation speeds and overpressures.
3/ Numerical modeling: adaptation of CFD codes (e.g., WALL-HY) to simulate flame-droplet interactions, building on previous work (G. Gai’s thesis).
This work aims to provide reference data and a predictive model to assess hydrogen-related risks in the presence of spray systems.
Glass for the Photovoltaic Power Plants of Tomorrow: links between ionic transport, high voltage, and aging
This thesis focuses on the aging of glass used in photovoltaic modules, a key challenge in the energy transition. High-power solar power plants operate at increasingly high voltages, which accelerates the migration of sodium from the glass into the cells—a phenomenon exacerbated by ambient humidity, known as potential-induced degradation (PID). This phenomenon limits the service life of the modules and restricts their use in medium- and high-voltage series installations. Recent studies show that adding other cations to the glass composition can reduce this migration, but the fundamental mechanisms of Na? ion transport remain poorly understood.
The objective of this thesis is to characterize these mechanisms in detail as a function of intrinsic parameters (composition, glass structure) and extrinsic parameters (voltage, humidity, temperature), first on the glass alone, then on modules fabricated with these glasses. The approach will combine experiments and modeling: preparation and physicochemical characterization of the glasses; high-temperature NMR (²³Na, ²?Si, ¹7O, ²7Al) to probe the local environments of the cations and activation energies; and molecular dynamics simulations (with and without an electric field, including on protonated systems) to study sodium mobility and directional ion transport.
The approach will follow a progressive increase in complexity: glass alone, then bifacial glass-encapsulant materials, and finally complete modules subjected to standardized aging tests used in the PV industry. We are seeking a candidate with a background in materials science, ideally with expertise in atomistic modeling. The dissertation will draw on collaborations between the CEA-Marcoule, INES (for module fabrication), CEA-Saclay, and CEMHTI (NMR), as well as Sorbonne University (simulation).