Numerical aand experimental study of nuclear fuel cracking and oxyde-cladding delamination
Sans objet.
Sans objet.
The development of nuclear reactors is part of a drive to improve safety, with, for example, the deployment of nuclear fuels with improved properties in terms of their behavior under accident conditions, the so-called E-ATF (Enhanced Accident Tolerant Fuel). Industrial operator FRAMATOME is developing the Cr2O3-doped UO2 fuel as E-ATF. However, very little data is available on the behavior of fission products from Cr-doped fuel under accident conditions.
This thesis proposes to develop a synthesis process for Cr-doped UO2 fuel that simulates irradiated fuel, in order to study the behavior of the elements (Cr and fission products) at different temperatures and under different oxygen partial pressures. The methodology is based on an experimental approach combining synthesis of model materials and in-depth chemical characterization, complemented by a theoretical approach (thermodynamic calculations) enabling thermal sequences to be dimensioned and the proposed reaction mechanisms to be confirmed.
The thesis will be carried out at CEA Cadarache (France), within IRESNE (Research Institute on Nuclear Systems for Low-Carbon Energy Production). The PhD student will be hosted in a laboratory dedicated to the study of uranium-based compounds in the Department of Fuel Studies (DEC). Depending on the densification processes chosen, experiments of various duration may be carried out in other laboratories in France or Europe.
The PhD student will have the opportunity to learn advanced characterization techniques for ceramic materials, gain access to experiments on large-scale instruments (synchrotron) and take part in exchanges with the academic community(CNRS, Universities, JRC). He or she will be able to promote his work through publications and participation in conferences.
At the end of this thesis, the PhD student will have acquired skills in materials science and solid state characterization that he/she will be able to use in various materials fields, as well as experience in the nuclear environment of interest to the nuclear industry.
In a context where the standards of high performance computing (HPC) keep evolving, the design of supercomputers includes always more frequently a growing number of accelerators or graphics processing units (GPUs) that provide the bulk of the computing power in most supercomputers. Due to their architectural departures from CPUs and still-evolving software environments, GPUs pose profound programming challenges. GPUs use massive fine-grained parallelism, and thus programmers must rewrite their algorithms and code in order to effectively utilize the compute power.
CEA has developed PLEIADES, a computing platform devoted to simulating nuclear fuel behavior, from its manufacture all the way to its exploitation in reactors and its storage. PLEIADES can count on an MPI distributed memory parallelization allowing simulations to run on several hundred cores and it meets the needs of CEA's partners EDF and Framatome. Porting PLEIADES to use the most recent computing infrastructures is nevertheless essential. In particular providing a flexible, portable and high-performance solution for simulations on supercomputers equipped with GPUs is of major interest in order to capture ever more complex physics on simulations involving ever larger computational domains.
Within such a context the present thesis aims at developing and evaluating different strategies for porting computational kernels to GPUs and at using dynamic load balancing methods tailored to current and upcoming GPU-based supercomputers. The candidate will rely on the tools developed at CEA such as the thermo-mechanical solver MFEM-MGIS [1,2] or MANTA [3]. The software solutions and parallel algorithms proposed with this thesis will eventually enable large 3D multi-physics modeling calculations of the behavior of fuel rods on supercomputers comprising thousands of computing cores and GPUs.
The candidate will work at the PLEIADES Fuel Scientific Computing Tools Development Laboratory (LDOP) of the department for fuel studies (DEC - IRESNE, CEA Cadarache). They will be brought to evolve in a multidisciplinary team composed of mathematicians, physicists, mechanicians and computer scientists. Ultimately, the contributions of the thesis aim at enriching the computing platform for nuclear fuel simulations PLEIADES.
References :[1] MFEM-MGIS - https://thelfer.github.io/mfem-mgis/[2]; Th. Helfer, G. Latu. « MFEM-MGIS-MFRONT, a HPC mini-application targeting nonlinear thermo-mechanical simulations of nuclear fuels at mesoscale ». IAEA Technical Meeting on the Development and Application of Open-Source Modelling and Simulation Tools for Nuclear Reactors, June 2022, https://conferences.iaea.org/event/247/contributions/20551/attachments/10969/16119/Abstract_Latu.docx, https://conferences.iaea.org/event/247/contributions/20551/attachments/10969/19938/Latu_G_ONCORE.pdf; [3] O. Jamond et al. «MANTA : un code HPC généraliste pour la simulation de problèmes complexes en mécanique », https://hal.science/hal-03688160
This thesis is part of the studies on pellet-cladding interactions in nuclear fuel rods used in NPP. The operator must ensure and demonstrate the integrity of rods in any situations. The mechanical stresses on the clad, the first safety barrier, are linked to the viscoplastic properties of the fuel. It is therefore necessary to know these behaviors and their evolution in operation.
The topic proposed will focuse on the characterization, in hot lab, of an irradiated fuel. One of the main difficulties is that the irradiated fuels in a reactor are multi-cracked, which makes their mechanical characterization particularly complex. However, an ongoing thesis (2022-25) has reached different steps: (i) the design of a specific thermomechanical testing machine, (ii) the partial qualification of this device, (iii) the implementation of tools and cracked sample extraction method, (iv) and a whole system model (digital twin).
The thesis will be the continuation of this work and will be built in four stages on three experimental platforms available at the CEA:
1. Getting the knowledge and improving existing digital and experimental tools,
2. Implementation of the device in hot-cell on an existing furnace,
3. Thermomechanical testing on irradiated fuel, a world first time in these conditions.
The tests will require dedicated post-processing based on simulation-experiments comparisons. Once the experimental base is sufficiently developed and interpreted, it will then be possible to confirm or revise the irradiated fuel behaviour laws. A link with the microstructure of materials could be addressed.
Throughout these stages, the PhD student will draw on skills and expertise of laboratories of the Fuel Research Department (IRESNE Institute, CEA Cadarache) and on a academic collaboration. This thesis also fits into the framework of the European project OPERA HPC and is a major issue.
The PhD student should have a strong taste for the experimental approach and some facilities for the use of digital tools. Knowledge of materials science is the minimum required. During the three years, the PhD student will improve his multiphysical skills in experimental device design and high-temperature material behavior, as well as in numerical simulation, which will facilitate his professional integration.
Uranium dioxide (UO2) fuels used in nuclear power plants are ceramics, for which solid-phase sintering is a key manufacturing step. The sintering stage involves heat treatment under controlled partial O2 pressure that induces coarsening of UO2 grain and then consolidation and densification of the material. Densification induces macroscopic shrinkage of the pellet. If the compact (powder obtained by pressing, manufacturing step before sintering) is highly heterogeneous density, a difference in densification within the pellet may occur, leading to differential shrinkage and the appearance of defects.
The PhD thesis aims at developing a Thermo-chemo-mechanical modeling of sintering to simulate the impact of the gas composition and properties on the pellet densification. This scale will enable us to take into account not only the density gradients resulting from pressing, but also the oxygen diffusion kinetics that have a local impact on the densification rate, which in turn impacts the transport process. Therefore, a multiphysics coupling phenomenon has to be modelled and simulated.
This thesis will be conducted within the MISTRAL joint laboratory (Aix-Marseille Université/CNRS/Centrale Marseille CEA-Cadarache IRESNE institute). The PhD student will leverage his results through publications and participation in conferences and will have gained strong skills and expertise in a wide range of academic and industrial sectors.
Carbon neutrality requires the development of low-carbon energy production systems, including nuclear power. The safety analysis of nuclear reactors requires the containment of fission products in all operating conditions, including the integrity of the first barrier made up of the fuel elements. For rod-type designs, which consist of a stack of fuel pellets in a metallic cladding, the mechanical behavior of uranium dioxide (UO2), pellet material, plays an important role in the cladding integrity assessment. During power transients, fuel-cladding contact increases mechanical stresses on the cladding, and fuel creep can accommodate swelling deformations, thereby reducing the stresses induced the cladding. One of the challenges is to understand and predict this phenomenon of UO2 creep, and in particular the mechanisms that drive it at the polycrystalline microstructural scale.
The main objective of the thesis will be to provide simulation methods and reference results in support of multi-scale modeling of the mechanical behavior of fuel at high-temperature, which is highly dependent on dislocation climbing mechanisms. To this end, a computational scheme will be developed, based on the coupling of a dislocation dynamics code (NUMODIS) and a code for solving nonlinear partial differential equations by FFT (AMITEX-FFTP), in order to describe the evolution of a dislocation microstructure (NUMODIS) under the effect of dislocation climbing induced by vacancy diffusion (AMITEX-FFTP). Simulations based on this approach will then be used to quantify the recovery of stored dislocation density with the effect of climbing mechanisms in different configurations (temperatures, stresses, etc.). Ultimately, this work will improve and validate the existing micromechanical modeling implemented in the CEA's PLEIADES simulation platform.
This thesis will be carried out under the joint supervision of the Département d'Etude des Combustibles (Institut IRESNE, CEA Cadarache) and the Département de Recherche sur les Matériaux et la Physico-chimie (Institut ISAS, CEA Saclay), and in collaboration with IM2NP at Aix Marseille Université. The thesis work will be carried out at the LM2C (Cadarache) and LC2M (Saclay) laboratories, in an environment that provides access to extensive expertise in multi-scale materials modeling. The research work will be promoted through publications and participation in international conferences in the materials field.
Context : Understanding the behavior of metals at high deformation rate [4] (between 104 and 108 s-1) is a huge scientific and technologic challenge. This irreversible (plastic) deformation is caused by linear defects in the crystal lattice : these are called dislocations, which interact via a long-range elastic field and contacts.
Nowadays, the behavior of metals at high deformation rate can only be studied experimentally by laser shocks. Thus, simulation is of paramount importance. Two approaches can be used : molecular dynamics and elastodynamics simulations. This thesis follows the second approache, based on our recent works [1, 2], thanks to which the first complete numerical simulations of the Peierls-Nabarro Equation (PND) [5] was performed. The latter equation describes phenomena at the scale of the dislocation.
PND is a nonlinear integrodifferential equation, with two main difficulties : the non-locality in time and space of the involved operators. We simulated it thanks to an efficient numerical strategy [1] based on [6]. Nevertheless, the current implementation is limited to one CPU –thus forbidding thorough investigations on large-scale systems and on long-term behaviors.
Thesis subject : There are two main objectives :
- Numerics. Based on the algorithmic method of [1], implement a HPC solver (High Performance Computing) for the PND equation, parallel in time and space, with distributed memory.
- Physics. Using the solver developped, investigate crucial points regarding the phenomenology of dislocations in dynamic regime. For exploiting the numerical results, advanced data-processing techniques will be employed, potentially enhanced by resorting to AI techniques.
Depending on the time remaining, the solver might be employed for investigating dynamic fractures [3].
Candidate profile : The proposed subject is multidisciplinary, between scientific computing, mechanics, and data-processing. The candidate shall have a solid background in scientific computing applied to Partial Differential Equations. Mastering C++ with OpenMP and MPI is recommended. Moreover, interest and knowledge in physics –especially continuum mechanics- will be a plus.
The PhD will take place at the CEA/DES/IRESNE/DEC in Cadarache (France), with regular journeys to Paris, for collaboration with CEA/DAM and CEA/DRF.
[1] Pellegrini, Josien, Shock-driven motion and self-organization of dislocations in the dynamical Peierls model, submitted.
[2] Josien, Etude mathématique et numérique de quelques modèles multi-échelles issus de la mécanique des matériaux. Thèse. (2018).
[3] Geubelle, Rice. J. of the Mech. and Phys. of Sol., 43(11), 1791-1824. (1995).
[4] Remington et coll., Metall. Mat. Trans. A 35, 2587 (2004).
[5] Pellegrini, Phys. Rev. B, 81, 2, 024101, (2010).
[6] Lubich & Schädle. SIAM J. on Sci. Comp. 24(1), 161-182. (2002).
The dynamic modeling of complex structures may require to take into account phenomena occurring at very different scales. However, a full refined modeling of this type of structure generally leads to prohibitive calculation costs. Multiscale modeling then presents an alternative solution to this problem, taking into account each phenomenon at the most appropriate scale.
We are interested here in slender structures subjected to mechanical stresses with frictional contacts between the structure and the retaining elements. The behavior of slender structures is in general represented by beam models, but accurately taking into account all the local contact/friction requires massive 3D models.
The originality of the work proposed here is to build an efficient multiscale and multimodel approach between beam and massive models which makes it possible to locally take into account the friction contact of slender structures. We are therefore moving towards the use of local multigrid (or multilevel) methods which naturally allow a non-intrusive multiscale coupling. The accuracy of these methods depends on the choice of transfer operators between scales, which must be carefully defined. It will also be necessary to take into account the incompatibility of the meshes supporting the models on the various relevant scales. Hence, the final model will consist in an enriched beam model taking into account local contact phenomena.
The developed model will be compared with experimental results obtained during test campaigns already carried out, and with reference numerical solutions, of increasing complexity, intended to finely validate the relevance of the proposed multiscale approach.
The strong potential of the targeted multiscale approaches, applied in this subject to the nuclear field, could be exploited by the candidate for other industrial issues such as those of aeronautics or the automotive industry.
This thesis will take place within the framework of the joint MISTRAL laboratory between the CEA and the LMA (Laboratoire de Mécanique et d’Acoustique) in Marseille. The PhD student will carry out the major part of his thesis within the CEA (IRESNE institut, Cadarache) in teams specialized in numerical methods and dynamic modeling of complex structures. The doctoral student will travel regularly to Marseille to discuss with the university supervisors.
Controlling the behavior of fission gases in nuclear fuel (uranium oxide) is an important industrial issue, as fission gas release or precipitation limit the use of fuels at extended burn-ups. The gas behavior is strongly influenced by the material’s microstructure evolution due to the aggregation of irradiation-induced defects (gas bubbles, dislocation loops and lines). Cluster dynamics (CD) (a kind of rate theory model) is relevant for modelling the nucleation/growth of the defect clusters, there gas content and the gas release. The current model has been parameterized following a multiscale approach, based on atomistic calculations (ab initio or empirical potentials). This model has been successfully applied to annealing experiments of UO2 samples implanted with rare gas atoms and has emphasized the impact of the irradiation damage on gas release. The aim of this PhD thesis is now to improve the model, particularly the damage parameterization, and to extend its validation domain through in depth comparison of simulation with a large set of recently obtained experimental results, such as gas release measurement by annealing of sample implanted in ion beam accelerator, bubble and loop observation by transmission electrons microscopy, and positron annihilation spectroscopy. This global analysis will finally yield an improved parameterization of the CD model.
The research subject combines a “theoretical” dimension (improving the model) with an “experimental” one (interpreting existing experiments or designing some new ones). The variety of techniques will introduce the candidate into the experimental world and thus broaden his scientific skills. The candidate will also have to manage collaborations for the experiments analysis, for the model development and for the specification of additional atomistic calculations. He will be at the interface of atomistic techniques, large-scale simulation and various experimental techniques. Therefore, he will develop a broad view of irradiation effects in materials and of multi-scale modelling in solids in general.
This project is an opportunity to contribute to the overall development of numerical physics applied to multi-scale modeling of materials, occupying a pivotal position and adopting a global viewpoint. This will allow experiencing oneself the way computed fundamental microscopic data finally helps solving complex practical issues.
Further readings:
Skorek et al. (2012). Modelling Fission Gas Bubble Distribution in UO2. Defect and Diffusion Forum, 323–325, 209.
Bertolus et al. (2015). Linking atomic and mesoscopic scales for the modelling of the transport properties of uranium dioxide under irradiation. Journal of Nuclear Materials, 462, 475–495.
The behavior laws used in numerical simulations describe the physical characteristics of simulated materials. As our understanding of these materials evolves, the complexity of these laws increases. Integrating these laws is a critical step for the performance and robustness of scientific computations. Therefore, this step can lead to intrusive and complex developments in the code.
Many digital platforms, such as FEniCS, FireDrake, FreeFEM, and Comsol, offer Just-In-Time (JIT) code generation techniques to handle various physics. This JIT approach significantly reduces the time required to implement new simulations, providing great versatility to the user. Additionally, it allows for optimization specific to the cases being treated and facilitates porting to various architectures (CPU or GPU). Finally, this approach hides implementation details; any changes in these details are invisible to the user and absorbed by the code generation layer.
However, these techniques are generally limited to the assembly steps of the linear systems to be solved and do not include the crucial step of integrating behavior laws.
Inspired by the successful experience of the open-source project mgis.fenics [1], this thesis aims to develop a Just-In-Time code generation solution dedicated to the next-generation structural mechanics code Manta [2], developed by CEA. The objective is to enable strong coupling with behavior laws generated by MFront [3], thereby improving the flexibility, performance, and robustness of numerical simulations.
The doctoral student will benefit from guidance from the developers of MFront and Manta (CEA), as well as the developers of the A-Set code (a collaboration between Mines-Paris Tech, Onera, and Safran). This collaboration within a multidisciplinary team will provide a stimulating and enriching environment for the candidate.
Furthermore, the thesis work will be enhanced by the opportunity to participate in conferences and publish articles in peer-reviewed scientific journals, offering national and international visibility to the thesis results.
The PhD will take place at CEA Cadarache, in south-eastern France, in the Nuclear Fuel Studies Department of the IRESNE Institute [4]. The host laboratory is the LMPC, whose role is to contribute to the development of the physical components of the PLEIADES digital platform [5], co-developed by CEA and EDF.
[1] https://thelfer.github.io/mgis/web/mgis_fenics.html
[2] MANTA : un code HPC généraliste pour la simulation de problèmes complexes en mécanique. https://hal.science/hal-03688160
[3] https://thelfer.github.io/tfel/web/index.html
[4] https://www.cea.fr/energies/iresne/Pages/Accueil.aspx
[5] PLEIADES: A numerical framework dedicated to the multiphysics and multiscale nuclear fuel behavior simulation https://www.sciencedirect.com/science/article/pii/S0306454924002408