ED Sciences Chimiques
Development of novel praseodymium oxide-based infiltrated oxygen electrodes for high-temperature electrolysers
by Clara PIOGER (ICMCB - Institut de Chimie de la Matière Condensée de Bordeaux)
The defense will take place at 14h00 - Amphithéâtre Institut de Chimie de la Matière Condensée de Bordeaux (ICMCB-CNRS) 87, Avenue du Docteur Schweitzer 33608 PESSAC France
in front of the jury composed of
- Jean-Marc BASSAT - Directeur de recherche - Institut de Chimie de la Matière Condensée de Bordeaux (ICMCB-CNRS) - Directeur de these
- Christel LABERTY-ROBERT - Professeure des universités - Laboratoire de Chimie de la Matière Condensée de Paris (LCMCP-CNRS) - Examinateur
- Cyril AYMONIER - Directeur de recherche - Institut de Chimie de la Matière Condensée de Bordeaux (ICMCB-CNRS) - Examinateur
- Clément NICOLLET - Chargé de recherche - Institut des Matériaux de Nantes (IMN-CNRS) - Examinateur
- Janick BIGARRé - Directeur de recherche - Commissariat à l'énergie atomique et aux énergies alternatives (CEA) - CoDirecteur de these
- Christophe TENAILLEAU - Maître de conférences - Centre Inter-universitaire de Recherche et d'Ingénierie des Matériaux (CIRIMAT-CNRS) - Rapporteur
- Patrice TOCHON - Directeur de recherche - Genvia - Examinateur
- Marlu Cesar STEIL - Ingénieur de recherche - Laboratoire Electrochimie et Physicochimie des Matériaux et Interfaces (LEPMI-CNRS) - Rapporteur
High‑temperature steam electrolysis is a promising route for the production of low‑carbon hydrogen. When implemented in solid oxide electrolysis cells (SOECs), this technology nevertheless sees its performance limited by the oxygen electrode, whose optimisation remains one of the main levers for further advancement. In this context, this doctoral work focused on the development of oxygen electrodes infiltrated with praseodymium oxide, materials that exhibit excellent oxygen exchange activity but intrinsically low electronic conductivity. To enhance their performance, several porous scaffolds were prepared in order to tailor their microstructure and transport properties. Praseodymium oxide infiltration was also implemented through different strategies to control their distribution, morphology and composition within the supporting scaffold. Electrochemical characterisations in symmetrical cells, relying on electrochemical impedance spectroscopy (EIS) and distribution of relaxation times (DRT) analysis, enabled the identification of the rate-limiting mechanisms and guided the joint design of the infiltrated electrodes. The electrode identified as the most promising was integrated into a full SOEC. At 750 °C, it achieved a current density of -1.57 A·cm-2 at the thermoneutral voltage, and under a flow rate of 18 NmL·min-1·cm-2 (corresponding to a steam conversion of 64%). These results demonstrate the potential of praseodymium‑based infiltrated electrodes for the development of more efficient SOEC systems. This work, carried out in collaboration with Genvia, the CEA and the ICMCB‑CNRS, provides new insights and improvements for the design of high‑performance infiltrated oxygen electrodes.
ED Sciences Physiques et de l'Ingénieur
Contribution of learning algorithms to optimize reconfigurable manufacturing systems
by Amine CHIBOUB (Laboratoire de l'Intégration du Matériau au Système)
The defense will take place at 14h00 - Amphi Jean-Paul DOM 351 Cours de la Libération, 33405 Talence Cedex, France.
in front of the jury composed of
- Rémy DUPAS - Professeur - Université de Bordeaux - Directeur de these
- Ali SIADAT - Professeur - ENSAM - Campus de Metz - Rapporteur
- Uday VENKATADRI - Professeur - Dalhousie Univeristy - Rapporteur
- Olga BATTAIA - Professeure - KEDGE Business School - Examinateur
Reconfigurable Manufacturing Systems aim to provide adaptability and responsiveness to changing production requirements. Within this context, the Facility Layout Problems play a central role, as layout decisions directly affect transportation cost, material flow efficiency, and operational performance. This thesis investigates the use of Reinforcement Learning and Deep Reinforcement Learning for solving static, stochastic, and dynamic Facility Layout Problems, including both weak and strong dynamic forms. A unified modelling and experimental framework is developed, integrating layout generation, Autonomous Mobile Robot trajectory construction, and discrete-event simulation. Static FLPs are first analysed to study representation effects, scalability, and algorithmic performance. Several value-based and policy-based Deep Reinforcement Learning methods are evaluated and compared with Simulated Annealing. The analysis is then extended to stochastic and dynamic environments through a novel Multi-task Deep Reinforcement Learning framework, termed Episode-level Task Switching with Shared Replay Buffer. The proposed approach enables a single network to learn across multiple transportation demands while preserving learning stability and task diversity. This approach provides a structured mechanism for handling demand variability within a unified reinforcement learning model and offers a novel reinforcement learning perspective on layout optimization in reconfigurable manufacturing contexts.