This thesis aims to develop new theoretical and computational tools to model light–matter interactions in nanophotonic systems while fully accounting for dissipation, which is unavoidable in real devices but remains difficult to describe in conventional quantum approaches. It will rely on a non-Hermitian formalism based on quasinormal mode theory to investigate resonant interactions between photonic or plasmonic resonators and quantum emitters. The goal is to provide a more realistic and physically transparent framework for the design of future nanophotonic and quantum technologies, including single-photon sources, quantum sensors, and strongly coupled light–matter systems.

A detailed description of the position is attached; it includes the application requirements and guidelines.

In the context of the ecological transition and efforts to reduce the carbon footprint of materials, the incorporation of bio-based components into composites represents a major strategic opportunity.
Among these solutions, the use of plant-based fibers (such as flax or hemp) as a substitute for synthetic reinforcements in composites is attracting growing interest. Indeed, these plant fibers exhibit competitive tensile properties, with specific stiffness levels comparable to those of glass fibers.
However, these natural fibers also have defects (kink bands, etc.) as well as natural variability in structure and morphology that can affect their fracture properties and the performance of parts in service. A thorough understanding of the damage mechanisms in plant fibers is therefore necessary to enable their use in high-value-added applications. A few studies in the literature have examined damage at the plant fiber scale, using in situ mechanical testing under X-ray tomography.
However, a number of questions remain unanswered to this day: where does damage initiation occur within the fiber? What are the characteristic lengths involved during fracture? How does the composite matrix interact with the fiber? To answer these questions, a number of obstacles must be overcome regarding (i) the manage of fracture tests down to the single-fiber scale, (ii) the characterization of damage at these small scales, and (iii) the identification of relevant parameters and lengths to be incorporated into fiber-scale damage models.

Applicant profile
A student currently enrolled in an engineering school or a Master’s 2 program in mechanical engineering or materials science, with a strong interest in experimentation, image analysis, and/or numerical modeling. Candidates with a background in both experimental and numerical methods will be particularly sought after. A specialization in bio-based composites, material damage, or digital image correlation would be a plus. Knowledge of X-ray tomography is also desirable but not a strict requirement.

Descriptif de la thèse : 

La thèse portera sur la génération de cavéoles, nano-objets biologiques calibrés et produits à façon pour le développement de micro dispositifs de détection et quantifications d’objets eucaryotes et procaryotes. Les cavéoles sont des nano-objets pouvant être utilisés en recherche fondamentale comme appliquée, notamment pour la caractérisation des protéines membranaires, cibles de plus de 70% des médicaments chimiques actuels et pour la production de biomédicaments. Ces nano-objets sont obtenus par expression de la protéine cavéoline dans la bactérie Lactococcus lactis. Ils présentent des tailles comprises entre 30 et 50 nm, équivalentes à celles des exosomes ou petits vésicules extracellulaires (EVs) eucaryotes et BEVs (EVs bactériennes). Il est nécessaire de mieux connaître et d'optimiser la production de ces cavéoles, mais aussi de simplifier l'isolement de celles ci, en évitant le gradient de sucrose qui permet, actuellement, d’isoler ces éléments des autres vésicules membranaires.

Profil demandé :

Le/la doctorant(e) devra posséder un Master 2 ou diplôme d’ingénieur en Biologie et Santé, Biochimie, Biotechnologie avec un intérêt pour les caractérisations de nano-éléments par des techniques biophysiques, optiques et en flux.

The rapid integration of Large Language Models (LLMs) into clinical workflows presents a fundamental tension: while generative agents offer unprecedented capabilities for processing unstructured medical narratives, their stochastic nature conflicts with the deterministic safety requirements of medical practice. Within the European and French regulatory landscape—defined by the EU AI Act, GDPR, and the HDS (Hébergeur de Données de Santé) certification—deploying clinical AI is simultaneously a performance challenge and a sovereignty mandate. Healthcare institutions require systems that operate on-premise while providing rigorous, auditable decision support.

The objective of this thesis is to propose a neuro-symbolic middleware that prevents direct LLM interpretation of raw EHR or medical data. In this design, LLM agents interact exclusively through a Clinical Semantic API backed by a formal Medical Knowledge Graph and governed by Semantic Contracts.

A detailed description of the position is attached; it includes the application requirements and guidelines.

The aim of this thesis in automatic control is to use the port Hamiltonian framework to develop a model and a robust control for systems actuated by dielectrophoresis for applications linked to biological cells manipulation and characterization.

This work will be carried out in the MOCOPHYS team at the AS2M department at the FEMTO-ST institute.

A detailed description of the position is attached; it includes the application requirements and guidelines.

The objective of this thesis is to develop generalized protocols for calibrating robotic micromanipulators, based on the use of microforce sensors together with joint encoders. The proposed strategies shall integrate micro-force sensors as an exteroceptive measurement technology, which is to measure forces that are external to the robot (palpation). The force measurement can either be done using on-board micro-force sensors or by scattering multiple sensors around the robot’s workspace. It would also be possible to use indirect proprioceptive measurements (robot joint measurements) synthesized using observers. All these measures shall be fused together in a coherent matter to correct the static robotic model of the micro-manipulator and to develop dynamic control strategies to reach precise contact positions for the calibration procedure and obtain a submicrometric level of precision. The microforce sensors used for this purpose may be based on one of many different measurement technologies used for micromanipulation, such as piezoelectric sensors, piezoresistive force sensing or capacitive measurement among others.

The final goal of developing such procedures is on the one hand, to significantly improve the precision for the positioning and rotational control of these systems, and on the other hand, to improve the accuracy of microforce measurements using micromanipulators regardless of the direction of the force or the robot’s configuration. The fulfilment of the latter is of particular importance, so that the procedure can be used in various applications such as the characterization of mechanical properties of microstructures under different loading conditions like compression, tension and shear forces.

This PhD thesis aims to develop a new generation of soft microrobots to address major challenges in healthcare and advanced manufacturing. It leverages cutting-edge technologies such as 4D printing and two-photon stereolithography to create smart, responsive 3D microrobots. These light-activated microrobots, integrated at the tip of optical fibers, will enable precise, multi-degree-of-freedom actuation in extremely confined spaces.

The hired PhD student will actively participate in the newly launched national research network in miniature robotics (PEPR Miniro "miniature robotics", 2025-2032) and will have full access to equipements and facilities notably provided by the CMNR (Center for Micro and Nano Robotics – National infrastructure for cutting-edge robotics research) and the Mimento clean-room technological facility (National network of major technology hubs for Basic Technological Research).

Project description

At the cutting edge of the Second Quantum Revolution, the scaling of entanglement schemes has become a central challenge in Quantum Sciences and Technologies (QSTs) in general and in quantum information processing and quantum metrology in particular.

The goals of this PhD project are to prepare and stabilize highly entangled Dicke states in a large ensembles of NV centers placed in microwave cavities and to explore their utilization as an entanglement source for quantum information processing tasks or for quantum metrology. This involves triggering and controlling superradiance (SR) in the microwave domain (~3 GHz) and quantifying the collective quantum properties using dispersive measurements. 

The outcomes will advance our understanding of quantum coherence in solid-state systems, contribute to the development of scalable quantum technologies and provide innovative tool for exploring collective effect in Cavity QED.

Key Responsibilities:

• Simulation and Optimization: Design and optimize microwave resonator able to trigger the superradiance of NV centers.
• Experimentation: Develop the experimental setup and protocols to trigger and stabilize the collective superradiance in NV centers placed in microwave cavities.
• Quantum Characterization: Quantifying the Dicke states properties by mean of both local measurements and advanced dispersive measurements techniques.
• Theoretical Collaboration: Collaborate with theorists to model the dynamics of superradiance.
• Team Collaboration: Work with the rest of the team to investigate applications in quantum metrology and quantum information processing.
• Dissemination: Publish your findings in high-impact journals and present your work at international conferences.

The objective of this Ph.D. thesis is twofold: first, to develop a nonlinear infinite-dimensional port- Hamiltonian (pH) model for flexible structures undergoing large deformations with strongly nonlinear (electro-)mechanical behavior; and second, to synthesize robust controllers with clear physical interpreta-tions to stabilize such structures at desired configurations and to follow desired trajectories.

A detailed description of the position is attached; it includes the application requirements and guidelines.

This thesis is conducted within the framework of the ANR PRCE Power-Twin project, which aims to develop a digital twin for the diagnosis and prognosis of power module components, with particular attention to bonding wire degradation under variable load profiles.

The main objective of this thesis is to extend the health state prediction capabilities of prognostic models to conditions that are poorly represented in training data or entirely unseen. This will be achieved through three complementary research directions, each structured around a core research question and a set of concrete tasks.

A detailed description of the position is attached; it includes the application requirements and guidelines.