HAL Portal IOGS (nstitut d'Optique Graduate School)
Not a member yet
12589 research outputs found
Sort by
Reactive Atomistic Insights into Oxidation Dynamics and Wettability of Ultrafast Laser-Treated Titanium Surfaces
No full textInternational audienceUltrafast laser surface treatment offers remarkable advantages over other surface treatment. However, several experimental studies have provided insights into the resulting morphological and chemical changes but the mechanisms governing these transformations and their impact on wettability at nanoscale remain poorly understood. Herein, we employed Molecular Dynamics (MD) simulations to investigate the early stage oxidation dynamics and wettability evolution of laser-treated titanium (Ti) surfaces. The effects of laser-induced oxidation on surface chemistry and wettability were analyzed through reactive MD simulations. These results reveal that laser treatment significantly modifies the surface chemistry by influencing the oxidation kinetics, surface roughness, defect formation and wettability
Femtosecond laser fabrication of plasmonic-magnetic Ni-Au nanoparticles with enhanced magnetic properties
No full textInternational audienceBimetallic nanoparticles, particularly Ni-Au, exhibit unique magneto-optical properties that can be fine-tuned through careful control of their morphology and composition. Femtosecond lasers have demonstrated significant potential in achieving these modifications by inducing rapid heating and alloying. In this study, we explore the laser-induced synthesis of Ni-Au nanoparticles, focusing on their enhanced magnetic properties. Using femtosecond laser ablation in liquids, we produced nanoparticles with varying sizes, shapes, and compositions, depending on the choice of solvent. Numerical simulations provided insights into the mechanisms underlying these transformations, highlighting the influence of laser parameters and solvent effects. These findings suggest a path toward the design of multifunctional nanoparticles for advanced applications in catalysis, biomedicine, and data storage
Préparation et caractérisation d'états de spin corrélés avec un simulateur quantique à atomes de Rydberg
No full textThis thesis focuses on the development of experimental tools and protocols for the study of correlated spin systems, from their preparation to their characterization. The experimental setup relies on trapping individual rubidium atoms in optical tweezers, arranged in one- or two-dimensional arrays. Once excited to Rydberg states, the atoms interact pairwise, effectively described by spin models—allowing us to perform quantum simulations of these models. The overarching goal is to better understand the mechanisms at play in strongly correlated states of matter, which are notoriously difficult to simulate when the number of spins becomes large. Implementing spin models requires precise control of both the internal and external degrees of freedom of the atoms. During my thesis, I improved the pre-existing setup by applying a Raman cooling technique to reduce the thermal positional fluctuations of atoms in optical tweezers. We also devised a method to perform local transitions between Rydberg states, enabling the preparation of arbitrary spin textures and the simultaneous readout of several spins in different bases. These techniques allowed us to observe collective quantum phenomena in various regimes, ranging from low-energy equilibrium properties to the dynamics of high-energy states.First, the adiabatic preparation of the ground states of the so-called XY spin model enabled us to observe the formation of ferromagnetic or antiferromagnetic spin order, measured through spin correlations, on a square lattice and a one-dimensional chain. Regardless of the dimensionality of the system, the dipolar nature of the interactions destabilized the antiferromagnetic phase while reinforcing ferromagnetic order. In one dimension, we observed Friedel oscillations near a localized defect and measured power-law correlation profiles, a hallmark of critical phases of matter. Next, we developed a novel technique to measure the dispersion relation of low-energy excitations. This method, termed quench spectroscopy, involves measuring the propagation of spin correlations starting from a low-energy out-of-equilibrium state, and applying a double Fourier transform (in space and time) to obtain the energy of elementary excitations as a function of their wavevector. Applying this method to the aforementioned cases, we verified the linearity of the dispersion relation at low wave vectors, characteristic of spin waves in the XY model—except for the notable case of ferromagnetic order in two dimensions, where the energy evolves as the square root of the wavevector due to dipolar interactions. Finally, we exploited off-diagonal van der Waals interactions to implement a spin-hole model known as the t-J-V model, which describes the motion of holes in a spin lattice. Realizing this model with Rydberg states provides access to parameter regimes that were previously unattainable with single-particle resolution, in particular the possibility of long-range tunneling over several lattice sites. By studying the dynamics of a doped magnet, we identified signatures of this effective long-range tunneling, and we demonstrated the influence of the spin background on hole transport.Cette thèse porte sur le développement d'outils et de protocoles expérimentaux pour l'étude de systèmes de spin fortement corrélés, depuis leur préparation jusqu'à leur caractérisation.Le dispositif expérimental est basé sur le piégeage d'atomes de rubidium individuels dans des pinces optiques, selon des structures uni- ou bi-dimensionnelles ordonnées. Une fois excités dans des états de Rydberg, les atomes interagissent deux à deux, ce qui peut être décrit de manière effective par un modèle de spin --- donnant lieu à des simulations quantiques de ces modèles. L'objectif général est de mieux comprendre les mécanismes à l'œuvre dans les états fortement corrélés de la matière, qui sont notoirement difficiles à simuler numériquement lorsque le nombre de spins devient grand.La mise en œuvre des modèles de spin nécessite un contrôle précis des degrés de liberté tant internes qu'externes des atomes. Durant ma thèse, j'ai amélioré le montage préexistant, en appliquant une technique de refroidissement Raman pour réduire les fluctuations de position thermiques des atomes dans les pinces optiques. Nous avons également conçu une méthode de transitions locales entre états de Rydberg, permettant la préparation de textures de spin arbitraires et la lecture simultanée de plusieurs spins dans des bases différentes. Ces techniques nous ont permis de mettre en évidence des phénomènes quantiques collectifs dans des systèmes magnétiques avec interactions dipolaires de longue portée. Nous avons étudié plusieurs régimes, depuis les propriétés d'équilibre à basse énergie, jusqu'à la dynamique d'états de haute énergie.Tout d'abord, la préparation adiabatique des états fondamentaux du modèle de spin dit XY a permis d'observer la formation d'un ordre ferromagnétique ou antiferromagnétique entre spins, mesuré au moyen de leurs corrélations, sur un réseau carré et sur une chaîne unidimensionnelle. Quelle que soit la dimension du système étudié, le caractère dipolaire des interactions s'est traduit par une déstabilisation de la phase antiferromagnétique, et un renforcement de l'ordre ferromagnétique. En une dimension, nous avons observé des oscillations de Friedel à proximité d'un défaut localisé et mesuré des profils de corrélations en loi de puissance, signature d'une phase critique de la matière. Nous avons ensuite mis au point une nouvelle technique de mesure de la relation de dispersion des excitations de basse énergie. Cette méthode, appelée quench spectroscopy, consiste à mesurer la propagation des corrélations entre tous les spins à partir d'un état hors-équilibre de basse énergie, et à appliquer une double transformée de Fourier (spatiale et temporelle) pour obtenir l'énergie des excitations élémentaires en fonction de leur vecteur d'onde. Appliquant cette méthode aux cas précédents, nous avons vérifié la linéarité de la relation de dispersion à faible vecteur d'onde, caractéristique des ondes de spin du modèle XY ; à l'exception notable du cas ferromagnétique en deux dimensions, où l'énergie évolue comme la racine carrée du vecteur d'onde du fait des interactions dipolaires.Enfin, nous avons exploité les interactions de van der Waals non-diagonales pour mettre en œuvre un modèle de spin et de trous appelé modèle t-J-V, décrivant le mouvement de trous dans un réseau de spins. La réalisation de ce modèle avec des états de Rydberg permet d'accéder à des régimes de paramètres jusqu'à présent inaccessibles avec une résolution à l'échelle de la particule individuelle; en particulier, les interactions dipolaires équivalent à un effet tunnel de longue portée, autorisant le saut d'une particule entre deux sites distants. En étudiant la dynamique d'un aimant dopé par des trous localisés, nous avons trouvé des signatures de cet effet tunnel effectif de longue portée, et mis en évidence l'influence des états de spin sur le transport des trous
Temporal pulse shaping technique for nonlinear frequency shifting
No full textInternational audienceWe present a spectral phase-only shaping technique allowing to generate arbitrary temporal shapes and demonstrate its use for ultrafast ytterbium-based laser sources. This technique is based on the introduction of a nonlinear chirp that is determined solely by the input spectral intensity and the target temporal intensity. We describe its working principle and experimentally demonstrate the generation of triangle-shaped pulses at the output of a laser source, which delivers 150 μ J, sub-200 fs pulses at 150 kHz. These pulses are then passed through a nonlinear multipass cell to observe wavelength shifting via self-phase modulation. A spectral tunabibility of 14 nm is achieved around the input central wavelength of 1034 nm
A Bregman Proximal Viewpoint on Neural Operators
No full textInternational audienceWe present several advances on neural operators by viewing the action of operator layers as the minimizers of Bregman regularized optimization problems over Banach function spaces. The proposed framework allows interpreting the activation operators as Bregman proximity operators from dual to primal space. This novel viewpoint is general enough to recover classical neural operators as well as a new variant, coined Bregman neural operators, which includes the inverse activation operator and features the same expressivity of standard neural operators. Numerical experiments support the added benefits of the Bregman variant of Fourier neural operators for training deeper and more accurate models
Breast Cancer Diagnosis With Explainable Artificial Intelligence (XAI): Uncovering Strengths and Biases
No full textInternational audienceBreast cancer is one of the most common malignancies afflicting women globally, necessitating the use of cutting-edge AI approaches in diagnostic procedures to improve patient outcomes drastically. However, one key difficulty with the most recent AI models is a lack of transparency, making it difficult for medical practitioners to implement these technologies to increase diagnostic accuracy. Many explainable AI (XAI) solutions have been developed to overcome this issue. This survey focuses on applying XAI approaches in breast cancer detection and diagnosis, particularly emphasizing their role in increasing model transparency and clinical decision-making. The article also provides insights into the inherent biases in the most recent machine learning models towards specific XAI approaches, such as the compatibility of Convolutional Neural Networks (CNNs) with visual explanation methods and tree-based models with feature significance evaluations. Finally, the article covers the obstacles to using XAI technologies in clinical practice and the significance of defining standard measures for assessing their performance
A diffusion model for light scattering in ejecta
No full textInternational audienceWe derive a diffusion equation for light scattering from ejecta produced by extreme shocks on metallic samples. This model is easier to handle than a more conventional model based on the Radiative Transfer Equation (RTE) and is a relevant tool to analyze spectrograms obtained from Photon Doppler Velocimetry measurements in the deep multiple scattering regime. We also determine the limits of validity of the diffusive model compared to the RTE, based on a detailed analysis of various ejecta properties in configurations with increasing complexity
Unveiling Habitable Worlds with high precision astrometry for next-generation Direct Imaging missions like LIFE
No full textInternational audienceThe future space mission Large Interferometer For Exoplanets (LIFE) has been designed to characterize the atmospheres of temperate terrestrial exoplanets and search for potential bio-signatures. To reach this ambitious objective, a precise pre-characterization of target planets is essential—particularly their masses and orbital parameters.This is where Theia, a high-precision astrometric mission, plays a critical complementary role. Theia aims to detect Earth-mass planets within the Habitable Zones (HZ) of nearby F, G, and K-type stars, providing exact measurements of their true masses and full 3D orbital architectures—data currently beyond the reach of any existing or planned observatory. This information will be crucial for optimizing the target selection for LIFE, eliminating uncertainties and maximizing the scientific return of direct imaging and atmospheric spectroscopy.This presentation will give an overview of Theia recent technical developments. Notable innovations include its compact diffraction-limited 0.8-meter Korsch Three-Mirror Anastigmat (TMA) telescope, large-format CMOS detectors, and a simplified in-flight calibration approach using Gaia reference stars, eliminating the need for laser metrology. Operating from the Lagrange Point L2, Theia is designed to deliver sub-microarcsecond astrometric precision over a 4-year mission.By providing a robust and efficient target list, Theia would lay the groundwork for LIFE’s future search for life beyond Earth, illustrating the scientific synergy between astrometric and direct imaging missions
Experimental Investigation of a Bipartite Quench in a 1D Bose gas
No full textLong wavelength dynamics of 1D Bose gases with repulsive contact interactions can be captured by Generalized HydroDynamics (GHD) which predicts the evolution of the local rapidity distribution. The latter corresponds to the momentum distribution of quasiparticles, which have infinite lifetime owing to the integrability of the system. Here we experimentally investigate the dynamics for an initial situation that is the junction of two semi-infinite systems in different stationary states, a protocol referred to as `bipartite quench' protocol. More precisely we realise the particular case where one half of the system is the vacuum state. We show that the evolution of the boundary density profile exhibits ballistic dynamics obeying the Euler hydrodynamic scaling. The boundary profiles are similar to the ones predicted with zero-temperature GHD in the quasi-BEC regime, with deviations due to non-zero entropy effects. We show that this protocol, provided the boundary profile is measured with infinite precision, permits to reconstruct the rapidity distribution of the initial state. For our data, we extract the initial rapidity distribution by fitting the boundary profile and we use a 3-parameter ansatz that goes beyond the thermal assumption. Finally, we investigate the local rapidity distribution inside the boundary profile, which, according to GHD, presents, on one side, features of zero-entropy states. The measured distribution shows the asymmetry predicted by GHD, although unelucidated deviations remain
Mise en œuvre de protocoles itératifs pour la génération d'états quantiques encodés sur des impulsions lumineuses
No full textThis PhD work is part of a broader effort to develop quantum technologies, and especially the photonic platform, using continuous variables. The experimental realization of an efficient source of non-classical states of light is a key step in developing such a platform. The main focus of the present work is to take part in the development of one particular architecture for such a source. This source uses a quantum memory to produce non-classical states of light in an iterative fashion. At each step, a projective measurement is performed on an entangled state that was obtained through the interaction of a state that was stored in the memory and a resource state on a beam splitter. The resource states that we use are heralded single photons that are produced through spontaneous parametric down conversion. The first step in the development of the source is to produce “even Schrödinger cat states” from a photon coming from a single photon source, and a photon that was produced by the same source and stored in the quantum memory. The first results were obtained and published at the beginning of this PhD project. A theoretical work consisting of the optimization of the emission modes of the single photon source has then been conducted. In the meantime, closed-loop controlled stabilization systems, and electronic control optimizations were implemented in the experimental setup. This optimized setup allowed us to generate and publish new results in which the fidelity of the even cat states was improved and where the generation rate was multiplied by ten. The optimizations of the experimental setup also paved the way toward the generation of “odd Schrödinger cat states”. Different odd cat states generation protocols are proposed, and preliminary results of the implementation of one of these protocols are presented in this manuscript.Ce travail de thèse s'inscrit dans le cadre du développement des technologies quantiques en général et dans le cadre du développement d'une plateforme photonique en variables continues en particulier. Le développement d'une telle plateforme nécessite la réalisation de sources efficaces d'états non classiques de la lumière. Ce travail de thèse se concentre principalement sur le développement d'une architecture particulière pour une telle source. Cette dernière génère des états non classiques de la lumière à l'aide d'une mémoire quantique, de manière itérative, en réalisant à chaque étape une mesure projective sur un état intriqué produit à partir d'un état stocké en mémoire et d'un état ressource interagissant sur une lame séparatrice. Nous utilisons comme états ressources des photons uniques annoncés produits par fluorescence paramétrique. Dans un premier temps, nous cherchons à générer des états « Chat de Schrödinger pair » à partir d'un photon provenant de la source de photons uniques, et d'un photon stocké en mémoire, provenant initialement de la même source. Une première preuve de principe a été publiée au début de cette thèse. Une fois ces résultats obtenus, un travail théorique d'optimisation des modes d'émission de la source de photons uniques a été conduit. Parallèlement, des travaux expérimentaux de stabilisation et d'asservissement de l'expérience, ainsi qu'une optimisation du contrôle électronique du montage ont été réalisés. Grâce à cette version plus aboutie de l'expérience, de nouveaux résultats ont été obtenus et publiés au cours de la thèse. Ces nouveaux résultats présentent une meilleure fidélité et un taux de génération dix fois supérieur comparés aux résultats publiés en début de thèse. Au-delà des nouveaux résultats, les améliorations du dispositif expérimental mises en place nous ouvrent la voie vers la génération d'états plus complexes. Cette thèse propose donc des protocoles de génération d'états « Chat de Schrödinger impair ». Des résultats préliminaires de la mise en œuvre de ces protocoles sont présentés en fin de ce manuscrit