HAL Portal IOGS (nstitut d'Optique Graduate School)
Not a member yet
12589 research outputs found
Sort by
Disordered optical metasurfaces: basics, properties, and applications
No full textInternational audienceOptical metasurfaces are conventionally viewed as organized flat arrays of photonic or plasmonic nanoresonators, also called metaatoms. These metasurfaces are typically highly ordered and fabricated with precision using expensive tools. However, the inherent imperfections in large-scale nanophotonic devices, along with recent advances in bottom-up nanofabrication techniques and design strategies, have highlighted the potential benefits of incorporating disorder to achieve specific optical functionalities. This review offers an overview of the key theoretical, numerical, and experimental aspects related to the exploration of disordered optical metasurfaces. It introduces fundamental concepts of light scattering by disordered metasurfaces and outlines theoretical and numerical methodologies for analyzing their optical behavior. Various fabrication techniques are discussed, highlighting the types of disorder they deliver and their achievable precision level. The review also explores critical applications of disordered optical metasurfaces, such as light manipulation in thin film materials and the design of structural colors and visual appearances. Finally, the article offers perspectives on the burgeoning future research in this field. Disordered optical metasurfaces offer a promising alternative to their ordered counterparts, often delivering unique functionalities or enhanced performance. They present a particularly exciting opportunity in applications demanding large-scale implementation, such as sustainable renewable energy systems, as well as aesthetically vibrant coatings for luxury goods and architectural designs
Méthode de formation de nanoparticules à la surface de substrats
No full textLa présente invention se rapporte à un procédé de fonctionnalisation de surface permettant de générer des nanoparticules à la surface d'un matériau. En particulier, ces nanoparticules ont une composition chimique différente de celle du matériau traité, et elles présentent un ancrage mécanique dans le matériau de base. Le traitement obtenu est localisé tant en termes de profondeur du matériau concerné qu'en position sur la surface traitée.</div
Suppression of Bogoliubov momentum pairing and emergence of non-Gaussian correlations in ultracold interacting Bose gases
No full textInternational audienceStrongly correlated quantum matter -- such as interacting electron systems or interacting quantum fluids -- possesses properties that cannot be understood in terms of linear fluctuations and free quasi-particles. Quantum fluctuations in these systems are indeed large and generically exhibit non-Gaussian statistics -- a property captured only by inspecting high-order correlations, whose quantitative reconstruction poses a formidable challenge to both experiments and theory alike. A prime example of correlated quantum matter is the strongly interacting Bose fluid, realized by superfluid Helium and, more recently, ultra-cold atoms. Here, we experimentally study interacting Bose gases from the weakly to the strongly interacting regime through single-atom-resolved correlations in momentum space. We observe that the Bogoliubov pairing among modes of opposite momenta, emblematic of the weakly interacting regime, is suppressed as interactions become stronger. This departure from the predictions of Bogoliubov theory signals the onset of the strongly correlated regime, as confirmed by numerical simulations that highlight the role of non-linear quantum fluctuations in our system. Additionally, our measurements unveil a non-zero four-operator cumulant at even stronger interactions, which is a direct signature of non-Gaussian correlations. These results shed light on the emergence and physical origin of non-Gaussian correlations in ensembles of interacting bosons
Platforms for the realization and characterization of Tomonaga–Luttinger liquids
No full textInternational audienceThe concept of a Tomonaga-Luttinger liquid (TLL) has been established as a fundamental theory for the understanding of one-dimensional quantum systems. Originally formulated as a replacement for Landau’s Fermi-liquid theory, which accurately predicts the behaviour of most 3D metals but fails dramatically in 1D, the TLL description applies to a even broader class of 1D systems,including bosons and anyons. After a certain number of theoretical breakthroughs, its descriptive power has now been confirmed experimentally in different experimental platforms. They extend from organic conductors, carbon nanotubes, quantum wires, topological edge states of quantum spin Hall insulators to cold atoms, Josephson junctions, Bose liquids confined within 1D nanocapillaries and spin chains. In the ground state of such systems, quantum fluctuations become correlated on all length scales, but, counter-intuitively, no long-range order exists. In this respect, this review will illustrate the validity of conformal field theory for describing real-world systems, establishing the boundaries for its application and, on the other side will discuss the spectacular demonstration of how the quantum-critical TLL stategoverns the properties of many-body systems in one dimension
Copper’s Encore: (Dry)Etching a Path from Stardom to Survival in Next-Gen Interconnects
No full textInternational audienceCopper, once the undisputed star of semiconductor interconnects, now risks being written out of the script as next-generation technologies demand ever-tighter pitches. This chapter explores copper’s most infamous diva trait: its stubborn resistance to dry etching. Where aluminum took direction with grace, copper clings to the stage, its low-volatility byproducts refusing to bow out cleanly. We trace decades of creative etching attempts—using non-halogens, halogens, and hydrocarbons—all with mixed reviews. Rather than delivering a polished performance, copper dry etching has often culminated in a cluttered stage. But the show must go on. Drawing from our own work and that of the broader research community, we look beyond chemistry to spotlight the understudied role of copper’s microstructure—its grain boundaries, surface oxides, and crystallographic quirks—as critical players in this performance. The verdict? It may not be a curtain call for copper just yet. With the right script—balancing chemistry, anisotropy, and process integration—it could still earn a standing ovation. After all, even the most capricious divas deserve an encore
LED-pumped room-temperature solid-state maser
No full textInternational audienceAbstract Room-temperature MASERs (Microwave Amplification by Stimulated Emission of Radiation) amplify electromagnetic waves at microwave frequencies with minimal noise. We demonstrate a cost-effective LED-pumped maser using pentacene-doped para-terphenyl as the gain medium. Here, we show that LED light, which is brightness-enhanced and guided via a cerium-doped yttrium aluminium garnet luminescent concentrator, achieves persistent maser emission at 1.45 GHz with a duration of 200 µs and a microwave output power of 0.014 mW, surpassing previous non-laser pumped systems. Operating at low voltage, the LED-pumped maser ensures safety, reduced costs, and simple integration. Potential applications include sensitive magnetic resonance imaging, portable atomic clocks, quantum technologies, and enhanced deep-space radio astronomy
Atomistic insights into ultrafast laser-driven dynamics for silica nanostructuring
No full textInternational audienceThe advancement of micro- and nano-technologies demands precise and controllable material structuring [1]. Ultrashort laser processing enables high-accuracy nanoscale modifications by optimizing key parameters to match material responses. Achieving this precision requires understanding the fundamental electronic processes that govern nonlinear laser propagation, local field enhancement, and anisotropic energy deposition. The resulting relaxation mechanisms—thermal or hydrodynamic—drive structural transformations, including amorphization, polymorphic transitions, and the formation of nanovoids and nanogratings.This work focuses on the interaction of ultrafast laser excitation with wide-bandgap materials like silicon dioxide (fused silica, α-quartz) at the atomic scale, using ab initio calculations. Beyond conventional optical resolution, nanoscale precision can be achieved by controlling light energy distribution and thermodynamic pathways to direct absorption confinement, crystal-to-amorphous transitions, and rapid energy dissipation. To explore these dynamics, we employ multiphysics simulations that capture the interplay between electronic structure modifications and structural/hydrodynamic relaxation under extreme nonequilibrium conditions.Under ultrafast photoexcitation, transient electron densities in the conduction band increase dramatically, leading to band distortion and pronounced bandgap renormalization [2]. Using Density Functional Theory (DFT) within the GW approximation, along with Time-Dependent DFT and molecular dynamics, we investigate bandgap evolution across different excitation regimes, revealing shifts of several eV within femtoseconds. The redistribution of excited carriers weakens silica bonds, triggering structural reconfigurations that alter material cohesion as well as optical and thermal properties [3-4].At later stages, complex absorption dynamics promote the emergence of nanostructures formation, driving further structural rearrangements. Coupled electromagnetic and molecular dynamics simulations provide deeper insights into volume structuring, revealing the fundamental interaction between light and evolving material states. The underlying mechanisms stem from the formation of nanopores that can self-organize, leading to the formation of polarization-aligned nanogratings, exhibiting birefringence, and demonstrating the reshaping of nanoscale plasmas under strong fields and potential resonance effects. These findings contribute to a more comprehensive understanding of ultrafast laser-material interactions, supporting the development of refined laser processing techniques and opening new routes for controlled material transformations at the nanoscale. [1] R. Stoian, and J.P. Colombier, « Advances in ultrafast laser structuring of materials at the nanoscale », Nanophotonics 9(16), 4665-4688 (2020).[2] A. Tsaturyan, E. Kachan, R. Stoian, and J.P. Colombier, « Ultrafast bandgap narrowing and cohesion loss of photoexcited fused silica», The Journal of Chemical Physics 156 (22), 224301 (2022). [3] A. Tsaturyan, E. Kachan, R. Stoian, and J.P. Colombier, Unraveling the electronic properties in SiO2 under ultrafast laser irradiation, NPJ Computational Materials, 10 (1), 1-10 (2024). [4] A. Tsaturyan, E. Kachan, R. Stoian, & J.P. Colombier, « Excited‐state dynamics and optical properties of silica Under ultrafast laser irradiation», Advanced Physics Research, 2400106 (2024)
Visual emotion analysis using skill-based multi-teacher knowledge distillation
No full textInternational audienceThe biggest challenge in Visual Emotion Analysis (VEA) is bridging the affective gap between the features extracted from an image and the emotion it expresses. It is therefore essential to rely on multiple cues to have decent predictions. Recent approaches use deep learning models to extract rich features in an automated manner, through complex frameworks built with multi-branch convolutional neural networks and fusion or attention modules. This paper explores a different approach, by introducing a three-step training scheme and leveraging Knowledge Distillation (KD), which reconciles effectiveness and simplicity, and thus achieves promising performances despite using a very basic CNN. KD is involved in the first step, where a student model learns to extract the most relevant features on its own, by reproducing those of several teachers specialized in different tasks. The proposed Skill-based Multi-teacher Knowledge Distillation (SMKD) loss also ensures that for each instance, the student focuses more or less on the teachers depending on their capacity to obtain a good prediction, i.e. their relevance. The two remaining steps serve respectively to train the student's classifier and to fine-tune the whole model, both for the VEA task. Experiments on two VEA databases demonstrate the gain in performance offered by our approach, where the students consistently outperform their teachers, and also state-of-the-art methods
Targeted Test Time Adaptation of Memory Networks for Video Object Segmentation
No full textInternational audienceSemi Automatic Video Object Segmentation (SVOS) aims to segment few objects in a video based on the annotation of these particular objects in the first frame only. State-of-the-art methods rely on offline training on a large dataset that may lack specific samples and details directly applicable to the current test video. Common solutions are to use test-time adaptation to finetune the offline model with the single annotated frame or by relying on complex semi-supervised strategies. In this paper, we introduce targeted test-time adaptation of memory-based SVOS providing the benefits of finetuning with much smaller learning effort. Our method targets specific parts of the model to ensure improved results while maintaining robustness of the offline training. We find that targeting the bottleneck features and the masks that are saved in memory provide substantial benefits. The evaluation of our method shows a significant improvement for video segmentation on DAVIS16 and DAVIS17 datasets