HAL UVSQ
Not a member yet
57319 research outputs found
Sort by
First JWST thermal phase curves of temperate terrestrial exoplanets reveal no thick atmosphere around TRAPPIST-1 b and c
International audience72 pages, 4 main text Figures, 20 Extended Data Figures, 6 Supplementary Figures. Accepted for publication in Nature Astronom
Leveraging Text-to-Text Transformers as Classifier Chain for Few-Shot Multi-Label Classification
International audienc
Polyoxométallates super réduits ˸ formation, structures et propriétés
Polyoxometalates (POMs) are polynuclear anions builded from transition metals in their highest oxidation state (VV, MoVI, WVI). Known for their remarkable redox properties, they can exchange large numbers of electrons, making them potential candidates as electro-active components for the design of electrochemical energy storage systems. It has been known since the 1960s that their advanced reduction leads to the formation of intramolecular metal bonds, but few fundamental studies have since been devoted to this subject. We have enriched them by studying the reduction of different POM architectures. Characterization of the super-reduced species has been implemented using electrochemical methods, single-crystal X-ray diffraction, multinucleus NMR (1H, 11B, 95Mo, 183W, 31P, 29Si), UV-vis and infrared spectroscopies. The manuscript begins with a study of the reduction processes of Keggin-type POMs [XW12O40]n- (X: H2, B and Si), leading to the obtention of several twelve-electron reduced derivatives.We have also studied the reduction of polyanions composed of 18 tungstic centers, following the formation of metal bonds in the species [H2AsWVI15WIV3O57(H2O)3]7-. The 18-electron reduction of the Dawson ion [P2W18O62]6- was also explored, leading to the formation of metal bonds on the structure's two apical triads. Finally, we have devoted part of this work to study the advanced stages of reduction of heterometallic W/Mo POMs. This study shows that molybdenum atoms are preferentially involved in the formation of metal-metal bonds, as indicated by the presence of the heterometallic cluster {MoW2O13} in the monosubstituted species [SiWVI9WIV2MoIVO37(H2O)3]4-. Finally, tracking the twelve-electron reduction of the POM [P2W15Mo3O62]6-, we draw on X-ray absorption spectroscopy (EXAFS) data supported by theoretical calculations (DFT) and show that its {Mo3O13} triad can accommodate up to 9 electrons, imposing in a novel way the formation of MoIII centers in a polyoxometalate.Les polyoxométallates (POMs) sont des anions polynucléaires construits à partir de métaux de transition dans leur plus haut état d'oxydation (VV, MoVI, WVI). Reconnus pour leurs propriétés redox remarquables, ils peuvent échanger un grand nombre d'électrons, ce qui en fait des candidats potentiels comme composants électro-actifs pour la conception de systèmes de stockage de l'énergie électrochimique. Il est connu depuis les années 1960 que leur réduction à un stade avancé conduit à la formation de liaisons métalliques intramoléculaires, mais peu d'études fondamentales s'y sont depuis lors consacrées. Nous les avons enrichies en étudiant la réduction de différentes architectures de POMs. La caractérisation des espèces super-réduites a été mise en place par l'usage de méthodes électrochimiques, de la diffraction des rayons X sur monocristal, la RMN multinoyaux (1H, 11B, 95Mo, 183W, 31P, 29Si) et les spectroscopies UV-vis et infrarouge. Le manuscrit débute par l'étude des processus de réduction de POMs de type Keggin [XW12O40]n- (X : H2, B et Si), conduisant à l'obtention de plusieurs dérivés réduits à douze électrons.Nous avons aussi étudié la réduction de polyanions composés de 18 centres tungstiques, en suivant la formation de liaisons métalliques dans l'espèce [H2AsWVI15WIV3O57(H2O)3]7-. La réduction à 18 électrons de l'ion de Dawson [P2W18O62]6- a par ailleurs été explorée, aboutissant notamment à la formation de liaisons métalliques sur les deux triades apicales de la structure. Enfin, nous avons consacré une partie de ce travail à l'étude des stades avancés de réduction de POMs hétérométalliques W/Mo. Cette étude montre que les atomes de molybdène sont préférentiellement impliqués dans la formation de liaisons métal-métal comme l'indique la présence du cluster hétérométallique {MoW2O13} dans l'espèce monosubstituée [SiWVI9WIV2MoIVO37(H2O)3]4-. Enfin, en suivant la réduction à douze électrons du POM [P2W15Mo3O62]6-, nous nous sommes appuyés sur des données de spectroscopie d'absorption des rayons X (EXAFS) supportées par des calculs théoriques (DFT) et montrons que sa triade {Mo3O13} peut accueillir jusqu'à 9 électrons, imposant de façon inédite la formation de centres MoIII dans un polyoxométallate
Photoaccumulation of Long-lived Reactive Electrons in the Microporous Ti(IV) Oxocluster Based Metal-organic Framework for Light and Dark Photocatalysis
The microporous Ti12 oxocluster based metal-organic framework MIP-177(Ti)-LT exhibits excellent stability and photoactivity, making it highly promising for photocatalytic applications (LT stands for low temperature synthesis). This study employs transient and photoinduced absorption spectroscopies to analyse the behaviour of reactive electrons in MIP-177(Ti)-LT across femtosecond-to-second timescales. Over these timescales, MIP-177(Ti)-LT shows effective charge separation and relatively slow decay kinetics, with photogenerated charges persisting through to microsecond-second timescales and exhibiting significantly higher yields and slower decay kinetics than two benchmark metal(IV) oxoclusters based photoactive MOFs, MIL-125(Ti)-NH2 and UiO-66(Zr)-NH2. Photogenerated holes in MIP-177(Ti)-LT are found to be able to oxidise water to produce oxygen at a yield of 335 µmol g-1⋅ h-1 in the presence of electron scavengers. Meanwhile, long-lived photogenerated electrons accumulate under continuous irradiation, with this accumulation being enhanced by the presence of a hole scavenger, methanol. Notably, these photogenerated electrons can persist for over 48 hours post-photoexcitation under sealed conditions in argon, causing a reversible colour shift from white to black. These accumulated long-lived electrons are shown to be redox-active, efficiently reducing both added oxygen and methyl-viologen. The subsequent dark addition of a platinum co-catalyst to the photocharged MIP-177(Ti)-LT results in hydrogen evolution with a yield of circa 300 µmol/g (~58 C/g), corresponding to an accumulated electron density of one electron per 12 Ti atoms (i.e., one electron per titania oxocluster). These results highlight the photocharging properties of MIP-177(Ti)-LT and its potential for sustainable and efficient photocatalytic processes
Evidence for Subtropical Dynamical Changes in Reunion Ozone Trends (1998-2021)
International audienceWe provide trend estimates for total, stratospheric, and tropospheric ozone columns over Reunion (21.1° S, 55.5° E) from 1998 to 2021, using only Syst`eme d’Analyse par Observation Z´enithale (SAOZ) and Southern Hemisphere Additional OZonesonde (SHADOZ) observations. Trends are derived using Trend-Run, a multiple linear regression (MLR) model, and a dynamic linear model (DLM) to identify potential turning points. Overall, total ozone shows signs of recovery (3.0 ± 1.5 DU/decade), with increases in both stratospheric (1.1 ± 1.6 DU/decade) and tropospheric ozone (2.2 ± 1.0 DU/decade). DLM identifies a turning point in stratospheric ozone in 2008, with a clear decrease in stratospheric ozone before this point and an increase afterward. We also determined changes in the lapse rate tropopause (LRT), the subtropical barrier position, and ERA5 wind and geopotential fields during the same period to investigate possible links between mid-tropospheric ozone increase and transport-related perturbations. Although trends in LRT height and temperature are barely significant, they suggest a recent deepening of the troposphere, indicative of climate change. Intensification of the anticyclonic gyre over Southern Africa and a weakening of the Mascarene anticyclone are found. This suggests that, independent of possible changes in ozone precursor emissions over Africa or South America, dynamics are driving increases of ozone and ozone precursors over Reunion from 1998 to 2021. Furthermore, the rate of Reunion’s free tropospheric trends exceeds that observed at all other southern hemisphere ozonesonde stations, including those in tropical, subtropical and mid-latitude regions
MERCURY: A fast and versatile multi-resolution based global emulator of compound climate hazards
High-impact climate damages are often driven by compounding climate conditions. For example, elevated heat stress conditions can arise from a combination of high humidity and temperature. To explore future changes in compounding hazards under a range of climate scenarios and with large ensembles, climate emulators can provide light-weight, data-driven complements to Earth System Models. Yet, only a few existing emulators can jointly emulate multiple climate variables. In this study, we present the Multi-resolution EmulatoR for CompoUnd climate Risk analYsis: MERCURY. MERCURY extends multi-resolution analysis to a spatio-temporal framework for versatile emulation of multiple variables. MERCURY leverages data-driven, image compression techniques to generate emulations in a memory-efficient manner. MERCURY consists of a regional component that represents the monthly, regional response of a given variable to yearly Global Mean Temperature (GMT) using a probabilistic regression based additive model, resolving regional cross-correlations. It then adapts a reverse lifting-scheme operator to jointly spatially disaggregate regional, monthly values to grid-cell level. We demonstrate MERCURY's capabilities on representing the humid-heat metric, Wet Bulb Globe Temperature, as derived from temperature and relative humidity emulations. The emulated WBGT spatial correlations correspond well to those of ESMs and the 95% and 97.5% quantiles of WBGT distributions are well captured, with an average of 5% deviation. MERCURY's setup allows for region-specific emulations from which one can efficiently "zoom" into the grid-cell level across multiple variables by means of the reverse lifting-scheme operator. This circumvents the traditional problem of having to emulate complete, global-fields of climate data and resulting storage requirements
Multi-site modelling and reconstruction of past extreme skew surges along the French Atlantic coast
Appropriate modelling of extreme skew surges is crucial, particularly for coastal risk management. Our study focuses on modelling extreme skew surges along the French Atlantic coast, with a particular emphasis on investigating the extremal dependence structure between stations. We employ the peak-over-threshold framework, where a multivariate extreme event is defined whenever at least one location records a large value, though not necessarily all stations simultaneously. A novel method for determining an appropriate level (threshold) above which observations can be classified as extreme is proposed. Two complementary approaches are explored. First, the multivariate generalized Pareto distribution is employed to model extremes, leveraging its properties to derive a generative model that predicts extreme skew surges at one station based on observed extremes at nearby stations. Second, a novel extreme regression framework is assessed for point predictions. This specific regression framework enables accurate point predictions using only the "angle" of input variables, i.e. input variables divided by their norms. The ultimate objective is to reconstruct historical skew surge time series at stations with limited data. This is achieved by integrating extreme skew surge data from stations with longer records, such as Brest and Saint-Nazaire, which provide over 150 years of observations
Inter-annual snow accumulation and meter-scale variability from trench measurements at Dome C, Antarctica
International audienceThe central regions of the East Antarctic ice sheet contains some of the oldest ice on earth, due to low snow accumulation rates and consistently cold conditions. One consequence of low accumulation however is that the little snow amount which is deposited at irregular times is more strongly affected by erosion and re-deposition by wind, inducing local mixing and loss of snowfall events in the snowpack. This discontinuous deposition leads to highly disturbed snow layers, limiting the interpretation of climate records from ice cores in these regions to time scales larger than decades. In order to interpret climate records at higher temporal resolution it is crucial to assess and quantify the local patterns of snow accumulation leading to stratigraphic noise. Here, we reconstruct the spatial and temporal variability of snow accumulation in Central Antarctica using chemical composition and physical properties in 35 vertical profiles sampled in a 50 m long snow trench at Dome C. We show that a high resolution alignment of the chemistry profiles is a suitable method for inter-annual dating of the single trench profiles, allowing the reconstruction of accumulation time series with a 1 year resolution over the last 20 years. This reconstruction shows annually-varying past surface configurations, with about 10 % of the surface subject to accumulation hiatus. More persistent patterns with timescales of years are also evidenced, causing difference in snow age of up to 4 years at similar depth in neighboring profiles, highlighting the complex dynamics of the snow accumulation in central Antarctica
Review of physical processes and parameterizations of sea ice in a changing climate
International audienceSea ice is a crucial component of polar climate systems, and is undergoing substantial changes in both the Arctic and the Southern Ocean due to evolving climatic conditions. The Arctic Ocean is undergoing a transition from being perennially ice-covered to be- ing only seasonally ice-covered. The shrinking and thinning of ice cover have been at- tributed to changes in atmospheric and oceanic forcing, as well as to changes in the phys- ical properties of pack ice. The Southern Ocean also shows signs of change, reflected in recent minima and amplified seasonality. As global warming persists in the coming decades, the role of sea ice in polar climate systems is poised for further transformation. How- ever, current theoretical frameworks and parameterization methods used in numerical models to simulate sea ice evolution and its interaction with the oceanic and atmospheric boundary layers are largely rooted in measurements from an earlier era characterized by thicker multiyear ice. In this review we aim to summarize the evolution of physical drivers governing the dynamics and thermodynamics of sea ice, snow, and ocean/atmosphere boundary layers. Furthermore, we describe and assess existing parameterization meth- ods of sub-grid scale processes, discuss recent observational findings, and propose essen- tial improvements
Adéquation Algorithme Architecture pour la localisation temps réel basée sur la fusion de données issues de l'odométrie visuelle et des mesures inertielles
Research on autonomous vehicle navigation is a major interest for the automotive industry and research organizations. Fully automating a vehicle's navigation requires devices or functions that help it perceive, analyze, and interpret its surroundings to locate itself precisely and identify all stationary or moving objects while predicting their movements over time. This automation relies on multiple sensors generating a large volume of data for processing and analysis. Therefore, the electronic system's size, the choice of an appropriate architecture, and the efficiency of the embedded software are crucial for the real-time implementation of such a system. We examined existing visual-inertial simultaneous localization and mapping (VI-SLAM) solutions for visual odometry perception and state estimation by inertial units. This analysis led us to focus on the front-end of this type of system, specifically image pre-processing, as a critical part in terms of precision and extensive computation to match the sensor frequency as closely as possible. We developed a method for detecting and tracking corners and line segments using a proposed multi-level contour detector. We planned for parallel tasks and independent scheduling between different blocks at the design phase to achieve a suitable implementation in an embedded heterogeneous architecture, taking advantage of GPU and FPGA accelerators. Using the A3, we proposed an appropriate implementation on CPU-GPU and CPU-FPGA architectures, which showed promising improvements for SLAM applications. Additionally, we proposed an observer based on an extended Kalman filter (EKF) to merge visual and inertial data for estimating the vehicle's position.La recherche sur la navigation autonome des véhicules est un intérêt majeur pour l'industrie automobile et les organismes de recherche. L'automatisation complète de la navigation d'un véhicule nécessite des dispositifs ou des fonctions qui l'aident à percevoir, analyser et interpréter son environnement pour se localiser précisément et identifier tous les objets stationnaires ou en mouvement tout en prédisant leurs mouvements dans le temps. Cette automatisation s'appuie sur de multiples capteurs générant un grand volume de données à traiter et à analyser. Par conséquent, la taille du système électronique, le choix d'une architecture appropriée et l'efficacité du logiciel embarqué sont cruciaux pour la mise en œuvre en temps réel d'un tel système. Nous avons examiné les solutions existantes de localisation et de cartographie simultanées visuo-inertielles (VI-SLAM) pour la perception visuelle de l'odométrie et l'estimation d'état par des centrales inertielles. Cette analyse nous a conduit à nous concentrer sur le front-end de ce type de système, en particulier le prétraitement d'images, comme une partie critique en termes de précision et de calcul extensif pour correspondre au plus près à la fréquence du capteur. Nous avons développé une méthode de détection et de suivi des coins et des segments de ligne à l'aide d'un détecteur de contours multi-niveaux proposé. Nous avons prévu des tâches parallèles et une planification indépendante entre différents blocs lors de la phase de conception pour obtenir une implémentation adaptée dans une architecture hétérogène embarquée, en tirant parti des accélérateurs GPU et FPGA. En utilisant l'A3, nous avons proposé une implémentation appropriée sur les architectures CPU-GPU et CPU-FPGA, qui ont montré des améliorations prometteuses pour les applications SLAM. De plus, nous avons proposé un observateur basé sur un filtre de Kalman étendu (EKF) pour fusionner les données visuelles et inertielles afin d'estimer la position du véhicule