HAL Portal ESPCI (Ecole Supérieure de Physique et de Chimie Industrielles)
Not a member yet
13939 research outputs found
Sort by
Non-monotonic conductivity of aqueous electrolytes: beyond the first Wien effect
No full textInternational audienceThe conductivity of strong electrolytes increases under high electric fields, a nonlinear response known as the first Wien effect. Here, using molecular dynamics simulations we show that this increase is almost suppressed in moderately concentrated aqueous electrolytes due to the alignment of the water molecules by the electric field. As a consequence of this alignement, the permittivity of water decreases and becomes anisotropic, an effect which can be measured in simulations and reproduced by a model of water molecules as dipoles. We incorporate the resulting anisotropic interactions between the ions into a Stochastic Density Field Theory and calculate ionic correlations as well as corrections to the Nernst-Einstein conductivity, which are in qualitative agreement with the numerical simulations.</div
Room-temperature efficient single-photon generation from CdSe/ZnS nanoplateletes
No full textInternational audienceIn the search for materials for quantum information science applications, colloidal semiconductor nanoplatelets (NPLs) have emerged as a highly promising new class of materials due to their interesting optical properties, such as narrow emission linewidth and fast photoluminescence (PL) lifetimes at room temperature. So far only few works focused on the quantum properties of their emission, however, NPLs, with their atomic-scale thickness and one-dimensional quantum confinement, are promising candidates for single-photon sources. Here, we demonstrate room-temperature single-photon emission from core/shell CdSe/ZnS NPLs, which feature 820 nm surface area and 1 nm shell. The limited surface area ensures effective Auger non-radiative recombination, resulting in highly efficient single-photon generation with values of photon purity as low as . The observed long-period blinking and bleaching, typical of such thin shells, can be easily reduced by increasing the shell thickness. This work establishes NPLs as new single-photon sources very well suited for integration into quantum photonic systems
In vivo autofluorescence lifetime imaging of the Drosophila brain captures metabolic shifts associated with memory formation
No full textInternational audienceAbstract Neuronal energy regulation is increasingly recognized as a critical factor underlying brain functions and their pathological alterations, yet the metabolic dynamics that accompany cognitive processes remain poorly understood. As a label-free and minimally invasive technique, fluorescence lifetime imaging (FLIM) of coenzymes NADH and NADPH (collectively referred to as NAD(P)H) offers the possibility to resolve cellular metabolic profiles with high spatial precision. However, NAD(P)H FLIM’s capacity to detect subtle changes in neuronal metabolism associated with cognition has not been demonstrated. In this study, we applied NAD(P)H FLIM to map the metabolic profiles of Drosophila neurons in vivo across multiple scales, focusing on the primary centers for associative memory: the mushroom bodies (MBs). At a broad scale, we obtained an overview of the metabolic signatures of the main brain tissue and identified a marked difference between neuropil and cortex areas. At a finer scale, our findings revealed notable heterogeneity in the basal metabolic profiles of distinct MB neuron subtypes. Measurements performed after associative olfactory learning also uncovered a subtype-specific metabolic shift associated with memory formation, demonstrating the utility of NAD(P)H FLIM in detecting physiology-driven changes linked to brain function. These results establish a promising framework for studying cerebral energy dynamics in vivo
DNA topoisomerase I acts as supercoiling sensor for bacterial transcription elongation
No full textInternational audienceDuring transcription, RNA polymerase (RNAP) continuously unwinds and rewinds DNA, generating negative and positive supercoils upstream and downstream, respectively. Using single-particle cryo-EM, we elucidated how bacterial RNAP and DNA topoisomerase I (TopoI), which relaxes negative supercoils, operate in close spatial proximity. TopoI binds to relaxed DNA upstream of RNAP, and this involves a conformational switch in the TopoI functional domains. This suggests that TopoI exerts a sensing role before the formation of negative supercoils. On DNA substrates mimicking negatively supercoiled DNA, TopoI threads one strand into the active site for cleavage and binds the complementary strand with an auxiliary domain. Transcriptomic and phenotypic analyses suggest that mutations affecting conformational changes in TopoI impact gene expression and operon polarity in bacteria. In summary, we propose a comprehensive model for DNA relaxation in the proximity of active bacterial transcription
STARCH AND STARCH-HYDROXYAPATITE AEROGELS AND XEROGELS: HIERARCHICALLY STRUCTURED BIO-BASED MATERIALS FOR POTENTIALCANCELLOUS BONE REPAIR
No full textInternational audiencePorous starches are promising bio-based materials for biomedical use, notably as scaffolds and drug carriers. Potato starch gels were obtained via starch dissolution, retrogradation, solvent exchange and either supercritical drying CO₂ (aerogels) or evaporative drying (xerogels). Incorporating 100-200 µm hydroxyapatite (HA) based-beads produced porous starch-HA xerogels with preserved nanostructure, suitable for bone regeneration. Cytotoxicity tests confirmed material safety. 3Dprinting enabled hierarchical control, making these constructs viable candidates for cancellous bone substitutes
QMCkl: A Kernel Library for Quantum Monte Carlo Applications
No full textQuantum Monte Carlo (QMC) methods deliver highly accurate electronic structure calculations but are computationally intensive. The quantum Monte Carlo kernel library (QMCkl) provides a modular, portable collection of high-performance kernels implementing the core building blocks of QMC calculations. It offers a C-compatible API, supports the TREXIO standard for input, and covers essential QMC kernels including atomic and molecular orbitals, cusp corrections, Jastrow factor, and the necessary derivatives also to perform variational and structural optimization. QMCkl separates algorithmic development from hardware-specific tuning by combining human-readable reference implementations with performance-optimized kernels that produce identical numerical results. The library enables consistent, efficient, and reproducible simulations across different QMC codes and architectures, and achieves substantial speedups in the evaluation of the energy and its derivatives. Beyond QMC, QMCkl can accelerate deterministic quantum chemistry workflows and visualization tools, promoting cross-code interoperability and simplifying high-performance scientific software development
SCOPE: A Self-supervised Framework for Improving Faithfulness in Conditional Text Generation
No full textLarge Language Models (LLMs), when used for conditional text generation, often produce hallucinations, i.e., information that is unfaithful or not grounded in the input context. This issue arises in typical conditional text generation tasks, such as text summarization and data-to-text generation, where the goal is to produce fluent text based on contextual input. When fine-tuned on specific domains, LLMs struggle to provide faithful answers to a given context, often adding information or generating errors. One underlying cause of this issue is that LLMs rely on statistical patterns learned from their training data. This reliance can interfere with the model's ability to stay faithful to a provided context, leading to the generation of ungrounded information. We build upon this observation and introduce a novel self-supervised method for generating a training set of unfaithful samples. We then refine the model using a training process that encourages the generation of grounded outputs over unfaithful ones, drawing on preference-based training. Our approach leads to significantly more grounded text generation, outperforming existing self-supervised techniques in faithfulness, as evaluated through automatic metrics, LLM-based assessments, and human evaluations
Improving Diversity in Language Models: When Temperature Fails, Change the Loss
No full textInternational audienceIncreasing diversity in language models is a challenging yet essential objective. A common approach is to raise the decoding temperature. In this work, we investigate this approach through a simplistic yet common case to provide insights into why decreasing temperature can improve quality (Precision), while increasing it often fails to boost coverage (Recall). Our analysis reveals that for a model to be effectively tunable through temperature adjustments, it must be trained toward coverage. To address this, we propose rethinking loss functions in language models by leveraging the Precision-Recall framework. Our results demonstrate that this approach achieves a substantially better trade-off between Precision and Recall than merely combining negative log-likelihood training with temperature scaling. These findings offer a pathway toward more versatile and robust language modeling techniques
Genetics of prelingual isolated deafness and Usher syndrome in the Maghreb and Jordan: Harnessing the potential of homozygosity
No full textInternational audienceThe molecular genetic diagnosis of prelingual sensorineural hearing impairment (HI) is essential for genetic counseling and patient management. Effective diagnosis requires a knowledge of the genetic architecture of HI, which is often lacking. We established a cohort of 450 unrelated patients with familial (at least two affected relatives) severe-to-profound bilateral prelingual HI in five countries with high consanguinity rates: Tunisia, Jordan, Algeria, Morocco, and Mauritania (the TJAMM cohort). Recessive and dominant inheritance were observed in 92% and 8% of cases, respectively; 14% were syndromic. Genome analysis detected 211 different mutations (36% not reported before) in 49 deafness genes, and fully resolved 90% of cases of autosomal recessive isolated deafness (DFNB forms), 89% of the mutations being homozygous. The deafness genes involved were similar in different countries, but their mutations, except a few in GJB2 and LRTOMT , differed considerably, suggesting an overrepresentation of private mutations. Biallelic missense mutations in MYO7A , CDH23 , PCDH15 , USH1C cause either DFNB forms or Usher syndrome type 1 (USH1) ( USH1/DFNB genes). Such mutations were overrepresented (13% of patients), highlighting the importance of distinguishing between these two mutation classes. We hypothesized that current difficulties might stem from the misclassification of certain mutations. By studying the 65 USH1/DFNB missense mutations reported to cause DFNB in the homozygous state, we identified some that, when associated with a loss-of-function mutation, resulted in USH1, a characteristic pattern of some recessive hypomorphic mutations. This reappraised classification of USH1/DFNB mutations has the potential to improve molecular diagnosis and patient management significantly