HAL-Paris 13
Not a member yet
33893 research outputs found
Sort by
Overcoming the strength-ductility trade-off in metallic thin films engineering their atomic and microstructure
No full textInternational audienceThe current trend toward miniaturization in devices components in key technologies such as microelectronics, energy production, sensors and wear protection requires the development of high-performance thin films (TFs) with superior mechanical properties combining mutually excluding mechanical properties such as high yield strength and ductility. Achieving such performance, relies on leveraging microstructure-induced mechanical properties, requiring the precise control of TF’s microstructure, atomic composition, grain size and thickness based on nanoengineering design concepts.Here, I will present recent results for several class of advanced TF materials including nanostructured metallic glasses (ZrCu, ZrCuAl) [1, 2] high entropy alloys (CoCuCrFeNi, AlxCoCuCrFeNi) and nanolaminates (fully amorphous, amorphous/crystalline, FCC/BCC) [3, 4], highlighting how the control of the microstructure affect the and micro-scale mechanical behavior and enable ultimate mechanical properties. Among the main results, I will show the fabrication of fully amorphous Zr24Cu76/Zr61Cu39 nanolaminates by magnetron sputtering with controlled nanoscale periodicity (Λ, down to 5 nm), local chemistry and glass-glass interfaces [3]. I will show how the shear band instability can be mitigated when Λ ≤ 50 nm, reaching remarkably large elastic/plastic deformation (16%) and yield strength (~2GPa) by micro-pillar compression [3]. Then, I will present the potential of Pulsed Laser Deposition (PLD) as a novel technique to synthetize nanostructured cluster-assembled metallic glass and HEA-based TF. I will show how PLD-deposited ZrCu metallic glass can reach ultimate yield strength (>3 GPa) and ductility (>6 %). Then, I will focus on the synthesis of crystal/glass ultrafine nanolaminates by PLD [4] in which ∼4 nm Al (crystalline) separate 6 and 9 nm-thick Zr50Cu50 metallic glass nanolayers, showing high yield strength (3.4 GPa) in combination with enhanced elastoplastic deformation (>6%, in compression), while managing to effectively block the percolation of shear bands even at >15% deformation [4]. Finally, I will present, the synthesis of CoCrCuFeNi HEA-TFs with a wide range of microstructures and nanosized grains, enabling strong Hall-Petch strengthening and thermal stability.In the last part of the seminar, I will present recent developments within LSPM, highlighting new research activities dealing with the synthesis of nanostructured metallic films and the development of a new platform for in situ SEM mechanical-electrical characterization.References[1] M. Ghidelli et al., Acta Mater. 213 (2021) 116955.[2] C. Poltronieri et al., Acta Mater. 258 (2023) 119226.[3] A. Brognara et al., Small Struct. (2024) 2400011.[4] F. Bignoli et al., ACS Appl. Mater. Inter. 16(27) (2024) 35686-35696
Purified bone xenografts: A novel and efficient animal bone substitute derived from an optimized supercritical CO2 treatment
No full textInternational audienceBone xenografts represent a promising alternative to autologous or allograft transplants, yet antigenicity in animal-derived tissues remains a major limitation to their clinical use. To provide any risk of contamination or allogenic rejection, the Supercrit® process was developed to treat allogeneic human bone combining a supercritical CO 2 treatment followed by a chemical treatment using high quantities of different solvents. The aim of this study was to produce a xenogeneic bone substitute thanks to the development of a new one-step supercritical process, 'Goxcrit', and to test it in vivo. This new process reduces the use of solvents by injecting them under pressure into the supercritical CO 2 flow, while maintaining the cleaning quality of the bone matrix and better preserving its inner structure, essential for its future bone integration.Porcine derived bone samples were treated using Goxcrit or Supercrit®, and compared with human bone treated with Supercrit®, the commercialized bone allograft. In vitro analyses demonstrated the absence of cytotoxicity and of the alpha-gal epitope mainly responsible for cross-species immunogenicity. Additionally, in vivo experiments revealed improved bone formation in rats critical calvarial defects (BV/TV and von Kossa analyses) implanted with Goxcrit samples, with bone remodeling (TRAP/ALP stains), compared to those treated with Supercrit®. These results can be attributed to the less aggressive chemical process of the Goxcrit, which preserves the bone's inner structure critical for remodeling.Our study highlighted the interest of using a porcine bone source treated with the Goxcrit process to meet the growing demand for reliable and effective bone substitutes.</div
Advancements in dislocation reduction for large area diamond substrates: toward scalable high performance materials
No full textInternational audienceDiamond’s exceptional properties such as high thermal conductivity, high breakdown voltage and high charge carrier mobility make it a prime candidate for next-generation of power electronic devices and high-energy particle detectors [1- 2]. The properties of specific defects in diamond such as NV centers stands as a highly promising platform for quantum technologies. It offers excellent spin properties, including millisecond coherence times at room temperature, paving the way for the development of ultra-sensitive, high-performance quantum sensors, offering new avenues for exploring characteristics inaccessible to conventional devices. However, translating these benefits into real-world applications is hampered by three main challenges: controlling defect densities, producing sufficiently large substrate surfaces, and ensuring reliable material availability for viable market use cases.HiQuTe Diamond is a founded start-up emerging from the LSPM-CNRS, an internationally recognized research center with over 30 years of expertise in diamond growth. Our goal is bridging the gap between state-of-the-art diamond research and commercial product development. By focusing on advanced process engineering and scalable manufacturing approaches, we aim to overcome the hurdles of defect control, substrate size, and material supply.As an example of recent developments through joined effort between HiQuTe and LSPM, we have demonstrated a two-order-of-magnitude reduction in dislocation density from 107 to 105 dislocations/cm² on 1 cm² single-crystal diamond substrates [3] . This progress enables up to 80% of the surface to reach or stay below that threshold. This key milestone paves the way for our exploratory phase on 1-inch substrates, where iterative or combined techniques will be used to further enhance uniformity and yield.Through our dedication to both fundamental research and entrepreneurial agility, our start-up strives to unleash diamond’s full potential for high-performance electronic devices and other emerging applications, positioning diamond as a truly transformative material for the 21st century.References[1] J. Isberg, et al., Diamond and Related Materials, 13 (2004) 320-324.[2] C. Masante, et al., Journal of Physics D: Applied Physics, 54 (2021) 233002.[3] L. Mehmel, et al, Appl. Phys. Lett. 118, 061901 (2021
Diagnostic performance of artefact-reduced cone-beam CT images using a generative adversarial neural network
No full textInternational audienceGenerative adversarial neural networks (GAN) have been demonstrating efficacy in increasing the quality of tomographic images. However, there is a lack of information about its effectiveness in cone-beam CT (CBCT). This study evaluated the performance of GAN in reducing CBCT artefacts. A phantom was custom-made with a mandible covered with Mix-D. Forty teeth were endodontically instrumented and inserted in mandibular sockets. CBCT scans were obtained with a field-of-view of 5 × 5 cm, 90kVp, 3 mA, 0.08 mm voxel size and 9 s. All scans were repeated after enabling metal artefact reduction (MAR). Two conditions with high-density materials were simulated and additional scans were performed. GAN was tailored to use the acquired raw images for validation, training and testing. Structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), mean absolute error (MAE) and root mean square error (RMSE) were calculated. Then, all the images were reconstructed and noise and signal-to-noise ratio (SNR) were measured in the cervical, middle and apical third of the second left premolar root. Also, assessments of the diagnostic confidence, artefact interference and dental and bone preservation were performed by four observers. All the conditions were compared using ANOVA two-way (α = 0,05) and Wilcoxon signed rank test. The corrected images showed significantly higher SSIM and PSNR, lower MAE and RMSE, lower noise, higher SNR, higher diagnostic confidence, lower artefact interference and high dental and bone preservation. The CBCT images with MAR and higher presence of high-density materials showed significantly lower noise values. The developed neural network has promising performance to increase the image quality of CBCT raw and reconstructed images
Édito: Mère-patrie. Le genre au coeur des « politiques migratoires »
No full textInternational audienc
Redesigning deep neural networks: Bridging game theory and statistical physics
No full textInternational audienc
Door-In-Door-Out Times and Functional Outcomes after Stroke in a Large Canadian Hub-and-Spoke Network
No full textInternational audienceABSTRACT Background: The determinants of door-in-door-out metrics (DIDO) at centers referring acute ischemic stroke patients for endovascular thrombectomy (EVT) and the impact of DIDO on functional outcomes are unclear. Our primary objective was to study the association between DIDO and 90-day functional outcomes. Our secondary objective was to investigate the associations between patient clinical and workflow characteristics and DIDO. Methods: We conducted a province-wide multicentric retrospective cohort study in Québec, Canada, of adults with acute ischemic stroke who were transferred from a primary stroke center (PSC) to a comprehensive stroke center for EVT between 2017 and 2020. DIDO was calculated as the time spent in the PSC emergency department. Our co-primary outcomes, assessed 90 days after stroke, were a favorable functional outcome (modified Rankin score of 0–2) and death. We estimated associations between DIDO and co-primary outcomes and between patient characteristics and DIDO using logistic mixed models. Results: Among 790 included patients, the mean age was 69 (+/–14) years, and 400 (51%) were female. The median DIDO was 102 (80–135) minutes. DIDO was not associated with 90-day favorable functional outcome (aOR: 1.00, 95% CI [0.99–1.00], p = 0.54) or death (aOR: 1.00, 95% CI [0.99–1.01], p = 0.69). Arrival at the PSC outside daytime hours (aOR: 3.28, 95% CI [1.26–8.51], p = 0.01) was significantly associated with DIDO ≥ 60 minutes. Conclusions: Although DIDO are long in Québec, they are not associated with 90-day functional outcomes or mortality among patients transferred for EVT. Further research is required to develop strategies to improve modifiable determinants of DIDO, including workflow outside of daytime hours.RÉSUMÉ Les délais arrivée-départ et les résultats fonctionnels après un accident vasculaire cérébral dans un grand réseau en étoile au Canada. Contexte : On ne connaît pas très bien les déterminants des délais arrivée-départ (DIDO) dans les hôpitaux qui dirigent vers des centres spécialisés des patients ayant subi un accident vasculaire cérébral (AVC) ischémique aigu en vue d’une thrombectomie endovasculaire (TEV) ni l’incidence des délais DIDO sur les résultats fonctionnels. L’étude avait pour but principal d’examiner l’association entre les délais DIDO et les résultats fonctionnels au bout de 90 jours et, pour but secondaire, d’analyser les associations entre les caractéristiques cliniques des patients et du flux opérationnel, et les délais DIDO. Méthode : Il s’agit d’une étude de cohorte rétrospective, multicentrique, réalisée à la grandeur du Québec, au Canada, entre 2017 et 2020, chez des adultes ayant subi un AVC ischémique aigu, qui ont été transférés d’un centre de soins primaires des AVC vers un centre de soins intégrés (CSI) en vue d’une TEV. Le délai DIDO a été défini comme le temps passé au service des urgences du centre de soins primaires des AVC. Les deux principaux critères d’évaluation étaient les résultats fonctionnels favorables (sur l’échelle modifiée de Rankin : 0 2) et la mort, évalués au bout de 90 jours après un AVC. Les associations entre les délais DIDO et les deux principaux critères d’évaluation ainsi qu’entre les caractéristiques des patients et les délais DIDO ont été estimées à l’aide de modèles logistiques mixtes. Résultats : Au total, 790 patients ont été retenus dans l’étude, dont 400 étaient des femmes (51 %), et l’âge moyen s’élevait à 69 ans (écart type : 14). Le délai DIDO médian était de 102 minutes (80 135), et il n’a pas été associé à des résultats fonctionnels favorables (risque relatif approché rajusté [RRAR] : 1,00; IC à 95 % [0,99 1,00]; p = 0,54) ni à la mort au bout de 90 jours (RRAR : 1,00; IC à 95 % [0,99 1,01]; p = 0,69). L’arrivée au CSI hors des heures normales de travail le jour (RRAR : 3,28; IC à 95 % [1,26 8,51]; p = 0,01) a été associée de manière significative à un délai DIDO ≥ 60 minutes. Conclusion : Bien que les délais DIDO soient longs au Québec, ils ne sont pas associés aux résultats fonctionnels ou à la mortalité au bout de 90 jours chez les patients qui ont été transférés en vue d’une TEV. Aussi faudrait-il approfondir le sujet afin d’élaborer des stratégies permettant d’améliorer les déterminants modifiables des délais DIDO, y compris le flux opérationnel hors des heures normales de travail le jour
Physiologies et objets
No full textInternational audienceLes Physiologies sont à la mode en France sous la monarchie de Juillet. Ces petits volumes in-18 ou in-32, illustrés, d’une centaine de pages et à petit prix (1 franc), déferlent dans la capitale en déclinant plus de cent trente titres à partir de types sociaux, moraux et professionnels. L’objet ainsi mis à l’honneur participe à une textualisation du social dans le contexte d’émergence des sciences naturelles, de développement des enquêtes, de diffusion de l’hygiénisme et de réévaluation de pseudo-sciences comme la phrénologie et la physiognomonie. Au-delà du seul enjeu marchand et des velléités de thésaurisation, c’est dans ce faisceau sémiotique pluridisciplinaire qu’est saisi l’objet multiple des Physiologies
Évaluation sans référence de la qualité des nuages de points via la saillance visuelle 3D
No full textInternational audienceL'évaluation de la qualité des nuages de points sans référence (en anglais, No-Reference Point Cloud Quality Assessment : NR-PCQA) vise à estimer la qualité perceptuelle d'un nuage de points sans nécessiter de référence, en se rapprochant du jugement humain. Cependant, la plupart des approches existantes n'exploitent pas pleinement les informations visuelles perçues par l'observateur, alors qu'elles jouent un rôle clé dans l'évaluation de la qualité. Dans cet article, nous proposons une nouvelle méthode NR-PCQA intégrant la saillance visuelle 3D afin d'améliorer l'estimation de la qualité. Nous générons d'abord une carte de saillance 3D, qui sert à pondérer le nuage de points en accordant une importance accrue aux régions perceptuellement significatives, celles qui captent spontanément l'attention humaine. Le nuage de points pondéré obtenu est ensuite analysé par un encodeur 3D, qui traite directement les points pour prédire un score objectif de qualité. Les expérimentations menées sur deux bases de données publiques de PCQA montrent que notre approche surpasse la majorité des méthodes existantes. Mots ClefNuage de points, saillance visuelle 3D, évaluation sans références de la qualité.</div