HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Vers une synthèse plus verte de molécules π-conjuguées pour les cellules solaires à pérovskite
No full textThe thesis focuses on the design and synthesis of novel hole transport materials for perovskite solar cells, employing palladium/platinum nanoparticles as catalysts within a green framework. After highlighting the shortcomings of conventional HTMs their high costs, humidity-induced instability, and the complexity of organic and inorganic alternatives, the work centers on using Pd/Pt nanocatalysts to achieve up to 88% yields in coupling reactions while significantly reducing both waste and energy consumption. lntegrating these catalysts into the synthesis of dibenzofuran and dibenzothiophene core HTMs, conducted in benign solvents, streamlines the process by cutting down on the number of steps and the use of halogenated solvents. The resulting materials deliver a photovoltaic efficiency of 13. 76%. Moreover, by exploring precursors derived from lignin, this research paves the way toward more environmentally friendly HTM production and the industrialization of circular processes applicable to other optoelectronic technologies.La thèse porte sur la conception et la synthèse de nouveaux matériaux de transport de trous pour cellules solaires pérovskites, en utilisant des nanoparticules de palladium/platine comme catalyseurs dans une démarche résolument verte. Après avoir mis en évidence les faiblesses des HTM conventionnels, leurs coûts élevés, instabilité face à l'humidité et complexité des alternatives organiques ou inorganiques, le travail se concentre sur l'utilisation de nanocatalyseurs Pd/Pt pour atteindre des rendements allant jusqu'à 88 % pour les réactions de couplage, tout en réduisant significativement déchets et consommation énergétique. Leur intégration dans la synthèse de HTM à noyau dibenzofurane et dibenzothiophène, réalisée en solvants sains, simplifie le processus en diminuant le nombre d'étapes et l'usage de solvants halogénés. Les matériaux obtenus permettent d'atteindre une efficacité photovoltaïque de 13, 76 %. En explorant par ailleurs l'emploi de précurseurs issus de la lignine, cette recherche ouvre la voie à une production d'HTM plus respectueuse de l'environnement et à l'industrialisation de procédés circulaires applicables à d'autres technologies opto-électroniques
Tailored gaussian process modeling for polycristalline texture representation
No full textInternational audienceA tailored Gaussian Process Regression (GPR) model is proposed for the reconstruction of pole density functions in texture analysis. This approach incorporates spherical-periodic distance measures into conventional stationary kernels to effectively capture localized texture features. A key innovation is the introduction of a log-linear data transformation that ensures non-negativity of both interpolated function values and stochastic intervals, leading to physically consistent reconstructions. The proposed approach is systematically evaluated on synthetic texture datasets, examining the influence of distance measures, kernel selection, and hyperparameter optimization. Moreover, comparisons with the conventional spherical harmonics method will be presented during the conference
Eco-Efficient Synthesis and In Vitro Evaluation of Vanillin Derivatives for Antifungal Use against Phytopathogens
No full textInternational audienceVanillin is mainly synthesized chemically but holds significant potential for biomass valorization as it can be extracted from lignin. Recognized for its herbicidal, insecticidal, antimicrobial, and antioxidant activities, vanillin can undergo structural modifications that may reinforce these properties. An eco-efficient methodology was tested that allows the synthesis of vanillin derivatives with a yield of up to 60% of purified products. The derivatives were studied for their antifungal effect by inhibition tests on 11 phytopathogens notably infecting wheat and rapeseed. Three out of seven compounds inhibited in vitro mycelial growth by 80% for most of the studied microorganisms at a concentration of 50 mu g mL-1. They displayed in vitro antifungal activity similar to that of the systemic fungicide tebuconazole. Moreover, they showed no phytotoxicity and no cytotoxicity, with significantly higher IC50 values for human skin fibroblasts and bronchial cells than tebuconazole. The broad in vitro antifungal efficiency of these vanillin derivatives, combined with their measured innocuity for plants and human cells, indicates that they could be further studied as potential active substances for alternative non-petro-sourced fungicides to control crop fungal diseases
Automated Skill Decomposition Meets Expert Ontologies: Bridging the Granularity Gap with LLMs
No full textInternational audienceThis paper investigates automated skill decomposition using Large Language Models (LLMs) and proposes a rigorous, ontologygrounded evaluation framework. Our framework standardizes the pipeline from prompting and generation to normalization and alignment with ontology nodes. To evaluate outputs, we introduce two metrics: a semantic F1-score that uses optimal embedding-based matching to assess content accuracy, and a hierarchy-aware F1-score that credits structurally correct placements to assess granularity. We conduct experiments on ROME-ESCO-DecompSkill, a curated subset of parents, comparing two prompting strategies: zero-shot and leakage-safe few-shot with exemplars. Across diverse LLMs, zero-shot offers a strong baseline, while few-shot consistently stabilizes phrasing and granularity and improves hierarchy-aware alignment. A latency analysis further shows that exemplar-guided prompts are competitive -and sometimes faster -than unguided zero-shot due to more schema-compliant completions. Together, the framework, benchmark, and metrics provide a reproducible foundation for developing ontologyfaithful skill decomposition systems
Simulation of membrane-assisted electroseparation of binary suspension
No full textInternational audienceA two-dimensional Monte Carlo model for a description of electroseparation of a binary suspension of small and big particles placed between two membranes is developed. It is assumed that the membranes are permeable only to small particles with diameter d. The ratio of particle diameters D/d is varied within the interval 2–10. The direction of the electric field is changed with the period TE. The effects of the particles concentration and the distance between the electrodes and the electrophoretic velocity on the electroseparation kinetics are studied. Behavior of a total electroseparation time tf required for the elimination of all small particles from the system is discussed. At some optimal values of TE, the minimum value of tf was observed
Imposing Constraints in Probabilistic Circuits via Gradient Optimization
No full textInternational audienceProbabilistic Circuits (PCs) are a class of tractable models that allow a range of efficient and exact computations while achieving state-of-the-art performance in some domains. In this work, we propose a sample-based procedure to let the distribution encoded by a PC satisfy probabilistic propositional logic constraints. This sample-based method is proposed as a direct competitor to a mathematical approach previously introduced, which is based on optimizing a convex upper bound of the KL divergence. In our empirical study, we compare both methods in two different scenarios, where constraints are utilized to: i) apply fairness to a distribution; and ii) improve the performance of a PC model under scarce data. Our results indicate that although both methods are competitive to one another in the case of fairness, sample-based method has an advantage in scenarios with scarce data
Transcriptomic characterization of the synergy between human induced pluripotent stem cells-derived liver-and pancreas-on-chip coculture
No full textInteractions between the liver and pancreas are key features of the carbohydrate and lipid homeostasis in healthy and pathological patients. To investigate the crosstalk between the two organs, we have developed an organ-on-chip coculture model derived from human induced pluripotent stem cells. The presence of pancreaticderived tissue in the culture environment contributed to increase the CYP3A4 activity, the glycogen storage, and the expression of genes related to lipids, bile acids and sterol metabolism in the liver derived tissue. Concomitantly, the presence of liver cells led to increase the C-peptide secretion in pancreas. The coculture with liver modulated the pancreatic differentiation by increasing the activity of important transcription factors (REST, MAFB, PBX1) and by downregulating several hormone encoding genes (INS, GCG, TTR). The liver also stimulated the expression of genes involved in the response to inflammation in pancreas (via TGFb/SMAD pathway). In parallel we observed a pancreatic cell reorganization coupled with the activation of the cell proliferation related transcription factor (SCRT1) and the upregulation of cellular remodeling genes (FLNA, FLNB, FN1, COL4A5). Finally, the pancreatic lipid genes were also upregulated in presence of the liver tissue. Overall, our results reflect a complex synergy between both tissues. We believe that those results are an encouraging step toward the development of relevant human model using advanced organ on chip technology and stem cells sources.</div
A geometrically exact thin-walled rod model with warping and stress-resultant-based plasticity obtained with a two-level computational approach
No full textInternational audienceIn this work, we propose a weakly coupled multiscale-like approach to derive a seven degree-offreedom kinematically exact rod model for thin-walled members with elastoplastic-hardening constitutive equation. The novelty lies in the macro-level description, where the effects of coupled elastoplastic-geometrical local phenomena are fully characterized in terms of crosssectional stress resultants and generalized rod strains, in a fully 3D context. Torsion-warping degrees of freedom and arbitrary (plastic) failure mode capabilities are inherently present in the formulation, allowing for the modelling of complex structural behavior in thin-walled members. The microscale is based on a kinematically exact shell or 3D-solid model with usual von-Mises plasticity and linear hardening, weakly coupled to the macroscale. At this finer scale, simulations are performed in a pre-process stage, with the resulting equivalent stress-resultantbased hardening plastic parameters transferred to the macroscale. This rather phenomenological representation of complex local effects may satisfactorily replicate the overall behavior of thin-walled members consisted of ductile materials, such as, but not only, steel or aluminum beam/column profiles. Numerical solution is carried in the framework of operator split, whereby, local variables are solved in an element-wise fashion through numerical condensation, thus not bringing any extra DOFs to the macroscale. The model is implemented in an in-house finite element program for the analysis of flexible thin structures and is validated against reference solutions.</div
A simple geometrically exact finite element for thin shells: part 2—dynamics
No full textInternational audienceThis paper is the continuation of the previous article [1] which has introduced a new triangular nonlinear shell finite element, denoted as T6-3iKL, designed to handle large displacements and rotations. This second part focuses on adapting the dynamic algorithm from [2] and [3] to deal with long-term dynamics. The element features 6 nodes, a quadratic displacement field, and a linear rotation field using Rodrigues incremental rotation parameters, which have been the biggest contribution from previous work, resulting in 21 degrees of freedom. The kinematic model from present element integrates principles from shear-rigid shell theory . The way the rotation field is parameterized in this kinematic model enables rotation continuity between adjacent elements through a single scalar at midside nodes , facilitating multiple branch connections in the mesh without introducing artificial parameters like penalties or Lagrange multipliers. The numerical implementation of the model is validated through comparisons with different references, demonstrating the consistency and reliability of the formulation. The proposed triangular shell element, characterized by its versatility, simplicity in kinematics, low number of DOFs, no need for artificial parameter calibration, geometric exactness, and compatibility with 3D material models, offers an effective solution for shell simulation in countless engineering applications.</div
Nonlinear Dynamics and Control of Reissner's 2D Geometrically Exact Beam by Distributed Port-Hamiltonian System
No full textInternational audiencePort Hamiltonian systems formalism 1 is proposed for providing the general control theory for finite dimensional systems, with the models typically used in multibody dynamics (such as rigid components interconnected with flexible joints, or ports). Many present applications require better modeling of the system flexibility (and risk of damage), and one has to consider infinite dimensional systems. The nonlinear dynamics and control of such a system in terms of Reissner's geometrically exact beam is studied in this work. More precisely, we first present the theoretical formulation for nonlinear dynamics for 2D Reissner's beam constructed as port-Hamiltonian system. This results in highly nonlinear problem due to nonlinear beam kinematics capable of representing, finite displacements, rotations and strains. The port Hamiltonian formulation suitable for (nonlinear) control problem is then developed by selecting appropriate effort and flow variables, and the model is reformulated as a coupled system of first-order partial differential equations in a structure-preserving format in continuum setting. We then develop an expanded format required for nonlinear system and the corresponding variational formulation by using the principle of virtual power, with the boundary conditions defining the port variables that are used in control. The final step is the finite element discretisation by using finite element interpolations for such nonlinear port-Hamiltonian formulation, resulting with a set of nonlinear ordinary differential equations with nodal degrees that count displacements, rotation, linear and angular velocities, forces and moments, which provides the greatest flexibility in choosing control strategies. This set of differential equations is here integrated by the backward Euler scheme, resulting in a nonlinear system of algebraic equations. The consistent linearization of such system provides a robust performance for the proposed port Hamiltonian formulation. This is illustrated with the results of several numerical simulations that confirm the improved performance in energy conserving which is superior to those provided previously by energy conserving time integration schemes that have been constructed for fully discretised problems 2 .</div