Publications

2025

• NeuROM
  
  • Daby-Seesaram Alexandre
  • Škardová Kateřina
  • Genet Martin
  , 2025. No description provided. (10.5281/zenodo.13772740)
  DOI : 10.5281/zenodo.13772740
• Eyring theory for plasticity in amorphous polymers violates Curie's principle
  
  • Merlette Thomas C
  • Masnada Elian
  • Sotta Paul
  • Long Didier R
  Soft Matter, Royal Society of Chemistry, 2025, 21 (13), pp.2502-2508. (10.1039/d4sm00894d)
  DOI : 10.1039/d4sm00894d
• Remanent magnetic response of hard magnetorheological elastomer foams: Fabrication, microstructure characterization and modeling
  
  • Lin Zehui
  • Hooshmand Zahra
  • Danas Kostas
  • Bodelot Laurence
  Journal of Magnetism and Magnetic Materials, Elsevier, 2025, pp.172954. This work deals with the experimental, numerical and theoretical study of the purely magnetic response of hard magnetorheological elastomer (h-MRE) foams of variable particle and porosity content. First, the fabrication and experimental measurement of the remanent magnetic flux of the h-MRE foams are presented. We find that at lower particle content, the foam comprises closed-cell porosity, with the voids having a variable size and ellipsoidal shape. As the particle content in the matrix increases, the voids become smaller in size and more spherical in shape, while the porosity decreases. We show experimentally that the remanent magnetic flux is entirely independent of the shape and orientation of the voids. Image-based morphological analysis of the h-MRE foam microstructure subsequently allows to reconstruct numerically unit-cells that share the same statistics as those of the experimental foams. These unit-cells are used to construct an explicit theoretical model with magnetic dissipation. We show that the remanent magnetization is a linear function of the overall particle volume fraction in the foam. The model is further used to scale up the analysis and solve the experimental boundary value problem of a permanently magnetized h-MRE cube, and the numerical estimates show excellent agreement with the experiments. Finally, the numerical model is shown to match available analytical solutions for the remanent magnetic flux of parallelepiped magnets. (10.1016/j.jmmm.2025.172954)
  DOI : 10.1016/j.jmmm.2025.172954
• Wavelength selection in the twist buckling of pre-strained elastic ribbons
  
  • Kumar Arun
  • Audoly Basile
  Journal of the Mechanics and Physics of Solids, Elsevier, 2025, 196, pp.106005. A competition between short- and long-wavelength twist buckling instabilities has been reported in experiments on thin elastic ribbons having pre-strain concentrated in a rectangular region surrounding the axis. The wavelength of the twisting mode has been reported to either scale (i) as the width of the ribbon when the pre-strain is large (short-wavelength case) or (ii) as the length of the ribbon when the pre-strain is small (large-wavelength case). Existing one-dimensional rod or ribbon models can only account for large-wavelength buckling. We derive a novel one-dimensional model that accounts for short-wavelength buckling as well. It is derived from non-linear shell theory by dimension reduction and captures in an asymptotically correct way both the non-convex dependence of the strain energy on the twisting strain~$\tau$ (which causes buckling in a first place) and its dependence on the strain gradient $\tau'$. The competition between short- and long-wavelength buckling is shown to be governed by the sign of the incremental elastic modulus $B_0$ associated with the twist gradient $\tau'$. The one-dimensional model reproduces the main features of equilibrium configurations generated in earlier work using 3D finite-element simulations. In passing, we introduce a novel truncation strategy applicable to higher-order dimension reduction, that preserves positiveness of the strain energy even when the gradient modulus $B_0$ is negative. (10.1016/j.jmps.2024.106005)
  DOI : 10.1016/j.jmps.2024.106005
• A benchmark for elasto-plasticity in finite strain
  
  • Lesueur Louis
  • Thorin Anders
  • Weisz-Patrault Daniel
  , 2025. Finite strain elasto-plastic simulations are critical in fields such as materials science (metal forming, forging, additive manufacturing) and automotive engineering (crash simulations). These simulations are traditionally carried out using computationally intensive finite element analysis (FEA), which limits their use in optimization tasks (e.g., optimal control, design processes) and real-time applications (e.g., tele-operation, personnel training). In this work, we introduce a benchmark dedicated to highly non-linear elasto-plastic simulations, designed to evaluate and develop neural network models tailored for solving elasto-plastic problems under finite strain conditions, ultimately unlocking the potential for real-time optimization and interactive simulations. The datasets include simulations of 1D and 3D elements, featuring quasi-static sequences of applied loads on complex geometries, and the resulting computed quantities: displacements fields, plastic flow coefficient field, stresses. To specifically evaluate the impact of plasticity on different neural networks, the datasets also feature simulations with identical inputs but employing a purely elastic constitutive law.
• Gradient effects in discrete elastic networks captured by asymptotic homogenization
  
  • Ye Yang
  , 2025. Architected structures have been more and more popular in a wide range of engineering applications due to their exotic mechanical properties, such as lightweight, high-strength, wave transmission, and shock-absorbing features. Their unique mechanical behavior can be finely tuned by altering their topology. Recent advancements in additive manufacturing have made it possible to fabricate truss-based architected structures with precisely designed and customizable microstructures. A straightforward approach to predicting the mechanical behavior of these lattice structures is to carry out full discrete simulations. However, this approach is restricted by the scale and complexity of the structure, making it impractical to study large or intricate designs. This limitation highlights the necessity of adopting a homogenized model.Recent work shows Cauchy-continuum obtained by the classical homogenization method is limited, for example, it cannot capture the size effect and ignores the gradient effect. This Ph.D. explores a general energy approach to the higher-order homogenization of discrete elastic networks made up of linear elastic beams or springs in dimension 2 or 3. The homogenized energy achieves asymptotic accuracy two orders beyond the results of classical homogenization. The model is validated through a series of examples, including lattice with various topologies, and varying geometric or material properties, with or without prestrain/prestress, either in 2D or 3D. For each example, we utilize the shoal library, an open-source homogenization engine designed for the homogenization of discrete structures, to homogenize the target lattice and we compare the homogenization results with those extracted from full, discrete simulations to evaluate the accuracy of our homogenized model. Our findings show that the asymptotic homogenized model remains highly accurate even when the strain gradient effects are significant.We further extend the linear homogenization method to analyze the Kagome lattice with springs under finite deformation. The Kagome with springs has mechanisms due to its topology and recent work shows that Cauchy's continuum is not enough. By introducing an enriched kinematic variable and applying a leading-order two-scale expansion, we obtain an asymptotic, enrichment homogenized model for the Kagome lattice with finite deformation. We present a 1D and a 2D example to demonstrate that the model can effectively predict deformation under applied loads by solving the corresponding boundary value problems. Our effective model is capable of capturing some interesting phenomena, such as the (left-right) asymmetry in the deformation of the discrete structure under symmetric boundary conditions. Moreover, it predicts that this asymmetry progressively decreases when the microscopic length scale is refined, ultimately achieving perfect symmetry as the microscopic length approaches zero.The homogenization approach used in this Ph.D. can provide a versatile and efficient tool to analyze the mechanical performance of discrete structures and is powerful for designing and optimizing microstructures.
• What is personalized lung poromechanical modeling and how can it improve the understanding and management of fibrotic interstitial lung diseases?
  
  • Brillet Pierre-Yves
  • Peyraut Alice
  • Bernaudin Jean-François
  • Fetita Catalin
  • Nunes Hilario
  • Genet Martin
  Expert Review of Respiratory Medicine, Taylor & Francis, 2025. (10.1080/17476348.2025.2464886)
  DOI : 10.1080/17476348.2025.2464886
• Wear behavior of 316L stainless steel obtained by additive manufacturing (laser metal deposition process)
  
  • Zurcher Théo
  • Charkaluk Eric
  • Fridrici Vincent
  Wear, Elsevier, 2025, pp.205948. (10.1016/j.wear.2025.205948)
  DOI : 10.1016/j.wear.2025.205948
• mechanical and numerical modelling of transient shear wave elastography for the cornea
  
  • Merlini Giulia
  , 2025. Recent advances in dynamic elastography have enabled rapid, localized, and non-invasive acquisition of corneal mechanical data, making these techniques promising for in-vivo applications. The primary aim of this thesis is the development of a computational model that could reproduce transient elastographic measurements and give access to mechanical properties of the cornea as a diagnostic tool. As a first step, we derive a variational formulation for linear wave propagation in hyperelastic, nearly incompressible and preloaded solids. The numerical approximation of wave propagation problems in nearly or pure incompressible solids faces several challenges such as locking and stability constraints. In this work we propose a stabilized Leapfrog scheme based on the use of Chebyshev polynomials to relax the stability condition. A 3D nonlinear static model of the cornea is developed to simulate the tissue's preloaded state under intraocular pressure. The anisotropic behavior, given by the presence of a complex collagen network within the cornea, is described with a microsphere model. By linearization, the constitutive behavior of the tissue is derived for the wave propagation problem, which is then solved using the previously described numerical method. Furthermore, we address the case where the assumption of linear wave propagation may not be sufficient. We propose the application of a novel numerical method for solving nonlinear elastodynamic problems. This fully explicit, three-step method ensures stability through a posteriori energy criterion. The nonlinear static model of the cornea is then coupled with the nonlinear wave propagation model. Finally, the impact of fluid interaction on elastic waves is studied using guided wave theory, modeling the cornea as an incompressible waveguide and solving the problem as a generalized eigenvalue problem with boundary conditions for fluid interaction.
• Finite strain formulation of the discrete equilibrium gap principle: application to direct parameter estimation from large full-fields measurements
  
  • Peyraut Alice
  • Genet Martin
  Comptes Rendus. Mécanique, Académie des sciences (Paris), 2025, 353 (G1), pp.259-309. The Equilibrium Gap Method (EGM) is a direct model parameter identification method, i.e., that does not require any resolution of the model. It has been extensively studied in the context of small strains but not thoroughly investigated for large strains. In this article, we propose a novel formulation of the EGM, valid in large strains, and applicable to both boundary and body forces, when full-field measurements are available. Our formulation is based on a recently proposed continuous formulation and consistent discretization of the equilibrium gap principle. Additionally, we developed an estimation pipeline to quantify the robustness of our new EGM formulation to noise, and we compared its performance to other classical estimation methods, namely the Finite Element Model Updating (FEMU) method and the Virtual Fields Method (VFM). Our robustness quantification pipeline involves generating synthetic data from a reference model through two methods: by adding noise to the reference displacement, or by generating noisy images and performing motion tracking with the Equilibrium Gap principle used as mechanical regularization. While the quality of estimation using our new EGM formulation is poor with the first data generation method, it improves drastically with the second method. Since the second method of synthetic data generation closely mimics experimental processes, the EGM, when combined with motion tracking with Equilibrium Gap regularization, demonstrates reasonable noise robustness. Thus, it is a promising option for direct parameter estimation from full-field measurements. (10.5802/crmeca.279)
  DOI : 10.5802/crmeca.279
• Deformation of synthetic rock salt investigated by X-Ray micro-computed tomography: Effect of brine, confining pressure and loading rate
  
  • Du Nina
  • Bornert Michel
  • Dimanov Alexandre
  • Aimedieu Patrick
  • King Andrew
  , 2025, pp.438-447. The development of micro-cracks in rock salt has been studied in 3D. In order to investigate the influence of brine, dry and humid materials have been synthetized by powder compaction. We present results of in situ X-Ray micro-tomography triaxial tests performed on the PSICHE beamline of Synchrotron SOLEIL. We have explored different loading rates and confining pressures for both types of synthetic rock salt.
• Altering microstructure and enhancing mechanical properties during direct energy deposition of Ti-6Al-4V via in-process laser heat treatments
  
  • Abdesselam Kouider Abdellah
  • Gaudez Steve
  • van Petegem Steven
  • Honkimäki Veijo
  • Hallais Simon
  • Cornet Louis
  • Vallet Maxime
  • Upadhyay M V
  Materials & Design, Elsevier, 2025, 254, pp.113997. Le traitement thermique par laser en cours de fabrication est étudié comme une alternative aux traitements thermiques post-processus tels que le recuit, afin de modifier la teneur en phases hcp (α′ et αd) et bcc (β). Trois échantillons ont été fabriqués à l'aide d'une machine miniature de dépôt métallique par laser, le laser fonctionnant à 300 W. L'échantillon de référence n'a pas été soumis à un LHT. Pour les autres échantillons, chaque couche déposée a été soumise à un LHT supplémentaire à 100 W (LHT100) pour un échantillon et à 300 W (LHT300) pour l'autre, sans ajout de poudre. Le LHT100 a montré une amélioration globale du compromis résistance/ductilité. En revanche, le LHT300 a permis d'obtenir une résistance plus faible, mais une ductilité et une ténacité beaucoup plus élevées que celles des autres échantillons. L'analyse par diffraction des rayons X synchrotron des échantillons tels que construits a révélé une différence insignifiante entre la teneur en phases hcp et bcc entre les différents échantillons. Cependant, la microscopie électronique à balayage a révélé un effet significatif des LHT sur les fractions des différentes phases hcp. Parallèlement à la décomposition de α′ en αd et β, la preuve d'une transformation massive de bcc β en hcp αm a également été trouvée dans tous les échantillons. Les résultats montrent une amélioration prometteuse du compromis résistance/ductilité du matériau, démontrant le potentiel des LHT en cours de fabrication pour modifier les microstructures et adapter les propriétés mécaniques. (10.1016/j.matdes.2025.113997)
  DOI : 10.1016/j.matdes.2025.113997
• Development of a rabbit model of uterine rupture after caesarean section, Histological, biomechanical and polarimetric analysis of the uterine tissue
  
  • Debras Elodie
  • Maudot Constance
  • Allain Jean-Marc
  • Pierangelo Angelo
  • Courilleau Aymeric
  • Rivière Julie
  • Dahirel Michèle
  • Richard Christophe
  • Gelin Valérie
  • Morin Gwendoline
  • Capmas Perrine
  • Chavatte-Palmer Pascale
  Reproduction & Fertility, Bioscientifica Ltd, 2025, 6 (4), pp.e-250018. Uterine rupture is a major complication of caesarean section (CS) associated with a high foetal and maternal morbidity. The objective is to develop an in-vivo model of uterus healing and rupture after CS in order to analyse histological phenomena controlling scarring tissue development and potential cause of defects. Eighteen pregnant primiparous female rabbits were bred naturally. At caesarean, after 28 days of gestation, foetuses were either extracted through a longitudinal incision in one of the uterine horns (“CS horn”) or via a short incision at the tip of the contralateral horn (“control horn”). The uterine horns were sutured by single layer, all by the same surgeon. They were mated again 14 days later and euthanized at G28. Genital tracts were collected for histological, biomechanical and polarimetric analyses. Macroscopically, 2/18 presented a dehiscence and 1/18 a spontaneous rupture. The mean thickness of the scarred area was significantly lower 0.9 mm [0.7-1.4] that the non-scarring area on CS horns 2.2 [1.6-2.3] or control horns 2 [1.5-2.3] (p<0.0001). The scar zone was statistically more fibrous (p<0.0001), containing fewer vessels (p=0.03) and oestrogen (p<0.001) and progesterone receptors (p<0.0001). After balloon inflation, ruptured occurred in the scar zone in 8 out of 17 cases (47%). Polarimetry revealed that the scar zone was statistically inhomogeneous (73%). Multifactorial analysis allowed to identify groups with poor uterine healing and less resistant to rupture (balloon inflation) mostly in case of thin myometrium in the scar and a group with strong resistant to rupture and correct healing characteristics. Lay summary Caesarean section rates are rising across the world. When a caesarean section is carried out, it can lead to scarring on the uterus that can affect its resistance to pressure. During the next pregnancy, the uterus can tear, increasing risks to the mother and baby. We carried out caesarean sections in a rabbit, allowing us to analyse the scar on the uterus, the healing and tissue resistance. The scarred part of the uterus was statistically thinner, more fibrous and contained fewer vessels and hormone receptors than the area without scarring. Under similar conditions, poor healing was observed in some animals, reducing resistance in following pregnancies. These results suggest that individual and genetic factor have an effect on healing after a caesarean section. This study may enable us to improve our knowledge and management care for patients who have a caesarean section in order to reduce complications. (10.1530/raf-25-0018)
  DOI : 10.1530/raf-25-0018
• A continuum magneto-viscoelastic model for isotropic soft magnetorheological elastomers: experiments, theory and numerical implementation
  
  • Danas Kostas
  • Nakano Masami
  • Sebald Gaël
  Mechanics of Materials, Elsevier, 2025, 200, pp.105187. This study deals with the experimental, theoretical and numerical investigation of the nonlinear viscoelastic response of magnetically soft magnetorheological elastomers (commonly known as -MREs and denoted here simply as MREs) subjected to combined magnetic and simple shear loads. We consider a fairly soft mechanically MRE. The experiments show a strong effect of the magnetic field on the resulting viscosity and hence dissipated energy expanded by the material during a simple shear cycle. Moreover, the effect of frequency on the response is weak indicating strongly nonlinear viscous effects similar to non-Newtonian fluids. An analytical magneto-viscoelastic model is proposed exhibiting magneto-mechanical coupling at both equilibrium and non-equilibrium energies as well as on the dissipation potential. The model is calibrated by solving in a semi-analytical way a simplified boundary value problem (BVP) of an infinite thin MRE strip embedded in an infinite air domain. These simplified solutions are cross-validated by full-field finite element simulations of the experimental setup showing very good agreement between the experimental data and model estimates. This illustrates the validity of the simplified material model for the proposed experimental setup and sets the ground for a more universal experimental protocol to characterize properly the finite strain response of MREs more generally. (10.1016/j.mechmat.2024.105187)
  DOI : 10.1016/j.mechmat.2024.105187
• Solving inverse source wave problem from Carleman estimates to observer design
  
  • Boulakia Muriel
  • de Buhan Maya
  • Delaunay Tiphaine
  • Imperiale Sébastien
  • Moireau Philippe
  Mathematical Control and Related Fields, AIMS, 2025. In this work, we are interested by the identification in a wave equation of a space dependent source term multiplied by a known time and space dependent function, from internal velocity or field measurements. The first part of the work consists in proving stability inequalities associated with this inverse problem from adapted Carleman estimates. Then, we present a sequential reconstruction strategy which is proved to be equivalent to the minimization of a cost functional with Tikhonov regularization. Based on the obtained stability estimates, the reconstruction error is evaluated with respect to the noise intensity. Finally, the proposed method is illustrated with numerical simulations, both in the case of regular source terms and of piecewise constant source terms. (10.3934/mcrf.2025007)
  DOI : 10.3934/mcrf.2025007
• Activation of deformation mechanisms in natural salt rock: A microscale investigation using digital image correlation
  
  • Li Xinjie
  • Dimanov Alexandre
  • Bornert Michel
  • Gharbi Hakim
  • Hallais Simon
  , 2025, pp.P415-425. This study examines the deformation behaviour of two natural salt rocks during uniaxial compression, observed by high-definition optical imaging. In-plane local strain maps were analysed at different loading stages using Digital Image Correlation (DIC), revealing intense localization and strong heterogeneities in relation with the microstructures of the materials. Combined with microstructural characterization via Electron Backscatter Diffraction (EBSD), three key deformation mechanisms were identified: crystal slip plasticity (CSP), grain boundary sliding (GBS), and intra-crystalline cracking. The interaction of these mechanisms is evident in the localization of deformation bands within certain large grains and relative movement at grain boundaries. Quantitative analysis on the relative dominance of CSP and GBS and the heterogeneity of strain fields reveals the influence of microstructure on the complex interaction of micro-mechanisms. (10.1201/9781003637349)
  DOI : 10.1201/9781003637349
• Investigating multi-scale heterogeneity in multi-layer additive friction stir deposition of high-strength aluminum alloys
  
  • Girault Florian
  • Toualbi Louise
  • Barres Quentin
  • Charkaluk Eric
  Materials Science and Engineering: A, Elsevier, 2025, 927, pp.147979. This study investigates the application of multi-layer Additive Friction Stir Deposition (AFSD) for the manufacturing of an AA7075 wall. A particular focus is placed on the material’s structural integrity, including, to the best of our knowledge, the first detailed characterization of the interface between the substrate and the deposited material. The diversity of analytical techniques used provides a detailed understanding of the evolution of microstructure during deposition and as a function of material height. Scanning electron microscopy in conjunction with X-ray diffraction allows for the observation of the evolution of the microstructure, revealing a smooth transition linked to a mechanical gradient. A crystallographic analysis reveals inter- and intra-layer texture variations, indicating that dynamic recrystallization and restoration mechanisms are concomitantly at work in the deposited material zone, as a function of the vertical distance from the tool. Hardness and tensile measurements indicate a non-negligible evolution from the substrate to the last deposited layer, resulting from the overaging of the phase. Finally, a detailed analysis of the interface between the substrate and the deposited material is proposed, which reveals a disturbed microstructure characterized by local heterogeneities in hardness due to significant variations in texture, grain size, and precipitation. All the results are intended to provide highly instructive data regarding microstructural evolution due to thermal cycling both in the deposited material and in the substrate, particularly in the context of the application of the repair of damaged parts. (10.1016/j.msea.2025.147979)
  DOI : 10.1016/j.msea.2025.147979
• Inverse Uncertainty Quantification for Personalized Biomechanical Modeling: Application to Pulmonary Poromechanical Digital Twins
  
  • Peyraut A.
  • Genet M.
  Journal of Biomechanical Engineering, American Society of Mechanical Engineers, 2025, 147 (8), pp.081003. The development of personalized models is a key step for addressing various problems, especially in biomechanics. These models typically include many constants, introduced in the model material law or loading definition, and their estimation is crucial for the model personalization. However, performing solely the estimation does not yield any information on the estimation accuracy. Additionally, all parameters can typically not be estimated based only on clinical data: some parameters are identified, while others are fixed at generic values. The question of the identifiability of the parameters, along with the robustness of the estimation, notably to measurement errors and to model errors, is therefore crucial and should be quantitatively addressed in parallel to the model development. In this paper, we propose a general inverse uncertainty quantification pipeline based on the creation of synthetic data—for which the parameters ground-truth values are known—generated for different noise and model error levels. Estimation is then performed for many realizations of the noise or model errors, as well as parameter initializations, until convergence of the estimated parameters error distributions. This pipeline was applied to a poromechanical lung model for illustration and validation purposes. It provides quantitative information on the actual identifiability of the parameters, and any derived quantity of interest, in the clinical setting. In particular, it allows us to retrieve a confidence interval for each estimated parameter, which represents valuable information for diagnosis or prognosis use of the estimated values. This work is therefore a step toward improving the reliability of digital twins pipelines. (10.1115/1.4068578)
  DOI : 10.1115/1.4068578
• Creep of rock salt under a large range of deviatoric stresses: Insights from tests in mines and rock mechanics laboratories
  
  • Blanco Martín Laura
  • Jiménez-Camargo Jubier
  • Jaworowicz Jerzy
  • Gharbi Hakim
  • Dimanov Alexandre
  • Bornert Michel
  • Brouard Benoit
  , 2025, pp.3-12. The mechanical behavior of rock salt has been most often predicted using phenomenological laws fitted on laboratory tests conducted on centimeter-scale samples. To minimize the impact of stress and temperature extrapolation, such experiments should be conducted along loading paths relevant to underground operations. However, the viscoplastic response of rock salt has been primarily investigated using confined creep tests covering a differential stress range between 5 and 20 MPa, and varying temperatures. More recently, lower deviatoric levels have been investigated, primarily through unconfined creep tests in remote drifts in underground mines. The joint analysis of experimental data in the two differential stress ranges is often delicate due to the use of (i) different salt facies/geographical origins, (ii) different sample preparation and preconditioning methods, and (iii) different scales of displacement measurements. We present results of two creep tests conducted using very similar loading conditions on two salt samples of the same provenance and prepared using the same protocol. One test is performed in a mine drift and the other is performed in a rock mechanics laboratory. The results are consistent with each other, which is promising to enlarge the differential stress range used to develop constitutive models for rock salt
• Composition effect in the thermomechanical behavior of glasses, and its modelization
  
  • Alvarez-Donado R.
  • Sepulveda-Macias M.
  • Tanguy A.
  Physical Review Materials, American Physical Society, 2025, 9 (8). We employed molecular-dynamics simulations to explore comparatively the thermomechanical behavior of two glass materials-an oxide silica glass (SiO2) and a binary Cu-Zr-based metallic alloy (Cu50Zr50)-during shear and elongation deformation cycles. By calculating the energy balance and tracking the temperature evolution of both glasses under deformation cycles, we are able to propose, for each of them, a constitutive law that accurately reproduces the self-heating process due to plastic deformation. These relatively simple constitutive laws involve strain rate sensitivity and a nonlinear temperature dependence of the thermal dilatancy coefficients, as well as strain gradient plasticity. To identify the right parameters, both glasses are equilibrated at very low temperature (10 K), and two independent deformation rates were applied to each sample for each type of deformation. Thermal attenuation is greatly amplified in silica compared to the metallic glass. Moreover, using a precise atomic description of the instantaneous deformation, combined with an exact coarse-graining procedure, we show, in silica, that self-heating is mainly supported by inhomogeneous strain gradient plasticity with nanometric characteristic lengthscales. (10.1103/ghpm-62vt)
  DOI : 10.1103/ghpm-62vt
• Finite Element Neural Network Interpolation. Part I: Interpretable and Adaptive Discretization for Solving PDEs
  
  • Škardová Kateřina
  • Daby-Seesaram Alexandre
  • Genet Martin
  Computational Mechanics, Springer Verlag, 2025. We present the Finite Element Neural Network Interpolation (FENNI) framework, a sparse neural network architecture extending previous work on Embedded Finite Element Neural Networks (EFENN) introduced with the Hierarchical Deep-learning Neural Networks (HiDeNN). Due to their mesh-based structure, EFENN requires significantly fewer trainable parameters than fully connected neural networks, with individual weights and biases having a clear interpretation. Our FENNI framework, within the EFENN framework, brings improvements to the HiDeNN approach. First, we propose a reference element-based architecture where shape functions are defined on a reference element, enabling variability in interpolation functions and straightforward use of Gaussian quadrature rules for evaluating the loss function. Second, we propose a pragmatic multigrid training strategy based on the framework's interpretability. Third, HiDeNN's combined rh-adaptivity is extended from 1D to 2D, with a new Jacobian-based criterion for adding nodes combining h- and r-adaptivity. From a deep learning perspective, adaptive mesh behavior through rh-adaptivity and the multigrid approach correspond to transfer learning, enabling FENNI to optimize the network's architecture dynamically during training. The framework's capabilities are demonstrated on 1D and 2D test cases, where its accuracy and computational cost are compared against an analytical solution and a classical FEM solver. On these cases, the multigrid training strategy drastically improves the training stage's efficiency and robustness. Finally, we introduce a variational loss within the EFENN framework, showing that it performs as well as energy-based losses and outperforms residual-based losses. This framework is extended to surrogate modeling over the parametric space in Part II. (10.1007/s00466-025-02677-3)
  DOI : 10.1007/s00466-025-02677-3
• Avoiding cracks in multi-material printing by combining laser powder bed fusion with metallic foils: Application to Ti6Al4V-AlSi12 structures
  
  • Jamili A.M.
  • Jhabvala J.
  • van Petegem S.
  • Weisz-Patrault Daniel
  • Boillat E.
  • Nohava J.
  • Özsoy A.
  • Banait S.
  • Casati N.
  • Logé Roland
  Additive Manufacturing, Elsevier, 2025, 97, pp.104615. Laser powder bed fusion (LPBF) as an additive manufacturing (AM) technology has emerged as a powerful platform for producing multi-material metallic structures. The main drawbacks of using metallic powders for multi-material printing are related to technical issues (i.e. powder contamination reducing the reusability of the powder) and interfacial defects. This paper attempts to demonstrate the advantages of using a combination of metallic powders and thin foils for printing light titanium-aluminum multi-material structures. An AlSi12 powder was printed using the conventional LPBF process and the behavior of the second material feedstock was investigated using both Ti6Al4V powders and foils. The printing process was simulated numerically using a finite element model (FEM), and characterized experimentally through operando X-Ray diffraction (XRD). For the powder-powder combination, cracking near the interface between the two alloys was considered as a combined effect of residual stresses and the presence of brittle intermetallic compounds (IMCs); both were investigated using nanoindentation. Replacing the Ti6Al4V powder by a foil resulted in a thinner layer of Ti-Al IMCs near the interface, and eliminated the large interfacial cracks. The results from FEM and CALPHAD thermodynamic simulations, supported by operando XRD, indicated that the increased thermal conductivity of the foil, compared to powders, led to heat transfer within the foil and to the underlying LPBF structure, prior to local melting. The new thermal regime produced a flawless interface between Ti6Al4V and AlSi12, due to reduced residual stresses in the plane normal to the building direction, and lower volumes of brittle IMCs. It is concluded that using foils instead of powders mitigates cracking and enhances microstructures near the interface, due to changes in thermal regime and alloys mixing patterns. (10.1016/j.addma.2024.104615)
  DOI : 10.1016/j.addma.2024.104615
• Objective assessment of cardiac function using patient-specific biophysical modeling based on cardiovascular MRI combined with catheterization
  
  • Gusseva Maria
  • Castellanos Daniel Alexander
  • Veeram Reddy Surendranath
  • Hussain Tarique
  • Chapelle Dominique
  • Chabiniok Radomír
  AJP - Heart and Circulatory Physiology, American Physiological Society, 2025, 329 (5), pp.H118-H1191. Synthesizing multi-modality data, such as cardiovascular magnetic resonance imaging (MRI) combined with catheterization, into a single framework is challenging. Different acquisition systems are subjected to different measurement errors. Coupling clinical data with biomechanical models can assist in clinical data processing (e.g., model-based filtering of measurement noise) and quantify myocardial mechanics via metrics not readily available in the data, such as myocardial contractility. In this work we use a biomechanical modeling with the aim 1) to quantitatively compare model- and data-derived signals, and 2) to explore the potential of model-derived myocardial contractility and distal resistance of the circulation (Rd) to robustly quantify cardiovascular physiology. We used 51 ventricular catheterization pressure and cine MRI volume datasets from patients with single-ventricle physiology and left and right ventricles of patients with repaired tetralogy of Fallot. Ventricular time-varying elastance (TVE) metrics and linear regression were used to quantify the relationship between the maximum value of TVE (Emax) and maximum time derivative of ventricular pressure (max(dP/dt)) in data- and model-derived pressure and volume signals at p<0.05. Pearson’s correlations were used to compare model-derived contractility and data-derived Emax and max(dP/dt), and model-derived Rd and data-derived vascular resistance. All data and model-derived linear regressions were significant (p<0.05). Model-derived max(dP/dt) vs. data-derived Emax produced higher R2 than data-derived max(dP/dt) vs. data-derived Emax. Correlations demonstrated significant relationships between most data- and model-derived metrics. This work revealed the clinical value of biomechanical modeling to assist in clinical data processing by providing high-quality pressure and volume signals, and to quantify cardiovascular pathophysiology. (10.1152/ajpheart.00232.2025)
  DOI : 10.1152/ajpheart.00232.2025
• Experiments and modeling of mechanically-soft, hard magnetorheological foams with potential applications in haptic sensing
  
  • Lin Zehui
  • Hooshmand-Ahoor Zahra
  • Bodelot Laurence
  • Danas Kostas
  Journal of the Mechanics and Physics of Solids, Elsevier, 2025, 203, pp.106218. This study proposes a family of novel mechanically-soft and magnetically-hard magnetorheological foams that, upon deformation, lead to robust and measurable magnetic flux changes in their surroundings. This allows to infer qualitatively and even quantitatively the imposed deformation and, eventually from that, an estimation of the stiffness and average stress on the sample even in complex loading scenarios involving combinations of uniform or nonuniform compression/tension with superposed shearing in different directions. The work provides a complete experimental, theoretical and numerical framework on finite strain, compressible magneto-elasticity, thereby allowing to measure and predict coupled magneto-mechanical properties of such materials with different particle volume fractions and then use it to estimate and design potential haptic sensing devices. (10.1016/j.jmps.2025.106218)
  DOI : 10.1016/j.jmps.2025.106218
• Finite element neural network interpolation: Part II—hybridisation with the proper generalised decomposition for non-linear surrogate modelling
  
  • Daby-Seesaram Alexandre
  • Škardová Kateřina
  • Genet Martin
  Computational Mechanics, Springer Verlag, 2025. This work introduces a hybrid approach that combines the Proper Generalised Decomposition (PGD) with deep learning techniques to provide real-time solutions for parametrised mechanics problems. By relying on a tensor decomposition, the proposed method addresses the curse of dimensionality in parametric computations, enabling efficient handling of high-dimensional problems across multiple physics and configurations. Each mode in the tensor decomposition is generated by a sparse neural network within the HiDeNN framework, with an element-based approach presented in Part I, where network parameters are constrained to replicate the classical shape functions used in the Finite Element Method. This constraint enhances the interpretability of the model, facilitating transfer learning, which improves significantly the robustness and cost of the training process. As shown in Part I, the HiDeNN framework can be leveraged to find the optimal spatial and parametric discretisation dynamically during training, which accounts to optimising the network’s architecture on the fly. This hybrid framework offers a flexible and interpretable solution for real-time surrogate modelling. We highlight the efficiency of the proposed neural Network-PGD (NN-PGD) approach through 1D, 2D and 3D benchmark problems, validating its performance against analytical and numerical reference solutions. The framework is illustrated through linear and non-linear elasticity problems, showing the flexibility of the method in terms of changes in physics. (10.1007/s00466-025-02676-4)
  DOI : 10.1007/s00466-025-02676-4