Track E
Computational Mechanics of Materials Across the Scale
ABSTRACTS
Session E-1 Computational mechanics of nanoscale materials
E-1:IL02 Simulations of Structural Phase Transitions in Crystals Using Metadynamics
R. MARTONAK, Faculty of Mathematics, Physics and Informatics, Comenius University in Bratislava, Bratislava, Slovakia
Structural phase transitions in crystals induced by pressure or temperature are complex phenomena of great importance. Besides fundamental interest e.g. in Earth and planetary science, they also allow us to prepare new materials with unique properties. Most of them are reconstructive, thermodynamically first order, and involve breaking and creation of chemical bonds. In past two decades a great progress was achieved in prediction of crystal structures, but a reliable prediction of structural transformations still lags much behind. The main problem is kinetics which ultimately decides about the outcome of the transformation. Understanding kinetics requires a detailed information about free-energy barriers involved in the microscopic transformation pathways. An important ingredient is also nucleation whose understanding in solid-solid transitions is still very incomplete. One of effective approaches to study these phenomena is metadynamics (Laio and Parrinello 2002; Martonak, Laio and Parrinello 2003). Another important ingredient are machine-learning based potentials which nowadays allow to simulate 10^5 or more atoms with high accuracy. In the talk I will present the progress made during past 20 years, illustrate the main achievements and outline possible future directions.
E-1:IL03 Lubricity in Hard and Soft Matter Contacts
A. Silva*, CNR - Istituto Officina dei Materiali (IOM), International School for Advanced Studies (SISSA), Trieste, Italy
At well-characterized microscopic ‘mechanical’ contacts, the two distinctive operative frameworks of hard and soft matter can remarkably meet with a view to understanding the way frictional forces and dissipation mechanisms depend on the geometry and commensurability of the sliding interface more than upon system-dependent characteristics, such as the specific governing forces and their range [1]. This scenario underlies one of the most theoretically fascinating and technologically important concepts of modern tribology, i.e. structural superlubricity, a sliding state of vanishing friction [2]. Here, I plan to line up and discuss recent results related to both structural and dynamic properties of ‘soft’ 2D trapped colloidal systems [3-5] and the nanomanipulation of ‘hard’ graphene suspended [6,7] or deposited on surfaces [8,9].
*In collaboration with: A. Vanossi, C. Bechinger, A. Benassi, T. Brazda, X. Cao, L. Gigli, R. Guerra, S. Kawai, A. Khosravi, D. Mandelli, N. Manini, E. Meyer, E. Panizon, E. Tosatti, M. Urbakh, J. Wang
[1] Vanossi, Bechinger, Urbakh, Structural lubricity in soft and hard matter systems, Nat. Comm. (2020); [2] Wang et al, Colloquium: Sliding and pinning in structurally lubric 2D material interfaces, Rev. Mod. Phys. (2024); [3] Brazda et al, Experimental observation of the Aubry transition in two-dimensional colloidal monolayers, PRX (2018); [4] Cao et al, Orientational and directional locking of colloidal clusters driven across periodic surfaces, Nat. Phys. (2019); [5] Cao et al, Moiré-pattern evolution couples rotational and translational friction at crystalline interfaces, PRX (2022); [6] Mescola et al, Anisotropic rheology and friction of suspended graphene, PRMaterials (2023); [7] Wang et al, Bending stiffness collapse, buckling, topological bands of freestanding twisted bilayer graphene, PRB Letter (2023); [8] Gigli et al, Graphene nanoribbons on gold: understanding superlubricity and edge effects, 2D Materials (2017); [9] Silva, Tosatti, Vanossi, Critical Peeling of Tethered Nanoribbons, Nanoscale (2022)
E-1:IL05 Stretching and Breaking of Ppolymeric Nanofibres
E. Bering1, A.S. de Wijn2, 1Department of Physics and PoreLab, NTNU, Trondheim, Norway; 2Department of Mechanical and Industrial Engineering and PoreLab, NTNU, Trondheim, Norway
Bundles of polymeric materials are ubiquitous and play essential roles in biological systems, and often display remarkable mechanical properties. With the experimental advances in control and manipulation of molecular properties on the nanometric level follows an increasing demand for a theoretical understanding at this scale. This regime of nano-scale bundles of small numbers of molecules has not been investigated much theoretically; here chain–chain interactions, surface effects, entropy, nonlinearities, and thermal fluctuations all play important roles. I will present a broad exploration by molecular-dynamics simulations of single chains and bundles under external loading, using polyethylene oxide (PEO) as a prototype material. Stretching and rearrangements of chains are investigated, as well as their breaking. We have also studied the limits of thermodynamics of stretching in these small systems. If time allows I will also show how this type of simulation can be used to study the dissolution of cellulose.
E-1:IL06 How Temperature- and Electric-field-driven Chain Reorientation affects Friction
M.M. GIANETTI, Dept of Mechanical and Industrial Engineering (MTP), Norwegian University of Science and Technology (NTNU), Trondheim, Norway; R. Guerra, Center for Complexity and Biosystems, Dept of Physics, University of Milan, Milan, Italy; A. Vanossi, CNR-IOM, Consiglio Nazionale delle Ricerche - Istituto Officina dei Materiali and International School for Advanced Studies (SISSA), Trieste, Italy; M. Urbakh, Dept of Physical Chemistry, School of Chemistry, The Raymond and Beverly Sackler Faculty of Exact Sciences and The Sackler Center for Computational Molecular and Materials Science, Tel Aviv University, Tel Aviv, Israel; N. Manini, Dipartimento di Fisica, Università degli Studi di Milano, Milan, Italy
In this talk I will introduce a model for a bilayer of packed zwitterionic monolayers and investigate its tribological response to changes in applied load, sliding velocity, temperature and transverse electric field. The model exhibits different regimes of friction depending on how the chains in the monolayers rearrange according to the mechanical and thermodynamic conditions. The bilayer exhibits a surprising nontrivial nonmonotonic dependence on friction as a function of both temperature and the strength of the applied transverse electric field. We explain these observations by the formation and rupture of transient bonds between molecules belonging to opposite sliding layers. These findings are expected to be relevant to nanoscale rheology and tribology experiments of locally-charged lubricated systems such as, e.g., experiments performed on zwitterionic monolayers, phospholipid micelles, or confined polymeric brushes, as were carried out in surface-force-apparatus setup. These ideas can also be relevant for opening a route to the reversible control of friction forces by modifying the electric polarization of the sliding surfaces.
E-1:IL07 In situ Formation of Superlubricious Surfaces by Mechano-chemical Decomposition of Organic Friction Modifiers
TAKUYA KUWAHARA, Osaka Metropolitan University, Sakai, Osaka, Japan
Superlubricity is a state where the friction coefficient is below 0.01, and achieved by lubricating hard materials, such as amorphous carbon and ceramics, with organic friction modifiers in severe boundary lubrication. In the boundary lubrication, the friction between two sliding bodies is dominated by chemical interactions between two surfaces, and can be improved by utilizing mechano-chemical reactions of the organic friction modifiers resulting in the formation of a low-friction, passivated interface. Here we show recent quantum chemical studies on in situ synthesis of superlubricious carbonaceous nanofilms on amorphous carbon and ceramic materials. Our simulations reveal that decomposition of the lubricant molecules are accelerated by the presence of multiple reactive centers and reaction rates of the lubricant molecules can be controlled by tailoring functional groups. Shear-induced transformation of the lubricant-derived surface layers leads to a drastic decrease of the friction coefficient.
E-1:L09 The Mechanism of Strong Reinforcement of Si Nano-powders by thin Continuous SiC Coatings
K. KAYANG, A.N. VOLKOV, Department of Mechanical Engineering, University of Alabama, Tuscaloosa, AL, USA
The deposition of thin ceramic coatings on nano-powders is known to significantly improve mechanical properties, but the mechanism of the reinforcement is not well understood. The goal of the present work is to reveal the effects of coatings on the mechanisms of plastic deformation and mechanical properties of nano-powders. For this purpose, atomistic simulations of compression and tension of close-packed and random systems of Si nanoparticles with SiC coatings are performed. The simulations show a multifold increase in the elastic and inelastic material properties at deposition of 1-5 nm thick SiC coatings. Such heterostructures, like the raw nano-powders, demonstrate predominantly ductile deformation behavior but have the Young’s modulus which is only two times smaller than the Young’s modulus of bulk SiC. To reveal the mechanism of this unexpectedly strong reinforcement, additional simulations of hollow structures composed of SiC coatings are performed. These simulations show that the reinforcement is associated with the strong interaction between Si cores and SiC coating. The mechanical load is effectively redistributed through Si cores. This results in a homogeneous distribution of stress in thin SiC coating and prevents their loss of stability and collapse.
E-1:IL10 SEM2: A Coarse-grained Particle Framework for Multiscale Cell Mechanics
S. CHATTARAJ, F. PASQUALINI, University of Pavia, Pavia, Italy
How do human tissues and organs build themselves in the body? Modeling morphogenetic processes to answer this question is hard. In fact, continuum mechanics deals poorly with non-linear, large deformations caused by individual cells migrating or proliferating. However, discrete Potts models that more naturally deal with cellular events fail to capture the underlying multiscale mechanics. Here, we present an enhanced version of the subcellular element modeling that incorporates mechanics - a framework we call SEM2 - to tackle this problem. SEM models tissue mechanics by describing cells as ensembles of particles whose interactions are governed by empirically defined potentials. In this contribution, we chose a potential that preserves single-cell rheology to demonstrate how eSEM offers multiscale mechanics in cell stretching experiments, as well as in migration and proliferation in traditional and engineered culture substrates.
E-1:IL11 Tribologically Induced Nanoscale Materials Transformations
G. MORAS, T. Reichenbach, M. Moseler, Fraunhofer IWM and MikroTribologie Centrum µTC, Freiburg, Germany; L. Pastewka, University of Freiburg, Germany
We present the results of a reactive molecular dynamics study of mechanically induced phase transitions at tribological interfaces between Si crystals. The simulations reveal that the shear strain is accommodated through the plastic deformation of an amorphous sliding interface that results from a shear-driven amorphization process. At the same time, the atomic mobility provided by the shear deformation of the amorphous Si enables a thermodynamically driven recrystallization process, which would be otherwise kinetically hindered at temperatures well below the melting temperature. The interplay between shear-driven amorphization and recrystallization results in a steady state in which the amorphous shear interface has a constant thickness. Interestingly, different shear elastic responses of the two anisotropic crystals can lead to the migration of the amorphous interface normal to the sliding plane, causing the crystal with lowest elastic energy density to grow at the expense of the other one. This “triboepitaxial” growth can be achieved by crystal misorientation or exploiting elastic finite-size effects. We propose a model validation experiment that could enable the direct deposition of homoepitaxial silicon nanofilms via mechanical scanning-probe nanolithography with a Si tip.
E-1:L12 Probing the Solute Effect on Twin Embryo Growth in Mg Alloys
YANG HU, D.M. KOCHMANN, Mechanics & Materials Lab, ETH Zurich, Zurich, Switzerland
Twinning plays an essential role in the plastic deformation of Magnesium (Mg) and its alloys. Optimizing twin structures can enhance the strength and ductility of Mg and Mg alloys. At an early growth stage of a twin, after nucleation and before mature twins of micrometer size, twin embryos exist and are bound by various twin facets. The growth of twin embryos is an extremely fast process occurring at the nanoscale, which can be simulated using atomistic simulations. In this study, we use the molecular dynamics method to perform systematic studies on the effect of nine alloying elements and different solute compositions on twin embryo growth. We demonstrate that all tested solutes reduce the motion of facets on the dark side of the twin embryo. While for facets appear on the bright side, particularly TBs, adding solutes promotes their motion. Alloying elements such as Li, Y, and Nd place an opposite effect on the motion of the dark side versus the bright side, leading to different twin morphology. Regarding the evolution of the twin size, only in Mg with Y, Li, and 4 at.% of Nd, the twin embryo reaches a larger volume than the one in pure Mg. Our findings contribute new insight for developing more durable, stronger, lighter, and safer structural materials.
Session E-2 Computational mechanics of nanodevice applications
E-2:IL01 Design and Simulation of Micro- and Nano-technology Tools for Biomedical Applications
E. CIMETTA, Università degli Studi di Padova, Padova, Italy
Engineers can lead groundbreaking discoveries advancing our understanding of diseases and improving human health. While using distinct terminology and methodologies compared to life science researchers, a common ground resides in the fundamental principles of thermodynamics, physics, and mathematics, which also apply to biological phenomena. We hypothesize that micro- and nano-technology tools exploiting classical engineering principles will solve the limitations of existing classical culture models. In our bodies cells reside in a complex microenvironment, regulating their fate and function. Most of this complexity is lacking in standard laboratory models, leading to readouts poorly predicting the in vivo situation. This is particularly felt in cancer research, as tumors are extremely heterogeneous and can induce changes both at short and long distances. Our devices generate time and space-resolved gradients, support fast dynamic changes and reconstruct complex interactions between cells while performing multifactorial and parallelized experiments. We expect that our technologies, paired with strong mathematical approaches, will bridge the gap between in vitro techniques and in vivo biological phenomena, shedding light on previously unexplored scenarios.
E-2:IL02 Molecular Dynamics Investigation of Cross-linked Gold Nanoparticle Thin Film
Kai-Chih Yeh, Ya-Yun Tsai, Shu-Wei Chang, Department of Civil Engineering, National Taiwan University, Taipei, Taiwan
Nanoparticle thin films have been shown to achieve exceptional electronic, optical, and mechanical properties. They composed of noble metal cores and organic ligands, and have been widely used for applications including strain gauges, touch sensors, and vapor sensors. In this talk, I will present our recent studies on the molecular dynamics investigation of the nanoscale features and mechanical properties of alkanedithiol cross-linked gold nanoparticle (GNP) thin films. We proposed a computational framework for constructing self-assembled cross-linked GNP thin films with a variety of ligand chain length, nanoparticle core size, and ligand grafting ratio to understand the molecular mechanism underlying the tunable mechanical properties of cross-linked GNP thin films. Our results showed that the elastic modulus decreases by increasing the chain length of the ligands. We revealed that the number of bridge linkers in all-tans conformation dominates the values of the Young’s modulus of cross-linked GNP think films. The molecular insights gained in this study could enable the design of tailored properties of nanomaterials for a variety of engineering applications.
Session E-3 Computational mechanics at mesoscopic / macroscopic scale
E-3:IL01 A Mapping between the Non-linear Micromechanics of Glasses and Elasto-plastic Models
D. RICHARD, Laboratoire Navier, Champs-sur-Marne, France
In contrast to crystalline solids, characterizing glassy amorphous structures remains a challenging task for both condensed matter physicists and material scientists. In recent years, many new computational tools have been developed to this aim, from purely structural descriptions of local motifs, through machine learning based tools, to various analyses of the potential energy landscape. In the context of shear driven amorphous solids, we will discuss structural indicators for probing glassy heterogeneities and present a novel microscopic method that allows one to estimate local stress activation thresholds from bare (as-cast) static configurations. We will show how obtained yield rules can serve to parametrize elasto-plastic models and predict the stress-strain response of materials. We will illustrate such a mapping between atomistic simulations and mesoscale models in the context of strain localization for glasses featuring a wide range of mechanical stabilities.
E-3:IL02 Extreme Events on Structures. The Key Role of Multiphysics Simulation
A. LARESE1, 2, L. Moreno4, V. Singer3, N. Crescenzio1, R. Wuechner4, 1Dept. of Mathematics, Università di Padova, Padova, Italy; 2Institute for Advanced Studies of the Technical University of Munich TUM-IAS, Germany; 3Chair for Structural Analysis, Technical University of Munich, Germany; 4Institute of Structural Analysis, Technical University of Braunschweig, Germany
In recent years, the frequency and intensity of large gravitational mass movements, including landslides, debris flows, and mudflows, have risen due to climate change and related factors. These phenomena transport vast amounts of rocks and granular materials that can damage structures and landscapes, creating significant risks that can lead to loss of life and property damage. The classical Finite Element Method (FEM) has limitations in dealing with materials undergoing large deformations as is required in this case. In recent decades, alternative solutions have been proposed to overcome this limitation, one of which involves the use of particle-based methods. Among these, the Material Point Method (MPM) blends the advantages of both mesh-based and mesh-free methods. MPM avoids the problems of mesh tangling while preserving the accuracy of Lagrangian FEM. It is well suited for nonlinear solid mechanics and fluid dynamics problems. The presentation will highlight recent MPM advancements, including granular flow and large-deformation material simulation in both compressible and incompressible regimes. Additionally, partitioned coupling strategies with other techniques like FEM or DEM will be discussed for complex multiphysics simulations.
E-3:L03 Multifractal Mechanics and Thermal Transport of Solids: Theory, Experiments, and Uncertainty Analysis Across Scales
W.S. OATES, B. Pahari, M. Carvajal, Florida State University, Department of Mechanical Engineering, Tallahassee, FL, USA
We formulate and experimentally validate complex thermomechanical structure-property relations using fractal geometry and fractal/fractional order calculus to accommodate a broad range of time and length scales. Fractal geometry is well known in mathematics where responses repeat over all scales. This behavior accurately approximates observations from nature (e.g., turbulence, mountain ranges, ice crystals, arterial structure, etc.). This has important implications on advancing predictions of constitutive properties in the nonlinear mechanics of solids. We offer a modeling framework that connects the scale-free properties of fractals with thermomechanical properties in materials as a means to accommodate complexity. This framework uses power-law characteristics of (multi)fractals (i.e., random fractals) as a means to relate complex structure with nonlinear constitutive relations and thermomechanical balance equations. An information theoretic framework is implemented using entropy dynamics to derive both diffusion and wave propagation governing equations for fractal media. The results are compared to experiments and its uncertainty is quantified using Bayesian statistics.
E-3:IL04 Unveiling Microstructure Effects on Fracture: Atomistic Simulations, Mesoscale Models and Micro-mechanical Tests
E. BITZEK, Max-Planck-Institut für Eisenforschung, Düsseldorf, Germany
In any engineering alloy, a propagating crack will encounter a multitude of microstructural features like point defects, dislocations, precipitates, grain, and phase boundaries. All of these interactions will influence the crack and contribute to energy dissipation. Understanding the elementary interaction mechanisms between cracks and these defects is therefore of fundamental importance for the development of microstructure-sensitive micromechanical failure models. Here, we show examples in bcc metals of how atomistic simulations of crack–microstructure interactions can be used to explain micromechanical fracture tests and to inform mesoscale models. These include the study of how atomic-scale defects affect crack propagation, how grain boundary structure and segregation can impact not only intergranular but also transgranular fracture, and how voids and preexisting dislocations can trigger crack tip plasticity and lead to crack arrest. Such processes are being included in a newly developed code that couples discrete dislocations dynamics (DDD) with the extended finite element method (XFEM) for fracture simulations.
E-3:L06 Modeling the Deformation and Ductile Damage of Irradiated EUROFER97
R. RAJAKRISHNAN, E. Gaganidze, J. Aktaa, Karlsruhe Institute of Technology (KIT), Institute for Applied Materials, Eggenstein-Leopoldshafen, Germany
Due to its superior thermo-mechanical properties and swelling resistance, Reduced Activation Ferritic Martensitic (RAFM) steel EUROFER97 was selected as the reference structural material for future fusion reactors. However, irradiation reduces its strain-hardening and uniform elongation (UE) capabilities, which could make it unusable for deformations beyond UE under current design rules. The irradiated material's high strength and low ratio of uniform elongation to fracture strain, however, warrant further investigation into its applicability beyond UE. To establish the safest possible operating conditions for reactor components, a macroscopic scale continuum material model describing the tensile properties of irradiated material is required. Considering this, a thermodynamic framework for irradiated materials is proposed and used to build a coupled deformation and ductile damage model within the finite viscoplasticity framework, based on the models by Aktaa-Petersen (2011) and Tvergaard-Needleman (1984). Camera-monitored tensile tests at room temperature and 300°C are used to calibrate and validate the model. Model predictions are used to discuss the implications of irradiation's influence on current ductile failure design rules and the scope for relaxation of failure criteria.
E-3:IL07 Nonlocal Fracture in Elastomers: Experiments and Continuum Modeling
HANSOHL CHO, Jaehee Lee, Jeongun Lee, Korea Advanced Institute of Science and Technology, Yuseong Gu, Daejeon, South Korea
Predictive modeling of the failure in elastomers remains a significant challenge due to the complexity of the underlying physical mechanisms of the irreversible damage process across a wide range of length scales. Furthermore, the nonlocal nature renders an accurate modeling of the failure in elastomers more challenging. The main objective of this work is two-fold. First, we discuss a nonlocal gradient-theory-based framework that employs (1) entropy-driven or (2) internal energy-driven damage hypothesis which have been widely accepted for modeling the failure in elastomeric networks. We then discuss the size-dependent fracture behavior of an elastomer using experiments and nonlocal continuum modeling. To this end, we performed precisely controlled tension tests for notched specimens made of a photo-curable elastomer. We also made use of the nonlocal approach for modeling the damage processes in the material subjected to extreme stretch. Our numerical simulation was found to be able to capture salient features of the size-dependent fracture in the material revealed in experiments. Altogether, using both experiments and gradient-damage theory-based numerical simulations, our study provides some useful insight into the nonlocal nature of damage and fracture in elastomeric materials.
E-3:IL08 Multiscale Modelling of Ceramic Matrix Composites
E. BARANGER, Université Paris-Saclay, CentraleSupélec, ENS Paris-Saclay, CNRS, LMPS - Laboratoire de Mécanique Paris-Saclay, France
Over the past decades, non-linear models describing the damage and fracture of architectured materials have progressed, considering varied degradation scenarios under complex multi-axial loadings. Many of these models remain expansive to handle and difficult to understand numerically. A challenging task for the researcher is extracting information from that data quantity. The extraction of this information needs a language support. A classical framework of physics generally relies on state variables and the associated energetic potentials. This paper will present the automatic definition of state variables. Choosing a norm allows us to compare an approximate and a full representation of the state, which will first be discussed on a simple example. An anisotropic damage model is chosen and simplified automatically. The interest of different norms is first discussed including a classical L2 norm leading to the classical principal component analysis. Then, a complex microstructure of a ceramic matrix composite (CMC) at the scale of the fibre is described using a set of elementary patterns. They relate to the local morphology, elastic interactions, and first cracking patterns. The associated numerical strategy relies on the generalised finite element method (GFEM).
E-3:IL09 Modelling the Data of Nonlinear Mechanical Properties of Fabrics by Decomposing Friction
G. STYLIOS, L. LUO, Heriot Watt University, Scotland
Friction between fibers is a physical phenomenon inherent in yarns and fabrics. The coexistence of fibre friction and fiber elasticity enables the coherency (holding themselves together) of yarns and fabrics and defines their behaviour when they are deformed, their ability to take large curvatures and to wrap easily around 3D bodies. By separating the frictional and elastic components from the hysteresis of the fabric's tensile properties, we introduce a fabric tensile frictional stiffness parameter which is more important than using the hysteresis value only at a specific strain point of a specific load/unload curve loop. This new property is of great significance to the exploration of fabric mechanics, for 3D realistic fabric/clothing simulation, and for engineering desirable fabric performance. The first-order coefficients of the nonlinear pure elasticity curve also have a clearer physical meaning for the study of the initial Young's modulus of fabric stretch, which is currently not measured by any testing equipment. Its higher order coefficients also make it easier to quantify the non-linear properties of flexible fabrics. Since hysteresis is mainly formed by the friction between fibers, the frictional stiffness and the Coulomb friction curves can be purposely changed by the selection of the fiber material type, the spinning process, and the structure for the desired fabric, offering for the first-time the ability to truly engineer fabric behaviour, and hence classifying textile fabrics as engineering materials. In consequence, fabrics can be engineered as any other engineering material.
E-3:L10 Novel Tool to Perform Thermomechanical Characterisation on Refractory Microstructure Design using Discrete Element Method (DEM)
H. RANGANATHAN1, 2, D. ANDRE2, M. HUGER2, R. SOTH1, C. WÖHRMEYER1, 1Imerys Technology Center, Vaulx-Milieu, France; 2University of Limoges, IRCER, UMR CNRS 7315, Limoges, France
Refractory castable, despite appearing to be a homogeneous continuum medium at macroscopic scale, exhibits significant heterogeneities at microscopic scale due to the presence of various constituents. Understanding the influence of such heterogeneous microstructures, taking into account the intrinsic properties of each constituent, on the macroscopic thermomechanical behavior at high temperature thermal is quite complex. In such purpose, advance numerical techniques like Discrete Element Method (DEM) can provide very interesting new insight for a better visualization of each local effect between large aggregates and the fine part of the material that significantly affect it thermomechanical performance. Granoo, a numerical DEM platform, utilizes Voronoi tessellation models to enhance understanding of the mechanisms involving large diffused damage within microstructures designed to resist thermal shocks. This work aims to develop a virtual lab to demonstrate the potential of a lattice spring model with DEM approach for understanding and improving thermomechanical behavior of a arbitrary/hypothetical refractory microstructure. For purpose of demonstration of the robustness of this novel DEM tool, different cases of model materials will be investigated, considering aggregates as single crystal with isotropic or anisotropic properties.
Session E-4 Computational mechanics in simulated operating conditions
E-4:IL02 Novel Approaches to Computational Additive Manufacturing
D. SOLDNER, J. MERGHEIM, Institute of Applied Mechanics FAU Erlangen-Nürnberg, Erlangen, Germany
Powder bed-based additive manufacturing (AM) allows the realization of complex part geometries due to the combination of a layer wise built and local melting of powder material, whereby the part quality is drastically influenced by the thermal conditions. Selective laser sintering (SLS) of polymers represents a widely used AM technique. In contrast to metal AM, where the solidification of molten material appears rapidly after melting, crystallization of the molten polymer material appears slowly during the manufacturing process. It has been shown, that repeated crystallization followed by re-melting during the process may occur, which necessitates adapted crystallization models. Further, a dependency of the evolution of the crystallization front on the geometry of the specimen has been observed. Crystallization is accompanied by material shrinkage, which can lead to warpage, distortion and built up residual stresses. The present contribution extends the model presented in [1] by a visco-elastic visco- plastic material model at finite strains, to compute not only the temperature and degree of crystallization, but also deformations and residual stresses during the additive manufacturing of polymeric parts.
[1] Soldner et al. DOI: 10.1016/j.addma.2020.101676
E-4:IL03 Computational Mechanobiology Towards Applications in Tissue Engineering
J.H. HENDERSON, Syracuse University, Syracuse, NY, USA
Advances in programmable biomaterials are enabling new investigations and understanding in mechanobiology—the study of how physical forces at the cell and tissue level contribute to development, maintenance, wound healing, and disease. A major focus of our work in this area is the development and application of cytocompatible shape-memory polymers (SMPs) for the study and control of cell mechanobiology in basic science and tissue engineering applications. Current work includes emphasis on the development of advanced manufacturing approaches through which fabrication with SMPs has the potential to enable the paradigm-shifting creation of devices possessing spatially varying material functionality that cannot be achieved by any current approach. Here we will present integrated, interdisciplinary experimentation and simulation to address gaps in fundamental understanding at the interface of materials processing, materials science, and mechanobiology.
Session E-5 Advances in theory and computational methods
E-5:IL01 Recent Advances in Tribological Modelling and Simulations Across the Scales
D. DINI, Department of Mechanical Engineering, Imperial College London, South Kensington Campus, London, UK
Recent theoretical progress is making the accurate modelling of tribological interfaces a reality; the resulting methodologies provide an ideal foundation for the development of a framework in which a wide range of complex features can be captured successfully across the scales. In this talk, I will discuss methods and tools recently developed to address the challenges of predicting the behaviour of critical interfaces. Simulations are used to explore the mechanisms that control friction, tribofilm formation and surface wetting in different applications and environments, from aerospace to triboelectric devices [1-6]. The development of design strategies, which incorporate methods that enable accurate in-silico tribological experiments to be performed, will impact many sectors, including energy, aerospace, biomedical technologies, consumer goods and food industries.
[1] C. Ayestarán Latorre et al. (2021), Communications Chemistry 4, 1-11.
[2] S. Hu et al. (2021), ACS Applied Materials & Interfaces 13, 31310-19.
[3] E. Weiand et al. (2022), Soft Matter 18, 1779-1792.
[4] S.J. Eder et al. (2022), Applied Materials Today, 29, 101588.
[5] X. Zhang et al. (2022), Advanced Science 9, 2270154.
[6] S. Ntioudis et al. (2023), Computational Materials Science, 229, 112421.
E-5:IL02 Novel Approaches to Computational Continuum-atomistic Coupling for Polymers
S. PFALLER, L. Laubert, M. Ries, F. Weber, W. Zhao, Institute of Applied Mechanics, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany
Integrating simulation techniques addressing the atomistic or molecular structure of materials into macroscopic treatments is still a challenge in computational mechanics. Particularly in the field of polymers, where the chain-like structure of molecules can cause entanglements and crosslinks, specific scale-bridging techniques are required. Typically, strategies established for crystalline materials are not sufficient in this regard. The present contribution first briefly subsumes the theoretical foundations of multiscale domain-decomposition techniques, which, in general, use a fine-scale consideration only in distinguished regions of a sample. Such specific sub-domains may be necessary in the vicinity of crack tips or in the neighbourhood of filler particles of polymer-(nano)composites. Secondly, it introduces the Capriccio method which has been designed for amorphous thermoplastics and is currently applied to further amorphous materials like thermosetting polymers and silica glasses. It combines an atomistic treatment by molecular dynamics with a continuum solved by the finite element method and is capable to capture the highly inelastic effects arising in polymers. Finally, recent studies and findings are presented and discussed.
E-5:L03 Concurrent Multiscale Modeling of Boundary Lubrication, Enabled by Machine Learning
H. HOLEY1, 2, 3, P. Gumbsch1, 3, L. Pastewka2, 1Karlsruhe Institute of Technology, Karlsruhe, Germany; 2University of Freiburg, Freiburg, Germany; 3Fraunhofer IWM Freiburg, Freiburg, Germany
Friction and lubrication are inherent multiscale problems, particularly when the gap between contacting bodies is on the order of molecular interaction length scales. These conditions occur predominantly in the so-called boundary lubrication regime under high normal loads, low sliding speeds or with low viscosity lubricants. Modeling lubrication across scales beyond purely sequential approaches has so far remained elusive. In this talk, I will present a reformulation of the classical lubrication equations that allows straightforward coupling between continuum and molecular models. Instead of using fixed-form constitutive expressions that are parametrized a priori using atomistic methods, we build a surrogate model for the interfacial shear stress and normal pressure on-the-fly. Concurrent coupling is achieved by an active learning scheme based on Gaussian process regression, which allows for a data-efficient interpolation of the microscopic stresses obtained from molecular dynamics in high-dimensional parameter spaces. Furthermore, the Gaussian process posterior variance provides a transparent path to uncertainty quantification in lubrication modeling. We validate the proposed method for simple fluids and highlight its application potential for more realistic systems.
E-5:L04 Eigenstrain Representation of Defects, Dislocations, and Dislocation Networks
S.L. DUDAREV, P.-W. Ma, A.R. Warwick, M. Boleininger, L. Reali, UKAEA, Culham Science Centre, Oxfordshire, UK
The total (lattice) and elastic strains are the main variables entering the finite element model equations used for simulating deformations in reactor components exposed to irradiation. These variables differ fundamentally in that the total strain can be directly visualised and observed in diffraction experiments whereas the elastic strain, while not being an observable quantity, is a fundamental quantity of elasticity theory where it is related to elastic stress, a notion that enters a variety of structural integrity criteria. The source terms for the elasticity equations containing defects is given by the density of relaxation volumes of defects, which can now be computed using density functional theory, atomistic simulations, or mesoscopic formulae. The theoretical formulation of elasticity equations with sources enables applying it to the predictive treatment of reactor components on the engineering scale, with cases explored so far including ion irradiated foils, individual internal components of tokamak devices, and the entire structure of a tokamak device.
E-5:L05 Some Recent Advances and Applications in Isogeometric Analysis
A. REALI, Department of Civil Engineering and Architecture, University of Pavia, Pavia, Italy
Isogeometric Analysis (IGA) is a successful simulation framework originally proposed by T.J.R. Hughes et al., in 2005, with the aim of bridging Computational Mechanics and Computer Aided Design. In addition to this, thanks to the high-regularity properties of its basis functions, IGA has shown a better accuracy per degree-of-freedom and an enhanced robustness with respect to standard finite elements in many applications - ranging from solids and structures to fluids, as well as to different kinds of coupled problems - opening also the door for the approximation in primal form of higher-order partial differential equations. After a concise introduction of the basic isogeometric concepts, this lecture aims at presenting an overview of some IGA recent advances with a focus on interesting applications in structural and coupled problems from different fields like Mechanical, Civil, and, in particular, Biomedical Engineering, including also fluid-structure interaction in aortic valves, electro-mechanical cardiac muscle simulations, electro-mechano-fluid coupling in jellyfish motion, tumor growth.
E-5:IL06 Adaptive (Iso-)geometric Modeling for CAD/CAE Applications
C. GIANNELLI, University of Florence, Florence, Italy
In the field of numerical simulation, the framework of isogeometric analysis (IGA), which suitably exploits CAD standard B-splines and their adaptive extensions, opens new perspectives, allowing the incorporation of automatic optimization methods into the product development process. In view of the rigidity of tensor-product B-splines, however, localized mesh refinement can only be obtained by identifying a (non-standard) spline basis on different unstructured grid configurations. In this setting, the mathematical foundations of adaptive isogeometric analysis were recently established and together with related applications, they are now a rather large research area in computational mechanics and numerical analysis. Adaptive IGA methods are of interest for the simulation of different kinds of problems which necessarily require very fine meshes. The talk will discuss recent advances in the design and analysis of adaptive spline refinement and coarsening schemes by focusing on efficient processing and approximation for CAD model (re-)construction and (iso-)geometric modeling. Key applications to the phase-field modeling of evolving interface problems and to computational mechanics problems inspired by additive manufacturing will be presented.
E-5:IL07 Mechanics of Bioinspired, Bionic, Nano and Meta Materials
N. PUGNO, University of Trento, Italy
The Italian artist, inventor and scientist Leonardo da Vinci (1452-1519) can probably be considered the father of bio-inspired design, as illustrated for example by his artificial wings and flying machines, based on bird observation and dissection. Five centuries from his death, bioinspiration is attracting widespread attention worldwide, both in academia and industry. In this Invited Lecture I will review our (https://pugno.dicam.unitn.it/) recent research activity in the field of mechanics of bioinspired, bionic, nano and meta materials. The inspiration for this work derives not only from Nature and from da Vinci, but also from fundamental papers published by estimated colleagues of our community, all of whom I cannot mention here for brevity, starting from A. A. Griffith, who in 1921 published his seminal paper on Fracture Mechanics that is thus representing the fil rouge of this presentation.
E-5:IL08 The Forming, Function and Optimization of Bio-inspired Composites by Multiphysics Simulations and Generative Model
ZHAO QIN, Syracuse University, Syracuse, NY, USA
Generative artificial intelligence (AI) has recently demonstrated to play a significant role in text, sound, and image generation, leveraging its strengths to enhance and accelerate the development of innovative designs. Here, I will focus on the recent study of different biomaterials (e.g., bamboo, bone and mycelium) and discuss how generative AI can detect the hidden structure-mechanics correlations and be applied to material design. For example, we manage to rapidly generate bone-inspired bicontinuous composite structures together with the stress distribution in loading. We find that generative AI, enabled through fine-tuned Low Rank Adaptation models, can be trained with a few inputs to generate both synthetic composite structures and the corresponding von Mises stress distribution. The results show that this technique is convenient in data augmentation with useful mechanical information that dictate stiffness, fracture, and robustness of the material with one AI model, and such must be done by several different experimental or simulation tests before. Our research offers valuable insights for the improvement of composite design with the goal of expanding the design space and automatic design screening for optimized mechanical functions.