Figure - available from: Computational Mechanics
This content is subject to copyright. Terms and conditions apply.
![Damage propagation caused by steel projectile impact on the SiO2 surface. Left: photos taken from experiment [37], right: snapshots from numerical simulation](publication/341258976/figure/fig29/AS:960234239582210@1605949091279/Damage-propagation-caused-by-steel-projectile-impact-on-the-SiO2-surface-Left-photos.png)
Damage propagation caused by steel projectile impact on the SiO2 surface. Left: photos taken from experiment [37], right: snapshots from numerical simulation
Source publication
![Transition from fourfold-coordinated Si to sixfold-coordinated Si [24]](publication/341258976/figure/fig1/AS:960234206023680@1605949083744/Transition-from-fourfold-coordinated-Si-to-sixfold-coordinated-Si-24_Q320.jpg)
An atomistically-informed constitutive model based on multiplicative hyper-elasto-plasticity with damage is developed for the study of high-pressure induced densification of silica glass. At the atomistic level, a molecular dynamics (MD) representative volume element of amorphous silica (a-SiO2) is first constructed using the melt-quench approach....
Citations
... Computational methods aimed at resolving this remarkable SLG behavior range from linking atomistic and continuum scales [20,21,22,23,24] to coupled thermodynamically consistent thermo-mechanical glass model [25]. While the analysis of the mechanical behavior of soda-lime glass under hypervelocity impact has received considerable attention, the literature on the associated electromagnetic phenomena is rather limited. ...
... On the one hand, the shape of the cavity is modelled via a blast wave model, independently derived by Geoffrey I. Taylor [35,36], John von Neumann [37] and Leonid Sedov [38] during World War II. The self-similar solution predicted by such a model, together with the thermodynamic properties and a metal-like behavior observed in numerous studies [14,15,16,17,18,19,20,21,22,23,24,25] of soda-lime glass, thus allows to estimate the amount of vaporized target material and the state variables within the plasma during the impact. This model neglects the constitutive forces in a solid, but the energy density is so high at early stages of hypervelocity impact that material behaves largely in a fluid manner. ...
... The emitted radiation upon hypervelocity impact is therefore estimated here by informing Equation (30) with the total charge computed in Section 2.2 and by assuming n = 10 3 , while the speed ratio β is considered a free parameter. For the impact at υ p = 10 km/s, the energy 22 This is the author's peer reviewed, accepted manuscript. However, the online version of record will be different from this version once it has been copyedited and typeset. ...
A semi-analytic method is proposed to compute the produced plasma and the emitted Cherenkov radiation from hypervelocity impacts on soda-lime glass for various projectiles and impact velocities. First, the Taylor-von Neumann-Sedov blast wave model, coupled with the system of nonlinear Saha equations for multispecies, strongly coupled plasma, is adopted to estimate the hydro-dynamic profiles and the ionization state of the target material in the early stage of the impact. Second, the Frank-Tamm formula is considered to investigate the onset of the Cherenkov radiation and to compute the emitted energy. The present approach predicts a linear dependence of the produced total electric charge on the projectile density and a quadratic dependence on the projectile velocity, whereas the emitted Cherenkov radiation scales quadratically with the produced charge, if the onset conditions are met.
... Moreover, machine-learning algorithms also can be employed to enhance the datasets, such as the artificial neural network (ANN), symbolic regression (Kronberger et al., 2018), manifold learning (Lopez et al., 2018), and support vector machine (Yuan et al., 2017). In the work of Xu et al. (2020), the ANN has been successfully employed to tune up certain continuum history-dependent model parameters based on the database generated by molecular dynamics simulations, then a novel continuum-based multiplicative hyper-elastoplasticity-damage model that accounts for the anomalous densification behavior was developed (Xu et al., 2020). Finally, it is noted that iterative searches are the main part of the solution process. ...
... Moreover, machine-learning algorithms also can be employed to enhance the datasets, such as the artificial neural network (ANN), symbolic regression (Kronberger et al., 2018), manifold learning (Lopez et al., 2018), and support vector machine (Yuan et al., 2017). In the work of Xu et al. (2020), the ANN has been successfully employed to tune up certain continuum history-dependent model parameters based on the database generated by molecular dynamics simulations, then a novel continuum-based multiplicative hyper-elastoplasticity-damage model that accounts for the anomalous densification behavior was developed (Xu et al., 2020). Finally, it is noted that iterative searches are the main part of the solution process. ...
The present chapter overviews recent research trends of deep learning related to computational mechanics. In Sect. 3.1, we see the growing interest in deep learning in recent years based on the trend of the number of published papers on this topic, discussing how deep learning is applied to various fields in computational mechanics. In Sect. 3.2, we review the research trends from the list of papers on computational mechanics with deep learning published since 2018.
Fracture and damage ascribed to the intrinsic brittleness of amorphous oxide glasses are crucial problems for the daily use of glass products. Because the latest developments in glass and glass-ceramics technologies have further broadened their applications, the safety issues become increasingly important. Computational modeling and simulation are now indispensable in the design and analysis of glass quality and safety. This review, therefore, provides an overview of the state-of-the-art fracture modeling/simulation techniques ranging from atomistic scale to continuum scale. In addition to the fundamental theories, typical and recent studies using a variety of continuum methods are introduced. This review also covers the application examples of classical molecular dynamics (CMD) simulations and reactive CMD simulations to investigate the fracture and damage evolutions in glass and glass-ceramics. Advanced multiscale modeling techniques that bridge atomistic and continuum method are also introduced for modeling amorphous materials.
The mechanism-based theoretical models are presented to provide the theoretical explanations for strengthening and pressure-induced hardening behaviors in ultrafine-grained metals under high pressure. The grain boundary deformation model is extended to construct the relationship among the critical stress for grain boundary deformation, pressure and the grain size. The pressure-dependent critical twinning stress is derived on the basis of dislocation theory to describe the pressure-induced hardening behavior. The classic Hall-Petch for the grain size-dependent yield strength is modified through involving the contribution of partial dislocations. The simulation results demonstrate that the proposed models provide good descriptions of the strengthening and hardening behaviors in ultrafine-grained metals with respect to the high pressure. The critical grain size for the transition of deformation mechanisms is sensitive to the pressure and the grain boundary thickness. The predicted grain size-dependent yield strength and flow stress under high pressure are agreeable well with the experimental tests. These findings could shed some lights into understanding the plastic deformations of ultrafine-grained metals under high pressure.
The concept of multiscale modelling has emerged over the last few decades to describe procedures that seek to simulate continuum-scale behaviour using information gleaned from computational models of finer scales in the system, rather than resorting to empirical constitutive models. A large number of such methods have been developed, taking a range of approaches to bridging across multiple length and time scales. Here we introduce some of the key concepts of multiscale modelling and present a sampling of methods from across several categories of models, including techniques developed in recent years that integrate new fields such as machine learning and material design.
Silica-based glass may possess paradoxically high resistance to hypervelocity impact due to the experimentally observed phase change emanating from high pressure characteristic to hypervelocity impact combined with irreversible densification of the material, which leads to a highly efficient kinetic energy-absorption mechanism. In order to capture this extraordinary behavior of silica glass in hypervelocity impact, a coupled thermo-mechanical model is developed in the framework of thermodynamics with internal state variables. In addition to pressure induced densification (phase change), the proposed model is aimed at capturing the effects of dramatic increase in temperature, strain rate sensitivity and fragmentation/comminution of the material. The proposed model is based on the multiplicative decomposition of the deformation gradient into thermoelastic and plastic parts. The irreversible densification of silica glass is characterized by the plastic volumetric strain which is a basic internal state variable associated with a molecular structure rearrangement due to phase change. Evolution of the plastic deformation is described by a critical state plasticity model combined with damage evolution. In the absence of damage, the elastic domain is fully informed by molecular dynamics simulations of perfectly intact silica glass. With evolving damage, the atomistically informed elastic domain shrinks smoothly to another critical state plasticity elastic domain which serves as a granular description of the fragmented/comminuted state of the material. Thermo-mechanical coupling is considered where temperature rises as a result of mechanical dissipation while mechanical behavior depends on temperature through thermal softening. In addition, the model is capable of capturing both the material’s ductile behavior (featuring significant densification due to high pressure) in the vicinity of projectile–target contact interface and its characteristically brittle behavior exhibited elsewhere. This is achieved by introducing a brittle damage initial criterion based on the thermodynamic driving force for damage that is analogous to the energy release rate-based criterion for crack growth. Constitutive functions and material parameters in the model are determined from the molecular dynamics simulations. The proposed model has been implemented in the explicit coupled thermo-mechanical finite element code and validated against hypervelocity impact experiments.
