Fig 2 - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
Stress-strain curves for the pure MD computations, a pure FE simulation, and for various coupled cases; The strain rate is ˙ ε = 1% ns −1 for all simulations and the stress of the coupled systems is calculated by averaging over the nodes at boundary.
Source publication
In this contribution we present an extension of the multiscale Capriccio method towards inelasticity. This enables coupled simulations of a particle domain embedded into a continuum with particular focus on polymer systems. Starting from the method's initial implementation of pure elasticity, we substitute the nonlinear elastic continuum constituti...
Contexts in source publication
Context 1
... this context, a viscoelastic-viscoplastic continuum model has been developed recently that is able to reproduce the behaviour of the particle system also for larger strain ranges. In Figure 2, the stress-strain curve for the pure MD system is depicted for a strain rate of ˙ ε = 1% ns −1 (black solid line) up to 8% strain. The associated curve for the FE implementation of the viscoelastic-viscoplastic continuum model shows almost a perfect match between both description at the same strain rate (black dashed line). ...
Context 2
... apply a specific strain rate, the FE load step size and the associated time step length have to be chosen such that ˙ ε = ∆ε ∆t FE , which, in turn, restricts the relative choice of n iter , n MD , and ∆t MD . The results for n iter = 800, n MD = 2000, and ∆t MD = 5fs are given in in Figure 2: the stress strain curves for α = 0.5 (blue dashed line) and for an optimized choice of α as a linear function (red dashed line) show only very small deviations from the pure MD reference solution. In comparison, the original, purely elastic Capriccio method leads to large deviations for strains exceeding approx. ...
Similar publications
A simple and efficient method is proposed for the analysis of twist of rectangular box-girder bridges, which undergo distortion of the cross section. The model is developed in the framework of the Generalized Beam Theory and oriented towards semi-analytical solutions. Accordingly, only two modes are accounted for: (i) the torsional mode, in which t...
A six-dimensional force sensor with a three-beam structure was designed, and its strain and stress distribution under various load conditions were analyzed based on finite element simulation technology. According to the strain distribution, the Wheatstone bridge scheme of the sensor is designed, and the output voltage and sensitivity formulas in ea...
In order to investigate the force characteristic of different steel deck pavement under extreme temperature and load coupling situation, we established the three dimensional extreme high temperature and extreme low temperature field of simple supported beam bridge, the paving structure were double layered SMA, double layered epoxy concrete, SMA+GA,...
Citations
... To a certain extent, the amount of the material can be increased by CG techniques, which cluster groups of atoms and thus decrease the numerical effort, but an engineering scale considering millimeters and beyond is still not within reach. The simulation size can be further enlarged by partitioned-domain multiscale simulations, e.g., the Capriccio method developed in our group (Pfaller et al., 2013;Pfaller et al., 2019;Zhao and Pfaller, 2021), where only the area of interest is treated in MD resolution while the other areas are calculated using continuum model. Such multiscale models, however, require a sophisticated constitutive law for the continuum within a reasonable validity range, which we aim at in the present manuscript. ...
This contribution presents a phenomenological viscoelastic-viscoplastic constitutive model informed by coarse-grained (CG) molecular dynamics (MD) simulations of pure glassy polystyrene (PS). In contrast to experiments, where viscoplasticity is caused by various effects simultaneously, these effects can be decomposed in MD simulations by adjusting the MD system. In the MD simulations considered here, neither bond breakage nor cross-links are introduced; instead, we focus on the intermolecular interaction of polymer chains. We employ a thermo-dynamically consistent generalized Maxwell framework parallelly comprising an elastic, a viscoelastic, and several elasto-viscoplastic modules with different yield stress to capture the viscoelastic and the viscoplastic mechanical behavior simultaneously. The yield stresses decrease with the maximum deformation the MD system has experienced. The constitutive model presented here is based on 10 material parameters, which can be identified by a few data sets and fits well the CG MD simulations of PS under uniaxial and biaxial deformation with time-proportional and cyclic loading conditions in a wide range of strain rates (0.1%/ns-20%/ns).
(Free access link until July 16, 2021:
https://authors.elsevier.com/a/1d8MX_UECrkz1)
This contribution presents a partitioned‐domain particle‐continuum coupling method for amorphous polymers with multiple particle‐based domains. The coupling method treats the particle‐based domains with molecular dynamics (MD) simulations and the continuum domain discretized by the Finite Element (FE) method. In the continuum domain, a viscoelastic‐viscoplastic (VE‐VP) constitutive model derived from MD simulation results of the polymer at molecular resolution is employed. The effects of the minimum distances between the domains, the distribution and the number of the MD domains as well as the strain rates are studied under uniaxial tension. This method is a precursor for multiscale simulations of polymer‐based nanocomposites (PNC).
In this contribution, we present a partitioned‐domain method coupling a particle domain and a continuum domain for multiscale simulations of inelastic amorphous polymers under isothermal conditions. In the continuum domain, a viscoelastic‐viscoplastic constitutive model calibrated from previous molecular dynamics (MD) simulations is employed to capture the inelastic properties of the polymer. Due to the material's rate‐dependence, a temporal coupling scheme is introduced. The influence of the time‐related parameters on the computational cost and accuracy is discussed. With appropriate parameters, multiscale simulations of glassy polystyrene under various loading conditions are implemented to showcase the method's capabilities to capture the mechanical behavior of polymers with different strain rates and with non‐affine deformations of the MD domain.
