Adaptive generalized mathematical homogenization framework for nanostructured materials

Abstract

SUMMARYWe present an adaptive generalized mathematical homogenization (AGMH) framework for modeling nanostructured materials with evolving defects at a finite temperature. By this approach molecular dynamics model is employed in the vicinity of defects whereas constitutive equation-free continuum model is used away from the defects. The proposed framework consists of the following salient features: (i) a constitutive law-free cohesive element whose behavior is modeled by the atomistic representative volume element (ARVE) and (ii) a dislocation detection band embedded in the ARVE aimed at detecting dislocations passing from or into the interior of the nanograins. These features of the model are critical to studying deformation of nanocrystals where only a small portion of the problem domain requires molecular resolution.

Coupling the finite element method and molecular dynamics in the framework of the heterogeneous multiscale method for quasi-static isothermal problems

“…However, the modelling of the representative volumes gives a molecular dynamics-like problem and is not integrated for the full atomistic motion. A framework that combines the serial and concurrent coupling in the generalised mathematical homogenisation method is given by Fish et al (2010). Chockalingam and Wellford (2011) give a homogenisation procedure with an emphasis on thermal problems.…”

Computational continua for linear elastic heterogeneous solids on unstructured finite element meshes

“…As the coarse-scale strain gradients become sufficiently large, the hypothesis of the constant deformation gradient fails to account for the variation of the coarse-scale solution gradients over the UC domain. In higher-order theories, [14][15][16][17] the UC is subjected to nonconstant coarse-scale deformation gradients. For instance, in second-order theories, the coarse-scale deformation gradient is assumed to vary linearly over the UC domain.…”

An automatic method for generating carbon nanostructure atomistic models using hexagonal meshes with properly distributed defects

“…Therefore, molecular dynamics (MD) simulation is an indispensable approach for studying carbon nanostructures (CNS). Because modeling is the primary step in the MD simulation , it has been a recurrent topic in the last 20 years. Existing modeling methods and software packages can automatically generate simple atomistic models, but as the research on nanomaterials continues to develop and its applications continue to extend, increasingly complicated CNS are designed theoretically or discovered in the laboratory.…”