Multiscale modeling in nanomaterials science
Nanoscience is an area with increasing interest both in the physicochemical phenomena involved and the potential applications such as silicon carbide films, carbon nanotubes, quantum dots, MEMS etc. These materials exhibit very interesting properties (electronic, optical, mechanical) at various length/time scales necessitating better insight. Modem fabrication techniques, such as CVD, also require better understanding in a wide range of length/time scales, in order to achieve better process control. Multiscale modeling is a new, fast developing and challenging scientific field with contributions from many scientific disciplines in an effort to assure materials simulation across length/time scales. In this paper we present a brief review of recent advances in multiscale materials modeling. First, a classification of existing simulation methods based on time and length scales is presented along with basic principles of the multiscale approach. More specifically, we focus on electronic structure calculations, classical atomistic simulation with molecular dynamics or monte carlo methods at the nano/micro scale, Kinetic Monte Carlo for larger system/time scales and finite elements for very large scales. Then, we present the hierarchical and the hybrid strategies of multiscale modeling to couple these methods. Finally, we deal with selected applications concerning thin film CVD deposition and mechanical behavior of carbon nanotubes and we conclude presenting an overview of future trends of multiscale modeling. (c) 2006 Elsevier B.V. All rights reserved.