Atomistic simulation of rare events using gentlest ascent dynamics

Amit Samanta

Princeton Univeristy, Program in Applied and Computational Mathematics, Princeton, USA

Dynamics of complex systems often involve thermally activated barrier-crossing events that allow a system to move from one basin of attraction to another on the free energy surface. Such events are ubiquitous, but challenging to simulate using conventional simulation tools like molecular dynamics. I will show how a set of dynamical equations obtained from a simple modification of molecular dynamics can be used to determine transition states and escape paths from basins of attraction on a potential energy landscape. Mathematically, it can be shown that the stable fixed points of such a dynamical system are the saddle points of the energy surface. The method provides the flexibility of sampling an ensemble of saddle configurations. As an example of application, I will discuss results of deformation processes such as dislocation nucleation, surface diffusion, vacancy-adatom formation, etc. taking place during nano-indentation. Using these results, a deformation mechanism map is proposed for the complex atomistic processes taking place during nanoindentation.

Back