|
Long time scale simlulations are carried out using the adaptive kinetic Monte Carlo (AKMC) method within the harmonic transition state theory approximation. The mechanism and rate of possible transitions from each state are found in an unbiased way by searching for low lying saddle points on the potential energy rim using the minimum mode following method. Systematic coarse graining of the energy landscape is used to speed up the simulation, and distributed computing is used to make use of a large number of loosely connected computers. Application to grain boundary structure and diffusion, surface diffusion and growth, and nanocluster annealing will be presented. Extensions to quantum mechanical tunneling and spin systems will also be introduced. For references see: http://www.hi.is/~hj/publications.html |
|