| The development and application of a new approach for evolving complex atomistic systems will be presented. The approach employs an off-lattice (relaxed) atomistic kinetic Monte Carlo (KMC) method to evolve the system, with the evolution at each time step dictated by a saddle point search that identifies possible atom movements. The concept of 'active volumes' is introduced to accelerate the time dynamics. In contrast to conventional KMC models that require all reactions to be predetermined, the new approach is self-evolving and any physically relevant motion or reaction may occur. In order to estimate the fidelity of dynamics and accuracy of the simulation time, a general criterion is proposed and applied to representative problems, including the diffusion of interstitials and vacancies in bcc iron. Direct comparisons are made with MD simulations which demonstrate that this method can be used to investigate dynamic processes and generate results that formerly could be obtained only through MD. The correlation factor for SIA in dumbbell configuration, which is difficult to obtain using MD, is predicted using SEAKMC. The strengths and limitations of SEAKMC are discussed. |
|