Investigation of the Pt-Cu surface alloy formation with the Self-learning kinetic Monte Carlo algorithm
M.V.Lomonosov Moscow State University, Faculty of Physics, Leninskie Gory, Moscow 119991, Russia
Self-learning kinetic Monte Carlo (SLkMC)  algorithm is widely used for the simulation of the formation of nanostructures. It is relatively simple for implementation and has low resource cost. However, sometimes on the energy surface there are areas with low diffusion barriers of jumps inside of the area and high barriers to jump out. Such areas are called energy basins. When program gets inside the energy basin it models diffusion of the system inside of the energy basin during numerous steps. Simulated time don't grow and the algorithm becomes very inefficient.
There are various methods for acceleration of the SLkMC . They range in accuracy, resource cost and complexity. Their common feature is that they require knowledge of the energy basin's structure. In this work we propose our method of finding energy basins. We suppose that it will be useful for modeling of complicated systems with various degrees of freedom.
To test our algorithm we simulated formation of the surface alloy during deposition of Pt atoms onto stepped Cu(111) surface. Diffusion barrier for jump of the Cu atom near embedded to the step edge Pt atom is quite low. Therefore, there is a need for acceleration of the algorithm. Our results are in a good qualitative agreement with experimental work wherein such process was investigated .
1. Trushin O., Karim A., Kara A., Rahman T.S. Self-Learning Kinetic Monte Carlo Method: Application to Cu(111), Phys. Rev. B, 72, 2005. pp. 115401.
2. Puchala B., Falk M.L., Garikipati K., An energy basin finding algorithm for kinetic Monte Carlo acceleration, J. Chem. Phys., 132, 2010. pp. 134104.
3. Lucci F.R., Lawton T.J., Pronschinske A., Sykes E., Charles H., Atomic Scale Surface Structure of Pt/Cu(111) Surface Alloys, J. Phys. Chem. C, 118, 2014. pp. 3015-3022.