The DFT Study Of Diffusion Barrier And KMC Simulation Of Long-range Interactions On Cu(111) Surface

Ordered nanostructure thin films on substrate are widely used in scientific and technological fields, and the study fruits of thin film science are increasingly translated into productivity. The understanding of the initial stage of thin film process such as diffusion and nucleation from atomic-level have much significance in optimizing the growth technics and capability. In experiments, scanning tunnel microscope(STM) and atom force microscope (AFM) are used to observe the atom actions in the thin film process. However, in theory, different methods are fit for different time and space scale, and the methods which match the actual growth process must be selected. In this paper, the method that combined density-functional theory(DFT) and kinetic Monte Carlo(kMC) simulation is proposed. Firstly, the diffusion barrier of adatoms on Cu(111) surface is calculated using DFT, and then the self-organization and nucleation rule of magnetic adatoms on Cu(111) surface are simulated using kMC method in which long-range interactions are considered.The diffusion barrier of Cu on Cu(111) surface is calculated by means of geometrically optimizing the absorption models for fcc and hcp absorbed sites, and then calculating the transition state from fcc site to hcp site, which is based on the DFT and supercell approach. The results show that the model of diffusion barrier is accurate when the supercell is of six total layers with two relaxed layers, the coverage is 1/4ML (monolayer) and the vacuum space is 1.0nm. In addition, the diffusion barriers and magnetic moments of magnetic adatoms, such as Fe,Co,Ni on Cu(111) surface are obtained with the calculation model above.Then the self-organization of Mn adatoms on Cu(111) surface is simulated by kinetic Monte Carlo method with the activation barrier obtained above and the long-range interactions between Mn adatoms. The simulation results show that the self-organization of Mn adatoms is sensitively dependent on the coverage and temperature. At low coverage the Mn adatoms are likely to form linear chains at first and to locally ordered nano-superstructures when it becomes steady at about 12K and 18K. If the number of Mn adatoms is larger than the saturated coverage, the periods of the superstructures will decrease with the increase of coverage. Analyzing the formation mechanism of superlattice, we find that the size of the repulsive ring, the distance to the first minimal potential, and the activation barrier are crucial parameters to form superlattice. At last, with the diffusion barrier obtain above the nucleation rule of Ni adatoms with the long range interactions on Cu(111) surface are simulated with kinetic Monte Carlo method. The results show that the nucleation is delayed, the island density is increased and the island size becomes much uniform, which distributes in the range of 2-6 atoms if the repulsive potential of the long-range interactions is considered. In addition, the relationship between island density and substrate temperature becomes much nonsensitive and no longer obeys the classical nucleation law.