Monte Carlo Simulation For The Growth Of Magnetron Sputtering Cu Films

Thin solid film technologies have been widely used in modern scientific and technological fields, and the fruits of the thin films science have also been increasingly translated into productivity. The growth mechanism of thin films has been studied deeply via theoretical and experimental methods, and computer simulation technology has become one of important and effective methods because of the limitations of the contemporary micro-testing technology. Since the growth process of thin film is of the random state, Monte Carlo method has naturally been used in the process of computer simulation for the growth of thin films.Firstly, the theories of thin films growth and current simulation research level of films were overviewed in the present thesis, in which the current research situations, simulation models and calculation methods, especially Monte Carlo simulation, were introduced. A description and analysis between molecular dynamics and Monte Carlo method were carried out. Then the application of virtual science, as well as computer graphics in the simulation of thin film growth process was illustrated.Secondly, Monte Carlo method and Q-State Potts model were applied in the simulation of the grains growth of magnetron sputtering Cu films with two-dimensional image. The relation between the growth time and the size of the Cu grains was studied during the simulation. The magnetron sputtering technique was used to prepare the copper films, and the comparison analysis between simulation and experimental results was done. The results indicated that the growth of the grains was of dynamic state, and the number of the grains in unit area was reduced with the growing of average size. However, the number of the grains became to be in the state of dynamic balance after 500 MCS, and the average size was found to be approximately 9.1 units, which was consistent with the experimental results.Finally, a three dimensional Kinetic Monte Carlo method was adopted to simulate the dynamic process of Cu films growth on the substrate with different surface active points and temperatures. The model involves the adsorption, diffusion and desorption of incident atoms. It was investigated some key parameters represented the quality of the films such as surface morphology, deposition rate, RMS roughness and relative density based on the study of the relationship between substrate activity and the growth of the thin Cu films. The results showed that initial surface activity was a vital factor for the growth of films, especially to the crystal lattice structure and morphology of the films. It is indicated that the more activity points the substrate has, the higher the absorption rate of deposition is. However, the activity point is not the sole governing factor. When the activity distribution was shown with fcc crystal lattice structure, the films had perfect morphology, low roughness, splendid relative density and brilliant crystal. Self-repairing phenomenon was also observed via simulating Cu films growth on different substrates. However, it could not eliminate the congenital defects completely.