Computer Simulation For The Metallic Nanostructure Formation Mechanism On Liquid Surfaces

In this dissertation, in order to study the formation mechanisms of metallic nanoclusters and nanocrystals on the liquid surfaces, three computer simulation models by using the Monte Carlo simulation method are established. Then simulations are carried out systematically. The morphology, size distribution and growth mechanism of the nanosystems with different parameters are studied. The simulation results are in good agreement with the experimental findings. Finally, several suggestions for further research in this field are presented.In order to study the aggregation mechanism of nanoclusters on liquid surfaces, based on the traditional cluster-cluster aggregation (CCA) picture, a modified growth model (namely, RCCA model) is established. N particles are deposited randomly on a square lattice substrate (avoiding overlap). Then the clusters (including single particles and their aggregates) move randomly with diffusion step length l and aggregate irreversibly when they meet. If the number of particles contained in a cluster is larger than a critical size, the particles at the edge of the cluster possess a possibility to jump onto the upper layer, which results in the crossover from 2-dimensional to 3-dimensional aggregations. The main results are as follows:(1) the coverage increases linearly with the particle number when the coverage is smaller than 0.06 ML; (2) as the particle number further increases, a nonlinear dependence between the coverage and the particle number is observed. (3) The average height of the clusters increases with the particle number and the diffusion length and then approaches a saturation value. The results are accordance with the experimental data.Based on the experimental observation, we establish a new cluster condense model (namely, CAC model), in which, if the size of a cluster is larger than the critical size, then it may have certain possibility to collapse. This process leads to the decrease of the cluster volume and increase of the cluster density. Our simulation results reveal that a random fluctuation behavior of the coverage occurs when the particle number further increases. More precisely speaking, when the particle number is small (< 0.06 ML), most of the clusters are smaller than the critical cluster, which leads to the linear increase of the coverage; when the particle number is larger (> 0.08 ML), the average size of the clusters goes beyond the critical size, then the collapse phenomenon of the clusters happens. Finally, the average cluster density increases gradually and reaches a stable value. These results may be used to explain the experimental findings successfully.Based on the hypothesis of the preferential growth direction of crystals, a one-dimensional crystal growth model (namely, OCG model) is established, which is performed on a square lattice substrate. First, particles are deposited homogeneously and, as a result, each of the lattice sites is occupied by one particle. In the subsequent stage, N nuclei (seed number) are selected randomly on the substrate, then the growth process starts by adsorbing the surrounding particles along the preferential growth directions of the crystals. Finally, various one-dimensional crystals with different length and width form. The statistic analyses of the simulation results prove that the length and width distributions of the crystal rods are in accordance with the lognormal distribution. When the seed number N increases, both the length and width distribution peaks are gradually concentrated. The average length and width decrease with N, and are proportional to (1/N)0.60 and (1/N)0.28, respectively. In a word, the preferential growth direction phenomenon is commonly existed in various crystal growth processes, our simulation results explain the growth mechanism of the zinc crystal rods formed on the isotropic liquid surfaces successfully.The dissertation contents are organized as follows:In chapter 1, we introduce the classic experimental facilities, characterization methods, physical properties and the basic growth theories of thin films. The recent development of the experimental and theoretical studies in this field is presented. In particular, the computer simulation studies of the film growth mechanism are reviewed. Finally, we point out the purpose and the meaning of this study.In chapter 2, we establish an improved CCA model, assuming that the edge particles possess a possibility to jump onto the upper layers when the cluster is larger enough. The model reveals the crossover from 2-dimensional to 3-dimensional aggregation of the metallic clusters on liquid surfaces. The dependence of the coverage and average height of the clusters on different parameters are studied.In chapter 3, a cluster condensation model is established on a nonlattice substrate. In this model, if the size of a cluster is larger than the critical size, then it may have certain possibility to collapse. The evolution of the coverage and the average cluster density is systematically investigated.In chapter 4, based on the experimental observation of the preferential growth direction phenomenon in crystals, a one-dimensional aggregation and crystal growth model on a square lattice substrate with periodic boundary conditions is established. The evolution of the length and width distributions with the seed number and other properties are simulated. As a result, the formation mechanism of zinc crystal rods on isotropic liquid surfaces is well explained. Finally, we show the potential applications of the new growth mechanism in the future.In chapter 5, we conclude the research results of this dissertation. Then a proposal for the next work and the prospect of the future research are presented.