Au/Cu(001)异质外延岛演化的分子动力学研究

利用分子动力学弛豫方法模拟了Au/Cu(001)异质外延生长初期Au异质外延岛的形貌演化，分析了Au外延岛演化过程中的局域应力及与基体结合能随表面岛尺寸的变化. 研究结果表明：当异质外延岛小于7×7时，外延岛原子分布呈现赝Cu点阵形貌；当外延岛达到8×8后，外延岛内开始出现失配位错，失配位错数量随外延岛尺寸的增加而增加. 局域压力分析指出，外延岛上原子之间的近邻环境不同导致了所受应力的差异，而外延岛的形变则是由外延岛原子的应力分布所决定. 研究还发现，失配位错的产生导致错位原子与基体原子之间的结合强度减弱，但相对增加了非错位原子与基体原子之间的结合强度. 关键词： 异质外延 表面形貌 局域压力 分子动力学模拟     

晶体相场法模拟异质外延过程中界面形态演化与晶向倾侧

采用晶体相场模型研究了异质外延过程中失配应变与应力弛豫对外延层界面形态演化的影响, 并对由衬底倾角引起的外延层晶向倾侧进行了分析.研究结果表明: 在有一定倾角的衬底晶体上进行外延生长时,若衬底和外延层之间失配度较大 (ε>0.08),外延层中弹性畸变能会以失配位错的形式释放, 最终薄膜以稳定的流动台阶形式生长且外延层的晶向倾角与衬底倾角呈近似线性关系. 而当衬底和外延层之间失配度较小(ε<0.04)不足以形成失配位错时, 外延层中弹性畸变能会以表面能的形式释放,最终使薄膜以岛状形态生长. 在高过冷度条件下,衬底倾角和失配度较大时,衬底和外延层之间会形成由大量位错规则排列而成的小角度晶界从而显著改变外延层的生长位向.     

采用元胞自动机法模拟二元规则共晶生长

基于元胞自动机方法,建立了二元规则共晶生长的数值模拟模型.该模型耦合宏观温度场,考虑了溶质扩散、成分过冷以及曲率过冷等重要因素,实现了二元规则共晶的稳态层片生长.以共晶模型合金为对象,研究了定向凝固条件下不同过冷度、初始层片间距、温度梯度及凝固速率对共晶两相生长形貌及层片间距的影响,再现了两相在溶质扩散及界面能的相互作用下通过形核、分岔、湮没及合并等机理实现共晶层片间距的调整.模拟得到的结果与Jackson-Hunt模型及前人实验结果规律一致.将模型扩展到三维系统,验证了二元规则共晶生长三维模拟的可行性. 关键词： 规则共晶 生长形貌 层片间距 CA方法     

多晶凝固及后续调幅分解过程的晶体相场法模拟

采用晶体相场模型,模拟了二元合金多晶凝固及后续调幅分解全过程.结果表明,晶体相场模型可完整再现包括形核、生长、粗化、晶界形成等多晶生长过程以及圆满完成从凝固到调幅分解的多相变过程. 关键词： 晶体相场模型 多晶凝固 调幅分解 组织演化     

BCC枝晶生长原子堆垛过程的晶体相场研究

通过在自由能泛函中引入各向异性参数得到了一个基于高斯内核的改进晶体相场模型, 并采用该模型研究了体心立方结构(BCC)枝晶生长的原子堆垛过程. 结果表明, 在BCC由正十二面体平衡形貌演化为枝晶组织过程中, 形核位置经历了由面心({110}面)到尖端(<100>取向)的转移, 进而发生界面失稳形成枝晶组织; 枝晶生长过程中, 新的固相原子首先在枝晶尖端附近形核, 并快速向尖端及根部生长, 枝晶尖端被新原子完全包覆后将再次诱发液相原子附着形核及生长; 随初始液相密度的增加, 固-液界面移动速率增加, 速率系数的各向异性也增强.     

双模晶体相场研究形变诱导的多级微结构演化

本文采用双模晶体相场模型,计算了双模二维相图;模拟了形变诱导六角相向正方相转变过程的多级微结构演化,详细分析了位相差、形变方向对位错、晶界、晶体结构、新相形貌的影响规律.模拟结果表明:形变方向影响正方相晶核的形核位置和生长方向,拉伸时正方相优先在变形带上形核,垂直于形变方向长大,而压缩时正方相直接在位错和晶界的能量较高处形核,平行于形变方向长大;位相差对形变诱发晶界甄没过程有显著影响,体现在能量峰上为,小位相差晶界位错的攀滑移和甄没形成一个能量峰,大位相差晶界位错攀滑移和甄没因分阶段完成而不出现明显的能量峰;形变诱导相变过程中各种因素相互作用复杂,是相变与动态再结晶的复合转变.     

高温下金属薄膜生长初期的模拟研究

采用实际的生长模型和物理参量,用Monte Carlo方法对高温下金属薄膜的生长过程进行了模拟研究.综合考虑了原子沉积、扩散、成核、生长和扩散原子的再蒸发、原子沿岛周界扩散和岛的合并等众多过程后,模拟得到与实验结果相当一致的薄膜生长形貌及其相应的定量结果.通过动态统计薄膜生长过程中的岛数目及薄膜生长率,得到实验中不易直接获得的高温下薄膜生长的许多细节,如岛数目和薄膜生长率随表面温度、覆盖度变化的详细情况等 关键词： 薄膜 Monte Carlo模拟 成核 岛密度 薄膜生长率     

外加应力作用下UO2中空洞演化过程的相场模拟

本工作建立了外加应力作用下UO₂中空洞演化的相场模型.首先,使用摄动迭代法求解了弹性平衡方程,对外加应力下单个空洞周围的应力分布进行了计算,结果表明空洞边缘有应力集中现象,模拟得到的应力分布和解析解一致.然后,利用相场方法模拟了不同外加应力下单个空洞的演化过程,结果表明随着外加应力的增大,空洞的生长速度加快.最后,研究了外加应力对多晶体系中晶粒长大和空洞演化的影响,结果表明,不同晶粒内的应力大小不同,应力越小的晶粒越容易长大,尺寸越大的空洞的边缘应力也越大.晶间空洞与弯曲晶界存在相互作用,一方面晶界附近的空洞会生长成透镜状,另一方面空洞对晶界也有钉扎作用,能减缓晶界的迁移.此外,外加应力会加速多晶系统中空洞的生长,并且本文计算得到了外加应力与空洞半径的关系,发现外加应力越大,空洞的生长越快.     

非均匀基底上三维薄膜生长的模拟研究

考虑原子在基底表面的扩散、沿岛周界的扩散和不同层间的扩散以及非均匀基底上表面吸附能分布的各向异性，建立起非均匀基底表面上原子扩散和三维薄膜生长的动力学蒙特卡罗模型.模拟得到在不同生长条件下出现的层状生长、岛状生长和混合生长三种生长模式和相应的多层薄膜生长形貌图.通过统计三维薄膜中原子在各层的分布，计算薄膜的表面粗糙度，得到薄膜生长模式与生长条件之间的关系. 关键词： 薄膜生长 非均匀基底 动力学蒙特卡罗模拟     

横向外延过生长磷化铟材料的生长速率模型

分别考虑气相扩散和掩膜表面扩散过程，建立了金属有机化学气相沉积条件下横向外延过生长的速率模型. 在砷化镓衬底上外延磷化铟条件下，模拟得到了生长速率随掩膜/窗口宽度(m/w)变化的关系. 通过讨论掩膜/窗口宽度的影响，说明了掩膜宽度、窗口宽度以及有效掩膜宽度是决定生长速率的关键因素. 以上结论与实验结果一致. 关键词： 横向外延 生长模型 扩散 生长速率     

Two-scale modeling of adsorption processes at structured surfaces

We present an algorithm for the simulation of vicinal surface growth. It combines a lattice gas anisotropic Ising model with a phase-field model. The molecular behavior of individual adatoms is described by the lattice gas model. The microstructure dynamics on the vicinal surface are calculated using the phase-field method. In this way, adsorption processes on two different length scales can be described: nucleation processes on the terraces (lattice gas model) and step-flow growth (phase field model). The hybrid algorithm that is proposed here, is therefore able to describe an epitaxial layer-by-layer growth controlled by temperature and by deposition rate. This method is faster than kinetic Monte Carlo simulations and can take into account the stochastic processes in a comparable way.     

A phase-field/Monte-Carlo model describing organic crystal growth from solution

In this paper work we present a phase-field/Monte-Carlo hybrid algorithm for the simulation of solutal growth of organic crystals. The algorithm is subsequently used for an investigation of diffusion effects on the growth mechanisms. This method combines a two-scale phase-field model of the liquid phase epitaxial growth and a Monte-Carlo algorithm of the 2D nucleation and thus is faster than previous purely Monte Carlo simulations of crystal growth. The inclusion of supersaturation and diffusion in the method allows the study of crystal growth under various growth conditions. Parameters used in the hybrid algorithm are bound to the energetic parameters of crystal faces, which can be estimated from a detailed study of the actual crystal structure based on a connected nets analysis, which allows the prediction of the shape and morphology of real crystals. The study of the diffusion effect is carried out based on an example of a hydroquinone crystal, which grows from the water solution at various supersaturations. The dependencies of the growth rate and the nucleation rate on the supersaturation indicate the change of the growth mechanism from spiral growth to 2D nucleation. The difference in the growth rate for various faces is in agreement with the crystal morphologies derived from the attachment energy method and observed experimentally. The main result of the simulation is the evaluation of engineering limits for choosing appropriate external process conditions.     

薄膜生长的随机模型

利用Monte Carlo模型研究了薄膜生长初始阶段岛的形貌与基底温度之间的关系,同时还研究了它们与汽相粒子入射剩余能量之间的关系．模型中考虑了三种动力学过程：粒子入射、吸附粒子扩散和粒子脱附,与以前薄膜生长模型的不同之一在于把入射过程看作独立于其他过程,而扩散和脱附过程是相互关联的．结果表明随基底温度的升高,岛的形貌经历了一个从分散生长、分形生长到凝聚生长的变化过程．低温下随汽相粒子入射和剩余能量增加,岛的形貌也经历了同样的变化过程. 关键词：     

Nucleation and growth mechanisms of Fe on Au(111) in the sub-monolayer regime

The growth of Fe on Au(111) at 300 K in the sub-monolayer regime has been investigated using scanning tunneling microscopy, focusing on the mechanisms of nucleation, coalescence and interlayer diffusion. Below a coverage of 0.1 ML, Fe growth proceeds in a well-ordered fashion producing regular arrays of islands, while approaching the island coalescence threshold (above 0.35–0.4 ML), we observed a consistent increasing of random island nucleation. These observations have been interpreted through rate equation models for the island densities in the presence of preferred nucleation sites. The evolution of the second layer fraction has also been interpreted in a rate equation scheme. Our results show that the ordered to random growth transition can be explained by including in the model bond breaking mechanisms due to finite Fe–Fe bond energy. A moderate interlayer diffusion has been inferred from data analysis between the second and the first layer, which has been used to estimate the energy barrier of the adatoms descending process.     

Scaling of submonolayer island growth with reversible adatom exchange in surfactant-mediated epitaxy

Surfactant-mediated epitaxial growth is studied with a realistic model, which includes three main kinetic processes: diffusion of adatoms on the surfactant terrace, exchange of adatoms with their underneath surfactant atoms, and reexchange in which an exchanged adatom resurfaces to the top of the surfactant layer. The scaling behavior of nucleus density and island size distributions in the initial stage of growth is investigated by using kinetic Monte Carlo simulations. The results show that the temperature dependence of nucleus density and island size distributions governed by the reexchanging-controlled nucleation at high temperatures exhibits similar scaling behavior to that obtained by the standard diffusion-mediated nucleation at low temperatures. However, at intermediate temperatures, the exchanging-controlled nucleation leads to an increase of nucleus density with temperature, while the island size distribution scales to a monotonically decreasing function, showing nonstandard scaling behavior.     

Kinetic Monte Carlo simulations of Au clusters on Si(111)-7 × 7 surface

Self-assembled growth of Au nanoclusters on the Si(111)-7 × 7 surface has been studied using kinetic Monte Carlo simulations. A model considering various atomic processes of deposition, adsorption, diffusion, nucleation, and aggregation is introduced, and the main energetic parameters are optimized based on the experimental results. The evolution of surface morphology during Au growth is simulated in real time, from which the atomic behaviors of Au could be really captured. Most of Au atoms diffuse on the substrate in the very early stage of growth, and Au clusters nucleate and grow with the increasing coverage. The competition among various atomic processes results in the distinct distribution of Au clusters under different coverages. The growth conditions are further optimized, showing that the higher uniformity of Au clusters would be obtained at a low deposition rate and an optimal substrate temperature of about 380 K.     

Unified phase-field model for epitaxial growth of multi-scale three-dimensional islands on insulating surfaces

In this paper we present a unified phase-field model for non-equilibrium growths of various three-dimensional metal islands on insulating surfaces. We introduce a phase-field variable to distinguish the island from the non-island regions and substrate and a density variable to describe local density of deposited adatoms. Two partial differential equations with appropriate boundary conditions, as the governing equations, are used to describe the evolution of the three-dimensional metal islands and the diffusion of adatoms. We solve the equations by using an adaptive mesh refinement method so that we can simulate the non-equilibrium growth of three-dimensional metal islands from tens of nanometers to several micrometers. We investigate the dependence of simulated results on the model parameters and experimental conditions. Equilibrium shape of such islands can be obtained through sufficient post-deposition relaxation. Experimental trends of island size and shape on various scales are obtained with reasonable parameters. This method should be a good approach to non-equilibrium growths of multi-scale three-dimensional metal islands.     

Nonlinear evolution of surface morphology in InAs/AlAs superlattices via surface diffusion

Continuum simulations of self-organized lateral compositional modulation growth in InAs/AlAs short-period superlattices on InP substrate are presented. The results of the simulations correspond quantitatively to the results of synchrotron x-ray diffraction experiments. The time evolution of the compositional modulation during epitaxial growth can be explained only including a nonlinear dependence of the elastic energy of the growing epitaxial layer on its thickness. From the fit of the experimental data to the growth simulations we have determined the parameters of this nonlinear dependence. It was found that the modulation amplitude does not depend on the values of the surface diffusion constants of particular elements.