Tavazza F, Nurminen L, Landau D P, Kuronen A, Kaski K
Center for Simulational Physics, The University of Georgia, Athens, Georgia 30602-2451, USA.
Phys Rev E Stat Nonlin Soft Matter Phys. 2004 Sep;70(3 Pt 2):036701. doi: 10.1103/PhysRevE.70.036701. Epub 2004 Sep 14.
A classical, hybrid Monte Carlo-molecular dynamic (MC-MD) algorithm is introduced for the study of phenomena like two-dimensional (2D) island stability or step-edge evolution on semiconductor surfaces. This method presents the advantages of working off lattice and utilizing bulk-fitted potentials. It is based on the introduction of collective moves, such as dimer jumps, in the MC algorithm. MD-driven local relaxations are considered as trial moves for the MC. The algorithm is applied to the analysis of 2D Si islands on Si(001). Results on early stages of island formation, island stability versus temperature and system size, and step-edge evolution are presented. In all cases good qualitative agreement with experimental results is found.
为了研究诸如二维（2D）岛稳定性或半导体表面台阶边缘演化等现象，引入了一种经典的混合蒙特卡罗 - 分子动力学（MC - MD）算法。该方法具有非晶格工作和使用体拟合势的优点。它基于在蒙特卡罗算法中引入集体移动，如二聚体跳跃。分子动力学驱动的局部弛豫被视为蒙特卡罗的试探移动。该算法应用于对Si(001)上二维硅岛的分析。给出了岛形成早期阶段、岛稳定性与温度和系统尺寸的关系以及台阶边缘演化的结果。在所有情况下，都发现与实验结果有良好的定性一致性。
