An effective method to calculate atomic movements in 3D objects with tuneable stochasticity (3DO-SKMF)

We present an effective computer simulation method, called 3D object stochastic kinetic modelling framework (3DO-SKMF), to calculate atomic movements in 3D objects including surface segregation and Gibbs–Thomson effect (surface curvature). Objects with any kind of shapes can easily be considered thanks to the flexibility and versatility of the model and code. Accordingly, the model and the computer code can be used in a wide variety of applications: nanoparticles, nanorods, nanotubes, nanopillars, etc. To increase the versatility of the model, it includes stochastic noise in a tuneable manner. This means that if the noise level is zero, the model is completely deterministic (mean-field), whereas by increasing the noise level the result gets closer and closer to that obtained by a kinetic Monte Carlo calculation. This allows us to calculate processes with activation barriers. Besides demonstrating the capabilities of the model, we also reproduce an experimental result showing decomposition of Ag–Cu nanoparticles.我们提出了一种高效的计算机模拟方法——三维对象随机动力学建模框架（3D object stochastic kinetic modelling framework，缩写3DO-SKMF），用于计算三维对象内的原子运动，涵盖表面偏析与吉布斯-汤姆逊效应（表面曲率）。得益于该模型与代码的灵活性与通用性，任意形状的三维对象均可轻松纳入计算范围。因此，该模型及配套计算机代码可应用于诸多场景：纳米颗粒、纳米棒、纳米管、纳米柱等。为进一步提升模型的通用性，其内置了可调节的随机噪声项。这意味着当噪声水平为0时，模型为完全确定性的平均场（mean-field）模型；而随着噪声水平提升，计算结果将愈发接近动力学蒙特卡洛（kinetic Monte Carlo）方法所得的结果。这使得我们能够计算存在活化能垒的物理过程。除了展示该模型的性能之外，我们还复现了一项关于银-铜（Ag–Cu）纳米颗粒分解的实验结果。