Soft Condensed Matter, Debye Institute for NanoMaterials Science, Utrecht University, Princetonplein 5, 3584 CC Utrecht, The Netherlands.
J Chem Phys. 2011 Jul 21;135(3):034510. doi: 10.1063/1.3609103.
We use computer simulations to investigate the crystallization dynamics of sedimenting hard spheres in large systems (hundreds of thousands of particles). We show that slow sedimentation results primarily in face-centered cubic (fcc) stacked crystals, instead of random hexagonal close packed or hexagonal close packed (hcp) crystals. We also find slanted stacking faults, in the fcc regions. However, we attribute the formation of fcc to the free energy difference between fcc and hcp and not to the presence of these slanted stacking faults. Although the free energy difference between hcp and fcc per particle is small (only 10(-3) times the thermal energy), it can become considerable, when multiplied by the number of particles in each domain. The ratio of fcc to hcp obtained from dynamic simulations is in excellent agreement with well-equilibrated Monte Carlo simulations, in which no slanted stacking faults were found. Our results explain a range of experiments on colloids, in which the amount of fcc increases upon lowering the sedimentation rate or decreasing the initial volume fraction.
我们使用计算机模拟来研究在大型系统(数十万粒子)中沉降硬球的结晶动力学。我们表明,缓慢的沉降主要导致面心立方(fcc)堆积晶体,而不是随机六方密堆积或六方密堆积(hcp)晶体。我们还在 fcc 区域发现了倾斜的堆垛层错。然而,我们将 fcc 的形成归因于 fcc 和 hcp 之间的自由能差,而不是这些倾斜的堆垛层错的存在。尽管每个粒子的 hcp 和 fcc 之间的自由能差很小(仅为热能的 10(-3)倍),但当乘以每个域中的粒子数时,它可能变得相当大。从动态模拟中获得的 fcc 与 hcp 的比例与没有发现倾斜堆垛层错的充分平衡蒙特卡罗模拟非常吻合。我们的结果解释了一系列胶体实验,其中在降低沉降速率或降低初始体积分数时,fcc 的量增加。