Can a computer crystallize a liquid? Molecular simulation of continuous trajectories from liquid to crystalline n-hexane

Can a computer crystallize a liquid? Molecular simulation of continuous trajectories from liquid to crystalline n-hexane
复制标题

计算机可以使液体结晶吗?

DOI:
--
复制
发表时间:
2011
期刊:
影响因子:
--
通讯作者:
A. Gavezzotti
A. Gavezzotti
中科院分区:
--
文献类型:
--
作者:
A. Gavezzotti

文献摘要

被引文献

相似文献

一种新的算法描述的组织过程,导致从各向同性液体的晶体结构的模拟。该算法是基于一个混合,伪蒙特卡罗技术,让分子组装演变的大都会条件下,迫使减少不对称参数,量化偏离分子间的对齐。将该程序应用于正己烷中液-固相变的模拟。从液体正己烷与一个显着的人口的gauche分子构象,产生一个结晶分子组装与平行的全反式脂肪链。这种结构与真实的晶体结构非常相似,除了层内分子取向的微小差异。该过程显然涉及几kJ mol−1的活化能和非常小的体积活化,这可能与密度波动有关,因为长分子经历大的构象重排时需要额外的空间。虽然模拟的一些详细的定量方面可能是开放的讨论,这是一个连续的全原子轨迹连接的液体和固体状态的一个大的多原子分子的个性化的非常少的例子之一。结果证实了对称性偏置的可行性,为新的晶体结构的产生,并提供有价值的工作机制的结晶假说。
A new algorithm is described for the simulation of the organization processes that lead from an isotropic liquid to a crystal structure. The algorithm is based on a hybrid, pseudo-Monte Carlo technique that lets a molecular assembly evolve under Metropolis conditions subject to forcing the decrease of asymmetry parameters that quantify the deviation from intermolecular alignment. This procedure was applied to the simulation of the liquid–solid transformation in n-hexane. Starting from liquid n-hexane with a significant population of gauche molecular conformations, a crystalline molecular assembly with parallel all-trans aliphatic chains is generated. This structure is very similar to the real crystal structure, except for a small difference in molecular orientation within layers. The process apparently involves an activation energy of a few kJ mol−1 and a very minor volume activation that could be associated with density fluctuations due to the need for extra space as long molecules undergo a large conformational rearrangement. Although some detailed quantitative aspects of the simulation may be open to discussion, this is one of the very few examples of individuation of a continuous all-atom trajectory linking the liquid and the solid state of a large polyatomic molecule. The results confirm the viability of the symmetry bias for the generation of new crystal structures, and provide valuable working hypotheses on the mechanism of crystallization.