CIESC Journal

• 综述与专论 •    下一篇

分子模拟在化工应用中的若干问题及思考

朱宇;陆小华;丁皓;王俊;王延儒;时钧   

  1. 南京工业大学化学化工学院,江苏 南京 210009

  • 出版日期:2004-08-25 发布日期:2004-08-25

MOLECULAR SIMULATION IN CHEMICAL ENGINEERING

ZHU Yu;LU Xiaohua;DING Hao;WANG Jun;WANG Yanru;SHI Jun   

  • Online:2004-08-25 Published:2004-08-25

摘要: 随着高技术学科的飞速进步,化工学科在多年来已形成的理论和实验研究之外,又产生了一种完全独立而新颖的研究手段——分子模拟.目前化学工业受关注的新技术涉及聚合物、电解质等复杂物质,临界、超临界等复杂状态,界面、膜、溶液等复杂现象.实现化学工业从产品到过程设计完全自动化,在这些方面除了准确的物性数据外,更要对各种复杂现象的机理有深刻了解.分子模拟被认为是实现这一目标的关键技术之一.本文以分子动力学为主,结合计算量子化学,对分子模拟在化工应用中的若干问题进行讨论.

Abstract: With the development of high-technology,molecular simulation,which is novel both in theory and experiment,has been applied in chemical engineering.The new technologies in chemical industry include such complex materials as polymer and electrolyte,such complex conditions as critical and super critical,and such complex phenomenon as interface, membrane and solution.To achieve the vision of completely automated product and process design in chemical industry, the properties of the materials and the mechanism of the phenomena should be accurately obtained.The expectations for molecular simulation are very high, because it has been identified as one of the key methods for determining these properties and phenomena.In this paper, the key problems of molecular simulation in chemical engineering especially based on molecular dynamics (MD) and quantum chemistry (QC) are discussed.They include potential model of MD, scale of MD, the connection of muti-scale MD, analysis of MD results, construction of a reasonable initial configuration and software & hardware of MD.Several promising guides are proposed to solve the above key issues.Computational quantum chemistry software such as Gaussian 98 could be utilized to acquire the missing interaction parameters.As for potential model it is not necessary to use the most complex one to describe the interaction between particles.Sometimes simple models are good enough to calculate certain properties.Some new techniques are developed to enlarge the system of MD at different scales.For example, the density functional theory (DFT) made the QC calculation for hundreds of atoms possible while dissipative particle dynamics (DPD) extended the system of MD to mesoscale.The general trend of molecular simulation is to combine QC, MD and mesoscale dynamics more closely, and many muti-scale MD methods such as Car-Parrinello MD (CPMD) would definitely become stronger and stronger.It is indispensable to define a suitable statistical variable to analyze the trajectory of MD, and the trick of definition is to deal with the specific situation in a special way.Normally initial configurations of MD are not the images of real systems but the abstract of real ones.Real systems must be reduced to different extents according to different goals.Sharing the MD codes and applying parallel computers could be regarded as the most promising solutions within the limitations of software & hardware of MD.