刺激响应性智能材料的分子动力学模拟

This research aims for theoretical designing and engineering novel stimuli-responsive system. The challenge lies in the accurate simulation of the dynamic switching process of the stimuli-responsive material. We plan to tackle this challenge by developing the efficient molecular dynamics simulation methods based on polarization model and ab initio calculations, respectively. To yield reliable predictions on the critical electric field to trigger the switching, the polarization and induced electric field will be considered in our polarization models. The introduction of the enhanced sampling techniques and polarizable force field based molecular dynamics simulation into the fragment-based ab initio molecular dynamics method will enlarge the conformational space obtained by the ab initio calculations. The employment of data mining and graph segmentation techniques will allow the automatic construction of fragment based coarse grain electrostatic model. The relationship between the structure and function will be established from the factor analysis of the molecular dynamics trajectories. The changes in entropy and free energy of molecular aggregates during the switching process will also be studied in this project, providing useful information to guide the assembly of smart materials.

设计与制备对电场或光照等激励具有响应性的智能材料是一个重要的课题. 如何精确模拟激励-响应性智能材料的动态开关行为至关重要.本项目拟考虑电场下的静电极化和诱导内电场，完善可极化静电模型；将增强采样方法引入从头算分子动力学模拟，混合使用基于可极化力场的分子动力学方法，增大搜索的构象空间，提高基于分子片从头算分子动力学模拟的效率；借鉴数据挖掘和图像识别等计算机技术，自动构建粗粒化静电模型，利用主成分因子分析处理上述各模拟方法得到的轨迹，探讨智能材料结构与性质之间的关系；研究光照、电场作用下，分子构象变化过程中的熵和自由能的变化， 模拟材料组装过程中电荷、偶极矩间相互作用的动态变化，为智能材料的可控组装提供理论预测。

在刺激-响应性材料设计中，需要通过理论计算模拟，研究材料分子在光或者电的刺激下的动态开关行为。本项目重点发展反应性力场，研究偶氮苯衍生物的异构化过程。发展可极化静电模型，并将增强采样方法引入分子动力学模拟，混合使用基于可极化力场的分子动力学方法，研究凝聚相中复杂分子体系的构象变化；运用机器学习方法，研究分子构象变化及材料的性质，探讨智能材料结构与性质之间的关系，为材料的设计提供理论预测。部分光响应材料的预测得到了实验的验证。