Van der Waals Approach to Systematic Coarse-Graining

宾夕法尼亚州立大学的William Noid得到了化学部化学理论、模型和计算方法计划的支持，该计划开发用于模拟液体和聚合物的热力学和界面性质的理论方法和计算工具。计算机模拟为解决技术挑战和指导具有所需性能的新型材料的设计提供了强有力的见解。虽然使用原子细节模型的模拟通常非常准确，但它们不能提供必要的效率来解决与材料设计相关的较长的长度和时间尺度。相比之下，分辨率较低的粗粒度模型提供了更高的效率，但目前提供的精度有限，并且通常需要为每个新系统和环境进行广泛的参数化。Noid教授和他的同事开发了理论方法来提高粗粒度模型的准确性，并扩大它们的适用范围。此外，诺德教授组织了一个代际科学俱乐部，在更广泛的社区内促进终身科学教育，同时将所有年龄段的学生纳入科学事业。该项目开发并验证了van der Waals(VDW)方法，用于改进自下而上多尺度方法的可转换性和热力学性质。由平均力的多体势能的精确分解驱动，这是粗粒度(CG)模型的正确势能。该模型完美地再现了原子模型的结构和热力学性质。VDW方法：(1)采用领先的变分方法来确定准确描述结构性质的有效势；(2)发展新的变分方法来精确地模拟热力学性质。特别是，这项工作扩展了压力匹配变分原理来模拟非均匀和界面系统中的密度涨落。通过对不同变分方法得到的势的比较，可以洞察其熵分量，并能够对其可转移性进行预测性估计。最终，这项工作可能会开发出预测性CG模型，用于非常准确和高效的中尺度水界面自组装和聚合物电池导电性的模拟研究。

William Noid of the Pennsylvania State University is supported by an award from the Chemical Theory, Models and Computational Methods Program in the Chemistry Division to develop theoretical methods and computational tools for simulating the thermodynamic and interfacial properties of liquids and polymers. Computer simulations provide powerful insight for addressing technological challenges and for guiding the design of novel materials with desired properties. While simulations with atomically detailed models are often very accurate, they do not provide the necessary efficiency for addressing the long length and time scales that are relevant for materials design. In contrast, lower resolution coarse-grained models provide superior efficiency, but currently provide limited accuracy and often require extensive parameterization for each new system and environment. Professor Noid and his coworkers develop theoretical methods for improving the accuracy of coarse-grained models, as well as for extending their range of applicability. Additionally, Professor Noid organizes an intergenerational science club that promotes life-long scientific education within the broader community, while integrating students of all ages into the scientific enterprise. This project develops and validates a van der Waals (vdW) approach for improving both the transferability and also the thermodynamic properties of bottom-up multiscale methods. Motivated by an exact decomposition of the many-body potential of mean force, which is the correct potential for a coarse-grained (CG) model. This model perfectly reproduces both the structural and thermodynamic properties of an underlying atomistic model. The vdW approach: (1) adopts leading variational methods to determine effective potentials that accurately describe structural properties and (2) develops new variational methods to accurately model thermodynamic properties. In particular, the work extends a pressure-matching variational principle for modeling density fluctuations in inhomogeneous and interfacial systems. The comparison of potentials obtained from different variational approaches provides insight into their entropic components and allow for predictive estimates of their transferability. Ultimately, this work may develop predictive CG models for exceedingly accurate and efficient simulation studies of mesoscale self-assembly at aqueous interfaces and conduction in polymer-based batteries.