Development of Molecular Simulation Methods to Compute Phase and Interfacial Properties of Complex Fluids

纽约州立大学布法罗分校的杰弗里·埃林顿和安德鲁·舒尔茨获得了化学部化学理论、模型和计算方法项目的支持，以开发预测复杂分子体系性质的计算方法和工具。该项目由工程部化学、生物工程、环境和运输系统中的计算和数据使能科学与工程计划共同资助。该项目的重点是相态(气体、液体、固体)和界面(例如，空气和水相遇的地方)的性质。这些性质在许多自然现象和许多工业过程中发挥着关键作用。对科学家和工程师特别有价值的是理解微观相互作用之间的关系，例如原子水平上的相互作用，以及它所表现出的宏观行为之间的关系。这样的信息可以用来调整系统的分子级细节，以获得所需的行为。从理论上讲，分子模拟为研究复杂流体的相和界面行为提供了理想的工具。虽然在这一领域已经取得了巨大的进步，但仍然需要高效和有效的计算方法。Errington和Schultz教授和他们的团队正在开发强有力的新策略，通过分子模拟来研究复杂流体的相和界面性质。说明这种方法必要性的例子包括分离技术、储能装置、碳捕获策略和表面涂层的设计。这项研究的结果将在更广泛的研究界免费获得的软件中实施。这项研究的重点是发展分子模拟方法，使人们能够推导出复杂流体的本体和界面性质。目前正在寻求两个方法学上的进展：(1)一种计算流体整体液-气饱和性质的新的严格策略；(2)一种基于力的策略，以确定等温-等压系综内的扩散界面势。首先，在跨越气-液共存区域的多种密度下收集维里压力测量，并随后用于构建等温-等压系综内的体积概率分布。该方法提供与通常使用的平面直方图方法相同的信息水平，不需要分子插入/删除，并且可以在分子动力学框架内实现。第二个方法学进步是为了增加扩散界面电位法的可及性。界面电势提供了有关系统润湿行为的定性和定量方面的重要见解。一般的方法提供了一种方法来确定液滴在母蒸汽中的固体衬底上的接触角。该项目解决了在常用的等温-等压系综内实施该方法所面临的挑战。标准方法会导致高度拉长的模拟框，在实际操作中很难使用。研究小组已经确定了一种方法，通过使用虚拟盒子来显著减小所需的模拟盒子的大小。该方法利用基于力的策略来计算界面势。与以前的发展相结合，这一进展为确定模型系统的润湿特性提供了一种严格、高效和可访问的方法。在第三个努力中，开发了工具来促进蒙特卡洛算法和分子动力学算法在单个分子模拟框架内的耦合。这种耦合提高了上述方法的效率。具体地说，该小组正在为公开和免费提供的LAMMPS分子动力学模拟器贡献更多的蒙特卡罗移动类型。这些贡献预计将有益于广大LAMMPS用户社区。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Jeffrey Errington and Andrew Schultz of SUNY at Buffalo are supported by an award from the Chemical Theory, Models and Computational Methods program in the Division of Chemistry to develop computational methods and tools to predict properties of complex molecular systems. This project is co-funded by the Computational and Data-Enabled Science and Engineering in Chemical, Bioengineering, Environmental and Transport Systems Program in the Division of Engineering. The project focuses on phase (gas, liquid, solid) and interfacial (e.g. where air and water meet) properties. The properties play a key role in many natural phenomena and in numerous industrial processes. Of particular value to scientists and engineers is an understanding of the relationship between the microscopic interactions, for example at the atomic level, and the macroscopic behavior it exhibits. Such information can be used to tune the molecular-level details of a system to obtain a desired behavior. In principle, molecular simulation provides an ideal tool for studying the phase and interfacial behaviors of complex fluids. Although tremendous advancements have been made in this area, there is still a huge need for efficient and effective computational methods. Professors Errington and Schultz and their groups are developing robust new strategies for interrogating the phase and interfacial properties of complex fluids via molecular simulation. Examples that illustrate the need for such methods include the design of separation technologies, energy storage devices, carbon capture strategies, and surface coatings. The results of this research will be implemented in software that is freely available to the broader research community. The focus of this research is to develop molecular simulation methods that enable one to deduce the bulk and interfacial properties of complex fluids. Two methodological advances are being pursued: (1) a new rigorous strategy for computing the bulk liquid-vapor saturation properties of fluids and (2) a force-based strategy to determine the spreading interface potential within an isothermal-isobaric ensemble. For the first, virial pressure measurements are collected at multiple densities that span the liquid-vapor coexistence region, and subsequently used to construct a volume probability distribution within the isothermal-isobaric ensemble. The method provides the same level of information as commonly-used flat histogram approaches, does not require molecule insertions/deletions, and can be implemented within a molecular dynamics framework. The second methodological advance is aimed at increasing the accessibility of the spreading interface potential method. The interface potential provides important insight regarding qualitative and quantitative aspects of a system's wetting behavior. The general approach provides a means to determine the contact angle of a liquid droplet on a solid substrate in a mother vapor. The project addresses challenges associated with implementing the method within the commonly-used isothermal-isobaric ensemble. The standard approach results in highly elongated simulation boxes that are difficult to work with in practice. The research team has identified a means to significantly reduce the size of the simulation box required via use of a virtual box. The approach leverages force-based strategies to compute the interface potential. When combined with previous developments, this advance provides a rigorous, efficient, and accessible approach for determining the wetting properties of model systems. In a third effort, tools are developed to facilitate coupling of Monte Carlo and molecular dynamics algorithms within a single molecular simulation framework. Such a coupling enhances the efficiency of the methods noted above. Specifically, the group is contributing additional Monte Carlo move types to the publically- and freely-available LAMMPS Molecular Dynamics Simulator. These contributions are expected to be beneficial to the broad LAMMPS user community.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.