Multiscale modeling of compounds and nanoparticles at liquid and soft matter interfaces

人们普遍认为，溶剂化对于广泛的科学和技术是至关重要的。特别是，液体界面在化学、电化学、生物催化、生物学和药理学的各种技术应用中都是至关重要的。在液体界面，水和有机溶剂相遇，并创造出具有独特性质的环境，可以微调以促进反应性和自组装。软物质纳米结构是由相互竞争的化学功能形成的复杂的液体界面。一个重要的例子是活细胞和细胞内结构中生物膜的脂质双层。对液体界面的结构和功能进行建模提出了一项挑战，这一挑战已经在不同的复杂程度上进行了尝试，从药物跨生物膜的传输估计为水和辛醇的大块液态之间的分配，到对纳米颗粒跨脂双层传输的粗粒度模拟。目前，特别是在化学和生物分子纳米系统和纳米材料的局域限制中，需要能够充分考虑溶剂分子结构的溶剂化理论模型。这种理论模型对扩展分子模拟的能力有很大的需求，分子模拟到目前为止还不足以解决从快到慢的整个尺度上的复杂现象和过程，在感兴趣的大型纳米系统中。基于统计力学的溶剂化理论模型可以耦合到预测性多尺度模型的所有部分，包括对化学结构和活性中心的量子描述，对纳米结构的分子动力学(MD)和耗散粒子动力学(DPD)，以及对系统功能的粗粒度模型。建立在统计力学第一原理基础上的液体积分方程式理论正变得越来越流行，因为它为处理溶液中复杂的化学和生物分子系统提供了坚实的平台。特别是，溶剂化的3D-RISM-KH分子理论(具有Kovalenko-Hirata闭合关系的三维参考相互作用位模型)已经显示出在溶液中的许多系统取得了实质性的成功，否则无法进行分子模拟或连续溶剂化处理。目前已进入实用化阶段，多尺度方法与量子化学、分子动力学和DPD模拟方法以及配体对接协议相结合，主要软件包有：阿姆斯特丹密度泛函(ADF)计算化学程序包、Amber MD程序包、AutoDock对接工具套件[9]和分子操作环境(MOE)程序包。此外，Lovett-Mou-Buff-Wertheim(SS-LMBW)积分-微分方程式的位置-位置推广构成了液体界面的分子理论，该理论已在非极性和复杂缔合液体的平面界面上得到验证。该项目将3D-RISM-KH和SS-LMBW方法结合在液体和软物质界面的新的精确分子理论中，并将其应用于复杂的化学和生物分子问题，包括：(I)功能化纳米颗粒在液体界面的化学反应，(Ii)药物跨生物膜和血脑屏障的传输，以及(Iii)纳米颗粒跨生物膜向不同尺寸和形状的纳米颗粒的纳米毒性数据库的转移的预测，以及表面功能化。新的方法学发展将被添加到教育部(加拿大)和ADF(荷兰)软件包的现有实施中。三名博士生将参与并接受该项目的培训。

It is widely recognized that solvation is critical for a broad range of science and technology. In particular, liquid interfaces are of paramount importance in a variety of technological applications of chemistry, electrochemistry, biocatalysis, biology, and pharmacology. At liquid interfaces, aqueous and organic solvents meet and create an environment with unique properties that can be fine-tuned to promote reactivity and self-assembly. Soft matter nanostructures are complex liquid interfaces formed by competing chemical functionalities. An important example is a lipid bilayer of biomembranes in living cell and intracellular structures. Modeling structure and functions of liquid interfaces poses a challenge, which has been attempted at separate levels of complexity, from drug transport across a biomembrane estimated as partitioning between bulk liquid phases of water and octanol, to coarse-grained simulations of nanoparticle translocation across a lipid bilayer. At present, there is a need of theoretical models of solvation that adequately include molecular structure of solvent, in particular, in local confinement of chemical and biomolecular nanosystems and nanomaterials. Such theoretical models are of great demand to extend the capabilities of molecular simulations which are by far not sufficient to address complex phenomena and processes on the whole spectrum of scales from fast to very slow in large nanosystems of interest. Theoretical models of solvation based on statistical mechanics can be an essential element coupled to all parts of predictive multiscale modeling including quantum description for chemical structure and active centers, molecular dynamics (MD) and dissipative particle dynamics (DPD) for nanostructures, and coarse-grain models for system functions. Integral equation theory of liquids based on the first principles of statistical mechanics is becoming increasingly popular, as it provides a firm platform to handle complex chemical and biomolecular systems in solution. In particular, the 3D-RISM-KH molecular theory of solvation (three-dimensional reference interaction site model with the Kovalenko-Hirata closure relation) has shown substantial success for a number of systems in solution that otherwise were not amenable to molecular simulation or continuum solvation treatment. It has been advanced to the state of practical applicability and coupled in multiscale methodology with quantum chemistry, MD and DPD simulation methods, and ligand docking protocols, the developments implemented in major software packages: Amsterdam Density Functional (ADF) computational chemistry package, Amber MD package, AutoDock suite of docking tools [9], and Molecular Operating Environment (MOE) package. Furthermore, the site-site generalization of the Lovett-Mou-Buff-Wertheim (SS-LMBW) integro-differential equation constitutes a molecular theory of liquid interfaces which has been validated on planar interfaces of non-polar and complex associating liquids. This project will couple the 3D-RISM-KH and SS-LMBW approaches in a new accurate molecular theory of liquid nd soft matter interfaces, and will apply it to complex chemical and biomolecular problems, including: (i) chemical reactivity of functionalized nanoparticles at liquid interfaces, (ii) drug transport across biomembranes and the blood-brain barrier, and (iii) prediction of translocation of nanoparticles across biomembranes towards nanotoxicity database for nanoparticles of different size and shape, and surface functionalization. The new methodological developments will be added to the existing implementations in the MOE (Canada) and ADF (Netherlands) software packages. Three PhD students will be involved and trained in this project.