Study and development of new metal-organic frameworks based membranes for separation of multicomponent gas mixtures

该项目的重点是开发用于多组分气体混合物纯化/分离的新型金属有机框架（MOFs）材料。我们进行计算建模，以设计新的MOFs膜的工业应用的重要性，包括沼气分离和氢气吸附。我们开发了原子尺度的计算模型，直接连接到现实条件下的气体吸附，并允许同时解决其基本方面。这些模型是在多层次工作流程的框架下实现的，结合了量子和经典方法，这为研究MOF在宽压力范围内的吸附行为提供了机会。虽然经典力场覆盖常压区域，但量子水平方法用于研究低压下的吸附过程，经典方法往往失败得很惨。我们非常关注量子化学方法的实现，因为众所周知，由于典型的大尺寸晶胞和过渡金属d-壳层的多参考性质，MOFs对于量子理论考虑来说是非常具有挑战性的对象。这反映在这样的事实中，即常规使用的量子方法在应用于M0 F时遭受收敛问题和低计算效率。因此，我们计划开发一种基于群函数概念的混合量子化学模型，允许将系统有效地分离为多参考效应必不可少的部分和环境，允许单参考甚至力场处理。提出的方法将提供一个有效的和强大的工具建模的MOFs可能的应用超越吸附性能。

This project is focused on development of new metal-organic frameworks (MOFs) based materials for purification/separation of multicomponent gas mixtures. We perform computational modelling in order to design new MOFs membranes for applications of industrial importance including biogas separation and hydrogen adsorption. We develop atomic-scale computational models directly connected to the sorption of gases under realistic conditions and allowing to address its fundamental aspects at the same time. These models are realized in the frame of multi-level workflow, combining both quantum and classical approaches, which provides an opportunity to investigate sorption behaviour of MOFs at a wide range of pressures. While classical force fields cover the normal-pressure region, quantum-level methods are used to study sorption processes under low-pressures, where classical approaches tend to fail quite miserably. We pay significant attention to implementation of the quantum chemical approaches, since MOFs are well-known to be very challenging objects for quantum theory consideration due to the typical big sizes of unit-cell combined with multireference nature of transition metals d-shell. This is reflected in the fact, that routinely used quantum approaches suffer from convergence problems and low computational effectiveness, when applied to MOFs. Therefore, we plan to develop a hybrid quantum chemical model, based on group function concept, allowing effective separation of the system into part, where multireference effects are essential, and environment, allowing single-reference or even force field treatment. Propose methodology will provide an effective and powerful tool for modeling of MOFs with possible applications going beyond sorption properties.