Multiscale modelling quantum mechanical-molecular

Zhang X, Zhao Y, Lu G. Recent development in quantum mechanics/molecular mechanics modeling for materials. Int. J. Multiscale Comput. Eng. 2012 10 65-82. 

Today, it is indispensable and conunon in modem chemistry to deal with molecules as the quantum systems that consist of a couple of classical mechanical (CM) nuclei and quantum mechanical (QM) electrons, for understanding chemical phenomena deeply. Such QM approaches can provide us the microscopic information such as the stmctural information (e.g. stable state (SS) and transition state (TS)) and chemical properties (e.g. electric or magnetic external fields and internal perturbations such as a nuclear or electron spin) of chemical reaction systems. However, from the point of view of computational efforts, it remains difficult to directly apply the QM approaches to large reaction systems such as the solution (or biological) ones that we are interested in. Thus, to treat these whole reaction systems in solution and biological environment, it is very useful in many cases to employ a multiscale model such as the quantum mechanical/molecular mechanical (QM/MM) methods, which are often combined with molecular dynamics (MD) or Monte Carlo (MC). 

Existing quantum mechanics and molecular dynamics methods have not yet advanced to the stage where water can be reliably modeled by itself [148-151], much less when involved in electrochemical reactions at a surface [152]. An important requirement ofany general multiscale systems framework is that it must be able to enable the resolution of the unknowns in complex heterogeneous mechanisms. 

We have recently started to explore a type of calculations in which DFT treatment of the quantum mechanical (QM) site is combined with either continuum electrostatics treatment of the protein, or with microscopic molecular mechanics/dynamics treatment of the protein, or with a combined molecular mechanics and continuum electrostatics treatment of the protein in a truly multiscale type of calculations. All these calculations have a spirit of QM/MM (quantum mechanics combined with molecular mechanics) method, which is currently in wide use in protein calculations. The DFT and the solvation energy calculations are performed in a self-consistent way. The work aims at both improving the QM part of p/ calculations and the MM or electrostatic part, in which of the protein dielectric properties are involved. In these studies, an efficient procedure has been developed for incorporating inhomogeneous dielectric models of the proteins into self-consistent DFT calculations, in which the polarization field of the protein is efficiently represented in the region of the QM system by using spherical harmonics and singular value decomposition techniques [41,42]. 

FIGURE 1.5 Multiscale modeling in computational pharmaceutical solid-state chemistry. Here DEM and FEM are discrete and finite element methods MC, Monte Carlo simulation MD, molecular dynamics MM, molecular mechanics QM, quantum mechanics, respectively statistical approaches include knowledge-based models based on database analysis (e.g., Cambridge Structure Database [32]) and quantitative structure property relationships (e.g., group contributions models [33a]). 

For the case of batteries, the majority of the reported multiscale models focuses on the understanding of the operation and the impact of the stmctural properties of LiFeP04 or graphite electrodes onto the global cell efficiency. And in the other hand, quantum mechanics and molecular dynamics models focalize on the understanding of the impact of the materials chemistry onto their storage or lithium transport properties at the nanoscale. It is now crucial to develop multiscale models that are able to incorporate both stracture and chemical databases, in other words, that they are able to mimic the materials behavior in realistic electrochemical environments. Within this sense, further intercalation and conversion... 

The hybrid Quanmm Mechanical/Molecular Mechanical (QM/MM) approach is one of multiscale models for complex chemical reactions in solution and in proteins. In the QM/MM method, it is common that only the reactive parts in the whole solution reaction system are treated quantum-mechanicaUy, while the other parts are molecular-mechanically [12-16]. The effective Hamiltonian H of the whole system consists of three terms ... 

Besides the classical techniques for structural determination of proteins, namely X-ray diffraction or nuclear magnetic resonance, molecular modelling has become a complementary approach, providing refined structural details [4—7]. This view on the atomic scale paves the way to a comprehensive smdy of the correlations between protein structure and function, but a realistic description relies strongly on the performance of the theoretical tools. Nowadays, a full size protein is treated by force fields models [7-10], and smaller motifs, such as an active site of an enzyme, by multiscale approaches involving both quantum chemistry methods for local description, and molecular mechanics for its environment [11]. However, none of these methods are ab initio force fields require a parameterisation based on experimental data of model systems DPT quantum methods need to be assessed by comparison against high level ab initio calculations on small systems. 

---