Molecular tailoring: An art of the possible for ab initio treatment of large molecules and molecular clusters

Abstract

Divide-and-conquer (DC) type methods are being actively developed in order to break the bottleneck of high scaling order of ab initio calculations of large molecules. Molecular Tailoring Approach (MTA) is one of such early attempts, which scissors the parent molecular system into subsystems (fragments). The properties of these subsystems are stitched back in order to estimate those for the parent system. Inclusion-exclusion principle from set theory is incorporated into MTA, which allows accurate estimation of electronic energy, energy-gradients and Hessian. This Chapter summarizes the algorithm, equations as well as basic parameters for obtaining an optimal fragmentation for a given molecule. The fragmentation in MTA is exclusively based on distance-criterion allowing its application to a general class of molecules. Further, the versatility of this method with respect to the level of theory [Hartree-Fock (HF) method, Møller-Plesset second order perturbation theory (MP2) and Density Functional Theory (DFT)] as well as the basis set is illustrated. Apart from earlier benchmarks, a few new test cases including geometry optimization of variety of molecules, benzene clusters, polyaromatic hydrocarbons, metal cluster and a protein with charged centers are presented in this Chapter.