2016/09/21 by Pedro E. M. Lopes, Lopes, Pedro E. M.
Engineering · Physics and Astronomy · #Advanced Chemical Physics Studies #Chemical Physics (physics.chem-ph) #FOS: Physical sciences #Phase Equilibria and Thermodynamics #Quantum, superfluid, helium dynamics
paper · pdf · doi:10.48550/arxiv.1609.06633
openalex publication_date 2016/09/21 · openalex created_date 2022/08/15 · openalex updated_date 2026/07/28
An alternative methodology to evaluate two-electron-repulsion integrals based\non numerical approximation is proposed. Computational chemistry has branched\ninto two major fields with methodologies based on quantum mechanics and\nclassical force fields. However, there are significant shadowy areas not\ncovered by any of the available methods. Many relevant systems are often too\nbig for traditional quantum chemical methods while being chemically too complex\nfor classical force fields. Examples include systems in nanomedicine, studies\nof metalloproteins, etc. There is an urgent need to develop fast quantum\nchemical methods able to study large and complex systems. This work is a\nproof-of-concept on the numerical techniques required to develop accurate and\ncomputationally efficient algorithms for the fast calculation of\nelectron-repulsion integrals, one of the most significant bottlenecks in the\nextension of quantum chemistry to large systems. All concepts and calculations\nwere developed for the three-center integral (pxApxB|pxCpxC) with all atoms\nbeing carbon. Starting with the analytical formulae, convenient decompositions\nwere tested to provide smooth two-dimensional surfaces that were easily fitted.\nThe approximating algorithm consisted of a multilayered approach based on\nmultiple fittings of two-dimensional surfaces. An important aspect of the new\nmethod is its independence on the number of contracted Gaussian primitives. The\nbasis set of choice was STO-6G. In future developments, larger basis sets will\nbe developed. This work is part of a large effort aimed at improving the\ninadequacies of existing computational chemistry methods, both based on quantum\nmechanics and classical force fields, in particular in describing large and\nheterogeneous systems (ex. metalloproteins).\n