Efficient Implementation of the Approximate Exchange Kernel Method

ORAL

Abstract

The approximate exchange kernel (AXK) method is the leading correction to the random-phase approximation (RPA) for the ground-state correlation energy functional within RPA-renormalized many-body perturbation theory. AXK alleviates the unphysical short-range behavior of RPA and improves the prediction of energetics for processes that do not conserve the number of electron pairs. However, the relatively high computational cost of AXK has hampered its widespread application in the past. Here we present a fast AXK algorithm with O(N4lnN) scaling. The algorithm takes advantage of the resolution-of-the-identity (RI) approximation and imaginary frequency integration technique. The new RI-AXK implementation in TURBOMOLE enables correlation energy calculations for molecules with well over 100 atoms on a single cluster node.

*This work is supported by the National Science Foundation under CHE-1464828.

–

Presenters

  • Guo Chen

    • Chemistry Department, University of California

Authors

  • Guo Chen

    • Chemistry Department, University of California

  • Matthew Agee

    • Chemistry Department, University of California

  • Filipp Furche

    • Chemistry, University of California, Irvine
    • Chemistry Department, University of California
    • University of California Irvine