Linearly scaling computation of ddPCM solvation energy and forces using the fast multipole method

20 June 2022, Version 1
This content is a preprint and has not undergone peer review at the time of posting.

Abstract

This paper proposes the first linear scaling implementation for the domain decomposition approach of the polarizable continuum model (ddPCM) for the computation of the solvation energy and forces. The ddPCM-equation consists of a (non-local) integral equation on the van der Waals (vdW) or solvent accessible surface (SAS) of the solute’s cavity resulting in a dense solution matrix and, in turn, one matrix-vector multiplication has a quadratic arithmetic complexity with respect to the number of atoms of the solute molecule. The use of spherical harmonics as basis functions makes it natural to employ the fast multipole method (FMM) in order to provide an asymptotically linear scaling method. In the present paper, we employ the FMM in a non-uniform manner with a clusterization based on a recursive inertial bisection. We present some numerical tests illustrating the accuracy and scaling of our implementation.

Keywords

polarizable continuum model
PCM
domain decomposition
ddPCM
multiscale
fast multipole method
FMM
linear scaling

Supplementary weblinks

Comments

Comments are not moderated before they are posted, but they can be removed by the site moderators if they are found to be in contravention of our Commenting Policy [opens in a new tab] - please read this policy before you post. Comments should be used for scholarly discussion of the content in question. You can find more information about how to use the commenting feature here [opens in a new tab] .
This site is protected by reCAPTCHA and the Google Privacy Policy [opens in a new tab] and Terms of Service [opens in a new tab] apply.