Harmonic surface mapping algorithm for molecular dynamics simulations of particle systems with planar dielectric interfaces

We have developed an accurate and efficient method for molecular dynamics simulations of charged particles confined by planar dielectric interfaces. The algorithm combines the image-charge method for near field with the harmonic surface mapping, which converts the contribution of infinite far-field...

Full description

Saved in:
Bibliographic Details
Published inThe Journal of chemical physics Vol. 152; no. 13; pp. 134109 - 134117
Main Authors Liang, Jiuyang, Yuan, Jiaxing, Luijten, Erik, Xu, Zhenli
Format Journal Article
LanguageEnglish
Published United States American Institute of Physics 07.04.2020
Subjects
Online AccessGet full text
ISSN0021-9606
1089-7690
1089-7690
DOI10.1063/5.0003293

Cover

More Information
Summary:We have developed an accurate and efficient method for molecular dynamics simulations of charged particles confined by planar dielectric interfaces. The algorithm combines the image-charge method for near field with the harmonic surface mapping, which converts the contribution of infinite far-field charges into a finite number of charges on an auxiliary spherical surface. We approximate the electrostatic potential of far-field charges via spherical harmonic expansion and determine the coefficients by fitting the Dirichlet-to-Neumann boundary condition, which only requires the potential within the simulation cell. Instead of performing the direct evaluation of spherical harmonic series expansion, we use Green’s second identity to transform the series expansion into a spherical integral, which can be accurately represented by discrete charges on the sphere. Therefore, the fast multipole method can be readily employed to sum over all charges within and on the sphere, achieving truly linear O(N) complexity. Our algorithm can be applied to a broad range of charged complex fluids under dielectric confinement.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0021-9606
1089-7690
1089-7690
DOI:10.1063/5.0003293