votca 2026-dev
Loading...
Searching...
No Matches
votca::xtp::AOGaussianSites Class Reference

Coupling of the QM region with Gaussian charges and dipoles (the induced charges and dipoles of charge-response sites). More...

#include <aogaussiansites.h>

Classes

struct  Source

Static Public Member Functions

static Eigen::MatrixXd Potential (const AOBasis &basis, const std::vector< Source > &sources)
static Eigen::MatrixXd PotentialAndGradientAt (const AOBasis &basis, const Eigen::MatrixXd &D, const std::vector< Eigen::Vector3d > &points, const std::vector< double > &widths)

Detailed Description

Coupling of the QM region with Gaussian charges and dipoles (the induced charges and dipoles of charge-response sites).

A site k of width s carries the charge density q g_k(r) + mu . (2 beta) (r - R_k) g_k(r), g_k(r) = (beta/pi)^3/2 exp(-beta |r - R_k|^2), beta = 1/s^2, i.e. a normalized Gaussian charge and the Gaussian dipole -mu . grad g_k (dipole moment mu). The integrals are three-centre Coulomb integrals of an AO pair with an s and a p primitive at R_k (libint2), so the QM->site potential and the site->QM Hamiltonian use the same integrals and the coupling is exactly symmetric. Width 0 is not allowed (use AOMultipole for points).

Definition at line 48 of file aogaussiansites.h.

Member Function Documentation

◆ Potential()

Eigen::MatrixXd votca::xtp::AOGaussianSites::Potential ( const AOBasis & basis,
const std::vector< Source > & sources )
static

<mu| phi |nu> with phi the electrostatic potential of the sources (positive next to a positive charge); the electrons' Hamiltonian gets minus this matrix.

Definition at line 743 of file libint2_calls.cc.

◆ PotentialAndGradientAt()

Eigen::MatrixXd votca::xtp::AOGaussianSites::PotentialAndGradientAt ( const AOBasis & basis,
const Eigen::MatrixXd & D,
const std::vector< Eigen::Vector3d > & points,
const std::vector< double > & widths )
static

For the charge density rho(r) = sum_mn D_mn chi_m(r) chi_n(r) (no sign: electrons are -D): per point k (rows) the potential of rho seen by a Gaussian of width widths[k] at points[k] (column 0) and its gradient with respect to points[k] (columns 1-3, = minus the field).

Definition at line 773 of file libint2_calls.cc.

The documentation for this class was generated from the following files: