Next: Principles Up: Preconditioned iterative minimization Previous: Introduction

# Formulation of the problem

A system of noninteracting particles in a potential is described by

 (1)

where is the single-particle Hamiltonian of the system, with energy eigenvalues and corresponding eigenstates . The eigenstates satisfy the orthogonality constraints given by

 (2)

for all and . For instance, within the Kohn-Sham scheme of density-functional theory [22,23,24], is the Kohn-Sham Hamiltonian and is the effective potential.

The total band-structure energy is given by

 (3)

where is the occupancy of state  [25]: at the energy minimum, all states below and above the Fermi-level have occupancy unity and zero, respectively.

In the case of linear-scaling calculations, the lowest extended eigenstates ( ) are expressed in terms of a set of localized functions ( ) that are generally nonorthogonal:

 (4)

where is a square ( by ), nonsingular matrix of coefficients, and can be equal to or greater than the number of occupied eigenstates. The overlap matrix of the localized functions is

 (5)

and on substitution of Eq. (4) into the orthogonality relation given by Eq. (2) we find that satisfies

 (6)

where a distinction has been made between contravariant and covariant quantities [26,27] through the use of superscript and subscript Greek suffixes, respectively.

Substituting Eq. (4) into the energy expression of Eq. (3), and defining

 (7) (8)

the band-structure energy becomes

 (9)

where is referred to as the density kernel [28].

We consider the localized functions to be represented in terms of a basis as follows:

 (10)

for some coefficients . As the basis functions may be in general nonorthogonal, the tensor properties must be taken into account through the use of superscript and subscript Greek suffixes.

Defining

 (11)

and using Eqs. (7)-(10), the energy may be written as

 (12)

Suffixes and run over the localized functions , and run over the basis functions and runs over the extended orthogonal orbitals . We have adopted the Einstein summation convention for all repeated Greek suffixes, and continue to do so from here on.

It is both convenient and physically meaningful to perform the minimization of the energy functional in two nested loops, as in the ensemble density-functional method of Marzari et al. [29]: in the inner loop we minimize the energy with respect to the elements of the density kernel using one of a number of methods [30,31,32] to impose the constraint that the ground state density matrix be idempotent and give the correct number of electrons; in the outer loop we optimize the localized functions with respect to their coefficients in the basis  [16].

Next: Principles Up: Preconditioned iterative minimization Previous: Introduction
Arash Mostofi 2003-10-28