Julian D. Gale,a
Emilio Artacho,b Alberto Garca,c
Javier Junquera,d Pablo Ordejn,e
Daniel Snchez-Portalf and Jos M.
Solerg
aDepartment of
Chemistry, Imperial College London, South Kensington, SW7 2AZ, U.K.; bDepartment
of Earth Sciences, University of Cambridge, Downing Street, Cambridge, CB2 3EQ,
U.K.; cDepartamento de Fsica de la Materia Condensada,
Universidad del Pas Vasco, Apt. 644, 48080 Bilbao, Spain; dInstitut
de Physique, Btiment B5, Universit de Lige, B-4000 Sart-Tilman, Belgium; eInstitut
de Cincia de Materials de Barcelona, CSIC, Campus de la UAB, Bellaterra, 08193
Barcelona, Spain; fDep. de Fsica de Materiales and
DIPC, Facultad de Qumica, UPV/EHU, Apt. 1072, 20080 Donostia, Spain; gDep.
de Fsica de la Materia Condensada, C-III, Universidad Autnoma de Madrid,
E-28049 Madrid, Spain (j.gale@imperial.ac.uk).
In order to
achieve the goal of being able to perform ab initio
electronic structure calculations on very large molecular and condensed matter
systems it is necessary to address the issue of the calculation scaling with
increasing system size. Here the details of the SIESTA method [1,2] will be
presented, which makes it possible to perform calculations that scale linearly
with increasing numbers of atoms, both in computational expense and memory
usage. This is achieved through the use of radially confined basis functions
that lead to sparse matrices, in combination with an auxillary basis of a real
space mesh for the evaluation of the Hartree and exchange-correlation
potentials. In addition, linear scaling requires the use of functional
minimization approaches to solve for self-consistency, though conventional
matrix diagonalisation is also available. Through the use of parallel
computing, and the above methodology, it is now quite feasible to perform ab initio calculations on
thousands of atoms.
Many standard observables
of electronic structure theory may be obtained including electron and spin
densities, optimized structural configurations, phonons, as well as dynamical
information. Examples will be given of the application of the technique, as
well as possible directions for future advancement.
1
Artacho, E., Snchez-Portal, D., Ordejn, P., Garca,
A., and Soler, J.M. (1999) Phys. Stat. Sol. B 215,
809-817.
2
Soler, J.M., Artacho, E., Gale, J.D., Garca, A.,
Junquera, J., Ordejn, P. and Snchez-Portal, D. (2002) J. Phys.: Condens.
Matter 14, 2745-2779.