Next: Primitive basis functions
Up: Basis sets
Previous: Basis sets
  Contents
  Index
OpenMX uses numerical pseudo-atomic orbitals (PAOs) as basis function to expand one-particle
Kohn-Sham wave functions. The PAO function is given by a product of a radial function and
a real spherical harmonic function as
where the radial function is a numerically defined one, and finite within a cutoff radius
in real space. In other words, the function becomes zero beyond the cutoff radius.
The PAO function calculated by ADPACK is called primitive function, and an optimized PAO
function is obtained by the orbital optimization method in OpenMX starting from
the primitive PAO function [26]. They are stored in a file with a file extension
of pao. When the OpenMX calculation is performed, the numerical data stored in the
file are read, and the value at any is obtained by an interpolation technique.
The files with the file extension of pao should be
stored in a directory, e.g., DFT_DATA11/PAO, where the directory without PAO can be
specified by the following keyword:
DATA.PATH ../DFT_DATA11 # default=../DFT_DATA11
Both the absolute and relative specifications are available, and the default is ../DFT_DATA11.
In an input file for the OpenMX calculation,
The basis set is specified by a keyword
'Definition.of.Atomic.Species'
as follows:
<Definition.of.Atomic.Species
H H5.0-s2p1 H_PBE11
C C5.0-s2p1 C_PBE11
Definition.of.Atomic.Species>
where an abbreviation, H5.0-s2p1, of the basis function is introduced.
H5.0 stands for the file name of the PAO functions without the file extension
which must exist in a directory specified by the keyword, DATA.PATH,
e.g., DFT_DATA11/PAO, and 5.0 implies the cutoff radius of the PAO functions.
Also, s2p1 means that two s-state radial functions and one p-state radial function
stored in the file are used.
In this case, totally five PAO basis functions (2x1+1x3=5) are assigned for 'H'.
Since optimized basis functions are available on the web site (http://www.openmx-square.org/)
as database Ver. 2011. We recommend for general users to use these optimized basis functions.
But for experts, both the primitive and optimized PAO functions are explained
in the subsequent sections.
Next: Primitive basis functions
Up: Basis sets
Previous: Basis sets
  Contents
  Index
2011-11-10