next up previous contents index
Next: PSCTR Up: The pseudopotential generation package Previous: The pseudopotential generation package   Contents   Index

N.B. This document is no longer maintained, please visit our wiki.


The AE-part of the program rhfsps is controlled by the V_RHFIN file. This file is strictly formatted, and you must be very careful, if you change the file. Typically the file might have the following contents:

Pd : s1 d9,  CA
   11  46.   .002000 106.42000 125.   .50E-05 .100  200FCA 36.00000
   .7  1.0   0
  1.0   .0   .5-1761.5171 2.0000
  2.0   .0   .5 -257.9015 2.0000
  2.0  1.0  1.5 -231.7505 6.0000
  3.0   .0   .5  -46.6977 2.0000
  3.0  1.0  1.5  -38.0485 6.0000
  3.0  2.0  2.5  -24.196610.0000
  4.0   .0   .5   -6.4877 2.0000
  4.0  1.0  1.5   -3.9976 6.0000
  5.0   .0   .5    -.3403 1.0000
  4.0  2.0  2.5    -.5091 9.0000
  5.0  1.0   .5    -.1000  .0000
The first line is a comment, which should contain the name of the element and the reference configuration for the valence electrons. The second line
   11  46.   .002000 106.42000 125.   .50E-05 .100  200FCA 36.00000
    J   Z   XION   N  AM         H    DELRVR   PHI  NC1 CH QCOR
gives the most important information about the atom. J is the number of orbitals, Z the ordering number. XION can be used to supply a degree of ionization, but normally this value is zero. N is the number of grid points, usually we use 2000, AM the atomic mass, which is used to calculate the innermost point for the logarithmic grid. H determines the spacing between the grid points. The grid points are given by

$\displaystyle r = r_{\rm small} {\rm e^\frac{ number}{H}}.$ (11.1)

We normally use H=125. DELRVR is the break condition for the selfconsistency loop and PHI the linear mixing parameter for the charge density. NC1 determines the maximum number of selfconsistency loops. If a V_TABIN file exists GREEN should be FALSE (F), if no V_TABIN exists set GREEN to T; in this case an appropriated start potential will be calculated. The parameter CH determines the type of the exchange correlation, the following settings are possible:

HL Hedin Lundquist (1971)
CA Ceperly and Alder parameterized by
  J.Perdew and Zunger
WI Wigner interpolation
PB Perdew -Becke
PW Perdew -Wang 86
LM Langreth-Mehl-Hu
91 Perdew -Wang 91

Among these, the last four are gradient corrected functionals. The parameter QCOR determines the number of core electrons (i.e. non valence electrons). The next line in the V_RHFIN file supplies less important information. The first parameter is the SLATER parameter used only in conjunction with the Slater-XC. The next parameter is no longer used, and the last one can be used the set up so called latter correction to the exchange correlation potential. Latter corrections must not be applied if pseudopotentials are calculated. The remaining J lines give information about each atomic orbital. The code is scalar relativistic, but the inputfile is compatible to a relativistic input format. The first value in each line is the main quantum number, the second one the l-quantum number, and the third one the j-quantum number ( $ j=l\pm 1/2$). The j-quantum number is not used in the program. The next value gives the energy of the atomic orbital, the last number is the occupancy of the orbital. The supplied energy is uncritical and only used as a start value for the calculation of the atomic orbitals. As a starting guess you might insert values obtained from an atom lying close to the atom of interest.

The program rhfsps writes two files V_RHFOUT and V_TABOUT. The V_RHFOUT file is compatible to V_RHFIN and can be copied to V_RHFIN, if V_TABOUT is copied to V_TABIN. In this case rhfsps will start from the fully converged AE-potential supplied in V_TABIN. This saves time, and generally we recommend this setting.

N.B. Requests for support are to be addressed to: