[Pw_forum] Potassium Ultrasoft pseudopotential with PBE

[Guido Fratesi]

Dear LiBin,

please notice that list rules ask posts to be signed with affiliation.

> Recently, my calculation needs Potassium PBE Ultrasoft pseudopotential. But I’m not sure about my pseudo generated by ld1.x. So is there existing K.pbe pseudo? Thank you very much!

I have a PBE K pseudo which I used to compute K adsorption properties on 
Cu with good agreement with experimental workfunction changes and 
diffusion energy barriers. That was generated with the Vanderbilt code, 
then converted, starting from a CA-LDA pseudo distributed with 
Vanderbilt's USPP code. In more detail, I have:

1)  taken input files for CA-LDA in USPP-7.3.5 distribution:
   uspp-735/Work/019-K/019-K-ca-sp-vgrp/k_ae_s1.adat
   uspp-735/Work/019-K/019-K-ca-sp-vgrp/k_ps.adat

2) changed:
   exfact 0 -> 5 (PBE)
   ifqopt 2 -> 3 (different augfun pseudiz; 2 not allowed for GGA)
   rinner 0.8 -> 0.95 (Determines the inner radii that are used
                                 for pseudizing the q-functions)

3) run the code
  $ make

4) converted to UPF
    $ echo k_ps.uspp | upftools/uspp2upf.x

I'm sending you the generated file in a separate email to not clutter 
the list. Please report whether that matches your needs, and especially 
if not, why.

HTH, Guido

-- 
Guido Fratesi

Dipartimento di Scienza dei Materiali
Universita` degli Studi di Milano-Bicocca
via Cozzi 53, 20125 Milano, Italy