[Pw_forum] About the upf2plotcore.sh in xspectra calculation

[huangxu1706 at sina.com]

Hi,
I have problem generating atomic-like wave functions from ultrasoft pseudopotential (USPP) using the upf2plotcore.sh script in QE-5.1 XSpectra package. I went over all of the examples of diamond, SiO2 and NiO xspectra calculations. Now I want try it in Fe2O3 hematite system using DFT-GGA+U method. The USPP I chose to use for Fe is called "Fe.pbe-nd-rrkjus.UPF". I down load it from QE website.
However, when I use the same command shown in the example script to generate atomic-like wave functions to get Fe.wfc:....../espresso-5.1/XSpectra/tools/upf2plotcore.sh ~/my_pseudo_dir/Fe.pbe-nd-rrkjus.UPF > ./Fe.wfc
I got an empty Fe.wfc file, which only contains one line:# number of core states 0 =
I didn't get this problem when I was working on those examples of diamond, SiO2 and NiO using provided pseudopotentials in the package like:C_PBE_TM_2pj.UPFSi_PBE_USPP.UPFNi_PBE_TM_2pj.UPFThey are in ....../espresso-5.1/XSpectra/examples/pseudo. They could generate proper C.wfc, Si.wfc and Ni.wfc for following xspectra calculations.
So I'm wondering if there is any specific "requirement" for the choose of USPP in xspectra calculations. There are many different Fe-USPP in QE website. Do you know which of them can be used to generate Fe.wfc?
Thank you,Xu Huang

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150122/c564693b/attachment.html>