[QE-users] "a" "b" values in "xanes.save" show "NaN" in xspectra calculation

[田飞飞]

Dear QE user,

I am doing a O K-edge XAS calculation on Pt/O surface system using xspectra in QE 6.7.
I need a half-core-hole pseudopotential (*starh1s*) for Oxygen. But I haven't found required pp on any pp library recommended in https://www.quantum-espresso.org/pseudopotentials.
So I tried generating my own pp (O.starh1s-pbe-n-rrkjus_gipaw.UPF) using ld1.x according to the tutorial in http://ewins2016.ijs.si/slides/tutorial-XSpectra.pdf . Then I calculated the XAS successfully using this half-core-hole pp. (see dir "mypp_starh1s")
After that I wanted to do a full-core-hole calculation. So I generated my full-core-hole pp (O.star1s-pbe-n-rrkjus_gipaw.UPF) using the same method, But after the scf completed successfully, the xspectra calculation failed. The "a" "b" coefficients in "xanes.save" show "NaN" after 4 or 5 steps of lanczos recursion.  (see dir "mypp_star1s")
I also found a full-core-hole pp (O.star1s-pbe-van_gipaw.UPF) in https://www.quantum-espresso.org/upf_files/O.star1s-pbe-van_gipaw.UPF . And the computation using this pp completed successfully.  (see dir "qepp_star1s")
So I don't know what's wrong when I generated my own pp using ld1.x?
Or is there any library providing half-core-hole pseudopotentials?
All the related files have been uploaded for reference. I would appreciate for any suggestions.
(see https://drive.google.com/file/d/1MAQ469i-FGW2lSy6x92XnqA3z72jxWWE/view?usp=sharing )

#=====================================>>
# ld1.x input file (in "mypp_star1s/O.star1s-pbe-n-rrkjus_gipaw.UPF")
#-----------------------------------------------------
&input
title='O.star1s',
zed=8.,
rel=1,
config='1s1 2s2 2p4 3s0 3p0',
iswitch=3,
dft='PBE'
/
&inputp
!   lpaw = .true.,
pseudotype=3,
file_pseudopw='O.star1s-pbe-n-rrkjus_gipaw.UPF',
author='Feifei, Tian',
lloc=-1,
rcloc=1.1
which_augfun='PSQ',
rmatch_augfun_nc=.true.,
nlcc=.true.,
new_core_ps=.true.,
rcore=0.7,
tm=.true.
lgipaw_reconstruction=.true.
/
4
2S  1  0  2.00  0.00  1.00  1.30  0.0
2S  1  0  0.00  1.00  1.00  1.30  0.0
2P  2  1  4.00  0.00  0.90  1.35  0.0
2P  2  1  0.00  0.05  0.90  1.35  0.0
/
&test
/
4
2S  1  0  2.00  0.00  1.00  1.30  0.0
3S  2  0  0.00  1.00  1.00  1.30  0.0
2P  2  1  4.00  0.00  0.90  1.35  0.0
3P  3  1  0.00  0.05  0.90  1.35  0.0
#----------------------------------------------------------------------<<

#===================================>>
# xanes a & b values (in "mypp_star1s/xspectra.epsilon010/xanes.save")
#-----------------------------------------------------
...
  1.73822111871875        8.80947137216540        27.8977091531785
   15.7058216460591        37.8310916947681                          NaN
                     NaN                     NaN                     NaN
                     NaN                     NaN                     NaN
                     NaN                     NaN                     NaN
...
#----------------------------------------------------------------------<<


Feifei Tian
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20210623/a25490fd/attachment.html>