[Pw_forum] How to generate old-version UPF files

[Chang Chen]

Dear Dr. Fortmann,
    Thank you for your reply. There is another question. For example, there is an input file "Gd.rel-pbe-spdn-rrkjus_psl.1.0.0.in" in pslibrary.1.0.0.tar.gz. The keyword "rmatch_augfun_nc=.ture." is not recognized by atomic-4.2.1. I change "rmatch_augfun_nc=.ture." to "rmatch_augfun=1.3". The .UPF files generated by atomic-5.0.2 based on the above two keywords are exactly the same. After the test, I add "upf_v1_format=.ture." and generate the .UPF file by atomic-4.2.1.
    When I use virtual.x in upftools of espresso-4.2.1 to mix the generated pseudopotentials, there is a error message:
  IOS=            0           1           4
  Reading pseudopotential file in UPF format...
 ikk2         954         954         954         954         954         954
         959         959         959         959           0           0
           0           0           0           0

 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
     Error in routine read_pseudo_nl (1):
     inconsistent angular momentum for Q_ij
 %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

The .in file is shown below. I do not know why the generated .UPF is wrong. Is there anything wrong in my .in file? Thanks for your time.
Regards,
Chang Chen


.in file:

 &input
   title='Gd',
   zed=64.0,
   rel=2,
   config='[Xe] 4f7.0 5d1.0 6s1.5 6p0.5',
   iswitch=3,
   dft='PBE'
 /
 &inputp
   lpaw=.false.,
   pseudotype=3,
   file_pseudopw='Gd.rel-pbe-spdn-rrkjus_psl.1.0.0.UPF',
   author='ADC',
   lloc=-1,
   rcloc=2.0,
   which_augfun='PSQ',
   rmatch_augfun=1.3,
   upf_v1_format=.ture.,
   nlcc=.true.,
   new_core_ps=.true.,
   rcore=1.1,
   tm=.true.
 /
6
5S  1  0  2.00  0.00  1.30  1.70  0.0
6S  2  0  1.50  0.00  1.30  1.70  0.0
5P  2  1  6.00  0.00  1.30  1.80  0.0
6P  3  1  0.50  0.00  1.30  1.80  0.0
5D  3  2  1.00  0.00  1.30  2.00  0.0
5D  3  2  0.00  2.00  1.30  2.00  0.0







At 2015-01-22 21:31:58, "Carsten Fortmann" <carsten.fortmann at quantumwise.com> wrote:
>you can use
>upf_v1_format=.true.
>in an input for for atomic
>
>here's a link to the documentation of all input variables :
>
>ftp://ftp.ictp.it/shared/qetutorial/espresso-4.2.1/Doc/INPUT_LD1.html
>
>good luck.
>
>
>On 01/22/2015 02:25 PM, Chang Chen wrote:
>> Dear Doctors,
>>     I want to calculate some doping effect based on virtual crystal approximation. The program virtual.x in espresso-5.0.2/upftools can mix two pseudopotentials. However, it seems that virtual.x can only read old-style UPF files. When I use files with <UPF version="2.0.1">, there is always a error message "Error in routine scan_begin (1):                  No HEADER block". I find pseudopotentials generated by atomic-5.0.2.tar.gz and pslibrary.1.0.0.tar.gz are always with a head line <UPF version="2.0.1">.
>>     Can I generate old-style UPF files by "atomic" program or can I convert new type of UPF to the old version?
>>     Thanks!
>> Regards,
>> Chang Chen
>> 
>> 
>> 
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>> 
>
>-- 
>Carsten Fortmann
>Scientific Software Developer
>QuantumWise A/S
>Lersø Parkalle
>2100 Copenhagen
>Denmark
>
>Phone: +45 699 01 888
>Fax:   +45 698 02 801
>skype: carsten.fortmann
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20150123/91d5c0de/attachment.html>