The ABINIT Users Mailing List ( CLOSED )

[matthieu verstraete <matthieu.jean.verstraete@gmail.com>]

you force nspinor 1 in the input file (don't touch the code), and that
precludes the calculation of the SO.

Matthieu

2009/7/27 Dr. Thanusit Burinprakhon <thanusit@kku.ac.th>:
>>> Dear all
>>>
>>> I've performed GW calculation for Si band structure using the H-G-H psp
>>> and got
>>> the following ERROR. The Abinit version I used is 5.7.3 (MPI version,
>>> prepared
>>> for a x86_64_linux_gnu4.1 computer)
>>>
>>> ======================================================================
>>>  Calculating and writing out Kohn-Sham electronic Structure file
>>>  Using conjugate gradient wavefunctions and energies (kssform=3)
>>>
>>>  outkss: ERROR -
>>>   Variable mpspso should be 1 !
>>>   Program does not stop but _KSS file will not be created...
>>> -P-0000  leave_test : synchronization done...
>>> ================================================================================
>>>
>>> The cause of the above error has been clearly explained in the forum a
>>> long
>>> while ago. That is it's due to the "mpspso" parameter in  h-g-h psp
>>> file
>>> is 2,
>>> which is not applicable to the GW routine in Abinit. It was also
>>> mentioned
>>> that
>>> this constraint could be  removed in the later versions of abinit.
>>> However, the
>>> above result seems to show that this problem hasn't been yet fixed in
>>> Abinit
>>> 5.7.3. Or  if it has been, how can we get GW work with hgh psp?
>>>
>>
>>  Dear Thanusit,
>>
>>  GW calculations with nspinor==2 are not yet available for production.
>>  Besides, the correct inclusion of the spin-orbit term will require
>>  several changes to the code.
>>
>>  You can use HGH pseudos with kssform=3 but only to perform GW
>> calculations
>>  with nspinor=1 such that the SO term will not be included.
>>
>>  Best Regards
>>  Matteo Giantomassi
>>
>>
>
> Dear Matteo
>
> Thank you for your reply. May I ask a couple more questions?
>
> "To perform GW calculations with nspinor=1 such that the SO term will not
> be included", does it mean that we have to have the HGH pseudos with
> nspinor=1, or that we have to force nspinor=1 (somewhere in the code) and
> the existing HGH pseudos with nspinor==2 is applicable?
>
> Sorry to ask the above questions but I haven't got any experience on the
> SO cases.
>
> Best regards,
> Thanusit
>
>



-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Matthieu Verstraete

European Theoretical Spectroscopy Facility (ETSF)
Dpto. Fisica de Materiales,
U. del Pais Vasco,
Centro Joxe Mari Korta, Av. de Tolosa, 72,   Phone: +34-943018393
E-20018 Donostia-San Sebastian, Spain        Fax  : +34-943018390

Mail : matthieu.jean.verstraete@gmail.com
http://www-users.york.ac.uk/~mjv500