[Pw_forum] k point selection

[Stefano Baroni]

On Jun 14, 2006, at 2:30 AM, Luke Thulin wrote:

> Eyvaz and Stefano,
>
> Thank you very much for the instruction on k point sampling,  I'll  
> look into getting a copy of the Monkhorst Pack paper.  One more  
> quick issue:  To save on computation time, does the automatic k  
> point sampling for the scf calculation determine by itself which  
> points comprise the irriducible brillouin zone?

try by yourself. run a calculation for, say, an fcc structure and see  
whether or not you find two k points which are equivalent by any  
bubic symmetry operation. (any operation of the cubic group can be  
seen as the product of a parmutation of the cartesian coordinates -  
there are seix such permutations - and the change of sign of any of  
the three coordinates - there are 2^3=8 such operations - for a total  
of 48=6x8 symmetry operations)

> Or is it at least general practice to reuse the scf calculation  
> results among different nscf calculations along different k paths?

I do not quite get the question

> Sorry if this question has been asked, but I seem to be  
> incompentent when it comes to searching the archives.

never mind: the important thing is to keep learning (and learning how  
to search archives is not that difficult!)

S.

>
> Thanks,
> Luke
>
> eyvaz_isaev at yahoo.com wrote:
>> Hi,
>>
>>
>>> Therefore, I'm currently staying with automatic
>>> generation.
>>>
>> This one you have to use for SCF-calculations.
>>
>>
>>> To do band structure plots, is it necessary to
>>>
>> choose > the  same number of k points for both the scf
>> and
>>
>>> nscf  calculations, or can less be used for the scf?
>>>
>> You can use the same set of k-points (you should
>> extract them from scf_output_file) them manually , but
>> the result  would be very ugly. You will not be able
>> to plot because k-points  are not aligned along high
>> symmetry directions.
>>
>> So, the answer is "No". You have to generate k-points
>> along a given directions and then calculate band
>> energies for these k-points
>>
>>
>>> If I try to enter a list of k points manually so
>>> that I can choose my
>>> path, how do you determine the weights?
>>>
>> Put them as 1.0, they are not used for band structure
>> calculations.
>>
>> Actually, I am going to redraw a small set of programs
>>  to get band structure (using Gnuplot) for cubic and
>> hexagonal lattices, at least. Hopefully it would be
>> useful for beginners.
>>
>> Bests,
>> Eyvaz.
>>
>>
>>
>>
>> __________________________________________________
>> Do You Yahoo!?
>> Tired of spam?  Yahoo! Mail has the best spam protection around
>> http://mail.yahoo.com
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>>

---
Stefano Baroni - SISSA  &  DEMOCRITOS National Simulation Center -  
Trieste
[+39] 040 3787 406 (tel) -528 (fax) / stefanobaroni (skype)

Please, if possible, don't  send me MS Word or PowerPoint attachments
Why? See:  http://www.gnu.org/philosophy/no-word-attachments.html



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20060614/03ad3fec/attachment.html>