[Pw_forum] ambiguous results with vc-relax

[Stefano de Gironcoli]

please note that the two outputs that you submitted ARE NOT examples of 
vc-relaxation but of structural relaxation of the internal coordinates 
only. Your question is anyway legittimate...
Are you sure to have converged the total energy of your system with 
respect to cutoff to a sufficient level ?
when comparing  different volumes one is using different basis sets in 
the two cases. If convergence w.r.t. cutoff is not sufficient this  can 
lead to sudden decreases of the energy when increasing the volume.

stefano 
SISSA and DEMOCRITOS

Guda Karthik wrote:
> Hi,
>
>     Here are the input and output files.
>
>     b19_Acklatpos_new.* - original structure
>
>     b19_pwscf.* - scf calculation with a structure obtained after vc-relax
>
>     Please have a look if they are of help.
>
> regards,
> Karthik
>
> On Sat, Aug 14, 2010 at 6:27 AM, Paolo Giannozzi <giannozz at democritos.it>wrote:
>
>   
>> On Aug 4, 2010, at 4:41 , Guda Karthik wrote:
>>
>>     
>>> I am trying to relax an orthorhombic structure using vc-relax
>>> (celldofree - xyz). I start with a structure which is at a Pressure
>>> of -20.5 Kbar and at the end of the vc-relax I get a structure
>>> which is at a pressure of -0.21 Kbar. Surprisingly, the structure
>>> at pressure closer to zero is higher in energy than the original
>>> structure by 8 * 10^-4 eV/formula unit. All the symmetry elements
>>> have been conserved during relaxation and the initial and final
>>> structures are the same. I am not able to make sense of this
>>> result. I observe this inconsistency in my B2 structure as well.
>>> Please help regarding this.
>>>       
>> hard to help without input and output
>>
>> P.
>> ---
>> Paolo Giannozzi, Dept of Physics, University of Udine
>> via delle Scienze 208, 33100 Udine, Italy
>> Phone +39-0432-558216, fax +39-0432-558222
>>
>>
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://www.democritos.it/mailman/listinfo/pw_forum
>>
>>     
>
>
>
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum