[Pw_forum] Fw : Help on ev.x procedures!!!

Mon Feb 15 15:20:28 CET 2010

--- En date de : Lun 15.2.10, Bertrand SITAMTZE <siyouber at yahoo.fr> a écrit :

De: Bertrand SITAMTZE <siyouber at yahoo.fr>
Objet: Help on ev.x procedures!!!
À: "Bertrand SITAMTZE" <siyouber at yahoo.fr>
Date: Lundi 15 février 2010, 15h19

Dear all, I have posted on the forum for ev.x procedures but still waiting for any help in understanding the ev.f90 file.

please, your help is welcome.

*******************************
Bertrand SITAMTZE YOUMBI
Laboratory of Material Science
University of Yaounde I-Cameroon
*************************************

--- En date de : Mer 10.2.10, Bertrand SITAMTZE <siyouber at yahoo.fr> a écrit :

De: Bertrand SITAMTZE <siyouber at yahoo.fr>
Objet: Re: [Pw_forum] Fw : ev.x procedures
À: "PWSCF Forum" <pw_forum at pwscf.org>
Date: Mercredi 10 février 2010, 11h38

Dear Sandro and all,

I check the ev.f90 file and the way those EOS are written do not make it easy to write the analytic form. Please can somebody help in giving the four EOS use in this file?

Also, the fitting procedure leads to an output  file containing a fitting pressure. But the problem I have is that there are negative pressures. What is the unit of the fitting pressure in this file? How to convert it in Pa? Is it correct to have negative pressure?

Thanks for your answers

*******************************
Bertrand SITAMTZE YOUMBI
Laboratory of Material Science
University of Yaounde I-Cameroon
*************************************

--- En date de : Jeu 4.2.10, Sandro Scandolo
 <scandolo at ictp.it> a écrit :

De: Sandro Scandolo <scandolo at ictp.it>
Objet: Re: [Pw_forum] Fw : ev.x procedures
À: "PWSCF Forum" <pw_forum at pwscf.org>
Date: Jeudi 4 Février 2010, 0h21

Dear Bernard,

>
> So, if I want to plot the P-V diagram, I can just calculate the 
> pressure for each volume using the EOS,since the fitting parameters 
> are already known. Is it correct?
>

Yes, correct.

>
> Also, when implementing the fitting procedure iteratively with ev.x, 
> on has to choose between the Murnagham, Birch 1 and Birch 2. According 
> to me, there is nothing in the litterature known as Birch 1 and Birch 
> 2. I thing they refer to Second order and Third order Birch-Murnagham 
> EOS. I would like you to
 clarify it for me.
>

I presume so, but you may wish to check what's inside ev.f90 and let us

know. :-)

Regards,
Sandro

>
>
>
> --- En date de : *Mar 2.2.10, Sandro Scandolo /<scandolo at ictp.it>/* a 
> écrit :
>
>
>     De: Sandro Scandolo <scandolo at ictp.it>
>     Objet: Re: [Pw_forum] Fw : ev.x procedures
>     À: "PWSCF Forum" <pw_forum at pwscf.org>
>     Date: Mardi 2 Février 2010, 16h19
>
>     Dear Bertrand,
>
>     pressure is not an input of your pw calculation. It is an output
>     of the
>     ev.x calculation.
>
>     What ev.x does is to fit the results of your E(V) pw calculations
>     to
 a
>     chosen EOS functional form, with a number of parameters. Most
>     functional
>     forms (e.g. Birch-Murnaghan) include the parameters K and K' (see
>     http://en.wikipedia.org/wiki/Birch-Murnaghan_equation_of_state ).
>     P(V)
>     can then be calculated from the chosen E(V) (and the fitted
>     parameters)
>     as P(V)=-dE(V)/dV .
>
>     Notice that the P(V) calculated form the EOS with the fitted
>     parameters
>     may be slightly different with respect to the one calculated by
>     pw.x at
> 
    the volumes where you did your pw.x calculations. But this is another
>     story (related to
>     http://www.quantum-espresso.org/wiki/index.php/Methodological_Background#Stress
>
>     ).
>
>     So, coming to your question, K' is not obtained by finite
>     differences,
>     but rather directly as a result of the E(V) fit.
>
>     Sandro
>
>
>     Bertrand SITAMTZE wrote:
>     >
>     >
>     > --- En date de : *Mar 2.2.10, Bertrand SITAMTZE
>     /<siyouber at yahoo.fr </mc/compose?to=siyouber at yahoo.fr>>/*
>     > a écrit :
>     >
>     >
>     >     De: Bertrand SITAMTZE <siyouber at yahoo.fr
>     </mc/compose?to=siyouber at yahoo.fr>>
>     >     Objet: ev.x procedures
>     >     À: "pw code" <pw_forum at pwscf.org
>     </mc/compose?to=pw_forum at pwscf.org>>
>     >     Date: Mardi 2 Février 2010, 11h04
>     >
>     >     (SORRY, THERE
 WAS A SMALL MISTAKE IN MY PREVIOUS MESSAGE)
>     >
>     >
>     >     Dear all,
>     >
>     >     As I know, scf and nscf in pw.x are with zero pressure (please
>     >     correct me if I am
 wrong). Thus, when evaluating the EOS
>     >     parameters with ev.x, the code uses the E vs VOLUME data to
>     >     evaluate the bulk modulus throught the second derivative.
>     But, how
>     >     does the code proceed in order to evaluate the pressure
>     derivative
>     >     of the bulk modulus K', without varing the pressure? If
>     otherwise,
>     >     can somebody tell me the range of pressure used to evaluate K'?
>     >
>     >     Please, note that K' is defined as K'=(dK/dP) at P=0.
>     >
> 
    >
>     >     Thanks very much for your answers.
>     >
>     >
>     >
>     >     ***************************************
>     >     Bertrand SITAMTZE YOUMBI
> 
    >     Laboratory of Material Sciences
>     >     University of Yaounde I-Cameroon
>     >     *************************************
>     >
>     >
>     >
>     >
>     ------------------------------------------------------------------------
>     >
>     > _______________________________________________
>     > Pw_forum mailing list
>     > Pw_forum at pwscf.org </mc/compose?to=Pw_forum at pwscf.org>
>     > http://www.democritos.it/mailman/listinfo/pw_forum
>     >   
>
>     -- 
>     Sandro
 Scandolo, ICTP | http://www.ictp.it/~scandolo
>     <http://www.ictp.it/%7Escandolo>
>
>     _______________________________________________
>     Pw_forum mailing list
>     Pw_forum at pwscf.org </mc/compose?to=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
>   

-- 
Sandro Scandolo, ICTP | http://www.ictp.it/~scandolo

_______________________________________________
Pw_forum mailing list
Pw_forum at pwscf.org
http://www.democritos.it/mailman/listinfo/pw_forum

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20100215/a728c835/attachment.html>