[Pw_forum] clarifications regarding structure optimization

[janardhan H.L.]

Dear all,
 
 I have attached input file and also output file with respect
 to that I have few doubts.
 As I wanted to study band structure and DOS coordinates for
 Input file were taken from cif file from ICSD database and
 performed relaxed calculations as mentioned in the previous
 discussion. After a long calculation time, to my surprise,
 releaxed positions were completely different. Please see the
 input and out put file, let me know where I have done a
 mistake. 
 
 with respectful regards
  
 Janardhan H L
 Research Fellow
 PPISR
 Bangalore,
 www.ppisr.res.in
 
 
 
 
 
 ------------------------------
 
> Message: 3
> Date: Mon, 22 Sep 2014 01:44:53 +0800
> From: "janardhan H.L." <janardhanhl at yahoo.com>
> Subject: [Pw_forum] why do we need to relax the atomic
> coordinates
>     obtained    from cif file
> To: "pw_forum at pwscf.org"
> <pw_forum at pwscf.org>
> Message-ID:
>     <1411321493.75997.YahooMailNeo at web190402.mail.sg3.yahoo.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Dear all,
> 
> I am an experimentalist wanted to use QE for obtaining DOS
> and band structure. As I was recently got introduced to QE
> and also most of the theory, I have a small question. I
> searched in PW archives and didnt find the answer looking
> for. My question is as follows,
> 
> While solving a crystal structure we take lots of parameter
> and refine the crystal structure to at most accuracy. If we
> take this atomic coordinates for QE scf calculation does it
> affect the results obtained? 
> 
> If yes why, and relaxing the atomic positions may result in
> different coordinates which may be different from the input
> file which may not give the similar XRD patterns. Hence by
> relaxing the atomic coordinates from cif file and using for
> further calculations may not match to experimental results.
> Please comment on this.
> 
> 
> 
> 
> with respectful regards
> 
> Janardhan H L
> Research Fellow
> Poornaprajna Institute of Scientific Research 
> Bangalore India
> www.ppisr.res.in
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: http://pwscf.org/pipermail/pw_forum/attachments/20140922/c41681e2/attachment-0001.html
> 
> 
> ------------------------------
> 
> Message: 4
> Date: Sun, 21 Sep 2014 17:20:42 -0400
> From: Axel Kohlmeyer <akohlmey at gmail.com>
> Subject: Re: [Pw_forum] why do we need to relax the atomic
> coordinates
>     obtained from cif file
> To: "janardhan H.L." <janardhanhl at yahoo.com>,
> PWSCF Forum
>     <pw_forum at pwscf.org>
> Message-ID:
>     <CADTmJ6Eoau8opwq=7d=4gt_7F1Ssy7rJLnBQjPXcXL0nhg9Cfg at mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
> 
> On Sun, Sep 21, 2014 at 1:44 PM, janardhan H.L. <janardhanhl at yahoo.com>
> wrote:
> > Dear all,
> >
> > I am an experimentalist wanted to use QE for obtaining
> DOS and band
> > structure. As I was recently got introduced to QE and
> also most of the
> > theory, I have a small question. I searched in PW
> archives and didnt find
> > the answer looking for. My question is as follows,
> >
> > While solving a crystal structure we take lots of
> parameter and refine the
> > crystal structure to at most accuracy. If we take this
> atomic coordinates
> > for QE scf calculation does it affect the results
> obtained?
> 
> yes.
> 
> > If yes why, and relaxing the atomic positions may
> result in different
> > coordinates which may be different from the input file
> which may not give
> > the similar XRD patterns. Hence by relaxing the atomic
> coordinates from cif
> > file and using for further calculations may not match
> to experimental
> > results. Please comment on this.
> 
> there are assumptions in *both*, your DFT calculations and
> your x-ray
> diffraction pattern refinement. your "most accurate" results
> may not
> that be as accurate on a absolute scale, they are for the
> most part
> *consistent* with the model that you use to refine your
> crystal
> structure. the same is true for DFT calculations and wanting
> a
> geometry that is *consistent* with the model is usually
> desirable.
> using the unoptimized coordinates may result in a
> wavefunction that is
> "squeezed" into a situation that it doesn't really want to
> be and then
> will "act funny".
> 
> in general, the differences in relative geometries are
> fairly small,
> since in both cases a lot of error cancellation is
> happening. if you
> cannot get a good agreement between the experimental results
> and the
> DFT calculation, that is usually an indication that
> something is going
> badly wrong (or that you may be on the brink of a nobel
> prize worthy
> discovery)
> 
> HTH,
>     axel.
> 
> >
> >
> >
> > with respectful regards
> >
> > Janardhan H L
> > Research Fellow
> > Poornaprajna Institute of Scientific Research
> > Bangalore India
> > www.ppisr.res.in
> >
> > _______________________________________________
> > Pw_forum mailing list
> > Pw_forum at pwscf.org
> > http://pwscf.org/mailman/listinfo/pw_forum
> 
> 
> 
> -- 
> Dr. Axel Kohlmeyer  akohlmey at gmail.com 
> http://goo.gl/1wk0
> College of Science & Technology, Temple University,
> Philadelphia PA, USA
> International Centre for Theoretical Physics, Trieste.
> Italy.
> 
> 
> ------------------------------
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ZincHexahydroxostannate.w50.0.in-second
Type: application/octet-stream
Size: 4987 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20141001/ba7a531c/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: final coordinates
Type: application/octet-stream
Size: 3995 bytes
Desc: not available
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20141001/ba7a531c/attachment-0001.obj>