[Wien] difference in forces from mini and scf calculations

[Md. Fhokrul Islam]

Dear Javad,

     Thanks for your explanation and the reference. I think I know what is the difference between the
two cases. I mixed up partial forces with total force on the atom.

Regards,
Fhokrul



From: hashemifar at cc.iut.ac.ir
Date: Wed, 22 Jun 2011 09:07:03 +0430
To: wien at zeus.theochem.tuwien.ac.at
Subject: Re: [Wien] difference in forces from mini and scf calculations

Dear Fhokrul

Since basis sets in APW/LAPW methods depend on the atomic positions, force calculations in these methods needs a 

special correction called Pulay force [Physics Letters A, Volume 299, 2002, 413] Since calculation of this important correction is somewhat time consuming, in normal scf calculations that accurate forces are not crucial, wien2k does not calculate pulay force and print the calculated forces without this correction (partial forces). But when you are doing mini calculation or if you change the first line switch of case.in2 input file from TOT to FOR in a normal scf run, then the Pulay correction is applied to the calculated force (total force).

RegardsS. Javad Hashemifar
======================================
Seyed Javad  Hashemifar 
Physics Department, Isfahan University of Technology
84156-83111 Isfahan, Iran


Tel: +98 311 391 2375     Fax:+98 311 3912376
Email:  hashemifar at cc.iut.ac.ir
Homepage:  http://hashemifar.iut.ac.ir
---------------------------------------------------------------------------





2011/6/22 Md. Fhokrul Islam <fislam at hotmail.com>







Hi Marks,

      Thanks for your reply. Let me clarify my question little bit more. 

1) I have optimized the slab using mini. 

2) I have took this optimized case.struct file from mini calculation, initialized  


    a job for wein2k and run scf calculation.

But when I check the force on the same atom here is what I get.

case-1: From mini calculation: 
grep :FGL073 *mini|tail -1
:FGL073:  73.ATOM                 0.233000000     0.045500000     4.378500000 total forces



case-2: From scf calculation: 
grep :FGL073 *scf | tail -1
:FGL073:  73.ATOM                -1.725500000    -0.171500000  -301.077000000 partial forces

My question is why there is difference in force even though in both cases I am using the same


structure. In the 2nd case, I have slightly larger RMT and k points. Is it the reason for this 
difference or I am comparing totally different things.

Thanks,
Fhokrul


Date: Tue, 21 Jun 2011 22:18:38 -0500


From: L-marks at northwestern.edu
To: wien at zeus.theochem.tuwien.ac.at
Subject: Re: [Wien] difference in forces from mini and scf calculations



You do not provide enough information, so all amyone can do is guess.

My guess: you are confusing the valence corrected forces (total at the end of case.scfm) and the uncorrected forces (partial). Mini works with the corrected forces.

On Jun 21, 2011 8:40 PM, "Md. Fhokrul Islam" <fislam at hotmail.com> wrote:
> 
> Hi Prof Blaha,
> 
>     I optimized a slab using mini and it worked fine (forces on atoms were reasonably small) . 



> But when I used the optimized structure from mini calculations and run scf calculations (with 
> energy and charge convergence) the forces on atoms increases even after energy is converged. 
> So my questions are:



> 
> 1. why forces increases even though I am using exactly the same structure from mini calculations?
> 
> 2. how does it effect the calculated properties such as DOS, magnetic moments? I mean should I



>     trust these numbers that I have calculated from scf calculations?
> 
> I have tried looking for the answer in the mailing list but unfortunately I couldn't find one. I would
> appreciate any comment from you.



> 
> 
> Thanks,
> Fhokrul
> 
>  		 	   		  


_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien 		 	   		  

_______________________________________________

Wien mailing list

Wien at zeus.theochem.tuwien.ac.at

http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien





_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien 		 	   		  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20110622/3c39110f/attachment.htm>