[Pw_forum] Graphene Structural Relaxation

[meysam pazoki]

---------- Forwarded message ----------
From: meysam pazoki <meysam.pazoki at gmail.com>
Date: Tue, Jul 20, 2010 at 11:46 PM
Subject: Graphene Structural Relaxation
To: pw_forum-request at pwscf.org


Dear PWSCF users

I have a problem in relaxation of grahene super cell with 24 atoms. In
relaxation procedure ,graphene hooney-comb structure completely destroys and
atoms have no regular ordering after relaxation procedure.I use pbe-rrkjus
psudopotential,4 3 3 kpoint grid,30 rydberg cutff energy,300 rydberg
ecutrho.
I use both ion_damps & bfgs algorithms and get similar results.
If i change the ecutrho to 100, atoms have an irregular positions in 3
dimension,but when ecut rho set to 300 atoms have irregular position
in-plane.
furthermore if i set :trust_radius_ini=.1D0,trust_radius_max=.3D0,bfgs_ndim
= 3,
I have the honeycomb structure but some of c atoms exprience great forces
like:

atom 19 type 1 force = 11948.08469282 -219.69802882 0.00000000

atom 20 type 1 force = 11782.00343549 12.01959200 0.00000000

atom 21 type 1 force = 11748.09287051 115.25218004 0.00000000

atom 22 type 1 force = 11241.09464255 -222.56999757 0.00000000

atom 23 type 1 force = -0.50941109 0.26861287 0.00000000

atom 24 type 1 force = 11905.77649683 -91.34763119 0.00000000

My question is what is wrong in my structure relaxation procedure that
result in such non-physical data.

It would appreciated if anyone can help me in this problem

Thanks in advance

meysam pazoki
Ph.D student of physics
Sharif university of technology
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20100817/0d9480dc/attachment.html>