[Wien] Convergence problem for the slab calculation!
Michael Gurnett
michael.gurnett at kau.se
Tue Dec 20 14:55:38 CET 2005
Wouldn't this type of test be possible to incorporate in the scf so that
these parameters are adjusted "on the fly"
Michael
-----Original Message-----
From: "L. D. Marks" <L-marks at northwestern.edu>
To: A Mailing list for WIEN2k users <wien at zeus.theochem.tuwien.ac.at>
Date: Tue, 20 Dec 2005 07:34:12 -0600 (CST)
Subject: Re: [Wien] Convergence problem for the slab calculation!
> Several things:
>
> 1) What is the "Warning" that you are getting? You need to look at the
> scf
> code to see what this is due to. Please check the archives (and/or the
> user guide) about what you find.
>
> 2) Inspecting the energy (:ENE) and charge (:DIS) are relics of "the
> bad
> old days", as is reducing the mixing coefficient to values like 0.01.
> The
> correct way to look at these problems is to look at the behavior of the
> plane waves (:PLANE) and the angle between the Broyden and Pratt steps
> (:DIRB); run the script attached. If this angle is large (say > 50
> degrees) you need to reduce the plane wave scaling in the mixer. You
> should also look at the total plane wave interstitial charge (grep -e
> "CTO
> "), sometimes this just takes time to converge particularly for a
> surface.
>
> 3) There are a few cases (rare) when reducing the plane wave scaling
> does
> not work (by itself) and one has to also reduce the mixing coefficient
> to,
> say, 0.05 or even less. I suspect that most of these are cases where
> the
> physics is not properly defined in the calculation, e.g. one should be
> using more k-points or running spin polarized.
>
> On Tue, 20 Dec 2005, ²Ì ÃÏÇï wrote:
>
> > Dear wien2k user
> >
> > As a test, I am calculating the BaTiO3 surface with nine layers slab
> and the
> > same thickness vacuum. I used a Rkmax of 7 and gmax of 14 with the
> mixing
> > coefficient of 0.1 and even 0.01. Moreover, I correct the
> E-parameters in
> > case.in1 to replace all "0.3" by appropriate values 0.1 (below our EF
> about
> > 0.25). But I get energy oscillations in the scf cycle.
> > :ENE : ********** TOTAL ENERGY IN Ry = -177777.148950
> > :ENE : ********** TOTAL ENERGY IN Ry = -177776.243123
> > :ENE : ********** TOTAL ENERGY IN Ry = -177776.971894
> > :ENE : *WARNING** TOTAL ENERGY IN Ry = -177775.571717
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.294152
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.176350
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.170225
> > :ENE : *WARNING** TOTAL ENERGY IN Ry = -177775.117307
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.182107
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.115802
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.106464
> > :ENE : *WARNING** TOTAL ENERGY IN Ry = -177775.124568
> > :ENE : *WARNING** TOTAL ENERGY IN Ry = -177775.107141
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.191016
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.104630
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.107498
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.241324
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.097320
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.095030
> > :ENE : ********** TOTAL ENERGY IN Ry = -177775.119517
> >
> >
> > Could exist any other problem?
> >
> > Thank you!! for your time and advise.
> >
> > Dr. Cai
> >
> > _________________________________________________________________
> > ÓëÁª»úµÄÅóÓѽøÐн»Á÷£¬ÇëʹÓà MSN Messenger:
> http://messenger.msn.com/cn
> > _______________________________________________
> > Wien mailing list
> > Wien at zeus.theochem.tuwien.ac.at
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >
>
> Note: if you have an old email address for me, please note that "nwu"
> has
> been changed to "northwestern".
> -----------------------------------------------
> Laurence Marks
> Department of Materials Science and Engineering
> MSE Rm 2036 Cook Hall
> 2220 N Campus Drive
> Northwestern University
> Evanston, IL 60201, USA
> Tel: (847) 491-3996 Fax: (847) 491-7820
> email: L-marks at northwestern dot edu
> http://www.numis.northwestern.edu
> -----------------------------------------------
>
More information about the Wien
mailing list