[Wien] about optimize the hexagonal cell .the force is increasing.

[Laurence Marks]

The answer is to do nothing and just let PORT run WITHOUT changing the
delta parameter. In fact, if you have changed it you have to restart
the minimization (not the full calculation) by using "x pairhess" and
copying .minpair to .minrestart and .min_hess. I suggest that you copy
case.inM_st to case.inM as well. (Caveat, pairhess is not implimented
for rhombohedral unit cells.)

To explain, the code is minimizing the energy, not necessarily
minimizing the forces. If you are far from the solution (which you are
since your forces are high and the code gave you a warning about the
curvature condition) it is perfectly physical for the forces to
increase as the energy drops. Let the code handle this, in most cases
it knows what it is doing better than you do!

For more information see:
http://www.wien2k.at/reg_user/textbooks/Optimization-Notes.pdf
http://xxx.lanl.gov/abs/physics/0608160
http://dx.doi.org/10.1016/j.commatsci.2007.01.004

On 7/27/07, Mary White <us.white at yahoo.com> wrote:
> Dear users,
>         I use PORT option to optimize the hexagonal structure including cell
> parameters and atom position. But the force on an atom is too big ,as
> follows.
>
> :FGL001:   1.ATOM               -57.904000127     0.000000127
> 0.000000000
> :FGL001:   1.ATOM               -59.863000131     0.000000131
> 0.000000000
> :FGL001:   1.ATOM               -68.146000149     0.000000149
> 0.000000000
>
> :FGL002:   2.ATOM               -32.135000070     0.000000070
> 0.000000000
> :FGL002:   2.ATOM               -31.330000068     0.000000068
> 0.000000000
> :FGL002:   2.ATOM               -28.205000062     0.000000062
> 0.000000000
>
> :ENE  : ********** TOTAL ENERGY IN Ry =      -158751.858644
> :ENE  : ********** TOTAL ENERGY IN Ry =      -158751.860470
> :ENE  : ********** TOTAL ENERGY IN Ry =      -158751.867677
>
>
> The force on atom1 is increasing. I have decreased the delta in case.inM.
> This file
> is as follows:
> PORT 2.0 0.20      # PORT/NEWT;  tolf, Initial Trust Radius
> 0.2 0.0 0.0 1.0   #Atom  1 Generated by pairhess
> 0.2 0.0 0.0 1.0   #Atom  2 Generated by pairhess
>
>      The force is still increasing. And
>  " :WARNING, CURVATURE CONDITION FAILED
>  :WARNING, S.Y WAS -1.050783136317473E-005
>  :WARNING, CURVATURE CONDITION FAILED
>  :WARNING, S.Y WAS -1.872153352633030E-004 " is appear.
> I don't know how to deal with this case.
> If any one know about this,please give some suggestion.Any comments will be
> appreciated.
> Thank you in advance.
>
>                          Mary
>
>
>
>
>  ________________________________
> Fussy? Opinionated? Impossible to please? Perfect. Join Yahoo!'s user panel
> and lay it on us.
>
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>


-- 
Laurence Marks
Department of Materials Science and Engineering
MSE Rm 2036 Cook Hall
2220 N Campus Drive
Northwestern University
Evanston, IL 60208, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
email: L-marks at northwestern dot edu
Web: www.numis.northwestern.edu
EMM2007 http://ns.crys.ras.ru/EMMM07/
Commission on Electron Diffraction of IUCR
www.numis.northwestern.edu/IUCR_CED