[Wien] sign change in Mini

[L. D. Marks]

The UG refers to the (older) newt method and needs to be updated. In the
current version of port (soon to be changed) delta is ignored. In general
I suggest using 1.0 1.0 1.0 for the 1st three unless you want to constrain
an atom position -- don't change it in the middle of a run, leave it be!

The oscillation is standard; port should never diverge. If you do grep -e
:ENE or grep -e :G you should see the energy/gradient going down cleanly.
You might see an increase in the gradient in some steps -- don't worry,
gradients can go up (so long as the energy goes down).

On Mon, 16 Aug 2004, Jack London wrote:

> Dear all
>
> Using "port" to minize 9 layers Pd(117) slab, I find
> the force signs change from "positive" to "negtive" in
> some step during the Minimization step (it does not
> finish yet). So the force has a little bit
> oscillation.
>
>
> :FOR001: 1.ATOM   21.231  7.775  0.155   19.756
> :FOR001: 1.ATOM   21.230  7.611  0.150   19.818
> :FOR001: 1.ATOM   14.736  5.304  0.135   13.748
> :FOR001: 1.ATOM   5.597   0.440  0.244   5.574
> :FOR001: 1.ATOM   2.194  -2.164  0.301  -0.200
> :FOR001: 1.ATOM   4.510  -3.423  0.266  -2.925
>
> :FOR002: 2.ATOM   4.994  -3.314  0.159   3.732
> :FOR002: 2.ATOM   3.863  -2.554  0.161   2.894
> :FOR002: 2.ATOM   3.161  -2.371  0.093   2.089
> :FOR002: 2.ATOM   8.019   3.403  0.135   7.260
> :FOR002: 2.ATOM  10.365   5.999  0.182   8.451
> :FOR002: 2.ATOM   9.685   6.127  0.150   7.499
>
> :FOR003: 3.ATOM  13.149  -5.650  0.462  11.864
> :FOR003: 3.ATOM  12.744  -6.958  0.477  10.666
> :FOR003: 3.ATOM   7.683  -1.765  0.153   7.476
> :FOR003: 3.ATOM   6.239  -5.681 -0.155  -2.574
> :FOR003: 3.ATOM   9.523  -7.118 -0.334  -6.318
> :FOR003: 3.ATOM   8.574  -4.624 -0.479  -7.204
>
> :FOR004: 4.ATOM  13.336 -10.788  0.424  -7.828
> :FOR004: 4.ATOM  14.113 -11.064  0.465  -8.749
> :FOR004: 4.ATOM   8.813 -7.202   0.214  -5.075
> :FOR004: 4.ATOM   5.114  0.017   0.122  -5.113
> :FOR004: 4.ATOM   6.266  1.913   0.081  -5.966
> :FOR004: 4.ATOM   5.693  3.427   -0.043 -4.546
>
> >From the WIEN2k website minimization example, if the
>
> grep :for02 case.scf_mini
>     :FOR02: 2.ATOM 31.846 0.000 0.000  31.846
>     :FOR02: 2.ATOM 29.846 0.000 0.000 -29.846
> DELTA was too large (force changed sign !!!,), it
> could even diverge.
>
> But in my case, the sign is changed in the 4th or 5th
> step, and the absolute value of force is not so large
> like in the sample. I don't know if I should also
> change DELTA to a smaller one. Can the force sign
> changes during minimization?
>
> the file of case.inM:
>
> PORT 5.0           #(PORT, NEWT,
> 0.3 0.3 0.3 0.5    # Atom1
> 0.3 0.3 0.3 0.5    # Atom2
> 0.3 0.3 0.3 0.5    # Atom3
> 0.3 0.3 0.3 0.5    # Atom4
> 0.0 0.0 0.0 0.0    # Atom5
>
> Thanks
>
> Zhang
>
>
>
>
> __________________________________
> Do you Yahoo!?
> Y! Messenger - Communicate in real time. Download now.
> http://messenger.yahoo.com
>
> _______________________________________________
> 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
2225 N Campus Drive
Northwestern University
Evanston, IL 60201, USA
Tel: (847) 491-3996 Fax: (847) 491-7820
mailto:L-marks at northwestern.edu
http://www.numis.northwestern.edu
-----------------------------------------------