[Wien] Version 7.1 - Charged supercells

Tue Jan 30 14:36:52 CET 2007

On 1/30/07, E B Lombardi <lombaeb at science.unisa.ac.za> wrote:
> Thank you for your reply.
>
> I would be very interested if the details could be included on the
> Wien2k web page.
> Do you have any suggestions for determining V_0(Q)?

I think the only way you can get it is by trying to establish a
reference for something like a deep core level or (perhaps) the
VCOUL-ZERO within the RMT. You will need to do computer experiments to
try and determine a good way, and not rely upon what I say (which may
be wrong because not everything has been tested).

> Presumably the Madelung correction must still be added to the total
> energy provided by Wien2k?

I suspect that one does not need an additional Madelung correction,
but I might be wrong. This needs to be verified by computer
experiments.

> The corrections that converge as O(L^-1) and O(L^-3) discussed in the
> literature - are they included in the -Q*V_0(Q)/2 term or are they
> included in the new Wien2k E-tot?

Again, I think most of these are included in -Q*V_0(Q)/2. No
correction terms are built into the new Wien2k E-tot.

>
> I presume old charged calculations should be rerun from scratch, and not
> just running additional SCF cycles  to convergence with the new version?
> Is there any reason to doubt geometry optimization of charged supercells
> from previous versions of Wien2k?

Certainly not! I would restart the old calculations with a small
mixing term (since I've found that charged cells tend to be
non-convex). The geometry optimization may be a bit different, I think
you need to test this.

>
> Thank you
> Best regards
> Enrico Lombardi
>
>
> Laurence Marks wrote:
> > Essentially, the charged cell calculations in the earlier versions of
> > Wien were wrong, certainly for the energy and probably for the
> > electronic structure due to an inappropriate addition of charge
> > compensation. Peter corrected this, and I tested it in some detail.
> >
> > There is still an energy correction which I am 98% confident is
> > -Q*V_0(Q)/2 where Q is the excess charge of the cell (-ve for a
> > negative cell) and V_0(Q) is the vacuum Coulomb potential in the
> > calculation which is a function of Q. I am reasonably confident that
> > this includes most additional correction terms that people have talked
> > about in the literature. The details of how I got this have not yet
> > been published, and might never be, but in principle could be posted
> > on the Wien page if there is an interest. This does require the user
> > to find, somehow, V_0(Q) which is not so easy.
> >
> > On 1/29/07, E B Lombardi <lombaeb at science.unisa.ac.za> wrote:
> >
> >> Dear Prof. Blaha
> >>
> >> I am interested to know more about the fixes for charged cells in lapw0:
> >>
> >> Specifically, whether the comments on the Wien list below still apply
> >> and whether the corrections for charged cells are included in this new
> >> version?
> >>
> >> What differences can be expected in old Wien2k06 vs new Wien2k07 charged
> >> supercell calculations?
> >>
> >> I assume the procedure outlined in the FAQ still applies unchanged?  It
> >> would also be useful to include charged supercell limitations/fixes in
> >> the FAQ.
> >>
> >> Thank you in advance
> >> Best regards
> >> Enrico Lombardi
> >>
> >> PS: the Wien2k usersguide available under "Textbooks" seems to be an
> >> older version than included in the SRC directory.  All versions of the
> >> guide state "November 2001" and "revised edition June 2002".
> >>
> >>
> >>> I would NOT trust these energies either!
> >>> There is no correction in E-tot for adding this background charge nor the
> >>> global correction for charged cells (described in literature).
> >>>
> >>> Use these charged cells only for "DOS" or core-level spectroscopy,...
> >>>
> >>>
> >>>> I've always had some reservations about using the energy number that comes out
> >>>> of a charged cell calculation for anything. Since I did not find this
> >>>> discussed anywhere on the list, maybe someone can educate me as to whether it
> >>>> is useful (with perhaps some corrections) or to be avoided like the plague.
> >>>>
> >>> Does the energy that Wien2k gives for a charged cell require any
> >>> additional corrections beyond the standard ones in the literature,
> >>> e.g. Leslie and Gillan, J. Phys. C 18, 973 (1985)?
> >>>
> >>
> >> Peter Blaha wrote:
> >>
> >>> Dear WIEN2k users,
> >>>
> >>> A new version, WIEN2k_07.1 is ready for download.
> >>>
> >>> Highlights:
> >>> batch-mode for init_lapw;
> >>> hybrid-methods for correlated electrons (may replace LDA+U)
> >>> SRC_lapw0: fixes for charged cells, electric field, speedup
> >>> SRC_lapwdm: more accurate orbital and dipolar hyperfine fields
> >>> SRC_lcore: increased precision in printout. Much more stable (in
> >>> particular for heavier elements), but may change :ENE in comparison to
> >>> previous versions
> >>>
> >>> Please see the update page of www.wien2k.at for details.
> >>>
> >>> Thanks to all contributions
> >>>
> >>> Best regards
> >>>
> >>> PS: Since a new year has started a reminder: Please submit your papers
> >>> where WIEN2k was used into our website at  http://www.wien2k.at/papers
> >>> _______________________________________________
> >>> Wien mailing list
> >>> Wien at zeus.theochem.tuwien.ac.at
> >>> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >>>
> >>>
> >>>
> >> --
> >> Dr E B Lombardi
> >> Physics Department
> >> University of South Africa
> >> P.O. Box 392
> >> UNISA 0003
> >> Pretoria
> >> South Africa
> >>
> >> Tel: +27 (0)12 429 8654 / 8027
> >> Fax: +27 (0)12 429 3643
> >> E-mail: lombaeb at science.unisa.ac.za
> >>
> >> _______________________________________________
> >> Wien mailing list
> >> Wien at zeus.theochem.tuwien.ac.at
> >> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> >>
> >>
> >
> >
> >
>
> --
> Dr E B Lombardi
> Physics Department
> University of South Africa
> P.O. Box 392
> UNISA 0003
> Pretoria
> South Africa
>
> Tel: +27 (0)12 429 8654 / 8027
> Fax: +27 (0)12 429 3643
> E-mail: lombaeb at science.unisa.ac.za
>
> _______________________________________________
> 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
http://www.numis.northwestern.edu