[Wien] Optimization of fraction of exact exchange

[Laurence Marks]

My approach. You chose some valid set of experimental knowns, e.g.
atomization energies of clusters, heats of formation, reactions using
relevant atoms. You then vary the fraction to find the best fit. The value
will only be "best" to about 0.05.

An alternative that in a few cases worked surprisingly well (with -eece) is
to minimize the forces using the experimental bulk positions, allowing for
a lattice parameter change.

I do not consider fitting the band gap to be valid, as this is an excited
state property. It is valid with +U as this is implicitly a quasi-excited
method since it is defined via change in occupancy.

---
Professor Laurence Marks
"Research is to see what everybody else has seen, and to think what nobody
else has thought", Albert Szent-Gyorgi
http://www.numis.northwestern.edu
Corrosion in 4D http://MURI4D.numis.northwestern.edu
Partner of the CFW 100% gender equity project, www.cfw.org/100-percent
Co-Editor, Acta Cryst A


On Sep 23, 2017 07:36, "Wien2k User" <wien2k.user at gmail.com> wrote:

> Dear Wien2k users:
>
> In a recently published work using Hybrid Funtional I read that the
> authors have optimized the fraction of exact exchange but I did not
> understand how to choose its optimal value
>
> My question is how to optimize the fraction of exact exchange (alpha)
> *. *
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20170923/790c9670/attachment.html>