[Wien] mBJ calculation freezes for slab calculation

[Chang-Jong Kang]

Dear Wien2k users,

I'm trying to get mBJ (modified Becke-Johnson) results for a slab
calculation.
I'm using Wien2k program with version 11.
I converged a usual LDA calculation successfully,
When I start mBJ loop, after "x lapw0 -grr" succeeds, "x lapw0" freezes.
When I track the problem, It seems that the problem comes
from the evaluation of interstitial potential: XCPOT3

As Professor Blaha pointed out recently,
it might be related to the unphysical evaluation
of mBJ potential near vacuum region.

I want to try the "trick" Prof Blaha suggested - using bulk grr file. -
but I dont' knot how to cook that.

Assume that I have the following unit cell.

(Vacuum------A1-----A2-----A1-------Vacuum)  : slab
(---A1---)                             : bulk

then I get bulk potentials :  vns_grr and vsp_grr.
they consist of expansions of spherical harmonics in MT(A1) and
interstitial potential.
then In the case of slab, should I substitute all the MT potentials in slab
with
MT potential(A1) and its copies plus interstitial potential of bulk?
And should I replace the potential for every iteration in mBJ loop?
I'm somehow confused here.

Thanks,
Chang-Jong Kang
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20130228/949ece9e/attachment.htm>