[Wien] new wien2wannier version 0.96 including woptic to compute optical conductivity with Wannier functions

[Philipp Wissgott]

Dear wien2k-users,

There is a new*wien2wannier* version (0.96) available for download at

http://www.wien2k.at/reg_user/unsupported/wien2wannier

Beside some bug-repairs, it includes the post-processing tool *woptic 
*to compute the *optical conductivity*, dc-conductivity and thermopower 
in the basis of Wannier functions computed by wien2wannier.

The main purpose of the algorithm is to compute these observables 
employing a self energy sigma(w) from many-body calculations.

Some features of woptic:
1) It uses the wien2k momentum matrix elements(dipole matrix elements) 
computed by optic rotated in the basis of Wannier functions
2) To include correlation effects one can use the self energy on the 
real axis sigma(w) e.g. from DMFT calculations.
3) It uses adaptive k-mesh refinement, i.e. in an iterative process, the 
algorithm identifies regions in k-space where a larger portion of the 
k-integration error comes from and resolves these regions with a larger 
number of k-points thus decreasing the error.
4) To compare with state-of-the-art LDA+DMFT, one can also use the 
Peierls approximation instead of the full momentum matrices.
5) One can also use the algorithm for pure LDA cases. It is, however, 
often less efficient than optic-joint-kram. Since woptic uses a direct 
linear response formula without a Kramers-Kronig transform, it provides 
a good way to check the results.
6) Optical transistions are not limited to the Wannier basis, but one 
can also include transitions between the wien2k bands outside of the 
wannier90 energy window and the Wannier orbitals (which are usually 
adapted by correlation effects).

Note that it is the first time that the full momentum matrix elements 
are used in the LDA+DMFT framework. For results, computed with woptic, see

P. Wissgott, J. Kunes, A. Toschi and K. Held  PRB  85, 205133 (2012)
Dipole matrix element approach versus Peierls approximation for optical 
conductivity

For details on the usage see the wien2wannier userguide.

For those of you who already downloaded wien2wannier at an earlier 
stage: there is an old woptic version already included in the older 
wien2wannier versions which was however rather buggy. This release 0.96 
should therefore represent the first "official" version of woptic.

Best regards,
Philipp Wissgott


-- 
Philipp Wissgott
Institute for Solid State Physics, TU Wien
phone ++43 1 58801 13769
wissgott at ifp.tuwien.ac.at
http://www.ifp.tuwien.ac.at/cms

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20120619/db481b80/attachment.htm>