[Wannier] exclude_bands vs energy windows

[Mostofi, Arash]

Dear Temok,

The short answer to your question is no, you cannot use exclude_bands in lieu of disentanglement. Entangled bands, for example, often cross each other in the Brillouin zone, whereas the exclude_bands parameter is not k-dependent, i.e.,
exclude_bands = 3,4,5
excludes the 3rd, 4th and 5th lowest bands at every k-point, irrespective of crossings etc.

Best wishes,

Arash

—
Arash Mostofi — www.mostofigroup.org<http://www.mostofigroup.org>
Reader in Theory and Simulation of Materials
Imperial College London
Director, Thomas Young Centre @Imperial

On 3 Mar 2016, at 16:25, Cuauhtemoc Salazar <salazar at physics.utoronto.ca<mailto:salazar at physics.utoronto.ca>> wrote:

Dear Arash,

Thanks for your kind reply. Along your explanation you make the example of focusing / excluding attention to isolated groups of bands, like valence or conduction bands of a semiconductor. But but my concern is on whether exclude_bands is an effective, equivalent way of separating entangled bands. From the definition of exclude_bands in the W90 User Guide, Sec 2.6.6, it seems implicit that I could *always* use exclude_bands (instead of defining energy windows) regardless of whether the bands are entangled or are part of an isolated group. Is this correct?

(I’d like to extract four MLWFs associated with the first four conduction bands that are entangled to higher bands)

Thank you,
Temok



On Mar 3, 2016, at 04:51, Mostofi, Arash <a.mostofi at imperial.ac.uk<mailto:a.mostofi at imperial.ac.uk>> wrote:

Dear Salazar,
The exclude_bands list is used to tell the electronic structure code to avoid calculating the matrix elements associated with certain bands that you know you are not going to be interested in. As such, it is a less flexible way of focussing your interest on, e.g., the valence vs conduction bands of a semiconductor than simply computing all of the overlaps (i.e., including all the bands) and using the energy window parameters. Conversely, there are cases in which exclude bands can be useful, e.g., by saving computational time in avoiding the calculation of overlaps that you know are not of interest.
Hope this helps,
Arash

—
Arash Mostofi — www.mostofigroup.org<http://www.mostofigroup.org/>
Reader in Theory and Simulation of Materials
Imperial College London
Director, Thomas Young Centre @Imperial

On 2 Mar 2016, at 15:05, Cuauhtemoc Salazar <salazar at physics.utoronto.ca<mailto:salazar at physics.utoronto.ca>> wrote:

Dear All,

I wonder if setting
 exclude_bands
to a proper range of bands is equivalent to (i.e. can be used instead of)
dis_win_min
dis_win_max
dis_froz_min
dis_froz_max.

I am interested in extracting MLWF's associated to the 4 upper valence bands and the first 4 conduction bands of a semiconductor.
Those 4 conduction bands are entangled with higher energy bands.

Thanks,
Salazar

PhD Candidate
Physics Department
University of Toronto
_______________________________________________
Wannier mailing list
Wannier at quantum-espresso.org<mailto:Wannier at quantum-espresso.org>
http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier


_______________________________________________
Wannier mailing list
Wannier at quantum-espresso.org<mailto:Wannier at quantum-espresso.org>
http://mailman.qe-forge.org/cgi-bin/mailman/listinfo/wannier

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/wannier/attachments/20160304/0148f276/attachment.html>