Why non-linear optical properties not available for PAW?  [SOLVED]

Moderators: jzwanzig, jolafc

Why non-linear optical properties not available for PAW?

Postby kurtg » Tue Dec 06, 2016 2:48 am

Hi all,

I am wondering why it is not available to calculate non-linear optical properties for PAW PBE atomic datasets.

As I understand, as long as we have the electron transition matrix <psi_i|psi_j>, we can calculate the non-linear optical susceptibility. What is the main difference from using a norm-conserving pseudopotential? Is the transition matrix element calculated with PAW PBE incorrect?

Best,
Kurt
kurtg
 
Posts: 3
Joined: Tue Dec 06, 2016 2:29 am

Re: Why non-linear optical properties not available for PAW?  [SOLVED]

Postby ebousquet » Wed Dec 07, 2016 1:40 pm

Dear Kurt,
The problem is that the equations are not the same and have to be solved and implemented by someone for each case. The derivatives with GGA are more cumbersome than the ones with LDA and PAW is also more cumbersome than norm-conserving pseudopotentials and more you go high in the expansion, more complicated it is.
Look at the DFPT+PAW paper for example: http://journals.aps.org/prb/abstract/10.1103/PhysRevB.73.235101
Things are under development, see for example the log files of the different abinit versions to follow the status of these developments.
Best wishes,
Eric
ebousquet
 
Posts: 238
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium


Return to Theory

Who is online

Users browsing this forum: No registered users and 1 guest