BSE with finite momentum Q

Various technical topics such as parallelism and efficiency, netCDF problems, the Yambo code structure itself, are posted here.

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan, Nicola Spallanzani

Post Reply
Franz Fischer
Posts: 43
Joined: Wed Jul 20, 2022 9:36 am

BSE with finite momentum Q

Post by Franz Fischer » Wed Nov 01, 2023 3:04 pm

Hi all,

I did some finite center of mass momentum (Q) BSE calculations.
To my surprise the number of exchange G-vectors used in order to compute the G-sum for the electron-hole exchange matrix elements increased drastically from 23005 (for Q=0) to 111631 (for Q .ne. 0).
Furthermore I could not fully understand how the FFT grid for Q .ne. 0 (38 x 38 x 287) can get bigger than the FFT grid given by Quantum Espresso (36 x 36 x 270).

Why is this the case?

In the file "src/bse/K_driver.F" I found these 4 lines:

Code: Select all

if ( then
       	iG_BSE =max(BS_n_g_exch,BS_n_g_W)
Why do you multiply the qindx_B_max by 4? Can you elaborate on this?

Thanks in advance.
Franz Fischer
PhD student / IMPRS-UFAST fellow
Institute of Physical Chemistry
University of Hamburg

User avatar
Davide Sangalli
Posts: 617
Joined: Tue May 29, 2012 4:49 pm
Location: Via Salaria Km 29.3, CP 10, 00016, Monterotondo Stazione, Italy

Re: BSE with finite momentum Q

Post by Davide Sangalli » Thu Nov 02, 2023 6:29 pm

Dear Franz,
I checked the code, and I think you can safely remove the lines you reported in your message and recompile the code.

Also, if you want, you can test this branch of the code which has some improvement for BSE. ... -dichroism

Davide Sangalli, PhD
CNR-ISM, Division of Ultrafast Processes in Materials (FLASHit) and MaX Centre

Post Reply