The convergence problem of BndsRnXp and NGsBlkXp

[Ying Chen]

Dear Developers,

I am using YAMBO to calculate the quasi-particle band structure of the MoS2/WS2 heterojunction. But when the BndsRnXp reaches '1 | 120 |' and the NGsBlkXp reaches 9 Ry, the band gap still cannot converge. How can I speed up the convergence of the band gap? Is there anything in particular that needs to be paid attention to in calculating the quasiparticle band structure of the heterojunction? Thank you!

Best regards,
Ying Chen

[Daniele Varsano]

Dear Ying Chen,
please sign your post with your name and affiliation, this is a rule of the forum and you can do once for all by filling the signature in the user profile.

It seems to me that something weird happened in your calculation using 110 and 120 bands, such big oscillations are not expected and moreover, you end up in a negative gap, which does not male much sense. I suggest you check in the report and try to understand what went wrong.

In order to accelerate the convergence with respect to the number of bands, you can use the so-called terminator technique (see sections 3.3 and 4.3 of the latest yambo paper).

You can activate it by adding verbosity when building the input file (the -V all)

Code: Select all

yambo -p p -g n -r -V all

and setting:

Code: Select all

XTermEn= 40.81708      eV      # [GW] GW terminator energy (only for kind="BG")
XTermKind= "BG"

to accelerate the convergence with respect BndsRnXp,
and

Code: Select all

GTermEn= 40.81708      eV      # [GW] GW terminator energy (only for kind="BG")
GTermKind= "BG"

to accelerate with respect GbndRnge.
Note that in this way you accelerate the calculation with respect to the number of bands, but the calculations will be more demanding.

Best,
Daniele