Hi Roland,

I would highly recommend using psml pseudos: 
https://urldefense.com/v3/__http://www.pseudo-dojo.org/__;!!D9dNQwwGXtA!RrBjENfaENy8ZMudSEjcZAXbM2fn5MOyEJ82THnRwER5QrSWiV4QR4EMegYaKl394GIDwNSAlunD3CJh$
  . I have used the Bi pseudos from here (different system) and got good 
agreement with plane wave codes. You'll need to compile a version of siesta 
that can use psml pseudos, instructions 
here<https://personales.unican.es/junqueraj/JavierJunquera_files/Metodos/Pseudos/Siesta-abinit/Siesta-Abinit-PSML.pdf>.
 It will take a bit of work, but extremely worth it.

You can also use abinit with the psml pseudos (instructions Javier's slides), 
so you can compare the two codes and see if there is an issue with the siesta 
basis.

Before doing any serious tests, you can simply reduce the EnergyShift (try 
something small like 20 meV), and try larger basis sets: DZDP, TZP TZDP, etc. I 
think some are listed in the manual, but you can find a full list in 
siesta/Src/basis_specs.f, around line 1408

Best,

Daniel Bennett

________________________________
From: [email protected] <[email protected]> on behalf of Roland 
Coratger <[email protected]>
Sent: 25 April 2024 04:43
To: [email protected] <[email protected]>
Subject: [SIESTA-L] Hole pockets in Bi


Dear all,

I am currently engaged in investigating the electronic structure of Bismuth 
using Siesta4.1 b4. In the attached file, you will find my results concerning 
the band structure and electron pockets located at the L points of the 
Brillouin Zone (BZ). However, I've encountered an anomaly: hole pockets are not 
aligning with the trigonal axis (T point of BZ), but instead, they are slightly 
displaced along the H direction while the rest of the electronic structure is 
consistent with previous calculations of several authors. This displacement 
persists regardless of the basis set used (be it DZP or a more sophisticated 
one, as detailed in my fdf file).

While I'm open to the possibility of error, I think that a bad pseudo (my entry 
file for the Atom program is also in the attached file) or a poor basis set 
would result in very unsuitable electronic structures, not only at the T 
points. I'm reaching out to seek your insights. Perhaps one of you might have 
suggestions on what adjustments I could make to enhance the accuracy of my 
results and correctly obtain the structure at the T points.

Thank you sincerely for your attention and any assistance you can provide.

Regards,

Roland





-- 
SIESTA is supported by the Spanish Research Agency (AEI) and by the European 
H2020 MaX Centre of Excellence (http://www.max-centre.eu/)

Responder a