[salmon-user:00341] Calculatons for hexagonal CdS

[Ekaterina Migal]

Dear salmon users and developers,

I am trying to make ground state calculations for hexagonal CdS. I have made the following steps:

1) generate rectangular unit cell of hexagonal CdS with cif to salmon (https://github.com/uemoto1/salmon_inp/) (cif files from crystallography open database https://www.crystallography.net/cod/1011054.html)

2) download pseudopotentials for LDA (Fritz-Haber-Institute (FHI) pseudopotentials from https://www.abinit.org/miscellaneous.html   
3) manually change parametres of the unit cell (nelec and nstate) and r & k grids
However, calculations do not converge, band gap is not calculated at all (I get Gap = ***).

Do you have any ideas what is wrong? I would appreciate any help.

Sincerely yours,
Ph.D. Ekaterina А. Migal.
E-mail: migal...@list.ru, ea.m...@physics.msu.ru 
Phone: +79175350165 
Physics Department and International Laser Center
M.V. Lomonosov Moscow State University

[Yamada Shunsuke]

Dear Migal-san,

Thank you for the question. 

The parameters 'nstate' and 'nelec' may be too small. From the 2^nd line of the *.fhi files, the number of valence electrons are 6 and 12 for S and Cd, respectively.

Thus, the total number of electrons in the cell (i.e. nelec) is equal to 72 (=4*6+4*12). For spin-unpolarized systems, the number of states for calculation (=nstate) should be greater than nelec/2 for convergence.

Sincerely,

Shunsuke

---

差出人: Ekaterina Migal <ea.m...@physics.msu.ru>
送信日時: 2025年5月12日 16:54
宛先: salmo...@salmon-tddft.jp <salmo...@salmon-tddft.jp>
件名: [salmon-user:00341] Calculatons for hexagonal CdS

 

[Ekaterina Migal]

Dear Yamada-san,

Thank you for your reply.

In that case, calculations are "Killed". What does that mean? 

пн, 12 мая 2025 г. в 11:56, Yamada Shunsuke <yamada....@qst.go.jp>:

--

С уважением, 
к.ф.-м.н. Екатерина Александровна Мигаль. 

[Yamada Shunsuke]

Dear Migal-san,

I think that is due to insufficient memory. For such a large nstate, MPI parallelization with memory distribution may be required.

Sincerely,

Shunsuke Yamada

Kansai Institute for Photon Science, QST, Japan

山田俊介

量子科学技術研究開発機構 関西光量子科学研究所

---

差出人: Ekaterina Migal <ea.m...@physics.msu.ru>
送信日時: 2025年5月13日 2:44
宛先: salmo...@salmon-tddft.jp <salmo...@salmon-tddft.jp>
件名: [salmon-user:00343] Re: Calculatons for hexagonal CdS

 

[Ekaterina Migal]

Dear Yamada-san,

Could you give any approximate estimation of how much memory is required for such calculations?

Kind regards,

Ekaterina Migal.

вт, 13 мая 2025 г. в 03:17, Yamada Shunsuke <yamada....@qst.go.jp>: