depends on your system, is it magnetic ?
metallic aluminum or copper have spin compensated densities and you can use nspin=1.
nickel or iron are magnetic and nspin=2 is needed.

On 10/08/2017 08:29, Iryna Zaporozhets wrote:
Dear PW community,

I am working with an open-shell system which requires a lot of computational efforts. Is it possible to use other options than spin-polarized calculations (nspin = 2) to reduce computational cost, for example, restricted open-shell DFT? Is there any physical meaning to the results obtained with spin-unpolarized (nspin = 1) calculations in case of the open-shell system?

Thank you,
Iryna Zaporozhets,
Master student in Chemistry
Kharkiv National University, Ukraine

Pw_forum mailing list

Pw_forum mailing list

Reply via email to