RPA forces

[kdoblhoff]

Dear Vasp developers,
I wanted to ask about the status of RPA forces for metallic systems (finite temperature algorithm). I vaguely remember someone saying a while ago that those were under development. On the website, it still states that forces are not yet available for metallic systems. I assume that this is still the current status, right? (At least I am not aware of any publication related to this.)
Thank you for your reply,
Best regards,
Katharina

[merzuk.kaltak]

Dear Katharina,

Yes, you are right. RPA forces for metallic systems are still under development and there is no publication about this topic yet.
However, you can use the code in vasp-6.3.2 by setting LRPAFORCE =T and LFINITE_TEMPERATURE=T in the INCAR. Note, latter requires ISMEAR =-1

[kdoblhoff]

If I correctly understand your comment you imply that "at own risk" it can be used for tests, right?

[fabien_tran1]

Yes. There is still some possible uncertainty with the correctness of the implementation of RPA forces for metals.

A good idea would be to check that the geometry obtained after optimization is really a minimum by displacing the atoms (and hopefully observe an increase of the total energy). However, such a check is of course possible only for very simple systems.

Note that there is a test for RPA forces for a metal in the testsuite (FeAl_333_RPAFORCE).