Where to find LDA and hybrid pseudopotentials

External optional components of Abinit (BigDFT, Wannier90...) and codes using/providing data from/to Abinit (AtomPAW, ONCVPSP, EXC, DP, Yambo...)

Moderators: ebousquet, bxu

Locked
erichmond
Posts: 26
Joined: Fri Dec 23, 2016 12:00 am

Where to find LDA and hybrid pseudopotentials

Post by erichmond » Thu Mar 26, 2020 4:26 am

Somewhere in the ABINIT web-pages I read that one should not use Psps_for_tests pseudopotentials for production runs. With this in mind, I have been using the GGA pseuidopotentials in the ONCVPSP file set for my production runs to calculate the high precision work function of transition metals; but I would like to try a LDA or hybrid pseudopotential in my calculations. Where can I find these LDA and hybrid pseudopotentials?

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Where to find LDA and hybrid pseudopotentials

Post by ebousquet » Mon Mar 30, 2020 8:51 am

Dear Eric,
You can use the pseudos from pseudodojo projetc:
http://www.pseudo-dojo.org/

Or from the JTH table if using PAW:
https://www.abinit.org/downloads/PAW2

There is no Hybrid pseudopotentials at this stage, others can correct me if there are some alpha/beta available.

Best wishes,
Eric

erichmond
Posts: 26
Joined: Fri Dec 23, 2016 12:00 am

Re: Where to find LDA and hybrid pseudopotentials

Post by erichmond » Fri Apr 10, 2020 9:40 pm

Eric,
Thank you for your reply, but I have another question about the JTH-1.1 pseudopotential files. There are 3 folders within this folder - ATOMIUCDATA, ATOMICDATA-SP, and InputFiles. Would you please explain the difference between the pseudopotential files in these 3 folders.

About hybrid pseudopotentials, ixc = -402, -406, -427, -428 are listed as different hybrid XC. Are these still under development or are they usable now.
Eliezer

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Where to find LDA and hybrid pseudopotentials

Post by ebousquet » Wed Apr 15, 2020 6:29 pm

Dear Eric,
The folder ATOMIUCDATA contains all the pseudo files you can use. The folder ATOMICDATA-SP contains some elements with more semi-core states included in the valence, it is up to you test if you need them or not. The input folder contains all the input file for the generator, in case you want to change the pseudos yourself.

Regarding the Hybrid functional, I don't know, in the release note of the recent beta v9.0 there is no updtate on that:
https://docs.abinit.org/about/release-notes/
Lets see if somebody else can say something?

Best wishes,
Eric

erichmond
Posts: 26
Joined: Fri Dec 23, 2016 12:00 am

Re: Where to find LDA and hybrid pseudopotentials

Post by erichmond » Sun Apr 19, 2020 11:00 am

Thank you Eric.

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Where to find LDA and hybrid pseudopotentials

Post by ebousquet » Mon Apr 20, 2020 7:55 am

Dear Eric,
Sounds like the options -406, -456, -427 et -428 are working but not yet very optimized...
Best wishes,
Eric

erichmond
Posts: 26
Joined: Fri Dec 23, 2016 12:00 am

Re: Where to find LDA and hybrid pseudopotentials

Post by erichmond » Wed May 20, 2020 6:02 am

Eric,

What do you mean by "not very optimized".

Eliezer

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: Where to find LDA and hybrid pseudopotentials

Post by ebousquet » Sun May 24, 2020 7:04 pm

erichmond wrote:
Wed May 20, 2020 6:02 am
What do you mean by "not very optimized".
Might be more SCF or memory demanding...
Eric

erichmond
Posts: 26
Joined: Fri Dec 23, 2016 12:00 am

Re: Where to find LDA and hybrid pseudopotentials

Post by erichmond » Wed May 27, 2020 6:15 am

Eric,

Thanks for the information and explanations.

Eliezer

Locked