Hey Ehan Hemmasi

One of my favorite databases for pseudopotentials is PseudoDojo (https://www.pseudo-dojo.org/). But yeah, you’ve also got some other options like:

https://dalcorso.github.io/pslibrary/

https://discover.materialscloud.org/sssp-old/v0.7/

https://www.physics.rutgers.edu/gbrv/

https://www.materialscloud.org/discover/sssp/table/efficiency

Hope that helps,

Cheers

Gabriel Vinicius

The Quantum Espresso website provides pseudopotential files for most of the elements.

https://pseudopotentials.quantum-espresso.org/legacy_tables/ps-library/ce

##############################

Dear Ehan,

most standard pseudopotentials for Ce treat 4f electrons as part of the core. They dont explicitly include them in the valence ... so applying Hubbard U on 4f orbitals causes errors or nonsensical results.

You need a fully relativistic/scalar rlativistic pseudopotential including 4f elctrns in the valence; norm-conserving/ultrasoft/PAW ps-potential, depending on what you prefer and QE version compatibility. Also you have generated with the Hubbard correction in mind...so DFT+U-ready.

PseudoDojo provides hq norm-conserving pseudopotentls that often include 4f orbitals explicitly.

Use ld1.x in QE to generate a pseudopotential for Ce with 4f in valence explicitly included.

For the valence configuration, set something like Ce 5s2 5p6 5d1 6s2 4f1

So make sure you use relativistic or scalar-relativistic generation&choose an appropriate cutoff radius ensuring the 4f states are well described.

!!! Using Hubbard U

Ensure your input file specifies the correct atomic species/their Hubbard parameters:

&Hubbard

Hubbard_U(Ce) = 5.0

/

##############################################

Hope, it helps! :-}