GGA+U recommends the U value of rare earth elements

Materials Project
1

Hello everyone,

I am going to use GGA+U correction method to study ABO3 compounds, where the A site is a rare earth element and the B site is a transition metal. I searched the Materials Project Documentation and found that the U values of some elements were given in the following webpage:

Hubbard U Values | Materials Project Documentation

However, these elements only included transition metals and did not contain rare earth elements. Are there any recommended U values for rare earth elements on other websites of the materials project or research articles in authoritative journals?

Any assistance or clarification would be greatly appreciated!

2

I have used U values from 5 eV to 7.5 eV. Please see this paper about YbPtBi (https://journals.aps.org/prb/pdf/10.1103/PhysRevB.108.125110) and supplementary of this paper about CeCu2Si2 (https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.127.067002#supplemental). If you have experimental results such as ARPES spectra to compare with, it will perhaps be more precise.

3

Hi @kong_kong, you might want to take a look at performing self-consistent GGA +U calculations, described here for the d-block and here for the theory.

Like @fff said, you can always fit a U to an experimentally known quantity, but the self-consistent procedure is more theoretically rigorous and can be used in the absence of reference data.

4

OK, thank you very much for your answer！

5

Ok, I have understood, I will try more U to get close to the experimental value, Thank you for answering my question!