Dear QE developers, <br> I have a very naive question. It seems that there are two ways to calculate the charges on atoms in QE. One is to use the Lowdin orbitals to do the atomic projection and get the charges on different orbitals. The other way is to turn on Hubbard U and calculate the occupation on the localized orbitals. Recently, I tested LaNiO3 and found the two methods give different results of the charge on Ni d orbitals. I would like to know whether this is normal and if yes, is that because the Ni d orbitals are different in the two approaches?<br>
Thank you very much.<br><br>Hanghui Chen<br>Department of Physics<br>Yale University<br>