On specifying the spin occupations of the quantum chemistry problem.

[jiangtong1000]

Reno previously specified the total electron number qntot, for constructing the quantum chemistry model. The model then explored the ground state within the confines of this qntot space. Now, it requires the specification of the qn for spin-up and spin-down electrons.

A question arises: in the case of a given system, it's not common to have prior knowledge about the spin configurations of its ground state. Does the ground state has to be half up, half down? No, right? I was thinking if the original feature should be retained.

[liwt31]

Hi Dr. Jiang. I believe that currently there's no obstacle to specifying only one quantum number. in qc_model, two quantum numbers are assumed. Maybe an option can be added to the conserve_qn argument for the case that only the total number of electrons is conserved.

On the other hand, I think it is quite common to have prior knowledge about the spin configurations of its ground state. Just think of a PySCF or Gaussian calculation, you always have to specify the spin configuration, right?

Cheers

[jiangtong1000]

Oh no, my secret identity has been revealed. No need for the formal title next time lol.