The meaning (and validity) of fractional occupations in QM

[Einstein Mcfly]

Hello all. As I understand it, there's somewhat of a divide in the scientific community (basically between chemists and physicist) around the topic of solutions to the SE (or KS eqn) that give fractional occupations of molecular or KS orbitals. I myself see no physical reason why probability can't be allowed to leak into a number of orbitals in quantities less than one, particularly if it gives a lower energy solution. Is there anything that I should understand in this regard (ie, that these solutions violate some key property of QM such as <S**2> not being a good quantum #)?

Thanks for your thoughts.

 

[alxm]

I'm a quantum chemist, and this is the first I've ever heard of such a thing. (Or indeed any kind of fundamental disagreement between chemists and physicists.)

You'll have to elaborate on what you mean; I don't see the problem. On the contrary, I don't know why anyone would expect integer occupancies for an interacting Hamiltonian in a single-particle basis.