Some concept need clarifications.

maximos

During my my reading of Multiwfn manual, i faced some concept that i can't understand :
1-Promolecular densities approximation used in NCI analysis.
2-frozen core approximation
3-fuzzy atomic spaces
Of course, i did my research on google and i read some papers, but i still can't understand any thing.
I hope i can find some clarifications for a beginner in this field.


Many thanks in advance,
Maximos

sobereva

1 Namely using promolecular density for NCI analysis. Usually, NCI analysis is based on the electron density calculated by electron wavefunction, which is usually generated by quantum chemistry calculation. However, this task is quite expensive for huge system. Using promolecular density instead is an alternative choice, this kind of density is simply constructed by superposition of spherically averaged electron densities of atoms in their isolated states, clearly calculating promolecular density is significantly cheaper than calculating density using wavefunction, especially for very large systems (e.g. fragments of biological macromolecules). Multiwfn contains a built-in library of atomic electron density for constructing promolecular density. Of course, the quality of NCI analysis based on promolecular approximation is poorer than common NCI analysis.

2 In many kinds of quantum chemistry calculations involving multiple Slater determinants (SDs) or configurational state functions (CSFs), such as MP2, CISD, CCSD(T), TDDFT, frozen core approximation makes inner-core molecular orbitals do not participate in construction of excited SDs or CSFs, this reduces cost notably while commonly doesn't hurt accuracy evidently. If you are still confused, please read quantum chemistry books such as "Introduction to Computational Chemistry" (Frank Jensen).

3 Please check introduction given in Section 3.18.0 of Multiwfn manual.