A difficulty will already have become apparent: the solution of the Schrodinger

equation for is so complicated (even after making the Born-Oppenheimer

approximation) that there can be little hope that exact solutions will be found for
more complicated molecules. Therefore, we must resort to another approximation,
but use the exact solutions for as a guide. Another reason why making a further
approximation is quite sensible is that we already have available quite good atomic
orbitals for many-electron atoms, and it seems appropriate to try to use them as a
starting point for the description of many-electron molecules built from those atoms.

There are two major approaches to the calculation of molecular structure, valence
bond theory (VB theory) proposed by Heitler and London (1927) latter extended by
Linus Pauling, and molecular orbital theory (MO theory) by Friedrich Hund and
Robert S. Mulliken (1932).

The theories have their own unique approach to determine bonding and thus the
overall structure of a molecule.
Molecular orbital theory assumes that individual electron pairs are found in molecular
orbitals that are distributed over the entire molecule and every electron contributes
to the strength of every bond. Bonding electrons are delocalized throughout the
molecule.
In valence bond theory, a bond is regarded as forming when an electron in an atomic
orbital on one atom pairs its spin with that of an electron in an atomic orbital on
another atom. Bonding electrons are localized in the midway of two atoms.