Spin orbitals
In section 1.5.1 it was demonstrated that a product of spin orbitals,
a Hartree product, adequately describes a molecular system in which the
mutual repulsions of the electrons are neglected. Other than indicating
that a spin orbital is a one-electron wave function dependent upon both
the space and spin coordinates of the electron, section 1.5.1 states nothing
about the mathematical form of a spin orbital. A spin orbital, ,
consists of a spatial molecular orbital,
, multiplied by either the
or the
spin function to indicate
and
spin states respectively.
 |
(17) |
The variable
is an arbitrary spin variable commonly used for keeping track of integrations
over spin. The spin functions
and
are orthogonal; thus,
and
.
If the set of spatial orbitals are also orthogonal, then any two spin
orbitals generated from this set must also be orthogonal.
The spatial molecular orbital
contains all the information about the distribution of the electron in
space and is the only place where a set of variational parameters can
be included in a Hartree product wave function. Technically, the form
of the spatial orbital can be any mathematical function with a sufficient
degree of flexibility; however, theorists typically choose a wave function
based on a physical picture of the system. Relating the trial function
to the model of the physical system allows theorists to make useful chemical
interpretations of the trial wave function and many of the terms which
occur in the ab initio calculations. The most common method of
constructing spatial orbitals is to take a linear combination of atomic
orbitals to form the molecular orbitals with the variational parameters
being the linear combination coefficients, the so called LCAO-MO approach.
In ab initio methods two formalisms exist for combining the spatial
and spin functions: restricted and unrestricted. The restricted formalism
is closest to the standard picture taught in general chemistry where every
orbital contains both a spin-up and a spin-down electron, i. e. electrons
with opposite spin are selectively paired into a single spatial orbital.
In the unrestricted formalism,[8] electrons
with
and
spin are allowed to occupy completely separate spatial orbitals. The advantage
of the restricted formalism is that the wave functions which are constructed
in this fashion can be made to be eigenfunctions of the S2
operator. The unrestricted formalism has the advantage that there are
more variational parameters which allows for a lower expectation value
of the energy than the restricted formalism. However, the unrestricted
formalism is subject to ``spin contamination'' because the wave function
is not an eigenfunction of the S2 operator.
|