Chemistry Reference
In-Depth Information
3.3.3
Ligand Field Theory - Making Compromises
A problem we left a little while ago in CFT was how to explain the experimentally observed
order of ligands in the spectrochemical series, and issues such as water being a better ligand
than hydroxide and the neutral carbon monoxide being such a strong ligand. Molecular
orbital theory has given us another perspective on the d-orbital set, with the
t
2g
level seen as
dominantly nonbonding in character (which means that its energy should not be influenced
particularly by changing ligands) and the
e
g
∗
level thus carrying the responsibility for
altering the size of
o
as it responds to the presence of different ligands. Obviously, with
just two energy levels involved, increasing the size of
o
can be done in two ways - either by
raising the energy of the
e
g
∗
level or else by lowering the energy of the
t
2g
level. Any action
in the latter direction seems at odds with the
t
2g
level being nonbonding in character, or
having little to do with the ligands. The crystal field model focused on raising or lowering
of the
e
g
level through electrostatic interactions between ligands and metal d electrons.
Fortunately, the molecular orbital theory can provide us with a means for understanding
how the
t
2g
level may also be influenced and its energy manipulated by the ligands. If
we return to the spatial orientation of metal orbitals in the
t
2g
set, we are reminded that
they point not along the axes associated with ligand positioning, as for the 3d
e
g
∗
orbitals
or even 4s and 4p orbitals, but between the principal axis directions. Therefore, they can
have little to do with traditional
-bond is defined as one
where electron density is enhanced in a direct line joining the two atom centres, i.e. metal
and donor atom in the present case). The key to any interaction involving
t
2g
d orbitals is
whether the ligand donor has p or even
-bonding (as, by definition, a
orbitals directed orthogonal (sideways-on) to the
metal-donor bond direction. If this does occur, a further interaction between metal
t
2g
d
orbitals and ligand donor p (or
) orbitals can arise. In the usual MO manner, interaction
of a single d orbital and a single p orbital would lead to two MOs, one bonding (
,or
∗
, or out-of-phase) orbital of lower and higher energy
than the parent orbitals respectively - a mechanism for manipulating the energy of the
t
2g
set has thus been established, and one that depends on the properties of the ligand. This is
developed and exemplified in Section 3.3.4.
This capacity to model and account for the interaction of ligands capable of additional
in-phase) and one antibonding (
-bonding with metals provides an enabling mechanism for dealing with the experimental
spectrochemical series positioning of ligands. Of course, not all ligands will have the
capacity to undergo further
-type interactions with metal orbitals. For example, ammonia
has but one lone pair directed to
-bonding, and otherwise three internal N
H
-bonds;
it cannot undertake any further bonding. However, carbon monoxide (C
O) has an array
of
-bonds resulting from its multiple bond character, and is a clear candidate for further
-acid ligand, a group of key
importance in organometallic chemistry, with empty orbitals of symmetry appropriate for
overlap with a filled d
,or
t
2g
, metal orbital). The scene has now changed, so that we have
greater capacity to understand and predict the effects of ligands on the chemistry of metal
complexes - this is the strength of the LFT.
What we have in the end is a reasonably consistent set of models, but ones that differ
in their focus and assignment of importance to electrostatic and covalent character. What
is the 'real' situation, and how can we effectively assess the contribution of each compo-
nent? A key and indeed fairly simple approach to this came forward from Pauling, who
asserted that metals and ligands would adopt as far as practicable nett charges close to
zero, through metal-ligand bond polarization or
-type interactions (it is an example of a
-acceptor or
-type metal-to-ligand or ligand-to-metal
