4.

Crystal Field Theory

This theory largely replaced VB theory for interpreting the chemistry of
coordination compounds. It was proposed by the physicist Hans Bethe in
1929.
Subsequent modifications were proposed by J. H. Van Vleck in 1935 to allow
for some covalency in the interactions. These modifications are often referred
to as The Ligand Field Theory.
Ref. For a review on the evolution of bonding models see C. J. Ballhausen, J. Chem. Ed.
1979 56 194-197, 215-218, 357-361.

Pure crystal field theory assumes that the interactions between the metal ion
and the ligands are purely electrostatic (ionic). The ligands are regarded as
point charges. Although somewhat unrealistic, it uses symmetry
considerations which are valid for Ligand Field Theory as well as MO Theory.
It is of outmost importance to draw the d orbitals correctly in order to get an
idea of which of these orbitals will interact with the ligands (point charges).
The 5d orbitals in an isolated gaseous metal are degenerate. If a spherically
symmetric field of negative charges is placed around the metal, these orbitals
remain degenerate, but all of them are raised in energy as a result of the
repulsion between the negative charges on the ligands and in the d orbitals. If
rather than a spherical field, discrete point charges are allowed to interact with
the metal, the degeneracy of the d orbitals is removed (or, better said,
lifted).

The splitting of d orbital energies and its consequences are at the heart
of crystal field theory.

Octahedral Geometry

On going from a spherical to an octahedral symmetry, all d orbitals are raised

in energy, relative to the free ion. However, not all d orbitals will interact to the
same extent with the six point charges located on the +x, -x, +y, -y, +z and -z
axes respectively. The orbitals which lie along these axes (i.e. x2-y2, z2) will be
destabilized more that the orbitals which lie in-between the axes (i.e. xy, xz,
yz).

Referring to the character table for the Oh point group reveals that the x2-y2,
z2 orbitals belong to the Eg irreducible representation and xy, xz, yz belong to
the T2g irreducible representation.
The extent to which these two sets of orbitals are split (the eg and the t2g use

lower case for orbitals) is denoted by 0 or alternatively 10Dq. As the

barycenter must be conserved on going from a spherical field to an octahedral
field, the t2g set must be stabilized as much as the eg set is destabilized.

Illustration
[Ti(H2O)6]3+.
This is a d1 complex and the electron occupies the lowest energy orbital
available, i.e. one of the three degenerate t2g orbitals. The purple colour is

the result of the absorption of light which results in the promotion of this t2g
electron into the eg level.
t2g1eg0 > t2g0eg1
The UV-Vis absorption spectrum reveals that this transition occurs with a
maximum at 20300 cm-1 which corresponds to 0 243 kJ/mol.
(Remember that 1000 cm-1 = 11.96 kJ/mol or 2.86 kcal/mol or 0.124 eV.)

Typical 0 values are of the same order of magnitude as the energy of a

chemical bond.

What happens when more than 1 electron in d orbitals?

For d2-d9 systems the electron-electron interactions must be taken into
account (vide supra).
For d1-d3 systems, Hund's rule predicts that the electrons will not pair and
occupy the t2gset.
For d4-d7 systems, there are two possibilities:
Either put the electrons in the t2g set and therefore pair the electrons. This is
the so called low spin case or strong field situation.
Or put the electrons in the eg set, which lies higher in energy, but the
electrons do not pair. this is the so called high spin case or weak field
situation.

Therefore, there are two important parameters to consider

The Pairing energy (P)
and
the eg - t2g Splitting (referred to as 0, 10Dq)

For both the high spin (hs) and low spin (ls) situations, it is possible to
compute the Crystal Field Stabilization Energy as a function of electron
count and field strength.

The electron pairing energy P is composed of two terms

a) The Coulombic repulsion
This repulsion must be overcome when forcing electrons to occupy the same
orbital. As 5d orbitals are more diffuse than 4d orbitals which are more diffuse
than 3 d orbitals, the pairing energy becomes smaller as one goes down a
period.
As a rule, 4d and 5d transition metals are always low spin

b) The loss of exchange energy

The exchange energy which is konwn as Hund's rule is proportional to the
number of electrons having parallel spins. The greater this number, the more
difficult it is to pair electrons. Therefore
The d5 (Fe3+, Mn2+) configuration is the most enclined to form high spin
complexes

Tetrahedral Geometry
The tetrahedral symmetry can be derived from a cubic symmetry where only
four of the eight corners are occupied by point charges. In such a situation, it
is the xy, yz, xz orbitals which are destabilized as they point towards the
incoming point charges. the x2-y2 and z2 are stabilized so that the barycenter
is preserved.

Please note that the irreducible representations are t2 and e in the Td point
group.
The crystal field splitting in a tetrahedral symmetry is intrinsically smaller than
in the octahedral symmetry as there are only four ligands (instead of six
ligands in the octahedral symmetry) interacting with the transition metal ion.
The point charge model predicts that

t = 4/9 o

As a result, low spin configurations are rarely observed. Usually, if a strong

field ligand is present, the square planar geometry will be favoured.

Square Planar Geometry

In order to derive the square planar geometry, let us consider a tetragonal
distortion.
Removing both point charges from the z axis, stabilizes the z2 orbital of the eg
set. As the bonding of the remaining four point charges increases the x2-y2
orbitals rises in energy.
Again here, as the symmetry is lowered, the degeneracy is lifted and the
orbitals have different irreducible representations.

Other Geometries
orbital splitting patterns for various relevant geometries can be derived with
the crystal field theory

Factors which influence the magnitude of

a) Oxidation state of the metal

The higher the charge of the metal, the greater

. As a rule of thumb

increases by ca. 50% when the oxidation state increase by one unit.
b) Nature of the metal ion

increases on going from 3d < 4d < 5d

(ca. 50% going from Co to Rh. ca. 25% going from Rh to Ir.)
c) Number and geometry of the ligands

0 is ca. 50% larger that t.

d) Nature of the ligands

Based on similar data for a wide variety of complexes, it is possible to list

ligands in order of increasing field strength in the so-called spectrochemical
series

I- < Br- < S2- < SCN- < Cl- < N3-, F-< urea, OH- < ox, O2- < H2O <
NCS- < py, NH3 < en < bpy, phen < NO2- < CH3-, C6H5- < CN- < CO.

It is possible to arrange the metals according to a spectrochemical series as

well. The approximate order is

2+

< Ni2+ < Co2+ < Fe2+ < V2+ < Fe3+ < Co3+ < Mn3+ < Mo3+ < Rh3+ <
Ru3+ < Pd4+ < Ir3+ < Pt4+
Mn

Jorgensen has proposed a formula which allows to estimate the

f g
where f is a function of the ligand and g is a function of the metal

Applications of Crystal- and Ligand Field Theory

Ionic Radii. For a given oxidation state, the ionic radius decreases steadily on
going from left to right in a transition series. Populating antiboding orbitals (i.e.
filling the eg levels in an octahedron) leads to an increase in ionic radius.
Therefore, the ionic radius depends on the spin state of the metal (i.e. high
spin or low spin).

Hydration Enthalpy. Let us look at the variation of enthalpy of M2+ ions (NB.
Since water is a weak field ligand, the complexes are high spin).

M2+(g) + 6 H2O(l) = [M(O2H)6]2+(aq)

Plotting the enthalpy across the first transition series

This double-humped pattern is very frequent when properties are plotted for
transition metals acrsoss a transition series.
The straight line (full black circles) is obtained by substracting the CFSE for
the given electron count. The linear increase in stability on going through a
transition series is caused primarily by the increasing acidity (largely due to a
decreasing size of the metal cation and electrostatic effects). This forms the
basis of the Irving-Williams series.
The Irving-Williams series states that for a given ligand, the stability of the
complexes increases from Ba to Cu (and then drops for Zn). It should be
noted however that as one moves towards the right, the late transition metals
prefer softer ligands. (see the HARD-SOFT ACID-BASE Theory).

Geometry Preference as a Function of Electron Count. If one keeps in

mind that the t is about half that of o, it is possible to predict qualitatively
which electron counts should favour which geometry.

Crystal Field Theory (CFT), An Introduction.

Lecture 3. CHEM1902 (C 10K) Coordination Chemistry
In the ionic CFT, it is assumed that the ions are simple point charges. When applied
to alkali metal ions containing a symmetric sphere of charge, calculations of energies
are generally quite successful. The approach taken uses classical potential energy
equations that take into account the attractive and repulsive interactions between
charged particles (that is, Coulomb's Law interactions).
Electrostatic Potential is proportional to q1 * q2/r
where q1 and q2 are the charges of the interacting ions and r is the distance separating
them. This leads to the correct prediction that large cations of low charge, such as K+
and Na+, should form few coordination compounds.
For transition metal cations that contain varying numbers of d electrons in orbitals
that are NOT spherically symmetric, however, the situation is quite different. The
shape and occupation of these d-orbitals then becomes important in an accurate
description of the bond energy and properties of the transition metal compound.
To be able to understand and use CFT then, it is essential to have a clear picture of the
shapes (angular dependence functions) of the d-orbitals.

A superb graphical approach using CHIME has been developed by John Nash at
Purdue University for showing the spatial relationship of the different orbitals in
different stereochemistries. (Look for the pages under orbital orientation.) Since the
Gaussian surface files used in these displays are very large, a local copy of these
pages has been authorised for use by Mona students at our Intranet site.
A cube provides a convenient reference for displaying the coordination centre of
complexes since octahedral compounds have cubic symmetry i.e. the six anions sit at

the centres of the faces of the cube and for tetrahedral complexes the 4 ligands sit at
opposite diagonal edges.
You should be able to see that two of the d-orbitals, the dz2 and dx2-y2 meet the faces
of the cube, but the remaining three (dxy, dyz and dxz) point towards the edges of the
cube and actually have a node meeting the centres of the faces.
What happens then to the energy of the d-electrons if negatively charged ligands are
brought in towards the metal centre to give an octahedral complex - where the ligands
sit at the centre of the faces of the cube i.e. on the X, Y and Z axes? Since the d
electrons are themselves negatively charged they would experience repulsion and
their energies would be raised. For 's' and 'p' electrons this is of little consequence but
for 'd' electrons there will be an energy difference involved depending on the orbital
distribution and occupation.
Consider the simple example of TiCl63- in which six chloride ions octahedrally
surround the Ti3+ cation. There is only one d-electron to be allocated to one of the five
d- orbitals. If it were to occupy the dz2 or dx2-y2 orbital, both of which meet the face
of the cube and thus point directly towards the chloride ligands, it would be strongly
repelled. The geometry of these orbitals and their nodes would require the electron to
stay near the negatively charged ligands causing even more repulsion than a
spherically distributed electron would experience. On the other hand, if the electron
were to occupy the dxy, dyz or dxz orbital, it would spend less time near the ligands
than would a spherically distributed electron and would be repelled less.
This difference between for example the dx2-y2 orbital, the dxy orbital and a
spherical distribution can be graphically represented by

The result of these differences for the d-orbitals is an energy difference between the
dz2 and dx2-y2 orbitals compared to the dxy, dyz and dxz orbitals.

The CFT approach can be easily extended to other geometries and the next most
important case is the tetrahedron. To predict the splitting pattern of the d-orbitals
under a tetrahedal crystal field you may once again find it convenient to consider how
the ligands can fit into a cube to give a tetrahedron.
The next step is to consider how the d-orbitals interact with these incoming ligands.
Once again, the pages at Purdue should be viewed or (local copy for UWI, Mona
students). For tetrahedral complexes, those orbitals which point towards the edges
should now be raised to energy higher than those which point towards the faces. That
is, the exact opposite of the situation we just dealt with for the octahedral crystal field.
The end result is a splitting pattern which is represented in the splitting diagram
above.
The square planar case can be considered as an extension of the octahedral, where
we remove the two ligands from the Z axis. Consequently, repulsion of an electron in
the dz2 orbital will no longer be equivalent to that experienced by an electron in the
dx2-y2 orbital, and the end result is shown above. For this course, the only square
planar complexes will be for d8 complexes, i.e all four coordinate complexes are
tetrahedral except for d8 which may be tetrahedral or square planar.
Once we accept that the 5 'd' orbitals are no longer degenerate in a coordination
compound we can begin to explore some of the implications. Note that a different
CFT energy splitting diagram has to be applied for each stereochemistry. In this
course we will only be concerned with diagrams for octahedral, tetrahedral and
square planar complexes.
One of the important aspects of CFT is that not all ligands are identical when it comes
to causing a separation of the d-orbitals. For transition metal compounds, there is
clear evidence for this from the multitude of colours available for a given metal ion
when the ligands or stereochemistry are varied. For octahedral complexes this is a

reflection of the energy difference between the higher dz2, dx2-y2 (eg subset) and the
dxy, dyz, dxz (t2g subset).
It has been established that the ability of ligands to cause a large splitting of the
energy between the orbitals is essentially independent of the metal ion and the
SPECTROCHEMICAL SERIES is a list of ligands ranked in order of their ability
to cause large orbital separations.
A shortened list includes:
I- < Br- < SCN- ~Cl- < F- < OH- ~ ONO- < C2O42- < H2O
< NCS- < EDTA4- < NH3 ~ pyr ~ en < bipy < phen < CN- ~ CO
When metal ions that have between 4 and 7 electrons in the d orbitals form octahedral
compounds, two possible electron distributions can occur. These are referred to as
either weak field - strong field or high spin - low spin configurations.
Take for example, Fe2+- d6.

The diagram on the left represents the case for the aqua ion (small ) and on the right
that of the hexacyano ion (large ).
What this means is that if we use a technique that can detect the presence of unpaired
electrons in each compound, then for the first we should find 4, while in the latter
none. This accounts for the names high spin compared to low spin. The terms weak
field, strong field give an indication of the splitting abilities of the ligand. Water
always give rise to small splittings of d orbitals for first row transition metal ions and
hence is referred to as a weak field ligand. Conversely, CN- is a strong field ligand,
since it causes large splittings of the d-orbitals.
A simple JAVA applet that shows the electronic configuration for octahedral and
tetrahedral complexes and calculates the spin-only magnetic moment is available.
One method of determining the number of unpaired electrons is by looking at the
magnetic properties of the compounds. A simple technique to determine a magnetic
moment (the GOUY METHOD) involves weighing the sample in the presence and
absence of a strong magnetic field. By careful calibration using a known standard,
such as HgCo(SCN)4 the number of unpaired electrons can be determined.
To predict the magnetic moment, we can use the simple spin-only formula:

[4S(S+l)]

Bohr Magneton

(BM)

where S is the spin quantum number (1/2 for each unpaired electron).
An alternative representation is:

[n(n+2)]

Bohr Magneton

(BM)

where n is the number of unpaired electrons.

These simple formulae give good results for most first row transition metal
compounds and can be refined to include orbital contributions (COVERED IN MORE
DETAIL IN CHEM2101 (C 21J) ).
It is expected that all students will be able to calculate the spin-only magnetic moment
of any coordination compound. To do this, you must know the number of d-electrons
in the central metal ion, the stereochemistry and whether weak field-strong field
considerations are necessary.

metal ion
number of d electrons
stereochemistry
High Spin/Low Spin
# of unpaired electrons
magnetic moment

K3[Fe(oxalate)3] 3H2O
Fe3+
5
octahedral
High Spin
5
(35)

K2CuCl4
Cu2+
9
tetrahedral
Not relevant
1
(3) B.M

Note that if there are 1-3 or 8-9 d electrons in an octahedral complex, the spin-only
magnetic moment will have the same value irrespective of whether the ligands present
are considered weak field or strong field. For octahedral Co(III) complexes, we will
make the simplification that all are low-spin, so regardless of the types of ligands
present the magnetic moment will be =0, that is, diamagnetic.
Other examples can be found in the Tutorial paper for this course.