XII. ML₅ Complexes

In this Chapter we are going to build up the important valence orbitals of ML₅ and use them both as a molecule as well as a molecular fragment.. The approach we are going to take for this is a deconstructionist one. We are going to see what happens to the MOs of an octahedron when one ligand is removed:

The ML₅ orbitals then can be used as a starting point to get to the D_3h trigonal bipyramid or a number of other geometries. Just as we saw a relationship between the octahedron and the square plane there will also be a relationship between the octahedron and the ML₅ unit. Furthermore there is a relationship between this fragment and one where one ligand is removed from the square plane to give a C_2v ML₃ fragment. finally removing two and three ligands from the octahedron and square plane generates the other fragments shown below:

Therefore, it is very important to see where these relationships come from and how many electrons are contained within these sets of "important valence orbitals".

A. The C_4v ML₅ fragment

The removal of one ligand from an octahedron is shown below.

As shown there is only one change! The z² MO loses nearly 1/2 of it's antibonding so it is stabilized. Furthermore, it is hybridized by mixing some of the higher-lying 2a_1g and one component of 2t_1u MOs (see previous chapter) into it in a bonding way:

Contour plots of these MOs are shown below.

Notice that the CO p* orbitals mix extensively into these MOs. For this molecule there is another geometrical degree of freedom which is important to consider - the apical-metal-basal bond angle, plotted as q in the Walsh diagram on the next page. I've listed e(1) and e(2) terms for each MO.

e(1) = 0; e(2) = (+)

e(1) = (-); e(2) ~ 0

e(1) = 0; e(2) = 0

So the population of a₁ or b₁ favors q > 90· whereas population of the e set favors q = 90·. Some real molecules are:

molecule dⁿ MO configuration q

Cr(CO)₅ 6 (a₂)²(e)⁴ 90.0·

Ni(CN)₅^3- 8 (a₂)²(e)⁴(a₁ )² 101·

Co(CN)₅^3- 7 (a₂)²(e)⁴(a₁ )¹ 97.6·

dexyhemoglobin 6 (hs) (a₂)²(e)²(a₁ )¹(b₁)¹ 110·

For the hemoglobin example there are four heme units in it and this in turn is an Fe(2+) complex. Each heme consists of a phorphorin coordinated to the iron-

And there is also an imidazole ring coordinated in the apical position which is part of an amino acid that forms the protein backbone. The coordination geometry around the iron is then given by:

The reaction of O₂ with one heme is remarkable. First there must be a spin change since the resultant oxygenated heme is a singlet (recall O₂ is also a triplet). Secondingly the iron atom moves into the plane of the phorphorin and with it the imidazole ring and protein backbone undergoes a series of geometrical changes.

This structural distortion in the protein opens channels so that diffusion of the three other O₂ molecules is greatly facilitated. CO and HCN are much stronger s-donors so they bind more strongly to the Fe(2+) center...

B. Dimers

In a general sense one should be somewhat worried about what ligands are coordinated to the metal and what is the M-M distance in the dimer. Here are two extreme examples:

The molecule on the left has a bond order of one, whereas, that on the right has a bond order of FOUR!

Lets take two examples where there is an atom between the two metals and investigate the geometrical consequences.

1. The (OC)₅M-H-M(CO)₅ molecules

There are a series of (OC)₅M-H-M(CO)₅^- molecules where M = a d⁶, group 6 transition metal (Cr, Mo, and W).

The interesting thing is that the structures show a wide range of M-H-M bond angles - from 180· - 123·.

# M M-H-M M-H M-M

1. Cr 180·

2. Cr 158· 1.77 3.31

3. Cr 145· 1.73 3.29

4. Mo 136· 3.42

5. W 123· 1.90 3.34

This is an electron deficient three center - two electron bond.

2. Mixed valence complexes.

Here a halide lies between the two metals. Our concern is whether the halide, X, lies exactly midway between the two metals - a delocalized solution - or whether the X lies much closer to one metal than it does to the other - the mixed valence state, i.e.

The basic interaction diagram can be easily constructed:

We now carry out the distortion to an asymmetric geometry looking at the s_g* and s_u* MOs in particular:

The s_u* MO mixes into the lower s_g* MO stabilizing it when the geometry is shifted from the middle, symmetrical position. Likewise, s_g* mixes into and destabilizes the upper s_u* MO . The lower s_g* MO becomes concentrated on the five coordinate atom.

And the upper s_u* MO becomes the z² M-L antibonding orbital:

So for the d³/d⁴ case one electron is transferred in an opposite direction to the movement of the bridging halide -

This forms the basis for the inner-sphere electron transfer reaction. An example is given by:

C. Distortion to the D_3h trigonal bipyramid.

A Walsh diagram can be constructed by:

The e' set in the trigonal bipyramid deserves some comment. It is d in character and somewhat antibonding to the equatorial ligands. However, metal p mixes into it also to hybridize it away from the ligands. So e' stays at moderate energy. When p-acceptors are present they will also mix strongly into e'. An example is provided by Fe(CO)₅.

There are some preference wish favor D_3h versus C_4v in ML₅:

# d e's shape

0 - 2 C_4v

4 D_3h

6 C_4v

8 D_3h or C_4v