Chapter XIII. C_2v ML₃, ML₄ and ML₂

In this Chapter we are going to generate the important valence orbitals for three common fragments and focus on their inter-relationships along with metal-olefin bonding.

A. The C_2v ML₃ Fragment

One orbital is stabilized, just like for ML₅. That orbital is constructed as:

Contour plots for the orbitals of PtCl₃^- are:

Notice that the Cl p AOs mix in an antibonding way - they are p-donors.

The important valence orbitals of the C_2v ML₃ fragment were generated from removing one ligand from the square plane and those for a C_4v ML₅ fragment from removing a ligand from the octahedron. Since there is a relationship between the orbitals of the octahedron and the square plane, it is not surprising that there is a relationship between the

MOs of the two fragments. How this is so is illustrated below.

So a d⁶ ML₅ fragment is related to a d⁸ ML₃ one. A couple of examples will illustrate this.

Notice in both cases there is donation from ethylene p MO to an empty metal d orbital and back donation from a metal d orbital to ethylene p*.

Dick Schrock's group at MIT made a number of tungstenacyclobutadiene complexes. They have the structure shown below.

Cyclobutadiene it self is extremely reactive and has singlet and triplet electronic states that are accessible. On the other hand, Schrock's molecules are quite stable and always are singlets.

This is the full interaction diagram -

This can be divided into a s and a p system. Let us start with the s portion.

Very simply this has established the splitting pattern for a trigonal bipyramid. The p system adds some complexity-

B. The C_2v ML₄ Fragment

We can again use a decomposition from an octahedron; this time we remove two cis ligands. Consequently there should be two empty hybrids that form. Likewise removal of two cis ligands from a square planar complex generates the C_2v ML₂ fragment that we will consider next.

C. The C_2v ML₂ Fragment

There is a definite relationship between the two fragment:s:

C. Metal-olefin Complexes

We are going to develop the bonding in ethylene-Fe(CO)₄ and ethylene-Ni(PH₃)₂. There is a further conformational issue to be investigated:

In both conformations the 1a₁, 2a₁, and p orbitals form a three-orbital pattern. The bonding and nonbonding MOs are filled. Furthermore, they are at the same energy in both conformations. The reason for this is that they are all cylindrically symmetric, therefore, their overlaps, i.e. <1a₁|p> and <2a₁|p>, are constant with respect to ethylene rotation. This is not the case for ethylene p*. When it lies in the equatorial plane it has b₂ symmetry and when it is perpendicular to this it has b₁ symmetry. The question then is the p*+b₂ interaction stronger than the b₁+p* one?

Therefore, for energy gap and overlap reasons the p*+b₂ interaction is larger than the b₁+p* one. So the ethylene is more stable in the equatorial plane.

The same factors come into play for ethylene-Ni(PH₃)₂.

D. The Generalized Metal-olefin Bonding Picture

We have seen four cases of metal-olefin bonding now. All have two features in common:

But are these really metal-olefins or are they metallcyclopropanes? Is there a difference???

The metal-carbon bonds in a metallacyclopropane can be easily constructed by taking linear combinations of M-C s bonds-

But in the metallacyclopropane model we would expect the hydrogens to be bent back in a tetrahedral manner. Does the Dewar-Chatt-Duncanson model for the metal-olefin predict this structural feature? It does.

The hybridization in the p* MO causes its energy to drop and it becomes hybridized towards the metal so it is a better p-acceptor. What is clear is that the two components of bonding in the Dewar-Chatt-Duncanson model are not equal. When forward donation is much more important than back donation, then perhaps a metal-olefin picture is a better way to express the structure. When back donation becomes very important, a metallacyclopropane picture might be more appropriate.