Some terminology clarification, please.

[Beer-monster]

I'm a physicist trying to learn some chemistry for my research (on-surface reactions and assembly) but am having some trouble with getting the meaning of some of the terms straight. Often one source seems to contradict another, in my reading of it at least.

Specifically, the meaning of the term Ligand. My understanding is that it is a non-metallic group at has formed a coordinate bond to a metal centre by donating a pair of electrons. This is what forms a complex.

However, I have seen the term applied to groups added to a metal through oxidative addition. My understanding is that this process forms a direct covalent bond (not a coordinate bond) to the metal as the metal loses its electrons (is oxidized) rather than has some donated to it.

Is my understanding of this reaction incorrect or is the term ligand used more widely to describe any group bonded to a metal in any way?

 

[DrDu]

Is my understanding of this reaction incorrect or is the term ligand used more widely to describe any group bonded to a metal in any way?

Yes, I would also use it in that broader sense especially because there is no fundamental difference between a coordinate bond and a covalent (or partially ionic) bond. After all, you can obtain the product of an oxidative addition also starting from the correctly oxidized ion coordinating the already reduced ligands.

 

[Beer-monster]

Yes, I would also use it in that broader sense especially because there is no fundamental difference between a coordinate bond and a covalent (or partially ionic) bond.

Hmm...that does mean something is off in my understanding as I have seen metal-coordination bonds grouped as non-covalent interactions. I had thought that was why the oxidation state of a complex (electrons lost/gained) did not relate to the coordination number (number of atoms actually bonded to).

Is that totally wrong or just sort of?

 

[DrDu]

Besides maybe the bond between two identical atoms, there is no purely covalent bond and neither are there purely ionic bonds. Bonding in complexes always has both ionic and covalent character. When determining the oxidation number, you count the electrons to that atom with is more electronegative. This is a purely calculational device and does not tell you anything about the actual ratio of covalent to ionic character of the bond.

 

[JohnRC]

Yes, nature does not really know about covalent, co-ordinate, or other sorts of bonds; in fact it does not even know about bonds at all! All of these things are just convenient caricatures and accounting systems that chemists use. The trick is that if you use them consistently in the way that chemists do, you will nearly always get the right answer.

So the term "ligand" is used in a very broad sense in any situation where one or more sets of atoms is bonded around a metal atom in a central sort of location in the structure. The atom bonded to the metal must not be another metal atom, and especially not anything more electropositive than the central metal.

"Co-ordination" number is used in an even broader sense. In an ionic crystal lattice with the sodium chloride structure, a chemist would describe both sodium ions and chloride ions as having a co-ordination number of 6, and in a network structure with polar covalent bonds, a chemist might say that in quartz silicon has a co-ordination number of 4 while oxygen has a co-ordination number of 2 -- not a co-ordinate bond in sight in either of these last two examples.

With your stated example there is a really good reason for extending the concept of "ligand" to cover the case of covalent bonds formed by oxidative addition. Consider the case of oxidative addition of chloride ions to copper metal in concentrated aqueous hydrochloric acid:

Cu + 2 H⁺ + 4 Cl^– --> CuCl₄^2– + H₂

Formally, two of the Cu--Cl bonds are covalent and the other two are co-ordinate bonds. In practice, though, all four Cl atoms are symmetrically equivalent in this structure (it is not a perfect tetrahedral structure, but it is tetragonal -- a tetrahedral structure squashed a little towards square planar). So it is quite inappropriate to try to distinguish between the Cl atoms, and all four must count as ligands.

If you look at the Latin roots of the term, a ligand is simply "that which is tied on"!