Comments on: Ferrocene

By: The π-complex theory of metal-alkene compounds. « Henry Rzepa

The π-complex theory of metal-alkene compounds. « Henry Rzepa — Mon, 13 May 2013 19:08:25 +0000

[...] period 1951–1954 was a golden one for structural chemistry; proteins, DNA, Ferrocene (1952) and the one I discuss here, a bonding model for Zeise’s salt [...]

By: Aromaticity in the benzidine π-complex. « Henry Rzepa

Aromaticity in the benzidine π-complex. « Henry Rzepa — Sat, 19 Jan 2013 08:55:19 +0000

[...] resembles to remarkable degree the lowest π-MO of ferrocene which can be used to illustrate the 18-electron filled shells of the iron at the centre). in fact this is one of seven π-MOs that can be identified, making [...]

By: Trimethylenemethane Ruthenium benzene « Henry Rzepa

Trimethylenemethane Ruthenium benzene « Henry Rzepa — Thu, 18 Oct 2012 07:49:00 +0000

[...] which instantly makes an interesting point. Thus Ruthenium is ten electrons short of completing an 18-electron shell, and it can form a complex with benzene on one face and a ligand known as trimethylenemethane on [...]

By: Hafnium and Niels Bohr « Henry Rzepa

Hafnium and Niels Bohr « Henry Rzepa — Sun, 05 Jun 2011 09:38:02 +0000

[...] I note the striking contrast between Bohr’s article and Langmuir’s, written about a year earlier. Here, Langmuir sets out some postulates, the first of which is shown [...]

By: Henry Rzepa

Henry Rzepa — Sat, 23 Apr 2011 07:05:14 +0000

Responding to Steve, the ELF method does tend to distinguish between ionic and covalent bonding. If the electrons are located in disynaptic ELF basins, we might map that to covalent electrons, and if they are in monosynaptic basins, then the bonding is ionic. For ferrocene, there are only ~3.1 shared electrons located in the ten disynaptic basins between Fe and the carbons. The missing electrons are largely located in the C-C regions (~2.64 per C-C bond) and the C-H bonds (~2.2), which does indeed resemble an ionic Cp ring. So indeed one could reasonably say the interaction is predominantly ionic, but with a significant covalent component.

By: Steven Bachrach

Steven Bachrach — Fri, 22 Apr 2011 18:42:19 +0000

Might one be tempted to consider the interaction between the central iron and each cp ring as largely electrostatic. i.e. the attraction of the iron (ii) cation to the cyclopentadienyl anion?

By: Henry Rzepa

Henry Rzepa — Thu, 21 Apr 2011 07:07:04 +0000

A footnote to this post, I allude that Langmuir was the first to suggest that elements of the first transition series formed stable, and as he called it complete compounds if their shared valence (he referred to it as the sheath of the atom) contained 18 electrons. We now call this the 18-electron rule for transition elements. Langmuir gives examples of this genre as Fe(CO)₅ and Ni(CO)₄. He was of course spot on! But this article has other points of interest.

1. Langmuir presents the 18-electron (and for good measure the 32-electron) rules as postulates. There is absolutely no indication of how he arrived at these postulates. With Fermat (and his last theorem), at least we have a footnote! Langmuir simply proceeds on the basis that such a postulate explains a lot of chemistry. It was probably not until around 1927, when the Schoedinger equation was solved for the hydrogen atom, that a theoretical basis emerged.

2. Langmuir’s second postulate was that two atoms may be coupled together (we say bonded) by one or more duplets (we now say electron pairs) held in common by the completed sheaths of the atoms (we now say filled valence shells). By more, he means two, but near the end of his article, speculates that some types of exceptional compounds, citing HC≡CH, N₂, CO, CN^- as examples may merit a triple bond (he cites Pease as having suggested this first). This is clearly such a radical idea (in 1921!), that he cautions that it will need careful study. I cannot resist include this link to a diagram produced in 1860 by Loschmidt, in which he very clearly draws a triple bond in the cyano group!

3. Also of interest is what Langmuir does not say. For example, he cites nitric oxide, NO, as an exceptional molecule, bundling it in with the triply bonded systems listed above. However, he fails to mention that it has one unpaired electron, since he has introduced only the electronic duplex as a bonding concept. The reasons for this may originate from G. N. Lewis’ idea of shared covalency in the form of an octet of electrons arranged in a cube, whereby one shared duplet corresponds to a shared edge, and two shared duplets a shared face. This pictorial concept does not easily deal with sharing an odd number of electrons (five in the case of NO), or come to that six electrons (for triple bonds). In fact Pease had solved the triple bond problem by recasting Lewis’ 8 electrons at the corners of a cube by a tetrahedron of four electron pairs, in which a shared face corresponds to a triple bond.

4. Memories of Lewis’ cube model might have also been responsible for Langmuir’s failure to anywhere in his article mention benzene. Also in 1921, J. J. Thomson had in fact published his thoughts on benzene, where he clearly indicates his belief that the C-C bonds had three shared electrons. It was in fact becoming clear that fixed cubes or tetrahedra as models for bonding would have to be abandoned, but again we had to wait until quantum mechanics had matured (in particular the concept of σ/π separability by Hückel) to explain the spatial arrangement of the three electrons in benzene (in effect, we now treat each term of the series 2(1 + 3 + 5 + 7 +…) as having different angular solutions).

I mention one final precient observation, this time by Lewis himself. He had speculated that colour in molecules might be due to weakly held electrons. Again, spot on!