While learning about ferrocene I've come across this molecular orbital diagram. What particularly confused me is the nature of the a'1g orbital. My professor taught that a'1g and e2g are considered HOMO bonding orbitals, but this MO diagram clearly shows the a'1g as an MO with antibonding character.
To make sure it wasn't just a bad MO diagram, I looked for other MO diagrams, but they consistently display a'1g as a higher energy orbital compared to Iron's $3dz^2$ orbital.[a][b] So, slight anti-bonding character, right?
The problem is, looking at bond-length comparison, the a'1g orbital is slightly bonding. The Fe-Cp distance for ferrocene is 1.66 Å; for ferrocenium, it's 1.68 Å.[c] Given that the removal of an electron from the HOMO a'1g orbital resulted in the lengthening of Fe-Cp distance, it should have bonding character.
So what is going on here? Why is there this discrepancy? How are these orbitals actually interacting with each other?