Metal-Metal Bonding

From bondslam
Jump to: navigation, search

LIST OF PARTICIPANTS Bernard Silvi(not attending)


What is the nature of the Fe-Fe interaction?,

Previous studies by Bauschlicher MO analysis does not give any evidence of a direct Fe-Fe interaction, dπ* donation into the bridging CO 2π* orbital and bridging CO σ donation into the empty dπ orbital[1]. Gatti and Lasi report a delocalization index δ(Fe,Fe)=0.33 and no bcp.[2], finally Reinhold et al have used a combined AIM/MO approach; they conclude "It turns out that the questioned Fe-Fe bond is not critically related to the existence or the lack of the corresponding bcp, because this appears to be dependent on the computational details (basis sets and functionals), but to a specific interaction between d orbitals."[3]

--CarlosMF (talk) 15:25, 14 August 2017 (CEST)If one uses classic "electron counting" as is taught at the undergraduate level for the Fe2(CO)9 complex, we would obtain for each iron atom:

# electrons per Fe atom
Terminal CO 2 (x3)
Bridging CO 1 (x3)
Fe 8
Fe-Fe bond 1
Total: 18 e

If the Fe-Fe single bond is not taken into account then we would not have 18e complexes. This argument is usually taught to bachelor students in order to decide whether there is or not a Metal-Metal bond in a certain organometallic compound. However, what can we learn from theory?

--Silvi (talk) 10:11, 16 August 2017 (CEST) ELF populations enable to check magic number rules. For Fe2(CO)9 one has

# electrons per Fe atom
Terminal V(Fe,C) 2.85 (x3)
Bridging V(Fe,C) 1.58 (x3)
C(Fe) -18 5.4
Total: 18.69 e

The 18 e rule is almost verified. The main difference between the standard electron count and the ELF electron count is that the former does not account for the mesomery of the CO molecule. A description of CO consistent with the ELF analysis of the free molecules implies four mesomeric structures yielding 2.5 electrons for the carbon lone pair. With 2.5 and 1.25 e for terminal and bridging COs, the electron count is 3x2.5 + 3x1.25 +8 +1 = 20.25 showing that there is no need for a Fe-Fe single bond.

ELF=0.8 localization domains


  1. C. W. Bauschlicher Jr., J. Chem. Phys., 1986,84,872-875, DOI: [1]
  2. C. Gatti, D. Lasi, Farad. Discuss., 2007,135, 55-78, DOI: [2]
  3. J.Reinhold, O. Kluge, C. Mealli, Inorg. Chem.,2007,17,7142-7147, DOI: [3]