We're going to build a vanadocene sandwich, in a nearly idealized D5h structure, similar to ferrocene.
Construct the sandwich in AMSinput by clicking on the benzene at the bottom, and then select [Metal complexes]:[Sandwiches]:[Ferrocene Sandwich] and click to create it
Select the metal, [Right-Click] and select [Change Atom Type]: change it to vanadium
Choose the following setup:
Unrestricted: [Yes]
Spin polarization: 3.0
XC functional: OPBE (GGA)
Relativity: Scalar
Basis set: TZP
Frozen core: Small
Numerical quality: Good
Make it C2v symmetric by changing one V-Xx-C angle to 95° at top ring, and the same for bottom ring
Enforce C2v symmetry by clicking the [☆] at the bottom
(optional) First, optimize the structure
Go to [Model]:[Regions]
Select the metal, and click on the [+] sign
Select the two rings, and click again on [+]
Go to [Multilevel]:[Fragments], and select [Use fragments]:[Yes]
Make sure that [Check charge and spin] is also set to [Yes]
For the metal fragment, set [Charge] to +2, [Spin] to 3
For the rings fragment, set [Charge] to -2
Save
Run
Analysis
Once the job has run, go to [SCM]:[Levels], to see how the molecular orbitals of the two fragments are combined into the MOs of the sandwich compound
Go to [SCM]:[Output] to see the output, and then go to [Properties]:[Bonding Energy Decomposition] to inspect the bonding energy (-725.9 kcal·mol-1), and its components (which you can compare with the reference data from ICA 2007)
Further reading
M. Swart
"Metal-ligand bonding in metallocenes: differentiation between spin state, electrostatic and covalent bonding" Inorg. Chim. Acta2007, 360, 179-189
10.1016/j.ica.2006.07.073
"Inorganic Chemistry - The Next Generation" issue
A.C. Castro, M.P. Johansson, G. Merino and M. Swart
"Chemical bonding in supermolecular flowers" Phys. Chem. Chem. Phys.2012, 14, 14905-14910
10.1039/c2cp42045g
Predicting new molecules by quantum chemical methods issue