Page 1 of 1

strange bond

PostPosted: Mon Sep 11, 2017 6:28 am
by mnahali
Dear All

CO is adsorbed on Cu5Ag cluster and the adsorption energy is about -0.4 eV. I have already posted a topic about ionic bonding on the NBO Forum. There is not any
covalent (bonding orbital) between CO and Cu ( number 2 ) in Cu5Ag cluster so it would be an ionic bond between C (in CO) and Cu (in Cu5Ag) but the point is that the NBO charge of C and Cu are both positive and this repulsion seems not to be responsible for making a bond. As the energy of the adsorption is negative how could we explain such bond between C and Cu.
Attached you can find input and output of our NBO calculation.
I appreciate your help in advance.

Best Wishes
Masoud Nahali
Sharif University of Technology

Re: strange bond

PostPosted: Mon Sep 11, 2017 8:55 pm
by ericg

I ran your calculation myself. This is what I see in the NBO output.

There is no NBO (bond) identified between C(8) of the carbonyl ligand and Cu(2) of the metal cluster. However, perturbative analysis reveals are a rather strong charge transfer interaction (~76 kcal/mol) between the C lone pair and the Cu 4s valence orbital (LV). 76 kcal/mol (~3.3 eV) is stronger than the 0.4 eV absorption energy you report. You might view the interaction between ligand and cluster as arising from this CT interaction, but weakened by the repulsive electrostatic interaction between ligand and cluster.

More detail regarding the interaction could be obtained by performing natural energy decomposition analysis (NEDA) on this system. NEDA partitions the interaction into electrical interaction, charge transfer, and core repulsion. However, NEDA only works with GAMESS, not with Gaussian.