I am doing DFT calculations on CdSe clusters (Cd15Se15) connected with a molecular bridge. When I change the composition of the bridge, these clusters move close and merge. The linker is composed of an aromatic ligand that attaches to the particle (two, one for each) bridged with a linker to create a conjugated system. When I change the type of linker from the one containing C=C (double bond) to benzene ring, the clusters bend the linker and merge. With C=C linker, they just remain apart with the linker stretched out between them (distance is about 1.5nm, about twice the size of each cluster).
I am guessing that introduction of benzene created strong electron coupling between clusters, but I am not sure. I will do TD-DFT on the system, but it will have to be single-point calculation, because I want to look at the separated particles. The values of HOMO-LUMO gap from these single-point calculations are very small (1.36eV for 2-particle system and 1.05eV for 3-particle system, compared to 2.92 for optimized single particle and 2.68 and 2.65 for optimized 2- and 3-particle systems with C=C linker, respectively), so I don't think they are a realistic representation of the band gap change. Although, these systems were created using optimized 2- and 3-particle systems with C=C linker, so they are as close to optimized structure I can get at this point.
Any advice on this would be much appreciated.