Hi Felix, thank you for your quick response about BSSE calculations but I still have some doubts. I am using the DMol3 module to perform the calculations. I used the BSSE correction and I got the following results (A is the atom and B is the surface):

Total DFT-D energy = -1615.5445226Ha

Total energy A(B) = -86.847072

Total energy B(A) = -1528.575822

Total energy A = -86.950460

Total energy B (Surface) = -1528.577920 Ha

Applying the equation found in http://vergil.chemistry.gatech.edu/notes/cp.pdf and taking into account that the surface structure changes a little during the adsorption process, I got:

ECPbind(AB) = ECPint (AB) + EBdef(B)= ECPint (AB)+[EB(isolated B, but with geometry of B as it appears in the dimer)-EB(isolated B and geometry optimized)

The first term is obtained from the DMol3 calculation (my BSSE corrected calculation) but my doubt is about the energy of the isolated surface, I mean the EB (isolated B, but with geometry of B as it appears in the dimer) term. If I have understood correctly, when one of the molecules suffers a deformation, one must do an energy calculation using the geometry of the molecule as it appears in the dimer in its monomer basis.

Must I perform a separate calculation on this structure to get its energy or can I use the term Total energy B (Surface) = -1528.577920 Ha (listed in the DMol3 output)? When I perform a separate calculation on this structure using the same parameters as those used for the dimer (the same functional, convergence criteria, dispersion correction,…) and I compare both, the results are very different (-1528.7203389 Ha (separate calculation)/ -1528.577920 Ha (BSSE calculation).

I feel I am doing something wrong or, quite simply, I have not understood a thing… :-(

Thanks again.