I am afraid I won't dig into this now due to time reasons (I have to go back and read up stuff), but I just would like to give a remark regarding the concept you have sketched out:
The different states you describe do not have to be on different energy levels. A catalyst provides an alternative reaction mechanism that might have a different activation energy. It can result in a molecule of similar energy state, and, yes, you can have a reaction towards an energetically higher state.
I am sure there is many a publication that models the electronic transitions that occurs when gas molecules react with metals. Funky stuff, which involves number crunching with models of atomic/molecular orbitals or bands, and even more funky spectroscopic real time measurements of reactions.
Chemistry has gone a long way since alchemy, and there are many abstraction layers to get through if you want to have an explanation that would satisfy a physicists. At the end, you will move from empirical knowledge towards a world of quantum chemistry/physics. That is the beauty (or curse) of the discipline that is chemistry.
Now, if you are curious how such a thing works, I suggest look up the "Wilkenson catalyst" (a Rhodium based catalytic hydrogenation agent) as an example. It gives you an idea how organometallic compounds work at various levels of depth.