Thanks for the reply.

We are dealing with the reactants and products having multi reference character and from our chemical intuition, we also are expecting that at the transition state, complex won't have that much multi reference character so, we can use less CAS space to study the transition state. Here, we have 2 molecules as reactants and two molecules as products and we have done constrained optimization by placing the 2molecules 20 Angstroms far apart for reactants and products by using same CAS space (19/15).

Because of large CAS space, searching transition state with this CAS space will take so much time so, we want to do transition state calculation with a lesser CAS space. If that is the case then how can I compare the energies of reactants and transition state.

Can U suggest any better methodology to do CASSCF study of this type of systems..??

Hi, if still interested, or maybe for other readers, I would recommend doing it in few steps - in fact it might be a general way how to treat bigger molecules without "obvious" active space.

I would first do a RASSCF calculation for reactants and TS with quite big AS for each, allowing say double excitations only (possibel to try more, depends on the resources available). Then carefully select the orbitals involved in transitions and compare/combine the reactants and TS results . Move these orbitals to be the highest occupied and lowest unoccupied. These orbitals will then form the common smaller AS for the actual CASSCF calculations. After that it will be time to consider CASPT2 importance if not just looking for the "mechanism" but also for the energetics.