Active space and basis set effects in CASPT2 models of the 1,3-butadiene-ethene cycloaddition and the 1,3-butadiene dimerization

In complete active space studies on the butadiene-ethene cycloaddition and the butadiene dimerization, no systematic consideration so far has been given to the selection of either active space or basis set and their effect on the corresponding reaction barrier. Benchmark calculations indicate that the dependence of reaction barriers on basis set and active space size is minimal. CASPT2 results also suggest a preference for the concerted mechanism for this reaction. The first step of the asynchronous biradical, stepwise biradical, and concerted mechanisms of the 1,3-butadiene Diels-Alder reactions with both ethene and itself was studied using CASPT2 to determine the influence of basis set and active space on reaction barriers. CASPT2(6,6) with the cc-pVDZ, 6-311+G(3df,2p) and cc-pVTZ basis sets provided the best results with average errors below 3.1 kJ mol-1 with respect to the experimental result. Increasing the active space size also had little effect on the calculated reaction barriers. With respect to experimental results, uncontracted multireference averaged quadratic coupled cluster (MRAQCC) produced superior barriers than internally contracted MRAQCC by 16.1-39.3 kJ mol-1. The inability of CASSCF to locate transition states for some of the cycloadditions across the butadiene-ethene and butadiene dimerization reactions is also rationalized. CASPT2 suggests a preference for the concerted mechanism of the butadiene-ethene reaction, however, no basis set yielded a preference for any of the butadiene dimerization pathways.