Journal of Physical Chemistry A, Vol.107, No.25, 5049-5057, 2003
Ab initio quantum chemical investigation of the spin states of some chain and monocyclic diradicals
Ab initio calculations have been performed on nine organic diradicals to find the spin multiplicities in their electronic ground states. Three diradicals, namely, trimethylene methane (TMM), tetramethylene ethane (TME), and m-xylylene, were previously investigated in detail by various authors. These have been used as test cases so as to establish the reliability of the calculated results. The basis sets used in this work are mainly STO-3G and 4-31G. For the smaller molecules, the 6-31G basis set has also been tried. In every case, the molecular geometry has been optimized at the unrestricted Hartree-Fock (UHF) level for each basis set and for each spin state. Calculations have also been performed at the post-Hartree-Fock Moller-Plesset (MP) and coupled-cluster (CCSD) levels. Results have been quoted only for the UHF, UCCSD, and UCCSD(T) levels because the MP-level calculations do not generate reliable singlet-triplet energy gaps. The UHF calculation generally yields an unrealistically large splitting, but the UCCSD and especially the UCCSD(T) methods reduce the gap to a significant extent. From a comparison with the best results already reported on the test cases, we find that the UCCSD(T) singlet-triplet energy gaps are of the correct order. In fact, the singlet-triplet energy gaps calculated by the UCCSD(T) method using the 4-31G basis sets are more or less in agreement with the results of previous detailed investigations on TME and m-xylylene. Six more diradicals, two of them of the linear chain type (3-methylene pentane-2,4-diyl and tris(methylimino)methane) and four of the monocyclic variety (2-isopropylidene cyclopentane-1,3-diyl, 2,3-bis(methylene) cyclohexane-1,4-diyl, 3-methylene phenoxyl, and tetramethyl m-xylylene), have been investigated here. To our knowledge, this work is the first report on an ab initio post-Hartree-Fock calculation of the spin states of these six species. Out of the six diradicals, one has a singlet ground state, and the rest are ground-state triplets. These findings agree with the experimental observations without fail. The UCCSD(T)/4-31G results on tris(methylimino) methane, for which the hyperfine splitting spectrum is available, can explain the number of lines as well as the average hyperfine coupling constant. The molecule 2,3-bis(methylene) cyclohexane-1,4-diyl has been found to have almost degenerate singlet and triplet states, as observed experimentally. The UCCSD(T)/4-31G singlet-triplet gap is -0.84 kcal mol(-1) for this species. The UHF spin-density plots show that the ground states of all nine diradicals can be successfully predicted by the rule of spin alternation in the UHF treatment.