Is the Gaussian-2 technique reliable for determining the energetics of small molecular dications? Available to Purchase

We present a quantum chemical study of 39 molecular dications comprising between two and four atoms. The calculations, which employ the G2 procedure and several variants (G1, G2(MP2), G2Q, and G2Q(QCI)), allow an assessment of the performance of these popular “model-based” techniques when applied to molecular dications. Of the dications included here, all except $HeAr2+$ and $AlCl2+$ have been studied previously by different theoretical methods: both $HeAr2+$ and $AlCl2+$ are calculated to be metastable against dissociation into atomic ions. Overall, the results of the present study indicate that G2 and its variants are less well suited to the characterization of molecular dications than of neutral or singly charged species, and that such theoretical methods should therefore be used with caution in determining the properties of dications. Several factors (including adequacy of the level of theory implemented in geometry optimizations; adequacy of the basis sets employed for single-point total energy calculations; and additivity of the basis-set extensions implicit in G2 and related methods) are responsible for the lapses in performance of the ab initio methods surveyed in the present study. Particularly severe discrepancies are evident in the calculated enthalpies of formation of $SO2+,$ $OOO2+,$ and $AlNC2+,$ which for the G2 and G2Q methods differ by 30, 98, and $162 kJ mol−1,$ respectively; other dications which are poorly treated at some or all of the levels of theory employed here include $NeBeNe2+,$ $OSO2+,$ and $F22+.$