We are proposing a new computational thermochemistry protocol denoted W3 theory, as a successor to W1 and W2 theory proposed earlier [Martin and De Oliveira, J. Chem. Phys. 111, 1843 (1999)]. The new method is both more accurate overall (error statistics for total atomization energies approximately cut in half) and more robust (particularly towards systems exhibiting significant nondynamical correlation) than W2 theory. The cardinal improvement rests in an approximate account for post-CCSD(T) correlation effects. Iterative T3 (connected triple excitations) effects exhibit a basis set convergence behavior similar to the T3 contribution overall. They almost universally decrease molecular binding energies. Their inclusion in isolation yields less accurate results than CCSD(T) nearly across the board: It is only when T4 (connected quadruple excitations) effects are included that superior performance is achieved. T4 effects systematically increase molecular binding energies. Their basis set convergence is quite rapid, and even CCSDTQ/cc-pVDZ scaled by an empirical factor of 1.2532 will yield a quite passable quadruples contribution. The effect of still higher-order excitations was gauged for a subset of molecules (notably the eight-valence electron systems): T5 (connected quintuple excitations) contributions reach 0.3 kcal/mol for the pathologically multireference X 1Sigmag+ state of C2 but are quite small for other systems. A variety of avenues for achieving accuracy beyond that of W3 theory were explored, to no significant avail. W3 thus appears to represent a good compromise between accuracy and computational cost for those seeking a robust method for computational thermochemistry in the kJ/mol accuracy range on small systems.

Core-valence basis sets for the alkali and alkaline earth metals Li, Be, Na, Mg, K, and Ca are proposed. The basis sets are validated by calculating spectroscopic constants of a variety of diatomic molecules involving these elements. Neglect of (3s, 3p) correlation in K and Ca compounds will lead to erratic results at best, and chemically nonsensical ones if chalcogens or halogens are present. The addition of low-exponent p functions to the K and Ca basis sets is essential for smooth convergence of molecular properties. Inclusion of inner-shell correlation is important for accurate spectroscopic constants and binding energies of all the compounds. In basis set extrapolation/convergence calculations, the explicit inclusion of alkali and alkaline earth metal subvalence correlation at all steps is essential for K and Ca, strongly recommended for Na, and optional for Li and Mg, while in Be compounds, an additive treatment in a separate 'core correlation' step is probably sufficient. Consideration of (1s) inner-shell correlation energy in first-row elements requires inclusion of (2s, 2p) 'deep core' correlation energy in K and Ca for consistency. The latter requires special CCVnZ 'deep core correlation' basis sets. For compounds involving Ca bound to electronegative elements, additional d functions in the basis set are strongly recommended. For optimal basis set convergence in such cases, we suggest the sequence CV(D+3d)Z, CV(T+2d)Z, CV(Q+d)Z, and CV5Z on calcium. * Electronic address: comartin@wicc.weizmann.ac.il;

Heats of formation for a number of key C 1 and C 2 bromoalkanes and radicals have been calculated ab initio, both directly using an all-relativistic variant of W2 theory and indirectly using Douglas-Kroll relativistic CCSD(T)/Aug-VTZ reaction energies. For some of the bromoalkanes, our calculated values represent the first reliable data available. Bromine (3d) correlation contributes significantly to the molecular binding energies, but the effect of bromine (3s, 3p) correlation appears to be very small despite these orbitals lying above the carbon (1s) in energy. Thermodynamic functions have been obtained from molecular geometries and harmonic frequencies obtained at the B97-1/Aug-VTZ level and are given in the Supporting Information. These accurate thermodynamic parameters can be used to develop kinetic rate parameters.