In particular, the features of the potential energy surface of reaction (2) were also derived from BMK, M06-2X, and ωB97X-D models combined with a 6–311++G(3df,3pd) basis set. 8 and 19), and these may be employed in alternative kinetic analyses, we use the opportunity to compare their results with the results from the method described above. 18Īs we often employ simpler modeling methods (see, e.g., Refs.

17 Modest breakdown of the Born–Oppenheimer approximation is accounted for through the ∆ E DBOC term, for which we use the diagonal Born–Oppenheimer correction (DBOC) evaluated at the CCSD = full/aug-cc-pCVTZ level of theory. ∆ E rel reflects the scalar component of the impact of relativistic effects, and here we use the sum of the expectation values of the mass-velocity and Darwin terms computed with configuration interaction with single and double substitutions, CISD, and the cc-pwCVTZ basis set, using the Molpro 2010.1 program. No generalized scale factors were therefore employed to obtain the zero-point energy or fundamental frequencies.īoth energies were obtained with the cc-pVDZ basis set and the frozen core approximation, with only valence orbitals included in the correlation treatment, using the MRCC extension of CFOUR. In effect, DFT values of x ij were used to correct the higher level ω i, rather than computing x ij with CCSD(T) theory as in the HEAT method. 15 Following the hybrid procedure of Barone and coworkers, 16 we added the anharmonic corrections for zero-point energy and fundamental frequency from DFT to the harmonic quantities from CCSD(T). Density functional theory (DFT) at the B2PLYP/aug-cc-pVTZ level was applied to evaluate anharmonic coupling coefficients x ij via vibrational second-order perturbation theory using third and fourth energy derivatives, as implemented in the Gaussian 16 code. Harmonic frequencies ω i were obtained with this level of theory. (12)where represents the Hartree–Fock energy at the infinite basis set limit, represents the correlation energy at this limit evaluated via coupled-cluster theory with double electron substitutions and a perturbative triples contribution, E zpe is the vibrational zero-point energy, ∆ E CCSDT(Q) is a further correction for electron correlation with up to quadruple excitations treated perturbatively, ∆ E rel accounts for scalar relativistic effects, and the ∆ E DBOC term addresses the Born–Oppenheimer approximation, details follow.įirst, the geometry was optimized at the CCSD(T) = full/cc-pVTZ level of theory, where “full” indicates that all orbitals were included in the electron correlation treatment, so that core–valence correlation effects were included implicitly. This justifies new theoretical calculations of this rate constant as described in the following section. Without going into further details of the radical reactions governing later stages of the reaction, one concludes that the experimental information on k 2 so far is inconclusive. For example, under typical shock wave conditions with equimolar mixtures of 1000 ppm CF 4 and H 2 in Ar, = 5 × 10 –5 mol cm –3, and at temperatures near 2400 K, k 1 ≈ 3 × 10 3 s –1 again would be considerably larger than k 5 ≈ 1.0 × 10 1 s –1 such that the dominant source for H atoms here would be the sequence of reactions (1), (9), and (10) instead of reaction (5) (following reaction (1), hydrogen atoms again are formed by reactions (9) and (10) on a μs timescale). The situation hardly changes for higher temperatures. 5 thus may not have been the dissociation of H 2, but that of CF 4 followed by reactions (9) and (10). The dominant primary source for hydrogen atoms in the experiments of Ref. 5 is in the fall-off range, at 1 bar of bath gas pressure (for M = Ar) and T = 1473 K from Ref.

Accounting for the fact that reaction (1) under the conditions of Ref. This question can be answered with the results of Ref. One, therefore, must ask whether CF 4 dissociation can compete with H 2 dissociation. 10, one realizes that H atoms are formed almost instantaneously after F and CF 3 have been produced (effective first-order rate constants larger than 10 5 s –1). As the rate constants for reactions (1), (5) (with M = H 2), (9), and (10) by now are known, one may compare the rate of hydrogen atom formation by reaction (5) with that of the sequence of reactions (1), (9), and (10). (11)and further reactions of CF 2 then carry on the reaction.

Cf4 molecular geometry