Dichlorine peroxide (ClOOCl), chloryl chloride (ClCl(O)O) and chlorine chlorite (ClOClO): very accurate ab initio structures and actinic degradation.

The structural parameters of the three most stable isomers with formula Cl2O2, dichlorine peroxide, chloryl chloride and chlorine chlorite, were determined by high-level ab initio theory. The effects of core-valence electronic correlation as well as relativistic corrections were included in the calculations, and vibrational averaging of the so-obtained structures was performed. The bond distances agree with experimental data, where available, to within 0.01 Å, an unprecedented accuracy in particular for the floppy dichlorine peroxide molecule. The UV spectra of the three molecules were computed and decay pathways investigated. Under actinic UV radiation at 248 nm dichlorine peroxide decomposes principally according to ClOOCl → 2Cl + O2, in a synchronous concerted mechanism. In the low-energy tail region of this signal, the decay proceeds in a non-synchronous manner. There is also a low probability of the decay channel towards ClOO + Cl, whereas ClO molecules were not found. Chloryl chloride absorbs strongly around 300 nm. It disintegrates into ClO2 + Cl, where ClO2 seems to be formed mainly in excited electronic states which decompose further into Cl + O2. Hence chloryl chloride, though much less abundant in the stratosphere than dichlorine peroxide, also contributes to ozone depletion.

Accurate theoretical characterization of dioxygen difluoride: a problem resolved.

Dioxygen difluoride is a tough molecule that has defied accurate theoretical description for many decades. In the present work we have identified the reason for this resistance: the flatness of the OO, and more important OF, stretching potential energy curves, which make it difficult to localise the global minimum. It is not related to the weak multi-reference character. Using high-level CCSD(T)-F12/VTZ-F12 ab initio theory, the global minimum has been properly located and vibrationally averaged bond lengths obtained. These vibrationally averaged parameters agree with experimental data to within 0.01 Å. Averaging was found essential to achieve this unprecedented accuracy. We have then simulated the IR and UV spectra, which compare well with experimental data and permit identification of the observed transitions.

Tropospheric Reactions of Triazoles with Hydroxyl Radicals: Hydroxyl Addition is Faster than Hydrogen Abstraction.

We present the results of a systematic investigation, at the BHandHLYP/AVTZ density functional (DFT) level, of the tautomeric equilibria of 1,2,3- and 1,2,4-triazoles and their reactions with hydroxyl radicals in the gas phase. A total of twenty-six chemical reactions has been studied, and thermodynamical data and rate constants are reported. The reactions can be classified in two categories: hydrogen abstraction and OH addition. Nine of these reactions are favourable at room temperature. It was found that OH addition proceeds more rapidly than hydrogen abstraction, by several orders of magnitude. For the most stable tautomers, which presumably dominate in the gaz-phase, the fastest reactions are OH addition to 2H-1,2,3-triazole and site-specific OH addition to carbon atom 5 of 1H-1,2,4-triazole. In absolute values, however, the rate constants are rather small, k=5.82×10-20  cm3 s-1 and k=4.75×10-18  cm3 s-1 , respectively, at room temperature. Therefore, under the conditions of the troposphere, triazoles cannot be eliminated by reactions with OH. The accuracy of the computational approach has been demonstrated by studying the tautomeric equilibria of 1,2,3- and 1,2,4-triazoles in the gas phase. The computed equilibrium constants are in excellent agreement with those derived from spectroscopic observations. Additional assessment of the quality of the computed data was made by comparison with the results from high-level CCSD(T)-F12/AVTZ ab initio calculations for three exemplary structures. This comparison demonstrates the high accuracy of our DFT results of 0.29 kcal mol-1 for energy differences between stable isomers.

The gas-phase structure of dimethyl peroxide.

There has been a disagreement amongst experimentalists and between experimentalists and theoreticians as to the gas-phase structure of dimethyl peroxide. We have investigated this problem with high-level CCSD(T)-F12 and MRCI procedures. There can be no doubt anymore that, at the minimum of the potential energy surface, the COOC fragment has a trans-structure. The dynamical structure of the molecule can, however, be different and be explained by the very slow torsional motion. We have analysed the dynamical structure using numerical wavefunctions of the torsional motion and a fully optimized potential curve of MP2/aug-cc-pVTZ quality. Computational and all experimental results are shown to be in complete agreement. The problem that has persisted for more than thirty years, highlighted in a recent review article by Oberhammer titled "Gas phase structures of peroxides: experiments and computational problems", has been resolved.