Two different theoretical approaches are used to study the OH radical attack on toluene: the Møller-Plesset perturbation theory and the B3LYP density functional method. The critical points of the potential energy surface for the OH addition to toluene are determined, and rate-equilibrium relationships are discussed. A stable structure corresponding to a prereactive complex which is formed when the OH radical is at about 2.5 Å from toluene is obtained. The existence of this loosely bound system is necessary to explain the experimentally observed negative activation energy. The geometry of transition states and products are determined for addition at different positions in the ring, including the ipso position, which has not been considered in previous works. Energy results at the MP4 and coupled cluster levels calculated at the optimized MP2 and B3LYP geometries confirm that the ipso adduct is more stable than the ortho adduct by about 0.5 kcal/mol. Several routes are proposed for the subsequent reactions of the ipso adduct, which could explain the very high yield of o-cresol with respect to the other cresol isomers.

Maya blue is a marvelous pigment with extraordinary properties. It was invented by the Maya around VII-VIII century and used by many Mesoamerican peoples in prehispanic times. It is made by encapsulating natural indigo into an inorganic clay matrix of palygorskite. The palygorskite-indigo mixture becomes acid-resistant when a moderate thermal treatment is applied. The chemical reasons of the unusual stability of the pigment and the exact mechanism of interaction between the indigo and the clay are not well understood. We present a Raman study of different preparations of Maya blue and other mixtures of indigo with other inorganic materials. We found that the unheated mixture of indigo with palygorskite presents the same Raman spectrum as Maya blue, indicating that the differences with respect to the indigo spectrum are not due to the interaction produced during the thermal treatment, which makes the mixture acid-resistant. Moreover, indigo mixed with other clays, like sepiolite or montmorillonite, presents a Raman spectrum very similar to that of Maya blue. Some chemical mechanisms that could explain these spectra, and the suitability of Raman spectroscopy for identifying Maya blue are discussed.

OH radical reactions with benzene and toluene have been studied in the 200-600 K temperature range via the CBS-QB3 quantum chemistry method and conventional transition-state theory. Our study takes into account all possible hydrogen abstraction and OH-addition channels, including ipso addition. Reaction rates have been obtained under pseudo-first-order conditions, with aromatic concentrations in large excess compared to OH concentrations, which is the case in the reported experiments as well as in the atmosphere. The reported results are in excellent agreement with the experimental data and reproduce the discontinuity in the Arrhenius plots in the 300 K < T < 400 K temperature range. They support the suggestion that the observed nonexponential OH decay is caused by the existence of competing addition and abstraction channels and by the decomposition of thermalized OH-aromatic adducts back to reactants. We also find that the low-temperature onset of the nonexponential decay depends on the concentration of the aromatic compounds and that the lower the concentration, the lower the temperature onset. Under atmospheric conditions, nonexponential decay was found to occur in the 275-325 K range, which corresponds to temperatures of importance in tropospheric chemistry. Branching ratios for the different reaction channels are reported. We find that for T > or = 400 K the reaction occurs exclusively by H abstraction. At 298 K, ipso addition contributes 13.0% to the overall OH + toluene reaction, while the major products correspond to ortho addition, which represents 43% of all possible channels.

The OH abstraction of a hydrogen atom from both the side chain and the ring of toluene has been studied in the range 275-1000 K using quantum chemistry methods. It is found that the best method of calculation is to perform geometry optimization and frequency calculations at the BHandHLYP/6-311++G(d,p) level, followed by CCSD(T) calculations of the optimized structures with the same basis set. Four different reaction paths are considered, corresponding to the side chain and three possible ring hydrogen abstractions, and the branching ratio is determined as a function of temperature. Although negligible at low temperatures, at 1000 K ring-H abstraction is found to contribute 11% to the total abstraction reaction. The calculated rate coefficients agree very well with experimental results. Side chain abstraction is shown to occur through a complex mechanism that includes the reversible formation of a collisionally stabilized reactant complex.