A direct measurement of the pH of water in contact with supercritical CO2 was made by observing the spectra of a pH indicator with a UV-vis spectrophotometer. The pH was analyzed under pressures of 70-200 atm and temperatures of 25-70 °C. The measured pH varied from 2.80 to 2.95, and relative standard deviations of <1.5% were achieved. The effects of pH on the efficiency of supercritical fluid extraction of metals and ionizable organic species in water-containing systems are discussed.Supercritical fluid extraction (SFE) has become an attractive alternative to conventional solvent extraction for the recovery of organic compounds and metal chelates from solid and aqueous samples.1•2 Carbon dioxide is the gas of choice for SFE because of its moderate critical constants, inertness, low cost, and availability in pure form. In many environmental applications, water is often present in an SFE system, either as a part of the original sample or added deliberately. The presence of water has been shown to facilitate the extraction of metal ions from solid materials using the in situ chelation technique in supercritical C02.2 3 The role of water in the SFE of organics and metals is not well understood. Water in contact with carbon dioxide becomes acidic due to the formation and dissociation of carbonic acid: C02 + H20 -H2C03 -»H+ 4-HC03D etermining the pH of water in contact with supercritical C02 is essential to understanding SFE of both metal ions and ionizable organic species. For SFE of metal ions, complexing agents are used to neutralize the metal charge and to transport the metals into the CO2 phase. The degree of dissociation of the complexing agents is dependent on the pH of the solution. If the complexing agents are not ionized at the solution pH, complexation and

A grand canonical ensemble molecular dynamics (GMD) simulation method has been adapted to examine the thermodynamics of clay-mineral hydration. In the GMD method, the number of water molecules in the system is treated as a continuous variable for which an equation of motion is established. Fluctuations in the water content at constant chemical potential are investigated using trajectories of this particle number variable. A bias potential may be used to modify the free energy contour along the particle number coordinate. This catalyzes particle fluctuations and greatly improves simulation convergence. Adaptation of the GMD method to treat hydrated clay minerals included the introduction of a local-control technique that fixes the water chemical potential in the clay interlayer region. In addition, a bias-potential feedback algorithm was implemented to improve particle fluctuation efficiency. Information pertaining to the free energy contour, generated during the course of the simulation, was used periodically to enhance the bias potential. This allowed for the utilization of a single input bias potential under a broad range of simulation conditions. The method was used to investigate swelling of a cesium–montmorillonite clay. Measured disjoining pressures showed oscillations that are indicative of crystalline-swelling phase transitions. Integration of the disjoining pressures yielded a swelling free energy profile with distinct free-energy minima for the one- and two-layer hydrates. The results may be compared qualitatively with both clay swelling and surface force apparatus experiments, and with previous simulation studies of simple fluids in slit pores.

Alkyl nitrates (RONO 2 ) are minor products formed in the atmospheric reactions of alkyl peroxy radicals (RO 2 •) with nitric oxide. The major products are alkoxy radicals (RO•) and NO 2 . The alkyl nitrate channel is important in the troposphere because RONO 2 formation results in removal of NO x and trapping of free radicals; both effects reduce the rate of ozone production. We have used electronic structure calculations at the G3 and B3LYP/6-311++G** levels to calculate the geometries, energies, and vibrational frequencies for major stationary points on the potential energy surfaces for R ) H, CH 3 , C 2 H 5 , n-C 3 H 7 , i-C 3 H 7 , and 2-C 5 H 11 . Selected calculations have been made at the G2 and QCISD(T)/cc-pVTZ levels. Reaction energies are found to be rather insensitive to the size of the alkyl group. Corrections to the reaction energies are estimated, and a generic set of reaction energies are suggested. The B3LYP/6-311++G** barriers for the isomerization of ROONO to RONO 2 are found to be much too high to account for observed nitrate formation.

Master equation calculations are used to model gas-phase literature experimental data for alkyl nitrate formation via the following reaction system of reversible reactions:  (1) RO2 + NO ↔ ROONO, (2) ROONO ↔ RO + NO2, (3) ROONO ↔ RONO2, and (4) RONO2 ↔ RO + NO2 for R = CH3, i-C3H7, and 2-C5H11. The structures and thermochemistry of the stable species are based on electronic structure calculations described in the preceding companion paper in this issue (Lohr et al. J. Phys. Chem. A 2003, 107, xxx−xxx). Literature data for recombination rate constants are used to constrain the model calculations. Several transition state models and a range of energy transfer parameters are investigated. The results for R = CH3 show that a wide variety of plausible transition state models for k - 4 gives good agreement with experiment for reaction (−4), because changes in assumed energy transfer parameters can compensate for differences between the transition state models. It is concluded that recombination reactions are good sources of absolute energy transfer parameters only when transition state properties are known with great accuracy. Although satisfactory models are obtained for the individual systems, the parameters cannot be transferred reliably from one system to another. Master equation models can be made to reproduce the experimental 2-pentyl nitrate yields from the title reaction as long as 〈Δ E〉down, the average energy transferred in deactivating collisions, is assumed to be surprisingly, and perhaps unphysically, small (∼25 cm-1), regardless of assumptions about the barrier to isomerization reaction 3. Several critical assumptions in the master equation models are examined, but none of them accounts for the small value of 〈Δ E〉down. It is concluded that new experiments should be carried out to verify or possibly revise the pressure-dependent alkyl nitrate yield data currently available in the literature.

All species involved in the multichannel decomposition of gas-phase dimethyl methylphosphonate (DMMP) were investigated by electronic structure calculations. Geometries for stationary structures along the reaction paths, were fully optimized with the MP2 method and the B3LYP and MPW1K DFT functionals, and the 6-31G*, 6-31++G**, and aug-cc-pVDZ basis sets. The geometries determined by the B3LYP and MPW1K functionals are in very good agreement with the MP2 values. Increasing the basis set size from 6-31G* to aug-cc-pVDZ does not significantly alter this result. Single point energy calculations were carried out with highly accurate but computationally more expensive CBS-QB3 theory. DMMP has three conformers, which lead to the four primary product channels, (O)P(CH(2))(OCH(3)) + CH(3)OH, (O)P(CH(3)) (OCH(3))(OH) + CH(2), c-(O)P(CH(3))OCH(2) + CH(3)OH, and (O)P(CH(3))(OCH(3))(OCH) + H(2). The first channel has the lowest energy barrier and is expected to be the most important pathway. It occurs via C-H and P-O bond cleavages accompanied by O-H bond formation. The other three channels have higher and similar energy barriers, and are expected to have smaller and similar rates. The product (O)P(CH(3))(OCH(3))(OCH) undergoes a secondary decomposition to form (OH)P(CH(3))(OCH(3)) + CO.