Hydrothermal chemistry of methylene chloride and MTBE : experimental kinetics and reaction pathways

Abstract

With the growing desire for sustainable technologies, reactions in benign solvents, such as hydrothermal water and supercritical fluids, have become the focus of many investigations. Hydrothermal water has been used as a medium for chemical reactions, where the enhanced dissociation constant of water led to both acid- and base-catalysis without added reagents. Supercritical water oxidation (SCWO) has been proposed as an alternative technology for the treatment of aqueous-organic waste streams. At typical SCWO conditions (T = 550-650°C and P = 250-300 bar), mixed organic waste streams can be completely mineralized (>99.99%) in residence times of less than one minute. In an SCWO process, the waste stream is preheated to temperatures of >400°C in the absence of oxygen prior to the reactor. In the preheater, hydrolysis reactions may occur, significantly changing the composition of the reactor feed. To properly model these processes, a fundamental understanding of the reaction pathways and associated rates is essential. The rates of methylene chloride and methyl tert-butyl ether (MTBE) hydrolysis in sub- and supercritical water have been measured experimentally at 250 bar over a range of temperatures from 100 to 600°C. The rate constants for both compounds showed a local maximum below the critical temperature of water (374°C) followed by a local minimum just above the critical temperature. This behavior was qualitatively attributed to the changes in the solvent properties of water, shifting from a polar solvent in the subcritical region to a nonpolar solvent in the supercritical region. One of the primary objectives of this thesis was to develop a better understanding of the molecular-level effects of the solvent on the reaction rates and mechanistic pathways. The effects of water as a solvent on the hydrolysis reaction of CH2Cl2 were modeled as a dielectric continuum using Kirkwood theory. A correction factor, obtained from ab initio calculations, was applied to adjust the activation energy in order to account for differences in the free energy of solvation of the reactant and the transition state. Application of the Kirkwood correction to the empirical rate expression fit to data from 100-250°C yielded a model that quantitatively agreed with the experimentally measured reaction rate over the entire temperature range (from 100 to 500°C). To explain the extrema in the rate constant measured for MTBE hydrolysis, an acid-catalyzed mechanism was proposed. A new empirical rate expression was determined with a first-order dependence on the concentrations of H+ and MTBE. For the entire temperature range studied from 150 to 600°C, the empirical rate expression quantitatively modeled the experimentally measured decomposition rate within the uncertainty of the experiments. Further experiments were conducted with added HCl or NaOH that validated the acid-catalyzed hydrolysis pathway. The experimentally observed dependence on the concentration of H+ was slightly smaller than predicted by the proposed mechanism. Ab intio tools were employed to determine the relative contribution of a unimolecular decomposition pathway, which concluded that the pathway was not significant below 550°C. The unimolecular decomposition pathway set a lower limit on the overall reaction rate, which was observed experimentally under basic conditions where the acidcatalyzed pathway was effectively shut off. In addition to the kinetic measurements, two new experimental tools were developed to improve the capabilities of the supercritical fluids laboratory. Firstly, a new, large-bore tubular reactor system was built to address limitations in the current flow reactors in the supercritical fluids laboratory. The reactor was designed with the following advantages: 1) large diameter to minimize wall effects; 2) direct organic feed to eliminate hydrolysis during preheaters; 3) movable sampling probe; 4) sapphire windows to allow optical accessibility. The reactor was tested in preliminary runs up to 600°C and 250 bar. Secondly, a new reactor system was built to allow optical accessibility for in situ Raman spectroscopic measurement in supercritical fluids. The system was used in a study to probe local solvent effects in supercritical carbon dioxide. The effect of temperature, pressure, and density of CO2 on the vibrations of benzene and methylene chloride were investigated. As the density of CO2 increased, the vibrations shifted to lower frequency initially, and then leveled off at moderate densities. This leveling off may be due to local clustering of solvent molecules around solutes in these systems.