The goal of this research is to understand the speciation of mercury in the flue gases of coal combustion. As the flue gas cools, thermochemical equilibrium calculations indicate that elemental mercury, Hg⁰, is converted to oxidized mercury, Hg²⁺, in the form of HgO or HgCl₂. Hg⁰ is insoluble in water, HgO has low solubility in water and HgCl₂ is highly soluble in water. Since HgCl₂ is water-soluble, it can be captured in wet chemical scrubbers to prevent its release to the atmosphere. Therefore, the understanding of the mechanisms of mercury's oxidation in flue gases is paramount when considering mercury capture. This research attempts to elucidate the mechanisms of oxidation through a detailed kinetic and thermodynamic analysis. The current research focuses specifically on the oxidation of mercury via chlorine-containing compounds. Future research will involve the oxidation via oxygen-containing compounds and the effect of SO₂ and NOₓ compounds on mercury's oxidation. Quantum chemistry is used to determine accurate transition structures, which are required for the calculation of activation energies and rate constants from theory. Simultaneous to the theoretical work, an experimental apparatus has been designed and fabricated with the inclusion of a quadrupole mass spectrometer. The mass spectrometer is used in conjunction with a laminar flow reactor to simulate the oxidation of mercury via chlorine-containing compounds in flue gases. The ultimate goal of this research is to obtain a potential mercury oxidation mechanism based upon theoretically predicted kinetic parameters, which are then validated through concentration profiles obtained from experimental measurements. In addition, results from the experimental work indicate that at ambient conditions, the oxidation of mercury via chlorine may result as a consequence of heterogeneous reactions involving the Pyrex reactor surface. This work not only allows for a more thorough understanding of mercury's speciation in the flue gas environment, but also questions current sampling devices and their potential interference with reactivity measurements involving mercury-chlorine species.

This dissertation presents the development of a method for integrating two-phase flow into the vector formulation of the One Dimensional Turbulence model (ODT). The novel ODT model is an unsteady turbulent flow simulation model implemented on a one-dimensional domain, representing flow evolution as observed along a line of sight through a 3D turbulent flow. Overturning motions representing individual eddies are implemented as instantaneous rearrangement events. They obey applicable conservation laws and emulate the multiplicative increase of strain and decrease of length scales associated with the turbulent cascade. Eddy occurrences are random, with likelihoods proportional to a local measure of shear kinetic energy. These events punctuate conventional time advancement of viscous transport. In the present study, the ODT configuration used to simulate turbulent channel flow is augmented by a representation of particles coupled to the fluid by a drag law, with one-way coupling. It is straightforward to implement this drag coupling using the vector wall-normal fluid velocity profile evolved by ODT, but motion (displacement by eddy events) and velocity are distinct in ODT, so this procedure violates physical requirements such as correct representation of the marker-particle limit. Instead, a particle-eddy interaction mechanism is introduced. ODT eddies are instantaneous, so this interaction is defined by integrating the drag law over the lifetime of the corresponding physical eddy, but applying the resulting particle location and velocity change at the instant of eddy occurrence. A subtraction procedure is used to prevent double-counting of particle-eddy interaction due to subsequent viscous time advancement over the same time interval. The net outcome is a particle-eddy interaction that obeys correct limiting behaviors and transitions smoothly between these limits. This formulation introduces a free parameter that multiplies a scaling estimate of the eddy lifetime. Numerical simulations were run with turbulent friction Reynolds numbers ranging from 180 to 1395. Validation was achieved by comparing (1) wall-normal profiles of particle statistics with DNS, LES, and experiments; (2) wall deposition for particles from the inertial range of (Stokes number) 0.3 <= Tau+ <= 55,000 to DNS, LES, and experiments; (3) the non-inertial, Brownian Motion, regime was demonstrated by comparison with experiments and DNS.

The emission of semi-volatile toxic metals from boilers and incinerators is a major concern. One potential method to control semi-volatile metal emissions is high temperature sorbent injection. In this work, binary combinations of two toxic metals and three sorbents--kaolinite, hydrated lime, and a paper waste-derived sorbent (PWDS), were examined qualitatively and quantitatively. PWDS is a novel material produced from the waste sludge of the paper recycling process. An aerosol size-fractionation technique was developed to discriminate between metal vapor and metal that is either condensed on or reacted with particles in the furnace. A rapid-quench sampling system forces homogeneous nucleation to occur in the presence of other particles. Small nuclei generated from metal vapor are easily distinguished from metal reacted or condensed to larger particles. The aerosol size-fractionation technique was used to explore metal partitioning in the combustor. The longitudinal evolution of two mixtures, lead/cadmium, and cadmium/nickel, were examined. Based on this study, condensation and coagulation control aerosol processes in the furnace. Screening and parametric tests were performed to investigate the effect of injection temperature, sorbent/metal molar ratio and chlorine concentration on the capture of toxic metals by sorbents. In general, kaolinite and PWDS reacted rapidly with lead and achieved &sim;100% lead sorption in a short time while cadmium was absorbed rapidly by lime only. The lead/lime reaction appeared to be slow compared to the other sorbents. In the presence of chlorine, the amount of lead captured by kaolinite and PWDS was significantly reduced. However, the small amount of lead absorbed by the lime seemed unaffected by chlorine. The reaction between lead and kaolinite was studied in detail. Lead capture increases with increasing sorbent feedrates and decreases with increasing chlorine concentration. The lead/kaolinite reaction appeared to be inhibited by higher temperatures. A simple global reaction model was proposed where the lead/kaolinite reaction product inhibits further reactions. Kinetic parameters were estimated from experimental results. The model was then modified to account for the effect of chlorine, the results of the model predict super-equilibrium concentrations of lead chloride vapor.

The chemical mechanical planarization (CMP) process is critical in fabricating ultra large scale integrated (ULSI) circuit devices in semiconductor industry. In a typical aluminum damascene interconnect scheme, Al is usually blanket deposited over a liner layer to fill contact holes and vias. After deposition, the excess of Al is removed by CMP, leaving Al only in the holes and vias to form interconnects. Since the slurries used for aluminum CMP typically contain an oxidant and other chemical additives, the electrochemical behavior of Al and the liner may be expected to affect the polishing rates. In addition, when the excess of Al is removed, a surface transition from Al to liner occurs. Since Al and the liner may exhibit different electrochemical behaviors in the slurry, galvanic coupling between Al and the underlayer is a possibility. Such a coupling may lead to localized corrosion or rate control problems. The objective of this research was to characterize the fundamental electrochemical behavior of thin aluminum-0.5%copper, titanium and aluminum/titanium stack films before, during and after abrasion in a commercially available alumina based slurry containing iodate as an oxidant. A special apparatus in which electrochemical tests can be carried out during polishing was fabricated and used for this research. It was found that the electrochemical corrosion rates during abrasion were much smaller than the actual polishing rates obtained with the simulated CMP apparatus, indicating that the mechanism of Al removal by the iodate based slurry may not be dominated by electrochemical factors. A sharp rise in corrosion potential (Ecorr) during the transition from Al to Ti film was measured during polishing of the Al/Ti film stack. This potential change during transition was of the order of 1V on the Al film deposited at room temperature. The transition was much sharper with the low-temperature (25°C) Al film than the high-temperature (475°C) Al thin film. The slower transition in OCP in high-temperature films is most likely due to a Ti-Al intermetallic compound formed at the Al/Ti interface. The galvanic corrosion between Al and Ti during polishing and Al post-polishing corrosion issues were also investigated. It was found that the galvanic corrosion rate between Al and Ti is 6*10⁻⁴ A/cm² and the corrosion potential is -0.24 V. Also, the corrosion current density for Al after abrasion and immersion in de-ionized water is lower than that in the slurry. In addition, the post polishing corrosion of Al in after abrasion in the iodate based alumina slurry was also investigated. It was found that the corrosion of Al in DI water after abrasion was insignificant.

Dewatering of silica and alumina suspensions was accomplished using electrodecantation and electrocoagulation. Electrocoagulation was found to occur in high-conductivity alumina suspensions (250-1300 μS/cm), while electrodecantation was found to be the separation mechanism in low-conductivity suspensions of alumina and silica ( < 20 μS/cm). With these low-conductivity suspensions, a clear fluid layer developed on the surface of the suspension. A clear fluid layer did not develop in high conductivity silica suspensions, 250 μS/cm, even though electrodecantation was found to dominate the separation. Spatial variations in the pH and conductivity were measured at the completion of electrodecantation experiments. A boundary-layer model was developed to quantitatively establish the principles of electrodecantation. This model provides an explanation for the formation of the clear fluid layer on the surface of colloidally stable suspensions and provides an understanding of how buoyancy driven motion redistributes ions produced/consumed at the electrodes, which results in the formation of pH and conductivity gradients. The growth rate of the clear layer at the top of the chamber is initially slower than that predicted by the model; at later stages the theory and experiments are in agreement. Numerical simulations were performed to support the boundary-layer model and were used to incorporate important features such as electrode reactions, ion gradients, and cell geometry omitted from the model. Two-dimensional simulations were performed to study the effects of buoyancy driven motion in the absence of any ion gradients. Due to limited computer resources, one-dimensional simulations were used to show that a clear fluid layer would not necessarily be expected in high-conductivity suspensions and to study the effects of electrode reactions in the absence of any fluid motion. To characterize properties important to electrodecantation, two techniques were developed to measure particle diffusion coefficients, size, and electrophoretic mobility. Taylor-Aris dispersion measurements are shown to provide accurate diffusion coefficient measurements for colloidal particles up to about 0.3 μm in diameter and capillary electrophoresis is used to establish a novel method for measuring electrophoretic mobilities of colloids that compares favorably with existing methods.

To fully utilize coal as a long-term energy source, pollution prevention technologies must be developed to mitigate the negative environmental and health impacts of coal combustion. NOₓ emissions are of particular concern due to their role in forming ground-level ozone, photochemical smog, fine particulates and acid rain. A systematic evaluation of near-flame aerodynamics was conducted to determine how burner operating parameters and oxygen partial pressure influence flame attachment and coal ignition, two properties essential for proper low NOₓ burner operation. A laboratory scale (17kW) 2m tall, 0.5m diameter electrically heated furnace and axial burner with adjustable secondary combustion air annuli and primary fuel jets were used in the study. Transport air oxygen partial pressure (PO₂), coal particle size distribution, primary and secondary jet velocity, and wall temperature were varied independently to determine the effect of each variable on flame attachment and NOₓ. NOₓ emissions from the furnace were similar to those from full-scale tangentially-fired boilers. The tendency for flame attachment increased with velocity ratio (Θ), wall temperature, PO₂, and coal fines. Θ's greater than 1 were required for stable combustion. Increasing Θ reduced flame standoff distances and NOₓ for always-detached flames. NOₓ increased with Θ for always-attached flames. Increasing PO₂ reduced NOₓ by up to 50% by promoting flame attachment. However both oxygen enrichment and increasing fines had little impact on NOₓ for always-attached and always-detached flames. Wall temperature and excess air in leakage were the dominant variables affecting NOₓ. Furnace exhaust oxygen levels increased when operating under a slight vacuum with corresponding increases in NOₓ. Emissions for detached flames increased with wall temperature 3 times faster than attached flames. Emissions data obtained from the furnace under slight positive furnace pressure increased linearly with wall temperature. A novel dual flame was produced at high Θ and reduced PO₂ consisting of one flame attached to the burner and one stabilized 18" below the burner. This configuration is similar to staged combustion but without separate over-fire air. Emissions from the dual flame were significantly below those observed from conventional Type-O attached and detached flames.

The effect of particle size on ilmenite (FeTiO₃) reduction was studied by experimentally determining the intrinsic activation energy and modeling the data using a shrinking core model. The model accounted for both kinetics and ash diffusion control and allowed for a variable diffusivity as a function of conversion. The intrinsic activation energy was determined by reducing ilmenite particles ranging from 60 to 400 mesh with H₂ and CO partial pressures between .131 and .156 atm over a temperature range of 812 to 1173°C. The activation energy for hydrogen reduction was found to be 22.3 kcal/mol and the activation energy for carbon monoxide reduction was found to be 16.1 kcal/mol.

The onset of electrohydrodynamic motion in a thin layer of liquid has been investigated for the case in which the electrical conductivity varies linearly across the liquid layer. It is shown that a finite conductivity gradient can be quiescently sustained, provided certain requirements are satisfied. Dimensional analysis shows, and numerical computations confirm, that the conditions for the onset of motion in the layer can be characterized by two dimensionless parameters: the electrical Rayleigh number, and the dimensionless conductivity gradient. Interest in the problem stems from experiments on isoelectric focusing (IEF), an electrically-driven separation process used to purify peptide and protein mixtures. Previous experiments had led to the postulate that, as conductivity gradients developed during IEF, electrohydrodynamic convection would arise to disrupt the separation process. The theory developed here supports such a postulate.

The bauxite, a mineral rich ore, is an important source of metallic aluminum in industry. The Bayer process, developed in 1888, involves the extraction of alumina (Al2O3) from bauxite. One of the key stage of Bayer process is the crystallization of aluminum trihydroxide (Al(OH) 3), an intermediate product from which alumina is obtained by dehydration. Crystallization behavior and the morphology of aluminum trihydroxide depend strongly on the reactor temperature, the supersaturation defined by the concentration ratio of soluble [Al2O3]/[Na2O], noted as A/C ratio, and the presence of impurities. This crystallization behavior of aluminum trihydroxide was examined at various temperatures, initial concentration ratios of [Al2O3]/[Na2O] and concentrations of titanium oxide and ferric oxide as impurities. The X-ray diffraction analysis showed that only gibbsite, a polymorph of Al(OH)3, was produced at the end of all the experiments. An increase in the temperature, as well as an increase of the initial A/C ratio, resulted in an augmentation of the slurry density and the particle size distribution. The Scanning Electron Microscope (SEM) images of aluminum trhydroxide crystals revealed that the presence of titanium oxide favored the production of the small single crystals minimizing agglomeration. In the case of ferric oxide, the agglomerates were formed but showed a decrease of the average diameter as compared with the crystals formed without any impurities. Zeta-potential and settling velocity of aluminum trihydroxide in the presence and absence of impurities were also measured.

Biosorption, a process of passive metal binding to biomass, may be used for purification of metal bearing effluents. This work investigates the binding of heavy metal ions (Cd, Cu, Zn), light metal ions (Ca, Na) and protons to biomass of the brown alga Sargassum. / The mechanism of metal binding was confirmed to be ion exchange. A novel multi-component sorption isotherm model for cation binding was derived to aid in predicting the biosorption performance of the new biosorbent. This model considers chemical binding to free sites on the biomass and ion exchange (1:2 stoichiometry for divalent ions). It is based on chemical equilibrium constants and assumes competition of all cations for the same binding sites. Two main binding sites (carboxyl and sulfate) were characterized in terms of their respective quantities and pK$ rm sb{a}.$ Only two model parameters had to be determined for each metal cation (binding constants) and one additional parameter for each binding site (site quantity). / The two-site model successfully described metal and proton binding at different pH in mono- and di-metal systems. It was possible to predict the complete equilibrium sorption state, residual metal concentration in solution and metal uptake by the biosorbent, from the known initial state for varying amounts of biomass and different initial pH values. In order to account for the effect of ionic strength and electrostatic attraction, the above mentioned biosorption model was expanded by incorporating a version of the Donnan model and/or a Gouy-Chapman double layer model. The charge density of the biomass was characterized and intrinsic binding constants were derived. Correlations to account for biosorbent particle swelling were established. For the specific case of linear increase of swelling with the number of free sites, an explicit sorption isotherm equation was derived that includes the Donnan model in an easy-to use-way. / Using the parameters obtained from pH titrations at different ionic strength, it was possible to predict the effect of ionic strength on Cd binding. The influence of Ca on Cd binding was predicted from experiments with Cd and Ca, respectively, in mono-metal systems.