Browsing by Author "George W. Roberts, Committee Member"

Now showing 1 - 6 of 6

Analysis of the Factors Affecting the Competition Between Filaments and Floc-formers in Activated Sludge
(2006-07-25) Lou, In Chio; Francis L. de los Reyes III, Committee Chair; George W. Roberts, Committee Member; Joel J.Ducoste, Committee Member; Detlef R. U. Knappe, Committee Member
Filamentous bulking, the most common solids separation problem in activated sludge systems, is caused by the excessive growth of filamentous bacteria over floc-forming bacteria, resulting in decreased sludge settling ability. The competition between the two types of organisms has been historically described using kinetic selection. However, it has been suggested that other factors, such as the presence of a filamentous "backbone", differences in decay rates, bacterial storage abilities, and substrate diffusion limitation, may also affect the microbial selection. In this research, various hypotheses that integrated several of those factors were tested using modeling, reactor and molecular studies, and a new conceptual qualitative model combining kinetics and diffusion explaining bulking was developed. As a first step, a bacterial competition model integrating kinetic selection theory, filamentous backbone theory, decay rates and storage abilities of filaments and floc formers was set up to predict and explain coexistence in a completely mixed reactor. Sensitivity analysis showed that the kinetic parameters μmax and Ks, storage rate constants and backbone coefficient had the greatest effect on the simulation results. Monte Carlo simulation showed the effect of storage, and the ranges of dilution rates wherein one group outcompete the other were delineated. Since bacterial storage was an important factor in microbial selection, respirometry-based kinetic parameter measurement was reevaluated by considering cell storage. Substrate uptake tests combined with metabolic modeling were used to include bacterial storage in determining the kinetic parameters for bulking and non-bulking sludge. It was found that non-bulking sludge had higher maximum substrate uptake rates than bulking sludge, consistent with results from respirometry. Quantitative fluorescence in situ hybridization (FISH) showed that the filaments Eikelboom Type 1851, Type 021N and Thiothrix nivea were dominant in bulking sludge, comprising 42.0 % of mixed liquor volatile suspended solids (MLVSS), with 61.6% of the total filament length extending from flocs into bulk solution. Only low levels of Type 1851 filament length (4.9% of MLVSS) occurred in non-bulking sludge, 83.0% of which grew inside the flocs. This result seemingly supported the kinetic selection theory, but contradicted our previous experimental data that showed that bulking and non-bulking sludge have similar levels of total filaments length, and thus supporting the diffusion limitation hypothesis. To resolve this contradiction, a new conceptual qualitative framework was developed in this study. We hypothesize that the growth rates of filaments and floc formers are affected by the combination of kinetic selection and substrate diffusion limitation. Three different regions (bulking, transitional and non-bulking region) based on substrate concentration are suggested. In the bulking and non-bulking regions, kinetic selection controls the growth rate process and favors filaments and floc formers, respectively. However, in the transitional region, substrate diffusion limitation, determined by the floc size, plays an important role in causing bulking. To test this framework, sequencing batch reactors (SBRs) were operated with various influent substrate concentrations, and sludge settleability was measured at various floc sizes induced by different mixing strength. A model integrating both mechanisms was developed to simulate the substrate concentrations at different floc sizes. The modeling results showed the occurrence of diffusion limitation inside the flocs at a certain range of activated sludge floc sizes, and the experimental data supported this framework in the bulking and transitional regions.
Charge Defects in Low Temperature Silicon Nitride/Silicon Interfaces for Application in Computational Clothing and Electronic Textiles
(2002-07-26) Park, Kie Jin; Gregory Parsons, Committee Chair; George W. Roberts, Committee Member; David F. Ollis, Committee Member
The purpose of this research has been to 1) explore materials prepared using plasma enhanced chemical vapor deposition (PECVD) for amorphous silicon thin film transistors (TFTs) fabricated on large area flexible polyimide substrates, and 2) develop new concepts to make smart fabrics for computational clothing using the flexible TFTs. For item 1), silicon nitride films, as gate dielectric of TFTs, were deposited using various processing gases having different NH3/SiH4 gas ratio with constant temperature and various temperature at constant processing gas ratio. It was shown that as NH3/SH4 ratio increases, NH/SiH ratio increases. Apparent leakage current decreased but the flat band voltage was shifted with increasing NH/SiH in the films. It was proposed that the decrease in apparent leakage current with increasing NH/SiH ratio was related to charge screening effect as well as film improved insulating quality. The interface charge and bulk charge densities for SiNx films with different processing gas composition have been calculated. The interface charge density increases with increasing NH/SiH causing flatband shift and increasing total charge density. It was believed that the interface charge is generated by stress build up at Si/SiNx interface and increases with NH/SiH ratio. We found that as substrate temperature increases the NH/SiH ratio remains constant, the apparent leakage current increases and the flat band voltage shifts. Because the net total charge is compensated as substrate temperature changes, charge screening effects were believed to be less important than effects of composition change under the conditions studied. Interface charge density increased also with temperature. This was consistent with flat band voltage shift with temperature. For item 2), amorphous silicon TFTs were formed successfully on large area polyimide substrate using back channel inverted staggered structure and novel masks design. Linear and saturation mobility are 0.026 and 0.059 cm2/V•s, respectively. To address TFTs into clothing, conductive thread contact was investigated. Finally, using novel method, we attempted to form an electronic NOR gate using thin film transistors on polyimide, woven directly into a cotton fabric.
Computer Simulation of Chemical Reactions in Porous Materials
(2002-08-21) Turner, Christoffer Heath; Gregory N. Parsons, Committee Member; Keith E. Gubbins, Committee Chair; Carol K. Hall, Committee Member; George W. Roberts, Committee Member
Understanding reactions in nanoporous materials from a purely experimental perspective is a difficult task. Measuring the chemical composition of a reacting system within a catalytic material is usually only accomplished through indirect methods, and it is usually impossible to distinguish between true chemical equilibrium and metastable states. In addition, measuring molecular orientation or distribution profiles within porous systems is not easily accomplished. However, molecular simulation techniques are well-suited to these challenges. With appropriate simulation techniques and realistic molecular models, it is possible to validate the dominant physical and chemical forces controlling nanoscale reactivity. Novel nanostructured catalysts and supports can be designed, optimized, and tested using high-performance computing and advanced modeling techniques in order to guide the search for next-generation catalysts - setting new targets for the materials synthesis community. We have simulated the conversion of several different equilibrium-limited reactions within microporous carbons and we find that the pore size, pore geometry, and surface chemistry are important factors for determining the reaction yield. The equilibrium-limited reactions that we have modeled include nitric oxide dimerization, ammonia synthesis, and the esterification of acetic acid, all of which show yield enhancements within microporous carbons. In conjunction with a yield enhancement of the esterification reaction, selective adsorption of ethyl acetate within carbon micropores demonstrates an efficient method for product recovery. Additionally, a new method has been developed for simulating reaction kinetics within porous materials and other heterogeneous environments. The validity of this technique is first demonstrated by reproducing the kinetics of hydrogen iodide decomposition in the gas phase, and then predictions are made within slit-shaped carbon pores and carbon nanotubes. The rate constant is found to increase by a factor of 47 in carbon nanotubes, as compared to the same reaction in the bulk gas phase. Overall, the results of these simulation studies demonstrate improvements in chemical reaction yield and chemical kinetics that are possible by understanding the nature of confined reactions, and applying this knowledge to catalyst design.
Hybrid Particle-Nonwoven Membrane Materials for Bioseparations
(2009-07-07) Herigstad, Matthew Omon; Ruben G. Carbonell, Committee Chair; George W. Roberts, Committee Member; Jason M. Haugh, Committee Member; Behnam Pourdeyhimi, Committee Member; Patrick V. Gurgel, Committee Member
Adsorption separations performed in feed streams containing large particulates pose interesting problems, the solution of which would aid in many fields of bioseapartions. Production of biologically derived protein products is one of the most rapidly expanding sectors in the global economy. The capture and purification of these products has, of late, become the bottleneck of the industry and can account for approximately 50-80% of the production costs. The biopharmaceutical industry has begun to focus on improving overall economics by merging two or more separation schemes into one. The majority of the emphasis has been on combining the initial protein capture and host cell clearance steps; however, many of the currently available methods have shown little efficacy at large-scale. Additionally, interest in the clearance of pathogenic activity, most importantly infectious prions, from blood and blood derived products has grown over the past decade with the increased threat of blood-transfusion of variant Creutzfeldt-Jakob disease. This work characterizes the transport and binding properties of a novel hybrid particle-nonwoven membrane medium in which a polymeric chromatographic resin is entrapped between layers of a nonwoven polypropylene membrane (a particle-impregnated membrane or PIM). This membrane-supported resin construct offers the advantage of increased interstitial pore diameter to allow passage of cells and other debris in the feed, while providing sufficiently high surface area for product capture within the resin particles. Columns packed with stacked disks of PIM displayed excellent flow distribution, and had an interstitial porosity of ÃŽÂµb = 0.48 Ã‚Â± 0.01, a 25-60% increase over those typically observed in a packed bed. These columns were able to pass over 95% of E. coli cells and human red blood cell concentrate (RBCC) in 30 column volumes, while maintaining a pressure drop significantly lower than that of a packed bed. The dynamic binding capacity of the chromatographic resin entrapped in the PIM packed column for bovine serum albumin (BSA) was essentially the same as that observed with the same volume of resin in a packed bed. Additionally, the binding of prion was characterized to PIM constructs containing an affinity ligand for PrPSc, in saline, RBCC, and human IgG solutions. The General Rate (GR) model of chromatography was used to analyze experiments indicating that the breakthrough and elution behaviors of the PIM column are predictable, and very similar to those of a normal packed bed. These results indicate that PIM constructs can be designed to process viscous mobile phases containing particulates while retaining the desirable binding characteristics of the embedded chromatographic resin. The PIM systems could find uses in adsorption separation processes from complex feed streams such as whole blood, cell culture, and food processing and could offer a process alternative to expanded beds.
Kinetics of Photocatalytic Degradation Using Titanium Dioxide Films
(2008-04-25) Chin, Paul; Eric A. Stone, Committee Member; Christine S. Grant, Committee Member; H. Henry Lamb, Committee Member; George W. Roberts, Committee Member; David F. Ollis, Committee Chair
Titanium dioxide (TiO₂), a semiconductor metal oxide, has been used in heterogeneous photocatalysis for the destruction of organic, inorganic, and biological materials. The chief objectives of my doctoral research are to generate kinetic data and to develop engineering models for photocatalytic oxidation (PCO) using TiO₂ thin films for current challenges in "solid"-solid and air-solid environmental remediation. Three topics are studied in detail: i. TiO₂ Photocatalytic Oxidation for Formaldehyde Removal from Air (Air-Solid) Formaldehyde (CH₂O) is a toxic air contaminant present in industrial, commercial, and residential buildings. A novel rotating honeycomb adsorbent coupled with a PCO reactor was demonstrated by F. Shiraishi and coworkers for CH₂O oxidation. They showed that their cyclic adsorbent ⁄ PCO reactor could oxidize CH₂O to concentrations below the WHO guideline, but they made no attempt to model the system. In this project we modeled their batch system at transient and steady states. In addition, we applied the batch model kinetic parameters to design a continuous system for typical residential home challenges. ii. TiO₂ Photobleaching of Dye Layers as a Field Analysis Method ("Solid"-Solid) Technological advances in the past decade allow glass manufacturers to deposit thin, photoactive, nano-sized TiO₂ layers on glass billed as "self-cleaning" surfaces. Field installation of such "self-cleaning" window glass for office buildings or residential homes will require the creation of field analysis methods to characterize the initial and continuing catalyst activity variations with time and environmental conditions. In this project we characterized the PCO properties of commercial Pilkington Activ™ glass by oxidizing deposited organic dyes (a) to show visual decolorization and recovery of the aesthetic clarity of the glass and (b) to determine the light-driven reaction kinetics on Activ™ glass. iii. TiO₂ Photooxidation of Deposited Soot Layers ("Solid"-Solid) A major contribution to visual degradation of exterior surfaces in urban environments, especially the major cities of industrialized countries, is the deposition of particulate soot. Destruction of deposited soot layers by TiO₂ photocatalysis had been reported recently, but there was (a) an inability to deposit a soot layer of uniform thickness on the TiO₂ surface and (b) a lack of rigorous kinetic modeling. In this project we developed a method to apply a uniform, thin soot layer or a model soot on TiO₂ thin films. We also executed laboratory studies to collect kinetic data for PCO of soot deposits and demonstrated applicability of a series ⁄ parallel reaction network model to describe the kinetics of TiO₂ PCO on these porous carbonaceous layers.
Removal of Polar and Emerging Organic Contaminants by Alternative Adsorbents.
(2008-12-05) Rossner Campos, Alfred Armin; George W. Roberts, Committee Member; Detlef Knappe, Committee Chair; Dean L. Hesterberg , Committee Member; Joel Ducoste, Committee Member
The removal of polar and ionizable organic contaminants from drinking water sources is a challenge for water utilities. Prescription and non-prescription drugs are present at detectable levels in many US surface waters, and conventional water treatment processes do not provide an effective barrier for many emerging contaminants. Therefore, effective treatment technologies are required to successfully remove these contaminants from drinking water. In this study the effectiveness of alternative adsorbents for the removal of MTBE was evaluated. Additionally, adsorbent, adsorbate, and water characteristics that affect ionizable organic contaminant adsorption were identified. In the first phase of this research, the MTBE removal effectiveness of activated carbon, a carbonaceous resin and a silicalite zeolite was compared. Isotherm and short bed adsorber tests were conducted in ultrapure water and river water to obtain parameters describing MTBE adsorption equilibria and kinetics and to quantify the effect of natural organic matter (NOM) on MTBE adsorption. Both the silicalite zeolite and the carbonaceous resin exhibited larger MTBE adsorption uptakes than the tested GAC. Results showed that GAC was the most cost-competitive adsorbent when considering adsorbent usage rate only; however, the useful life of an adsorber containing silicalite zeolite was predicted to be ~5-6 times longer than that of an equally sized adsorber containing GAC. Pilot column results also showed that NOM preloading did not impair the MTBE removal efficiency of the silicalite zeolite. Thus, it may be possible to regenerate spent silicalite with less energy-intensive methods than those required to regenerate GAC. One activated carbon, one carbonaceous resin, and two high-silica zeolites were studied to evaluate their effectiveness for the removal of an emerging pollutant of concern (EPOC) mixture from lake water. Adsorption isotherm experiments were performed with environmentally relevant concentrations of the 28 targeted EPOCs (~200-900 ng/L). Among the tested adsorbents, activated carbon was the most effective, and activated carbon doses typically used for taste and odor control in drinking water (1-10 mg/L) were sufficient to achieve a 2-log removal for most of the tested EPOCs. The results of this study demonstrate that heterogeneity in pore size and shape along with a large pore volume in the 6-9 Ã… size range are important adsorbent characteristics when an effective barrier against a broad spectrum of EPOCs is desired. Five carbonaceous adsorbents were evaluated to adsorbed sulfamethoxazole (SMX) and trimethoprim (TMP) at different solution pHs. Results indicated that activated carbon AC1230C more effectively removed both compounds from ultra-pure water (UPW), Tar River water (TRW), and Lake Mead water (LMW). TMP and SMX were more effectively removed at pHs at which both SMX and TMP were primarily in the neutral form. Normalizing the equilibrium liquid-phase concentration of SMX and TMP by the pH-dependent solubility of the compounds was effective. Only at solution pHs where repulsive electrostatic interactions proved to be important the normalization procedure was not able to account for the pH-dependency of the adsorption capacity of the ionizable organic compounds. Finally, the effect of preloading reduced TMP and SMX adsorption capacities similarly at pH 7.8. SMX adsorption capacities for fresh F400 GAC at pH 3.6 and 5.8 were similar. In contrast for the preloaded GAC the SMX adsorption capacity at pH 5.8 was lower than at pH 3.6. This was related to the decrease in pHPZC and increase in acidic groups of the activated carbon surface with preloaded NOM.