New analytical and numerical approaches for calculating the gas holdup and drift velocity in bubble columns
SUPERVISOR: Dr. Siamak Elyasi (Chemical Engineering)
ABSTRACT: The two-phase flow phenomenon is of prime importance in the chemical, petroleum, nuclear and power industries. The two-phase flow encountered in nuclear and power industries is usually boiling and condensation two-phase flow while those practiced in the chemical and petroleum industries is essentially non-boiling in nature. The installation of two-phase flow lines has been proven economical in the oil and natural gas transportation system. The two-phase flow is used in chemical industry to enhance the mass transfer rate as in case of the bubble columns. Processes based on the contact between gas and liquid phases are commercially used for performing a variety of chemical reactions. Although different types of reactors are used for this purpose, bubble column reactors have received more attention during the past decade since they offer some unique advantages, such as ease of operation, high rates of heat and mass transfer, and lower maintenance costs due to the absence of moving parts. The design and scale-up of a bubble column reactor require a complete understanding of its complex hydrodynamics, which is influenced by the physical properties of the phases, the operating variables, and the design parameters. Current design procedures for bubble columns involve several steps of pilot-plant experimentation using equipment of different scales, which is expensive and time consuming. One of the critical unknown parameters involved in designing any gas–liquid system is the gas holdup which is the volume of space occupied by the gas. The gas holdup distribution varies dramatically throughout the flow regimes and determining its magnitude from input conditions in a given pipe is complicated as a result of the relative motion between the gas and the liquid phases. Owing to the complexity and lack of understanding of the basic underlying physics of the problem, the majority of the analyses were more inclined towards empirical correlations. The most of the empirical models have some form of restrictions attached to them. As a result, the design engineer is faced with the difficult task of choosing the right correlation. One of the most practical and accurate models for calculating void fraction is known as drift-flux theory. This model is of considerable importance because of its simplicity and easy applicability to a wide range of two-phase flow problems. Because of the practical significance, the focus of this study is on the drift-flux model as a simple and precise solution for determining the gas holdup.Furthermore, a numerical model to study the hydrodynamic behavior of a large-diameter bubble column, using the commercial CFD code ANSYS® FLUENT 16.2, was performed. The model was run for both semi-batch (stationary liquid) and continuous bubble columns. An Eulerian- Eulerian approach was used, and the effect of the superficial gas velocity on the flow regime, distribution parameter and drift velocity was examined. Based on the obtained results, a correlation was proposed to predict the gas holdup in semi-batch reactors. Verification of all simulation results and the suggested correlation against available experimental data showed a satisfactory agreement.A comprehensive literature study was carried out for the available drift-flux based gas holdup correlations and experimental data taken from researches conducted in different flow, operating and geometrical conditions. The performance of the existing correlations for the distribution parameter and drift velocity was evaluated against diverse data in small- and large-diameter bubble columns. Comparisons between the correlations showed that most of the developed correlations are very restricted in terms of handling a wide variety of data sets. A new approach to calculate the drift velocity that specifies the relative motion between phases was developed, and subsequently, a gas holdup correlation was proposed. A comparison of the models with various experimental data over various flow regimes and a wide range of flow parameters shows a satisfactory agreement.An analytical method was also developed for calculating the drift velocity. The model considers the axial flow through a fully developed bubble column. The reliability of this method was investigated and the evaluation of the results against the existing experimental data, considering different column diameters and flow conditions, indicated a reasonable prediction within ±20% error bands for large-diameter bubble columns while low accuracy for small ones.
Principles of superposition in viscoelastic finite-difference modelling
SUPERVISOR: Dr. Samuel Pichardo (formerly with Chemical Engineering)
ABSTRACT: The research presented in this thesis concerns the use of finite-difference time-domain (FDTD) numerical methods for use in modelling viscoelastic and acoustic wave propagation through complex media. FDTD models of viscoelasticity are widely used in the use of ultrasound for therapeutic applications, and this work includes two examples of the use of FDTD models for medical applications. The focus of the material presented is primarily concerned with a specific kind of numerical error - called the staircase error - which arises when boundaries between media do not align with the grid used in numerical simulations, and which require prohibitive amounts of time and computational resources to address. This thesis provides characterization of the numerical artifacts that arise due to the staircasing error, and presents an original numerical formulation to address these issues called the `Superposition Method'. The superposition method is characterized and shown to be an effective means of reducing these errors that does not become overly burdensome with computational time and resource requirements.
Pathogen-host omics analyses of human papillomavirus type 16 sub-lineages in a human epithelial organoid model
SUPERVISOR: Dr. Ingeborg Zehbe (Biology)
ABSTRACT: Pathogens such as human papillomaviruses (HPVs) have co-evolved with their hosts and form a molecular basis for common diseases. Persistent infection with the “high-risk” HPV type 16 (HPV16) is a potent cause of anogenital and oropharyngeal cancers. Taxonomic HPV16 sublineages, based on geographic origin of discovery, are noteworthy due to their variable tumourigenicity. In this dissertation, I present basic research and the resulting biotechnologies we developed, improved, and utilized to study their fascinating pathogen-host relationship with human stratified epithelia. A small number of variations in the E6 gene of HPV16, found in the D2 and D3 sub-lineages, lead to increased tumourigenic risk compared to the prototype A1 sub-lineage. Using an organotypic human epithelial model (or in vitro organoid) we recapitulated the viral life cycle and used “-omics” analyses to assess viral and host molecular differences due to sub-lineage variation. Sub-lineage variants of E6 were associated with host genome instability and viral integration into host DNA. Following these initial findings, I provide perspective on epithelial organoids, namely that the trade-off between model complexity and feasibility should be sensibly considered based on its utility for answering the biological research question at hand. Model applications and improvements are presented, including time-series epithelial stratification measurements, strategies for introducing full-length sub-lineage HPV16 genomes into host keratinocytes, and experiments to study innate immune evasion. These wet-lab works are accompanied by software to aid biologists in analyzing sequencing data. As well, we present current work using The Cancer Genome Atlas to test the association between HPV16 sub-lineage and integration. Overall, this interdisciplinary and interconnected collection has significance for basic researchers, providing insight on how a small number of natural viral variations can lead to increased tumourigenic risk, as well as for experimentalists to gain insight on organoid modelling and novel bioinformatics tools. More broadly, characterizing these molecular interactions between pathogen and host enables us to form a basis for diagnosis, treatment, and ultimately prevention of disease. Future research should aim to closely integrate biological and computational sciences for improving experimental approaches and our ability to make meaningful biological interpretations given the complexity and variability of biological systems.
Anaerobic membrane bioreactor for malting wastewater treatment and energy recovery under different temperature conditions
SUPERVISOR: Dr. Baoqiang Liao (Chemical Engineering)
ABSTRACT: In this dissertation, the novel application of anaerobic membrane bioreactor (AnMBR) technology for treating malting wastewater and resource recovery in the form of biogas under mesophilic (36±1 °C), room (23±1 °C), and psychrophilic (21-17 °C) temperature conditions is investigated. For each temperature condition, the effect of changing the organic loading rate (OLR) or the hydraulic retention time (HRT) on the biological treatment was evaluated by measuring the COD removal rate and efficiency, the effluent quality, as well as the biogas production yield and composition. During each study, the membrane performance was monitored by measuring the transmembrane pressure (TMP) and the permeate flux. Besides, the membrane fouling was characterized by comparing the permeability tests results, the scanning electron microscope (SEM) images, and the energy dispersive X-ray (EDX) spectra of the fouled membrane after cleaning to that of the virgin membrane. Additionally, the polymerase chain reaction-denaturing gradient gel electrophoresis (PCR–DGGE) technique was employed to investigate the bacterial community shifts as a result of changing the OLR or the HRT for each temperature condition study. The OLR ranges, during the mesophilic, room, and psychrophilic temperature studies, were 1.3-3.2, 1.2-2.9, and 1.3-2.3 kg COD/m³/d, respectively. The results of this study demonstrated that treating malting wastewater and biogas production in an AnMBR under psychrophilic (21-17 °C) temperature conditions is feasible. The results of the psychrophilic study are comparable with the performance at mesophilic conditions. Also, the feasibility of treating malting wastewater and biogas production in an AnMBR under mesophilic and room temperature conditions was confirmed. For all studies, changing the OLR and fluctuations of the influent COD did not influence the COD removal efficiencies significantly. During all studies, the AnMBR adapted readily to the changes in OLR and HRT and the BOD₅ removal efficiency was above 99%. The COD removal efficiencies ranges of mesophilic and room temperature studies were 89-96% and 86-95%. Corresponding biogas yield for mesophilic and room temperature conditions were 0.31-0.34 and 0.14-0.29 L at STP/g CODremoved. Methane content of biogas varied between 68-73% for mesophilic and 54-73% for room temperature conditions. The AnMBR operation at psychrophilic conditions resulted in 86-94% COD removal efficiency, 0.22-0.30 L at STP/g CODremoved biogas yield, and 0.92-3.26 L/d biogas with 76-84% methane content. During room temperature and psychrophilic conditions the membrane permeability results, scanning electron microscope (SEM) images, and energy dispersive X-ray (EDX) spectra implied that cake layer was the predominant fouling mechanism. However, during room temperature study, when the membraned was reused and worked for an extended duration, irreversible fouling had happened. The results of the membrane permeability recovery for all studies suggested that physical and chemical cleanings were able to recover the permeability of the fouled membrane to a large extent.Furthermore, the PCR–DGGE results of all studies showed that the microbial community actively changed due to changes in OLR, HRT, as well as the progress of the process. The literature review of the kinetic studies of conventional anaerobic process and anaerobic membrane bioreactors (from 1992 to 2017) showed that the first-order kinetic model had been used for both substrate removal and biogas production in batch bioreactors. The Michaelis-Menten model, the Monod equation, the Stover-Kincannon model, and the Grau’s second-order model were used to describe the kinetics of continuous bioreactors. The Monod equation is the most popular model used for AnMBR systems and Contois model as well as Chen and Hashimoto model were occasionally used.
Development of nanostructured material-based electrochemical sensors for food safety and quality control
SUPERVISOR: Dr. Aicheng Chen (formerly with Chemistry)
ABSTRACT: The issue of foodborne related illnesses due to additives and contaminants poses a significant challenge to food processing industries. Electrochemical-based strategies offer simple and robust analytical tools, which are ideal for food safety and the quality assessment process, in contrast to conventional instrumentation methods. The development of nanomaterials based electrochemical sensors has garnered significant attention due to their capacity for accurate analytical quantification, which has strong potential toward the replacement of conventional techniques by offering advantages such as high sensitivity and selectivity, real-time monitoring, and ease of use. During my Ph.D. study, four distinct types of nanostructured materials were used to develop electrochemical sensors for the detection of food preservatives in food and beverage products. The consumption of excessive amounts of nitrite (NO2-) can be detrimental to the human body. In light of this, we developed an electrochemical sensor based on cobalt oxide nanosheets and gold nanoparticles (Co3O4/Au) for NO2- sensing. The nanomaterial was synthesized through the electrodeposition of gold (Au) on Co3O4 nanosheets. The Co3O4/Au/GCE was capable of electrooxidizing nitrite with a higher anodic peak current, and the sensor exhibited excellent linearity with a limit of detection (LOD) value of 0.11 µM. A nanoporous gold microelectrode was synthesized for the determination of contaminants (hydrazine, N2H4) and preservatives (sulfite (SO32-), nitrite (NO2-)). The fabricated microelectrode was characterized via scanning electron microscope (SEM) and energy dispersive X-ray spectroscopy (EDX). The nanoporous gold microelectrode exhibited excellent electrochemical performance for the simultaneous electrochemical oxidation of N2H4, SO32-, and NO2-. In addition, the nanoporous gold microelectrode possessed high selectivity and stability. The performance of the electrochemical sensor was further validated using actual samples such as water, wine, apple cider beer, and beef with good recovery rates, thereby confirming its potential for food safety and quality control applications. A novel electrochemical sensor was developed using fluorine-doped graphene oxide (F-GO) for the detection of caffeic acid (CA). The fabricated nanomaterial was systematically characterized using SEM and X-ray photoelectron spectroscopy (XPS). The electrochemical investigation of F-GO/GCE for CA oxidation revealed that it demonstrated high electrocatalytic activity compared with other electrodes (e.g., bare GCE and GO/GCE). The analytical quantitation of CA recorded with the F-GO/GCE produced a stable oxidation signal over the selected CA concentration range (0.5 μM to 100.0 μM, R2 = 0.9960) with a LOD value of 0.018 μM. The fabricated sensor successfully exhibited the capacity to directly detect CA in assorted wine samples without pretreatment. To further explore the applications of the F-GO, a nanocomposite material synthesized with Au and F-GO was employed for the development of an Au/F-rGO/GCE sensor for the detection of vanillin. The electrochemical performance and the analytical capabilities of this novel electrochemical sensor were investigated using electrochemical techniques such as CV and DPV. The excellent sensitivity, selectivity, augmented electrocatalytic activity, and reproducibility of these developed electrochemical sensors can be attributed to the high conductivity of the nanostructured materials. The dimensions and morphologies of the developed nanomaterials played a critical role in enhancing the electrochemical performance of these sensors.
A novel metadata analysis approach for analyzing and understanding wood-decaying mechanisms exhibited by fungi
SUPERVISOR: Dr. Wensheng Qin (Biology)
ABSTRACT: Fuel has become an essential commodity in our day to day life. Increase in global population and decreasing fuel reserves have forced mankind to look for other fuel alternatives. Forest biomass serves as a potential renewable resource for substituting the conventional fossil-based fuels. In the last few decades, several chemical, physical and microbial based methods were being developed for the breakdown and conversion of lignocellulosic components to commercially valuable products including bioethanol and other platform chemicals. The separation of lignocellulosic biomass plays a significant role in conversion of lignocellulosic biomass to ethanol and other valuable products respectively. Naturally, lignocellulosic components are arranged in intricate networks leading to its high recalcitrance nature. Over years research groups around the world have isolated and characterized several lignocellulose degrading microorganims. Naturally, fungi play a crucial role in maintaining the geo-carbon cycle by decaying all the dead organic matter on the earth’s surface. Majority of the wood-decaying fungi are grouped under Basidiomycota division. Based on their decay patterns the basidiomycetous fungi were classified into white-rot, brown-rot and soft-rot fungi. Understanding these natural fungal decaying mechanisms will benefit the growing biofuel, biorefining and bioremediation industries. Next generation sequencing techniques have significantly enhanced our present day’s knowledge about various biological mechanisms. Phanerochaete chrysosporium was the first basidiomycetous fungi with complete annotated genome sequence, which has inspired the whole-genome sequencing studies of different wood-decaying fungi. Increasing whole genome sequencing studies have led to the enrichment of public repositories especially JGIMycoCosm, 1000 fungal genome project, Hungate 1000 projects have played a significant role in supporting these sequencing projects. As of today, there are 443 published and completely annotated fungal genome sequences in the JGI-MycoCosm repository. Thus, the availability of whole annotated genome sequences has significantly helped in designing genome-wide transcriptomic studies for understanding the molecular mechanisms underlying the process wood-decay. A total of 11 genome-wide transcriptomic studies of P. chrysosporium were reported and are publicly available under the NCBI-GEO repository. However, transcriptomic studies give a snapshot of gene expression at a given growth conditions and time period. Thus, we have developed a robust and efficient metadata analysis approach for re-analyzing gene expression datasets of P. chrysosporium and Postia placenta for understanding the common significant gene expression patterns of this model white and brown rot fungi cultured on different growth substrates (simple customized synthetic media and natural plant biomass media). We have reported a significant list of genes encoding for various lignocellulolytic enzymes significantly expressed among all the gene expression datasets. Based on the common gene expression patterns obtained from our analysis we have tentatively derived the molecular network of genes and enzymes employed during the breakdown and conversion of lignin, cellulose, hemicellulose components of plant biomass. For the first time, we have reported, and classified lignin degrading genes expressed during the ligninolytic conditions of P. chrysosporium. It is well-known that fungi experience a significant amount of stress during the process wood-decay. Plant secondary metabolites such as quinones, tannins, stilbenes, flavonoids and other phenolic compounds exhibits a fungicidal activity. Thus, wood-decaying fungi wood decaying fungi would have developed an efficient detoxification and stress responsive mechanisms for sustaining this effect. Using the results obtained in our study we have reported and classified as phase-I and phase-II metabolic genes involved in detoxification and stress responsive mechanisms respectively. Compared to P. chrysosporium, P. placenta lacks several copies of genes encoding for cellulolytic, hemicellulolytic, ligninolytic and pectinolytic enzymes. However, our gene expression metadata analysis has reported that P. placenta is strongly dependent on Fenton’s reaction for the degradation of lignocellulosic components. We have also observed that genes encoding for several hemicellulolytic enzymes were differentially expressed even during cellulolytic conditions. Based on the present metadata analysis we have also tentatively developed cellulose and hemicellulose metabolic mechanism. In the last few decades, genome sequencing studies of wood-decaying fungi have been extensively reported. Presently, the JGI-MycoCosm database resides 1165 whole genome sequences of fungi, out of which about 443 fungal genome sequences are published. It is hard to choose a particular fungus specifically for the degradation of plant biomass. Thus, we have developed an efficient metadata analysis pipeline for comparing and understanding the genome-wide annotations of fungi. The metadata analysis pipeline reported in our study can be used for selecting a wood-decaying fungus for the Invitro degradation of studies. We have compared the genome-wide annotations of about 42 wood-decaying basidiomycetous fungi (white-rot, brown-rot and soft-rot fungi) and reported a tentative comparison method explaining the total cellulolytic, hemicellulolytic, ligninolytic and pectinolytic abilities. Similarly, we have also specifically compared and analyzed the genome wide annotations of anaerobic fungi belonging to Neocallimastigomycota division fungi. This study has reported the complete genetic makeup of these peculiar fungi and their carbohydrate degrading abilities (plant cell wall carbohydrates), as they completely lack lignin degrading enzymes. We have also delineated and compared the genes coding for structural and functional components of cellulosomes and hydrogenosomes. We have also performed an extensive homology modeling and protein docking study of white-rot, brown-rot and soft-rot fungal laccases protein sequences using 6 different types of lignin model compounds. This study has revealed the structural and functional variations of white, brown and soft rot fungal laccases. Results obtained in this study reported that white and brown rot fungal laccases reported higher catalytic efficiencies compared to that of soft rot fungal laccases. This study also reported that soft rot fungal laccases exhibited small but significant variations in its structural and physicochemical properties. However, further molecular dynamic simulation and high throughput proteomic studies must be performed to understand the structural and functional properties of these laccases. The metadata analysis work frame reported in this thesis can be extended to understand the natural wood-degrading mechanisms of various microorganisms (e.g. fungi, bacteria). The highly reactive significant list of proteins obtained in this study can be used in vitro for developing highly resistant enzyme mixes used for the breakdown and conversion of various organic compounds including plant biomass. The above reported strategies also can be used as a preliminary analysis for designing large scale experiments.