This thesis presents the development and optimization of a high-throughput fluorescence microbead based approach for multiplexed, large scale medical diagnostics of biological fluids. Specifically, different sizes of semiconductor nanocrystals, called quantum dots, are infused into polystyrene microspheres, yielding a set of spectrally unique optical barcodes. The surface of these barcodes are then used for sandwich assays with target molecules and fluorophore-conjugated detection antibodies, changing the optical spectra of beads that have associated with (or captured) biomolecular targets. These assayed microbeads are analyzed at a single bead level in a high-throughput manner using an electrokinetic microfluidic system and laser induced fluorescence. Optical signals collected by solid state photodetectors are then processed using novel signal processing algorithms. This document will discuss developments made in each area of the platform as well as optimization of the platform for improved future performance.

This thesis discusses the preliminary development of a microfluidic system for low concentration vapour analysis incorporating a novel analyte preconcentration method to extend vapour detection limits. The topicality of this subject is evidenced by the urgent requirement to detect vapours released by explosives or their manufacturing byproducts, allied to recent reports of gas phase detection of pathogen-related chemical markers. Commercially available, non-microfluidic, sensitive, delayed response, broadly specific, gasphase analysis methods have been developed recently. However microfluidic analysis offers the prospect of both the improved specificity of liquid phase analytical methods and increased sensitivity with fast response times. The necessary conditions to achieve a viable microfluidic vapour analysis system are discussed from collection, sampling, assay and measurement perspectives. Efficient, rapid, vapour collection into a liquid phase is predicated by large surface area to volume ratio phase-interfaces, as occur within microfluidic devices. Accordingly, research has focussed on stable, segmented gas and liquid microflows. The literature has concentrated on fixed structures and precise flow rate control to produce such segmented flow. In contrast, we have investigated pressure driven flow and small active valves in combination with precision patterned passive valves to provide deterministic control over flow and thus define gas and liquid segment sizes. This has allowed introduction of larger, precise gas volumes and hence gas/liquid ratios while still maintaining more stable flow patterns than those previously reported in the literature. Ethanol was employed as a completely soluble, volatile, ‘model’ analyte to assess collection efficiency. Research into detection focussed on a number of optical methods utilising either ‘wet’ or enzymatic chemistries. The Phase-to-Phase Extraction via a Chemical Reaction to give Lower Limits of Chemical Detection hypothesis (for the purpose of brevity this is shortened to ‘Chemical Amplification’ within this dissertation) was proposed. Thorough testing of the hypothesis using an enzyme catalysed reaction scheme has demonstrated its validity, and potential value if applied to ‘real world’ systems, particularly those for detecting low solubility analytes such as the explosive 2,4,6-trinitrotoluene (TNT) or its byproduct 2,4- dinitrotoluene (2,4-DNT).

Many developed nations depend on oil for the production of gasoline, diesel, and natural gas. Meanwhile, oil shortages progress and bottlenecks in oil productions continue to materialize. These and other factors result in an energy crisis, which cause detrimental social and economic effects. Because of the impending energy crisis, various potential energy sources have developed including solar, wind, hydroelectric, nuclear, and biomass. Within the biomass sector for renewable energy sources, algae-based biofuels have become one of the most exciting, new feedstocks. Of the potential plant biofuel feedstocks, microalgae is attractive in comparison to other crops because it is versatile and doesn’t pose a threat to food sources. Despite its many advantages, the process to convert the microalgae into a biofuel is very complex and inefficient. All steps within the algae to biofuel production line must be optimized for microalgal biofuel to be sustainable. The production of biofuels from algae begins with selecting and cultivating an algae strain and giving it all the necessities to grow. The algae is then harvested and processed for specific uses. It is the harvesting or lysing step, which includes the extraction of the algal lipids, which is the biggest hindrance of algae being used as a cost effective energy source. The lysing step within the microalgal biofuel processing is of particular interest and will be the focus of this work. This work discusses the optimization of the biofuel production from microalgae biomass through computational and experimental approaches. With atomic force microscopy (AFM), a key mechanical property that would aid in the computational modeling of mechanical lysis in the in-house computational fluid dynamics (CFD) code, Particle-Surface Analysis Code (P-STAC), was determined. In P-STAC, various flow patterns were modeled that would most effectively lyse microalgal cells based on the shear stresses placed on the cells, which will be compared against microfluidic experiments using lipid specific dyes. These results would be influential in developing an energy-efficient method of processing microalgae for biofuel.

AC electrothermal flow (ACEF) is the fluid motion created as a result of Joule heating induced temperature gradients. ACEF is capable of performing major microfluidic operations, such as pumping, mixing, concentration, separation and assay enhancement, and is effective in biological samples with a wide range of electrical conductivity. Here, we report long-range fluid motion induced by ACEF, which creates centimeter-scale vortices. The long-range fluid motion displays a strong voltage dependence and is suppressed in microchannels with a characteristic length below similar to 300 mu m. An extended computational model of ACEF, which considers the effects of the density gradient and temperature-dependent parameters, is developed and compared experimentally by particle image velocimetry. The model captures the essence of ACEF in a wide range of channel dimensions and operating conditions. The combined experimental and computational study reveals the essential roles of buoyancy, temperature rise, and associated changes in material properties in the formation of the long-range fluid motion. Our results provide critical information for the design and modeling of ACEF based microfluidic systems toward various bioanalytical applications. (C) 2016 Elsevier Ltd. All rights reserved.

The analysis of biological samples in clinical or research settings often requires measurement of analytes from complex and limited matrices. Immunoaffinity separations in miniaturized formats offer selective isolation of target analytes with minimal reagent consumption and reduced analysis times. A prototype capillary-based microfluidic system has been developed for immunoaffinity separations in biological matrices with laser-induced fluorescence detection of labeled antigens or antibodies. The laboratory-constructed device was assembled from two micro syringe pumps, a microchip mixer, a micro-injector, a diode laser with fused-silica capillary flow cell, and a separation capillary column. The columns were prepared from polymer tubing and packed under negative pressure with a stationary phase that consisted of biotinylated antibodies attached to streptavidin-silica beads. A custom software program controlled the syringe pumps to perform step gradient elution and collected the signal as chromatograms. The system performance was evaluated with flow accuracy, mixer proportioning, pH gradient generation, and assessment of detectability. A direct labeling/direct capture immunoaffinity separation of C-reactive protein (CRP) was demonstrated in simulated serum. CRP, a biomarker of inflammation and cardiovascular disease risk assessment, was fluorescently labeled in a one-step reaction and directly injected into the system. A quadratic calibration model was selected and precision and accuracy were reported. Parathyroid hormone was also analyzed by the direct capture approach, but displayed nonspecific binding of human plasma matrix components that limited the useful assay range. Capillary sandwich assays of CRP in human serum and cerebrospinal fluid were performed using both capture and detection antibodies. The detection antibody was labeled and purified offline to minimize signal from labeled matrix components. Four parameter logistic functions were used to model the data and precision and accuracy were evaluated. During the study, 250 nL injection volumes 2.0 µL/min flow rates were employed, minimizing sample and reagent consumption. The microfluidic system was capable of separating antigens from biological matrices and is potentially portable for patient point-of-care settings. Additionally, the flexible design of the separation capillary allows for the analysis of different clinical markers by changing the antibodies and the low assay volume requirements could lead to less invasive patient sampling techniques.LabVIEW version 7 or later is required to open the attached files.

Dans ce travail de thèse, nous avons développé deux méthodologies permettant d'acquérir rapidement et facilement des propriétés physico-chimiques, cinétiques et thermodynamiques de fluides complexes. Nous nous sommes focalisés sur le rôle de la concentration sur ces propriétés. Les deux méthodes développées sont basées sur la concentration en continu d'une solution aqueuse par évaporation contrôlée du solvant. Le premier outil est une goutte de quelques microlitres confinée entre deux plaques dont la hauteur est de 100µm. Dans cette géométrie à deux dimensions, l'évaporation est entièrement décrite par un modèle que nous avons développé. L'observation du séchage de la goutte couplée à des mesures locales de concentration par spectroscopie Raman, permet d'accéder quantitativement au diagramme de phase d'une solution de copolymères, et de mesurer l'activité ainsi que d'estimer le coefficient d'interdiffusion de la solution. Le second outil est une puce microfluidique permettant de concentrer des solutions aqueuses grâce à la pervaporation de l'eau à travers une membrane. Cet outil permet avec quelques microgrammes de soluté, de bâtir un gradient de concentration stationnaire le long d'un microcanal. Les techniques de spectroscopie Raman et de diffusion des rayons X aux petits angles permettent à nouveau de mesurer des propriétés physico-chimiques de la solution mais également de mettre en évidence le caractère discontinu du coefficient d'interdiffusion en fonction de la concentration, dépendant des mésophases présentes. / In this work, we developed two methods in order to access rapidly and easily physico-chemical, thermodynamic and kinetic properties of complex fluids. We focused on the role of the concentration on these properties. The two methods that we developed are based on the continuous concentration of an aqueous solution thanks to the evaporation of the solvent. The first tool is a microliter droplet confined between two circular plates with a cell height of about 100 µm. Within this two dimensional cylindrical geometry, the evaporation of the droplet is totally described by a model that we developed. The observation of the droplet evaporation combined to local Raman spectroscopy measurements permits us to build a quantitative phase diagram, to measure the activity of the solution and to estimate its mutual diffusion coefficient. The second tool is a microfluidic chip in which water is removed through a thin membrane. This device permits us to build with a few micrograms of solutes a stationary concentration gradient along a microchannel. Raman confocal spectroscopy and small angle X-ray scattering give access to the quantitative phase diagram and also permit to evidence that the mutual diffusion coefficient is discontinuous at some of the phase boundaries.

Master of Science / Department of Biological & Agricultural Engineering / Mei He / Anemia affects about 25% of the world’s population and causes roughly 8% of all disability cases. The development of an affordable point-of-care (POC) device for detecting anemia could be a significant for individuals in underdeveloped countries trying to manage their anemia. The objective of this study was to design and fabricate a 3D printed, low cost microfluidic mixing chip that could be used for the diagnosis of anemia. Microfluidic mixing chips use capillary flow to move fluids without the aid of external power. With new developments in 3D printing technology, microfluidic devices can be fabricated quickly and inexpensively. This study designed and demonstrated a passive microfluidic mixing chip that used capillary force to mix blood and a hemoglobin detecting assay. A 3D computational fluid dynamic simulation model of the chip design showed 96% efficiency when mixing two fluids. The mixing chip was fabricated using a desktop 3D printer in one hour for less than $0.50. Blood samples used for the clinical validation were provided by The University of Kansas Medical Center Biospecimen Repository. During clinical validation, RGB (red, green, blue) values of the hemoglobin detection assay color change within the chip showed consistent and repeatable results, indicating the chip design works efficiently as a passive mixing device. The anemia detection assay tended to overestimate hemoglobin levels at lower values while underestimating them in higher values, showing the assay needs to go through more troubleshooting.

The focus of this project is to find a suitable method to determine the mixing ratio inbinary fluid mixtures in continuous-flow microfluidic systems because of thedifficulties in doing so for mixtures containing compressible fluids. Refractive indexand relative static permittivity are both properties that could be suitable, but methodsmeasuring the refractive index scales badly for microsystems. A microfluidic chip for measuring capacitance was placed on a PCB together with amixing structure with strain-relieved fluid and electrical interfaces. This PCB was builtinto a rig with two piston pumps and a backpressure regulator to makemeasurements of the relative static permittivity of air, ethanol, methanol, acetonitrile,liquid and gaseous carbon dioxide, as well as of several mixtures of ethanol andcarbon dioxide using a Network Analyzer. Several other measuring techniques were tried, but the Network Analyzer wassuperior in accuracy, stability and frequency range. It produced values within 4% ofthe theoretical, and the discrepancy could be explained by the approximations in theparallel plate capacitor formula, the capacitance contributions of the external parts ofthe system and surface roughness. The Network Analyzer is a good tool to determinethe mixing ratio in binary fluid mixtures in continuous-flow microfluidic systems.

Pancreatic Ductal Adenocarcinoma (PDAC) continues to have a dismally low survival rate due to late diagnosis and poor treatment options. Therefore, there is a need to understand the early stages and progression of the disease. PDAC is known to have two types of cells of origin: ductal cells or acinar cells. Since acinar-derived PDAC is thought to be the more malignant of the two, it was chosen as the focus of this work. Most studies of acinar cells as they relate to PDAC are accomplished by using animal models such as genetically engineered mouse models. While this method yields a large amount of insight into the progression of the disease and the role of specific genes, it has the drawbacks of being very time and resource intensive. The quicker and less costly alternative is <i>in vitro </i>culture. Specifically, here we have developed a microfluidic model which can incorporate a key aspect of the extracellular matrix (ECM), type I collagen, and mimics the 3D geometry of an <i>in vivo </i>acinus. Most attempts at <i>in vitro </i>culture have been limited by the fact that isolated acinar cells show a decrease in the amount of digestive enzymes they secrete as culture continues. For this reason, we are using a reprogrammed cancer cell line. These cells can be induced with doxycycline to express PTF1a, which allows the cells to adapt acinar characteristics, such as the production of digestive enzymes. We were able to successfully culture and induce PTF1a in these cells within our chip. We showed that the cells exhibit no invasion into the collagen matrix once PTF1a is expressed, thus eliminating a key aspect of cancer cell culture. The cells grown in the chip are confirmed to be producing PRSS2, the digestive enzyme trypsinogen. Collectively, this suggests that we have produced healthy acinar cells growing in the same configuration that they would <i>in vivo. </i>This has many applications in the study of pancreatic ductal adenocarcinoma, as we have developed way to culture reprogramed cancer cells as their benign precursors and maintain acinar characteristics <i>in vitro.</i> It will also have applications in the study of many other pancreatic diseases by providing an <i>in vitro</i> model of a healthy, functional acinus.

Thesis advisor: Michelle Meyer / The primary bacterial targets for most antibiotics are well known. To survive the stress of an antibiotic a bacterium must decrease the antibiotic to target binding ratio to escape from harmful effects. This can occur through a number of different functions including down-regulation of the target, mutation of the binding site on the target, and decreasing the intake or increasing the efflux of the antibiotic. However, it is becoming more evident that an antibiotic stress response influences more than just the primary target, and that a wave of secondary responses can be triggered throughout the bacterium. As a result resistance mutations may arise in genes that are indirectly affected by the initial interaction between the antibiotic and target. These indirect responses have been found to be associated with metabolism, regulation, cell division, oxidative stress, and other critical pathways. One technique recently developed in our lab, called transposon insertion sequencing (Tn-seq), can be used to further understand the complexity of these indirect responses by profiling growth rates (fitness) of mutants at a genome-wide level. However, Tn-seq is normally performed with large libraries of pooled mutants and thus it remains unclear how this may influence fitness of some independent mutants that may be compensated by others in the population. Additionally, since the original method has only utilized planktonic culture, it is also not clear how higher order bacterial structures, such as biofilms or microcolonies, influence bacterial fitness. To better understand the dynamics of pooled versus individual mutant culture, as well as the effect of community structure in microcolony development on the influence of fitness, we adapted a droplet microfluidics-based technique to encapsulate and culture single mutants. We were able to successfully encapsulate at least 7 different species of bacterial pathogens, including Streptococcus pneumoniae, and culture them planktonically, or as microcolonies, in either monodisperse liquid or agarose droplets. These experiments, however, raised an important challenge: the DNA yield from one encapsulation experiment is insufficient to generate samples for sequencing by means of the traditional Tn-seq method. This led us to develop a novel Tn-seq DNA library preparation method, which is able to generate functional Tn-seq library molecules from picogram amounts of DNA. This method is not ideal yet because fitness data generated through the new method currently does not correlate well with data from traditional Tn-seq library preparation. However, we have identified one major culprit that should be easily solvable. We expect by modifying the binding site of the primer used for linear amplification of transposon ends that the new preparation method will be able recapitulate results from the traditional Illumina preparation method for Tn-seq. This will enable us to prepare robust Tn-seq samples from very small amounts of DNA in order to probe stress responses in single mutants as well as in microcolonies in a high-throughput manner. / Thesis (MS) — Boston College, 2016. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Biology.