Three-dimensional spheroid arrays represent in vivo activity better than conventional 2D cell culturing. A high-throughput microfluidic chip may be capable of depositing cells into spheroid arrays, but it is difficult to regulate the path of individual cells for deposition. Droplets that encapsulate cells may aid in facilitating cell delivery and deposition in the return bend of a microfluidic chip. In this study, a low-cost method for fabricating polymer-cast microfluidic chips has been developed for rapid device prototyping. Computational fluid dynamic (CFD) simulations were conducted to quantify how a change in geometry or fluid properties affects the dynamics of a droplet. These simulations have shown that the deformation, velocity, and trajectory of a droplet are altered when varying the geometry and fluid properties of a multiphase microfluidic system. This quantitative data will be beneficial for the future design of a microfluidic chip for cell deposition into 3D spheroid arrays.

<p>Sample preparation is a crucial processing step required for molecular biological analysis of environmental samples like water that has a variety of constituents in it. Furthermore, large volumes of sample need to be processed as the prescribed limits of pathogens in water are extremely low. However, microfluidic biosensing devices that can perform rapid molecular biological analysis in the field are designed to handle small sample volumes. In such cases, there is a need for a sample processing device that can reduce (concentrate) a large sample volume into a small one while retaining the biological species present in it. Hollow fibers are appropriate for this purpose of sample reduction and serve as a macro to micro interface for the microfluidic device. The received concentrate from the hollow fiber device requires be further concentrated to several microliters and separated and sorted to various modular components within the microfluidic device. This requires a second stage microfiltration where an integrated membrane can sort based on particulate size. In this thesis, a two stage filtration was designed. A first stage hollow based fiber pre-concentration device is developed that is portable, low cost, has high retention efficiency, low elution volume and is rapid. The hollow fiber device has low elution volume of ~1-3 ml. Controlled experiments were performed to validate the recovery of the hollow fiber device. Simulated 250 ml E.coli contaminated samples were filtered to <5 ml from an original sample volume of 250 ml. No bacteria were present in the filtrate and nearly 100% was recovered at high bacterial concentrations. At low concentrations (~200 cells in the sample) the recovery was less (~50%). A second stage microfiltration device that can be integrated with the microfluidic device and that can reduce the sample still further from ~ 5 ml to 5 μl was designed. Plasma bonding of ultrafiltration and microfiltration membranes using fluorine ions was investigated for fabrication of this device. The bonding of PDMS channels with polysulfone membranes via SF6 plasma was tested via tensile pull tests, burst pressure tests, and analyzed through scanning electron microscopy and electron dispersive x-ray spectroscopy. Quantitative tests on 10kDa and 70kDa polyethersulfone membranes demonstrated increased operational bonding strength of 86.6 and 146.9 kPa increases with three hour plasma application. Microfiltration membranes (0.2 micrometer pore size polyethersulfone) bonded in such a way that was easier to permeate as compared to ultrafiltration membranes. This bonding technique is generic in nature and can be used for integration of other commercially available polyethersulfone membranes with microfluidic devices for applications such as bio separations. No filtration testing was performed with E.coli samples.</p> / Master of Applied Science (MASc)