High performance inductors are playing an increasing role in modern communication systems. Despite the superior performance offered by discrete components, parasitic capacitances from bond pads, board traces and packaging leads reduce the high frequency performance and contribute to the urgency of an integrated solution. Embedded inductors have the potential for significant increase in reliability and performance of the IC. Due to the driving force of CMOS integration and low costs of silicon-based IC fabrication, these inductors lie on a low resistivity silicon substrate, which is a major source of energy loss and limits the frequency response. Therefore, the quality factor of inductors fabricated on silicon continues to be low. The research presented in this thesis investigates amorphous Si and porous Si to improve the resistivity of Si substrates and explores amorphous Si as a structural material for low temperature MEMS fabrication. Planar inductors are built-on undoped amorphous Si in a novel application and a 56% increase in quality factor was measured. Planar inductors are also built-on a porous Si and amorphous Si bilayer and showed 47% improvement. Amorphous Si is also proposed as a low temperature alternative to polysilicon for MEMS devices. Tunable RF MEMS inductors and transformers are fabricated based on an amorphous Si and aluminum bimorph coil that is suspended and warps in a controllable manner. The 3-D displacement is accurately predicted by thermomechanical simulations. The tuning of the devices is achieved by applying a DC voltage and due to joule heating the air gap can be adjusted. A tunable inductor with a 32% tuning range from 5.6 to 8.2 nH and a peak Q of 15 was measured. A transformer with a suspended coil demonstrated a 24% tuning range of the mutual coupling between two stacked windings. The main limitation posed by post-CMOS integration is a strict thermal budget which cannot exceed a critical temperature where impurities can diffuse and materials properties can change. The research carried out in this work accommodates this temperature restriction by limiting the RF fabrication processes to 150°C to facilitate system integration on silicon.

This study investigated the behavior of porous silicon gas sensors under exposure to CO, NO, and NH3 gas at the part per million level. Parameters of interest in this study included the electrical, environmental, and chemi-resistive performance associated with various porous silicon morphologies. Based upon the variability of preliminary results, a gas pulsing method was combined with signal processing in order to analyze small impedance changes in an environment of substantial noise. With this technique, sensors could be effectively screened and characterized. Finally this method was combined with various post-treatments in order to improve the sensitivity and selectivity of individual sensors.

This dissertation describes the fabrication and operation of porous silicon gas sensors. The first chapter describes the motivation behind gas sensor research and provides the reader with background knowledge of gas sensors including the terminology and a review of various gas sensors. The following two chapters describe both how the porous silicon gas sensors are created and how they have been tested in the laboratory. Chapter 4 describes the steps required to create arrays of gas sensors to provide for a selective device through the application of patented selective coatings. Chapter 5 proposes a physical model that leads to a numerical solution for predicting the operation of the gas sensor. The next chapter builds from this model to analyze and optimize the experimental methods that are used to test both this and other gas sensors. The final chapter of this dissertation describes the prototype gas sensor system that has most recently been created, the company that was formed to further the development of that system, and the future applications of the porous silicon gas sensor.

Synthetic diamond thin films have potential for fabricating high-temperature semiconducting and optical devices because of its extraordinary properties. In this work, a microwave plasma chemical vapor deposition system has been setup. A two-steps deposition process will be applied for the growth of boron-doped diamond on silicon and on porous silicon. The effects of temperature, microwave power and of doping concentration of B2O3 have been studied by varying the growth parameters. The doping source of B2O3 solved in C2H5OH is applied with carrying gas of Ar. To vary the concentration of boron with the flow of Ar is controlled mixing into a reaction gas of CH4 and H2 mixture. Polycrystalline diamond thin films are examined by Raman, XRD and FTIR. In the SEM photograph a nano-wires structure has been found for higher doping of B2O3. A higher temperature the growth rate of the boron-doped diamond films will increase and the shape of crystallites will tend to polycrystalline. The diamond growth is in multi steps and the mechanism of deposition will change when the boron-doped diamond film grows up to a critical thickness. In this work a smooth diamond film was successfully grown on porous silicon without the step of nucleation.

Hepatocellular carcinoma (HCC) is the third most common cause of cancer-related deaths worldwide, accounting for over 600,000 deaths per year. The most common treatment strategy for intermediate and advanced stage unresectable HCC is transarterial chemoembolization (TACE), which involves the local administration of a chemotherapeutic drug combined with arterial occlusion resulting in ischemic tumor necrosis. However, TACE suffers from inadvertent exposure of noncancerous liver parenchyma to embolic agents resulting in liver injury. In some cases, over-embolization has lead to infection, necrosis of unaffected liver tissue, and even liver failure which suggests the need for a biocompatible, multifunctional embolic material which can deliver anticancer drugs with high target specificity. Our laboratory has recently developed a method to fabricate porous silicon (pSi) microparticles with defined physicochemical properties based on photolithography and anodic etching. These microparticles function as multistage drug delivery systems that can circumvent the biobarriers present in the systemic circulation enabling site-specific localization and release of chemotherapy and imaging agents. The versatility of the fabrication process enables the realization of microparticles ranging in size from 600nm to 116[mu]m in diameter with varying shapes, including discoidal, cylindrical and hemispherical, and varying porosity with pore sizes ranging from 6nm to greater than 50nm in diameter. Nanoparticles, such as quantum dots, siRNA-loaded nanoliposomes, gadolinium-based contrast agents, gold and iron oxide nanoparticles, are loaded in pSi microparticles by tailoring their pore sizes and surface chemistries. This thesis presents preliminary results on the applicability of biocompatible, engineered pSi microparticles as an embolic agent for HCC chemoembolization therapy. Hemispherical microparticles with 116[mu]m diameter were successfully fabricated and suspended in phosphate buffered saline (PBS). A microvascular construct was rapid prototyped in polydimethylsiloxane (PDMS) as an in vitro experimental platform to study the embolization behavior of pSi microparticles. Oxidized pSi microparticles were introduced into the microfluidic device at an appropriate flow rate and time-lapse images were taken showing the formation of occlusions at the bifurcation within minutes of administration. Furthermore, penetration through the bifurcation was completely hindered suggesting that pSi microparticles can potentially be used as a biocompatible, multifunctional chemoembolization agent. Although these results are promising, further investigations are warranted.

The rational and directed delivery of genetic material to the cell is a formidable tool to investigate the phenotypic effects of gene expression regulation and a promising therapeutic strategy for genetic defects. RNA interference constitutes a versatile approach to gene silencing. Despite the development of numerous strategies the transfection of small interfering RNA (siRNA) is highly dependent on cell type and conditions. Direct physical access to the intracellular compartment is a promising path for high efficiency delivery independently of cell type and conditions. Silicon nanowires grant such access with minimal toxic effects, and allow intracellular delivery of DNA when actuated by atomic force microscope. These findings reveal the potential for porous silicon nanostructures to serve as delivery vectors for nucleic acids due to their porous nature, elevated biocompatibility, and biodegradability. This dissertation illustrates the development a novel platform for efficient siRNA transfection based on an array of porous silicon nanoneedles. The synthesis of biodegradable and biocompatible porous nanowires was accomplished by a novel strategy for electroless etch of silicon that allows anisotropic etch simultaneously with porosification. An ordered array of cone shaped porous silicon nanoneedles with tunable tip size, array density and aspect ratio was obtained coupling this strategy with patterned metal deposition and selective reactive ion etch. This process also granted control over porosity, nanopore size and flexural modulus. The combination of these parameters was appropriately optimized to ensure cell penetration, maximize siRNA loading and minimize cytotoxic effects. Loading of the negatively charged siRNA molecules was optimized by applying an external electric field to the nanoneedles under appropriate voltage conditions to obtain a tenfold increase over open circuit loading, and efficient penetration of the siRNA within the porous volume of the needles. Alternative surface chemistry modification provided a means for effective siRNA loading and sustained release. siRNA transfection was achieved by either imprinting the nanoneedles array chip over a culture of MDA-MB-231 cells or allowing the cells to self-impale over the needles. The procedures allowed the needles to penetrate across the cell membrane without influencing cell proliferation. siRNA was successfully transfected and was effective at suppressing gene expression. / text

No description available.

No description available.

This thesis describes the fabrication of a set of bright, visible light-emitting silicon LEDs. These devices were fabricated in-house at the University of Saskatchewan using a custom plasma ion implantation tool, an annealing furnace, and a physical vapour deposition system. A high-fluence (F = 4 × 1015 cm^−2) implantation of molecular hydrogen ions extracted from an RF inductively coupled plasma at an energy of 5 keV was used to create a heavily damaged region in the silicon centered approximately 40 nm below the silicon surface with a width of approximately 56 nm. A matrix of annealing (e.g. thermal processing) processes at 400 ºC and 700 ºC and different durations (30 minutes and 2 hours) as well as an aluminum gettering procedure were tested with the goal of increasing the output electroluminescence intensity. Current-voltage characterization was used to extract information about the defect-rich nanocrystalline, light-emitting layer as well as the Schottky barrier height. This enabled comparison of these new devices with previous silicon LEDs based on porous silicon and other approaches. The processes which were used to fabricate these devices are compatible with standard CMOS processing techniques and could provide one solution to the problem of optical interconnect on multi-core chips. The scientific significance of this work is the demonstration of bright, visible light emission at mean photon energies ∼1.84 eV corresponding to a photon wavelength of λ ≈ 675 nm. This is remarkable given that ordinary crystalline silicon is an indirect bandgap material with a bandgap energy of 1.1 eV, in which band-to-band radiative recombination is forbidden by momentum conservation. The devices fabricated in this thesis have light emission properties similar to previous silicon LEDs based on nanocrystalline or nanoporous silicon. They have the advantage of being easily electrically driven. The nanocrystalline region which is the source of the light emission was nucleated from the ion-implanted layer below the surface of the silicon. This makes these devices mechanically robust and insensitive to environmental conditions. The engineering significance of this work is the production of CMOS compatible light emitters. This study demonstrated increased light emission efficiency at higher annealing temperatures which is likely due to enhanced diffusion and nucleation of silicon nanocrystals in the ion-implant damaged layer.

A novel approach to the design of multi-junction solar cells on silicon substrates for 1-sun applications is described. Models for device simulation including porous silicon layers are presented. A silicon bottom subcell is formed by diffusion of dopants into a silicon wafer. The top of the wafer is porosified to create a compliant layer, and a III-V buffer layer is then grown epitaxially, followed by middle and top subcells. Due to the resistivity of the porous material, these designs are best suited to high efficiency 1-sun applications. Numerical simulations of a multi-junction solar cell incorporating a porous silicon compliant membrane indicate an efficiency of 30.7% under AM1.5G, 1-sun for low threading dislocation densities (TDD), decreasing to 23.7% for a TDD of 10^7 cm^-2.