Typescript. / Digitized by Kansas Correctional Industries

Microsensors and microactuators are considered to be the most crucial elements of micro-electromechanical systems (MEMS) and devices. There has been growing interest in the development of new microactuator technologies with an increasing requirement for low cost microswitch arrays providing large air gap and large force at the same time. In particular, large air gap/large force microactuators are essential for high voltage switching in automobile electronics, test equipment switchboards and in network remote reconfiguration. The necessity to reduce the size of actuators and at the same time increase the force and the air gap has placed severe constraints on the suitability of current microactuator technology for various applications. This has led to the development of new actuator technologies based on novel materials or modifying existing systems. As an effort in this direction, this thesis presents the details of the work on the design, fabrication and testing of a new hybrid microactuator, combining electromagnetic and piezoelectric actuation mechanisms. The design and fabrication of electromagnetic actuators using planar coils and a soft magnetic core has long been established. However, in many instances these designs are constrained by difficulties in the fabrication of the multi layer planar coils, which is tedious, often resulting in a low yield. Hence device performance is limited by the maximum coil currents and thereby the maximum force able to be generated. In order to overcome these problems, a hybrid actuator combining the electromagnetic system along side of a piezoelectric actuation is proposed. This has been demonstrated to assist in enhancing the total force and consequently achieving larger actuator displacements. In this research a hybrid microactuator with a footprint of 10 mm2 was designed, fabricated and tested. It can generate 330 쎠force and cover 100 쭠air gap as a microswitch. Piezoelectric actuation has been used for many applications, due to its high precision and speed. In these applications, piezo-ceramic materials, such as PZT and ZnO were commonly used because they exhibit large piezoelectric coefficients. However, there are also some difficulties associated with their use. Piezoelectric ceramic materials are usually brittle, and have a relatively large Young?s modulus, thus limiting the achievable strain. Furthermore, the deposition technologies required for preparing thin/thick films of these ceramic materials need extensive optimization. Patterning these films into required structures is also difficult. Hence, piezoelectric polymer polyvinylidene fluoride (PVDF) is chosen in this work in spite of the fact that these materials have relatively lower piezoelectric coefficients. However, the low numerical Young?s modulus values of these polymers facilitates large strain in the piezoelectric actuators. The hybrid microactuator designed in this work comprises a piezoelectric composite polymer cantilever with a planar electromagnetic coil structure beneath. The composite cantilever consists of polarized piezoelectric polymer PVDF with an electroplated permalloy layer on one side. The device includes a permalloy core at the centre of a copper micro coil with a permanent magnetic film attached on the other side of the silicon wafer (substrate) and is aligned axially with the permalloy core. The cantilever is suspended from an electroplated 150 mm high nickel post. Initially the principle was tested using hand wound electromagnetic coils with permalloy wire as the core. The performance of such a hybrid actuator was evaluated. In the next stage, a microactuator was fabricated using completely planar micro technologies, such as high aspect ratio SU-8 lithography, laser micromachining, microembossing, as well as copper and permalloy electroplating. This micro device was designed by modelling and finite element method simulation using ANSYS 7.1 and CoventorWare electromagnetic and piezoelectric solvers respectively. This helped in understanding the critical aspects of the design at the same time leading to the determination of the optimum parameters for the cantilever, micro coils and the core. An analytical model has also been developed to validate the numerical results obtained from finite element analysis. The devices were tested and the experimental data obtained were compared with the simulation results obtained from both the finite element calculations and from the analytical model. Good agreement was found between the experimental results and the simulation.

A novel design, material, and fabrication method are presented to fabricate a passive micro-ball valve. Microvalves are critical components in microflow control devices used to control the fluid flows in microchannels. These microflow control devices can be integrated with microsensors to form micro analysis systems. Glass/silicon-based fabrication is complicated and expensive. Therefore, other materials and fabrication methods have been proposed. In this research, Melinex 453, a polyester film, and pressure sensitive adhesives were used to fabricate a micro-ball valve by a microlamination method. The valve was designed to have a 450 μm diameter glass ball floating inside a chamber size of 800 μm. The ball will permit flow in the forward direction and impede flow in the reverse direction. The fabrication method consists of three steps: patterning, registration and bonding. The patterning step was accomplished using laser micromachining. Registration and bonding were performed with the use of a pin-alignment fixture. Pressure sensitive adhesive was used in the bonding step using double-sided acrylic adhesive tape. The micro-ball valve has advantages over other microvalves in terms of little dead volume, simple design, disposability, low operating pressure in forward direction, and low leakage in reverse direction. The micro-bal1 valve was characterized by pressure drop testing at different flow rates from 1 to 7.5 ml/min. The experimental results tend to agree with a simple theoretical model of the pressure drop through an orifice. Moreover, an average pressure drop diodicity of at least 2980 has been achieved. / Graduation date: 2002

A self-powered neutron detector (SPND) is a device that, coupled with a current meter, provides a readout proportional to neutron population. This thesis discusses the design parameters of an array of such devices, their characteristics, and the use of these devices as a self-powered neutron spectrometer (SPNS) to provide information about the energy distribution in a neutron radiation field. Neutron absorption in an appropriate material produces subsequent beta emissions. In a SPND, some of these beta particles will cross a non-conducting region and stop in a collector material. A net exchange of charge between these regions can be read as a current flowing between the emission region and the collector region. One potential SPNS design was modeled using a Monte Carlo simulation of the device's interaction with a radiation field. The Monte Carlo program used predicts the beta flux which is proportional to the current that would be produced by an actual device. Various beta emitting materials were considered for this device, and a sensitivity study of each was included. The design considered is comprised of a concentric set of these cylindrical SPND detector elements which, in themselves, are currently available technology. / Graduation date: 1999

This thesis investigates two methods of optical excitation of alternating-current thin-film electroluminescent (ACTFEL) devices. The two experimental methods investigated in this thesis are the photo-induced charge (PIQ) and luminescence (PIL), and the subthreshold-voltage induced transferred charge (VIQ) techniques. PIQ/PIL experiments utilize an above-bandgap laser pulse to investigate the transport properties of photo-injected electrons and holes within the phosphor layer of the ACTFEL device. VIQ experiments use a broadband xenon lamp pulse to optically reset traps which are ionized by subthreshold bipolar voltage pulses. Both experiments characterize traps within the phosphor layer. PIQ/PIL experiments are performed on evaporated ZnS:Mn ACTFEL devices possessing phosphor layers with thicknesses of 950, 700, and 300 nm. From the PIQ/PIL experiment, an impact excitation threshold electric field for evaporated ZnS:Mn is found to be ~1 MV/cm. Evidence of hole-trapping is also obtained from the PIQ experiment. The holes in evaporated ZnS:Mn ACTFEL devices are found to possess a drift length of ~180 �� 70 nm, a hole lifetime of ~2 ps, and a capture cross-section of ~7 x 10������� cm��. It is speculated that the trap responsible for hole capture is a zinc vacancy or zinc vacancy complex. VIQ experiments are performed on evaporated, atomic layer epitaxy [ALE] (Cl), and ALE (DEZ) ZnS:Mn ACTFEL devices. Data obtained via VIQ experiments yield evidence for the generation of space charge below the EL conduction threshold, as well as providing a means of estimating the physical location, energy depth, density, and capture cross-section of traps responsible for VIQ. The depth of the traps responsible for VIQ in evaporated, ALE (C1), and ALE (DEZ) ZnS:Mn are estimated to be ~1.1, ~0.3, and ~0.8 eV, respectively. It is speculated that the traps responsible for VIQ are due to sulfur vacancies, chlorine, and oxygen, for evaporated, ALE (Cl), and ALE (DEZ), respectively. / Graduation date: 1999

Recent advances in high density low cost RF and microwave three dimensional integration technologies using LTCC(Low Temperature Cofired Ceramics), laminate and other multilayer hybrid and integrated circuits have increased interest in the design of embedded passive components such as inductors, capacitors and filters. The purpose of this study is to develop the design methodology of multilayer components such as coupled line filters in a multilevel inhomogeneous medium. Although multilayer assembly including simple components have been used in the past for digital and low frequency systems, RF and microwave circuits have been fabricated mostly in single level configurations. The use of multilayer three dimensional components and circuits makes microwave circuits more compact and the design more flexible. This thesis describes the basic principles and computational procedure for the design of multilayer components such as, planar single and two-level spirals for applications as an inductive elements for RF and MICs, and coupled line band-pass filter circuits consisting of multiple sections. It is shown that both the quality factor and the inductance values can be enhanced by using multilevel spirals. Design methodology for general multisection filter consisting of asymmetric and multiple coupled lines is formulated and presented. It is shown that given the filter specifications, e.g., bandwidth, selectivity, input and output impedances, single, two and multilevel coupled line filters can be physically realized. The design procedure for narrow band filters is formulated in the conventional manner by using the equivalent circuit with admittance inverters and the component values of the low-pass prototype for Butterworth, Chebyshev and other response functions. Examples of Butterworth and Chebyshev multisection filters are included to demonstrate the design procedure. The physical multilevel filter is then optimized by using the SPICE model for coupled multiconductor lines on commercial CAD tool like LIBRA. The optimized multilevel structure design has been validated by MOMENTUM commercial electromagnetic simulator tool. The design methodology is validated by comparing the theoretical results with measurement data for a strip line filter fabricated on FR-4. / Graduation date: 1998

Graduation date: 1998

The integration of photodetectors with IC circuits provides a significant improvement over conventional designs. Featuring noise reduction, extended frequency responses, lower power consumption, and data operations, these integrated devices open challenging opportunities for many applications. One type of photodetector has the potential for important applications in the life science and remote sensing fields -- a photodetecting device that detects modulated light while rejecting ambient light. A circuit that can reject very bright ambient light yet provide high AC gain for the best signal-to-noise ratio was simulated, constructed and tested by discrete components, and excellent results were obtained. Using 80 klux tungsten light, this device detected an 0.08 lux light signal modulated at 16 kHz, rejecting more than 120 dB of DC light. This circuit was demonstrated by application to a plant physiology study, and the results were also significant. Based on a 1.2 ��m n-well CMOS process, a monolithic device that rejects DC light was designed and simulated by using HSPICE and the SWITCAP2 programs. It was found that a rejection of about 112 dB of DC light may be realized by the CMOS monolithic device. A structure extending this sensor to an imaging device that rejects DC ambient light is also proposed. / Graduation date: 1997

A relatively recent development in the late 1980s in processors has been the superscalar processor. Superscalar processors use multiple pipelines in an attempt to achieve higher performance than previous generations of processors. Having multiple pipelines makes it possible to execute more than one instruction per cycle. However, since instructions are not independent of one another, but are interdependent, there is no guarantee that any given sequence of instruction will take advantage of the wider pipeline. One major factor that governs the ability of a processor to discover parallel instructions is the processor's mechanism for decoding and executing instruction. For superscalar processors with the central window design, the number of parallel instructions discovered is dependent on the size of the window. With a large window, the probability that the processor can find more parallel instructions is higher because there are more instruction to choose from. However, the larger the window the longer the critical path and thus lower clock speed. The major theme of this thesis is to find ways to have a large instruction window but still have clock speed comparable to a small instruction window processor. One way to achieve this is to apply the idea of memory interleaving to the processor's instruction window or reservation station design. With interleaving, there are multiple small instruction windows instead of one large window. In the first cycle the first window is used, and the second window is used in the second clock cycle. After all windows are used, the processor returns to the first window. Therefore with the interleaved design only a small portion of the whole instruction window is active at one time. In this way, there can be a large virtual window. Furthermore since the size of individual window is kept small, the clock speed is not affected. The rest of this thesis will explain how this interleaved instruction window scheme works and also list some simulation results to show its performance. / Graduation date: 1997

Decision feedback equalization (DFE) is a sampled-data technique used for data recovery in digital communications channels. Multi-level decision feedback equalization (MDFE) has been developed for channels using the 2/3(1,7) RLL code. The optimum detector for a digital communication channel affected by ISI and noise consists of a matched filter, followed by a symbol rate sampler and a maximum likelihood sequence estimator. The optimal detector is unrealizable for saturation recording channels. A compromise structure uses fixed filter types with adjustable parameters. The objective is to maximize the signal-to-noise ratio in order to minimize the error rate. The read-channel waveform is corrupted at sampling instants by noise generated by various sources. We use a continuous-time low-pass filter cascaded with an all-pass filter at the receiver front-end. The low-pass filter band-limits high-frequency noise before sampling, and the all-pass filter equalizes the signal. This thesis examines different structures of the receiver and their optimal parameter placing. A design methodology developed specifically for choosing the poles and zeros location of the linear front-end part of the receiver is presented. It makes use of nonlinear optimization, and a software package written in MATLAB for equalizer computer aided design (CAD) is included in the appendix. The optimization criterion usually mentioned in the literature for digital channel optimal design is the sum of the intersymbol interference and noise. A new objective function is proposed in the thesis, and the error rate probability is shown to decrease by 30%. Issues pertaining to digital simulation of continuous-time systems are discussed. Design results are presented for different receiver structures, and bit error rate simulations are used for design validation. / Graduation date: 1997