The main aim of this study is to develop three dimensional models for micro flow-field plate of PEMFC and use numerical simulations to discuss the reliability of micro flow-field plate which works in real. A crack exists in the plate is loaded by the shear force, which is produced by the fuel H2 enter from inlet, and will propagate. The Ag, which is used to collect the electrons, will peel off and the efficiency of fuel cell will decrease. The commercial package software ANSYS was used to simulate the stress state around crack tip. Three modes of stress intensity factors K£L, K£L£L and K£L£L£L, were calculated in order to describe the stressed behavior of crack. Finally, the inlet pressure, geometry of crack and channel size is changed and Taguchi method with ANOVA is used to find the factors which influence the stressed behavior of crack most. The simulation results show that K£L and K£L£L are influenced most by geometry of crack and K£L£L£L is influenced more by geometry of crack and channel size

Aggregate properties influence different aspects of asphalt pavement performance. Aggregate polishing characteristics are directly related to pavement surface frictional properties and thus to skid resistance. Aggregate resistance to degradation (abrasion and breakage) is another important property that influences pavement performance. Aggregate degradation could take place during production due to plant operations and during compaction, leading to change in aggregate characteristics and mix properties. In addition, aggregate resistance to degradation is important in mixes such as Stone Matrix Asphalt (SMA) and Open Graded Friction Course (OGFC) that rely on stone-to-stone contacts among coarse aggregates. Some aggregates in these mixes fracture due to the high stresses at contact points. Many test methods exist for measuring aggregate polishing and degradation, but a critical review of these methods reveals that they suffer from being time consuming, are unable to differentiate between aggregates with distinct resistance to polishing, or unable to differentiate between aggregate resistance to abrasion and breakage. New methodologies are needed to give better assessment of aggregate resistance to polishing, abrasion, and breakage. The thesis presents the development of new methods for measuring aggregate resistance to polishing, abrasion, and breakage. These methods rely on measurements using the Aggregate Imaging System (AIMS) and Micro-Deval. The new method for measuring aggregate resistance to polishing monitors change in aggregate texture as a function of polishing time. As such, it provides the initial texture, rate of polishing, and final texture. The new method for measuring aggregate degradation is capable of distinguishing between breakage and abrasion. In this method, abrasion is defined as the reduction in aggregate angularity, while breakage is defined by fracture of particles. The new methods are shown to be rapid and accurate, and they require reasonable training. Since both AIMS and Micro-Deval are used in the new methods, it was necessary to evaluate the repeatability of these two methods. Measurements using two AIMS units and two Micro-Deval machines were used to assess the variability. There was no statistical difference between the measurements of the two AIMS units or between the measurements of the two Micro-Deval units.

Many applications in microelectromechanical systems such as smart living space sensing system, microchannel system on chip and biomedical sensing system usually require instantaneously compensating or controlling the temperature of chip to acquire more linear and accurate output signal. So it is necessary to develop a micro temperature sensor or micro-hot-plate which has highly thermal isolation and low power characteristics. This thesis aims to design and fabricate a low power micro-hot-plate and a high-sensitivity temperature sensor for portable applications. This dissertation utilized a high power E-beam evaporator to deposit the platinum thin film as the material of temperature sensing and heating. The Pt layer is patterned using the lift-off technique. In addition, the micro-hot-plate can be released from the silicon substrate as a floating thin-plate using TMAH-based anisotropic etching technology. The floating structure can improve the thermal isolation and reduce the power loss through the silicon substrate. In this study, the higher temperature sensitivity (1914 ppm/¢J) and optimized sensing linearity ( > 99.9 %) of the platinum-based temperature sensor is demonstrated. On the other hand, the heating power of the floating micro-hot-plate developed in this research is only 14 mW when it be heated to 300 ¢J and the power efficiency is very high (18.3 ¢J/mW).

Before studying to increase the catalyst utilization is one way to improve the performance of a fuel cell. But because it wastes a lot of time on making the hydrophobic pillared micro structures (HMPS) process. So this study aims to develop a fast and effective manufacturing method in order to increase the reaction surface area and catalyst utilization. The experiment has used the metal mesh stamping method to make the specific structure so as to increase the reaction surface area. The size of the metal mesh was the line path 35£gm, and the net square 70£gm*70£gm. The pressure was 300kg/cm2 and 500kg/cm2 that made the stamping structure in order to increase the reaction surface area. When the reaction surface area nearly increased 27 %, the performance also nearly increased 27 %; And when the reaction surface area nearly increased 36 %, the performance also nearly increased 36 %. So the increment of its performance nearly accorded with the increment of the reaction surface area. This method has saved a lot of time in the production process. Furthermore, the catalyst loading of cathode is halved in this experiment, the performance of fuel cell have no obviously decreased or reduced by half. As a result, the utilization of catalyst is raised.

This paper presents an integrated column-type micro thermoelectric cooler (£g-TEC) constructed with serial connected p-type antimony-tellurium (Sb2Te3) and n-type bismuth-tellurium (Bi2Te3) micro pillars deposited by electrochemical deposited technology. To optimize the power factor, density and uniformity of the TE films and to enhance the reproducibility of £g-TEC device, a cathode with tunable rotary speed and with accurate current controller has been designed in the electroplating system of this research. The electroplating deposited Bi2Te3 and Sb2Te3 with an average thickness of 8 £gm, are connected using Cr/Au layers at the hot junctions and cold junctions. The measured Seebeck coefficient and electrical resistivity are -86 £gV/K and 16 £g£[-m for Bi2Te3 films after annealed at 250¢XC, and are 68 £gV/K and 30 £g£[-m for Sb2Te3 films after annealed at 200¢XC. The optimized power factors of the n-type (2.64¡Ñ10-4 W/K2m) and p-type (2.64¡Ñ10-4 W/K2m) telluride compounds have been demonstrated in this paper. Under 5 volts driven, the integrated £g-TEC device shows average cooling achieved is about 1.3 ¢XC.

The passive devices used in the wireless communication system ¡]including resistor, capacitor and inductor¡^usually need high quality factor and low power dissipation characteristics. This thesis aims to develop a micro tunable capacitor with high-quality-factor and wide-tuning-range using surface micromachining. In contrast with conventional low-tuning-rate parallel-plate tunable capacitors, this research presents a concave structure and eight-suspending-beams layout design of the top electrode to enhance the elastic rigidity and tuning rate. In addition, this study appropriately decreases the thickness of top electrode, the tuning rate of such device can be improved to 65~2100%. On the other hand, in order to substantially increase quality factor, this thesis adopted the glass substrate ¡]Corning 7740¡^to reduce the power dissipation of high frequency operating signal. The optimized quality factor of this work is approximately equal to 41 under 2.4 GHz operation frequency. The material of sacrificial layer and top electrode adopted in this dissertation is aluminum and gold respectively. To avoid any breakage of the vertical supporting beams during releasing process, this research appropriately increases the width of vertical supporting beams, however, keep the thickness of the suspending part of top electrode for the maintenance of high quality factor and low driving voltage.

none

This paper aims to design a high-quality-factor suspending micro-inductor and to establish its equivalent circuit model for performance optimization. Two commercial software (Ansoft HFSS and Agilent ADS) are adopts to analysis the influences of quality factor on the geometric parameters and substrate materials. The designed micro-inductors are constructed by one bottom GSG electrode, two supporting copper vias and a spiral suspending copper conducting layer. As the simulated results of this research, the quality factor of the suspending micro-inductor is increased with the height of air gap, the thickness and width of suspending copper conducting layer and decrease with the number of turns, line space and outer diameter of suspending copper conducting layer. The influences of different shapes of the spiral suspending copper conducting layers on the quality factor of micro-inductors were also investigated. The simulation results well match to the theoretical prediction. Finally, this thesis has successfully derived two experiential formulas based on the analysis results to estimate quickly the inductance of the suspending micro-inductors with circular and square shape. Compared with the simulation results and realistic measurement results, these experiential formulas demonstrate 94-95% and 90% accuracies respectively.

With the progress in VLSI technology, the external environment makes it easier for the interference affected the operation of microcontroller. The design of the recently microcontroller, not only the pursuit of speed and performance, also began the study of the various fault-tolerant technology to enhance the reliability and safety. This thesis, being designed for the Fault-tolerant microcontroller according market, presents a Robust Micro-Control Unit : RMCU for dual core architecture of ARM9 ISA. The RMCU provides two operation modes: synchronize mode and Processor test mode for fault-tolerant mechanism. In synchronize mode, both processors are executing the same program concurrently. The results generated by processors are compared, and every mismatch indicates a transient fault in one of the two processors. When the transient fault occurred, the two processors will use Instruction retry mechanism, recover system operation. If the same address's errors larger than the number of settings are considered permanent fault, processors will be held, and entered the processor test mode for processor functional test. In accordance with the test results to close the wrong processor and operating system back to normal. This approach to solve the traditional dual-core processor fault-tolerant architecture that can not be fixed to permanent-fault restrictions. In addition to the design of fault-tolerance mechanism, for the upgrading of software and hardware development and validation of this paper design of the RMCU debug platform. RMCU debug platform including JTAG-based OCD (On-Chip Debugging) unit, and debug interface program. In addition to providing read and write registers and memory, set Breakpoint, Watchpoint and single-step but also take the initiative to increase the external interrupt inserted to provide a more effective ISR (Interrupt Service Routine) debug. In the last of the thesis, we use the FPGA Implementation of the RMCU fault-tolerant mechanisms and debug platform. After simulation and testing, the results prove the feasibility of RMCU.

In this study, complex rolling technology is adopted to produce aluminum/copper clad metals . The aluminum alloy A1050 and the copper C1100 are used. The experimental plan : the roughness (RCu=3.5£gm¡ARAl=6.45£gm)¡A(RCu=1.0£gm¡ARAl =2.15£gm)¡A(RCu=0.35£gm¡ARAl=0.8£gm) the thickness ratio(tAl:tCu=3:1)¡A(tAl:tCu =2:2)¡A(tAl:tCu=1:3) reduction 60% and 70% are set. It shows that the influence of the thickness is more significant than the roughness on the curvature according to the experimental results. And the results of peeling tests show that the peeling strength for the reduction 70% is larger than that for 60%. And peel strength with the second time rolling is larger than that with only once. The average peeling strength of the specimen in the rolling direction is larger than that in the perpendicular direction. From the micro Vickers hardness tests , it is known that the larger of the reduction of copper is, the larger the micro Vickers hardness between the interface of the sheets is .