No abstract

This paper reports on the dynamic characteristics of a MEMS electrostatic harvester employing interdigitated gap-closing topology. Devices are fabricated using SOIMUMPS technology and are characterized with and without biasing voltages for a broad range of excitation accelerations. At low vibration amplitudes the presence of a dc bias causes the resonant frequency peak to shift to lower frequencies with increasing bias. At higher vibration amplitudes the dynamic response of the devices exhibits the behavior of a Duffing oscillator with spring softening due to nonlinear stiffness attributed to the effect of electrostatic forces. Specifically, the devices exhibit sweep direction hysteresis with jump phenomena due to the multivaluedness of the response curve. Amplitude sweeps at constant frequency and varying bias voltage also show jump phenomena, highlighting how slight differences in operating conditions dramatically affect device performance. Spring hardening effects are reported for devices contaminated with dust particles. The paper also discusses SOIMUMPS limitations, the importance of reducing off-axis vibration during testing, characterization methods, and the effect of grounding on parasitic capacitance.

Mechanical to electrical energy conversion employing variable capacitors is assisted by electronic circuits that can have synchronous or asynchronous architectures. The later does not require synchronization of electrical events with mechanical motion, which eliminates difficulties in gate clocking and the power consumption associated with intelligent control circuitry. However, implementation of asynchronous energy harvesting circuits with the mechanical-to-electrical converter can be detrimental to the performance of the converter when done without concurrent optimization of the mechanical device and the circuit, an aspect mainly overlooked in the literature. This paper carries out system level analysis of electrostatic micro-generators with asynchronous control and charge fly-back mechanism to optimize the useful energy generated by the harvester. Our theoretical and experimental investigations show that there is an optimum value for either the storage capacitor or cycle number for maximum scavenging of ambient energy via asynchronous electrostatic transduction. The analysis also indicates that the maximum power is extracted from the system when approaching synchronization of mechanical and electrical events. However, there is a region of interest where the storage capacitor can be optimized to produce almost 70% of the ideal power, taken as the power harvested with synchronous converters when neglecting the power consumption associated with synchronizing control circuitry. Theoretical predictions are confirmed by measurements on an asynchronous energy harvesting circuit implemented with a macro-scale electrostatic converter prototype.Index Terms-Ambient vibrations, asynchronous switched mode operation, electrostatic energy conversion, energy harvesting circuits, modeling and optimization.

This work explores the potential of microwave heating for applications requiring parallel DNA amplification platforms. Device characterization and thermal modeling is performed on 4.1μl microfluidic chamber fabricated in polycarbonate. Microwave power at 6GHz is delivered to the chamber via copper transmission line in a microstrip configuration. Microwave power reflection coefficient and temperature measurements are performed to characterize the power coupled to the chamber and rate of change in temperature. Temperatures up to 72°C are achieved with less than 400mW power applied at the input of the transmission line. Initial heating and cooling rates measured experimentally are ∼7 and ∼6°C∕s, respectively. These results suggest that microwave heating is an efficient, rapid heating technique suitable for programmable, parallel DNA amplification platforms to be empolyed in future genetic analysis systems.

No abstract