Yang J.A., Shuvra P.D., Lin J-T., Walsh K., McNamara S., Alphenaar B.
University of Texas at Austin, US
Keywords: memory, MEMS, modeling, optimization
Microelectronics and microelectromechanical systems (MEMS) have given rise to a large class of new de- vices, from ultra-miniature sensors for cell phones and automobiles to complex memory devices for computers. Recent research in memory devices has largely focused on designing new non-volatile forms of electronic mem- ory, yet a MEMS-based memory device has not been proposed in the literature. This work presents a novel MEMS memory device using an asymmetrical, bistable buckled beam. Relying on well-known MEMS structures and the piezoresistive effect, the MEMS memory model was first conceptually designed, then optimized for sig- nal strength using CoventorWare finite-element model- ing. In the optimization process, various aspects of the devices geometry were ranked in order of magnitude of influence on the devices output signal strength, then op- timized based on their importance. Simulation results indicate that the optimized device can generate signals up to 5.5uV with a supply voltage of 2.5V, a 27.5x im- provement over the initial design. The length and width of the beam were found to be the most influential fac- tors in controlling the signal output: increases in beam width lead to significant increases in signal when paired with the corresponding beam lengths. However, large beam widths caused the beam to buckle into higher- order modes when the beam was short, leading to sharp decreases in signal. Other geometric factors had only minor impacts on signal strength. The MEMS memory device paves the way for future low-power, radiation- hard MEMS memory models.
Journal: TechConnect Briefs
Volume: TechConnect Briefs 2019
Published: June 17, 2019
Pages: 330 - 333
Industry sector: Sensors, MEMS, Electronics
Topics: Nanoelectronics, Sensors - Chemical, Physical & Bio
ISBN: 978-0-9988782-8-7