Liquid gallium contact RF MEMS switch and VLSI MEMS switch

A self-healing RF MEMS switch, which utilizes liquid gallium contacts to take the place of the traditional solid contacts, is proposed. Electrostatic actuation is used to drive a silicon nitride diaphragm with upper electrodes. When the diaphragm is pulled down, small liquid gallium droplets work as an interface between the upper and lower contact electrodes. The loss of the gallium droplets can be avoided due to the unwettability of the material surrounding the contact electrodes. A hermetic package is designed to avoid gallium oxidation and reduce air damping of the diaphragm. The switch is designed to handle a 1 A hot switching current.; A stair-shaped electrostatic actuator that reduces driving voltage requirement is designed and simulated. A double exposure surface micromachining process is used. Compared to traditional cantilever electrostatic actuators, the pull-in voltage is reduced by 45.4%.; The design, fabrication and characterization of MEMS contact switches herein referred to as "VLSI MEMS switches" customized for VLSI are presented. The switches are fabricated using a surface micromachining process. Aluminum is employed as the contact metal. A stair-shaped electrostatic actuator is used to actuate the switch. The cantilever size is 23 mum by 3.2 mum. The pull-in voltage ranges from 10 to 18 V; and the nominal contact resistance ranges 2 to 20 O. A NOT gate is built using a VLSI MEMS switch with an integrated polysilicon resistor.; In a traditional asperity contact model it is assumed that the thickness of the planar contact electrode is infinitely large. This assumption is invalid for VLSI MEMS switches. A single asperity finite element model is constructed to simulate the contacts. Simulation results give that, for a 0.1-mum-radius asperity, Al-Al contact resistance increases from 0.5564 to 1.923 O as the thickness of the planar contact electrode reduces from 1 to 0.08 mum. A contact force and contact resistance measurement system is set up to verify the simulation results and to evaluate the influence of native oxide on contact resistance. The measured results agree with the simulation results.