Handle substrate Energy converter IC (sensor signal conditioning + RFID/ZigBee tag) Antenna Sensor Distributed sensor network A bulk silicon-based vibration-to-electric energy converter using an In-Plane Overlap Plate (IPOP) Mechanism Ayyaz M. Paracha 1 , Philippe Basset 1* , Peter C. L. Lim 2 , Frédéric Marty 1 , Tarik Bourouina 1 1 ESIEE-ESYCOM, Cité Descartes - BP 99 - 2 Bd, Blaise Pascal, 93162, Noisy-le-grand, Cedex, France 2 NANYANG POLYTECHNIC, 180 Ang Mo Kio Avenue 8, Singapore Abstract This paper focuses on the design, fabrication issues and characterization of a bulk silicon-based, vibration powered, electric energy generator. Its design is based on an In-Plane Overlap Plate (IPOP) configuration which is based on a parallel plate architecture. It uses electrostatic damping of a MEMS-based variable capacitor as the energy converting mechanism. The device is composed of a movable silicon proof mass with patterned electrodes on the underside, anodically bonded to a glass wafer with electrodes on top. High capacitance values can be achieved thanks to a low cost silicon-glass technology. By virtue of an in-plane movement of the mobile proof mass, a capacitance variation is achieved. Two architectures have been recently proposed [1]. This paper shows the results of full working devices. Electrical and mechanical characterization have shown that with a theoretically lossless electronic and a starting voltage of 5 V, power density of 59 µW/cm 3 is achievable at the mechanical resonance frequency of 290 Hz. Keywords: Energy harvesting, Vibration-to-electricity converters, Electrostatic transduction 1 - INTRODUCTION During the past, decade efforts have been made to reduce the power consumption of sensors. At the same time, various research works have been reported to make them autonomous i.e. independent of externally attached power sources [2]. Various techniques of harvesting energy from the environment to power these sensors have been studied. This concept can be helpful in many ways: it reduces the bunch of wires used for the electrical connections in between the power source and different sensors, device’s lifetime increases, cost maintenance decreases as there is no need to replace the external batteries. In addition, environmental polluting chemical materials can be eliminated. In order to provide autonomous nodes in a sensor network (cf. Fig. 1) and to power smart-dusts [3], ambient vibration is a power source that is largely considered, as there exist broadband vibration spectrum in applications like automotives [4], air planes, etc… The three main existing mechanisms to harvest energy from environmental ambient vibrations are electrostatic [1, 2, 4], piezoelectric [5] and electromagnetic [6]. Among all these, electrostatic-based energy harvesters are good candidates as they give good compatibility with CMOS process and are considered the best for miniaturization. Electrostatic converters need two sets of electrodes. First one is attached to a moving mass called as proof mass and second one is fixed to the substrate. A huge proof mass is normally used to target __________________________________ *Contact author: Tel. (+33) 1 45 92 65 86, email: p.basset@esiee.fr a low resonance frequency. External vibrations force the mobile part to move relatively to the substrate, leading to mechanical-to-electrical energy transduction if a constant charge is maintained on the electrodes while the capacitance decreases. 2 - CONVERTER DESIGN There exist four major topologies of electrostatic based energy harvesters: In-Plane Gap-closing Combs (IPGC) [2, 4], In-Plane Overlap Combs (IPOC) [7], Out-of-Plane Gap-closing Plate (OPGP) [8] and In-Plane Overlap Plates (IPOP) [1, 9]. Recently we have proposed two architectures Figure 1 – Project Overview: the vibration-to-electricity converter is the power source for nodes in a distributed network. Each node includes a sensor, a chip-size antenna [3] and an IC. - 169 - The Sixth International Workshop on Micro and Nanotechnology for Power Generation and Energy Conversion Applications, Nov. 29 - Dec. 1, 2006, Berkeley, U.S.A.