Monitoring Plasma Over-etching of Wafer-bonded Microstructures

Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work must be obtained from the Foundation for Sensor and SUMMARY We have demonstrated that wafer-level probing of electrostatic pull-in test structures provides a means to monitor MEMS process uniformity and integrity. Test structures, which have proved to be useful in determining MEMS material properties, are used here to identify and quantify a non-uniform plasma over-etch used in the final release of silicon wafer-bonded microstructures. MEMCAD simulations confirm that the changes in test structure geometry do indeed account for the behavior seen experimentally. INTRODUCTION Measurement of electrostatic pull-in voltage has been proposed as a method for material property measurement and process monitoring [1,2]. In our group's previous work [3,4], we presented numerical and analytical models for the electrostatic pull-in of cantilevers, fixed-fixed beams, and circular diaphragms, and demonstrated the use of some of these structures for monitoring process uniformity and for extraction of material properties. The approach used an analytical expression which relates the pull-in voltage to structural dimensions and material properties through two intermediate parameters, the stress parameter S n and bending parameter B n , where n indexes the structure type. These quantities are found by non-linear least-squares fitting of pull-in voltage as a function of structure length or radius. They depend linearly on residual stress (for S n) or YoungÕs Modulus (for B n), and on powers of critical dimensions varying with a given test structure. In this paper, we show how pull-in measurements on fixed-fixed beams of two different widths signaled a fabrication problem during the plasma-etch release step.