Fernando Bitsie

Interferometry of actuated microcantilevers to determine material properties and test structure nonidealities in MEMS

By integrating interferometric deflection data from electrostatically actuated microcantilevers with a numerical finite difference model, we have developed a step-by-step procedure to determine values of Youngs modulus while simultaneously quantifying nonidealities. The central concept in the methodology is that nonidealities affect the long-range deflections of the beams, which can be determined to near nanometer accuracy. Beam take-off angle, curvature and support post compliance are systematically determined. Youngs modulus is then the only unknown parameter, and is directly found. We find an average value of Youngs modulus for polycrystalline silicon of 164.3 GPa and a standard deviation of 3.2 GPa (/spl plusmn/2%), reflecting data from three different support post designs. Systematic errors were assessed and may alter the average value by /spl plusmn/5%. An independent estimate from grain orientation measurements yielded 163.4-164.4 GPa (the Voigt and Reuss bounds), in agreement with the step-by-step procedure. Other features of the test procedure include that it is rapid, nondestructive, verifiable and requires only a small area on the test chip.

Interferometric measurement for improved understanding of boundary effects in micromachined beams

Micromachined beams are commonly used to measure material properties in MEMS. Such measurements are complicated by the fact that boundary effects at the ends of the beams have a significant effect on the properties being measured. In an effort to improve the accuracy and resolution of such measurements, we are conducting a study of support post compliances in cantilever and fixed-fixed beams. Three different support post designs have been analyzed by finite element modeling. The results are then compared to measurements made on actual devices using interferometry. Using this technique, the accuracy of measurements of Youngs modulus has been improved. Continuing work will also improve the measurement of residual stress.

Small-area in-situ MEMS test structure to measure fracture strength by electrostatic probing

We have designed, fabricated, tested and modeled a first generation small area test structure for MEMS fracture studies by electrostatic rather than mechanical probing. Because of its small area, this device has potential applications as a lot monitor of strength of fatigue of the MEMS structural material. By matching deflection versus applied voltage data to a 3D model of the test structure, we develop high confidence that the local stresses achieved in the gage section are greater than 1 GPa. Brittle failure of the polycrystalline silicon was observed.

Optical performance of pivoting micromirrors

In this paper we describe optical and dynamic performance of tip/tilt micromachined mirrors fabricated using the SUMMIT V surface micromachining process. We find that the tilt angle for a given mirror design is determined by a combination of geometric factors and stiffness of the capacitive suspension. Switching speeds of ~40-50 microsecond(s) econds are measured for 50 micrometers -square mirrors. Finally surface roughness and curvature before and after metallization are obtained using white light interferometry.

Quantifying the Effect of Dynamic Boundary Condition Differences Between the Vibration Laboratory and the Field Operating Environment

Vibration qualification testing of components or subassemblies is performed on shaker tables. The uncertainty associated with the boundary condition difference between the field and the shaker table is usually ignored. It will be shown with analytical and hardware demonstrations that component response can be significantly different even if controlled at one accelerometer. The quantification of this uncertainty would be useful for characterizing margin associated with the laboratory boundary condition. Assuming that a rigid body representation of the field base input is known, a method is proposed that uses a modified version of structural modification, to attempt to correct for the difference using a modal characterization of the shaker table. Disappointing results are shown for a first time implementation with actual hardware. However, the experience provides important insights to guide future efforts for the quantification of the response uncertainty due to the boundary condition difference between the field and laboratory.

Microdiagnostic Lab on a Chip - LDRD Final Report

Polycrystalline silicon (polysilicon) surface micromachining is a new technology for building micrometer ({micro}m) scale mechanical devices on silicon wafers using techniques and process tools borrowed from the manufacture of integrated circuits. Sandia National Laboratories has invested a significant effort in demonstrating the viability of polysilicon surface micromachining and has developed the Sandia Ultraplanar Micromachining Technology (SUMMiT V{trademark} ) process, which consists of five structural levels of polysilicon. A major advantage of polysilicon surface micromachining over other micromachining methods is that thousands to millions of thin film mechanical devices can be built on multiple wafers in a single fabrication lot and will operate without post-processing assembly. However, if thin film mechanical or surface properties do not lie within certain tightly set bounds, micromachined devices will fail and yield will be low. This results in high fabrication costs to attain a certain number of working devices. An important factor in determining the yield of devices in this parallel-processing method is the uniformity of these properties across a wafer and from wafer to wafer. No metrology tool exists that can routinely and accurately quantify such properties. Such a tool would enable micromachining process engineers to understand trends and thereby improve yield of micromachined devices. In this LDRD project, we demonstrated the feasibility of and made significant progress towards automatically mapping mechanical and surface properties of thin films across a wafer. The MEMS parametrics measurement team has implemented a subset of this platform, and approximately 30 wafer lots have been characterized. While more remains to be done to achieve routine characterization of all these properties, we have demonstrated the essential technologies. These include: (1) well-understood test structures fabricated side-by-side with MEMS devices, (2) well-developed analysis methods, (3) new metrologies (i.e., long working distance interferometry) and (4) a hardware/software platform that integrates (1), (2) and (3). In this report, we summarize the major focus areas of our LDRD project. We describe the contents of several articles that provide the details of our approach. We also describe hardware and software innovations we made to realize a fully automatic wafer prober system for MEMS mechanical and surface property characterization across wafers and from wafer-lot to wafer-lot.