Designing Thermally Uniform Mems Hot Micro-Bolometers

  • PDF / 466,359 Bytes
  • 7 Pages / 612 x 792 pts (letter) Page_size
  • 45 Downloads / 301 Views

DOWNLOAD

REPORT


DESIGNING THERMALLY UNIFORM MEMS HOT MICRO-BOLOMETERS Nicholas Moelders, Martin U. Pralle, Mark P. McNeal, Irina Puscasu, Lisa Last, William Ho, Anton C. Greenwald, James T. Daly, Edward A. Johnson, Ion Optics, Inc., Waltham, MA 02452; Thomas George, Jet Propulsion Laboratory, Pasadena, CA 91109. ABSTRACT Here we describe the evolution of a silicon, MEMS-based chip design developed for infrared gas and chemical detection. The “Sensor-Chip,” with integrated photonic crystal and reflective optics, employs narrow-band optical emission/absorption for selective identification of gas and chemical species. Gas concentration is derived from attenuated optical power, which results in a change in device set point. This change in temperature results in a change in device resistance, via the TCR of the Si. Thermal non-uniformity across the device results in optical “noise” and accelerates localized thermal and electrical failures. This paper reports the influence of processing and design, on achieving uniformly heated, high reliability devices. Specifically, we examine the role of contacts, drive scheme, and device thermal distribution on chip design. Experimentally the temperature uniformity was characterized using an infrared camera. Experimental results indicate that the design of the contact areas in combination with the device design is essential for the reliable performance of the Sensor-Chip. Redesigned devices were fabricated and demonstrated as highly-selective gas and chemical sensors. INTRODUCTION Sensing carbon dioxide (CO2) gas is of profound importance because of its prevalence as both a physiological and industrial byproduct. A review of CO2 gas sensor applications includes respiration monitoring, combustion by-product monitoring, indoor air quality, and environmental monitoring needs. Respiration monitoring (capnography) is used to monitor patient CO2 exhalation levels under general anesthesia and/or during ventilation therapy. There are several traditional methods for gas and chemical detection. Those based on chemical reactions are generally classified as electrochemical or catalytic. Catalytic sensors measure a change in resistance due to an oxidation change on the surface of the sensor. Non-dispersive infrared (NDIR) spectroscopic gas sensors are based on the fact that many gases have unique infrared absorption signatures in the 2-14 µm region. The uniqueness of each gas absorption spectra enables conclusive identification and quantification of chemicals in liquid and gas phase mixtures. This work seeks to exploit MEMS-based technologies to integrate the functionality of NDIR laboratory-grade instruments, costing tens of thousands of dollars, onto a chip. This is accomplished by using a thermally isolated, narrow band emitter in combination with reflective optics. The emitter projects a beam of light through an optical cell to the reflector, which reflects the light back to the emitter (see Figure 1a). In the absence of any gas absorption in the optical cell, the filaments (and optics) rapidly reach thermal