Localized, In-Situ Vacuum Measurements For MEMS Packaging
- PDF / 329,603 Bytes
- 5 Pages / 612 x 792 pts (letter) Page_size
- 30 Downloads / 259 Views
A5.32.1
Localized, In-Situ Vacuum Measurements For MEMS Packaging Nicholas Moelders, James T. Daly, Anton C. Greenwald, Edward A. Johnson, Mark P. McNeal, Ramesh Patel, Martin U. Pralle, Irina Puscasu, Ion Optics, Inc., Waltham, MA.
Abstract MEMS devices have unique packaging considerations compared to conventional semiconductor devices. They tend to have relatively large die size and many architectures cannot tolerate elevated temperatures. Often these devices require a vacuum environment for efficient operation. While advances have been made in hermetic packaging of MEMS devices, vacuum packaging remains elusive. One significant problem in developing vacuum sealing has been the inability to determine, readily and non-destructively, the vacuum level inside the package. We have previously described the development of a silicon MEMS-based chip design, “SensorChipTM,” with integrated photonic crystal and reflective optics, which uses narrow-band optical emission and absorption for selective identification of gas and chemical species.1 Because the power consumption required to maintain a specific temperature is directly related to the vacuum level, these devices effectively serve as microscopic Pirani gauges – local vacuum sensors in the moderate vacuum range (0.01 to 1.0 torr) of interest to MEMS devices. Using the membrane itself as a vacuum gauge during sealing has proven to be an invaluable tool in developing a robust vacuum seal in a leadless chip carrier package. It has enabled us to optimize choice of design, materials and processing. Introduction Performance of some micro-electrical-mechanical systems (MEMS) containing mechanical oscillators, infrared bolometers, electron-emitters or heaters is enhanced by operation in a vacuum with pressure less than the molecular-viscous flow transition, preferably less than 10mtorr. In most cases, performance of the MEMS device itself will be an indicator of internal package pressure. However, connections to internal devices may be complex and operation of the device during the sealing process may not be possible. This paper presents the results of an experiment performed during development of a packaging process. We wanted to measure the pressure as a function of time during the sealing process, in-situ, so as to optimize the time-temperature cycle. We decided that a simple heater composed of a thin, suspended silicon membrane would serve well as a thermal-loss based pressure sensor. It is very rugged and can operate with only one connection (plus grounded package) during wide environmental changes (temperature, pressure). There are many examples of such a device.2, 3, 4 Experimental Procedure A schematic diagram of the MEMS device used for this work is shown in Figure 1. The outer dimension of the “frame” is 5 x 5mm square. There are two suspended filaments, each about 1.5 by 2.5mm, supported by two U-shaped arms. Starting material for fabrication of this device is a bonded SOI (silicon-on-insulator) wafer. The filaments and supporting arms are nominally 10-microns thick
Data Loading...
