No abstract

Research on contact characterization for microelectromechanical system (MEMS) switches has been driven by the necessity to reach a high-reliability level for micro-switch applications. One of the main failures observed during cycling of the devices is the increase of the electrical contact resistance. The key issue is the electromechanical behaviour of the materials used at the contact interface where the current flows through. Metal contact switches have a large and complex set of failure mechanisms according to the current level. This paper demonstrates the validity of a new methodology using a commercial nanoindenter coupled with electrical measurements on test vehicles specially designed to investigate the micro-scale contact physics. Dedicated validation tests and modelling are performed to assess the introduced methodology by analyzing the gold contact interface with 5 μm 2 square bumps at various current levels. Contact temperature rise is measured, which affects the mechanical properties of the contact materials and modifies the contact topology. In addition, the data provide a better understanding of micro-contact behaviour related to the impact of current at low-to medium-power levels.

Until now, the determination of microelectronic packages hermeticity has been related to the MIL-STD-883 method 1014 which is based on the helium leak detection method. But this method is no longer suited for small packages due to the resolution limit of the apparatus. Indeed, leaks induced by nonhermetic MEMS packages are often one order of magnitude smaller than the resolution of the helium leak tester. Consequently, characterization of MEMS packages requires new methodologies to measure hermeticity accurately. Two methods will be investigated in the context of this study: the membrane deflection measurement, when exposed to different pressures, using optical profilometry, and the measurement of the variation of gas concentration in a sealed silicon cavity by Fourier transform infrared spectroscopy (FTIR). The calculated leak rates are compared for samples where the standard fine leak test gave no results. The values obtained for the leak rates with the optical test and FTIR test for the same sample are identical, showing the relevance of these two methods. FTIR spectroscopy is a promising method which enhances standard detection limits. It can be used as a reliable process quality control tool.

No abstract

No abstract