The Effects of Silicone Oil Contamination on the Contact Resistance of MEMS Electrical Contact Surfaces

Abstract

Surface contamination has long been known to affect the performance of devices that utilize contacting electrodes. Electrical contact degradation is insensitive to the specific nature of the surface contamination, in that formation of any dielectric material at contact points will result in increased contact resistance. This phenomenon is particularly detrimental in microelectromechanical system (MEMS) electrical contacts, where contact forces are limited and may be insufficient to disrupt surface films. Increases in electrical contact resistance with cyclic operation is a major source of reliability problems associated with MEMS electrical contacts. Silicone oil can act as a highly effective lubricant for sliding MEMS surfaces, increasing operational lifetime for devices with interacting surfaces. However, silicone is also a known source of electrical contact surface contamination, readily decomposing into insulating species when sufficiently energized [1–3]. Even though silicone oil immersed electrical contacts have been successfully used in large contact force electrical contacts, the performance and reliability implications of using silicone-immersed low-force MEMS electrical contacts are not well characterized. The subject of this study was to determine if hot-switched metal contacts immersed in silicone oil will degrade similarly to contacts know to degrade in a non-immersed environment. Electrical contact resistance degradation originating from arcing or metal-bridge-evaporation induced decomposition of surface contamination has been observed previously [4]. Silicone oil immersed low-force electrical contacts were made using a modified nano-indentation apparatus. A schematic of the contact zone is shown in Fig. 1. The apparatus was able to measure electrical contact resistance and adhesion of Au-coated spheres contacting silicone oil-contaminated Au-metallized silicon wafers. The contact forces selected were similar to normal loads achievable in MEMS devices. Figure 2. shows the electrical contact resistance degradation of a silicone oil immersed gold-gold contact vs. the same uncontaminated contact obtained from the experimental apparatus. The data points are the averaged resistance values during the period of maximum applied load, 100 μN in this case. The calculated Hertzian contact area (neglecting roughness effects) was 2.1 μm. The open-circuit voltage was set at 3.3 V and the in-contact current was limited to 3 mA. An individual contact cycle data point taken from Fig. 2, displaying the contact force and resistance versus time, is shown in Fig. 3. The resistance averaged over the peak load remains ∼1.1 Ω, even though during periods of low contact force the contact resistance is several orders of magnitude higher than at peak load. The asymmetry of the contact resistance in Fig. 3 suggests that an interfacial contaminant layer was ruptured during loading, creating adherent metallic contacts and allowing for lower resistance at smaller contact loads. This load-supporting, dielectric layer continues to evolve until, by cycle 20, the conductivity of the contact surfaces has been completely inhibited. Surface analysis of the contaminated surfaces was performed in order to ascertain the composition of the electrical contact interface. Relationships between surface contamination, mechanical stress and electrical contact resistance degradation will be discussed relating to the use of silicone oil in MEMS electrical contacts.