Microelectromechanical Systems Switch Simulator

Unique microelectromechanical systems switch simulator uncovers material property mysteries and opportunities.

AFRL materials scientists developed a highly sophisticated laboratory instrument that simulates the effects of physical forces and electrical current on microelectromechanical systems (MEMS) switches. The simulator's performance has induced revolutionary insights into microscale switches—how they work and what causes them to fail.

MEMS switches offer substantial performance enhancement over current electromechanical (EM) and solid-state (SS) switches. They demonstrate significant potential for military and commercial use, especially for radio frequency (RF) applications. Realizing this potential, however, requires a greater understanding of both the contact physics involved with electrode materials and the material characterization properties at the nanoscale level.

The advantages of MEMS RF switches over conventional EM and SS switches include (1) higher linearity, (2) lower insertion loss, (3) lower power consumption, (4) reduced size, (5) higher shock resistance, (6) wider temperature range, (7) improved isolation, and (8) lower cost. The improved performance and reliability that MEMS RF switches (see figure) provide will benefit existing systems and create a new paradigm for future system development. Immediate applications include radar, aviation instrumentation, and cellular phones.

Scanning electron microscope view of a MEMS RF switch (photo courtesy of US Air Force)

Among the switch devices currently used in RF systems, EM relays offer the best high-frequency performance in terms of low insertion loss, high isolation, and good power handling (up to several watts). Unfortunately, EM devices are large, slow, and expensive, and they lack durability. Conversely, SS switches offer chip-level integration, small size, fast switching times, excellent durability, and low cost. However, they generally do not perform well in broadband applications; they have high insertion loss and poor isolation; and their high losses tend to nullify their size benefits due to the need for signal amplifiers, which increase power consumption and complexity. In choosing between EM and SS switching, designers are accustomed to accepting the necessary trade-off between the high-frequency performance offered by EM and the durability, size, low cost, and switching speed of SS. Therefore, the appeal of MEMS RF switches is that they offer the performance of EM switches with the durability, size, and low cost of SS devices.

Prior to AFRL's research and subsequent development of a switch simulator, scientists possessed a limited understanding of the factors determining the performance and reliability of these small devices. The team built the simulator to study contact resistance (Ω) and microscale surface forces in the gold contacts used in direct current MEMS switches. Applying fundamental properties to performance, they correlated the effects of contact force and electric current on Ω, microadhesion, reliability, and durability.

Experimentation identified microadhesion as the primary failure mechanism at low current (0-100 μA) and short circuits as the cause of failure at high current (1-10 mA). Results further revealed that electric current had a profound effect on deformation mechanisms, microadhesion, Ω, reliability, and durability. In addition, asperity creep, switch bouncing, and switching-induced adhesion occurred at low currents, whereas near-zero adhesion, asperity melting, and switch shorting via nanowire formation occurred at higher currents. The presence of a carbon/oxygen film, along with switch contact load versus resistivity, suggested that tunneling (electron passage through thin, insulating contact films) was a significant charge transfer mechanism at both low and high currents. The experiments also demonstrated that repeated switch activation atomically hammered the asperities flat, which resulted in greater adhesion, and that surface roughening via nanowire formation prevented adhesion at high current levels.

The principal conclusions of the research are as follows: microadhesion is the dominant failure mechanism at low current and becomes apparent as contact surfaces are flattened during actuation, whereas short circuits are the dominant failure mechanism at high current and occur as nonowires are formed from melted asperities. The research effort further aids MEMS designers by establishing quantitative relationships among switch actuation force, ambient environment, current density, and performance. In the past, switch designers knew switches failed; now, they have significant insight as to why failures occur. Armed with this data, designers can build more durable, robust switches. The research is similarly applicable to failures inherent to the simple pressure contacts used in connectors, sockets, and chip holders—a growing concern as the number of inputs and outputs increase and contact sizes decrease.

MEMS-based switches offer outstanding potential for the integration of sensors, actuators, signal processing, and communications. Nevertheless, the small size of MEMS devices results in large surface-area-to-volume ratios, which results in the surface contact forces dominating the performance. MEMS switches and relays thus represent a promising new technology that achieves control over these surface forces.

Dr. Jeffrey S. Zabinski, Dr. Steven T. Patton (University of Dayton Research Institute), and Dr. Peter S. Meltzer (Anteon Corporation), of the Air Force Research Laboratory's Materials and Manufacturing Directorate, wrote this article. For more information, contact TECH CONNECT at (800) 203-6451 or place a request at http://www.afrl.af.mil/techconn_index.asp . Reference document ML-H-05-01.