MEMS Failure Analysis
Far outside the realm of typical integrated circuits, there exists a myriad of devices with astonishing properties that are nonetheless fabricated in silicon and other semiconducting materials. Microscopic membranes deflect and deform under fluctuations in atmospheric pressure, creating an electrical signal that can be digitized as sound; an extensive array of silicon springs and plates shift under the influence of external forces, serving as an accelerometer; an infinitesimally small array of metallic mirrors actuated by electrostatic forces can serve as one of the primary components in a theater projector or television. Though these devices have widely divergent purposes, they all have one thing in common: all are Micro-Electro-Mechanical systems (MEMS), and all pose a significant challenge for a failure analyst.
In addition to the potential issues faced by any integrated circuit - electrical and mechanical stresses, contaminants, and so on - failure analysis of a MEMS is complicated by the fact that these devices are comprised of microscopic moving elements, adding an entirely new dimension to the analysis. For example, a device may fail as a result of “stiction” - an element failing to move properly, perhaps as a result of an incomplete release etch, a foreign particle, electrostatic attraction, or other mechanisms - or from mechanical fatigue, the natural wear and tear of constant motion resulting in a drift in performance over a product’s lifetime. Further complicating matters, many MEMS devices are securely sealed, as a way to protect their fragile elements; to successfully disassemble such a product, an analyst must be familiar with a wide variety of chemical and mechanical preparation methods, always keeping in mind the delicate nature of the sample. Many of an analyst’s tools can be adapted for MEMS work: infrared microscopy can often be used to inspect the MEMS elements for damage or anomalous material without breaking the device seal, as device lids are often made of silicon which is somewhat transparent to short wave infrared light; the focused ion beam can be used to cut windows in the lid of a MEMS to perform a high magnification inspection; and so on.
IAL has experience with a wide range of MEMS devices, including extensive work with accelerometers, microphones, and pressure sensors, among others. As long as we can develop a method for replicating a failure on our test bench, chances are good that we can delve into the microscopic beams and gears of your MEMS and come out with the root cause of failure.
- Analyzing “dead pixels” on a mirror-based imager array
- Troubleshooting bad outputs from a silicon accelerometer