What’s Up MEMS?

Strong segment growth. A whole slew of new devices on the horizon. A healthy pipeline of enabling critical problems to be solved. Has somebody been peeking at my Christmas list? Possibly yes, and thankfully so, because 2016 has been that kind of year in MEMS.

Taking a look at the numbers from Yole Developpement, they expect 2016 MEMS device segment revenue to be around $13B, with an estimated CAGR of approximately 13 percent through 2021. Yole expects unit volume for 2016 to come in at just under $17B, with an estimated CAGR of approximately 21 percent through 2021. Wow! The majority of device revenue has been generated by growth in the consumer and automotive markets, each representing more than 50 percent and 25 percent of total device revenue, respectively.

Source: Yole Developpement

In 2016, the industry saw an increase in fabless players in this space, which led to an increase in foundry activity. While many continue to prod at the 300mm glass ceiling (300mm MEMS was demonstrated in Europe and readied for production in Taiwan), 100 percent of production is still on 200mm wafers and is expected to remain so for the foreseeable future.

While the mainstays of the MEMS segment (accelerometers, gyroscopes, microphones, RF filters and pressure sensors) continue to be strong performers, 2016 saw a big push toward a bunch of new and potentially high-volume MEMS devices. We’re talking about gas sensors, fingerprint chips (FPC), next-generation microphones/RF filters, and even MEMS-based speakers. Piezo-based MEMS are also making a comeback as inkjet technologies increasingly look to embrace higher viscosity inks and applications like portable ultrasound, gesture recognition and high force actuators come to bare.

What’s behind the need for these new devices?

Many factors are driving the need for new MEMS device technologies. Noise cancellation and speaker recognition in automotive is creating a pull for new microphone architectures capable of tens of dB higher signal-to-noise ratios (SNR) than conventional devices. The general call for more ruggedized FPCs that can withstand moisture and dirt, and can offer security above the conventionally used capacitive-style devices is another influence. Greater electro-mechanical coupling (Tx/Rx efficiency) of next-generation RF filters used in 4G and 5G networks are required to enable steeper pass-band transitions so carriers can pack more frequencies into a given amount of bandwidth. These are but a few of the goings on behind the onset of a whole new generation of high-volume MEMS.

And while opportunity abounds, the challenges are plenty. Each new device technology is characterized by one or two key on-wafer film properties that unlock that magical device-level capability. For RF filters it’s thickness non-uniformity of less than 0.5 percent coupled with a stress range target of ±75Mpa, for a 1µm PVD film of Scandium doped (4-6 percent) aluminum nitride (ScAlN). For microphones, it’s once again ScAlN, however in concentrations up to 43 percent Sc. Based on the new cantilever type architecture, each 5 percent of in-film Sc provides an additional 1 dB of SNR! On the high force piezo side, we’re developing solutions for PVD lead zirconium titanate (PZT), and in conjunction with the CTO team here at Applied, we’re also looking into lead manganese niobate – lead titanate (PMN-PT). In both materials we’re focused on high-temperature (750°C – 800°C) PVD processes. The key on-wafer parameters sought after are d₃₃ and d₃₁ piezo-coefficients in excess of ≥100 pm/V and ≥150 pm/V, respectively.

My new motto is “Engage Early, Engage Often!” Many of these new materials require device-level characterization, so it’s imperative we engage with our customers early to characterize the on-wafer performance of these new materials. Combining our deposition and process optimization knowhow with our customers’ device-level test vehicles is the surest way to develop the solution they need when they need it. It’s what works and it’s something we strive for each time we look to deliver a new material or process solution into this space.