What little (devices) I remember: the genesis of MEMS design and development in Alberta, Canada

• Tags:
• CMUT
• MEMS
• MEMS Foundry
• microfluidic
• optical switching
• Silicon micromachining

As we celebrate the 20th anniversary of Micralyne as a private company (Micralyne was privatized on April 3rd 1998), we thought that it was the perfect opportunity to begin blogging and start with a review of the origins of MEMS development and manufacturing at Micralyne.

Micralyne originated in 1982 with the Government of Canada (Industry Canada, now the Ministry of Innovation, Science and Economic Development Canada https://www.ic.gc.ca/eic/site/icgc.nsf/eng/home ) forming a Centre of Excellence (COE) for microelectronics, sensors and actuators called the Alberta Microelectronic Centre (AMC) at the University of Alberta (Edmonton) and the University of Calgary. The intent was to mobilize a cluster of research scientists at the two universities and a handful of small start-up companies for micro and nanotechnology advancement. AMC offered circuit board design and fabrication services initially, but quickly built a repertoire of CMOS ASIC design and fabrication and higher functionality electronic design. At this time, the beginning of what would become Micralyne was formed as the Sensors and Actuators Laboratory. The lab was a modest start, with myself being the sole employee, thus the defacto manager with a large group of dedicated graduate students anxious to fabricate devices. The lab was a large collection of used semiconductor equipment in 1987, but we rapidly established a number of key MEMS processes and worked forward.

By 1991, the CMOS fab was combined with the lab to form a large comprehensive microfabrication and assembly facility – a MEMS foundry. We were fabricating a variety of devices for academic research such as split drain MOSFETs for magnetic flux measurement, early Ion Sensitive FETs (IsFETs), glass microfluidic devices, high mobility semiconductors, and single mode fibre waveguides. In order to make detectors for sub mm wave astronomy, we fabricated quantum tunnelling Superconductor-Insulator-Superconductor THz receivers. We developed high g force (100,000 gs) accelerometers. Our thin film focus led to development of a simulation package to model the growth of thin films over topography and eventually marketed it as SIMBAD (Simulation by Ballistic Aggregation Deposition). This eventually became an entire business division, and refinement of the models led to the development of structured GLAD (GLancing Angle Deposition) films. We started opening the doors to research groups in a collaborative manner, and were highly involved in the development of antimicrobial coatings for bandages and medical devices, which were commercialized in Alberta. Further commercial interests in microfabrication services led to us developing imaging systems for the first Computer to Plate (CtP) printing systems that proved to revolutionize the digital printing industry. This commercial interest gathered momentum, and the lab needed to separate academic research from proprietary customer development.

In 1998, the Alberta Microelectronic Corporation was born. After relocation from University of Alberta campus to our present 50,000 square feet location, we were then renamed to Micralyne Inc. The University of Alberta nanoFAB (http://ww.nanofab.ualberta.ca) was created circa this period to continue to provide high level of support for academic research and MEMS foundry services to early-stage companies in the region. This high level of interest in micro and nanofabrication in the area led to the National Research Council (http://www.nrc-cnrc.gc.ca) further investing in the local MEMS infrastructure through the establishment of National Institute for Nanotechnology (NINT http://www.nint-innt.ca/) in 2001.

 

Looking back now at the last 20 years as a private company, the steady progress path that the timeline chart implies is a bit of data smoothing, but for the most part, the greatest fits and starts came from rapid growth due to innovations which sped the path to market for a variety of device types.

In 2000, apart from the continued strength in the CtP optical switching platform, it would have been within reason to call ourselves a Life Sciences company, with a large offering of complicated microfluidic devices with embedded electrodes and engineered dielectrics to provide novel biological compatibility features (microfluidics whitepaper). We were able to use standard processes developed for glass micromachining for dielectrophoresis purposes and integrate quickly the same approach to make DWDM (Dense Wavelength Division Multiplexing) ROADM (Reconfigurable Optical Add Drop Multiplexing) WSS (Wavelength Selective Switches). We were the dominant supplier for these devices, which enabled Agile Optical Networks. This platform was also offered to university researchers as the MicraGEM process platform through the CMC (Canadian Microelectronic Corporation), the beginning of many standard processes and platforms designed to enable rapid design and implementation.

At the same time, in other optical switching applications, we developed electrically addressable high speed (10 MHz) spatial light modulators (whitepaper), which when used as a grating array, could switch hundreds of channels at high speed for imaging applications. Continued development in the WSS platform led to devices being made from multiple layers of silicon, which allowed comb drive high angle tilt mirrors. This process was also offered as a standard offering for designers through CMC as MicraGEM-Si. The ability to make the high-speed light modulators was coupled with silicon processing, and we began fabrication of multi layer SOI (Silicon on Insulator) structures with cavity SOI bonded to structured SOI. The application was for CMUTs (Capacitive Micromachined Ultrasonic Transducers), and the ability to bond under vacuum provided high speed (40 MHz) and high sensitivity devices to be used as ultrasound emitters and detectors.

Extending silicon micromachining technology, we designed a high sensitivity accelerometer structure using a large proof mass with silicon combs that were vacuum-sealed in package. We were in fairly high production volumes on this design and began development of the next generation, which is made using the MicraSilQ platform. This allowed wafer level processing, which meant that the vacuum packaged accelerometer would be in chip format, and the ball grid array of solder pads would address the device through doped polysilicon TSVs (Through Silicon Vias). The full suite offered by this platform can be modified to create a large range of potential device types. Presently development is underway on our metal oxide gas sensor technology, which takes advantage of many of the already developed film and standardized processes available at Micralyne. Coupled with our ISO13485 medical certification, we have made a number of devices for our customers that have successfully received FDA & CE clearance for implanting in humans.

This is a short overview of 30 years of MEMS development and commercialization at Micralyne. MEMS devices today are everywhere and Micralyne is one of the few foundries that can trace its history to the beginnings of product-ready MEMS. If you have ever read a glossy magazine or watched a video on YouTube in the first decade of this millennium, you have likely been utilizing Micralyne enabled optical devices, as did the LCROSS mission which found water on the moon. Today, some of the most innovative companies in the world are choosing to work with Micralyne in developing unique new-to-the-world MEMS and sensors that are revolutionizing the way we interact with electronics.