Today the University of Washington launched an exciting program that we hope will help trigger a new technology revolution around integrated silicon photonics chips. This program, which we call OpSIS (Optoelectronics Systems Integration in Silicon), will provide a silicon photonics wafer service to give researchers the ability to design and build experimental optical chips.
My team at Intel Labs has been researching silicon photonics devices for nearly a decade with a vision of creating silicon chips that can send trillions of bits per second across an optical fiber. After years of experimentation, I have come to believe that photonic integrated circuits will truly be a transformative technology, with a potential similar to the development of electronic integrated circuits and a diverse set of applications. Silicon ICs have, in only a few decades, been developed to address countless applications – they are used for computers, music amplifiers, sensors, radios, displays and many other devices. Likewise, silicon photonics will have impact to many other areas well beyond high-speed communications. These include sensing, wireless, and bio-medical devices.
With OpSIS, we are helping to develop the silicon photonics industry by reproducing a key factor in the success of the silicon IC. OpSIS is modeled in part after MOSIS (Metal Oxide Semiconductor Implementation Service), an IC wafer service for R&D started in 1981. MOSIS was (and still is) operated by the University of Southern California and assembles designs from multiple researchers on multi-project silicon wafers to share the cost of fabrication. OpSIS will do the same for silicon photonics chips. In fact, Carver Mead, the co-founder of MOSIS and coiner of the term “Moore’s Law” (among many other things) is speaking today along with Intel CTO Justin Rattnerin support of UW’s new optical wafer service.
OpSIS is the brainchild of UW professor Michael Hochberg, who will lead the program. Intel Labs has provided input and some funding to help launch this new service. The photonics technology lab here at Intel and I will also continue to help consult and provide assistance where possible to help OpSIS get operational. OpSIS has other significant partners. BAE Systems will provide fabrication facilities. The U.S. Air force is supporting Prof. Hochberg with funding to help OpSIS get started as well as for some special military interests in using silicon photonics for new radar technology. We have half a dozen researchers already signed up to hop on the first wafer fabrication service run.
As a final note – OpSIS will be important not just for innovation, but for education. In order to develop a new industry, we need to develop a new generation of engineers. OpSIS will provide university students with the ability to build their own silicon photonics experiments – a necessary capability that doesn’t exist today. These students of today will become the innovators and leaders of tomorrow that grow this nascent industry into one that touches people’s lives in ways we can’t even predict today. This is an exciting time for silicon photonics and the potential impact it will have to everyone’s lives.