What’s in a research blog header?

First, I want to thank you for taking the time to read this blog. Brian McCarthy and I will be taking on the roles of editors and keeping you apprised of the latest and greatest research at Intel. We work in a part of Intel’s central labs called Strategy and Planning which exists to facilitate interactions amongst all of the central labs, between these labs and other Intel groups, and between the labs and the outside world. In the case of this blog, we are working to facilitate a dialog between some of our key researchers and yourselves.

But, today I want to talk to you about a picture. Specifically, the header of the Research@Intel blog page.


I think the pic looks pretty cool. It was taken about a month ago by a friend of mine named Jeffrey Tseng, who has been a mechanical engineer in our photonics labs for years. He’s also a great photographer, and can make pretty much anything or anyone look really cool. But, though he took a couple hundred pictures in the labs that day, after all the sorting and vetting and voting we ended up with this one. And it’s not just a random object, it’s part of history still in the making. The chip in the middle is one of the latest examples of our Silicon Photonics research, which Ansheng Liu introduced in his first blog. Specifically, it’s a Raman Ring Laser. I’ll get back to what that is, but first I want to tell you where it came from.

Intel’s Silicon Photonics program has its roots in a debug tool developed by (now Intel Fellow) Mario Paniccia called the Laser Voltage Probe, a now standard tool which allows debug engineers to measure transistor signals on flip-chip (i.e. face-down) microprocessors by probing through the back side of the silicon with an infrared laser beam. When I first started working for Mario in 1997, I was an intern qualifying an anti-reflective coating tool which made the laser probing of the chips more effective.

About a year or so later, after I’d joined the team full time, Mario pitched an idea to senior management for a silicon-based telecom optical switch. Although he was able to convince Intel to fund this new project, it happened during the telecom buildup in the early 2000’s, and the bulk of the existing team was vectored to develop more traditional optical products to fuel the telecom explosion.

To make a long (and a bit painful) story short, Mario took his idea, his admin and myself to a near-empty building, where he built a new team. Mario ran the group like a start-up company within the big corporate machine. More researchers were hired — Ansheng being the first — and a few of his previous team even “crossed over” to rejoin him in the effort. New labs were refitted with optical equipment and before long test wafers were coming out of the fab.

It became apparent that the best use of this switching technology would be to create a fast modulator (an optical data encoder… like a really fast shutter), which did not exist yet in silicon above about 20 MHz. The initial goal was ~1 GHz. Amid much skepticism, the team not only met the goal, we exceed it. In fact the latest modulators which Ansheng has designed can send data at a rate of over 30 Gbps. During this timeframe, we were picked up as a seed project and finally integrated into Intel’s central labs.

Meanwhile, Mario had funded a small side-project in the group looking into the Raman Effect in silicon. The idea was that by using this effect, one might be able to amplify light or even create a laser, which had not been accomplished in silicon. This was during tough times and budget cuts, but Mario kept funding Raman project due to its potential for a breakthrough. Soon, there was a world wide race among researchers in this area to demonstrate amplification and lasing. In the end, we demonstrated the first world’s first (optically pumped) silicon laser that could emit a sustained beam of light.

So, the chip you see here is a next step in the Raman laser evolution – a Raman ring laser, in which light resonates in a racetrack shaped ring on the chip, building strength with each pass. Ring lasers such as this are essential to extending the laser’s wavelength further into regions of special interest for science and medicine, by passing them from one ring to another on the same chip.

After working with lasers for many years, I’ve moved on to help with the Intel’s Tera-scale Computing effort, of which silicon photonics is a part. But besides simply bringing back a wealth of memories, I think the photonics chip in the blog header exemplifies the spirit of the research labs here — perseverance in the face of many obstacles to do something cool which has never been done before.

Hope you enjoy the blog.


One Response to What’s in a research blog header?