Laser on Silicon

A new glass glue opens the path to optical computing

2 min read

Samuel K. Moore is IEEE Spectrum’s semiconductor editor.

Scientists at Intel and at the Univer­sity of California, Santa Barbara, have managed to combine an indium-­phosphide light emitter and a silicon chip to produce a hybrid laser that, years from now, could lead to cheap terabit-per-second connections within and around computers.

Lasers and other optoelectronic devices carry billions of bits through our telecommunications networks every second. But the materials they’re made from, exotic semiconductors such as indium phosphide, and the costly manufacturing techniques involved in their production have kept such gigabit-per-second connections largely confined to long-haul telecommunications. By integrating optoelectronic devices on silicon chips, Intel and other companies, notably Luxtera, in Carlsbad, Calif., and STMicroelectronics, in Geneva, hope to make optoelectronic bandwidths affordable enough for your average notebook computer.

It has been a difficult quest. Silicon is not a natural for producing and manipulating light. Nevertheless, Intel and Luxtera each have been able to produce silicon versions of optoelectronic components, such as waveguides and the modulators that encode data onto the laser. Intel even produced a silicon laser chip, but impractically—it had to be powered by light from a separate laser [see ”The Silicon Solution,” IEEE Spectrum, October 2005].

Now the researchers have been able to overcome that problem by binding a light emitter made from indium phosphide to a silicon laser cavity. The key was in making a kind of ”glass glue,” a thin layer of oxidized material, on both the indium-phosphide light emitter and the silicon laser, and then bonding them together [see illustration, "Stacked Up"]. Applying a voltage to the indium-­phosphide device produces light that passes through the glass into the silicon.

Intel still has some kinks to work out of its new laser. For one, the laser peters out if the temperature rises above 40 °C. Commercial chips typically must work at up to 80 °C. ”We have a pretty good idea of what the limitation is. We have a new test chip in design and hope to move [the working] temperature up above 70,” says Mario Paniccia, director of Intel’s photonics technology lab. His group is also looking to reduce the amount of current needed to get the laser shining. Right now, it takes 65 ­milliamps, but Paniccia hopes to get the device operating on less than 20 mA.

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions