Print 8 comment(s) - last by subhajit.. on Feb 8 at 5:49 AM

A diagram of the nanolaser's nanopillar structure, which was grown directly on silicon  (Source: University of California, Berkley)

The researchers took microscopic snapshots of the grown lasers.  (Source: University of California, Berkley)

They also imaged the laser in action.  (Source: University of California, Berkley)
Can we get one of these on our motherboard, please?

A major bottleneck to modern computing power is latency -- the time it takes for a CPU to communicate with other devices like a graphics chip, or memory.  One possible solution is to switch from using an electrical bus to an optical one.  In well-made glass, light can travel at nearly two-thirds its speed in the vacuum an astonishingly fast transfer speed.  The key, though, is that you need something to emit a high intensity burst of light (a laser pulse).

Researchers at the University of California, Berkeley have completed a work that inches the computer industry closer to this dream.  They've developed a new process that will allow laser-emitting nanopillars to be directly grown onto silicon at milder conditions than in past work.

Silicon itself is remarkably bad at emitting light, so researchers must use alternate semiconductors.  One of the most promising candidates is so-called "III-V" semiconductors, which consist of one transition metal element, and one semiconducting element from the fifth group of the periodic table, like gallium.

But growing these elements onto silicon using traditional process technologies has been pretty much impossible, due to the mismatch complexities and high temperature constraints.  

The study's lead author, Roger Chen [profile], a UC Berkeley graduate student in electrical engineering and computer sciences, explains [press release], "Growing III-V semiconductor films on silicon is like forcing two incongruent puzzle pieces together. It can be done, but the material gets damaged in the process."

Adds Connie Chang-Hasnain [profile], UC Berkeley professor of electrical engineering and computer sciences who served as the project's principle investigator; "Today's massive silicon electronics infrastructure is extremely difficult to change for both economic and technological reasons, so compatibility with silicon fabrication is critical. One problem is that growth of III-V semiconductors has traditionally involved high temperatures – 700 degrees Celsius or more – that would destroy the electronics. Meanwhile, other integration approaches have not been scalable."

The researchers developed a process to deposit nanopillars of indium gallium arsenide, a III-V semiconductor, at only 400 degrees Celsius.  The metal-organic chemical vapor deposition method used is already in commercial use in the solar cell and LED industry.

The resulting nanopillar structure is hexagonal and amplifies an infrared (950 nm) laser signal.  Spiral helically up the pillars, the laser is emitted at their ends.  This laser-cavity mechanism is complex theoretically, but the bottom line is the researchers have created an on-chip laser that's grown using more affordable process.

States Professor Chang-Hasnain, "This is the first bottom-up integration of III-V nanolasers onto silicon chips using a growth process compatible with the CMOS (complementary metal oxide semiconductor) technology now used to make integrated circuits. This research has the potential to catalyze an optoelectronics revolution in computing, communications, displays and optical signal processing. In the future, we expect to improve the characteristics of these lasers and ultimately control them electronically for a powerful marriage between photonic and electronic devices."

The study on the work was published [abstract] in the prestigious peer-reviewed journal Nature Photonics.  The work was funded by a grant from The Defense Advanced Research Projects Agency (DARPA) and a Department of Defense National Security Science and Engineering Faculty Fellowship.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

harks with frickin' laser beams
By BladeVenom on 2/7/2011 12:30:35 PM , Rating: 5
Can you put them on sharks?

Sorry, I can't help myself.

RE: harks with frickin' laser beams
By Flunk on 2/7/2011 1:04:16 PM , Rating: 3
I'm sure you could, but what's the point of putting microscopic lasers on sharks?

RE: harks with frickin' laser beams
By nayy on 2/7/2011 4:05:23 PM , Rating: 3
Actually these are perfect for my bio-engineered nano sharks!:D

Not as close as it seems...
By 3DoubleD on 2/7/2011 5:30:42 PM , Rating: 5
Reading the article I have the following observations to add to the article:

1) These lasers were not electrically pumped, but optically pumped. This doesn't help electric transistors send signals across the chip. Future work would focus on making this an electronically pumped laser.

2) The silicon substrates were annealed at 600 C before the laser "nano" pillers were grown. This isn't a very low temperature process.

3) It appears they had no control over where the pillars formed. This is a big problem if you are trying to mass-produce optical interconnection on a CMOS chip.

4) I say "nano" because general nano is used to refer to structures less than 100 nm, these pillars have diameters greater than 500 nm. No doubt these large dimensions were required to reduce surface recombination effects (even with the presence of the GaAs passivation layer). Attempts to reduce the dimensions of this structure may result in a poor performing device. This is a problem that many are currently working on.

I think the biggest reason why this paper was published was due to the unique laser cavity mode. To prove my point - this paper was published in Nature Photonics. The stuff about integrating III-Vs with Si is a bunch of fluff that most students add to their introductions as the root motivation behind their work.

RE: Not as close as it seems...
By subhajit on 2/8/2011 5:49:52 AM , Rating: 2
We are waiting for a optical computing breakthrough.

fishy and cheap
By zkln on 2/7/2011 8:07:53 PM , Rating: 1
cheap on-chip
What the hell were you thinking when you were typing this?
My native language is not English, but I'm literate enough to know that this is considered a literal disaster. It simply makes my eyes and ears bleed. You could've scratched a black board with your fingernails as well.
How about inexpensive on-chip lasers ?

RE: fishy and cheap
By someguy123 on 2/7/2011 10:03:09 PM , Rating: 2
What's the problem with using the word cheap?

optical interconnect MB
By kattanna on 2/7/2011 1:13:32 PM , Rating: 2
a motherboard where all components used an optical connection is a dream i have had for sometime, as it will really change the face of computing.

"So, I think the same thing of the music industry. They can't say that they're losing money, you know what I'm saying. They just probably don't have the same surplus that they had." -- Wu-Tang Clan founder RZA
Related Articles

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki