Skip to Content
Uncategorized

Graphene Could Make Data Centers and Supercomputers More Efficient

New research suggests graphene could enable highly efficient optical communication in chips for data centers and supercomputers.
October 2, 2013

Computer chips that use light, instead of electrons, to move data between electronic components and to other chips could be essential for more efficient supercomputers and data centers. Several industrial research labs are working toward such optical interconnects that rely on germanium to turn light into ones and zeros. But recent research suggests that graphene devices could be far better and cheaper.

Illuminating graphene: A colored scanning electron microscope image shows a graphene photodetector (purple) integrated with a silicon waveguide (the ridge in the middle). On each side sits a metal electrode (green).

An optical interconnect consists of a modulator that converts electrical signals into optical ones, and a photodetector, which does the reverse. Current iterations feature modulators made of silicon and photodetectors made of germanium. Intel recently announced plans to use such technology and begin manufacturing a product it calls “silicon photonics,” for use in data centers (see “Intel’s Laser Chips Could Make Data Centers Run Better”).

But graphene photodetectors have a good chance to equal or surpass the performance of germanium ones in several important aspects within a few years, says Dirk Englund, a professor of electrical engineering and computer science at MIT. Although graphene devices are still about an order of magnitude behind germanium in terms of capacity to generate current in response to the absorption of light, they have improved immensely in this area in just a few years.

Graphene has a number of potential advantages over germanium, says Englund. Because of its exceptional electronic properties, devices made of the material can work at very high frequencies, and could in principle handle more information per second. Also, graphene can absorb a broader range of wavelengths than germanium can. That property could be exploited to transmit more data streams simultaneously in the same beam of light. Further, unlike germanium detectors, graphene photodetectors work “quite well” without applied voltage, which could reduce the energy needed to transmit data, says Englund. Finally, he says, graphene detectors would in principle require a simpler and potentially less expensive process to integrate them on a silicon chip.

But graphene as a photodetector material has at least one big problem: it does not strongly absorb light. To address this issue, three groups—one led by Englund, one led by Thomas Mueller, a professor at Vienna Institute of Technology’s Institute of Photonics, and a team at the Chinese University of Hong Kong—have separately designed on-chip detectors consisting of a graphene sheet paired with a silicon waveguide, a component that confines and routes light and maximizes the interaction.  

The first graphene-based modulator was demonstrated in 2011. So this recent work suggests it might be possible to build an optical interconnect entirely out of graphene.

An important caveat to the new demonstrations is that, in each case, graphene was transferred to the silicon substrate by mechanical processes that are not conducive to large-scale manufacturing. There has been recent progress toward developing a large-scale process for growing graphene on desired substrates, but that option is not yet viable.

Further, graphene photodetectors are very far behind germanium ones in terms of technology development, cautions Solomon Assefa, a research scientist at IBM’s T.J. Watson Research Center. With graphene, he says, “there’s still a lot to be done. It could potentially be cost-effective, but we still have to find out.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.