Most modern information travels long distances in the form of infrared laser light, but when that data reaches its destination it requires devices called photodetectors to translate the optical language of data transfer into the electronic language of computation. This week, researchers from the Vienna University of Technology (VUT) managed to create a silicon chip with an integrated photodetector made of graphene. The wonder material that seems poised to revolutionize so many industries might just change the way we build computers and networks, too.
There are two key advantages to using graphene in this case. The first is simple: speed. Oodles and oodles of speed. Even advanced photodetectors made of rare elements lag behind graphene in terms of transmission speed and responsiveness (latency). That’s because graphene has a special sub-atomic arrangement that makes it “aromatic” — in other words, graphene has such a well-balanced distribution of electrons within its atomic structure that those electrons can flow with virtually no resistance or delay. This means that an exciting event like an incoming photon can be transformed into an electronic signal much more quickly than by any previous model. The responsivity of this newest integrated detector is actually eight times greater than previous graphene models which existed separate from the chip itself.
The second advantage to graphene is the incredibly small size at which it will perform this function. The team claims that their photodetector can be made so small that 20,000 of them could fit onto a chip just one square centimeter in size. This means that a sufficiently high-resolution transfer mechanism could theoretically provide that single chip with 20,000 independent lines of information. Whether the chip could handle such input is another matter entirely.
The team’s major hurdle wasn’t proving graphene’s amazing abilities in optoelectronics, but integrating those abilities into a chip itself; you can’t just swap in this new graphene photodetector and use it with any old processor. A working integrated photodetector means, among other things, that optical data transfer could be useful within a computer. This means that multiple cores in a single system could communicate more quickly, and would result in a machine that wastes far less electricity.
Graphene can also soak up and convert energy from the entire spectrum of light used in modern communications. Older photodetectors, and even advanced modern ones made of elements like germanium, can only absorb a single wavelength of incoming light. The researchers showed that graphene can absorb light from all across the usable spectrum equally well, from 1310nm to 1650nm. All it requires in this setup is a waveguide that directs incoming light onto the tiny graphene detector. (See: IBM creates first cheap, commercially viable, electronic-photonic integrated chip.)
The work from VUT was actually one of three graphene absorptivity studies published in Nature Photonics this week — the possible applications of stretched graphene continue to grow at an incredible pace. It has fundamental physical properties that surpass anything that’s come before — the same virtues that give graphene promise for superconducting cables, for instance, make it useful in photo-conversion. Though many in the public are becoming jaded about the sheer volume of proposed uses for graphene, the carbon rush will almost certainly continue for some time. (Read: The wonderful world of wonder materials.)
The main drawback of graphene for optical applications is its sensitivity; while the electronic transfer speed is unmatched, its ability to respond to low light intensities is quite poor — as much as 10 times worse than its competitor, germanium. The researchers remain confident that they have a method of addressing this problem, but their integrated chip is already enough to demand attention from anyone looking to predict the next frontier in computing.
Research paper: doi:10.1038/nphoton.2013.240 – “CMOS-compatible graphene photodetector covering all optical communication bands”