Research From Duke University Moves Us One Step Closer To Optical Computing

Microchips today use electrons to process and transmit data, but if they were to use photons, that information could be processed at a much faster rate. Creating a light source that can be turned on and off several billion times per second to replace the billions of transistors in current designs is something for which the scientific community has been searching. Lasers are capable of the task, but that technology is too power hungry and cannot be scaled down small enough for use in a microchip.

A team of researchers from the Pratt School of Engineering published a study in the Nature Communications journal called "Ultrafast Spontaneous Emission Source Using Plasmonic Nanoantennas." In the study, the team explained how it managed to push semiconductor quantum dots, which are six nanometer-wide spheres of semiconducting material, to switch between on and off states more than 90 billion times per second. Scientists believe devices using this technology could one day be used for optical computing chips or faster communication between traditional electronic chips.

The team shined a laser onto a 75 nanometer-wide silver cube. The free electrons found on the surface of the cube began oscillating together in a wave. This process creates its own light source, which further reacts with the electrons. Plasmon is the energy trapped in the surface of the nanocube in this fashion.

A thin sheet of gold is then placed next to the silver nanocube. When placed 20 atoms away from the silver nanocube, the plasmon energy creates an intense electromagnetic field. The Duke researchers used this field to interact with the quantum dots. The team said that by using this technique, the emission of photons can be directionally controlled, and turned on and off at 90 GHz, efficiently.

Fluorescent materials used in traditional LED designs are limited by a slow emission rate, lack the ability to direct photons and are inefficient, which has been the barrier for using LEDs in optical communications devices. Secure communication requires directed source photons, and the group said extremely secure quantum communication will require a single photon source.

The team is now working on creating a structure using plasmonics that can create such a direct source, and is currently experimenting with placing a single quantum dot between the gold foil and silver cube. The team is using precise orientation and placement of the quantum dot in an effort to maximize the fluorescence rate.

Maiken H. Mikkelsen, assistant professor of electrical and computer engineering and physics at Duke and a member of the research team, said that, as they have done with semiconductors, they can create new designer materials with almost any optical properties while tailoring the environment around the material.

"Ultrafast Spontaneous Emission Source Using Plasmonic Nanoantennas" was written by Thang B. Hoang, Gleb M. Akselrod, Chistos Argyropoulos, Jiani Huang, David R. Smith and Maiken H. Mikkelsen of the Pratt School of Engineering at Duke University. It was published in Nature Communications on July 27, 2015.

Update, 7/27/15, 7:32am PT: Fixed typos.

Follow Kevin Carbotte @pumcypuhoy. Follow us @tomshardware, on Facebook and on Google+.

 Kevin Carbotte is a contributing writer for Tom's Hardware who primarily covers VR and AR hardware. He has been writing for us for more than four years. 

  • Kewlx25
    "Fluorescent materials used in tradition traditional LED designs"
  • scolaner
    "Fluorescent materials used in tradition traditional LED designs"

    Fixed, thank you.
  • TallestJon96
    It seems to me that regular silicon is becoming harder and harder to utilize with quick growth. Intel's announcement of a tick-to k-tock cycle demonstrates that the leading CPU manufacturer cannot keep up with Moore's Law, and I would argue that Moore's law slowed even before this.

    I think in order to keep up with Moore's law new methods like this will have to be implemented. Obviously this process isn't nearly in a manufacturing stage, or even taped out for a practical CPU design, but if first generation optical computing elements can operate at 90ghz, then there is a lot of room to grow from our 4-5ghz.
  • Wisecracker
    What the article left out: The researchers from Dook went to school in Chapel Hill ...