Nanoscale optics for data transmission

Nanoscale optics for data transmission

Summary: Researchers from Rice University have gained new insights into nanoscale optics by discovering "a universal relationship between the behavior of light and electrons."

TOPICS: Processors

I've already written about plasnomics, this new field which aims to develop optical components and systems similar in size with current integrated circuits (check "A plasmonic revolution for computer chips?"). Now, researchers from Rice University have gained new insights into nanoscale optics by discovering "a universal relationship between the behavior of light and electrons." These new findings could be used to develop nanoscale antennae "that convert light into broadband electrical signals capable of carrying approximately 1 million times more data than existing interconnects." If this looks appealing, don't expect any kind of commercial product in the short term.

Light and electrons both can behave as waves or particles. But the Rice researchers at the Laboratory for Nanophotonics (LANP) have discovered how light-like waves, or plasmons, can interact with nanoparticles of gold or silver.

In recent years there has been intense interest in developing ways to guide and manipulate light at dimensions much smaller than optical wavelengths. Metals like gold and silver have ideal properties to accomplish this task. Special types of light-like waves, called plasmons, can be transmitted along the surfaces of metals in much the same way as light in conventional optical fibers.
When small metallic nanoparticles are positioned on the metal film, they behave like tiny antennae that can transmit or receive light; it is this behavior that has been found to mimic that of electrons. Until now, the coupling of light waves into extended nanoscale structures has been poorly understood.

So how did these researchers set up their experiments?

In the latest research, Halas' graduate student Nyein Lwin placed a tiny sphere of gold -- measuring about 50 nanometers in diameter, within just a few nanometers of a thin gold film. When a light excited a plasmon in the nanosphere, this plasmon was converted into a plasmon wave on the film, for certain specific film thicknesses.

The diagram below shows how looks like a plasmon oscillation for a sphere (Credit: J. Phys. Chem. B, 2003, 107, 668-677). The oscillation frequency is determined by four factors: the density of electrons, the effective electron mass, the shape of the charge distribution, and the size of the charge distribution.

Plasmon oscillation for a sphere

But when will see these nanoscale interconnects carrying approximately 1 million times more data than existing ones or plasnomic components running at frequencies 100,000 times greater than the ones of current microprocessors? It's really too early to tell.

For more information, you should check this page about plasmonics at LANP. And their latest research work has been published online by Nano Letters under the name "Plasmons in the Metallic Nanoparticle-Film System as a Tunable Impurity Problem" (September 14, 2005). It will appear in an upcoming print edition, but here is a link to the abstract.

Finally, here are two links to recent papers related to plasmons and nanoshells, even if they're not centered on electronics, Shining a Light on Cancer Research (PDF format, 3 pages, 472 KB) and Plasmonic Nanoshells (PDF format, 23 pages, 667 KB). The above illustration has been extracted from this presentation.

Sources: Rice University, September 14, 2005; and various web sites

You'll find related stories by following the links below.

Topic: Processors

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Gee Whiz!

    I have been involved in research and development regarding "optical computing" since 1989. These involved optical processors using free space interconnects which paralleled the Bell Labs and Stanford developments using SEED (Self-Electro-optic Effect Device). Generally the research community came to a very important and implementation agnostic result: optical computing can only compete with gated semiconductor technology if 1] a bistable optical gate (flip-flop) can be implemented using optical non-linearities (such as the SEED) and, more importantly, 2] the fan-in and fan-out for cascading decision blocks exceeds around 8 or more. The most important obstacle for optical based computing, to date, has been the fan in/out limitation not the scale of the physical-optical components involved. Fan in/out is required to allow for higher interconnection density of data and decision algorithms and for moving data along a multi-port access data (memory) bus.

    The biggest problem for a plasmon or photon is that you must destroy it in order to determine its state. You can't capture a photon or plasmon; they only exist when they're propagating. This is an inherent characterisitic of data riding on a carrier with zero rest mass. If other gates need that bit, then you must regenerate it for the next and so on. also, this approach requires a wave-guide to transport data from one point to the next. Creating that wave-guide will have the same challenges in routing and density as current microprocessor fabrication processors. So it's doubtful that the "gate" density will ever exceed that of the electronic processor.

    Until the Rice Researchers overcome the fan in/out and data path routing issues, then their science will only apply to very specialized algorithms and processors, at best. I won't be investing in a company that promises optical plasmon super-computers any time soon.
  • cool post !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!