Needle beam could eliminate signal loss in on-chip optics

September 7, 2012
Needle beam could eliminate signal loss in on-chip optics
Researchers led by Federico Capasso at SEAS demonstrated a cosine-Gauss plasmon beam, dubbed a "needle beam," that propagates without diffraction. The advance may assist the development of ultrapowerful microprocessors. (Courtesy of Patrice Genevet.)

(Phys.org)—An international, Harvard-led team of researchers have demonstrated a new type of light beam that propagates without spreading outwards, remaining very narrow and controlled along an unprecedented distance. This "needle beam," as the team calls it, could greatly reduce signal loss for on-chip optical systems and may eventually assist the development of a more powerful class of microprocessors.

Based at the Harvard School of Engineering and Applied Sciences (SEAS) and the Laboratoire Interdisciplinaire Carnot de Bourgogne, CNRS, in France, the applied physicists both characterized and created this needle beam, which travels efficiently at the interface of gold and air. Their findings were published online August 31 in the journal .

The needle beam arises from a special class of quasiparticles called surface plasmons, which travel in tight confinement with a . The metallic stripes that carry these surface plasmons have the potential to replace standard copper electrical in microprocessors, enabling ultrafast on-chip communications.

Needle beam could eliminate signal loss in on-chip optics
Top: A micrograph and diagram of the metallic gratings that produce the needle beam. Bottom: An approximation of the experimental setup. A laser is focused from the glass substrate side onto the device. Once the non-diffractive surface wave is created, detailed information on its intensity distribution is gathered using an ultrahigh-spatial-resolution near-field scanning optical microscope. (Images courtesy of Patrice Genevet.)

One of the fundamental problems that has so far hindered the development of such optical interconnects is the fact that all waves naturally spread laterally during propagation, a phenomenon known as diffraction. This reduces the portion of the signal that can actually be detected.

"We have made a major step toward solving this problem by discovering and experimentally confirming the existence of a previously overlooked solution of Maxwell's equations that govern all light phenomena," says principal investigator Federico Capasso, Robert L. Wallace Professor of and Vinton Hayes Senior Research Fellow in Electrical Engineering at SEAS. "The solution is a highly localized surface plasmon wave that propagates for a long distance, approximately 80 microns in our experiments, in a straight line without any diffraction."

The so-called needle beam, the technical term for which is a cosine-Gauss plasmon beam, propagates in tight confinement with a nanostructured metal surface. Lead author Jiao Lin, a visiting postdoctoral fellow at SEAS from the Singapore Institute of Manufacturing and Technology, and coauthor Patrice Genevet, a research associate in Capasso's group, found an ingenious way to demonstrate the theorized phenomenon. They sculpted two sets of grooves into a gold film that was plated onto the surface of a glass sheet. These tiny grooves intersect at an angle to form a metallic grating. When illuminated by a laser, the device launches two tilted, plane surface waves which interfere constructively to create the non-diffracting beam.

"Our French colleagues did a beautiful experiment, using an ultrahigh-resolution microscope to image the needle-shaped beam propagating for a long distance across the gold ," says Genevet.

Capasso's team hopes the finding will assist the development of more energy-efficient and powerful microprocessors.

Explore further: Scientists demonstrate laser with controlled polarization

More information: prl.aps.org/abstract/PRL/v109/i9/e093904

Related Stories

Scientists demonstrate laser with controlled polarization

April 13, 2009

Applied scientists at the Harvard School of Engineering and Applied Sciences (SEAS) in collaboration with researchers from Hamamatsu Photonics in Hamamatsu City, Japan, have demonstrated, for the first time, lasers in which ...

Building a more versatile laser

November 16, 2009

(PhysOrg.com) -- One of the drawbacks associated with using semiconductor lasers is that many of them can only produce a beam of a single wavelength, and can only send that beam in one direction at a time. There have been ...

Recommended for you

New blow for 'supersymmetry' physics theory

July 27, 2015

In a new blow for the futuristic "supersymmetry" theory of the universe's basic anatomy, experts reported fresh evidence Monday of subatomic activity consistent with the mainstream Standard Model of particle physics.

Rogue wave theory to save ships

July 29, 2015

Physicists have found an explanation for rogue waves in the ocean and hope their theory will lead to devices to warn ships and save lives.

Researchers build bacteria's photosynthetic engine

July 29, 2015

Nearly all life on Earth depends on photosynthesis, the conversion of light energy into chemical energy. Oxygen-producing plants and cyanobacteria perfected this process 2.7 billion years ago. But the first photosynthetic ...

4 comments

Adjust slider to filter visible comments by rank

Display comments: newest first

TheGhostofOtto1923
3.3 / 5 (19) Sep 07, 2012
This may or may not be related but surface plasmon polarities are what NASA is investigating re the widom-Larsen theory and LENR.
http://www.youtub...a_player
http://newenergyt...#summary

-You know, COLD FUSION.
drhoo
5 / 5 (3) Sep 07, 2012
Vaguely sounds like a waveguide, no???
TheGhostofOtto1923
3.4 / 5 (18) Sep 07, 2012
This may or may not be related but surface plasmon polarities are what NASA is investigating re the widom-Larsen theory and LENR.
http://www.youtub...a_player

-You know, COLD FUSION.
Thats uh polaritons. Spellcheck is as dumb as pussytard/esteban.
Mike_Massen
1 / 5 (1) Sep 07, 2012
Knew an engineer back in 1998 who wrote his thesis on optical comms in large scale integrated cpus as way to manage core to multiple cache latency issues - congratulations Andrew your work back then is being followed up with this new approach :-)

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.