Terahertz imaging made national headlines in the wake of the 9/ 11 attacks as a new approach for detecting explosives and nonmetal lethal devices. Terahertz fields with wavelengths longer than infrared radiation but shorter than radio frequencies are sensitive to a range of materials denser than clothing but not as dense as bone or metal, which can be picked up with X-rays.

But the radiation has been difficult to tame. No method for amplifying terahertz signals has been found, and imaging systems using the signals must work with fixed sources and detectors that need to be carefully calibrated to work at all.

A flexible waveguide similar to the optical fibers that were developed for shorter-wavelength lightwaves would greatly extend the flexibility and usefulness of terahertz imaging.

Project shifts gears

"The history of terahertz waveguides is mostly a history of people borrowing waveguide designs from other wavelength regimes, but what we have done has never been implemented at any frequency, so it is, in that sense, quite new," said Mittleman. "Originally, we were investigating the use of sharp metal tips for enhancing the spatial resolution of an image. The technique, known as apertureless near-field optical microscopy, has been demonstrated at microwave and infrared frequencies but had not been explored at terahertz frequencies."

By chance, the researchers illuminated the metal shaft from which the tip had been honed. They found that terahertz signals propagated down its length. "This led us to think about the [potential utility] of propagating radiation down a metal wire, and so we more or less abandoned the microscopy project in favor of this other idea," Mittleman said.

He has determined that a simple metal wire will perform essentially the same function as an optical fiber at visible wavelengths. The terahertz signal propagates down the wire and, when reaching the end, emerges in the same form as the original wave. The system forms a terahertz endoscope that can be directed at any desired target.

The physics behind the terahertz waveguide are distinct from longer-wavelength radio antennas or shorter-wavelength optical fiber. With antennas, the wavelength of the radiation is much longer than the antenna itself, so the entire length is immersed in the field.

"In our case, the length of the wire is many times the wavelength, so we are exciting the structure in a small region, and then that excited region moves down the wire at the speed of light," he said. The terahertz field moves along the wire near its surface. As it passes, it causes the conduction electrons to oscillate, creating an effect known as a surface plasmon polariton. At the end of the wire, the surface plasmon radiates its energy out into free space in the form of a terahertz electromagnetic wave.

This effect "wouldn't work at higher frequencies, because of the increasing losses associated with the decreasing conductivity as frequency increases," Mittleman said. "And it would not be practical at lower frequencies, because the diameter of the wire would be on the order of, or longer than, the wavelength, which would be really big. So, terahertz is really a sweet spot for this waveguide design."

The waveguides seem to be efficient, and Mittleman has not detected much variation with the diameter of the wire or its conductivity. The researchers tried wires ranging from 0.9 to 6 mm.

"We are now interested in what will happen when the wire diameter becomes much smaller than the wavelength of the terahertz wave," he said. "I don't know what to expect in that case, but we are setting up to do the experiment right now."

http://www.eet.com