New Light In Optics

 How a little-known ECE Lab scored a series of firsts

November 26, 2003

"Great laboratories are vertically integrated," says Harold Lessure. "In Edison's laboratory they did everything from investigating materials for light-bulb filaments to inventing the movie projector. And our lab is also vertically integrated. We can go from understanding the quantum mechanics of materials to developing complete commercial devices based on that understanding."

Lessure, a research professor, directs an ECE lab that has been quietly chalking up breakthroughs in optical sensing. The lab already has helped to develop a first-of-a-kind product: an optical methane detector (OMDTM) that utility companies are mounting on service vehicles to find natural gas leaks. A versatile handheld detector is now in the works, and in research on basic materials, Lessure's group has grown an extraordinary new kind of crystal that could have uses in fields from military surveillance to astronomy (see "The Quest for a Crystal").

All of this has come together fortuitously. The gas-detector work and the crystal research were initiated at the former Westinghouse Science and Technology Center. When the company folded the R&D center in 1996, these projects were taken over by Carnegie Mellon Research Institute. Key ex-Westinghouse people came aboard to join Lessure, who then was leading a nitrous-oxide sensor program at CMRI. And when CMRI itself was dissolved last year, ECE picked up Lessure and his staff, who by that time had become a potent team.

"We started with bright ideas that were still only partially developed," Lessure says. "We just keep working to take them to their full potential."

The Beauty of OMD

The gas detection story shows what optical sensing can do. Natural gas utilities maintain safety by searching for early traces of leaks from buried pipelines. Previously the best instruments were flame ionization detectors (FIDs), which work by drawing air samples over a hydrogen-fueled pilot light. But that method is slow and balky. A vehicle trolling for leaks has to crawl the streets at a few miles per hour and must carry an extra crew member to tend the FID, which can be fouled by wet or dirty air, or tricked into false alarms by other substances.

Enter the Optical Methane Detector, which Lessure calls "an agile spectroscopic system." It will detect less than a part per million of methane [in the air] within milliseconds," he says. "We've tested it at vehicle speeds to 60 miles per hour." The principle behind the OMD is simple: methane molecules absorb certain wavelengths of infrared light and thus have an optical signature. The instrument — mounted across the front of the vehicle — has a light source and tunable filter at one end and a receptor unit at the other, with an open light path in between. The filter is tuned so that transmissions periodically match the absorption spectrum of methane. Any methane drifting across the path will then block part of that signal and set off alarms — while other molecules will not.

"We've driven an OMD through the Squirrel Hill Tunnel at rush hour to demonstrate its resistance to false signals from exhaust gases," Lessure says. "We've driven through service stations to be sure it's not picking up gasoline vapors." The instrument works in rain, snow, fog and dust. It can be operated by the driver alone.

The Hard Road to Market

Getting the OMD to this wondrous state was not simple, however. "This thing has gone from a 250-pound unit mounted on a truck, using one and a half kilowatts, with laboratory electronics in the back, to a 35-pound, 60-watt self-contained unit that you just turn on and use," Lessure says.

Along with miniaturization, many problems had to be solved: perfecting the tunable filter and the control systems, writing software, and more. Work was sponsored by a consortium of utilities through the Gas Research Institute. The leading instrument maker Heath Consultants, Inc., was chosen to commercialize the OMD. Lessure's team produced the commercial optical and electronic designs and transferred the technology, working closely with Heath's engineers.

"Then we had to sell the regulatory agencies on it," Lessure recalls. Existing specs didn't allow for optical gas detection. So the team used a two-wheeled golf-bag carrier to hold their new OMD prototype and went off to an industry conference for a demo and test. "They asked us to find a minor leak in the neighborhood that they knew existed," Lessure says. "We found it. Plus another leak they didn't know they had."

Heath began selling the OMD in 1998. Gas Utility and Pipeline Industries magazine named it the hottest product of the year. Now used in 17 countries, OMDs are credited with saving millions of dollars while performing a crucial safety task better. Meanwhile Lessure's group, again with gas utility sponsors, is moving ahead.

Next In The Pipeline

Ready for demo this winter is a prototype of a handheld portable methane detector (PMD), a super-compact "walking stick" version of the OMD also sponsored through the Gas Research Institute. Patented new filtering and light-collection schemes enable all the gear to be placed along a thin aluminum shaft with a handle. The PMD weighs three and a half pounds, draws 10 watts from a battery pack and has a market potential far greater than the vehicle-mounted OMD. Utility crews use large quantities of portable detectors, carrying them over rough terrain or inside buildings both to pinpoint leaks and to "grade" them (i.e, measure the severity just above the source) — and they're currently using flame-based units that the PMD can outperform.

Better yet, as the lab's research progresses, future optical detectors will be tunable to trace both methane and ethane. Although ethane is harder to detect — natural gas contains only small amounts — its presence can help confirm that what you've got is a natural-gas leak rather than, say, methane from decay in a nearby landfill.

Others are now coming to market with optical gas detectors. But Lessure notes that most competing designs use tunable lasers. This, he says, makes them far more expensive than (and not as versatile as) the Carnegie Mellon designs, which use a simple non-collimated quartz light with a tunable filter. Many more possible applications beckon: detectors for many gases, in military and civilian settings. Lessure's group intends to play.

And, he adds, "we're by no means the only people in ECE doing photonics research." Citing other professors with groundbreaking work in various areas — David Casasent, Takeo Kanade, David Lambeth, Ed Schlesinger, Dan Stancil, Ozan Tonguz, Elias Towe — Lessure concludes: "This department is building the strength to really be a leader."

The Quest for a Crystal

Mounted in an optical test stand in Harold Lessure's ECE lab is a block of material, about half the size of a matchbox, that looks like clear glass. It is a polarizing prism made from a crystal of mercurous chloride. No one, until now, has been known to make such a thing — largely because few have succeeded in growing crystals of the required quality.

Mercurous chloride (Hg2Cl2) has been made for centuries in the form of white caustic powder called calomel. In recent decades, harnessing the crystalline form for electro-optics "has become a sort of holy grail," says Lessure, as the material would make an ideal wideband filter for all kinds of scanning and analysis.

It transmits long-wave infrared as well as visible light, whereas most optical crystals absorb strongly in the infrared. It is a powerful polarizer at all wavelengths — useful for bringing out "hidden" aspects of an image, much as polarized sunglasses or night-vision goggles do — whereas other crystals that transmit long-wave IR will not polarize it. And a mercurous chloride crystal is eminently tunable: the optical parameters change when you impart tiny acoustical vibrations.

With all these features in one package, potential uses are vast. "You could have a compact instrument that helps detect and ID an aircraft or other threat ," Lessure says. "By looking not only at the visible image but the IR, and rotating through angles of polarization, you may be able to pick up features that are camouflaged." Further uses could range from spectroscopic analysis in astronomy to geological surveys and communication systems.

The problem has been to grow optical-grade crystals. Difficulties abound. For instance, mercurous chloride sublimates easily, at around 400°C, so one must apply gradual heat under high vacuum. Linear blocks of Cl-Hg-Hg-Cl atoms form in long strands, creating a crystal that has strongly anisotropic properties — and cleaves readily in certain planes. Crystals produced in the past would have a yellowish or khaki color and be riddled with light-scattering microcracks and other flaws.

But Lessure's team has persisted. Lessure holds a Ph.D. in materials science and engineering from Carnegie Mellon and has had top former Westinghouse researchers working with him: Tom Henningsen, Zoltan Kun, Milton Gottlieb and Dennis Suhre. With long experimentation, the group has not only gotten better crystals more consistently but learned how to fabricate this tricky material once grown.

Prisms in recent tests are colorless. They'll transmit light at wavelengths from 300 nanometers to 20 microns, from the ultraviolet to the long-wave infrared, with very little absorption or scattering. Acousto-optical tuning is done by vibrating a thin piezoelectric transducer on the surface of the crystal. And thanks to the material's incredibly high "birefringence" — the ability to double-refract light into two divergent rays polarized in different planes — a small prism can do a lot of work.

Reprinted with permission, Electrical and Computer Engineering Currents, Fall 2003. Text by Mike Vargo. Photography by Tim Kaulen, Mellon Institute Photography/Graphic Services

In his lab at the Pittsburgh Technology Center, Lessure holds the prototype of a portable methane detector. Weighing only 3.5 lbs., it draws 10 watts from a battery pack and has a market potential far greater than the vehicle mounted OMD.

A prototype of the OMD mounted on a van (above) can detect less than a part per million of methane in the air within milliseconds without giving false alarms for other types of molecules.

A representation of the methane molecule.

Growing optical grade crystals of mercurous chloride has been very difficult. Like the crystal boule shown below, they were often yellowish and full of flaws. Later attempts produced crystals that yielded the flawless clear prism shown above.