Originally published in Business Digest, November 1997

Brilliant Technology

by Craig C. Bailey

Like many scientific discoveries, Audrey Engelsberg's was made serendipitously. But while the genesis of the company Engelsberg would help found in 1988 was accidental, the development of the firm has been the exact opposite. An intellectual property company with a somewhat involved structure, Cauldron Co., the parent company of Radiance Laboratories Inc. in South Burlington, has not integrated its high tech patent into the marketplace, yet the valuation of the firm has grown twentyfold in less than 10 years to $200 million. If you think this business is rolling forward on sheer power of potential and selling smoke along the way, you're only half right. Actually, it's selling light.

Engelsberg is clearly into lasers. From the conference room overlooking the X-Files-like softwall cleanroom at the Radiance facility -- "It's more like Buck Rogers," she contends -- Engelsberg occasionally pops up to sketch a diagram on the markerboard to explain her point. "This technology removes contaminants from a surface by using high-energy photons and any flowing inert gas, such as nitrogen, argon or even clean, dry air," she explains. "The contaminant is removed without damaging the underlying substrate, which is critical in a lot of applications."

Ironically, no one seems to be able to explain why the process works. "The experts are still arguing about that," says senior engineer Andy Johnson, the only other person to staff the company's 2,500-square-foot Green Tree Drive facility. It's a simpler explanation than is offered by company literature, which suggests photodissociation, phonon shock, and quasi-metastable states, among others.

[Audrey Engelsberg] Several years ago, Audrey Engelsberg, co-founder of Cauldron Co., the parent of Radiance Laboratories Inc. in South Burlington, turned down IBM's $50,000 offer for the rights to her laser cleaning process. Cauldron is now valued at $200 million. (Photo: Jeff Clarke)

"The nice thing about this technology is there's so many applications," Engelsberg adds. The company has focused its efforts on the semiconductor industry and industrial applications such as tire manufacturing, but the list of potential uses of cleaning with lasers is nearly endless, according to Engelsberg, and includes everything from urinal molds to parts used in manufacturing doorknobs.

"I've done things as small as the end of the spring that holds the read/write head on a disc drive," says Johnson, "and I've done things as big as 27-inch television tube models."

"This is basically a test bed system," Engelsberg adds, standing within the clear, plastic walls of the cleanroom, an oasis of light not much larger than a bathroom, within the larger laboratory, which is mostly empty. "It's very, very flexible. You just modify how you hold the substrate." Johnson, who hasn't donned his white bunny-suit for the experiment, has a circular piece of plastic attached to the workbench-like contraption, as he fills the laser with a krypton/neon/fluorine mixture that's shipped premixed from a firm in New Jersey. "It needs care and feeding," Engelsberg jokes.

The material in question this particular afternoon is a sample of plastic used by a manufacturer of ink jet printers. The test is simple: "They've given me some parts, and they put fluorescent dirt on it. They measure them ahead of time, so they know about what they look like," Johnson says. "I'm going to clean them, and send them back. And they're going to see how much of the dirt I got off." Operating at a spectrum lower than that visible with the naked eye, the laser beam itself cannot be seen. "But you can see the effects of it," Johnson remarks. "It makes things fluoresce, like black light."

Today's experiment will help determine if the patented system Engelsberg servicemarked as the Radiance Process -- "The Engelsberg Process didn't sound right," she says -- can be used by this particular manufacturer, whose name must remain "mum."

Engelsberg, born in New York City and raised on Long Island and in Boston, was working on her doctoral dissertation at Rensselaer Polytechnic Institute (RPI) in Troy, N.Y., in the mid '80s when she discovered the process. "I originally wanted to go to medical school, and decided that sheer memorization was not my bag," she says. "I decided I didn't want to go to law school, because I thought that lawyers were yucky." After a bachelor of science in chemistry earned at Framingham State College outside of Boston in 1983, she entered the chemistry department at RPI, where she started working with lasers.

She was trying to deposit a line of material using an excimer laser in 1987, when she found, much to her chagrin, she was removing material instead. Commiserating on the phone to her friend Joe Dehais, an RPI management Ph.D. candidate when she met him in September 1983, the benefits of her discovery weren't immediately apparent.

Before long, its significance, combined with Dehais's entrepreneurial experience from starting a company in 1985, led to visions of a patent. "We didn't know how big the technology was at the time," says Engelsberg, but the two did know the first step was to present the discovery to RPI, which, through a standard agreement, had the right to claim any patentable discovery of Engelsberg's, if it wished. "They didn't care about it," according to Engelsberg, citing an ignorance of the value of intellectual property on the behalf of educational institutions in the late 1980s. "They just gave me a piece of paper saying, 'It's yours.'"

Finishing her doctorate in physical chemistry in 1987, Engelsberg moved on to work for IBM's East Fishkill, N.Y., facility, while the patent process proceeded. In 1988, Dehais, who holds a bachelor's degree in chemical engineering from Princeton University and a master's from RPI, and Engelsberg formed Cauldron Co., holding company of Radiance Services Co., to hold the impending patent. They teamed with Ralph Benko, a lawyer associate of Dehais from his previous business endeavor, who was a finance attorney-adviser for the U.S. Department of Energy, and involved with the President's Commission on Privatization among other Washington endeavors.

In June 1990, Engelsberg moved to Vermont to join Matt Rutten, an RPI alum working at the IBM Essex Junction facility, whom she had started to date. Relying on the "RPI Mafia" in place at IBM, she got a job at the Essex firm, intending to work until her patent went through and Cauldron was ready to become a full-time occupation.

Rutten and Engelsberg became married and now live in Milton with their two cats, Max and O.C.

The patent was issued in June 1991, shortly after Donna Fitzpatrick, then assistant secretary of energy in Washington, was taken in to help with marketing. All four remain principals, owning 70 percent of Cauldron Limited Partnership, the financial arm of Cauldron Co., with another 175 limited partners owning the rest.

In less than a year, Engelsberg left IBM.

She explains the company has never strived to be a tool manufacturer, or, as she says, "bricks and mortar. We believed we could leverage the patent by doing a sufficient number of demonstrations of technology. That someone would license the technology or buy us out for a significant amount of money." It was the patent, the intellectual property that the company has always been interested in -- the notion of leverage, of using other companies' facilities to experiment and demonstrate the process, while keeping Cauldron Co. somewhat virtual. When the firm was formed, its offices were the various New York and Maryland domiciles of its founders. Now its offices are in Bethesda, Md., with Engelsberg in Vermont.

Johnson has fired up the laser, and a blue glow radiates from the lab outside the conference room as Engelsberg rattles off a half-dozen recent legal cases involving major high-tech companies battling over intellectual property. "Doesn't that tell you something about patents? He who owns the intellectual property has potential control of the market," she stresses. "It's not whether or not you can build the better hardware."

The company soon set about raising money by turning to the families of Benko and Fitzpatrick. "Standard blood money, which is typical of all start-up businesses. That's how we got our first million dollars," says Engelsberg, who refers to the whole finance side of the start-up as "a black art of arts." Ever since, the company has survived and grown on investors' interest in the company. She adds that industry awards the technology has garnished have continued to attract investment bankers.

While the benefits of a cleaning system that uses no dangerous chemicals -- in fact, not even water! -- are obvious today, they weren't as integral to the company's pitch in the early years. In the beginning, many semiconductor insiders claimed there wasn't a cleaning problem. Industry opinion changed in 1992, when an article published in a trade magazine made cleaning a hot topic. "When that article came out we were able to go from a $5 million valuation first offering to a $40 million valuation," says Engelsberg.

Eventually, after eight years of "leverage," including time spent using the Microelectronics Research Laboratory in Columbus, Md., and the lab of Jim Parker's Diffraction Ltd. in Waitsfield, the company decided it needed its own facility to appear more "real." In August 1996 it moved into its South Burlington space, and hired Johnson. With 24 years' experience with IBM and General Electric, he had been under contract with Diffraction, and had worked with Engelsberg in Fishkill.

Still, after nearly 10 years of experimenting and demonstrating, the Radiance Process hasn't been implemented in the field. The semiconductor industry is a prime market. "We're heading toward 300 cleaning steps to make a chip on a wafer," Engelsberg says. "This technology has the potential to probably displace maybe a third or more of the wet cleaning steps in a fab," but, she adds, the industry is slow to adopt new procedures.

"Basically what it comes down to is risk and reward," according to Todd Lizotte, director of research and development, new product development, at Neuman MicroTechnologies Inc., a technology research house and laser equipment manufacturer in Concord, N.H., licensed to build equipment for the process. Lizotte says there are yield losses for the semiconductor industry associated with cleaning, but no individual manufacturer is particularly concerned because all foundries share the loss equally. And no single manufacturer is eager to jump in and try a new technology for fear it will bear the brunt of development costs, the benefits of which will then be shared across the industry. Besides, by keeping yield lower, the price of chips remains high.

He is optimistic about the Radiance Process. "I think it could be pretty widespread. The applications of the laser -- just the things that we've been working on -- have gone from the automotive industry to semiconductors," he says.

"We've found that heavy industry or industrial applications are much more open to change. The first industrial application that's being seriously considered for the technology is for tire molds," Engelsberg says, before adding that Neumann recently completed successful demonstrations for a "major tire company," cleaning a tire mold in about a quarter of the time it takes through the traditional glass bead blasting method, without the negative effects of intense heat and undue wear on the mold. "The savings are considerable," she adds.

Engelsberg spends many of her days on airplanes and shuffling in and out of hotels, doing what she calls "technical marketing," while Johnson staffs the South Burlington lab. For now, she's enjoying a rare stretch in Vermont before family obligations take her to New Jersey within a week, with surely more business travel after that. "There comes a point in every company's life cycle when the founders have to leave," she offers, turning philosophical. "If I left tomorrow, I'd be perfectly satisfied with what I did. We have strong patents. The basic technology is there. Now it's just engineering."