Now, thanks to a new infrared camera whose heart was created in the Center for Photonics and Optoelectronics Materials (poem) at the School of Engineering and Applied Science, art historians can see where Renoir made changes to his masterpiece. His frustration was corroborated last year by the first technical study ever conducted on the painting at the Phillips Collection in Washington, D.C.

The camera reads paintings using infrared reflectometry, an imaging technique that penetrates through upper paint layers and reflects off other subsurface pigments, revealing earlier versions that the artist contemplated. In an essay about the study, Elizabeth Steele, the Phillips Collection's associate conservator, notes that the sitters' relationship to one another and to the viewer underwent radical changes, as did the table setting: "Strongly textured brush strokes that lie below the upper paint layers bear no correspondence to the shapes of the bottles or to the wine cask."

Stephen R. Forrest, a professor of electrical engineering and the former director of poem, conducted the basic research that made the camera possible. Sensors Unlimited Inc., a company on U.S. 1 in Princeton which markets the camera, used Forrest's work to develop the imaging detector and the camera itself. The first solid-state camera of its kind, it has "changed the way we look at the Boating Party," says Steele.

The infrared camera is just one example of new technologies being born out of poem and its links to the high-tech photonics community. The brainchild of Professor of Electrical Engineering Stuart C. Schwartz, poem was established in 1989 to expedite the transfer of photonic technologies from the lab to the marketplace and to develop collaborative relationships between faculty members, researchers, and students at Princeton and scientists in industry, particularly small and medium-sized firms. Currently, poem has relationships with about 50 companies, mostly in New Jersey, which boasts a high concentration of photonics-related industry.

Photonics deals with the technology of generating and harnessing light (a photon is a packet of light) and other forms of radiant energy. Photonic inventions have given us the clear sounds we hear from compact-disc players and the ability to image internal organs, and in the future they will make computers faster in transmitting information. Research at poem affects a broad range of fields, from telecommunications and entertainment to health care and the environment.

Poem has about 30 associated faculty members from five departments (physics and chemistry, and electrical, chemical, and mechanical and aerospace engineering), about 40 research scientists and staff members, and some 80 graduate students and 30 undergraduates. Several faculty members are also associated with the Princeton Materials Institute [profiled in the January 28 PAW].

AN "ESSENTIAL" RELATIONSHIP

Poem is jointly supported by the New Jersey Commission on Science and Technology, the federal government, the university, and its own industrial partners. According to poem's director, Professor of Electrical Engineering James C. Sturm '79, in 1997 the center generated for the university more than $5 million in sponsored research. Poem also accounted for more than a third of the "invention disclosures" (the first step in a patent application) filed by Princeton, according to its administrator, Joseph X. Montemarano.

In exchange for supporting research at poem, its industrial partners, including the David Sarnoff Research Center, Lucent Technologies, Xerox, and Sensors Unlimited, get "very fresh technology," explains Forrest. The companies gain access to ideas, to graduate students, faculty and staff, and to poem's laboratories. Depending on their level of funding, companies may send employees to poem for stints as visiting researchers.

Poem's industrial partners say the center has allowed their companies to grow in ways that they couldn't otherwise. The relationship is "essential for us," says Vladimir Ban, president of PD-LD Inc. in Montgomery, New Jersey, which develops new products in fiber-optic communications. "We could not do many of the things we do without poem." One product in development which grew out of the collaboraton between poem and PD-LD, says Ban, is a modulator for switching fiberoptic signals at very high speeds, so that more data can be transmitted faster.

Greg Olsen, the president of Sensors Unlimited, says, "It would be hard to imagine how we could get to where we are without the help of poem. They can do the research we can't do because of the expense. And we have the ability to produce and market a product which, as an academic institution, poem can't."

Most of poem's research is interdisciplinary. The technical problems are so complex, says Forrest, that more than one point of view is needed to tackle them. Forrest, an electrical engineer, works with chemists, physicists, and engineers, and "it's the mix that makes it all happen." He and two of his graduate students, Vladimir Bulovic '91 and Zilan Shen *97, have teamed with Mark E. Thompson, a chemist at the University of Southern California, and Paul Burrows, a physicist who works as a research scientist in the electrical-engineering department. Their goal is to develop organic light-emitting devices (OLEDs) that would lead to better display screens of the sort found in thousands of products, including TVs, cellular phones, computers, and wristwatches.

The dominant display technologies in today's market are cathode ray tubes (used in TVs and desktop-computer monitors) and liquid crystal displays (used in laptop computer screens and many handheld computer devices). Forrest's group wants to create displays that are smaller, sharper, brighter, faster, and consume less power. Thanks to OLEDs, in the future you might be able to hang your TV screen on the wall like a painting; turn on a map in the corner of your car's windshield; watch a baseball game or a video on a miniature projection display on your sunglasses; or read an electronic newspaper you could roll up and stick in your pocket.

OLED technology is based on vacuum-deposited, organic materials that emit intense light when electrically stimulated. An OLED is a solid-state thin film 1/1,000th the thickness of a human hair. The researchers have stacked separate red, blue, and green transparent OLEDs atop one another to create a single full-color pixel. (A pixel is a dot of light. Many thousands of pixels make up the image on a computer screen). Conventional displays, including those used in TVs, use individual red, green, and blue pixels in a side-by-side configuration, which has lower resolution than the stacked OLEDs. The devices developed by Forrest's group can be applied to a variety of surfaces, including glass and plastic. They emit light when turned on and are transparent when turned off. They are brighter and more efficient than liquid crystal displays, and the organic materials used to make them are cheap.

The exclusive licensee of OLEDs is Universal Display Corporation (UDC) of Bala Cynwyd, Pennsylvania, a firm founded three years ago to fund and commercialize Princeton's research in OLEDs. Its president, Steven Abramson, is enthusiastic about the poem-UDC collaboration. "You guys have the smartest people in the world," he says. Other technologies poem is working on include an imaging technique that will give physicians clearer pictures of areas inside the human body, devices for switching and routing fiber-optic communications, and organic lasers.

OVERCOMING CULTURAL BARRIERS

Princeton had to adapt to the creation of poem. "There are a lot of cultural barriers at places like Princeton to establishing such links with industry," says Forrest. At Princeton, he adds, there's a long history of intellectual pursuit as an end in itself. He believes the university needs to develop means to use its intellectual resources to benefit New Jersey and the larger engineering community.

Unlike Stanford and MIT, which in the 1950s began to embrace partnerships with companies that could commercialize the discoveries of their professors, Princeton didn't take this route until the 1980s, when federal funding for research leveled off. By then, academic research centers could no longer rely primarily on government funding, explains Stuart Leslie, a historian of technology at Johns Hopkins University. Today, a number of poem-like centers exist around the country, created in part to keep high-tech companies in their respective states.

Educationally, poem benefits students by exposing them to industrial scientists. "The expertise these people bring is not what I would be able to see in a regular graduate-student lab," says Bulovic. Brandon Collings, a graduate student in electrical engineering who has developed lasers for fiberoptic communications, spends several days a week in Lucent Technologies's labs, collaborating with scientists who are "among the best in the world," he says. "Frequently, if you have a problem in a particular area, an expert is right down the hall, or is somebody you eat lunch with." In poem's campus laboratories, the work is hands-on; students carry screwdrivers and get into the nitty-gritty of fixing equipment and preparing it for experiments. "About 75 percent of our time we are glorified car mechanics," says Bulovic, "and 25 percent of the time we do hard thinking, trying to figure out how things work." Some 30 undergraduates are paid assistants in poem labs, an experience that reinforces what they learn in the classroom.

AN INHERENT TENSION

When poem came into being a decade ago, there was some concern on campus that its embrace of industry would compromise the purity of faculty research. Forrest and others recognize the tension inherent in this collaboration. They don't deny that poem's industrial partners want to see the ideas developed in the university's labs lead to commercial products. "The companies draw us a little further down that path," says Forrest, "but from our labs to their products is a big gap." Sturm notes that sometimes faculty members are looking for intellectual problems to attack and find them by talking to their industrial counterparts.

Says Chris Dries, a fourth-year Ph.D. candidate who studies semiconductor lasers and devices used in fiber-optic communications, "What you don't want is industry dictating your graduate studies." He is quick to add that he doesn't see this happening, but that the partnership with industry does help to focus research on practical results. "We have the freedom to explore avant-garde ideas, but industrial collaboration keeps us grounded. When you design a device, you ask yourself, 'Can you actually envision this thing being built?'" Dries, who wants "to build devices that people can use," intends to put his skills to work after completing his doctorate by joining the several dozen other poem alumni hired in recent years by Bell Labs, Xerox, Hughes, and other New Jersey firms.

Sometimes poem is called on to be a problem solver. Emcore, a manufacturer of electronic processing equipment in Somerset, New Jersey, needed help understanding a process that worked on a semiconductor. Within two weeks of addressing the problem, poem had helped the firm achieve the result it was looking for. Consequently, the firm landed a contract with a major auto manufacturer, went public, and more than tripled its staff, from 60 to 200. "These are good-paying jobs that are desperately needed in the state," says Montemarano, "and they would not have been here if it hadn't been for the role played by poem."

Kathryn Federici Greenwood is PAW's staff writer.