CAMBRIDGE -- Like many great scientific discoveries, Teresa Baker's breakthrough in MIT's grimy Cryogenic Engineering Laboratory last October was punctuated by a memorable exclamation of victory. She raced upstairs from the first-floor lab and announced to her fellow graduate students: ''I made ice cream, come down and eat it!"
Baker's work involves liquid carbon dioxide, bulky stainless steel cylinders, heat exchangers, and vanilla ice cream mix, and it may change the way ice cream is made in the $20 billion-a-year industry. For consumers, the novel device could popularize a new type of frozen dessert that combines the chill of ice cream with the explosive fizz of soda pop.
''It's not ice cream in the usual sense," said John G. Brisson, an associate professor of mechanical engineering at the Massachusetts Institute of Technology. ''It has a carbonated bite, and it just kind of goes 'whoof' on your tongue."
Already, Brisson and professor Joseph L. Smith Jr., who both collaborated with Baker, have met with officials from several large ice cream companies -- they won't say which ones -- to demonstrate their technique. They hope to commercialize the apparatus, possibly with a partner. Their product does not yet have a name.
Jerry Dryer, president of Dairy & Food Market Analyst, a consulting firm in Delray Beach, Fla., said the dessert produced at MIT could take a bite out of the 21.5 quarts of ice cream consumed annually by the average American.
''This product is exciting," he said. ''The effervescence is unique. People are always looking for a new flavor, a new texture, a new mouth sensation. I'd think it would have pretty good commercial potential."
Brisson and Smith conceived their confection contraption several years ago after a dairy industry executive told them, ''We'd love to see a better way to make ice cream."
''We got to talking about it, and we thought, 'This just might work,' " said Smith, who has headed the cryogenics lab since 1964. Their invention stands alongside other MIT breakthroughs that include the development of microwave radar, the first living cell with man-made genes, the first atom laser, and the hologram that appears on many credit cards.
The two won $50,000 in funding from the Deshpande Center for Technological Innovation, an MIT program that funds early-stage emerging technologies, and began searching for a graduate student to lead the effort. That was no easy task, since the cryogenics program usually attracts fewer than five graduate students each semester. But along came Baker, a 24-year-old from Burbank, Calif., who had spent a year working at NASA's Jet Propulsion Laboratory in Pasadena, Calif. She was seeking a project involving thermodynamics, the science of converting heat into other forms of energy, which is exactly what happens when liquid ice cream mix solidifies.
The cryogenics laboratory is just steps from MIT's new Stata Center, the showplace designed by Frank Gehry that looks more like modern art than a building. The lab evokes an earlier age of engineering. Located in a drab 1923 building that originally was a laundry, it is crowded with lathes, drill presses, spouted oil cans, and ominous circuit breakers. Light filters through ancient Venetian blinds. In a back room, a mechanical helium liquefier chugs away, generating the syncopated puffing of a steam locomotive.
Nearby, the ice cream maker hangs from a platform. Stainless steel tubing snakes from two flasks; one contains liquid carbon dioxide, the other is filled with ice cream mix, a concoction of milk, cream, sugar, nonfat milk solids, and chemicals. The tubes pass through a picnic cooler filled with ice water, then through several valves and gauges before emptying into a solid steel cylinder held together with giant bolts. The nozzle controlling the flow of ice cream mix is the same as those that regulate the flow of heating oil into a furnace.
As Baker opened valves, high-pressure liquid carbon dioxide rushed through the tubing. It met the ice cream mix in a small steel chamber and the combination sprayed downward into the collecting cylinder with a soft hiss. There, the carbon dioxide expanded to a gas, instantly chilling to minus 40 degrees Fahrenheit. The droplets of ice cream mix solidify into tiny flakes, each about 30 microns across, smaller than the diameter of a fine hair.
After about three minutes, Baker and Brisson grabbed huge wrenches to undo the collecting chamber's bolts. Inside was a soup can filled with fine white powder. Cups and spoons were distributed and the sampling of the vanilla product began.
The powder is dry, extremely cold, and melts instantly in the mouth, releasing carbon dioxide vapor so quickly that the gas travels up the taster's nose, producing an unexpected but pleasant rush.
''It's like fresh fallen snow," said Baker, ''but it definitely gets sharp."
It is not the kind of work normally done in the lab. Smith has experimented with magnetic refrigerators, which can super-chill objects by fluctuating the magnetic fields. Brisson has explored the world of ultra-low temperatures, between absolute zero and 1 degree Kelvin, at which superconductivity and other extraordinary phenomena occur.
But ice cream takes the cake.
''Of all the stuff I've done, this is the most fun," said Brisson, who grew up eating the black raspberry ice cream at the old Cranberry Cove ice cream shop in Eastham.
The new method is also a long way from traditional ice cream manufacturing. In dairy factories, liquid ice cream mix flows into a chilled pipe and begins to freeze. Scrapers grab ice crystals and mix them into the liquid. Beaters mix air into the finished product to make it fluffier. The entire process takes about 30 seconds, after which the semi-soft product is ready for mix-ins or packaging.
The method hasn't changed much since the 1920s, said Bob Roberts, associate professor of food science at Pennsylvania State University.
But Roberts said the new method raises questions, including whether the product is actually ice cream. According to the government's Code of Federal Regulations, ''Ice cream is a food produced by freezing, while stirring, a pasteurized mix." MIT's concoction might not pass muster, he said, and could get tagged instead as a ''quiescently frozen confection."
But the cryogenics crew is not worried. Its next step is to build a more sophisticated device that could continuously blast out frozen dessert while enabling the operator to control its effervescence and flake size. Brisson also estimates that the carbon dioxide method could cut the energy used to make frozen desserts by as much as 40 percent. If the technique catches on, he and Smith would share in any royalties.
Smith said the frozen dessert apparatus illustrates why after more than half a century, he still loves working in the run-down lab.
''It's wonderful to do things that are new and different and to understand how they worked," he said. But, Smith will not be overindulging in the frozen treat. ''I'm lactose-intolerant," he said.
Jeffrey Krasner can be reached at firstname.lastname@example.org.