Plastic Food?

In 1967, the career advice given to a certain graduate played by Dustin Hoffman was “plastics.” Forty years later, Caltech chemical engineering professor Julia Kornfield (BS ’83, MS ’84) would add “shish kebabs.”

Shish kebabs are beautiful, tiny structures that can form when polymers crystallize during flow. When magnified a million times, they resemble a skewer running through a stack of bell peppers. Inside plastics, they make car body panels stiff and carpet fibers strong, and impart a nice, glossy finish. But they’re not without their problems—they might help you to resist a scratch, but they might also cause a layer to peel off. And that’s why people want to control them.

Kornfield, Yoshinobu Nozue at Sumitomo Chemical Company, and coworkers have upended the conventional wisdom about how shish kebabs form. Shish kebabs occur in polymers known as polyolefins, which make up half of all plastics used—over 100 million tons per year. In addition to being used for car parts, polyolefins are also used to make pipes, wire, cable, carpets, fabrics, disposable syringes, and many other things. Manufacturers can custom-design a polyolefin’s properties by varying its degree of crystallinity and the way the crystals come together. The result can be as hard as steel or as soft as a rubber band.

“The plastics industry can tailor-make molecular distributions, but we don’t know how to manipulate them,” Kornfield explains. “This discovery opens up a whole new neck of the woods that people didn’t know they could explore, and they’ll be able to create combinations of properties you couldn’t get before.”

Much as an inspiring leader can influence the action of thousands, the researchers discovered that some molecules—especially the long ones—can marshal many others to create the shish, which then direct the formation of kebabs. This knowledge will allow for greater control of the creation process itself.

“In other words,” says Kornfield, “you could make things by injection molding that you couldn’t make before, and injection molding is a very cheap, fast process—you can pop a plastic bumper for an automobile out of its mold in a couple of minutes. So you bring down the cost of manufacturing and at the same increase the throughput.”

A paper describing the work appeared in the May 18 issue of Science. The lead author is Shuichi Kimata, a former postdoc in Kornfield’s lab, who played a central role in linking Kornfield’s group at Caltech with Yoshinobu Nozue’s group at Sumitomo and collaborators at the University of Tokyo. —RT