Electronic displays can now be made on flexible materials, and they’re appearing in limited applications. But the high-speed processing power to run them still requires expensive – and rigid – silicon wafers. If all the components could be built onto the same flexible surface, though, it could save money, improve reliability, and perhaps allow for radical new designs.
Researchers have built working circuits now on plastic that are fast enough to make this integration possible. At Sarnoff Corporation in Princeton, NJ, and Columbia University, researchers have succeeded in operating circuits at 100 megahertz – as much as a hundred times faster than previous ones on plastic.
“To my knowledge, 100 megahertz is the fastest anyone has ever had any circuit working from transistors made directly on plastic,” says Michael Kane, a solid-state devices researcher at Sarnoff, who reported the results last week at the Institute of Electrical and Electronics Engineers meeting in Washington, DC.
“It is an impressive piece of work,” says Sigurd Wagner, electrical engineering professor at Princeton University, who also does research with flexible electronics, a field that’s focused in part on creating inexpensive ways to build large electronic devices – “much bigger than you could ever do on wafer-based silicon.”
The Sarnoff/Columbia advance could lead to displays measuring three meters or more diagonally that can also be rolled up and easily transported. One possible market: the Pentagon, which is interested in such a device for use in field command centers.
Fast transistors on plastic could also lead to portable phased-array antennas. Such antennas direct a transmission at a precise target, which saves power and makes communications harder to intercept. Today’s phased-array antennas cost $100,000 and take up at least one square meter, meaning they have to be mounted to a vehicle, according to Kane.
An antenna using Sarnoff’s new technology could cost a few thousand dollars and fit inside a backpack, to be unrolled on the ground when needed. What’s more, a soldier carrying such an antenna “could travel more lightly, because he or she won’t have to take as much power with him,” says Kane.
Currently, large-area displays and some flexible displays depend on a disorderly form of silicon, “amorphous” silicon, that can be fabricated at temperatures low enough to work with plastic. The researchers at Sarnoff and Columbia built their prototype by finding a way to crystallize amorphous silicon that is already deposited on plastic.
The core of the technology is a new laser-based process, developed by James Im, a materials science professor at Columbia University, that heats one narrow band of amorphous silicon at a time. This process makes well-aligned crystals that let electrons move quickly, allowing for the higher processing speeds. Researchers at Sarnoff helped adapt this process for use with a plastic substrate. For example, they introduced special barriers that spread out the heat from the laser, preventing the plastic from deforming.
The group has demonstrated the ability to form working circuits, a problem in previous attempts at high-performance transistors on plastic because the methods were prone to flaws. The next challenges is scaling up to larger circuits, which Kane says should be possible.
Sarnoff is already looking for ways to market the technology. One company is interested in using the circuits in sheets of tiny ultrasound detectors used to monitor the structural integrity of storage drums and other objects.
The devices could eventually be much more powerful, says Tayo Akinwande, an electrical engineering professor at MIT and co-chair of the session where the work was presented. The flexible circuits use the same CMOS technology as in today’s computers, except that the transistors are much bigger, akin to the size of state-of-the-art transistors in the 1970s. Akinwande says that just as computers have become dramatically faster as the size of transistors shrank, so processors on plastic will get faster as well.
Beyond displays and antennas, the technology might someday be incorporated into clothing, or speakers with microdevices that “shape sound,” Akinwande says. The most interesting applications, he says, may be completely different from anything we have now. “You’re not going to use it to try to make a microprocessor,” he says. “If you did that, you’d lose your shirt. But you could try to do some things that regular silicon cannot do right now. And that is left to our imagination.”
But the display applications alone may be enough to drive near-term research in high-performance flexible processing. “This display field is so huge now,” says Sigmund Wagner. “The business volume is about a third of all integrated circuits – that will push more and more R&D in that area.”
How AI is reinventing what computers are
Three key ways artificial intelligence is changing what it means to compute.
These weird virtual creatures evolve their bodies to solve problems
They show how intelligence and body plans are closely linked—and could unlock AI for robots.
A horrifying new AI app swaps women into porn videos with a click
Deepfake researchers have long feared the day this would arrive.
Surgeons have successfully tested a pig’s kidney in a human patient
The test, in a brain-dead patient, was very short but represents a milestone in the long quest to use animal organs in human transplants.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.