Like anyone who Designing computer chips for life, James Myers is basically a silicone type. “Silicon is great,” he says. Great because it’s natural semiconductor—It is also possible to conduct electricity and act as an insulator, depending on the conditions – and because it can be designed on a small scale. Great because it is the second most common element on Earth, which is currently glued to the soles and is easily produced by heating the sand. These attributes have become the foundation of almost every technology we use today. People like Myers, an engineer at a British semiconductor firm Arm, mostly spend time thinking about how to pack more silicon in less space – an exponential march of thousands transistors by the 1970s chip to billions today. With Moore’s law, we, as Myers says, “swim in silicon.”
In recent years, however, Myers has been looking at other materials, such as plastics, from silicon. That means starting from scratch. A few years ago, his team started designing plastic chips that contained dozens of transistors, then hundreds, and now, as reported in Nature On Wednesday, Tens of thousands. The 32-bit microprocessor contains 18,000 logic inputs – electrical switches obtained by combining transistors – and the basic lobes of the computer brain: processor, memory, controller, inputs and outputs, and so on. As for what he can do? Consider a desktop from the early 1980s.
Why turn back the technological clock? Because modern silicon chips are brittle, inflexible wafer electronics. They crunch under stress. And while silicon is cheap, and getting cheaper, there are some cases where it may never be cheap enough. Consider a computer chip housed in a milk carton, replacing the printed expiration date with a sensor that detects chemical signs of spoilage. Useful? Sorta! But it is worth adding billions of cartons of milk only if the costs are minimal. One application that Arm tests is a chip placed on a chest that monitors a patient for an arrhythmia – inconsistent, sedative heartbeat – and should be discarded after a few hours. For that, they want a computer that is cheap, but, more importantly, one that bends. “You need to move with you and not jump out,” Myers says.
Numerous materials could theoretically meet these needs. The researchers made the transistors from organic materials and designed the substrates – the coating that the transistors go into – from metal foils and even paper. The chip team described by Myers on Wednesday consists of “thin-film transistors” made of metal oxides – a mixture of indium, gallium and zinc – that can be made thinner than their silicone counterparts. The base is polyimide, a type of plastic, not a silicone tile. It’s cheap, slim and flexible – and it’s also a hassle for engineers. Plastic melts at a lower temperature than silicon, which means that some production techniques that involve heat are no longer usable. And thin transistors can contain imperfections, which means that energy does not move around the circuit in the way that chip manufacturers expect. Compared to modern chips, the design also uses much more power. These are the same issues that plagued chipmakers in the 1970s and ’80s, Myers points out. He can now sympathize with older colleagues.
Compared to the billions found in modern 64-bit silicon processors, 18,000 doors doesn’t sound too much, but Myers is proud to talk about them. Of course, the microprocessor doesn’t work much; it just runs some test code he wrote five years ago, which ensures that all the components work. The chip can run the same type of code as one of the usual Silicon-based Arm processors.
That consistency with silicone devices is key, explains Catherine Ramsdale, co-author of the research and senior vice president of technology at PragmatIC, a company that designs and manufactures flexible chips with Arm. Although the materials are new, the idea is to borrow as much as possible from the manufacturing process for silicon chips. This makes it easier to mass-produce chips and reduce costs. Ramsdale says these chips could cost about a tenth as much as similar silicon chips due to cheap plastic and reduced equipment needs. It’s, yes, a “pragmatic” way things can be handled, she says.