As the technology for making silicon circuitry smaller, faster and less power-thirsty approaches the limits of physics, scientists have tried out many materials in the search for an alternative to silicon. New research by a team at the US Department of Energy’s SLAC National Accelerator Laboratory may have put some other promising candidates into the race.
In a paper just published in the journal Nature Materials, researchers described their successful observation of electrical switching (that is, a forced switch from a non-conductive state to a conductive one) in magnetite, a naturally magnetic iron oxide. The ability to act as a switch that is either “on” (conducting) or “off” (non-conducting) is the basis for a transistor, which is the building block of any electronic circuit. And while magnetite itself isn’t slated to replace silicon, the work opens up the floor for other, similar materials to be studied.
Researchers showed that magnetite’s on-off electrical switch could be flipped in one-trillionth of a second—thousands of times faster than in transistors used currently. In theory, a computer made with magnetite chips instead of silicon would be that much faster than the machines we use today. But since magnetite has to be cooled to a chilly -190 °C (-310 °F) to lock its electrical charges into place, it’s not going to end up in your computer anytime soon.
But what’s cooler than the material itself is the method used to study it. Until now, researchers couldn’t observe the switching speeds of possible silicon competitors, because the optical lasers they used weren’t precise enough. Using intense X-ray pulses that lasted one-quadrillionth (that’s one-thousandth of one-trillionth) of a second, they finally saw the moment of switch from insulator to conductor. They also found that only some atoms of the material were “turned on,” with other portions remaining as insulating islands in the middle. But electrons were able to pass around these portions, showing that the switch from insulator to metal doesn’t need to be complete in order for the material to function as a transistor.
“We understand the process,” Hermann Dürr, the principal investigator of the experiment and senior staff scientist for the Stanford Institute for Materials and Energy Sciences, told Quartz, “so now it’s about optimizing the materials. For this to be practical, we need to explore other materials and other methods.”
Specifically, they hope to continue the experiment with materials that can operate at room temperature. One possibility is vanadium dioxide, which Dürr says the team is now working with. If they can get another material to behave the same way as magnetite did, but at a higher temperature, the next step is to find a way of inducing the change without a laser—hopefully using short, strong electrical pulses, like in a normal transistor.
“All we have to compare this process to is history,” Dürr said. “It took many decades from the first demonstration of a semiconductor transistor to the technological dominance this device has nowadays. And of course this dominance is the problem in finding an alternative. We need to generate a real winner if we want to transcend semiconductors.” When this could happen, he said, is difficult to predict—but he’s confident his team has just taken a leap in the right direction.