The Massachusetts Institute of Technology (MIT) team that created a new ferroelectric material in 2021 has used the same substance to build a transistor. This novel ultra-thin transistor is claimed to be superior to those used in today’s electronics. Specifically, the scientists boast of rapid nanosecond switching speeds and remarkable durability.
“In my lab, we primarily do fundamental physics,” says Pablo Jarillo-Herrero, the Cecil and Ida Green Professor of Physics who led the team behind the breakthrough (via Interesting Engineering). “This is one of the first, and perhaps most dramatic, examples of how very basic science has led to something that could have a major impact on applications.”
The most notable ability of this new transistor is the speed at which it can change its charge state. Current transistor technology switch states in the order of hundreds of nanoseconds, but this new material could potentially cut this down to a fraction of that. This is crucial for high-performance computing, especially as AI technologies require more and more data to process.
Since the material is so thin, manufacturers could potentially pack them more densely than current semiconductors. Aside from getting more performance per area, it would also lead to higher energy efficiency — a crucial factor in the future of AI processing, especially as power limitations are now serving as the primary bottleneck in expanding data centers.
Another important advancement the MIT team discovered is the increased durability offered by the new ferroelectric material. Current SSDs have a limited lifespan, with the top-of-the-line models capable of writing 700TB for every 1TB capacity. On the other hand, this transistor showed no signs of degradation even after 100 billion switches, potentially giving birth to archival flash storage.
At the moment, the team has made just a single transistor to demonstrate its capabilities. Thus it still faces several challenges before this technology makes its way to everyday devices. “There are a few problems. But if you solve them, this material fits in so many ways into potential future electronics,” says Ray Ashoori, a member of the group who built the transistor. “It’s very exciting.
Cornell University assistant professor Kenji Yasuda, who is the co-first author of the study, also added, “If people could grow these materials on the wafer scale, we could create many, many more.”
