With summer nearly upon us, temperatures are beginning to rise in the northern hemisphere and most of us know what happens when our smartphones, and other portable electronics, overheat.
Researchers at the University of Pennsylvania have unveiled an ultra-resistant memory device able to endure extreme temperatures, a development that bodes well not only for smartphones, but for AI devices running in harsh conditions.
A recently published study in the journal Nature Electronics from Deep Jariwala and Roy Olsson of the University of Pennsylvania, along with their engineering teams, demonstrates a memory technology capable of withstanding temperatures of up to 1,100°F. These high tolerance levels were maintained for over 60 hours, displaying extraordinary stability and reliability.
Megabytes only – for now
The team designed a non-volatile device, meaning it could retain information without an active power source. Unlike traditional silicon-based flash drives that start to fail around 392°F, the team’s device used ferroelectric Aluminum Scandium Nitride (AlScN). AlScN has the unique capability of being able to maintain specific electrical states, even at significantly higher temperatures.
The ultra-resistant memory device comprises a metal-insulator-metal configuration with a fine layer of AlScN sandwiched between nickel and platinum electrodes. This unique design was meticulously planned and executed to ensure compatibility with high-temperature silicon carbide logic devices, further enabling the memory device to function alongside high-performance computing systems meant for extreme temperatures.
“Conventional devices using small silicon transistors have a tough time working in high-temperature environments, a limitation that restricts silicon processors, so, instead, silicon carbide is used,” Deep Jariwala said.
“While silicon carbide technology is great, it is nowhere close to the processing power of silicon processors, so advanced processing and data-heavy computing such as AI can’t really be done in high-temperature or any harsh environments. The stability of our memory device could allow integration of memory and processing more closely together, enhancing speed, complexity, and efficiency of computing. We call this ‘memory-enhanced compute’ and are working with other teams to set the stage for AI in new environments.”
Despite the notable progress, the new technology will initially only be available in smaller capacities. Jariwala told us, “Based on the current size of devices and scalability of our fabrication process we can easily attain 10 Megabytes to 100s of Megabytes of storage capacity. Our goal is to commercialize these Megabyte scale chips through our startup company in the near to intermediate future.“
