This promises to revolutionize computing in extreme environments like jet engines, fusion reactors, and geothermal wells, offering unprecedented durability and energy efficiency.
Computer memory may soon withstand searing temperatures like those found in fusion reactors, jet engines, geothermal wells, and even sweltering exoplanets, thanks to a breakthrough solid-state memory device engineered by a University of Michigan-led team in collaboration with Sandia National Laboratories. This innovative memory technology operates at over 1,100°F (600°C)—exceeding the surface temperature of Venus and the melting point of lead.
The device’s resilience stems from its unique design, which moves negatively charged oxygen atoms instead of electrons, avoiding the heat-induced current disruptions common in silicon-based semiconductors. Traditional memory fails beyond 300°F (150°C), but the new device’s oxygen ions remain stable and functional, ensuring data retention even in extreme conditions.
The memory relies on tantalum oxide, a semiconductor, and metallic tantalum layers separated by a solid electrolyte. Platinum electrodes guide oxygen ions, enabling the material to alternate between insulating and conducting states to represent digital 0s and 1s. This electrochemical process mirrors a battery’s charge-discharge cycle but stores information instead of energy.
While the device currently stores one bit, researchers, led by Yiyang Li, an assistant professor at the University of Michigan, envision future versions capable of holding megabytes or gigabytes. The team mentions that it could enable electronic devices that didn’t exist for high-temperature applications before. However, new data can only be written above 500°F (250°C), posing a challenge for devices operating in varying temperatures—a limitation researchers suggest could be addressed with a built-in heater.
The memory boasts 24-hour stability at extreme heat and lower voltage requirements compared to alternatives like ferroelectric memory. Additionally, its potential for analog states opens doors to in-memory computing, which could revolutionize power-efficient data processing, especially in extreme environments. In-memory computing chips could help reduce the power demand in extreme settings where AI monitoring is essential.
