Memresistive crossbar based memory devices could be more efficient than traditional hardware for deep learning applications. However, it is important that they not only have high speed, but also have symmetric analog programming, small variability, compatible with CMOS processing, and have small nanometer scale footprints.
Recent research at the University of Illinois Urbana-Champaign by Jinsong Cui, Fufei An, Jiangchao Qian, Yuxuan Wu, Luke L. Sloan, Saran Pidaparthy, Jian-Min Zuo, & Qing Cao, demonstrates an electrochemical synaptic transistor that meets these goals by shuffling protons between a hydrogenated tungsten oxide channel and gate through a zirconium dioxide protonic electrolyte.
These devices were fabricated at the Holonyak Micro and Nanotechnology Laboratory utilizing an STS-Elionix 150 keV electron-beam lithography system.
For more information see the paper in Nature Electronics.
Image: False colored SEM image of 150 nm length, 150 nm width all-inorganic protonic electrochemical random access memory (ECRAM) device. (Courtesy Qing Cao).