Oxide-Based Resistive Memory and Neuromorphic Computing Devices
This community develops thin-film memory devices and artificial synapses that use resistive switching and ferroelectric effects to perform data storage and brain-inspired computation.
The work centers on hafnium oxide and other metal oxides, utilizing atomic layer deposition to create thin films with controllable oxygen vacancies. These materials form the basis for memristors and resistive random-access memory (RRAM) devices. A significant portion of the research applies these physical mechanisms to neuromorphic computing, implementing artificial synapses, spiking neural networks, and reservoir computing architectures. The devices are designed for in-memory computing, where data processing occurs within the memory array itself, reducing the need for separate processing units.
The largest share of the community's output is found in hafnium research, accounting for 31.0% of all hafnium research, and 8,709 papers here. Tantalum and tin also contribute significantly, representing 7.1% and 2.1% of their respective element research, with 2,700 and 3,279 papers in this group.
The community comprises 32,769 papers, publishing most frequently in Applied Physics Letters, Journal of Applied Physics, and Japanese Journal of Applied Physics.
Recent work focuses on integrating these memory devices into flexible and wearable systems for near-sensor computing, as well as developing ferroelectric transistors for low-power memory applications.