F. Zhao
Washington State University,
United States
Keywords: resistive random access memory, nonvolatile memory, natural organic, honey, neuromorphic computing, environment, energy efficiency
Summary:
As computing technology continues to advance rapidly, solving the environmental concerns due to electronic-waste and tremendous power consumption due to low energy efficiency of conventional computing architecture is becoming a critical priority. One potential solution to these challenges is “brain-like” neuromorphic computing with energy-efficient operation, eco-friendly disposals, and renewable source materials. Such neuromorphic computing requires hardware components that can mimic the functions of biological neurons and synapses – the building blocks of the brain. Among the memory technologies, resistive random access memory (ReRAM) is deemed as one of the competitive emerging non-volatile memory technologies for neuromorphic computing systems. ReRAM made of natural organic materials such as protein and carbohydrate has emerged for renewable nonvolatile memories due to their biocompatibility, environmental-friendliness, and biodegradability. Honey, a mixture of mono- and poly-saccharide based natural organic materials, has been formulated and processed into a resistive switching thin film in ReRAM devices. In this paper, we report the progress of honey-ReRAM development, including device design and fabrication, non-volatile memory properties such as bipolar resistive switching, retention, endurance cycles, long-term potentiation and depression, etc., and synaptic behaviors such as short-term and long-term memory, neural facilitation, spike-rate-dependent plasticity, etc. These characteristics with environmental benefits make honey a promising material for developing environmentally-benign electronic devices in neuromorphic computing. This work is supported by National Science Foundation (NSF) [Grant number 2104976].