While memcapacitors exhibit scalability, a sneak-path-free nature, and lower power consumption compared to memristors, the sensitivity and responsiveness of conventional memcapacitors are lower than expected due to their physical mechanisms. Quantum capacitance, which is controlled by the Fermi level and the density of states, holds the potential to enhance memcapacitor performance and address these issues. In this study, a unique memcapacitor based on the MoS2 quantum capacitance (MoS2 quantum memcapacitor, MQM) is proposed, and the “learning-forgetting” behavior of simulated synapses is discussed. Initially, the quantum capacitance memory effect of MoS2 was demonstrated. Subsequently, under gate voltage (Vg) control, the MQM exhibited a sensitive response and excellent repeatability. Finally, the MQM demonstrated a range of artificial synaptic behaviors consistent with biological synapses, including transformations from short-term potentiation to long-term potentiation and from short-term depression to long-term depression, as well as high pair-pulse-facilitation behavior. This showcases its excellent versatility and flexibility, thereby promoting the development of an integrated artificial neural network.
原文链接
分享
引用
