Characteristics of Synaptic Function of Mesoporous Silica-Titania and Mesoporous Titania Lateral Electrode Devices

In this paper, we have fabricated non-volatile memory resistive switching (RS) devices and analyzed analog memristive characteristics using lateral electrodes with mesoporous silica-titania (meso-ST) and mesoporous titania (meso-T) layers. For the planar-type device having two parallel electrodes, current-voltage (I-V) curves and pulse-driven current changes could reveal successful long-term potentiation (LTP) along with long-term depression (LTD), respectively, by the RS active mesoporous two layers for 20 similar to 100 mu m length. Through mechanism characterization using chemical analysis, non-filamental memristive behavior unlike the conventional metal electroforming was identified. Additionally, high performance of the synaptic operations could be also accomplished such that a high current of 10 6 Amp level could occur despite a wide electrode spacing and short pulse spike biases under ambient condition with moderate humidity (RH 30 similar to 50%). Moreover, it was confirmed that rectifying characteristics were observed during the I-V measurement, which was a representative feature of dual functionality of selection diode and the analog RS device for both meso-ST and meso-T devices. The memristive and synaptic functions along with the rectification property could facilitate a chance of potential implementation of the meso-ST and meso-T devices to neuromorphic electronics platform.

Automated summary

In this paper, non-volatile memory resistive switching (RS) devices with mesoporous silica-titania (meso-ST) and mesoporous titania (meso-T) layers were fabricated and analyzed for their analog memristive characteristics using lateral electrodes. The experimental results showed successful long-term potentiation (LTP) and long-term depression (LTD) in the RS active mesoporous layers, unlike conventional metal electroforming. The devices also exhibited high performance synaptic operations and rectifying characteristics, which make them a potential candidate for neuromorphic electronics platforms.