溶液処理されたSnO₂/SnS₂二層膜ベースの堅牢なメモリスタを用いた信頼性の高いニューロモルフィックコンピューティング

The development of scalable, low-power, and high-density resistive memory devices is crucial for next-generation computing architectures, particularly in neuromorphic applications. Here, we report solution-processed SnO₂/SnS₂ bilayer thin films as functional layers for memristors and synaptic devices. The incorporation of the SnO₂ layer enables the formation of oxygen-vacancy conductive filaments that act as virtual electrodes, which effectively guide the nucleation and rupture of sulfur-vacancy filaments in the two-dimensional (2D) SnS₂ layer. This synergistic mechanism significantly enhances resistive switching performance, yielding an ON/OFF ratio exceeding 200, stable endurance over 10⁴ cycles, and robust retention. Beyond conventional memory behavior, the bilayer devices emulate essential synaptic functions, including excitatory postsynaptic current (EPSC), paired-pulse facilitation (PPF), and spike-timing dependent plasticity (STDP), and achieve ∼93% inference accuracy in artificial neural network tasks.