Publication date: 9th January 2023
Inspired by the human brain, memristors have gained tremendous attention but there are still some challenges like low-power switching, robustness, and well-stabled devices for future growing computing. In the advancement of artificial intelligence (AI) and machine learning (ML), brain-inspired technology also known as neuromorphic computing play an important role. In this paper, we present Ruddlesden propper perovskite-based memristor devices in stable ambient air with mechanical flexibility without encapsulation shows ∼ 103 ON/OFF ratio. For further stability, the devices are wear resistance for 2500 cycles with a retention capacity of 600 seconds on bending at 5mm radii. The formation of conductance filament is based on charge transfer, here the perovskite layer and electrode interfacial layer play an important role. To explore and simulate synaptic functions, the device's conductance modulation behavior was used, which is akin to neural-inspired spike time-dependent plasticity (STDP) weight updating in real synapses. Pulsed potentiation-depression along with Long-term potentiation as well as Long-term depression (LTP/LTD) was also performed.