Organic flexible memristors have emerged as promising candidates for neuromorphic computing due to their wearability, simple fabrication, and low energy consumption. To efficiently handle complex tasks such as pattern recognition and neural network training, artificial synapse devices with large conductance variation ranges and good linearity are still required. Here, we reported a 1,4-diphenylbutyldiyne (DPDA) small-molecule flexible organic memristor, fabricated via spin-coating. The device exhibits femtojoule-level energy consumption and can implement various synaptic functions, including paired-pulse facilitation (PPF) and the fast Bienenstock–Cooper–Munro (BCM) learning rule. The device maintained high classification accuracy of 91.5% even after 1000 bending cycles. These findings highlight the potential of DPDA-based flexible memristors for wearable artificial neural networks and next-generation neuromorphic computing systems.
Cui et al. (Tue,) studied this question.