An organic artificial synaptic memristor for neuromorphic computing

Developing an artificial synaptic device that can simulate the learning and memory abilities of the human brain is a key step toward achieving neuromorphic computing. Although traditional transistors and emerging memristors are considered potential candidates for achieving these functions, their manufacturing often relies on nonrenewable semiconductor materials. Here, we have successfully fabricated a flexible artificial synaptic device with a typical memristive sandwich structure (Ag/PMMA/SLE/Ti) utilizing the cost-effective organic material sodium lignosulfonate (SLE) as the dielectric layer. This device effectively achieves short-term plasticity (STP), spike-number-dependent plasticity (SNDP), and long-term potentiation/depression (LTP/LTD). Furthermore, the conductance of the designed artificial synapse corresponds to synaptic weights, which can be recognized by neuromorphic computation on CYCLE and MNIST datasets (small/large sizes) with an accuracy of 33.8 % and 89.3 %/91.0 %, respectively. Therefore, this artificial synaptic device exhibits the flexibility to serve in various wearable scenarios, including intelligent electronic skin (e-skin). Additionally, the excellent biocompatibility of SLE aligns well with the concept of green electronics.