Abstract

Artificial intelligence (AI) is attractive due to its brain-like working pattern and efficient task processing. Light-stimulated synaptic devices show enormous potential in the field of neuromorphic computing. Here, we develop a kind of optical synaptic devices with a ternary photoactive material system consisting of organometal halide perovskite CH3NH3-PbBr3, organic dye Rhodamine B (RhB), and organic semiconductor pentacene. We found that the introduction of RhB can significantly improve the light absorption and photo-responsivity of the composite devices. The devices exhibit acceptable synaptic behaviors and realize a brain-like learning pattern toward various input signals. Notably, the devices can respond to light signals with a low intensity of 1.1 µW cm−2 and be operated at low working voltages. The devices achieve a low energy consumption of 1.25 fJ at a drain-source voltage of −50 µV, comparable to that of a biological synaptic event. Moreover, the devices simulate a self-learning process, indicating their significantly improved learning efficiency. Also, the devices can recognize subtle differences in Morse codes, showing potential in the practical application of information encryption. The excellent optoelectronic performance demonstrates that the proposed ternary synaptic devices are advantageous in developing new AI.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call