An optoelectronic synaptic transistor with efficient dual modulation by light illumination

Abstract

Inspired by biological neuromorphic systems, which can simultaneously perceive, remember, and process enormous information through parallel, energy-efficient processes, artificial synaptic transistors have shown great potential in paving a way to overcome the von Neumann bottleneck for neuromorphic computing. Artificial synapses capable of effectively emulating both the excitatory postsynaptic current (EPSC) and inhibitory postsynaptic current (IPSC) responses, and particularly reconfiguration, are crucial for building neuromorphic systems with desirable versatility. However, it is still challenging to realize emblematical neurobehavior in unipolar transistors owing to the unbalanced carrier concentration. Therefore, for the first time, a facile light-adjustable organic photoelectric synaptic transistor based on bulk heterojunction blends is developed. In addition, typical synaptic properties including the excitatory/inhibition postsynaptic current, paired pulse facilitation/inhibition, frequency-dependent characteristics, the transformation from short-term to long-term plasticity are successfully simulated and modulated by light illumination. Moreover, the mechanism of light illumination on the neuroplasticity of our synaptic device is discussed. Notably, the recognition accuracy of our synaptic device can be efficaciously modulated by the light intensity and achieves 86% accuracy with appropriate light exposure conditions. The artificial synaptic devices based on bulk heterojunction open up a whole new path for the urgent need of neuromorphic computation, in which light illumination can be utilized for modifying the circuit learning rate.