Optoelectronic Neuromorphic Transistors Using ZnO Quantum Dots and Multilayer Ionic Gate Dielectrics

Neuromorphic computing, inspired by the brain’s efficiency, offers a promising path beyond von Neumann limitations. This work presents a cost-effective synthesis of ZnO quantum dots with an average size of ∼4 nm, used in a low-voltage synaptic transistor featuring a bilayer ionic dielectric of LiInSnO4 and Li4Ti5O12. The device shows excellent performance: carrier mobility of ∼1.1 cm2·V–1·s–1, ON/OFF ratio of 8.6 × 105, near-zero threshold voltage, and a subthreshold swing of ∼96 mV·decade–1. Under UV light (295 nm), it mimics key synaptic behavior such as excitatory postsynaptic current, short-term-plasticity (STP)-to-long-term-plasticity transition, and potentiation–depression. Electrical synapse behavior is also demonstrated, showing STP. Notably, the device achieves ultralow energy consumption: 1.28 nJ/spike (0.25 fJ/μm2) for optical and 0.14 nJ/spike (0.01 fJ/μm2) for electrical synapse, outperforming many solution-processed oxide-based optoelectronic synaptic transistors. Additionally, it exhibits advanced functionalities like learning–relearning and classical conditioning. Integrated into neural network simulations with image preprocessing, the device achieves a high recognition accuracy of 97% for the optical synapse and 98% for the electrical synapse, underscoring its potential in future neuromorphic computing platforms.