Design of High Performance MXene/Oxide Structure Memristors for Image Recognition Applications

Recent popularity to realize image recognition by memristor-based neural network hardware systems has been witnessed owing to their similarities to neurons and synapses. However, the stochastic formation of conductive filaments inside the oxide memristor devices inevitably makes them face some drawbacks, represented by relatively higher power consumption and severer resistance switching variability. In this work, we design and fabricate the Ag/MXene (Ti₃C₂)/SiO₂/Pt memristor after considering the stronger interactions between Ti₃C₂ and Ag ions, which lead to a Ti₃C₂/SiO₂ structure memristor owning to much lower “SET” voltage and smaller resistance switching fluctuation than pure SiO₂ memristor. Furthermore, the conductances of the Ag/Ti₃C₂/SiO₂/Pt memristor have been modulated by changing the number of the applied programming pulse, and two typical biological behaviors, i.e., long-term potentiation and long-term depression, have been achieved. Finally, device conductances are introduced into an integrated device-to-algorithm framework as synaptic weights, by which the MNIST hand-written digits are recognized with accuracy up to 77.39%.