Energy-efficient resistive switching synaptic devices based on patterned Ag nanotriangles with tunable gaps fabricated using plasma-assisted nanosphere lithography

A neuromorphic device featuring a metal network structure has demonstrated the ability to mimic both the cognitive functionalities and structural characteristics of neurons. Recently, plasma-assisted nanosphere lithography was employed to fabricate a network of silver (Ag) nanotriangles with tunable nanogaps, tailored for neuromorphic applications. The smallest nanogap achieved was 15 nm, enabling a low threshold switching voltage of 0.86 V at 100 nA and ultra-low energy consumption of 8.4 femtojoules per synapse. The device successfully emulates synaptic behaviors, such as the transition from short-term to long-term plasticity and associative learning. The triangular Ag structures offer sharp edges that guide the formation of filamentary paths between the nanogaps. To predict the filamentary path, COMSOL Multiphysics simulations were performed, revealing strong electric field confinement within the nanogaps. Additionally, post-pulsing scanning electron microscopy (SEM) confirmed the traces of filamentary growth within the nanogaps, further validating the device’s performance.