An Ag/HfO2/Pt Threshold Switching Device with an Ultra-Low Leakage (< 10 fA), High On/OffRatio (> 1011), and Low Threshold Voltage (< 0.2 V) for Energy-Efficient Neuromorphic Computing

Threshold switching (TS) devices are of great interest for low power artificial neuron applications. In this work, we present an Ag/HfO2/Pt based TS device with an ultra-low leakage current (1011), fast turn-on and turn-off speeds (<; 300 ns and <; 250 ns, respectively), low threshold voltage (<; 0.2 V), and a stable TS performance. Furthermore, we demonstrate the integrate-and-fire (I&F) neuron functionality of our device experimentally. The inherent stochastic behavior and simple structure make TS devices an ideal candidate for replacing the complex and large-size CMOS (complementary metal-oxide-semiconductor) based neurons. This paves the way towards building energy- and area-efficient neuromorphic systems.