Lead-free AgBiI4 perovskite artificial synapses for a tactile sensory neuron system with information preprocessing function

Abstract

Organometal halide perovskites (OHPs) exhibit great potential in memristors and artificial synaptic devices. However, the toxicity of lead and instability set barriers for further applications in neuromorphic computing. Herein, we demonstrate an artificial synapse based on inorganic lead-free AgBiI4 perovskites, which exhibits typical analog resistive switching (RS) characteristics and synaptic behaviors. Several functions of biological synapses are realized including long-term potentiation (LTP), long-term depression (LTD), paired-pulse facilitation (PPF) and spike-timing-dependent plasticity (STDP). Furthermore, by integrating a tactile sensor array with a AgBiI4 synaptic device, an artificial sensory neuron system is demonstrated. Under pressure stimulations with various intensities, an electrical pre-synaptic signal is transferred to artificial synapses for further information preprocessing, realizing the capability of contrast enhancement and noise reduction. After information preprocessing with an artificial sensory neuron system, the recognition accuracy reaches up to almost 100% with only 500 training epochs by using the MNIST handwritten data set. The presented work provides a new candidate for the next generation of neuromorphic computation based on lead free perovskites.