Eco-Friendly All-Inorganic Perovskite Memristors and Artificial Synapses

Abstract

Artificial Synapse has been an emerging research field as emulating the functions of biological synapse. It is regarded as a basic building block for neuromorphic computing. Organic lead-based perovskites including CH3NH3PbI3 and CH3NH3PbBr3 have been researched as materials for synaptic devices based on Resistive Random Access Memory(RRAM). They have shown gradual increase and decrease of conductivity that represent the synaptic plasticity such as STP (Short Term Potentiation), LTP (Long Term Potentiation), and LTD (Long Term Depression) which imitate learning and forgetting in human brain. However, there have been some environmental concerns about using lead-based materials in large-scale applications because of the possibility of lead contamination. Also, it is still required to increase perovskites’ air-stability. All-inorganic halide perovskites are considered as promising alternatives for RRAM and synaptic devices because of their remarkable stability. Therefore, this study focused on fully inorganic CsPb(1-x)BixI3 perovskite to achieve lower lead content and better stability. Optimized bismuth content was discovered according to their morphology, grain size, and synaptic performances. CsPb(1-x)BixI3 emulates synapse by showing similar functions such as STP and LTP by different current change rate depending on the intensities and frequencies of pulses. Also, STD and LTD can be obtained by applying opposite pulses. Therefore, based on these synaptic functions that optimized CsPb(1-x)BixI3 achieves, it can be applied as a single synapse with low energy consumption for neuromorphic computing which works by parallel operation to connect and store information.