Research On The Preparation And Performance Of Metal Oxide Double-layer Synaptic Transistors

With the booming of the fields of big data,artificial intelligence?AI?and others,the demand for computing with low energy consumption and high performance is gradually growing.At present,the development of conventional computing paradigm based on CMOS logic and von Neumann architecture has encountered a bottleneck,which will become less efficient when the amount of information keeps exploding.Human brain has a massively parallel and reconfigurable architecture with a complex network which composed of a large number of neurons and synapses.Such architecture makes more robust,plastic and fault-tolerant learning/memory functions than any current digital computer.The synapse is the critical unit in the brain to achieve the information transfer and process.Hardware implementation of synapse by individual electric device is of great significance for realizing such neuromorphic computation.Among the synaptic bionic devices,the electric-double-layer thin-film transistors based on the ion conducting gate dielectric shows great potential for synapse emulating.The synaptic transistors can effectively achieve ultra-low voltage operation by using the large electric-double-layer capacitance of the electrolyte.In this dissertation,electrospinning and sol-gel methods were used to prepare metal oxide synaptic transistors.The main contents of this dissertation include:1)Electrospun metal oxide nanofibers have been regarded as promising blocks for large-area,low-cost,and one-dimensional?1D?electronic devices.In this work,zinc-tin oxide?ZnSnO?nanofibers were fabricated by electrospinning,and the synaptic transistors based on ZnSnO nanofibers were integrated.To mimic the synaptic behavior of human brain,LiClO₄ dissolved in polyethylene oxide was used as the gate electrolyte in the synaptic transistors.The plasticities,including short term and long term,were mimicked with the help of electrolyte ion dynamics under low and large bias voltage,and an energy consumption of as low as 0.44 pJ per event is observed.This work demonstrates a new approach for establishing the large-scale,energy-efficient 1D artificial synapses for neuromorphic networks.2)The indium oxide synaptic transistors based on polyimide substrates were fabricated by a nontoxic water inducement method at low temperature,and lithium perchlorate?LiClO₄?was dissolved in polyethylene oxide as the gate electrolyte.For water-inducement process,comparable electrical performances of the synaptic transistors can be achieved by prolonging the annealing time rather than high temperature annealing with relatively short time.The effect of the annealing time on the electrical performance of the electrolyte gated transistors annealed at various temperatures was investigated.It is found that the electrolyte gated synaptic transistor on polyimide substrate annealed at 200 ^oC exhibits high electrical performance and excellent mechanical stability.Due to the ion migration relaxation dynamics in the polymer electrolyte,various important synaptic behaviors such as the excitatory postsynaptic current,paired-pulse facilitation,high-pass filtering characteristics and long-term memory performance were successfully mimicked.The electrolyte gated synaptic transistors based on solution-processed In₂O₃ exhibit great potential in neuromorphological applications.