Research On C8-BTBT/InGaAs Heterojunction Photosynaptic Transistors

Brain-like synaptic devices have great potential to break through the Von Neumann bottleneck and continue Moore’s Law and exhibit excellent computing performance,biological adaptability,and low energy consumption.Brain-like synaptic devices are internal conductive processes(called "synaptic weights")capable of mimicking biological synapses to "update" and "memory" in response to changes in external electrical or optical stimuli.Current research is mainly focused on developing reliable artificial synaptic devices to mimic biological synaptic functions.In this paper,an artificial photosynaptic transistor based on 2,7-dioctyl[1]benzothieno[3,2-b]benzothiophene(C8-BTBT)/In Ga As nanowire heterojunction is proposed,which could mimic biological synapse functions.The main research contents of the paper are as follows:(1)The effect of the performance of the bottom-gate top-contact C8-BTBT thin-film transistors was studied,which the insulating layer modified by organic polymer materials including polymethyl methacrylate(PMMA),polyvinylpyrrolidone(PVP)and polystyrene(PS),the electrical properties and photoresponse were explored deeply.Structural features such as surface topography were analyzed.It was found that the hysteresis of the PVP-modified devices decreased and the threshold voltage shifted after multiple tests,which were due to the reorientation of the hydroxyl dipoles caused by the bias voltage and the built-in electric field induced by the trapping of carriers in the PVP-modified layer.(2)Based on PS interface modification,a multifunctional photosynaptic transistor with C8-BTBT/In Ga As nanowires type I heterojunction as the semiconductor active layer was constructed.Because of the energy levels of C8-BTBT completely cover the conduction and valence bands of indium gallium arsenide nanowires,it is indicated that the arrangement is favorable for hole transfer from C8-BTBT to indium gallium arsenide nanowires under the 365 nm laser,and electrons are trapped due to the matching energy levels of the two.In the C8-BTBT layer,the photogenerated carriers are more difficult to recombine.Based on the persistent photoconductive effect caused by charge trapping,biological synaptic behaviors,including excitatory postsynaptic currents,short-term plasticity,long-term plasticity and Pavlovian learning,were simulated successfully;and then a C8-BTBT/indium gallium arsenide nanowires synaptic transistor neural network was constructed,achieving high classification accuracy about 89.72% of the neural network.The research results provide a new idea for building high-performance optoelectronic synaptic device and neural network based on organic-inorganic heterojunction.