Cargando…
Near-Infrared Artificial Optical Synapse Based on the P(VDF-TrFE)-Coated InAs Nanowire Field-Effect Transistor
Optical synapse is the basic component for optical neuromorphic computing and is attracting great attention, mainly due to its great potential in many fields, such as image recognition, artificial intelligence and artificial visual perception systems. However, optical synapse with infrared (IR) resp...
Autores principales: | , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2022
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9698720/ https://www.ncbi.nlm.nih.gov/pubmed/36431733 http://dx.doi.org/10.3390/ma15228247 |
Sumario: | Optical synapse is the basic component for optical neuromorphic computing and is attracting great attention, mainly due to its great potential in many fields, such as image recognition, artificial intelligence and artificial visual perception systems. However, optical synapse with infrared (IR) response has rarely been reported. InAs nanowires (NWs) have a direct narrow bandgap and a large surface to volume ratio, making them a promising material for IR detection. Here, we demonstrate a near-infrared (NIR) (750 to 1550 nm) optical synapse for the first time based on a poly(vinylidene fluoride-trifluoroethylene) (P(VDF-TrFE))-coated InAs NW field-effect transistor (FET). The responsivity of the P(VDF-TrFE)-coated InAs NW FET reaches 839.3 A/W under 750 nm laser illumination, demonstrating the advantage of P(VDF-TrFE) coverage. The P(VDF-TrFE)-coated InAs NW device exhibits optical synaptic behaviors in response to NIR light pulses, including excitatory postsynaptic current (EPSC), paired-pulse facilitation (PPF) and a transformation from short-term plasticity (STP) to long-term plasticity (LTP). The working mechanism is attributed to the polarization effect in the ferroelectric P(VDF-TrFE) layer, which dominates the trapping and de-trapping characteristics of photogenerated holes. These findings have significant implications for the development of artificial neural networks. |
---|