Flexible Photonic Synapses Based on Graphene Quantum Dots for UV light Selective Detection
발표자
()
초록
내용
In the upcoming "Big Data" era, where the numerous types of unstructured data need to be processed, artificial neural networks have been developed to overcome the limitation of conventional von Neumann computing. Especially, artificial photonic synapses, consolidating the sensing and processing in one device, have received keen interest for emulating human visual system. Here, flexible photonic flash memory, selectively detecting and processing UV light stimulus, is designed. As a floating gate, UV light sensitive graphene quantum dots (GQD) blended in poly (4-vinylphenol) (PVP) are utilized. The GQD/PVP based device exhibits superior flexibility, maintaining synaptic performance during 200 bending cycle tests. Furthermore, photonic device successfully mimics synaptic behaviors such as short-term plasticity (STP), long-term plasticity (LTP) and transition of STP to LTP. Based on these synaptic behaviors, learning and forgetting behaviors are implemented with 3×5 pixelated templates.