2022/07/21 by Yaxin Ding, Peng Yuan, Jie Yu +10 · 5 citations
Engineering · Neuroscience · #Advanced Memory and Neural Computing #Ferroelectric and Negative Capacitance Devices #Neuroscience and Neural Engineering
paper · doi:10.1109/ted.2022.3191988
Two-terminal volatile NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> -based threshold switching memristor devices with electrical self-oscillation behavior have attracted tremendous interest for applications in both oscillators and neural networks. However, the forming process of NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> devices can be a burden and limit their application. In this study, a forming-free NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> -based threshold switching device is obtained by reducing the oxygen content of the NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> film, which simplifies the circuit design of the NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> device array and improves the device-to-device (D2D) uniformity due to the removal of the forming step. Furthermore, a low-threshold switching voltage is achieved in the forming-free device, indicating a low-power consumption operation. Finally, we demonstrate a low-energy-consumption artificial spiking afferent nerve based on a forming-free NbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><i>x</i></sub> device for spiking neural network (SNN) applications.