Choi Hyejung, Jung Heesoo, Lee Joonmyoung, Yoon Jaesik, Park Jubong, Seong Dong-jun, Lee Wootae, Hasan Musarrat, Jung Gun-Young, Hwang Hyunsang
Department of Material Science and Engineering, Gwangju Institute of Science and Technology, Gwangju, Korea.
Nanotechnology. 2009 Aug 26;20(34):345201. doi: 10.1088/0957-4484/20/34/345201. Epub 2009 Aug 4.
This paper describes the resistive switching of a cross-point cell array device, with a junction area of 100 nm x 100 nm, fabricated using ultraviolet nanoimprinting. A GdO(x) and Cu-doped MoO(x) stack with platinum top and bottom electrodes served as the resistive switching layer, which shows analog memory characteristics with a resistance ratio greater than 10. To demonstrate a neural network circuit, we operated the cell array device as an electrically modifiable synapse array circuit and carried out a weighted sum operation. This demonstration of cross-point arrays, based on resistive switching memory, opens the way for feasible ultra-high density synapse circuits for future large-scale neural network systems.
本文描述了一种采用紫外纳米压印技术制造的交叉点单元阵列器件的电阻开关特性，该器件的结面积为100 nm×100 nm。由顶部和底部铂电极以及GdO(x)和铜掺杂的MoO(x)堆叠构成的结构用作电阻开关层，其呈现出电阻比大于10的模拟存储特性。为了演示神经网络电路，我们将该单元阵列器件用作电可修改突触阵列电路并进行了加权求和运算。基于电阻开关存储器的交叉点阵列的这一演示，为未来大规模神经网络系统中可行的超高密度突触电路开辟了道路。
