Ultra-Low Power Robust 3bit/cell Hf 0.5 Zr 0.5 O 2 Ferroelectric FinFET with High Endurance for Advanced Computing-In-Memory Technology
Sourav De, Darsen D. Lu, Hoang‐Hiep Le, Soumen Mazumder, Yao‐Jen Lee, Wei-Chih Tseng, Bo-Han Qiu, Md. Aftab Baig, Po-Jung Sung, Chung-Jun Su, Chien-Ting Wu, Wen-Fa Wu, Wen-Kuan Yeh, Yeong‐Her Wang
Abstract
Scaled ferroelectric FinFET devices were fabricated with post fin formation surface engineering (SE) to remove the line-edge roughness (LER) from the silicon surface by dry etching. This facilitated 3bit/cell operations in 10 nm Hf0.5Zr0.5O2 based ferroelectric FinFETs along with on-state current (I ON ) to off-state current (I OFF ) ratio of 106, extrapolated 10-year retention and endurance above 1011 cycles. Further, we have evaluated its performance in all ferroelectric neural network, where ferroelectric FinFETs are used as synaptic devices or neurons for weight storage. Synaptic core built with optimized devices achieve software-comparable 97.91% inference accuracy on MNIST data and multi-layer perceptron network.