Monolithic 3D Integration of FeFET, Hybrid CMOS Logic and Analog RRAM Array for Energy-Efficient Reconfigurable Computing-In-Memory Architecture
Yiwei Du, Jianshi Tang, Yijun Li, Yue Xi, Bin Gao, He Qian, Huaqiang Wu
Abstract
In this work, we report a monolithically 3D integration of HfZrO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf> (HZO) ferroelectric FET (FeFET), analog computing-in-memory (CIM), hybrid back-end-of-line (BEOL) CMOS on top of standard Si-CMOS technology, namely M3D-FACT. The 1 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">st</sup> layer is Si CMOS circuits for control logic, and the 2 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">nd</sup> layer is an analog resistive random-access memory (RRAM) array for CIM. The 3 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">rd</sup> layer is a reconfigurable datapath (RCD), consisting of FeFETs with InGaZnO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x</inf> (IGZO) channel and hybrid CMOS logic based on carbon nanotube (CNT) PMOS and IGZO NMOS. The structure and functions of each layer were verified. Furthermore, a reconfigurable CIM architecture was implemented using the M3D-FACT chip, and the system-level benchmark against its 2D counterpart shows higher energy efficiency in three different network models (6.9$\times $ for VGG-8, 19.2$\times$ for DenseNet-121, and 9.9$\times$ for ResNet-18).