LOW THERMAL BUDGET FERROELECTRIC FIELD-EFFECT TRANSISTORS FOR HIGH-DENSITY EMBEDDED MEMORY AND NEUROMORPHIC COMPUTING SYSTEMS

Abstract

The rapid rise of IoT, AI, ML, and neuromorphic computing requires high-density, low-power embedded memory integration. 2D structures fall short, prompting the promising trend of monolithic 3D integration in CMOS back end of line. Challenges arise, particularly in high-density memory integration with limited thermal budget. FeFETs offer an attractive solution with low thermal budget, high-speed, low-power consumption, and non-volatile nature.

This work aims to optimize FeFETs for high-density embedded memory and neuromorphic computing. The study focuses on enhancing properties and fabrication of HZO ferroelectric material for reduced device variation. FeFETs with HZO gate oxide layers are created, assessing their impact on domain structure, switching characteristics, crystalline structure, electrical properties, and thermal stability.