Title:  Ferroelectric Non-volatile Memories and Their Applications for In-memory Computing

Committee: 

Dr. Yu, Advisor   

Dr. Shaolan Li, Chair

Dr. Khan

Abstract: The objective of the proposed research is to use ferroelectric capacitive crossbar arrays and ferroelectric embedded memories for in-memory computing. In-memory computing within resistive crossbar arrays has been intensively studied to accelerate deep learning algorithms. However, resistive crossbar arrays suffer from high static current/power, serious IR drop, and sneak paths. To overcome these challenges, “capacitive” crossbar array that relies on transient current and charge transfer is becoming attractive since it (1) consumes only dynamic power, (2) has no DC sneak paths, and avoids severe IR drop along wires, thus is selector-free, (3) can be fabricated on top of the CMOS circuits for potential 3D stacking. In this document, ferroelectric capacitive crossbar arrays are investigated from device physics, device/array measurement, circuit simulation, to system-level benchmarking. Furthermore, ferroelectric non-volatile memories as embedded storage units will also be investigated. More specifically, 1T1F FeRAM is investigated for its performance from 4K to 400K; 2T1F FeRAM is investigated for its ultra-compact and low-power characteristics compared to 1T1F FeRAM; Ferroelectric non-volatile SRAM is investigated due to its zero-leakage advantages for low-active-rate edge devices.