Ferroelectric Memristors - Materials, Interfaces and Applications

Sammanfattning: The backbone of modern computing systems rely on two key things: logic and memory, and while computing power hasseen tremendous advancements through scaling of the fundamental building block – the transistor, memory access hasn’tevolved as rapidly, leading to significant memory-bound systems. Additionally, the rapid evolution of machine learningand deep neural network (DNN) applications, has exposed the fundamental limitations of the traditional von Neumanncomputing architecture, due to its heavy reliance on memory access. The physical separation between the computing unitand the memory in von Neumann architectures is limiting performance and energy efficiency. A promising solution toaddress these challenges is the development of emerging non-volatile memory technologies that provide significant scalingand integration possibilities, fast switching speeds, and highly energy-efficient operations. Additionally, by integrating“memory resistors” (memristors) in large crossbar arrays, the computation can take place in-memory which can resolve thebottleneck in traditional von Neumann architectures.This thesis investigates the implementation of ferroelectric HfO2 in ferroelectric tunnel junctions (FTJs) and ferroelectricfield effect transistors (FeFETs) as potential candidates for emerging non-volatile memories and memristors.Initially, the thesis focuses on the integration of ferroelectric HfO2 onto the high mobility III-V semiconductor InAs forthe fabrication of metal-oxide-semiconductor (MOS) capacitors. Moreover, optimization of the processing conditions on thecritical interface between the semiconductor and high-k oxide is extensively studied using both electrical characterization andsynchrotron radiation techniques. After optimization of the annealing treatment and top electrode texturing, the fabricationof vertical InAs nanowire FeFETs is successfully implemented. The FeFET shows encouraging initial results with limitationssolvable by further process engineering.The fabrication of metal-insulator-metal (MIM) capacitors with a tungsten (W) top electrode enables ferroelectricity inHfxZr1?xO2 films down to 3.2 nm thickness. However, achieving ferroelectric properties in ultra-thin films requires anannealing temperature above the thermal budget for back-end-of-line (BEOL) integration. To combat this, nanosecond laserannealing (NLA) is introduced, where an ultrafast laser pulse confines the annealing both spatially and depth-wise. UsingNLA, we crystallize 3.6 nm-thick HfxZr1?xO2 films while still being BEOL compatible.The ability to fabricate thin ferroelectric HfO2 films opens up for the fabrication of FTJs, however, being constrained to aW top electrode is severely limiting the device design. By introducing the concept of a crystallization electrode (CE) and ametal replacement process, tuning of the FTJ device characteristics is achieved. We also highlight the impact of the postmetallizationannealing (PMA) temperature on the tunneling electroresistance ratio (TER) of the FTJ. Despite giving similarferroelectric properties, the PMA temperature strongly affects the interface quality which is key for FTJ performance.Partial polarization switching is utilized to achieve multi-state conductance levels in the FTJs, demonstrating its memristivecapabilities. The stable state retention and low variability are promising for the realization of in-memory computing usingcrossbar arrays. Finally, the impact of random telegraph noise (RTN) in ultra-scaled FTJs and the scalability of FTJ crossbararrays is assessed. The low conductance of FTJ memristors reduces the IR drop, while the self-rectifying current-voltageproperty relaxes the need for an external selector, results that encourage the realization of FTJ-based in-memory computingaccelerators.