Metal-Semiconductor Compound Contacts to Nanoscale Transistors
Semiconductor nanowires (NWs) and Fin structures are promising building blocks for next generation ultrascaled devices for electronic and optoelectronic applications. The detailed understanding of and control over the phase transformation that accompanies the formation of their compound contacts for lithography-free self-aligned gate design can accelerate the development of these ultra-scaled devices. Numerous aspects of nanoscale metallization technology were shown to exhibit significantly different behavior from their bulk counterparts. And up to now, the majority of the studies that explored nanoscale contact metallurgy focused on nanostructures of elemental semiconductors, i.e., Si and Ge NWs, and detailed contact reactions have not been uncovered in III–V NW channels or heterostructured NW channels at atomic resolution.
In the first and the major part of this thesis, I will focus on the narrow band gap, high electron mobility III−V semiconductor, InGaAs, motivated by its potential in sub-10 nm metal-oxide-semiconductor field-effect transistors (MOSFETs). In chapter 2, we reported the first study on the solid-state reaction between Ni and In0.53Ga0.47As nanochannels to reveal the reaction kinetics, formed crystal structure, and interfacial properties. In chapter 3 and 4, we developed a deeper understanding of this contact metallization process by utilizing the in-situ heating transmission electron microscopy (TEM) technique, and observed at atomic resolution the detailed ledge formation and movement behaviors in both the NW cross-section and along the NW channel.
In the second part, I will use the Ge/Si core/shell NW as a model system to talk about the compound contact formation in semiconductor heterostructures. In chapter 5, we managed to control the synchronous core/shell interface during the solid-state reactions between Ni and Ge/Si core/shell nanowires, and measured the strain evolution in ultra-short channels using in-situ TEM. These elevated compressive strains are expected to result in a non-homogeneous energy band structure in Ge/Si core/shell NWs below 10 nm and potentially benefit their transistor performance.
Finally, as appearing in chapter 6, I will introduce the ongoing electrical measurements of contact resistance for InGaAs transistors, and adapt the solid-phase-regrowth method to future reduce the contact resistance with locally introduced dopants. I will also talk about the collaborated work in fabricating the AlGaN/GaN Fin MOS-HEMTs for intrinsically linear power amplification devices.