Overview: The objective of this project is innovative research for developing III-nitride fin field-effect transistors on silicon substrate. The approach utilizes silicon support fins with (111)-oriented sidewalls for selective area growth of symmetric III-nitride heterostructures using metallorganic vapor phase epitaxy. Materials growth, nanofabrication, simulation and modeling, and electrical tests will be carried out to exploit the enormous flexibility in engineering the fin energy band structure, leading to unprecedented control of electronic properties. Intellectual merit: This proposal presents transformative ideas, leading to novel and effective electronic device architectures for high-power transistors on a silicon platform. The proposed innovations will address fundamental and applied research related to materials growth, nano- and micro-scale design and development of wide band gap semiconductor transistors, and fabrication principles for fin architectures on structured silicon substrates. The III-nitride/Si fins will exhibit new band structure physics that will be exploited for innovative enhancement-mode devices. Broader impacts: Efficient transistors are needed for high power radio-frequency modulation in wireless communications, DC power conversion, and imaging. Nanoscale biological and chemical sensors are needed for environmental monitoring, medical diagnostics, and homeland security. Normal-off state transistors for chip-scale miniature electronics are important for future ?green energy? applications. This proposal will enhance graduate and undergraduate education opportunities at Texas A&M and Texas Tech Universities and, strengthen diversity at each institution, lead to REU supplement requests, and initiate a program targeting junior and senior high students who are demographically underrepresented in science and engineering.
The objective of this project is innovative research for developing III-nitride fin fieldeffect transistors on silicon substrate. The approach utilizes silicon support fins with (111)- oriented sidewalls for selective area growth of symmetric III-nitride heterostructures using metallorganic vapor phase epitaxy. Materials growth, nanofabrication, simulation and modeling, and electrical tests will be carried out to exploit the enormous flexibility in engineering the fin energy band structure, leading to unprecedented control of electronic properties. We have carried out a systematic series of wet etching experiments for developing silicon nanofins with (111)-oriented sidewalls and (110)-oriented planar regions between the fins. The process is developed using microfins produced by optical lithography and SiO2 hard mask. Smooth sidewalls are obtained by virtue of the directional etching process. As in previous work for developing MEMS structures, doping the etchant with silicon (a sacrificial wafer, in our case) improves the surface roughness of the planar regions. Addition of surfactant is found to help in reducing the presence of shoulders where the (111) and (110) crystallographic surfaces meet. Control over the fin width, down to 30 nm, and high aspect ratio of 8:1 (fin height ~ 250 nm) is readily achieved using EBL and a SiO2 hard mask. The electron affinities and valenceband maxima for aq-HCl GaN, Al2O3 and HfO2 were measured using UV photoemission spectroscopy (UPS). Downward band bending of 0.9 eV on the aq-HCl GaN surface was deduced from UPS measurements. Band bending at the GaN/Al2O3 and GaN/HfO2 was determined using the shifts in the Ga 3d and N 1s core-level peaks measured using X-ray photoemission spectroscopy (XPS). Downward band bending of ~ 0.05 eV and ~ 0.9 eV existed at the GaN/Al2O3 and GaN/HfO2 interfaces. An interface dipole of ~ 0.4 eV for GaN/Al2O3 and ~ 1.6 eV for GaN/HfO2 was calculated. The data suggest that the interfacial oxide formed during the ALD of HfO2 may cause the dipole at the HfO2/GaN interface. The impact of dry etching using F-based plasma of AlGaN/GaN heterostructures was found to be neglgible in the density of interface states. It was also found that the major contributor to interface states in the 2DEG MOSHEMT devices is actually deep traps in the AlGaN, not defects at the dielectric/AlGaN interface. In conclusion, the research work completed for this project resulted in new methodologies for wet formation of nano-scale fins for use in 3D devices, improved band offset understanding of dielectrics in contact with nitride materials and low damage dry etching of nitrides with F-based chemistry. Finally, a new method of understanding defects in MOS-HEMTs was developed and used to attribute defects in the F-based dry etching as inherent to the heterostructure and not induced by plasma damage. These developments contribute to the understanding of how to form III-nitride 3D devices.
