Analytical Microscopy Applications to Wide-Bandgap Semiconductors and Nanocarbon-Metal Composite Materials

Thumbnail Image

Publication or External Link





Understanding the atomic structure of materials lies at the heart of materials science. Electron microscopy offers myriad techniques to both probe processing-structure-property relationships in materials, and to manipulate those relationships directly. In this thesis, analytical transmission electron microscopy (TEM) was used to investigate two distinct material systems with applications to energy-efficient technologies: wide-bandgap semiconductors and nanocarbon-metal composites.

In the first project, TEM and electron energy loss spectroscopy were used to investigate the structure, composition and bonding of metal-oxide-semiconductor devices based on silicon carbide (SiC) and gallium oxide (Ga2O3). The performance of SiC falls short of ideal due to electrically active interfacial defect states. This work confirms that boron doping at the SiC/SiO2 interface is feasible and improves the device channel mobility likely through a stress-relaxation mechanism. Separately, no adverse structural effects were found after antimony ion implantation into the SiC substrate, which independently raises mobility via a counter-doping mechanism.

Few atomic-scale studies on Ga2O3 have been reported to date; this thesis aims to bridge the knowledge gap by investigating gate oxide materials and process conditions from a structural perspective. Elevated annealing temperatures reduced interface quality for both SiO2 and Al2O3 gate oxides. Separately, amorphous Al2O3 layers were crystallized under moderate electron irradiation in TEM. One-fourth the dose was required for crystallization with 100-keV electrons compared to 200 keV, indicating an ionization-induced atomic rearrangement mechanism. This unexpected phenomenon will have implications for devices operating in extreme environments.

The second project investigated structure-property relationships in novel nano-carbon metal-matrix composites called covetics, which exploit the superior mechanical and electrical properties of carbon nanostructures such as graphene. Aluminum covetics were characterized using TEM and various spectroscopy techniques; complementary quantum-mechanical and effective-medium models were used to predict the performance of covetics with a range of structures. The models suggest that an electrical conductivity enhancement of ≈10% is feasible with a 5 vol.% carbon loading, but oxides and poor Al/C contact often diminish the performance of real covetics.