Synthesis and fabrication of wide bandgap semiconductors for high-power electronics

Abstract

Due to an ever-increasing demand for electricity, it is essential to invest in high-efficiency energy infrastructure. Wide bandgap materials (WBGs) are essential to the development of energy production, storage, transportation, and usage. WBGs (with bandgaps exceeding ~2 eV) typically have high breakdown fields, low voltage leakage in high electric fields, high maximum operating temperatures, ability to interact with high energy light, and chemical stability all while being able to maintain high levels of efficiency relative to their moderate bandgap counterparts, leading to significant energy and cost savings. This dissertation explores the development of WBG semiconductors for two applications: Power conversion and storage in beta-voltaic cells and efficient power transportation through diamond photocathodes. Boron suboxide (B6O) is a wide bandgap semiconductor with a direct bandgap between 1.8 – 3.0 eV and has garnered interest due to its ability to heal from radiation damage, its high neutron capture cross-section, chemical inertness, high-temperature stability, and ultra-hardness of 40 – 67 GPa, making the material ideal for radioisotope batteries. This work employs Density Functional Theory (DFT) to assess these material properties and shows a low diffusion barrier of 0.16 eV for boron diffusion through the lattice, allowing self-healing from Frenkel defects caused by radiation damage. The metal flux method was selected for crystal growth due to its ability to precipitate large-domain, high-purity crystals at atmospheric pressure without a substrate, eliminating crystal strain due to lattice mismatch. DFT results also predict that the synthesis of B6O from a mixture of B and CuO is thermodynamically favorable in oxygen-deficient environments. Unfortunately, this was not experimentally verified. Analysis of the crystals precipitated from a molten copper flux revealed only mixtures of β-boron and B2O3, indicating kinetic limitations during crystal growth. The difficulties in B6O synthesis caused the study to pivot to C2Al3B48 growth, which is a wide bandgap semiconductor expected to have similar properties to B6O. C2Al3B48 crystals up to 7 mm in diameter were grown from an aluminum-boron-carbon flux by varying the carbon concentration. The crystals were characterized using XRD, Raman, and PL. C2Al3B48 demonstrated a broad band edge between 2.4 – 2.7 eV, which is consistent with its yellow color. The second project involves vacuum electronics, which offer significant advantages over solid-state devices in high-power, high-frequency applications, but often require bulky cooling systems that prevent miniaturization. Diamond is well-suited for high power electronics due to its extreme hardness, ultrawide bandgap, high carrier mobility, and high electrical breakdown field and for photocathodes in particular due to its negative electron affinity (NEA) when terminated with hydrogen. Silicon carbide (SiC) was initially used to demonstrate photocathode enhancement, which utilized self-assembled Au nanodots to fabricate nanopillars on the cathode substrate using reactive ion etching. SiC devices demonstrated a field enhancement of 205 and a turn-on field of ~6.2 V/μm when exciting electrons with a 447 nm laser. The self-assembled Au nanodot fabrication process was transferred to diamond, and this work produced tip widths ranging from 40-460 nm, densities ranging from 0.5-53.5 pillars/μm2, and heights greater than 4.5 μm. The nanodot diameter and density can be tuned using the initial Au film thickness. After fabrication, the pillars were sharpened to tip widths less than 40 nm by annealing in air at 650°C. Finally, diamond was terminated with hydrogen by annealing in forming gas, and the sheet resistance of the diamond films was measured using CTLM. This work was partially performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344, LLNL-TH-2004327.