When it comes to an electric vehicles (EVs), the first material that comes to mind may be a lithium-ion battery. What about a photovoltaic (PV) system? Probably, silicon (Si), a material used for solar cells. These materials play a role in converting useless forms of energy into useful ones. In the field of eco-friendly energy, there is one thing we all know but does not rise to the surface the efficient use of eco-friendly energy. "Efficient use" herein does not refer to the energy efficiency of home appliances but the efficiency of conversion, transmission, and storage or the entire process, from electricity generation to service supply at home. For example, driving a motor with electrical energy from a lithium-ion battery or supplying electricity from the photovoltaic devices requires converting DC electricity into AC electricity or vice versa using a power device like an inverter, resulting in energy loss in the process. With an increasing demand for higher power-density devices for electrical vehicles, the size and weight of a power device are also becoming an issue. According to a recent report, electrical energy accounts for 40% of the total energy produced in the United States, and about 30% of electrical energy passes through power devices. In addition, this number is estimated to increase to 80% in the next decade as the demand for EVs and new and renewable energy increases. In other words, if the energy loss of power electronics decreases by only 5%, the amount of energy that can be saved is equivalent to one third of energy consumption in the United States. As such, power semiconductor devices that feature miniaturization and increased efficiency has gained much attention recently.
At present, Si power devices are used for industrial applications. However, it is difficult to produce highly reliable Si power devices with a high-power density that can match market requirements. To withstand high voltage in hundred- or thousand-volt range without dielectric breakdown, a semiconductor must have a high breakdown voltage, which is directly related to its bandgap size. In a driving environment with a high voltage of 1 kV or higher, the temperature also increases considerably. Semiconductor materials suitable for high-power density power devices that operate at high temperatures of 200°C or higher are called wide-bandgap (WBG) semiconductors, including gallium nitride (GaN), silicon carbide (SiC), and gallium (III) oxide (Ga2O3). The bandgap of GaN and SiC is 3 times wider than that of Si, exhibiting a breakdown voltage of 2.5-3.3 MV/cm, 10 times higher than that of Si. Thus, the former two WBG semiconductors enable a power device to be made 10 times smaller than Si at the same driving voltage. Because the mobility of the carriers is similar to that of Si while the saturation velocity is higher, they are suitable for power devices that work at high voltages and high frequencies. Moreover, because of these properties, remarkable advances in GaN and SiC have been achieved through research and development over the last 20 years, and they are commercially available in diverse industrial settings, such as 5G power amplifiers and inverters for EV inverters.
Current studies and prospects of Ga2O3 for power semiconductors
Ga2O3 has recently been spotlighted as one of the next-generation power semiconductor materials. This inorganic compound can have different crystal structures, such as the monoclinic beta-phase—the most thermodynamically stable structure—corundum-structured α-phase, hexagonal ε-phase, and orthorhombic κ-phase. Beta gallium oxide (β-Ga2O3) has a bandgap of about 4.8 eV, with a breakdown voltage of 8 MV/cm, 1.5 times and 2 times higher than those of GaN, respectively. Therefore, β-Ga2O3 has attractive physical properties for high-power density drive. Therefore, Ga2O3 is expected to be used for wind turbines that require a voltage of at least 3 kV, ship and train power devices, and smart grids. β-Ga2O3 can be grown into a substrate from its melt similar to a Si boule grown by the Czochralski method. The prospect of this high-quality, low-cost, and highly-scalable native substrate makes β-Ga2O3 more attractive than other phases for commercialization. A homoepitaxial growth of β-Ga2O3 means the substrate-level dislocation density (< 104 cm-2) in the epitaxial layers, thereby resolving degraded device performance caused by defects. Although still in the research and development stage, with the Ga2O3 epitaxial layers grown on a high-quality β-Ga2O3 substrate, a power device with a breakdown voltage of 3 kV has been demonstrated. Nevertheless, some are skeptical about the success of β-Ga2O3 Because of the intrinsic material properties that are detrimental to a high-power device. For example, as in other oxide semiconductors, p-type doping is impossible. This ultimately allows for only a unipolar device, not a bipolar device, leading to limited device structure and difficulty maximizing The device’s breakdown voltage to the material limit. Moreover, the thermal conductivity of Ga2O3 is about 10 times lower than that of GaN and SiC. At high driving voltage, the ambient temperature may increase to 400°C or higher. In such an environment, the reliability of the Ga2O3 device can be severely compromised.
For semiconductor applications, Ga2O3 is a somewhat new compared to GaN and SiC, so more research must be conducted. As the β-Ga2O3 substrate may not be the key to success, a heterogeneous substrate may be needed to overcome the limitations of its physical properties. Thus, fundamental studies of the material, such as correlations between defects and devices, should be carried out more actively in conjunction with research on crystal phases other than the β-phase. In particular, because the ε- and κ-phases feature spontaneous polarization, it is expected that a high electron mobility transistor based on 2-dimensional electron gas can be produced. A driving force for initial research on Ga2O3 was the commercially available β-Ga2O3 substrate, but the success may be led by studies of the Ga2O3 material growth and devices that can combine high breakdown voltage with mobility while bypassing the limitations of its physical properties.
Ultimately, as shown in Figure 1, WBG materials like GaN, SiC, and Ga2O3 have the unique properties of their own. These materials will help us resolve energy issues with the power devices optimized for each material.
Figure 1. Power semiconductor materials and applications depending on the driving voltage and output range
 Reese, S. B., Remo, T., Green, J., & Zakutayev, A. (2019). How much will gallium oxide power electronics cost?. Joule, 3(4), 903-907.
 Huang, X., Liao, F., Li, L., Liang, X., Liu, Q., Zhang, C., & Hu, X. (2020). 3.4 kV breakdown voltage Ga2O3 trench schottky diode with optimized trench corner radius. ECS Journal of Solid State Science and Technology, 9(4), 045012.
 Bosi, M., Mazzolini, P., Seravalli, L., & Fornari, R. (2020). Ga2O3 polymorphs: tailoring the epitaxial growth conditions. Journal of Materials Chemistry C, 8(32), 10975-10992.
Roy Byung Kyu Chung
Assistant Professor, Department of Materials Science and Engineering, Kyungpook National University