Gallium-Oxide may be a New Frontier for Power Semis
Several recent news stories seem to indicate that gallium-oxide power semiconductor devices may be on the commercial horizon. Kyma Technologies, Inc., of Raleigh, North Carolina, announced the addition gallium-oxide epiwafers with a beta orientation to its growing offering of advanced materials. In its announcement, the company points out that crystalline beta gallium oxide (Î²-Ga2O3) is a promising wide bandgap semiconductor material due in part to its large bandgap of 4.8 - 4.9 eV. It has a high breakdown field of 8 MV/cm, and a high dielectric constant of 10, along with an electron mobility of up to 300 cmÂ²/V-s.
In Kyoto, Japan, Flosfia Inc. also claims to be pioneering the development of next-generation wide-band gap power devices. Utilizing its proprietary Mist-CVD and all corundum structured gallium oxide based single crystal on sapphire, Flosfia expects to achieve industry-first, high breakdown voltage, high efficiency power devices with functional cost parity with silicon power devices. Later this year Flosfia plans to start to provide samples of its diodes to potential customers. Further ahead, company plans include diode production in 2018 followed by the development and launch of gallium-oxide transistors.
And, Japanâ€™s cross-ministerial Strategic Innovation Promotion (SIP) program next generation power electronics project "Research and Development on Fundamental Technologies of Gallium Oxide Power Devices" promoted by New Energy and Industrial Technology Development Organization (NEDO). Participants in this project include: National Institute of Information and Communication Technology, Silvaco Japan Co., Ltd, Tamura Corporation, Tokyo University of Agriculture and Technology and New Japan Radio Co., Ltd. The consortium believes that gallium oxide (Ga2O3) will be even more competitive than SiC and GaN for power devices because of its wider bandgap and other superior material properties and lower cost. Currently, Tamura produces single-crystal gallium oxide using the melt growth method "EFG" (Edge-defined Film-fed Growth method) with most stable Î²-type crystal structure.
According to the announcement from Kyma , the performance of gallium-oxide translates to a high voltage Baliga figure of merit (HV-BFOM) that is more than 3000 times greater than that of Si. Kyma says that the materialâ€™s Baliga figure of merit (HV-BFOM) is also more than eight times greater than that of 4H-SiC, and more than four times greater than that of GaN. Additionally, Kyma says that the materialâ€™s high-frequency Baliga figure of merit (HF-BFOM) is about 150 times that of Si, about 3 times that of 4H-SiC, and 50 percent greater than that of GaN.
Kymaâ€™s technical team recently began developing processes for the growth of ÃŸ-Ga2O3 on a number of substrates. Kyma received help through a long time collaborative partnership with leading scientists in the Sensors Directorate at Air Force Research Laboratory (AFRL) at Wright Patterson Air Force Base. Through the collaboration Kyma has demonstrated homoepitaxial growth on commercially available bulk ÃŸ-Ga2O3 substrates. The team has thus far achieved high growth rates (>3 microns/hr) and high-quality epilayers that are several microns in thick. Undoped films have demonstrated semi-insulating behavior, and n-type films have electron concentrations in the range of 1017 â€“ 1018 cm-3. The company intends to further test the materials to provide a more detailed characterization.
The anticipated superiority of Flosfiaâ€™s gallium-oxide devices stems from the materialâ€™s approximately 5â€“electron-volt bandgapâ€”way higher than that of gallium nitride (about 3.4 eV) or silicon carbide (about 3.3 eV). Bandgap is a measure of the energy required to kick an electron into a conducting state. A bigger bandgap enables a material to withstand a stronger electric field, making it possible to use a thinner device for a given voltage. Thatâ€™s important because the thinner the device, the lower its resistance, and thus the more efficient it is.
Gallium-oxide devices do not excel in all areas. One drawback is poor thermal conductivity. Another downside is that the crystal quality is not as high when grown on a sapphire substrate as it would be using a gallium-oxide substrate, which at the moment is quite expensive. GaN and SiC devices, on the other hand, are already cost and performance competitive with silicon devices in a growing number of applications.
John Palmour, CTO of SiC power device maker Wolfspeed (formerly Cree Power) sees the long term potential for a gallium oxide power device but, Palmour says, â€œHaving been through the commercialization of SiC power material and device technology at Cree, one needs to recognize the considerable time and resources required to provide reliable, cost effective commercial product with enough portfolio of devices and packages to become a market success. In addition, the incremental value of a higher bandgap gallium oxide device will be limited by its thermal conductivity disadvantages on sapphire versus monolithic SiC devices. The impact of higher forward voltage drop due to higher built-in voltage of the material versus SiC could further reduce its advantages.â€
Alex Lidow, CEO of GaN power device maker Efficient Power Conversion Corp. commented, â€œI think GaO is an unlikely replacement for silicon (or GaN for that matter). It is not very stable under normal semiconductor processing steps and tends to easily accumulate crystal imperfections (traps). I would put it in the same category as AlN and diamond as semiconductor candidates. They have crystal properties that seem great, but harvesting those characteristics is going to take a lot of time, money, and luck.â€
The process Flosfia uses for making gallium oxide devices was invented by company cofounder and Kyoto University professor Shozuo Fujita. In it, the sapphire substrate is heated and a fine mist of particles is swept into the chamber on a gust of nonÂreactive â€œcarrier gas,â€ hence the term â€œMist-CVDâ€ The mist, which contains metal compounds, decomposes when it hits the hot substrate and forms a film of gallium oxide. The whole process can be cycled through rapidly because, unlike with other methods, the chamber never has to be completely evacuated. And that drives down costs.
Engineers from Flosfia detailed the results of diodes made with this growth process in the February 2016 edition of Applied Physics Express. One device combines a 531-volt breakdown voltageâ€”the potential needed to reverse the flow of currentâ€”with an on-resistance of 0.1 milliohm per square centimeter, exceeding the limits of what is possible with silicon carbide.