Wide-Bandgap Semiconductors: When Research Becomes RealityFebruary 07, 2020 by Robert Keim
Silicon carbide and gallium nitride are gaining ground in a market that has long been dominated by silicon. What does the rise of wide-bandgap materials tell us about the relationship between research and engineering?
In 2004, the BBC reported that scientists had achieved atomic-scale “teleportation” using laser light. The article states that this phenomenon would allow for quantum data storage in the “computers of tomorrow.”
I’m not sure what exactly they meant by “tomorrow,” but that was 15 years ago, and I haven’t yet seen an advertisement for a microcontroller or memory chip that stores data as qubits.
The same year, the director of IBM’s Almaden Research Center told a Stanford journalist that spintronics could potentially “revolutionize the electronics industry, just as the transistor did 50 years ago.”
I wonder how many electrical engineers know what “spintronics” means. In any case, if the revolution exists it’s maintaining a very low profile.
And then there's the diamond. In the past, we've discussed how some researchers are vying for diamonds to replace silicon as a power and sensor semiconductor.
Though undoubtedly useful for saw blades, polishing devices, and marriage proposals, its status as the semiconductor of the future is still very much in the theoretical realm.
These examples remind us that many—or most?—engineering research projects don’t significantly influence the design work performed by professional engineers.
This doesn’t mean that such research has no value, but it does suggest that engineers who want to keep their knowledge and skills up-to-date would be much better off reading an app note or a newly released datasheet instead of a press release written by researchers.
Is WBG an Exception?
The term “wide-bandgap (WBG) semiconductor” refers to materials that can function as semiconductors but have a valence-band-to-conduction-band energy gap that is larger than that of silicon.
This valence-band-to-conduction-band energy gap is discussed in more detail in the Band Theory of Solids in the AAC textbook.
This means that it takes more energy to bounce electrons into a state that allows for the flow of electric current.
Image of electron band separation in semiconducting substances. Image from the Band Theory of Solids of the AAC textbook
The higher energy requirement makes a WBG material more like an insulator, and this might seem like a disadvantage.
However, in the context of semiconductor design, WBG materials bring major benefits by allowing devices to operate at higher temperatures, voltages, and frequencies.
Unlike so many other laboratory ideas that don’t end up in the engineer’s toolbox, silicon carbide (SiC) and gallium nitride (GaN) are truly changing the industry and allowing designers to achieve performance that wasn’t possible in the days when our options were mostly silicon, silicon, or silicon.
WBG Devices for High-Power and High-Frequency Applications
The most mature WBG material is SiC; it is already widely used in the fabrication of switching devices such as MOSFETs and thyristors.
GaN has potential as a semiconductor for power devices and is a major improvement over silicon in RF applications.
Cree introduced the first commercial SiC power MOSFET and affirms that this WBG material offers higher thermal conductivity, which allows for higher current in a smaller package, and higher critical breakdown field, which enables lower on-state drain-to-source resistance.
Microchip and ROHM have released new SiC MOSFETs and diodes, and we’re also seeing investment from Infineon, STMicroelectronics, and ON Semiconductor, especially in automotive power design.
We've previously discussed the pros and cons of Silicon Carbide (SiC) FETs by analyzing a MOSFET from Cree, the C3M0075120K. Image used courtesy of Wolfspeed
Analog Devices has produced GaN devices for high-frequency applications and believes that this material will help designers to reduce size and weight while achieving higher efficiency and extending bandwidth.
This graph conveys the high-power/high-frequency combination offered by wide-bandgap semiconductors. Image used courtesy of Analog Devices
The Road Ahead
At the beginning of October, Transparency Market Research published a market research report on wide-bandgap semiconductors.
They’re predicting a compound annual growth rate of 22% over the next eight years, with the strongest development occurring in the Asia Pacific and North American markets.
A significant proportion of this growth will be driven by the hybrid/electric vehicle sector, but applications such as power supplies, motor drives, and wind turbines will also figure prominently.
At this point, it looks like WBG semiconductors really are going to change the way that electrical engineers design circuits.
SiC and GaN devices are becoming more affordable and more widely available, and they offer performance that cannot be achieved with silicon, silicon germanium, or gallium arsenide.
What are your thoughts? Will the proliferation of WBG devices be an electronics “revolution,” or just one more chapter in the longstanding, gradual improvement of semiconductor technology?