What the Future Holds for WBG Devices

Article By : Maurizio Di Paolo Emilio

What is so special about these new semiconductor materials, and why are they being looked at as alternatives to silicon?

Silicon carbide (SiC) and gallium nitride (GaN) have witnessed increasing success in the semiconductor device market in recent years. GaN is now used in mobile device chargers and charging systems. Companies like Apple, Samsung, and Xiaomi have chosen GaN-based chargers that provide high power densities while maintaining, or even decreasing, the weight of these components. These chargers utilize power GaN high-electron–mobility transistor (HEMT) chips offered by companies like GaN Systems and Navitas Semiconductor.

On the other hand, SiC devices have primarily been used in the field of electric mobility. In 2017, electric-vehicle manufacturers like Tesla chose to use SiC-based motor controllers, which boosted the efficiency of their systems. This has kickstarted a race toward developing high volumes of SiC devices to meet the increasing amount of EVs that are being introduced into the market.

Their popularity begs the question: What is so special about these new semiconductor materials, and why are they being looked at as alternatives to silicon?

SiC and GaN vs silicon

As explained by Victor Veliadis in his July 28, 2022, PSMA webinar, “SiC Power Technology Status and Barriers to Overcome,” “SiC and GaN materials have a critical electric field that is about 10× higher than that of silicon, with a bandgap that is 3× higher. In a semiconductor system, the drift layer is what holds its rated voltage, which makes the thickness and doping levels of this layer determine the voltage capability of the device.”

For a specific rated voltage, the thickness of the drift layer is inversely proportional to the critical electric field. This implies that GaN and SiC devices of a particular voltage capability have drift layers that are 10× thinner than those of silicon devices. These factors drive design changes and have major implications in semiconductor design.

Due to their thinner drift layers, SiC devices are smaller in size, which decreases their capacitance. These devices can therefore efficiently switch at frequencies much higher than what is possible with silicon. As a result of the higher switching frequency, the size of passive components and magnetic devices like inductors also decreases. This leads to a significant reduction in the overall size of the system, which increases its power density. Furthermore, the large SiC bandgap and high thermal conductivity allow for high temperature operation with simplified cooling management, further decreasing system weight and volume.

None of this is to say that either SiC or GaN is superior or that silicon is obsolete. The choice of semiconductor material to be used will depend on the specifications of the application in which they are deployed. Silicon is still a strong contender in devices rated from 15 V to 650 V while also being much cheaper and more reliable, whereas GaN has been gaining popularity in low-power applications like mobile chargers and similar charging systems. As previously mentioned, GaN is the only viable wide-bandgap alternative to silicon in low-power applications, as SiC operation is impractical at voltages below 650 V.

Si, SiC, and GaN.
Figure 1: Si, SiC, and GaN (Source: Victor Veliadis)

Power factor correction

GaN enables a power factor correction (PFC) technology known as “totem-pole bridgeless PFC topology.” On the other hand, a traditional silicon boost solution would have a diode bridge where two of the diodes are constantly on. This would contribute to significant losses but is mitigated by GaN due to its essentially zero reverse recovery. 100-V GaN devices are also being deployed at data centers, as server racks are increasingly moving toward 48 V. Furthermore, 650-V GaN devices can also be deployed and run for PFC circuitry. SiC is suitable for higher-power applications than what is possible using GaN and is available in voltages ranging from 650 V to 3.3 kV, with higher-voltage devices being developed.

Stephen Russell, subject matter expert for power devices at Tech Insights, said during a company webinar, “Gallium nitride has truly found its killer app in replacing silicon and USB-C chargers for mobile devices. 2021 [was] a watershed year in market acceptance, and we only expect this momentum to continue. Gallium nitride’s real advantage, however, is its switching; it is the only viable wide-bandgap replacement for silicon at voltages less than 600 V.”

All of these devices compete heavily at the 650-V capacity, which is important, as these devices are used in the 400-V capacity bus for EVs.

EVs

EVs are a critical application for these newly adopted high-bandgap devices, as the market is expected to expand. This transformation is taking place owing to the rapid electrification across sectors and increased consciousness about emissions. They can be seen in motor drivers, DC/DC converters, on-board chargers, etc.

SiC is expected to have an edge in the EV sector, as more and more manufacturers are moving toward 800-V EV systems, due to its efficient high-voltage operational capability. Transitioning to higher-voltage systems enables higher power delivery while maintaining the same current levels. This allows copper conductors and other components to be smaller, lighter, and less expensive.

Manufacturers like Porsche, Audi, BYD, and Hyundai are already working on 800-V battery systems, while Lucid has a 900-V system under development. As Veliadis said, “Moving to 800 V while keeping the current the same doubles the power, with smaller losses. This reduces heavy copper cables, bringing lighter weight and space-saving advantages.”

Once adopted successfully into the EV space, the demand for SiC devices will further increase manufacturing. This will eventually scale down prices similar to silicon-based devices after mass production. The decrease in cost is an important step, as these devices are more expensive than silicon, with SiC material costing almost 2× to 3× as much as silicon.

Price and production

Apart from the high cost, manufacturing SiC has its own set of challenges, such as the presence of defects and slower fabrication times compared with silicon, and SiC devices are less rugged. This discourages people from adopting SiC-based systems and is a challenge to overcome. Due to their high-voltage potential, SiC devices are excellent candidates for deployment in power applications like HVDC transmission and renewable energy systems. For example, in the case of PV applications, although the SiC device cost is 3× higher than that of silicon, the overall system cost is lower due to the reduction in the size of the passive elements.

Market projections for the semiconductor industry

Despite the challenges they face, wide-bandgap devices are projected to be widely adopted across many industries and markets. Today, SiC and GaN are the only wide-bandgap semiconductor materials with commercially available power devices for a wide range of applications. Depending on their device power ratings, these materials can find applications in a variety of industries.

SiC fab infrastructures.
Figure 2: SiC fab infrastructures (Source: Victor Veliadis)

There are also projections showing that the SiC market is expected to be worth $6.5 billion by 2027. GaN devices will dominate the low-power mobile application industry, with more devices expected to reach the market with power densities higher than 20 W/in.3. These devices are expected to bring significant efficiency improvements and offer user convenience.

Unfortunately, SiC substrate and GaN epitaxy on silicon substrate production is more complicated and labor-intensive than that of silicon wafers, and this drives up cost. Moreover, the SiC and GaN market is much smaller, and it is a long way from a large-scale standardized division of labor, as the main process technologies are in the hands of a few select businesses. To overcome such issues, SiC and GaN must be mass-produced, which will bring economies-of-scale cost reductions.

 

This article was originally published on EE Times.

Maurizio Di Paolo Emilio has a Ph.D. in Physics and is a Telecommunications Engineer. He has worked on various international projects in the field of gravitational waves research designing a thermal compensation system, x-ray microbeams, and space technologies for communications and motor control. Since 2007, he has collaborated with several Italian and English blogs and magazines as a technical writer, specializing in electronics and technology. From 2015 to 2018, he was the editor-in-chief of Firmware and Elettronica Open Source. Maurizio enjoys writing and telling stories about Power Electronics, Wide Bandgap Semiconductors, Automotive, IoT, Digital, Energy, and Quantum. Maurizio is currently editor-in-chief of Power Electronics News and EEWeb, and European Correspondent of EE Times. He is the host of PowerUP, a podcast about power electronics. He has contributed to a number of technical and scientific articles as well as a couple of Springer books on energy harvesting and data acquisition and control systems.

 

Subscribe to Newsletter

Leave a comment