Why We Are So Excited About SiC Transistors
You may have read in our news releases or in our white papers that we are developing, jointly with GE Research, our own radio frequency (RF), ultra-high power converter system (patent pending). These are systems that take either DC or 60 Hz AC energy and convert it to RF energy that we then deliver to our underground RF XL system. You may have also heard that we have utilized silicon carbide (SiC) transistors to achieve unprecedented RF power levels and efficiencies. So why are we so excited about SiC transistors?
In the converter we are developing, transistors are used essentially as switches. For efficient operation, a switch needs to appear as a very high impedance (OFF-resistance) when the switch is open, and a very low impedance when the switch is closed (ON-resistance). Additionally, it needs to be able to switch very fast (high frequency) to generate RF! These three aspects are what is desired in a transistor to achieve low losses and hence, high efficiency. Remember Ohm’s law? The power loss in the device in watts is the product of the current flowing through the device and the voltage at the device terminals (real part to be exact). Another way to calculate power loss is the product of squared current and device resistance. You can see that when the switch is open, the OFF-resistance is so high that there is next to zero current, hence no loss. When the switch is closed, there is high current flowing through the device. To make the power loss small, we need the ON-resistance to be as low as possible. Then again, the product of current squared and resistance is very small, hence little power lost. The ideal situation, and our challenge, is to design the system such that when the transistor switches from open to close, we have either no current or no voltage on it. Without getting into too much detail, we achieve that with so called Zero Voltage Switching using resonant soft switching concepts. The key to achieving this is the ability of the transistor to switch fast. The most popular transistor used in high power applications, IGBT (insulated gate bipolar transistor), rely on injection of minority carriers to achieve low ON-resistance. That takes time and energy. It also takes time and energy to remove those carriers when we switch off the transistor. Not only does that limit the speed with which transistor can operate (and hence the frequency), but also increases loss significantly. In contrast the SiC MOSFET (metal–oxide–semiconductor field-effect transistor) eliminates these issues (it is a unipolar device) allowing significantly higher switching frequencies.
To make a high-power transistor it is advantageous to use a semiconductor material that can withstand very high voltage. High thermal conductivity is also beneficial as it allows the device to be smaller for the same amount of power, and smaller typically means faster. This is where the limits of silicon IGBT technology became apparent, and the features of SiC so exciting. SiC semiconductor has much higher dielectric breakdown voltage – that means we can apply much higher voltage to it. It also has almost 3 times higher thermal conductivity, so it can operate efficiently at much higher temperature. Importantly, it has ON-resistance that is two orders of magnitude lower than corresponding silicon IGBT device. These are incredible benefits. However, this has not been easy to achieve.
Compound semiconductors have long been recognized for their excellent potential (over silicon) in power and RF applications. Despite this positive potential, in the past, wide bandgap compound semiconductors, like SiC, but also gallium nitride (GaN) have been very hard to develop, hard to process and expensive. These technological challenges were overcome in the early 2010s, and we are now benefiting from that technological breakthrough. It allows us to achieve the conversion efficiency of close to 99%. Unheard of in RF generation technology. When we start with such an efficient RF converter building block, we can achieve very high efficiency overall in our RF XL technology.
About the Author:
Dr. Michal Okoniewski - Chief Scientific Officer & Co-Founder
Michal co-founded Acceleware in 2004 and is a renowned expert in applied computational electrodynamics and RF/antenna engineering, with a proven history of developing leading-edge scientific solutions for the electronics, medical and energy industries. With over 25 years of experience, Michal has pioneered hardware acceleration of computational electromagnetics, authoring over 350 technical publications and obtaining several patents. His innovations have revolutionized the engineering of modern computational devices, and he is now having the same impact on the science underlying heavy oil production from unconventional reservoirs. Prior to Acceleware, Michal provided consulting services for the electronic and biomedical industries in North America and Europe. He holds a Ph.D. in Electrical Engineering from the Technical University, Gdansk and is a Fellow of IEEE. He is a Professor for the Electrical and Computer Engineering Department with the Schulich School of Engineering at the University of Calgary, Canada.