Differences Between Gallium Nitride (GaN) and Silicon Carbide (SiC) Transistors

Silicon has dominated the transistor world for decades. But this situation has gradually changed. Compound semiconductors composed of two or three materials have been developed to offer unique advantages and exceptional properties. For example, with compound semiconductors, we have developed light-emitting diodes (LEDs). One type consists of gallium arsenide (GaAs) and gallium arsenide phosphide (GaAsP). Others use indium and phosphorous.

Silicon has dominated the transistor world for decades. But this situation has gradually changed. Compound semiconductors composed of two or three materials have been developed to offer unique advantages and exceptional properties. For example, with compound semiconductors, we have developed light-emitting diodes (LEDs). One type consists of gallium arsenide (GaAs) and gallium arsenide phosphide (GaAsP). Others use indium and phosphorous.

The problem is, compound semiconductors are harder and more expensive to manufacture. However, they have significant advantages over silicon. New, more demanding applications, such as automotive electrical systems and electric vehicles (EVs), are finding that compound semiconductors are better able to meet their stringent specifications.

Two compound semiconductor devices, gallium nitride (GaN) and silicon carbide (SiC) power transistors, have emerged as solutions. These devices compete with long-life silicon power laterally diffused metal oxide semiconductor (LDMOS) MOSFETs and superjunction MOSFETs. GaN and SiC devices are similar in some ways, but there are also significant differences. This article compares the two and provides some examples to help you decide on your next design.

Differences Between Gallium Nitride (GaN) and Silicon Carbide (SiC) Transistors
Figure 1. Shows the power capability versus switching frequency of popular high-voltage, high-current transistors and other devices, as well as major applications.

wide bandgap semiconductor

Compound semiconductors are known as wide bandgap (WBG) devices. Without commenting on meso-lattice structure, energy levels, and other headaches of semiconductor physics, let’s just say that the definition of WBG is a model that attempts to describe how electrical current (electrons) flows in compound semiconductors.

WBG compound semiconductors have higher electron mobility and higher bandgap energy, which translates into properties superior to those of silicon. Transistors made of WBG compound semiconductors have higher breakdown voltages and tolerance to high temperatures. These devices offer advantages over silicon in high voltage and high power applications.

Differences Between Gallium Nitride (GaN) and Silicon Carbide (SiC) Transistors
Figure 2. A dual-die dual field-effect transistor (FET) cascade converts a GaN transistor into a normally-off device,
Implements standard enhanced mode of operation in high power switching circuits

Compared to silicon, WBG transistors also switch faster and operate at higher frequencies. Lower “on” resistance means they dissipate less power, improving energy efficiency. This unique combination of features makes these devices attractive for some of the most demanding circuits in automotive applications, especially hybrid and electric vehicles.

GaN and SiC transistors are becoming readily available to meet the challenges of automotive electrical equipment. The main selling points of GaN and SiC devices are these advantages:

– High voltage capability with 650 V, 900 V and 1200 V devices.
– Faster switching speed.
– higher operating temperatures; and
– Lower on-resistance, minimal power dissipation and higher energy efficiency.

GaN transistors

In the field of radio frequency (RF) power, GaN transistors were found to have early opportunities. The nature of this material enabled the development of depletion-mode field-effect transistors (FETs). Depletion-mode (or D-type) FETs, known as pseudo-state high electron mobility transistors (pHEMTs), are naturally “on” devices; since there is no gate control input, there is a natural conduction path. The gate input signal controls the conduction of the channel and turns the device on and off.

Since in switching applications, enhancement-mode (or E-type) devices that are usually “off” are preferred, which has led to the development of E-type GaN devices. The first is the cascade of two FET devices (Figure 2). Now, standard e-type GaN devices are available. They can switch at frequencies up to 10 MHz and power up to tens of kilowatts.

GaN devices are widely used in wireless devices as power amplifiers for frequencies up to 100 GHz. Some major use cases are cellular base station power amplifiers, military radar, satellite transmitters, and general-purpose RF amplification. However, due to high voltage (up to 1,000V), high temperature and fast switching, they are also incorporated into various switching power supply applications such as DC-DC converters, inverters and battery chargers.

SiC transistor

SiC transistors are natural E-type MOSFETs. These devices can switch at frequencies up to 1 MHz at voltage and current levels much higher than silicon MOSFETs. The maximum drain-source voltage is up to about 1,800V and the current capability is 100 amps. Additionally, SiC devices have much lower on-resistance than silicon MOSFETs, making them more energy efficient in all switching power supply applications (SMPS designs). A key disadvantage is that they require higher gate drive voltages than other MOSFETs, but as designs improve, this is no longer a disadvantage.

SiC devices require a gate voltage of 18 to 20 volts to turn on devices with low on-resistance. Standard Si MOSFETs require less than 10 volts gate to fully turn on. Additionally, SiC devices require a -3 to -5-V gate drive to switch to the off state. However, dedicated gate driver ICs have been developed to meet this need. SiC MOSFETs are generally more expensive than other alternatives, but their high-voltage, high-current capabilities make them well-suited for automotive power circuits.

Competition for WBG transistors

Both GaN and SiC devices compete with other mature semiconductors, especially silicon LDMOS MOSFETs, superjunction MOSFETs and IGBTs. These older devices are gradually being replaced by GaN and SiC transistors in many applications.

For example, IGBTs are being replaced by SiC devices in many applications. SiC devices can switch at higher frequencies (100 kHz+ vs. 20 kHz), allowing the size and cost of any inductors or transformers to be reduced while improving energy efficiency. Also, SiC can handle larger currents than GaN.

To summarize the comparison of GaN vs SiC, here are the highlights:

・The switching speed of GaN is faster than that of Si.
・The working voltage of SiC is higher than that of GaN.
SiC requires high gate drive voltage; and
・Superjunction MOSFETs are gradually being replaced by GaN and SiC. SiC seems to be the favorite for on-board chargers (OBC). This trend will undoubtedly continue as engineers discover newer devices and gain experience with them.

automotive applications

Many power circuits and devices can be improved by designing with GaN and SiC. One of the biggest beneficiaries is the automotive electrical system. Modern hybrid and pure electric vehicles contain equipment that can use these devices. Some of these popular applications are OBCs, DC-DC converters, motor drives, and lidars (LIDAR). Figure 3 identifies the main subsystems in an electric vehicle that require high-power switching transistors.

Differences Between Gallium Nitride (GaN) and Silicon Carbide (SiC) Transistors
Figure 3. WBG onboard chargers (OBCs) for hybrid and electric vehicles. The AC input is rectified, Power Factor Corrected (PFC),
DC-DC conversion is then done (one output is used to charge the high voltage battery and the other is used to charge the low voltage battery).

DC-DC converter. This is a power circuit that converts the high battery voltage to a lower voltage to run other electrical equipment. Batteries now have voltage ranges up to 600 or 900 volts. A DC-DC converter reduces it to 48 volts or 12 volts, or both, for the operation of other Electronic components (Figure 3). In hybrid electric and electric vehicles (HEVEVs), DC-DC can also be used for the high-voltage bus between the battery pack and the inverter.

On-board chargers (OBCs). Plug-in HEVEVs and EVs contain an internal battery charger that can be connected to AC power. This allows charging at home without the need for an external AC−DC charger (Figure 4).

Main drive motor driver. The main drive motor is a high output AC motor that drives the wheels of the vehicle. The drive is an inverter that converts the battery voltage into three-phase alternating current to run the motor.

LIDAR. LIDAR refers to a technology that combines light and radar methods to detect and identify surrounding objects. It scans a 360-degree area with a pulsed infrared laser and detects reflected light. This information is translated into detailed 3D images at a range of about 300 meters with a resolution of a few centimeters. Its high resolution makes it an ideal sensor for vehicles, especially autonomous driving, to improve the ability to identify nearby objects. The LiDAR device operates in the DC voltage range of 12-24 volts, which is derived from a DC-DC converter.

Differences Between Gallium Nitride (GaN) and Silicon Carbide (SiC) Transistors
Figure 4. A typical DC-DC converter is used to convert high battery voltages to 12 and/or 48 volts.
IGBTs used in high-voltage bridges are gradually being replaced by SiC MOSFETs.

Due to their high voltage, high current and fast switching characteristics, GaN and SiC transistors provide automotive electrical designers with flexible and simpler designs and superior performance.

Author: Yoyokuo