Combining gallium (atomic number 31) and nitrogen (atomic number 7), gallium nitride (GaN) is a wide bandgap semiconductor material with a hard, hexagonal crystal structure. Bandgap is the energy needed to free an electron from its orbit around the nucleus and, at 3.4 eV, the bandgap of gallium nitride is over three times that of silicon, thus the designation ‘wide’ bandgap or WBG.
As bandgap determines the electric field that a material can withstand, the wider bandgap of gallium nitride enables the development of semiconductors with very short or narrow depletion regions, leading to device structures with very high carrier density. With much smaller transistors and shorter current paths, ultra-low resistance and capacitance is achieved, enabling speeds that are up to 100x faster.
The bottom line is that GaN technology can handle larger electric fields in a much smaller form factor than conventional silicon while delivering significantly faster switching. In addition, GaN technologies can operate at higher maximum temperatures than their silicon-based counterparts.
GaN is growing in importance because of its ability to offer significantly improved performance across a wide range of applications while reducing the energy and the physical space needed to deliver that performance when compared with conventional silicon technologies. In some applications where silicon as a power conversion platform has hit its physical limits, gallium nitride technologies are becoming essential, while in others the benefits of efficiency, switching speed, size and high-temperature operation combine to make GaN increasingly attractive.
As global need for energy increases, a move to GaN technology will help to meet demand while keeping carbon emissions to a minimum. Indeed, GaN design and integration has been shown to deliver next-generation power semiconductors with a carbon footprint ten times lower than older, slower silicon chips. To further support the case for GaN, it is estimated that a worldwide Si-to-GaN data center upgrade would reduce energy loss by 30-40%, which would translate as saving over 100 TWHr and 125 Million tons of CO2 emissions by 2030.