Next generation GaN power HEMTs: from device design to application

It is estimated that up to 25% of electrical energy is lost due to electronic energy conversions. In Europe, this means that up to 730 TWh is lost, or an equivalent of 365 megatons of CO₂. The electronic energy conversions (for example, from AC to DC, or DC to DC at different voltage levels) are realized by power electronic circuits that typically contain several power switches. These power switches are either realized as discrete semiconductor components such as (Schottky) diodes, MOSFETs, or IGBTs; or as part of an IC in which the power switches are incorporated with dedicated circuitry to drive and control the power switches.

The vast majority of power switches or power electronic circuits in general are – still today – using silicon as the active semiconductor. Yet wide band gap semiconductors such as SiC and GaN hold better cards when it comes to energy conversion. Material properties such as higher critical electric fields at breakdown, higher mobilities associated with two-dimensional electron gases, and higher saturation velocities outperform silicon’s properties. As a result, devices made in these wide band gap materials can operate at frequencies, voltages and current levels beyond the reach of silicon, opening new paths for more efficient and more dense power electronic switches and circuits. Silicon can rely on decades of technological advancements driven by the digital revolution. Wide band gap materials need to catch up and one of the main challenges is the growth of these crystals and their efficient use in power applications. GaN is typically grown epitaxially onto a silicon substrate whereby a complex layer stack is built consisting of several III-nitride layers.

The main objective of this PhD topic is to help optimize the next generation of GaN devices. The PhD candidate will have to understand thoroughly the device physics and collaborate with the GaN teams to improve the performance and reliability of the lateral GaN power switches. From device design towards application, including processing and characterization, the PhD student will be involved in all aspects. The focus will be on the device design (eventually including Technology CAD simulations) and on-wafer electrical characterization. The PhD student joins a team with over a decade of experience in processing, simulation, and characterization of GaN switches in a top-notch environment. The student will receive training by highly-skilled professionals and will be embedded in the GaN device team. Typical tools that will be used by the student include probe stations for on-wafer electrical characterization, software tools for data analysis (python), and TCAD Synopsys software suite.