Twist's Review #9 - Ron Instability of GaN HEMTs

Paper: Gate-Bias Induced RON Instability in p-GaN Power HEMTs

GaN is a wide bandgap semiconductor that has been attracting a lot of attention due to its potential in power electronics, specifically in its ability in delivering high power density. A popular device that is based on GaN is the High Electron Mobility Transistor (HEMT), which is a transistor that is based on a group of electrons in a 2D plane when a GaN heterojunction is formed with a material like AlGaN. This group of electrons is also called the 2-Dimensional Electron Gas, which exists in the channel and acts as the main driver of the device. This 2DEG provides the device with high mobility, thus the device is called the High Electron Mobility Transistor. The 2DEG itself forms due to some aspects of the two materials when stacked together, causing polarization that results in some band bending that then pushes the electron into a “well”.

This paper reveals that there is instability of Ron when a certain scheme of voltage is applied, specifically where the voltage applied is of different polarity consequently. Reliability is an important aspect of electronics devices and there are many instabilities that can occur from device operation. Ron is the resistance of the device when the device is in the on state. A lower Ron is better as it would let more current to flow through the device, thus letting the device do its job; actually turning on when switched on.

In this paper, voltages of (-6,6), (-6,2), (6,-6) and (6,-2) and the first three showed similar Ron instabilities where (6,-2) showed a less significant impact. After doing more measurements under different temperatures, it was found that the causes of this instability is due to hole traps that exists in the AlGaN where holes are being charged into from the gate and discharge to the gate. One interesting find is that the trap energy level of the (-6,6) and (6,-6) is very similar despite different trap mechanisms (charging and discharging).

Reference:
[1] A. Chini et al., “Gate-Bias Induced RON Instability in p-GaN Power HEMTs,” in IEEE Electron Device Letters, vol. 44, no. 6, pp. 915-918, June 2023, doi: 10.1109/LED.2023.3265503.

So I previously studied reliability in wide bandgap semiconductor devices, specifically those based on SiC and also learnt a little bit of GaN devices. I previously mainly focused on Bias Temperature Instability, a phenomenon where a device’s threshold voltage will shift depending on the applied gate voltage and temperature during device operation. There were other reliability aspects such as Hot Carrier Degradaton, Time Dependent Dielectric Breakdown, etc.

This paper stuck out to me, one due to its authors (very well-known authors in this field especially Nicolo Zagni and Giovanni Verzellesi), but also its study on Ron and the usage of bipolar stress/applied voltage. As also mentioned on the paper, Ron instability is arguably a more concerning phenomenon than threshold voltage shifts during operation and it is also a very important aspect of electronic devices, especially power devices. I have also previously seen some research on reliability problems occurring due to the usage of positive/negative voltages, so this was an interesting read.