Trap characterization of GaN-based HEMT devices based on a self-built test platform
The trap effect is one of the main factors affecting the performance of GaN-based high electron mobility transistor(HEMT)devices.To improve the accuracy and temporal resolution of trap characterization,this paper employs the transient voltage method and constructs a dedicated test platform for characterizing traps.This method suppresses the voltage drift phenomenon,improves the temporal resolution from the millisecond level to the microsecond level,expands the characterization range of traps,and extracts information such as the time constants of traps based on the Bayesian deconvolution algorithm.Based on this method,this paper studies the capture behavior of traps in GaN-based HEMT under different voltages and temperatures,characterizes their time constants and activation energies.The experimental results show that there are four types of traps in this device.In addition to the previously proven DP2 and DP3 traps on B1505 with activation energies of 0.058 eV and 0.041 eV respectively.This paper also discovers a new trap DP1 at the microsecond level with an activation energy of 0.063 eV.By constructing the test platform,this paper fills the gap in the characterization of microsecond-level traps to provid great convenience for accurate and rapid characterization of traps.
GaN-based HEMTtrap characterizationtransient voltage methodtime constant spectrumBayesian iterationself-built test platform
NETL
NSTL
万方数据
