In this letter, the mechanism for improvement of the dynamic performance of GaN-based light-emitting diodes with an InGaN insertion layer is investigated using the very fast electrical-optical pump-probe technique. Our measurements indicate that, when the bias current is relatively low (100 A/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), the device with the InGaN insertion layer (device A) exhibits a shorter response time than does the control (device B) without such a layer. However, when the bias current density reaches 0.5 kA/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> , devices A and B exhibit exactly the same response time during operation from room temperature to 200 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">°</sup> C. These results indicate that, under low current density (100 A/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ), the piezoelectric (PZ) field inside device A will be stronger, which should result in a lower effective barrier height with a shorter carrier escape time than is the case for device B. On the other hand, under high bias current density, both devices have the same internal response time, which indicates the screening of the PZ field inside due to injected carriers. These dynamic measurement results suggest that the origin of the efficiency droop in our device under low and high bias current densities is carrier leakage and the Auger effect, respectively.
Read full abstract