Understanding and controlling Ga contamination in InAlN barrier layers

•Quantifying the effect of GaN regrown layers on InAlN gallium contamination.•Model explaining quantitatively the behavior of gallium contamination in InAlN barrier layers.•The need of an optimum GaN regrowth thickness to have the best sheet resistance values. We have shown that in an AIXTRON Close...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of crystal growth 2019-02, Vol.507, p.139-142
Hauptverfasser: Mrad, M., Charles, M., Mazel, Y., Nolot, E., Kanyandekwe, J., Veillerot, M., Ferret, P., Feuillet, G.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Quantifying the effect of GaN regrown layers on InAlN gallium contamination.•Model explaining quantitatively the behavior of gallium contamination in InAlN barrier layers.•The need of an optimum GaN regrowth thickness to have the best sheet resistance values. We have shown that in an AIXTRON Close Coupled Showerhead MOVPE system, InAlN layers are strongly contaminated with gallium, giving increasing contamination when thicker GaN layers are grown below the InAlN. We have also shown that as the gallium contamination increases, there is reduced indium in the layers and thicker layers. To explain these phenomena we propose a quantitative model for these observations, suggesting that the TMIn precursor reacts with gallium on the showerhead surface to release TMGa or MMGa, which is then incorporated into the layers. We grew InAlN layers around 14 nm thick with a 1 nm AlN spacer to form High Electron Mobility Transistor (HEMT) structures. For the InAlN barrier layers grown directly on a GaN template, the 2D electron gas (2DEG) was degraded by the regrowth interface, but this 2DEG degradation was reduced by the growth of thin GaN layers. However, for thicker layers of GaN, there was a strong gallium contamination and so a change in alloy composition, which increased the sheet resistance. Thus we need an optimum regrowth thickness to have the best sheet resistance values.
ISSN:0022-0248
1873-5002
DOI:10.1016/j.jcrysgro.2018.10.039