Defect incorporation in In-containing layers and quantum wells: experimental analysis via deep level profiling and optical spectroscopy

Recent studies demonstrated that the performance of InGaN/GaN quantum well (QW) light emitting diodes (LEDs) can be significantly improved through the insertion of an InGaN underlayer (UL). The current working hypothesis is that the presence of the UL reduces the density of non-radiative recombinati...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of physics. D, Applied physics Applied physics, 2021-01, Vol.54 (2), p.25108, Article 025108
Hauptverfasser: Piva, F, De Santi, C, Caria, A, Haller, C, Carlin, J F, Mosca, M, Meneghesso, G, Zanoni, E, Grandjean, N, Meneghini, M
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Recent studies demonstrated that the performance of InGaN/GaN quantum well (QW) light emitting diodes (LEDs) can be significantly improved through the insertion of an InGaN underlayer (UL). The current working hypothesis is that the presence of the UL reduces the density of non-radiative recombination centers (NRCs) in the QW itself: during the growth of the UL, surface defects are effectively buried in the UL, without propagating towards the QW region. Despite the importance of this hypothesis, the concentration profile of defects in the quantum wells of LEDs with and without the UL was never investigated in detail. This paper uses combined capacitance-voltage and steady-state photocapacitance measurements to experimentally identify the defects acting as NRCs and to extract a depth-profile of the traps, thus proving the incorporation upon indium-reaction. Specifically: (i) we demonstrate that LEDs without UL have a high density (9.2 × 1015 cm−3) of defects, compared to samples with UL (0.8 × 1015 cm−3); (ii) defects are located near midgap (EC-1.8 eV, corresponding to Ei-ET ∼ 0.3 eV), thus acting as efficient NRCs; (iii) crucially, the density of defects has a peak within the QWs, indicating that traps are segregated at the first grown InGaN layers; (iv) we propose a model to calculate trap distribution in the QW, and we demonstrate a good correspondence with experimental data. These results provide unambiguous demonstration of the role of UL in limiting the propagation of defects towards the QWs, and the first experimental characterization of the properties of the related traps.
ISSN:0022-3727
1361-6463
DOI:10.1088/1361-6463/abb727