On the Origin of the Yellow Luminescence Band in GaN

Abstract

The yellow luminescence (YL) band with a maximum at 2.2 eV is the dominant defect‐related luminescence in unintentionally doped GaN. The discovery of the mechanism responsible for this luminescence band and related defects in GaN took many years. Eventually, a consensus has been reached that the CN acceptor is the source of the YL band (the YL1 band) in GaN samples grown by several techniques. Previously suggested candidates, such as VGa, VGaON, CNON, and CNSiGa, should be discarded. At the same time, other defects (such as the VN, BeGa, and CaGa) may cause luminescence bands with positions and shapes not much different from the YL1 band. In GaN containing high concentrations of gallium vacancies, oxygen, and hydrogen, complexes containing these species may also contribute in the red–yellow part of the photoluminescence spectrum. The main controversies related to the YL band are resolved.

Enhanced Efficiency InGaN/GaN Multiple Quantum Well Structures via Strain Engineering and Ultrathin Subwells Formed by V-Pit Sidewalls

“…The 5 K spectra of all three samples shown in Figure c include a weak yellow defect band centered at 550 nm, which was attributed to point defect-related deep states, especially those pertaining to Ga vacancies . To investigate its origin, RT PL excitation (PLE) spectra were acquired in the 500–650 nm range (Figure a), revealing that this yellow band is primarily associated with defects in the GaN layers: (i) the undoped GaN layer with a 3.42 eV band gap and (ii) the n -GaN layer beneath SLs with a narrower 3.39 eV band gap, as confirmed by the PLE spectra in Figure a.…”

Photoluminescence from defects in GaN

“…46 For example, the Cp for the CN acceptor is predicted to increase by two orders of magnitude as the temperature increases from ~50 to 600 K. 47 This dependence is attributed to the expected barrier for the hole capture. However, in the experiment, no variation of the Cp for the YL1 band could be found for temperatures between 18 and 500 K. 3 It appears that there are no substantial barriers for the hole capture at acceptors commonly contributing to PL in GaN. It is possible that PL bands from defects with significant barriers for hole capture are not observed because they are extremely weak.…”

“…The two main techniques in the PL experiment are steady-state PL (SSPL) and time-resolved PL (TRPL). 5,15 In this review and our publications since 2018, the as-measured PL spectra are not only corrected for the measurement system's spectral response but additionally multiplied by λ 3 , where λ is the light wavelength, to present the PL spectra in units proportional to the number of emitted photons as a function of photon energy. 5 Note that such corrections may change the shapes, relative contributions, and positions of broad PL bands.…”

“…In particular, the YL band with a maximum at 2.2 eV in undoped GaN was often attributed to the VGa or VGaON defects, 1 which contradicts the modern view. 3 Currently, the most common approach for the theoretical analysis of defects is the hybrid DFT, and the most popular hybrid functional used in this field is the Heyd-Scuseria-Ernzerhof (HSE) functional. 50 Several versions of hybrid functionals are used to describe defects in GaN, and the results of the calculations are slightly different.…”

“…In particular, the omnipresent yellow luminescence (YL) band in undoped GaN is unambiguously assigned to the isolated CN acceptor. 2,3 The ZPL at 2.59 eV and the characteristic phonon-related fine structure have been found, which now serve as fingerprints of the YL band (named YL1). 4 In addition to the −/0 transition level, the CN defect has a donor-like 0/+ level at 0.33 eV above the valence band maximum (VBM).…”