The exciton localization in wurtzite AlxGa1-xN alloys with x varying from 0.41 to 0.63 has been studied by deep-ultraviolet photoluminescence (PL) spectroscopy and picosecond time-resolved PL spectroscopy. Obvious S-shape temperature dependence was observed indicating that the strong exciton localization can be formed in high Al composition AlxGa1-xN alloys. It was also found that the Al composition dependence of exciton localization energy of AlGaN alloys is inconsistent with that of the excitonic linewidth. We contribute the inconsistency to the strong zero-dimensional exciton localization.
© 2013 Optical Society of America
Wurtzite AlGaN alloys have a very large direct energy gap ranging from around 3.4 eV to 6.2 eV, which makes them have many applications in deep-ultraviolet light emitting diodes (DUV-LEDs), laser diodes (LDs) and solar-blind UV detectors [1–4]. Solar-blind region (less than 290 nm) corresponds to the strong atmospheric absorption of solar UV waveband. AlxGa1-xN (x>0.4) material is an ideal alloy for realizing the luminescence and detection at the solar-blind region. These devices can be widely used in the areas of biochemical detection, water disinfection, air purification, and medical diagnostics. However, compared with InGaN based blue-LED, the external quantum efficiency of DUV-LED is still quite low . Multiple reasons can account for the poor external quantum efficiency (EQE) of AlGaN-based DUV-LEDs, such as the poor doping efficiencies of p-AlGaN with high Al composition and low extraction efficiency (EE) due to the changes in the valence band structure of AlGaN with the percentage of Al [5–7]. Besides, high dislocation density in the AlGaN active layer limits the internal quantum efficiency (IQE) . In order to enhance the IQE of DUV-LEDs, it’s very important to understand the carrier dynamics of AlGaN alloys. As is well known, a important factor affecting the optical properties of semiconductor alloys is the localization effect of carriers and excitons. It has been demonstrated that InGaN-based LEDs are highly efficient, which is attributed to the role of carrier localization in InGaN active region [9, 10]. And the localized states are caused by alloy inhomogeneity and/or quantum-dot-like In phase separation for InGaN-based light emitting diodes [11–14]. It is also reported that AlGaN alloys have some similarities to InGaN alloys in terms of temperature dependent near-band-edge emission behavior [15–17] and time-resolved photoluminescence [18, 19], which are attributed to the formation of bandtail states due to potential fluctuations in ternary alloys. Such potential fluctuations can lead to exciton localization at low temperatures. However, for a practical light emitting device, the active layer materials (AlxGa1-xN) with deeper potential fluctuations are required in order to obtain exciton localization at room temperature. Whether a large potential fluctuation in AlxGa1-xN alloys can be obtained is a key factor for designing high efficiency solar-blind region LEDs.
In this paper, the exciton localization in AlxGa1-xN alloys with x varying from 0.41 to 0.63 has been studied via DUV photoluminescence (PL) spectroscopy and picosecond time-resolved PL (TRPL) spectroscopy. We estimated the degree of exciton localization of AlxGa1-xN alloys through temperature dependent PL emission energy. The PL lifetimes and full width at half maximum (FWHM) of PL lines were also calculated for AlxGa1-xN alloys. The temperature dependence and dynamics properties of exciton localization in AlxGa1-xN alloys have been discussed.
The 1-μm-thick c-plane AlxGa1-xN epilayers were grown on c-plane AlN/sapphire templates by metal organic chemical vapor deposition (MOCVD). Structural characteristics of AlxGa1-xN alloy samples have been studied by high-resolution x-ray diffraction (HR-XRD), including (002)-, (102)-plane ω-2θ scan and rocking curve. The results indicated that the AlxGa1-xN films in all the samples were wurtzite single crystals and showed high qualities. Accordingly, the Al composition x was calculated from the c lattice parameter assuming the Vegard’s law. The composition x is 0.41, 0.47, 0.58, and 0.63 for AlxGa1-xN sample A, B, C, and D, respectively.
For optical investigation, PL spectra were measured from low temperature (7 K) to room temperature using a frequency-tripled mode-locked Ti:sapphire laser with a wavelength of 237 nm and an average power of 8 mW. All continuous wave PL spectra were measured by means of a single photon counting detection system together with a photomultiplier tube. And time-resolved PL (TRPL) was measured by a stand streak-camera acquisition system. The minimum time resolution was approximately 16 ps.
The structures, compositions, and PL peak positions at room temperature of four samples were summarized in Table 1.
3. Results and discussion
Temperature-dependent PL spectra ranging from 7 K to 270 K of sample A and sample B are shown in Fig. 1(a) and 1(b). Apart from a variation of the emission peak position with increasing the temperature for these two samples, it is also noticed that the full width at half maximum (FWHM) of PL line of sample A is smaller than that of sample B. Figure 1(c) and 1(d) show the temperature dependence of the main emission peak positions of sample A and B. The dotted lines in Fig. 1(c) and 1(d) are the least-squares fit of the experimental results with the Varshni equation at T > 200K for sample A and B.Fig. 1 (c) and 1(d) are the experimental data of sample A and B, which show an “S-shaped” emission shift behavior with increasing temperature. As is well known, the “S-shaped” temperature dependence is commonly attributed to potential fluctuation and band tail states inducing exciton localization in alloy material such as InGaN. At low temperatures, localized exciton emission caused by alloy fluctuation dominates the PL peak energy, and as increasing the temperature, it is red-shifted relative to the predicted energy following the temperature-induced bandgap shrinkage. Then the PL peak energy increased according to the temperature due to thermal activation of localized excitons. The anomalous blueshift is attributed to the transferring and filling processes at the band tail states or localized states. Finally, at high temperature region, the PL peak energies redshift follow the temperature dependence described by Varshni equation. Similar phenomenon has also been reported previously in InGaN/GaN multiple quantum wells [13, 14] and AlxGa1-xN alloys [15–17]. As shown in Fig. 1(c) and 1(d), we can describe the localization energy (Eloc) by the difference between the highest and lowest PL peak energy in the blueshift region, which reflects the blueshift energy in the “S-shaped” curves. For sample A, complete red -blue-red shift can be seen, and for sample B, at low temperature, the absence of the first redshift region due to its weak exciton localization is observed. The calculated results of blueshift energies indicate that the exciton localization caused by alloy fluctuation in sample A is much larger than that in sample B. At 10 K, for sample A with large localization energy, the FWHM of PL line is 48 meV, and for sample B with small localization energy, the FWHM of PL line is 102 meV. So the exciton localization energy is negative correlative with the excitonic linewidth. In sample A, the radiative recombination originating from carrier localization state caused by the potencial fluctuation owing to alloy disorder is not dominant. Other mechanism may play a key role in the recombination processes which will be discussed later.
In order to understand the above-mentioned results, TRPL was used to study the dynamics of exciton recombination process. The temperature dependent TRPL was measured for sample A and sample B. Figure 2(a) and 2(b) show the temperature dependent PL decay curves taken at the peak energy of sample A and B. For sample A, the PL decay curves especially at low temperatures can be fitted using a biexponential lineshape:19]. For sample B, at low temperatures, the PL decay curves follow well a single exponential law:Eqs. (2) and (3), respectively.
Through the PL lifetime (τPL) and internal quantum efficiency [IQE(T)], the radiative recombination lifetimes (τrad) and non-radiative recombination lifetimes (τnr) can be calculated with following equations:Fig. 2(c) and 2(d) for sample A and B, respectively. In order to analyze the degree of confined potential, which reflects the effects of exciton localization, the τrad of sample A and B in Fig. 2(c) and 2(d) are investigated. Previous theoretical calculation has discussed the dimensional property of the excitons, showing that τrad of a 0-dimensional (0D) exciton is almost independent of T, τrad of the 2D exciton is proportional to T, and that of 3D exciton increases proportionally to T 1.5 [20, 21]. As shown in Fig. 2, for sample A which has large exciton localization energy, it can be seen that there is not much variation of τrad at T < 120 K, indicating a typical characteristic of the 0D exciton. However, for sample B, an increase in τrad at T < 60 K was observed, exhibiting 3D behavior, which is attributed to exciton delocalization from bound or localized states. The possible reason for the decrease in τrad of sample B at 60 K < T < 120 K is that excitons may be released from band-tail to 3D space gaining the wavefunction overlap of the electrons and holes . Finally, both sample A and B show an increase in τrad at T > 120 K, and it can be explained by the τrad of excitons in 3D space increases proportionally to T 1.5, as discussed above. Therefore, an obvious 0D strong exciton localization induced radiative recombination was observed in sample A.
Figure 3 shows the variations of the exciton localization energy (Eloc), and full width at half maximum (FWHM) of the PL emission line with Al composition x in AlxGa1-xN sample A, B, C, and D. For the results of sample B, C, and D, the FWHM is proportional to the Eloc, similar to that in Ref. 15. Temperature dependent TRPL was also measured in sample C and D. Combined with the dynamics results of sample B, in these three samples, the τrad of excitons shows the same tendency of 3D excitons, indicating that the weak localized carriers introduced by alloy fluctuations diffuse into free/extended states. Generally, alloy fluctuations also lead to a statistical distribution in the excitonic transition energies, which causes the emission linewidth broadening.
On the contrary, sample A shows very large exciton localization energy, and the narrow FWHM of the PL emission line exhibits a different mechanism of light emission, caused by the strong 0D exciton localization. Even at room temperature, the dominant emission mechanism would originate from the strong localized states. Furthermore, the strong exciton localization will lead to the quantized energy levels, resulting in the shrinkage of emission linewidth. Obviously, the strong exciton localization will lead to a quantum-dot-like behavior of light emission, and effectively suppress the non-radiative processes, resulting in a high efficiency, similar to the mechanism of luminescence in InGaN-QW LEDs.
N. Nepal et al. found the Eloc increases with Al content x for x≤0.7 , and decreases with x for x≥0.8. And K. B. Lee et al. found the Eloc increases with Al content x for 0.08≤x≤0.52 . However, with x varying from 0.41 to 0.63 in this letter, the Eloc decreases with x, in contrast to the results in these references. The opposite variation tendency may be caused by different growth conditions. The composition fluctuations of InxGa1-xN and AlxGa1-xN show the same tendency as the fluctuations in an ideal solid solution, , indicating the maximum at x = 0.5 . Compared with AlxGa1-xN, the composition fluctuations are larger in the InxGa1-xN alloys, due to the immiscibility between InN and GaN. This immiscibility implies that localization state can be introduced through the phase separation in InGaN. Even though GaN and AlN show the good miscibility, large composition fluctuations can be introduced by a non-equilibrium growth (far from equilibrium growth condition). For MOCVD, decreasing the V/III flow ratios and the growth pressures will result in the compositional inhomogeneities of AlxGa1-xN alloys . And for MBE, the Ga-riched “liquid phase” epitaxy will lead to a strong band-structure potential fluctuations [26, 27]. Of course, other growth conditions, such as varied templates, maybe also affect the uniformity of AlxGa1-xN alloy leading to alloy fluctuation.
In conclusion, the temperature dependence and dynamics properties of exciton localization in wurtzite AlxGa1-xN alloys with x varying from 0.41 to 0.63 have been studied by DUV-PL spectroscopy and picosecond TRPL spectroscopy. For sample A with large localization energy (30 meV), the FWHM of PL line is only 48 meV at 10 K, while for sample B with small localization energy (15 meV), the FWHM of PL line is as large as 102 meV at the same temperature, indicating that the exciton localization energy is not correlative with the excitonic linewidth. We believe that this inconsistency is due to the 0D behavior of strong localized exciton recombination. The large localization energy in sample A shows that the strong exciton localization can be achieved in high Al composition AlxGa1-xN alloys, which is similar to the quantum-dot-like behavior of InxGa1-xN alloy in blue-LEDs. The results suggest that the efficient UV-LEDs can be obtained by strong exciton localization in control of suitable growth conditions.
This work was supported by National Basic Research Program of China(Nos:2012CB619306 and 2012CB619301)and National High Technology Research and Development Program of China (2011AA03A111).
References and links
1. A. Khan, K. Balakrishnan, and T. Katona, “Ultraviolet light-emitting diodes based on group three nitrides,” Nat. Photonics 2(2), 77–84 (2008). [CrossRef]
2. H. Hirayama, S. Fujikawa, N. Noguchi, J. Norimatsu, T. Takano, K. Tsubaki, and N. Kamata, “222-282 nm AlGaN and InAlGaN-based deep-UV LEDs fabricated on high-quality AlN on sapphire,” Phys. Status Solidi 206(6), 1176–1182 (2009) (a). [CrossRef]
3. M. S. Shur and R. Gaska, “Deep-ultraviolet light-emitting diodes,” Electron Devices, IEEE Transactions on 57(1), 12–25 (2010). [CrossRef]
5. C. G. Van de Walle and J. Neugebauer, “First-principles calculations for defects and impurities: Applications to III-nitrides,” J. Appl. Phys. 95(8), 3851–3879 (2004). [CrossRef]
6. K. Nam, J. Li, M. Nakarmi, J. Lin, and H. Jiang, “Unique optical properties of AlGaN alloys and related ultraviolet emitters,” Appl. Phys. Lett. 84(25), 5264–5266 (2004). [CrossRef]
7. R. Banal, M. Funato, and Y. Kawakami, “Optical anisotropy in -oriented AlxGa1-xN/AlN quantum wells (x> 0.69),” Phys. Rev. B 79(12), 121308 (2009). [CrossRef]
8. K. Takeda, F. Mori, Y. Ogiso, T. Ichikawa, K. Nonaka, M. Iwaya, S. Kamiyama, H. Amano, and I. Akasaki, “Internal quantum efficiency of GaN/AlGaN‐based multi quantum wells on different dislocation densities underlying layers,” Phys. Status Solidi 7(7-8c), 1916–1918 (2010). [CrossRef]
9. K. O'Donnell, R. Martin, and P. Middleton, “Origin of luminescence from InGaN diodes,” Phys. Rev. Lett. 82(1), 237–240 (1999). [CrossRef]
10. H. Yang, P. Kuo, T. Lin, Y. Chen, K. Chen, L. Chen, and J.-I. Chyi, “Mechanism of luminescence in InGaN/GaN multiple quantum wells,” Appl. Phys. Lett. 76(25), 3712–3714 (2000). [CrossRef]
11. R. Martin, P. Middleton, K. P. O’Donnell, and W. Van der Stricht, “Exciton localization and the Stokes’ shift in InGaN epilayers,” Appl. Phys. Lett. 74(2), 263–265 (1999). [CrossRef]
12. Y.-S. Lin, K.-J. Ma, C. Hsu, S.-W. Feng, Y.-C. Cheng, C.-C. Liao, C. Yang, C.-C. Chou, C.-M. Lee, and J.-I. Chyi, “Dependence of composition fluctuation on indium content in InGaN/GaN multiple quantum wells,” Appl. Phys. Lett. 77(19), 2988–2990 (2000). [CrossRef]
13. Y.-H. Cho, G. Gainer, A. Fischer, J. Song, S. Keller, U. Mishra, and S. DenBaars, ““S-shaped” temperature-dependent emission shift and carrier dynamics in InGaN/GaN multiple quantum wells,” Appl. Phys. Lett. 73(10), 1370–1372 (1998). [CrossRef]
15. N. Nepal, J. Li, M. Nakarmi, J. Lin, and H. Jiang, “Exciton localization in AlGaN alloys,” Appl. Phys. Lett. 88(6), 062103 (2006). [CrossRef]
16. E. Kuokstis, W. Sun, M. Shatalov, J. Yang, and M. Asif Khan, “Role of alloy fluctuations in photoluminescence dynamics of AlGaN epilayers,” Appl. Phys. Lett. 88(26), 261905 (2006). [CrossRef]
17. A. Bell, S. Srinivasan, C. Plumlee, H. Omiya, F. Ponce, J. Christen, S. Tanaka, A. Fujioka, and Y. Nakagawa, “Exciton freeze-out and thermally activated relaxation at local potential fluctuations in thick AlxGa1-xN layers,” J. Appl. Phys. 95(9), 4670–4674 (2004). [CrossRef]
18. H. Kim, R. Mair, J. Li, J. Lin, and H. Jiang, “Time-resolved photoluminescence studies ofAlxGa1-xNalloys,” Appl. Phys. Lett. 76(10), 1252–1254 (2000). [CrossRef]
19. T. Onuma, S. F. Chichibu, A. Uedono, T. Sota, P. Cantu, T. M. Katona, J. F. Keading, S. Keller, U. K. Mishra, and S. Nakamura, “Radiative and nonradiative processes in strain-free AlGaN films studied by time-resolved photoluminescence and positron annihilation techniques,” J. Appl. Phys. 95(5), 2495 (2004). [CrossRef]
20. H. Gotoh, H. Ando, T. Takagahara, H. Kamada, A. Chavez-Pirson, and J. Temmyo, “Effects of dimensionality on radiative recombination lifetime of excitons in thin quantum boxes of intermediate regime between zero and two Dimensions,” Jpn. J. Appl. Phys. 36(Part 1, No. 6B), 4204–4208 (1997). [CrossRef]
21. J. Huang, X. Dong, X. Luo, X. Liu, Z. Xu, and W. Ge, “Localized exciton dynamics in AlInGaN alloy,” Solid State Commun. 126(8), 473–477 (2003). [CrossRef]
22. S. Chichibu, K. Hazu, T. Onuma, and A. Uedono, “Collateral evidence for an excellent radiative performance of AlxGa1-xN alloy films of high AlN mole fractions,” Appl. Phys. Lett. 99(5), 051902 (2011). [CrossRef]
23. K. Lee, P. Parbrook, T. Wang, F. Ranalli, T. Martin, R. Balmer, and D. Wallis, “Optical investigation of exciton localization in AlxGa1-xN,” J. Appl. Phys. 101(5), 053513 (2007). [CrossRef]
24. L. Teles, J. Furthmüller, L. Scolfaro, J. Leite, and F. Bechstedt, “First-principles calculations of the thermodynamic and structural properties of strained InxGa1-xN and AlxGa1-xN alloys,” Phys. Rev. B 62(4), 2475–2485 (2000). [CrossRef]
25. M. Miyoshi, M. Sakai, H. Ishikawa, T. Egawa, T. Jimbo, M. Tanaka, and O. Oda, “MOVPE growth and characterization of high-Al-content AlGaN/GaN heterostructures on 100-mm-diameter sapphire substrates,” J. Cryst. Growth 272(1-4), 293–299 (2004). [CrossRef]
26. Y. Liao, C. Thomidis, C.- Kao, and T. D. Moustakas, “AlGaN based deep ultraviolet light emitting diodes with high internal quantum efficiency grown by molecular beam epitaxy,” Appl. Phys. Lett. 98(8), 081110 (2011). [CrossRef]
27. E. Francesco Pecora, W. Zhang, L. Zhou, D. J. Smith, J. Yin, R. Paiella, L. Dal Negro, and T. Moustakas, “Sub-250 nm room-temperature optical gain from AlGaN/AlN multiple quantum wells with strong band-structure potential fluctuations,” Appl. Phys. Lett. 100, 061111 (2012). [CrossRef]