Open Access
Add to BibSonomy
Abstract
We investigated the optical and electrical properties of red AlGaInP light-emitting diodes (LEDs) as functions of chip size, p-cladding layer thickness, and the number of multi-quantum wells (MQWs). External quantum efficiency (EQE) decreased with decreasing chip size. The ideality factor gradually increased from 1.47 to 1.95 as the chip size decreased from 350 μm to 15 μm. This indicates that the smaller LEDs experienced larger carrier loss due to Shockley-Read-Hall nonradiative recombination at sidewall defects. S parameter, defined as ∂lnL/∂lnI, increased with decreasing chip size. Simulations and experimental results showed that smaller LEDs with 5 pairs of MQWs had over 30% higher IQE at 5 A/cm2 than the LED with 20 pairs of MQWs. These results show that the optimization of the number of QWs is needed to obtain maximum EQE of micro-LEDs.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Micro-pixel light-emitting diodes (μ-LEDs) are of increasing importance because of their use in various applications including self-emissive micro-displays [1], visible light communications [2] and optogenetics [3,4]. For display application, μ-LED-based arrays have been demonstrated with pixel dimensions as small as 12 μm [5–7]. It was reported that μ-LEDs showed higher output performance than conventional LEDs due to better strain relaxation [8], improved light extraction efficiency (LEE) [9,10], and uniform current spreading [11]. The μ-LEDs also exhibited improved thermal effects [12] and operation at higher current density [13] as compared with conventional LEDs. The effect of the chip size on the performance of blue, green, and red LEDs has been investigated by many researchers [12,14–19]. For example, Gong et al. [12], investigating the effect of chip size on the light output, spectral shift, and self-heating of 400 nm InGaN-based LEDs, reported that smaller LED pixels could yield higher power densities and withstand higher current densities. The LEDs experienced blue-shift at the low current density and red-shift at the high current density. Olivier et al. [15], investigating the electrical and optical properties of blue GaN-based μ-LEDs as a function of chip size, found that smaller LEDs exhibited lower maximum efficiency. Furthermore, at low current levels, light emission was homogeneous across the surface of the LEDs, while at higher current densities, emission was inhomogeneous. This difference was attributed to degraded electrical injection caused by fabrication process damage. Hwang et al. [16] also investigated the size-dependent performance of blue InGaN-based μ-LEDs and reported that the peak external quantum efficiencies (EQEs) of smaller μ-LEDs were lower than the larger ones. The different EQE was explained in terms of nonradiative recombination caused by etching damage. The smallest μ-LED showed improved current spreading and so experienced less efficiency droop than the largest one. Royo et al. [17], investigating the optical efficiency of red AlGaInP-based micro cavity LEDs (μ-LEDs) as a function of chip size, found that EQE of μ-LEDs largely decreased with decreasing chip size. In addition, μ-LEDs exhibited about 55% lower apparent internal quantum efficiency (IQE) at 55 A/cm2 than reference LED. This was attributed to the poor current injection and electron leakage current of μ-LEDs. In this study, we investigated the effects of the chip size, the thickness of a p-cladding layer, and the number of multi-quantum wells (MQWs) on the output efficiency of red AlGaInP-based LEDs to elucidate the efficiency degradation mechanisms and hence to optimize the output performance of μ-LEDs. All LEDs were fabricated from the same epitaxial wafer. In addition, 3-D simulation was conducted to investigate the effect of the number of MQWs on the efficiency of LEDs.
2. Experimental procedure
A low-pressure metal organic chemical vapor deposition system was used to grow AlGaInP-based epilayers on n-GaAs substrates. The LED structure consisted of n-GaAs, 4.0 μm-thick n-(Al0.5Ga0.5)0.5In0.5P, 50 nm-thick AlInP diffusion barrier (DB), 20 pairs of Ga0.5In0.5P (7 nm)/(Al0.7Ga0.3) 0.5In0.5P (14 nm) multi-quantum wells (MQWs), 50 nm-thick AlInP DB, 200 nm-thick p-Al0.5In0.5P, 0.5 μm-thick p-GaP, and 20 nm-thick p++-GaP. After growth, LED structures with six different chip sizes were processed. The samples were treated with acetone and deionized (DI) water for 5 min per cleaning agent, and blown dry in N2 ambient, after which a 60 nm-thick ITO layer as a p-type Ohmic contact was deposited by radio frequency magnetron sputtering. The mesa structures were fabricated using an inductively coupled plasma reactive ion etching (ICP-RIE) system, followed by etching away the p-GaP, p-AlInP, MQWs, and n-AlGaInP to expose n-GaAs. AuGe as an n-Ohmic contact was evaporated. A SiO2 passivation layer was deposited using a plasma-enhanced chemical vapor deposition system. For small chips, a 250 nm-thick ITO bridge was formed to connect the chips in parallel. The geometries of the devices, whose size varied from 15 μm to 350 μm, were summarized in Table 1, where the sidewall surface ratio represents the exposed sidewall surface area divided by the chip area. Simulations were performed on the basis of surface carrier loss mechanism to understand the size-dependence of EQE. A simulation was conducted using a SpeCLEDTM 3-D LED simulator to investigate the effect of the MQW pairs on the performance of LEDs [18,19]. For the simulations, an ambipolar diffusivity of20 cm2/s and a surface recombination velocity of 105 cm/s were employed [20], and the active region characteristics were calculated by a SiLENSeTM simulator.
Table 1. Summary of the size, area, number, and sidewall surface ratio of LEDs
3. Results and discussion
Figure 1(a) represents EQE curves for six different LEDs as a function of current density. It is obvious that EQE is dependent on the LED size. The larger LEDs exhibit maximum EQE at lower current density than the smaller LEDs. In other words, the maximum EQE is shifted to higher current densities as the chip size decreases. The shift of maximum EQE can be related to leakage current and/or an increased Shockley-Read-Hall (SRH) non-radiative recombination at sidewall defects in the smaller geometries [14,16]. A sidewall-surface ratio, which is defined as an exposed sidewall quantum well surface area/chip area, decreases with increasing chip size (Table 1). It is noted that the total chip area of the sample #3 is larger than those of the sample #4 and #5, but #3 experiences a more severe reduction in the EQE than #4 and #5. Notwithstanding the large difference in the total chip areas, #6 reveals similar EQE to those of #4 and #5. Thus, considering that #1, #2, and #3 has larger sidewall surface ratio (0.96 to 3.2%) than #4 and #5 (0.48 and 0.32%, respectively) (Table 1), their lower EQE can be attributed to the increased SRH recombination at sidewall defects.
Fig. 1 (a) EQE curves for six different LEDs as a function of current density. (b) Normalized EQEs at 5 A/cm2 for the two sets of different-size LEDs.
Figure 1(b) exhibits normalized EQEs at 5 A/cm2 for the two sets of six different LEDs. The second set is the same as the first set except the p-type cladding layer which is 1.5 μm-thicker than that of the first set. It is evident that the EQE gradually decreases with decreasing chip size. It is noteworthy that at the chip sizes smaller than 50 μm, the second set of the samples shows somewhat higher EQE than the first set. At the moment, the exact mechanism for this improvement is not clearly understood. Several factors, such as the improved light extraction caused by a thicker p-cladding layer, carrier confinement effect, and current spreading effect, may be responsible for the improved performance, which remains yet to be confirmed. Furthermore, the simulated result is in good agreement with the experimental results. Thus, by considering that the simulation was performed based on surface carrier loss mechanism, the size-dependence of the light output power can be explained in terms of carrier loss caused by increased SRH non-radiative recombination at sidewall defects, which are dominant in the smaller LEDs [14].
Figure 2(a) illustrates the ideality factor (nideality) for LEDs as a function of chip size, which were estimated using the relationship given as [21,22],
where q is the elementary charge, k is the Boltzmann constant, and T is the temperature. The ideality factors were obtained as a function of current density and their minimum values were illustrated in Fig. 2(a). To measure the ideality factors of the sample #1, #2, and #3, 10 small chips were connected in parallel and the I−V curves were then obtained as a function of current. The ideality factors were assessed using the I−V curves. The ideality factor is dependent on the chip size. In other words, the ideality factor gradually increases from 1.47 to 1.95 as the chip size decreases from 350 μm to 15 μm. The smaller LEDs reveal higher ideality factor. It is known that the ideality factor denotes recombination mechanisms [23]. In other words, an ideality factor of 1.0 is associated with the band-to-band radiative recombination, while an ideality factor of 2.0 originates from the Shockley-Read-Hall (SRH) recombination via defect levels and those exceeding 2.0 is caused by the defect-assisted tunneling phenomenon [24]. This indicates that the smaller LEDs suffer from carrier loss due to SRH nonradiative recombination at sidewall defects [14,25].Fig. 2 (a) Size dependence of minimum ideality factor for LEDs, which was estimated as a function of current density. (b) Variation of S for different LEDs as a function of driving current density. (c) Size dependence of the I−V and (inset) J−V curves of LEDs.
Figure 2(b) exhibits the variation of S for the samples, which is defined as ∂lnL/∂lnI, as a function of driving current density which can be given as [23],
where L is the light output power, η is the coupling efficiency, B is the radiative recombination coefficient, and N is the carrier concentration in the MQW region. If the SRH recombination is dominant on the assumption that the injection efficiency (the ratio of the recombination current in the active region to the total driving current) is constant, the value of S should be close to 2.0 [26]. It can be seen from Fig. 2(b) that S gradually increases as the current density decreases. Furthermore, S increases with decreasing chip size. This means that the SRH recombination is dominant in the smaller LEDs, which is consistent with the ideality factor behavior.Figure 2(c) shows the current-voltage (I−V) and current density-voltage (J−V) characteristics of LEDs as a function of chip size. At the same current, the series resistance and forward voltage increase with decreasing chip size. Although the chip size decreases down to 15 μm, the forward and reverse leakage current characteristics are not degraded. This indicates that the lower EQE of small LEDs is not related to degradation in the current injection efficiency caused by the increased leakage current. Moreover, the J−V characteristics (the inset) reveal that the series resistance and forward voltage of the LEDs decrease with decreasing chip size. These I−V and J−V results exhibit that the current injection efficiency is improved with decreasing chip size. All of the LEDs have the same epilayer structures and so they are expected to have the same radiative recombination efficiency. On the one hand, the LEE may be improved with decreasing chip size, resulting in increase in the EQE. On the contrary, however, the EQE is lowered with decreasing chip size, as shown in Fig. 1(a). This is indicative of the occurrence of more SRH non-radiative recombination at the QW sidewall of smaller chips.
To investigate the effect of the number of MQWs on the light output efficiency of LEDs, 3-D simulations were performed with a SpeCLEDTM simulator. In other words, the number of MQW pairs indicates the exposed sidewall area of MQWs generated by inductively coupled plasma etching, e.g. the more QW pairs the larger QW sidewall area, resulting in larger surface recombination carrier loss. Figure 3(a) displays the internal quantum efficiency (IQE) of LEDs (chip size: 30 × 30 μm2) with different numbers of MQWs as a function of current density. Below 5 A/cm2, IQE increases with a decrease in the number of MQWs. In particular, the LEDs with a single QW and 5 pairs of MQWs show over 30% higher IQE at 5 A/cm2 than the LED with 20 pairs of MQWs. This improvement can be attributed to the combined effects of the reduced QW sidewall area due to the decreased pairs and the increased volumetric current density induced by the reduced volume of MQWs.
Fig. 3 (a) IQE of LEDs (chip size: 30 × 30 μm2) with different numbers of MQWs as a function of current density. (b) Relative EQE experimentally obtained from different-size LEDs with different numbers of MQW pairs as a function of current density.
Figure 3(b) exhibits relative EQEs from different-size LEDs with different numbers of MQW pairs as a function of current density. Notably, EQE is dependent on the chip size and number of MQWs. For instance, the small-size LEDs with 5 pairs of MQWs give higher EQEthan the LED with 20 pairs. For the larger LEDs, the 5 pair-MQW LED shows higher EQE below 2 A/cm2, above which the 20 pair-MQW LED exhibits higher EQE. These experimental results are consistent with the simulated IQE results [Fig. 3(a)]. The results show that a more exposed sidewall in the smaller chip sizes causes the occurrence of more nonradiative recombination (e.g., surface recombination). The surface recombination velocity and carrier diffusion length of AlGaInP was known to be much larger than those of InGaN [20]. Thus, red AlGaInP-based LEDs could suffer from a large reduction in the IQE as the chip size decreases because of the dominant sidewall surface area to chip volume ratio (Table 1). Considering that all LEDs with different chip sizes were processed from the same wafer, these results imply that the lower maximum EQE for smaller LEDs is attributed to the increased nonradiative surface recombination at the sidewall surfaces. To minimize the surface carrier loss and hence to maximize the EQE of micro-LEDs, it is important to optimize the QW sidewall area caused by the number of MQW pairs and volumetric current density induced by the volume of MQWs and a p-cladding layer.
4. Conclusion
The EQE of red AlGaInP LEDs was characterized as functions of chip size, p-cladding layer thickness, and the number of MQWs. The EQE decreased with decreasing chip size. For μ-LEDs (<50 μm), a thicker p-cladding layer produced higher EQE. The ideality factor and S parameter results revealed that smaller LEDs (with larger sidewall area ratio) experienced larger carrier loss due to SRH nonradiative recombination. Simulation and experimental results showed that the μ-LEDs with less than 5 pairs of MQWs had much higher IQE than the LED with 20 pairs of MQWs. These results imply that optimization of the QW sidewall area caused by the number of MQW pairs and current density induced by the volume of MQWs and p-cladding layer is essential for the achievement of maximum EQE of micro LEDs.
Funding
LG Innotek Co., Ltd.; Global Research Laboratory program through the National Research Foundation of Korea funded by the Ministry of Science and ICT (NRF-2017K1A1A2013160).
Acknowledgments
The authors gratefully acknowledge financial support from LG Innotek Co., Ltd. and National Research Foundation of Korea.
References and links
1. J. Day, J. Li, D. Y. C. Lie, C. Bradford, J. Y. Lin, and H. X. Jiang, “III-Nitride full-scale high-resolution microdisplays,” Appl. Phys. Lett. 99, 031116 (2011).
2. J. J. D. McKendry, R. P. Green, A. E. Kelly, Z. Gong, B. Guilhabert, D. Massoubre, E. D. Gu, and M. D. Dawson, “High-Speed Visible Light Communications Using Individual Pixels in a Micro Light-Emitting Diode Array,” IEEE Photonics Technol. Lett. 22(18), 1346–1348 (2010).
3. V. Poher, N. Grossman, G. T. Kennedy, K. Nikolic, H. X. Zhang, Z. Gong, E. M. Drakakis, E. Gu, M. D. Dawson, P. M. W. French, P. Degenaar, and M. A. A. Neil, “Micro-LED arrays: a tool for two-dimensional neuron stimulation,” J. Phys. D Appl. Phys. 41(9), 094014 (2008).
4. N. Grossman, V. Poher, M. S. Grubb, G. T. Kennedy, K. Nikolic, B. McGovern, R. Berlinguer Palmini, Z. Gong, E. M. Drakakis, M. A. A. Neil, M. D. Dawson, J. Burrone, and P. Degenaar, “Multi-site optical excitation using ChR2 and micro-LED array,” J. Neural Eng. 7(1), 16004 (2010). [PubMed]
5. H. X. Jiang and J. Y. Lin, “Nitride micro-LEDs and beyond--a decade progress review,” Opt. Express 21(9Suppl 3), A475–A484 (2013). [PubMed]
6. Z. Gong, E. Gu, S. R. Jin, D. Massoubre, B. Guilhabert, H. X. Zhang, M. D. Dawson, V. Poher, G. T. Kennedy, P. M. W. French, and M. A. A. Neil, “Efficient flip-chip InGaN micro-pixellated light-emitting diode arrays: promising candidates for micro-displays and colour conversion,” J. Phys. D Appl. Phys. 41(9), 094002 (2008).
7. J. Herrnsdorf, J. J. D. McKendry, S. L. Zhang, E. Y. Xie, R. Ferreira, D. Massoubre, A. M. Zuhdi, R. K. Henderson, I. Underwood, S. Watson, A. E. Kelly, E. Gu, and M. D. Dawson, “Active-matrix GaN micro light-emitting diode display with unprecedented brightness,” IEEE Trans. Electron Dev. 62(6), 1918–1925 (2015).
8. L. Dai, B. Zhang, J. Y. Lin, and H. X. Jiang, “Comparison of optical transitions in InGaN quantum well structures and microdisks,” J. Appl. Phys. 89(9), 4951–4954 (2001).
9. H. W. Choi, C. W. Jeon, M. D. Dawson, P. R. Edwards, R. W. Martin, and S. Tripathy, “Mechanism of enhanced light output efficiency in InGaN-based microlight emitting diodes,” J. Appl. Phys. 93(10), 5978–5982 (2003).
10. S. X. Jin, J. Li, J. Y. Lin, and H. X. Jiang, “InGaN/GaN quantum well interconnected microdisk light emitting diodes,” Appl. Phys. Lett. 77(20), 3236–3238 (2000).
11. T. I. Kim, Y. H. Jung, J. Song, D. Kim, Y. Li, H. S. Kim, I. S. Song, J. J. Wierer, H. A. Pao, Y. Huang, and J. A. Rogers, “High-efficiency, microscale GaN light-emitting diodes and their thermal properties on unusual substrates,” Small 8(11), 1643–1649 (2012). [PubMed]
12. Z. Gong, S. R. Jin, Y. J. Chen, J. McKendry, D. Massoubre, I. M. Watson, E. Gu, and M. D. Dawson, “Size-dependent light output, spectral shift, and self-heating of 400 nm InGaN light emitting Diodes,” J. Appl. Phys. 107(1), 013103 (2010).
13. S. Lu, W. Liu, Z. H. Zhang, S. T. Tan, Z. Ju, Y. Ji, X. Zhang, Y. Zhang, B. Zhu, Z. Kyaw, N. Hasanov, X. W. Sun, and H. V. Demir, “Low thermal-mass LEDs: size effect and limits,” Opt. Express 22(26), 32200–32207 (2014). [PubMed]
14. P. F. Tian, J. J. D. McKendry, Z. Gong, B. Guilhabert, I. M. Watson, E. D. Gu, Z. Z. Chen, G. Y. Zhang, and M. D. Dawson, “Size-dependent efficiency and efficiency droop of blue InGaN micro-light emitting diodes,” Appl. Phys. Lett. 101(23), 231110 (2012).
15. F. Olivier, S. Tirano, L. Dupre, B. Aventurier, C. Largeron, and F. Templier, “Influence of size-reduction on the performances of GaN-based micro-LEDs for display application,” J. Luminescence 191(B), 112– 116 (2017).
16. D. Hwang, A. Mughal, C. D. Pynn, S. Nakamura, and S. P. DenBaars, “Sustained high external quantum efficiency in ultrasmall blue III–nitride micro-LEDs,” Appl. Phys. Express 10(3), 032101 (2017).
17. P. Royo, R. P. Stanley, M. Ilegems, K. Streubel, and K. H. Gulden, “Experimental determination of the internal quantum efficiency of AlGaInP microcavity light-emitting diodes,” J. Appl. Phys. 91(5), 2563–2568 (2002).
18. K. A. Bulashevich and S. Yu. Karpov, “Impact of surface recombination on efficiency of III-nitride light-emitting diodes,” Phys. Status Solidi RRL 10(6), 480– 484 (2016).
19. S. S. Konoplev, K. A. Bulashevich, and S. Y. Karpov, “From Large-Size to Micro-LEDs: Scaling Trends Revealed by Modeling,” Phys. Status Solidi., A Appl. Mater. Sci. 11, 1700508 (2017).
20. J. M. Shah, Y. L. Li, Th. Gessmann, and E. F. Schubert, “Experimental analysis and theoretical model for anomalously high ideality factors in AlGaN/GaN p-n junction diodes,” J. Appl. Phys. 94(4), 2627–2630 (2003).
21. D. Zhu, J. R. Xu, A. N. Noemaun, J. K. Kim, E. F. Schubert, M. H. Crawford, and D. D. Koleske, “The origin of the high diode-ideality factors in GaInN/GaN multiple quantum well light-emitting diodes,” Appl. Phys. Lett. 94(8), 081113 (2009).
22. D. P. Han, C. H. Oh, D. G. Zheng, H. Kim, J. I. Shim, K. S. Kim, and D. S. Shin, “Analysis of nonradiative recombination mechanisms and their impacts on the device performance of InGaN/GaN light-emitting diodes,” Jpn. J. Appl. Phys. 54(2), 02BA01 (2015).
23. C. Sah, R. N. Noyce, and W. Shockley, “Carrier Generation and Recombination in P-N Junctions and P-N Junction Characteristics,” Proc. IRE45(9), 1228−1243 (1957).
24. D. P. Han, D. G. Zheng, C. H. Oh, H. Kim, J. I. Shim, D. S. Shin, and K. S. Kim, “Nonradiative recombination mechanisms in InGaN/GaN-based light-emitting diodes investigated by temperature-dependent measurements,” Appl. Phys. Lett. 104(15), 151108 (2014).
25. K. S. Kim, D. P. Han, H. S. Kim, and J. I. Shim, “Analysis of dominant carrier recombination mechanisms depending on injection current in InGaN green light emitting diodes,” Appl. Phys. Lett. 104(9), 091110 (2014).
26. M. Boroditsky, I. Gontijo, M. Jackson, R. Vrijen, E. Yablonovitch, T. Krauss, C. C. Cheng, A. Scherer, R. Bhat, and M. Krames, “Surface recombination measurements on III-V candidate materials for nanostructure light-emitting diodes,” J. Appl. Phys. 87(7), 3497–3504 (2000).
