Open Access
Add to BibSonomy
Abstract
The quantum efficiency of GaN-based micro-light-emitting diodes (micro-LEDs) is of great significance for their luminescence and detection applications. Optimized passivation process can alleviate the trapping of carriers by sidewall defects, such as dangling bonds, and is regarded as an effective way to improve the quantum efficiency of micro-LEDs. In this work, an AlN passivation layer was prepared by atomic layer deposition to improve the electro-optical and photoelectric conversion efficiency in GaN-based micro-LEDs. Compared to conventional Al2O3 passivation, the AlN passivation process has a stronger ability to eliminate the sidewall defects of micro-LEDs due to the homogeneous passivation interface. Our experiments show that the AlN-passivated device exhibits two orders of magnitude lower forward leakage and a smaller ideality factor, which leads to significantly enhanced external quantum efficiency (EQE). For 25*25 μm2 micro-LEDs, the EQE of the AlN-passivated device was 18.3% and 57.7% higher than that of the Al2O3-passivated device in luminescence application and detection application, respectively.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
In the past two decades, the GaN-based micro-light-emitting diode (micro-LED) was widely studied for next-generation display and visible light communication (VLC), due to its higher brightness, lower power consumption, higher frequency response, and longer lifetimes [1–3]. Recently, GaN-based micro-LED was used as a photodetector (PD) and co-integrated with light-emitting unit to realize a combination of reception and transmission, which paves the way of the intelligent display and communication system in the future [4–6]. Whether in luminescence or detection applications, passivation technology is essential to improve the performance of micro-LEDs. Since the top-down process of LEDs by dry etching, micro-LEDs have a larger specific sidewall area than traditional LEDs with large size [7]. A large number of sidewall damage and defects, such as dangling bonds, lead to higher non-radiative recombination, larger leakage current density, and reduced quantum efficiency, which severely deteriorate the micro-LED's luminescence and detection performance [4,8–10]. The most direct purpose of the passivation process is to eliminate sidewall defects and reduce SRH non-radiative recombination [7].
Many passivation schemes have been explored for the fabrication of GaN-based micro-LEDs, such as wet corrosion [11,12], thermal annealing [13,14], and dielectric film passivation [15–19]. Among them, dielectric film passivation is preferable due to its mature technology. Learning from the production schemes of commercial GaN-based LEDs, the SiO2 film grown by plasma enhanced chemical vapor deposition (PECVD) was used for the passivation of micro-LEDs [15,16]. However, the SiO2 grown by PECVD is considered to be unable to completely cover the micro-LED surface due to its poor compactness [13]. In order to meet the passivation requirements, an atomic layer deposition (ALD) technique has been employed as a tool for a denser dielectric film. Wong et al. used ALD to grow 50 nm SiO2 as the passivation layer, and increased the external quantum efficiency (EQE) of the 20*20 µm2 micro-LED by 37.5% [17]. Lee et al. studied the double-layer passivation process of ALD-grown Al2O3/PECVD-grown SiO2, which improved the EQE of 10 µm-sized devices by 22.3% [18]. As far as we know, heterogeneous materials were adopted in almost all reported passivation schemes. However, from the perspective of crystallography, the use of homogeneous materials or lattice-matched materials as the passivation layer is desirable. Based on this principle, GaN should be the best passivation material. However, due to its high doping concentration and low electrical resistivity, GaN is not considered as a suitable passivation material for micro-LEDs. It is known that AlN has been used as passivation layer for GaN power devices, due to its high dielectric constant and smaller lattice mismatch with GaN [20,21]. AlN was also reported to passivate p-GaN surface of LED to increase hole injection and improve the energy conversion efficiency [22]. However, the AlN film was not yet focused on the application of passivating the etched sidewall of micro-LEDs.
In this work, we report an InGaN/GaN-based micro-LED with AlN film passivation on the sidewall. The enhancement of electro-optical and photoelectric conversion efficiency was demonstrated when the fabricated micro-LED was used as light emitting device and as photodetector, respectively.
2. Device fabrication
InGaN/GaN-based blue LED epilayer structures were grown on 4-inch Si (111) substrate by a metal organic chemical vapor deposition (MOCVD) system. The LED structure is composed of a 200 nm-thick AlN nucleation layer, a 1.5 μm-thick undoped GaN layer, a 2.5 μm-thick Si doped n-GaN layer with InGaN/GaN shallow wells (SWs), active layer (9 periods In0.13Ga0.87N/GaN MQWs), a 100 nm-thick p-AlGaN electron blocking layer, and a 250 nm-thick p-GaN layer. After the MOCVD growth, the epitaxy wafer was annealed at 700 °C for 10 minutes under N2 ambient to activate the p-GaN. Then, a thin Ni/Au (5/5 nm) metal stack was deposited on the p-GaN by e-beam evaporation followed by a rapid thermal annealing (RTA) in an atmospheric ambient at 570 °C for 3 minutes, to form the micro-LEDs’ transparent current spreading layer (CSL) as well as the p-type ohmic contact. After that, the micro-LED chips were fabricated by UV photolithography, wet etching with aqua regia and dry etching with an inductively coupled plasma (ICP) system. The etching depth of the micro-LED was about 800 nm. Afterwards, a 30 nm-thick AlN film was deposited by plasma-enhanced ALD at 250 ℃ using trimethyl aluminum (TMA) and NH3 as precursors, followed by depositing 200 nm-thick SiO2 by PECVD. Then, the n-electrode area and p-electrode area were hole opened sequentially by wet etching with buffered oxide etchant (BOE). Finally, the Ti/Au (50/150 nm) metal stack as p-electrodes and n-electrodes were deposited on the transparent CSL and n-GaN, respectively. The micro-LEDs were fabricated in squares with the edge length ranging from 25 µm, 50 µm, 75 µm, and 100 µm. Figure 1(a) shows the cross-section schematic of the fabricated micro-LED. The fabricated micro-LEDs with different size can be clearly observed from the scanning electron microscopy (SEM) images in Fig. 1(b). The cross-section SEM image of the passivated sidewall was presented in Fig. 1(c). The sidewall of micro-LED mesa was covered by the AlN/SiO2 stack, which was in consistent with the design in Fig. 1(a). Figure 1(d) shows the electroluminescence (EL) spectra of devices of different sizes at a current density of 100 A/cm2. The center wavelength is 460 nm, which corresponds to the In composition of the quantum well. As a comparison, reference micro-LEDs passivated with 30 nm-thick ALD-grown Al2O3 and 200 nm-thick PECVD-grown SiO2 were prepared from the same epitaxial wafer. The growth conditions of the Al2O3 film are the same as that of the AlN film, except that the NH3 precursor was replaced with O2.
Fig. 1. (a) Schematics of the micro-LED design; (b) SEM images of the fabricated micro-LEDs with different sizes; (c) Cross-section SEM images of the passivated sidewall of the micro-LEDs; (d) EL characteristics of the fabricated micro-LEDs with different sizes at 100 A/cm2. The inset is the luminous microscope images of the 50*50 μm2 device.
3. Results and discussion
The current density-forward voltage (J-V) characteristics of the micro-LEDs was measured by semiconductor parameter analyzer (Keysight B1500A), shown in Fig. 2(a). The Al2O3-passivated and AlN-passivated micro-LED devices were denoted as Sample-1 and Sample-2, respectively. These two samples present overlapped J-V curve and similar turn-on voltage. As the size of the micro-LEDs increases, the J-V curves of the devices gradually disperse, which has been observed in previous reports [8]. This effect is explained by improved thermal and current spreading inside smaller devices. Figure 2(b) displays the J-V curves by logarithmic current density scale. Before the device was turned on, Sample-1 yielded more than two orders of magnitude higher forward leakage current density than Sample-2, indicating the AlN sidewall passivation was more effective in suppressing leakage current.
Fig. 2. (a) (b) Current density-forward voltage characteristics, (c) ideality factor versus voltage characteristics of the micro-LEDs with different passivation and different sizes; (d) ideality factor distribution of micro-LEDs with different passivation.
The LED ideality factors were extracted from the current-voltage characteristics using Eq. (1)
where n is the ideality factor, q is the elementary charge, k is the Boltzmann constant, T is the temperature in Kelvin, I is the current, and V is the voltage. The calculated ideality factor versus voltage was plotted in Fig. 2(c). It has been reported that the J-V curve can be divided into three ranges according to different conductive processes [23], which were illustrated by green dashed ellipses in Fig. 2(b). Both the shunt resistance at low current range and series resistance at high current range cause a significant deviation from the rectifying behavior of an ideal diode [24]. The space charge region dominates the current-voltage characteristics in the intermediate range, which results in the minimum value of ideality factor. Figure 2(d) plotted the minimum ideality factor of Sample-1 and Sample-2, that was about 2.5 and 1.6, respectively. It is known that ideality factors exceeding 2.0 is suggested to originate from the trap-assisted tunneling and carrier leakage [25]. The difference between the ideality factors in the two samples can be attributed to the AlN sidewall passivation process.The light output power (LOP) of micro-LEDs with different sizes and different passivation methods has been measured at a forward current density ranging from 5 to 1000 mA/cm2. Due to the limitation of our measurement system, we can only obtain the relative LOP versus current density as plotted in Fig. 3(a). It can be seen that the LOP increases nonlinearly with current density. This was attributed to the fact that efficiency droop [9]. For all mesa sizes, the LOP of Sample-2 is higher than that of Sample-1 at the same current density. Figure 3(b) shows the external quantum efficiency (EQE) versus current characteristics for these two micro-LED samples. The inset shows the EQE in the current range of 0∼100 A/cm2. At low current densities (<10 A/cm2), the values of EQE decrease as the device size decreases. This is because defect recombination dominates the EQE performance at a small injection current [9,13]. When the current density gradually increases, small-sized devices show greater EQE values. This can be attributed to the fact that the more uniform current spreading in small-sized devices has become the main factor affecting the EQE performance. Moreover, the EQE peaks decrease first and then increase as the micro-LED mesa shrinks. In addition to the impact of sidewall defects on EQE, the difference in light extraction rate of devices of different sizes is also responsible for this irregular phenomenon. Nevertheless, there are significant differences between the EQE of sample 1 and sample 2. Compared to Sample-1, Sample-2 has improved EQE at different mesa sizes. Furthermore, this improvement has a size effect, as shown in Fig. 3(c). When the size expands from 25*25 to100*100 μm2, the EQE increase decreases from 18.3% to 1.6%.
Fig. 3. (a) LOP and (b) EQE versus current density characteristics of micro-LED with different passivation and sizes. The inset shows the EQE in the 0∼100 A/cm2 range. The dashed lines represent Sample-1 and the solid lines represent Sample-2. (c) The comparison of EQE at 100 A/cm2. The inset shows the EQE increase of sample-2 relative to sample-1 versus micro-LED sizes. (d) The ball and stick models for the passivation interface of the two micro-LEDs.
In order to interpret the difference in electro-optical characteristics between Sample-1 and Sample-2, the ball and stick models were adopted to depict the passivation interface of the micro-LEDs. As shown in Figs. 3(d), the passivation layers were assumed to have crystalline structure, and the Al2O3 and AlN layer bonded with the micro-LED by Ga-O and Ga-N, respectively. According to the lattice constants of related materials (a(Al2O3) = 0.514 nm, a(AlN) = 0.311 nm, a(GaN) = 0.319 nm), the lattice mismatch between Al2O3 and GaN is significantly larger than that between AlN and GaN. In this case, there are more dangling bonds formed at the interface of the micro-LED and the Al2O3 film, which served as the non-radiative recombination center and leakage path in the etched sidewall. In essence, AlN film effectively eliminates the defect states and increases the radiation recombination efficiency near the etched sidewall. Therefore, the overall LOP and EQE of the micro-LED with AlN passivation process were improved.
The photoelectric detection performance of the fabricated micro-LEDs was also investigated. The light response spectrum of the two samples was measured using a system consisting of a xenon lamp, a monochromator, and a semiconductor parameter analyzer. Figure 4(a) shows the light response spectrum of 50*50 μm2 devices. The response spectrum is mainly concentrated in the range of 200 to 500 nm, in which the responsivity of sample 2 is higher than that of sample 1. Among them, the wavelength corresponding to the highest responsivity is around 360 nm. The 365-nm ultraviolet (UV) light was used to excite InGaN/GaN multiple quantum wells to show a sufficient light response. The power of UV lamp was adjusted to 0.4 mW/cm2. The current-voltage characteristics of Sample-1 and Sample-2 was measured under the UV light, as shown in Fig. 4(b). It can be found that, for all sizes, the photocurrent produced by Sample-2 is higher than that produced by Sample-1. In order to eliminate the influence of signal vibration and extract accurate responsivity, we use the fabricated micro-LEDs as self-powered ultraviolet photodetectors for transient light response testing at 0 V bias. As illustrated in Figs. 4(c) and 4(d), the inflow current density of Sample-1 and Sample-2 was measured when the UV lamp periodically switched on (Ilight) and off (Idark). It can be seen that the photocurrent density generated by micro-LEDs of different sizes is inconsistent. This can be attributed to different sidewall defect effects and disproportionate photosensitive area. It is worth noting that the Sample-2 exhibited larger photocurrent at all mesa sizes. Moreover, as the size of the device shrinks, the photocurrent difference between the two samples is getting larger and larger. The corresponding values of EQE were extracted using Eq. (2)
Fig. 4. (a) Response spectrum of the 50*50 μm2 samples at 200∼500 nm. (b) The current-voltage characteristics of micro-LEDs at 365-nm UV with power density of 0.4 mW/cm2. The transient photo-response behavior of the Sample-1 (c) and Sample-2 (d). (e) EQE versus micro-LED sizes when the Sample-1 and Sample-2 used as photodetector. (f) Schematic diagram of the working principle of LED as an UV photodetector.
Based on the aforementioned explanation of the trap defects in the passivation interface, the schematic diagram of micro-LED as an UV photodetector was depicted in Fig. 4(f). The UV radiation excites the InGaN/GaN active region to yield photo-generated carriers, and the carriers drift under the action of the built-in electric field to generate photocurrent. However, part of the carriers is trapped by a large number of sidewall traps during drifting, resulting in a decrease in photocurrent and quantum efficiency. In addition, when the size of the micro-LED increases, the proportion of the sidewall area reduces, and the effect of defect trapping gradually decreases. This can explain the size-dependent passivation effect on the micro-LEDs both in light-emitting and light-detecting application.
4. Conclusion
In summary, we have reported the GaN-based micro-LEDs with AlN passivation layer prepared by ALD. The electro-optical and photoelectric properties were investigated when the fabricated micro-LED were respectively used as a light source and an UV detector. Compared to the conventional Al2O3-passivated devices, the AlN-passivated devices show an improved EQE. For a given device area of 25*25 μm2, the EQE has increased by 18.3% and 57.7% for light-emitting and detection applications, respectively. Moreover, the increase in EQE has a size effect, that is, the smaller the device size, the greater the increase in EQE. These results suggest that the design of AlN passivation is desirable for the future intelligent display and communication system based on micro-LEDs with enhanced electro-optical and photoelectric conversion efficiencies.
Funding
National Natural Science Foundation of China (51861135105, 61874034, 62027818); Natural Science Foundation of Shanghai (18ZR1405000).
Acknowledgments
A portion of this work was done in the Fudan Nano-fabrication Lab.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper may be obtained from the authors upon reasonable request.
References
1. X. Zhou, P. Tian, C.-W. Sher, J. Wu, H. Liu, R. Liu, and H.-C. Kuo, “Growth, transfer printing and colour conversion techniques towards full-colour micro-LED display,” Progress in Quantum Electronics 71, 100263 (2020). [CrossRef]
2. Z. Liu, C. H. Lin, B. R. Hyun, C. W. Sher, and J. H. He, “Micro-light-emitting diodes with quantum dots in display technology,” Light Sci Appl 9(1), 83 (2020). [CrossRef]
3. V. W. Lee, N. Twu, and I. Kymissis, “Micro-LED Technologies and Applications,” Inf. Disp. 32(6), 16–23 (2016). [CrossRef]
4. X. Liu, R. Lin, H. Chen, S. Zhang, Z. Qian, G. Zhou, X. Chen, X. Zhou, L. Zheng, R. Liu, and P. Tian, “High-bandwidth InGaN self-powered detector arrays toward MIMO visible light communication based on micro-LED arrays,” ACS Photonics 6(12), 3186–3195 (2019). [CrossRef]
5. Q. Lyu, H. Jiang, and K. M. Lau, “Monolithic integration of ultraviolet light emitting diodes and photodetectors on a p-GaN/AlGaN/GaN/Si platform,” Opt. Express 29(6), 8358–8364 (2021). [CrossRef]
6. J. Yin, X. An, L. Chen, J. Li, J. Wu, Y. Luo, Q. Wang, H. Yu, and K. H. Li, “Phosphor-based InGaN/GaN white light-emitting diodes with monolithically integrated photodetectors,” IEEE Trans. Electron Devices 68(1), 132–137 (2021). [CrossRef]
7. M. S. Wong, S. Nakamura, and S. P. Denbaars, “Review-progress in high performance III-nitride micro-light-emitting diodes,” ECS J. Solid State Sci. Technol. 9(1), 015012 (2020). [CrossRef]
8. Z. Gong, S. Jin, Y. 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). [CrossRef]
9. P. Tian, J. J. D. Mckendry, Z. Gong, B. Guilhabert, I. M. Watson, E. Gu, Z. Chen, G. 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). [CrossRef]
10. A. Daami, F. Olivier, L. Dupre, F. Henry, and F. Templier, “Electro-optical size-dependence investigation in GaN micro-LED devices,” SID Symp. Dig. Tech. Papers49(1) 790–793 (2018).
11. Y. Yang and X. A. Cao, “Removing plasma-induced sidewall damage in GaN-based light-emitting diodes by annealing and wet chemical treatments,” J. Vac. Sci. Technol. B 27(6), 2337–2341 (2009). [CrossRef]
12. J. M. Lee, C. Huh, D. J. Kim, and S. J. Park, “Dry-etch damage and its recovery in InGaN/GaN multi-quantum-well light-emitting diodes,” Semicond. Sci. Technol. 18(6), 530–534 (2003). [CrossRef]
13. M. S. Wong, C. Lee, D. J. Myers, D. Hwang, J. A. Kearns, T. Li, J. S. Speck, S. Nakamura, and S. P. Denbaars, “Size-independent peak efficiency of III-nitride micro-light-emitting-diodes using chemical treatment and sidewall passivation,” Appl. Phys. Express 12(9), 097004 (2019). [CrossRef]
14. W. H. Choi, G. You, M. Abraham, S. Y. Yu, J. Liu, L. Wang, J. Xu, and S. E. Mohney, “Sidewall passivation for InGaN/GaN nanopillar light emitting diodes,” J. Appl. Phys. 116(1), 013103 (2014). [CrossRef]
15. P. Zuo, B. Zhao, S. Yan, G. Yue, H. Yang, Y. Li, H. Wu, Y. Jiang, H. Jia, J. Zhou, and H. Chen, “Improved optical and electrical performances of GaN-based light emitting diodes with nano truncated cone SiO2 passivation layer,” Opt Quant Electron 48(5), 288 (2016). [CrossRef]
16. P. W. Chen, P. W. Hsiao, P. C. Shen, H. J. Chen, Y. A. Chen, R. H. Horng, and D. S. Wuu, “Improved performance of passive-matrix micro-LED displays using a multi-function passivation structure,” IEEE Photon. J. 12(4), 99 (2020). [CrossRef]
17. M. S. Wong, H. David, A. I. Alhassan, C. Lee, R. Ley, S. Nakamura, and S. P. Denbaars, “High efficiency of III-nitride micro-light-emitting diodes by sidewall passivation using atomic layer deposition,” Opt. Express 26(16), 21324–21331 (2018). [CrossRef]
18. D. H. Lee, J. H. Lee, J. S. Park, T. Y. Seong, and H. Amano, “Improving the leakage characteristics and efficiency of GaN-based micro-light-emitting diode with optimized passivation,” ECS J. Solid State Sci. Technol. 9(5), 055001 (2020). [CrossRef]
19. R. T. Ley, J. M. Smith, M. S. Wong, T. Margalith, and M. J. Gordon, “Revealing the importance of light extraction efficiency in InGaN/GaN microLEDs via chemical treatment and dielectric passivation,” App. Phy. Lett. 116(25), 251104 (2020). [CrossRef]
20. L. Arivazhagan, D. Nirmal, J. Ajayan, D. Godfrey, J. S. Rajkumar, and S. B. Lakshmi, “Enhancement of drain current in AlGaN/GaN HEMT using AlN passivation,” AIP Conf. Proc.2201, 020009 (2019).
21. S. Yang, Z. Tang, M. Hua, Z. Zhang, Y. Lu, and K. J. Chen, “Investigation of sinx and aln passivation for AlGaN/GaN high-electron-mobility transistors: role of interface traps and polarization charges,” IEEE J. Electron Devi. 8, 358–364 (2020). [CrossRef]
22. K. Kim, M. Hua, D. Liu, J. Kim, K. J. Chen, and Z. Ma, “Efficiency enhancement of InGaN/GaN blue light-emitting diodes with top surface deposition of AlN/Al2O3,” Nano Energy 43, 259–269 (2018). [CrossRef]
23. E. F. Schubert, Light-Emitting Diodes - 2nd EditionCambridge University Press, UK, (2006). [CrossRef]
24. D. Zhu, J. 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), 3394 (2009). [CrossRef]
25. K. Mayes, A. Yasan, R. Mcclintock, D. Shiell, S. R. Darvish, P. Kung, and M. Razeghi, “High-power 280 nm AlGaN light-emitting diodes based on an asymmetric single-quantum well,” Appl. Phys. Lett. 84(7), 1046–1048 (2004). [CrossRef]
