Open Access
Add to BibSonomy
Abstract
In this work, the scattering mechanism by nano-patterned sapphire substrate (NPSS) for flip-chip AlGaN-based deep ultraviolet light-emitting diodes (DUV LEDs) has been investigated systematically via three-dimensional finite-difference time-domain (3D FDTD) method. It is found that for the conventional DUV LED with a thick p-GaN layer, the NPSS structure can enhance the light extraction efficiency (LEE) for the transverse magnetic (TM)-polarized light because the TM-polarized light with large incident angles can be scattered into escape cones. However, the LEE for the transverse electric (TE)-polarized light is suppressed by NPSS structure because NPSS structure scatters some TE-polarized light out of the escape cones. Moreover, the highly absorptive p-GaN layer also seriously restricts the scattering efficiency of NPSS structure. Therefore, to reduce the optical absorption, meshed p-GaN structure is strongly proposed to greatly enhance the LEEs for both TM- and TE-polarized light of DUV LEDs grown on NPSS. Compared to the DUV LED with only NPSS structure and that with only meshed p-GaN layer, the LEE for the TE-polarized (TM-polarized) light for DUV LEDs with the combination of NPSS structure and meshed p-GaN structure can be enhanced by 124% (5 times) and 112% (4 times), respectively.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
AlGaN-based ultraviolet light-emitting diodes (UV LEDs) have a wide range of application in the scopes such as water purification, medical treatment and environmental protection [1–4]. However, the external quantum efficiency (EQE) of DUV LED on sapphire substrate is still severely limited by the poor crystal quality and the low light extraction efficiency (LEE). The poor crystal quality is partly due to the large lattice and thermal mismatches between Al-rich AlGaN layer and sapphire substrate. The low LEE arises from a large amount of transverse magnetic (TM)-polarized light [5], the highly absorptive p-GaN ohmic contact layer [6], the total internal reflection (TIR) and Fresnel loss caused by the large refractive index contrast between AlGaN and air [7]. Some advanced technologies have been reported to suppress the threading dislocation density (TDD) in Al-rich AlGaN layer that is grown on sapphire substrate including migration-enhanced metal-organic chemical vapor deposition (ME-MOCVD) [8–10] and pulsed-flow multilayer AlN buffers growth technique [11]. Moreover, various methods such as inclined/roughened sidewalls [12–14], advanced reflectors [15–17], meshed p-type contact electrode [18,19], and photonic crystals [20,21] have ever been investigated for improving the LEE. It has also been proven that nano-patterned sapphire substrate (NPSS) can enhance the internal quantum efficiency (IQE) by epitaxial lateral overgrowth (ELO) and decrease the thickness for the coalesced AlN layer [22–25]. Although it is well known that NPSS can not only enhance the IQE but also effectively improve the LEE for GaN-based blue LEDs [26–29], we are not sure if NPSS can improve the LEE for AlGaN-based DUV LEDs at moment. On one hand, the reported EQE for DUV LEDs grown on NPSS has reached 20% because of the improved LEE [30]. On the other hand, it is also reported that NPSS is not always beneficial for increasing the LEE for the transverse electric (TE)-polarized light [31]. Therefore, it is essential to explore the effect of NPSS on the LEE for DUV LEDs and the scattering mechanism by NPSS in detail.
In this study, we have investigated the effect of NPSS on LEE for DUV LEDs based on three-dimensional finite-difference time-domain (3D FDTD) method. Our results show that, the scattering effect by NPSS causes more light to be absorbed by the p-GaN layer, and thus the LEE for TE-polarized light cannot be enhanced by NPSS structure. Nanoscale meshed p-GaN structure is then strongly proposed to improve the LEE for the TE-polarized light when DUV LEDs are grown on NPSS. It is also found that the combination of NPSS and meshed p-GaN layer can significantly improve the LEE for the TM-polarized light when compared with the conventional flip-chip DUV LEDs grown on flat sapphire substrate.
2. Model and simulation methods
In this work, flip-chip AlGaN-based DUV LEDs on NPSS are designed and investigated. The flip-chip DUV LED model comprises a 2-µm thick sapphire substrate, a 3-µm thick AlN buffer layer, a 1-µm thick n-AlGaN layer, a 100-nm thick multiple quantum well (MQW) layer, a 125-nm thick p-AlGaN layer, a 100-nm thick p-GaN layer and an Al-based metal reflector [see Fig. 1(a)]. The models for the calculated DUV LEDs are similar except that different substrates are utilized, i.e., flat sapphire substrate and NPSS. Specifically, we also show the schematic structure for NPSS in Figs. 1(b1) and 1(b2) for the simulated DUV LEDs. We design concave conical-shaped patterns, which is widely used and can produce good crystal quality and thin coalesced AlN layer according to the report by Dong et al [23]. The diameter of cone and period of array for the NPSS are set to be D and a, respectively as shown in Fig. 1(b1). The filling factor for the conical patterns is defined as the ratio of diameter to period (D/a). The apex angle for the cones is set to 60° according to the wet-etched fabricated NPSS in Refs. [23,32,33].
Fig. 1. (a) Schematic cross-sectional view of 3D FDTD computational models for flip-chip DUV LEDs, (b1) top view and (b2) cross-sectional view for the nano-pattern distribution.
TE- or TM-polarized single dipole source with the peak emission wavelength of 280 nm is placed in the middle of the MQW region [34]. The reflectivity of the Al reflector in the UV spectral range is chosen to be 92% [14]. Due to the limited computational memory, the lateral dimension for the simulation model is set to 8 µm × 8 µm, which is obviously smaller than the actual size of DUV LED. Therefore, we set the boundary conditions for the four lateral boundaries as metal with 100% reflectivity such that the limited lateral dimensions can be deemed as infinity [35]. The bottom boundary condition is set as perfect-matched-layer (PML), which completely absorbs the electromagnetic energy [14]. Furthermore, a non-uniform mesh is applied, and the smallest mesh size is set to 5 nm, which provides good accuracy when calculating LEE. The absorption coefficients for the AlGaN layer, the MQWs and the GaN layer are chosen to be 10 cm−1, 1000 cm−1 and 170000 cm−1, respectively [36,37]. The refractive indices for the sapphire substrate, the AlN layer, the AlGaN layer and the GaN layer are set to be 1.8, 2.16, 2.6 and 2.9, respectively [37]. The LEE is defined as the ratio of the total extracted power collected from the power monitor to the total light power emitted from dipole source [38,39]. To study the light trapping process in sapphire, the power monitor is placed in the sapphire at 500 nm from the AlN/sapphire interface which is larger than the 280 nm wavelength studied in this work. Consequently, it is not affected by electromagnetic waves from AlN/sapphire interface [18]. The power monitor can collect the received power transmission and record the near-field electric field radiation. By carrying out the Fourier Transformation, the near-field electric field is converted to the far-field electric field.
3. Results and discussion
We first investigate the impact of the D/a for NPSS on the LEE of DUV LED. Figure 2(a) illustrates the LEE values in the air (LEEair) of the TE- and TM-polarized light for DUV LED on NPSS at D = 500 nm as a function of D/a. For comparison, LEEair values of the TE- and TM-polarized light for conventional flip-chip DUV LED with flat sapphire substrate (Reference LED) are also plotted as the dotted line in Fig. 2(a). It is well known that the NPSS can enhance the LEE of blue LED due to the strong scattering property. However, it can be found from Fig. 2(a) that the LEEair of the TE-polarized light for DUV LED on NPSS decreases as the increased D/a, and this value is always inferior to that of the Reference LED. This means that the NPSS suppresses rather than improving the LEEair of the TE-polarized light for DUV LEDs, which therefore is different from the observations for GaN-based blue LEDs [28]. However, Fig. 2(a) demonstrates that the LEEair of the TM-polarized light for DUV LEDs on NPSS is higher than that for the Reference LEDs in the probed range of D/a, and increases with the increase of D/a. Compared with Reference LED, the DUV LED on NPSS enhances the LEEair for the TM-polarized light by 62.5% at D/a = 0.7. Figures 2(b) ∼ (e) show the far-field radiation patterns of TE- and TM-polarized light for Reference LED and DUV LED on NPSS at D/a = 0.7. It can be found that for TE- and TM-polarized light, the scattering effect by NPSS is very strong and changes the light distribution obviously. However, when we compare Figs. 2(b) and 2(c), it can be found that the scattering effect by NPSS for the TE-polarized light makes the overall light intensity decrease, which results in a lower LEEair. However, by the comparing the far-field radiation patterns for the TM-polarized light between the Reference LED in Fig. 2(d) and the DUV LED on NPSS in Fig. 2(e), we can get that the overall light intensity of TM-polarized light for DUV LED on NPSS in Fig. 2(e) is slightly higher than the counterpart. As a result, the LEEair of the TM-polarized light for DUV LED on NPSS is higher than that of Reference LED. These results indicate that the scattering effect by NPSS on the TM-polarized light and the TE-polarized light is different.
Fig. 2. (a) LEEair values of TE- and TM-polarized light for Reference LED and DUV LED on NPSS structure as a function of D/a for NPSS. Far-field radiation patterns of TE-polarized light for (b) Reference LED and (c) DUV LED on NPSS at D/a = 0.7. Far-field radiation patterns of TM-polarized light for (d) Reference LED and (e) DUV LED on NPSS at D/a = 0.7.
To explore the underlying reason, we firstly investigate the impact of NPSS on the light propagation into the sapphire substrate before further revealing the light extraction process from the sapphire substrate to the air. Then, a power monitor is placed in the sapphire at the distance of 500 nm from the AlN/sapphire interface [see inset of Fig. 3(a)], by doing so, we are able to calculate the LEE in sapphire (i.e., LEEsapphire) for DUV LEDs on NPSS. The simulated results are shown in Figs. 3(a) and 3(c), which show that both the LEEsapphire values for the TE- and the TM-polarized light in sapphire for DUV LEDs on NPSS are larger than that for the Reference LED, and the number increases as the increased D/a. When we compare LEEair for the TE-polarized light in Fig. 2(a) with the LEEsapphire for the TE-polarized light in Fig. 3(a), it is shown that NPSS significantly increases LEEsapphire for the TE-polarized light at the cost of reducing the LEEair for the TE-polarized light. The investigations into Fig. 3(c) infer that NPSS also makes more TM-polarized light trapped in the sapphire substrate. To further verify it, the cross-sectional profiles of the electric field for the TE-polarized light of Reference LED and DUV LED on NPSS at D/a = 0.7 are shown in Figs. 3(b1) and 3(b2), respectively. It is obviously shown that the electric field intensity in sapphire (the dotted red area) for DUV LEDs on NPSS is stronger than that for Reference LED. The cross-sectional profiles of the electric field for the TM-polarized light of Reference LED and DUV LED on NPSS at D/a = 0.7 are also shown in Figs. 3(d1) and 3(d2), respectively. It can be seen that the electric field intensity distribution in the NPSS is also stronger than that in the planar sapphire substrate. If we refer to Fig. 2(a), we find the abnormal phenomenon that, although the NPSS increases the LEEsapphire values for both the TE- and TM-polarized light by making them propagate into sapphire from the AlGaN layer, the NPSS makes more TM-polarized light and less TE-polarized be extracted into the external space from the sapphire.
Fig. 3. LEEsapphire values of (a) TE- and (c) TM-polarized light for Reference LED and DUV LED on NPSS as a function of D/a for NPSS. Inset of (a): the schematic diagram of computational model for calculating the LEEsapphire. Cross-sectional profiles of the electric field for TE-polarized light of (b1) Reference LED and (b2) DUV LED on NPSS at D/a = 0.7. Cross-sectional profiles of the electric field for the TM-polarized light of (d1) Reference LED and (d2) DUV LED on NPSS at D/a = 0.7.
To explain the above abnormal phenomenon, Fig. 4 presents the angle distribution patterns of the TE- and TM-polarized light that propagates into sapphire substrates for Reference LED and DUV LED on NPSS at D/a = 0.7, respectively. The angle distribution pattern in the sapphire substrate can be obtained by the near-to-far field transformation (NTFF) method [7]. According to Snell’s Law, the critical angle of sapphire/air interface can be calculated by using θ = arcsin(nair/nsapphire) ≈ 34° [see the black dashed line in Figs. 4(a) and 4(b)], where nair and nsapphire are the refractive indices for the air and the sapphire, respectively [40]. Therefore, the LEEair depends on the fraction of light beams with propagating angles smaller than 34°. The fractions of TE-polarized light with propagating angles smaller than 34° can be calculated to be 38.45% and 35.88% for Reference LED and DUV LED on NPSS, respectively. In addition, it can be found that compared with Reference LED, there is more TE-polarized light at large angles (e.g. 35° ∼ 50°) in sapphire for DUV LED on NPSS that cannot escape into external space. Namely, for TE-polarized light, NPSS not only makes more light scattered into sapphire but also causes more light with propagating angles larger than 34° in sapphire. Therefore, the LEEair for the TE-polarized light of DUV LED on NPSS is lower than that for Reference LED although the LEEsapphire in sapphire for DUV LED on NPSS is larger than that for Reference LED. For TM-polarized light, it can be seen from Fig. 4(b) that when compared with Reference LED (8.66%), the fraction (12.19%) of light with propagating angles smaller than 34° for DUV LED on NPSS is larger. Hence, the LEEair and LEEsapphire of TM-polarized light for DUV LED on NPSS are larger than that for Reference LED.
Fig. 4. Angular distribution patterns of (a) TE- and (b) TM-polarized incident light in sapphire substrate for Reference LED and DUV LED on NPSS at D/a = 0.7.
To further understand the scattering effect of NPSS on the light with different incident angles, a simplified model with the plan wave as incident light source is simulated and the modeled structure is shown in the inset of Fig. 5. The inclination angle of incident light source is α and lateral boundary conditions are period boundary. Figure 5 shows the LEEair values for DUV LED on NPSS and Reference LED versus incident angle α of plane wave source. As shown in Fig. 5, the LEEair for Reference LED quickly decreases as the increased incident angle when the angle of emitted light is beyond 23°. Note, the critical angle of AlGaN/air is 23°. In addition, the LEEair for DUV LED on NPSS is significantly lower than that for Reference LED when the inclination angle α is smaller than 23°. It means that NPSS structure makes partial light with emitting angle smaller than the critical value scatter out of the escape cone. As a result, these light beams will be totally internally reflected by the sapphire/air interface and will be finally absorbed by the p-GaN and other materials. The TE-polarized light mainly propagates perpendicularly to the out-light plane, such that the emitting angle of most TE-polarized light is within the escape cone. However, the NPSS will scatter some TE-polarized light that has the emitting angles smaller than the critical angle out of the escape cone. The result is consistent with Fig. 4(a) when compared with the planar substrate, such that the NPSS structure decreases the LEEair of the TE-polarized light when the propagating angle is within 34°. Therefore, the scattering effect of NPSS makes the LEEair for the TE-polarized light decrease. Furthermore, it can be found from Fig. 5 that when the emitting angles are from 23° to 35°, the LEEair for the DUV LED on NPSS is larger than that for Reference LED, which implies that some light with the emitting angle beyond the critical angle can be scattered into the escape cone by NPSS. The propagation direction of the TM-polarized light is mainly parallel to the out-light plane. Namely, the emitting angles for most of the TM-polarized light are beyond the critical angle. However, NPSS can scatter some of the TM-polarized light that has the emitting angles within 23° ∼ 35° into the escape cone, which agrees well with Fig. 4(b) when compared with the planar substrate. Hence, the DUV LED on NPSS increases the LEEair of the TM-polarized light that has propagating angle within 34° in sapphire. As a result, the scattering effect by the NPSS helps to increase the LEEair for the TM-polarized light. Undoubtedly, the fraction of the TM-polarized light with emitting angles within 23° ∼ 35° is very low because the TM-polarized light propagates laterally. Meanwhile, the LEEair of these light beams is not more than 20% as shown in Fig. 5. Thus, the LEEair for the TM-polarized light of DUV LED on NPSS is still very low, which is smaller than 1% as shown in Fig. 2(a).
Fig. 5. LEEair of Reference LED and DUV LED on NPSS at D/a = 0.7 versus incident angle α of plane wave source. Inset: the schematic diagram of computational model with the plan wave as incident light source.
According to the previous discussions, it can be concluded that the LEEair is difficult to be enhanced merely through the scattering effect by NPSS. For DUV LEDs with thick p-GaN film as the ohmic contact layer, the light that cannot be scattered out into air by the NPSS will be finally absorbed by the p-GaN layer and other materials. Therefore, the effect of the p-GaN layer thickness on the LEEair for DUV LEDs on NPSS is further investigated. Figures 6(a) and 6(b) illustrate the LEEair values of the TE- and TM-polarized light as a function of the p-GaN layer thickness for Reference LED and DUV LED on NPSS. As expected, LEEair values of TE- and TM-polarized light for DUV LED on NPSS increase as the decreased p-GaN layer thickness. Figure 6(a) further shows that the LEEair of the TE-polarized light for DUV LED on NPSS can be increased to 25% when the p-GaN layer thickness is 0 nm, and this number is 15% for reference LED. Figure 6(b) indicates that the LEEair of the TM-polarized light for DUV LED on NPSS can be increased to 5% when the p-GaN layer thickness is set to 0 nm, and this number is 0.4% for reference LED. Moreover, when the p-GaN layer thickness is set to 0 nm thick, the LEEair values of the TE-polarized light for the DUV LED on NPSS and the Reference LED are 4.6 and 2.6 times larger than that when the p-GaN layer is set to 100 nm thick, respectively for both devices; the LEEair values of the TM-polarized light for the DUV LED on NPSS and the Reference LED are 11 and 1.3 times larger than that when the p-GaN layer is set to 100 nm thick, respectively for both devices. These results manifest that compared with Reference LED, the LEEair values of TE- and TM-polarized light for the DUV LED on NPSS is more sensitive to the p-GaN layer thickness as shown in Figs. 6(a) and 6(b). That is to say, the p-GaN layer thickness can even more impact the LEEair for DUV LEDs on NPSS than that for Reference LED. It is because when the absorption in DUV LEDs is low, the light in DUV LED can be scattered many times by NPSS leading to the increased escaping probability for light. The cross-section electrical field distributions for TM-polarized light for Reference LED (DUV LED on NPSS) with the 0-nm thick p-GaN and with the 100-nm thick p-GaN layers are shown in Figs. 6(d1) and 6(d2), Figs. 6(e1) and 6(e2), respectively. It can be found that the electrical field intensity in the whole LED with 0-nm thick p-GaN layer [see Figs. 6(d1) and 6(e1)] is stronger than that with 100-nm thick p-GaN layer [see Figs. 6(d2) and 6(e2)], which indicates that the absence of strong optical absorption makes the light experience multiple reflections from the various interfaces of the LED with 0-nm thick p-GaN layer. Figure 6(c) shows the electric field intensity in air versus the position in Figs. 6(d1), 6(d2), 6(e1) and 6(e2). Note, the electric field intensity in the right half for the simulated structures is shown in the Fig. 6(c) because of the symmetric optical profiles in Figs. 6(d) and 6(e). It can be seen that although the absorption caused by the p-GaN layer is eliminated, the LEEair of TM-polarized light for Reference LED is not obviously improved [see the blue line in Fig. 6(c)]. This is attributed to the fact that the lateral propagation of TM-polarized light can easily form guided modes in the planar sapphire substrate. Thus, thinning the p-GaN layer for Reference LED does not obviously improve the LEEair for TM-polarized light. However, the optical intensity for DUV LED on NPSS without p-GaN layer (the black solid line) is significantly increased than that for Reference LED (the blue solid line). The multiple scattering by the NPSS structure can change the propagation paths for some photons and can eventually enter the escape cones. Therefore, we suggest fabricating DUV LEDs on NPSS substrate with properly thin p-GaN layer to enhance the LEEair.
Fig. 6. LEEair values of (a) TE- and (b) TM-polarized light as a function of the p-GaN layer thickness for Reference LED and DUV LEDs on NPSS. (c) Optical intensity of extracted light of the TM-polarized light for the two devices versus the position. Cross-section electrical field distributions of the TM-polarized light for Reference LEDs with (d1) 0-nm and (d2) 100-nm thick p-GaN layers. Cross-section electrical field distributions of the TM-polarized light for DUV LEDs on NPSS with (e1) 0-nm and (e2) 100-nm thick p-GaN layers.
However, at present, it is less possible to remove the p-GaN layer from the current DUV LED architectures. A cost-effective method to suppress the optical absorption by the p-GaN layer is to utilize meshed p-GaN structures [18,19]. Here, the meshed p-GaN layer is combined with NPSS for DUV LEDs and in-depth investigation is conducted. The p-GaN layer is designed into nanorods with a triangular array which are embedded in the Al reflector. The schematic diagram of meshed p-GaN structure is shown in the inset of Fig. 7(a). The p-GaN layer thickness is set to 100 nm, and the diameter of p-GaN nanorod is fixed to be 500 nm. Here, the filling factor for meshed p-GaN structure is defined as the ratio of the nanorod diameter to the array period (D/a). A larger filling factor is enabled by using smaller spacing among the nanorods. Smaller spacing may better spread the electric current but also increase the optical absorption. The LEEair values of the TE- and TM-polarized light as a function of the filling factor for DUV LEDs with meshed p-GaN structure are shown in Fig. 7. It can be seen from Figs. 7(a) and 7(b) that, for Reference LED with meshed p-GaN structure, the LEEair values for the TE- and TM-polarized light (the blue solid line) increase and then decrease with the increased D/a of the meshed p-GaN structure. It is because the scattering effect by the meshed p-GaN structure increases first and then decreases with the increased of D/a, which is consistent with the results in Ref. [18]. However, for DUV LEDs on NPSS and with meshed p-GaN structure, the LEEair values for TE- and TM-polarized light (the black solid line) monotonically decrease as the increased D/a. It can be inferred that the scattering effect by the meshed p-GaN structure and NPSS is monotonically decrease as the increased D/a due to the increased optical absorption. It is because the utilization of the meshed p-GaN structure with small D/a can significantly suppress the optical absorption, and by doing so the function of the NPSS in scattering the light can be fully presented. In addition, we can see from Figs. 7(a) and 7(b) that even Reference LED utilizing the meshed p-GaN structure, the LEEair values for the TE- and TM-polarized light are always lower than that for the counterpart DUV LEDs on NPSS at each D/a for meshed p-GaN structure. Such small LEEair for Reference LED is well attributed to the poor scattering effect by the planar substrate. Therefore, we can infer that scattering effect by meshed p-GaN structure and NPSS is stronger than that by only meshed p-GaN structure. To further confirm it, the far-field radiation patterns of TE-polarized light for DUV LED on NPSS and Reference LED at D/a = 0.7 both of which have meshed p-GaN structures are shown in Figs. 7(c) and 7(d), respectively. It can be observed that, because of the higher LEEair, the electric field intensity for DUV LED on NPSS is stronger than that for Reference LED. Moreover, the light distribution for the DUV LED on NPSS is more uniform than that for Reference LED, which indicates that the coupled scattering effect by both the NPSS and the meshed p-GaN structure is much stronger than that by merely using the meshed p-GaN structure. As a result, we can conclude that the LEEair for DUV LEDs can be significantly enhanced by combining the NPSS and the meshed p-GaN structures, which is a have-to technical approach, e.g., when D/a is set to 0.2, the LEEair of the TE-polarized light for the DUV LED on NPSS and with meshed p-GaN structure can be enhanced by 112% and 124% when compared with Reference LED with meshed p-GaN structure and the DUV LED on NPSS without meshed p-GaN structure, respectively. These numbers are 4 and 5 for the LEEair of the TM-polarized light according to Fig. 7(b).
Fig. 7. LEEair values of (a) TE- and (b) TM-polarized light for DUV LEDs with meshed p-GaN structure as a function of the D/a for meshed p-GaN structure. Inset of (a): top view and cross-sectional view for the meshed p-GaN structure. Far-field radiation patterns of TE-polarized light for (c) DUV LED on NPSS and (d) Reference LED at D/a = 0.7 for meshed p-GaN structure.
4. Conclusions
In conclusion, we have numerically investigated the effect of nano-patterned sapphire substrate (NPSS) structure on the light extraction efficiency (LEE) for AlGaN-based flip-chip DUV LEDs. For DUV LEDs with a thick p-GaN layer, NPSS cannot fully enhance the LEE for the TE-polarized light, and this is because the scattering effect of NPSS makes more TE-polarized light absorbed by the optically absorptive p-GaN layer. For the case of TM-polarized light, the NPSS can scatter some light with large incident angles into escape cones. However, the enhanced LEE of the TM-polarized light for DUV LEDs on NPSS is not that obvious because the strongly absorptive p-GaN layer makes negative impact. Therefore, the meshed p-GaN structure is proposed for DUV LEDs on NPSS. The NPSS can fully show the positive function for scattering the light. As a result, the LEEs for both the TE-polarized light and the TM-polarized light are remarkably increased.
Funding
National Natural Science Foundation of China (61975051, 62074050); Natural Science Foundation of Hebei Province (F2020202030); research fund by State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology (No. EERI_PI2020008); joint research project for Tunghsu Group and Hebei University of Technology (HI1909).
Disclosures
The authors declare no conflicts of interest.
References
1. D. B. Li, K. Jiang, X. J. Sun, and C. L. Guo, “AlGaN photonics: recent advances in materials and ultraviolet devices,” Adv. Opt. Photonics 10(1), 43–110 (2018). [CrossRef]
2. M. Kneissl, T.-Y. Seong, J. Han, and H. Amano, “The emergence and prospects of deep-ultraviolet light-emitting diode technologies,” Nat. Photonics 13(4), 233–244 (2019). [CrossRef]
3. S. J. Zhou, H. H. Xu, H. P. Hu, C. Q. Gui, and S. Liu, “High quality GaN buffer layer by isoelectronic doping and its application to 365 nm InGaN/AlGaN ultraviolet light-emitting diodes,” Appl. Surf. Sci. 471, 231–238 (2019). [CrossRef]
4. M. Kneissl, T. Kolbe, C. Chua, V. Kueller, N. Lobo, J. Stellmach, A. Knauer, H. Rodriguez, S. Einfeldt, Z. Yang, N. M. Johnson, and M. Weyers, “Advances in group III-nitride-based deep UV light-emitting diode technology,” Semicond. Sci. Technol. 26(1), 014036 (2011). [CrossRef]
5. P. Zhao, L. Han, M. R. McGoogan, and H. P. Zhao, “Analysis of TM mode light extraction efficiency enhancement for deep ultraviolet AlGaN quantum wells light-emitting diodes with III-nitride micro-domes,” Opt. Mater. Express 2(10), 1397–1406 (2012). [CrossRef]
6. J. Yun and H. Hirayama, “Investigation of the light-extraction efficiency in 280 nm AlGaN-based light-emitting diodes having a highly transparent p-AlGaN layer,” J. Appl. Phys. 121(1), 013105 (2017). [CrossRef]
7. Q. Y. Yue, K. Li, F. M. Kong, J. Zhao, and M. Liu, “Analysis on the effect of amorphous photonic crystals on light extraction efficiency enhancement for GaN-based thin-film-flip-chip light-emitting diodes,” Opt. Commun. 367, 72–79 (2016). [CrossRef]
8. D. Lee, J. W. Lee, J. Jang, I.-S. Shin, L. Jin, J. H. Park, J. Kim, J. Lee, H.-S. Noh, Y. Kim, Y. Park, G.-D. Lee, Y. Park, J. K. Kim, and E. Yoon, “Improved performance of AlGaN-based deep ultraviolet light-emitting diodes with nano-patterned AlN/sapphire substrates,” Appl. Phys. Lett. 110(19), 191103 (2017). [CrossRef]
9. R. Jain, W. Sun, J. Yang, M. Shatalov, X. Hu, A. Sattu, A. Lunev, J. Deng, I. Shturm, Y. Bilenko, R. Gaska, and M. S. Shur, “Migration enhanced lateral epitaxial overgrowth of AlN and AlGaN for high reliability deep ultraviolet light emitting diodes,” Appl. Phys. Lett. 93(5), 051113 (2008). [CrossRef]
10. J. P. Zhang, H. M. Wang, M. E. Gaevski, C. Q. Chen, Q. Fareed, J. W. Yang, G. Simin, and M. Asif Khan, “Crack-free thick AlGaN grown on sapphire using AlN/AlGaN superlattices for strain management,” Appl. Phys. Lett. 80(19), 3542–3544 (2002). [CrossRef]
11. H. Hirayama, T. Yatabe, N. Noguchi, T. Ohashi, and N. Kamata, “231–261 nm AlGaN deep-ultraviolet light-emitting diodes fabricated on AlN multilayer buffers grown by ammonia pulse-flow method on sapphire,” Appl. Phys. Lett. 91(7), 071901 (2007). [CrossRef]
12. J. W. Lee, J. H. Park, D. Y. Kim, E. F. Schubert, J. Kim, J. Lee, Y. I. Kim, Y. Park, and J. K. Kim, “Arrays of truncated cone AIGaN deep-ultraviolet light-emitting diodes facilitating efficient outcoupling of in-plane emission,” ACS Photonics 3(11), 2030–2034 (2016). [CrossRef]
13. Y. H. Zhang, R. L. Meng, Z. H. Zhang, Q. Shi, L. P. Li, G. X. Liu, and W. G. Bi, “Effects of inclined sidewall structure with bottom metal air cavity on the light extraction efficiency for AlGaN-based deep ultraviolet light-emitting diodes,” IEEE Photonics J. 9(5), 1600709 (2017). [CrossRef]
14. Y. H. Zhang, Y. X. Zheng, R. L. Meng, C. Sun, K. K. Tian, C. Geng, Z. H. Zhang, G. X. Liu, and W. G. Bi, “Enhancing both TM- and TE-polarized light extraction efficiency of AlGaN-based deep ultraviolet light-emitting diode via air cavity extractor with vertical sidewall,” IEEE Photonics J. 10(4), 8200809 (2018). [CrossRef]
15. X. X. Liu, Y. Mou, H. Wang, R. L. Liang, X. Z. Wang, Y. Peng, and M. X. Chen, “Enhanced light extraction of deep ultraviolet light-emitting diodes by using optimized aluminum reflector,” Appl. Opt. 57(25), 7325–7328 (2018). [CrossRef]
16. S. J. Zhou, X. T. Liu, Y. L. Gao, Y. C. Liu, M. L. Liu, Z. Y. Liu, C. Q. Gui, and S. Liu, “Numerical and experimental investigation of GaN-based flip-chip light-emitting diodes with highly reflective Ag/TiW and ITO/DBR Ohmic contacts,” Opt. Express 25(22), 26615–26627 (2017). [CrossRef]
17. S. J. Zhou, X. T. Liu, H. Yan, Z. W. Chen, Y. C. Liu, and S. Liu, “Highly efficient GaN-based high-power flip-chip light-emitting diodes,” Opt. Express 27(12), A669–A692 (2019). [CrossRef]
18. Y. X. Zheng, Y. H. Zhang, J. Zhang, C. Sun, C. S. Chu, K. K. Tian, Z. H. Zhang, and W. G. Bi, “Effects of meshed p-type contact structure on the light extraction effect for deep ultraviolet flip-chip light-emitting diodes,” Nanoscale Res. Lett. 14(1), 149 (2019). [CrossRef]
19. N. Lobo, H. Rodriguez, A. Knauer, M. Hoppe, S. Einfeldt, P. Vogt, M. Weyers, and M. Kneissl, “Enhancement of light extraction in ultraviolet light-emitting diodes using nanopixel contact design with Al reflector,” Appl. Phys. Lett. 96(8), 081109 (2010). [CrossRef]
20. J. G. Kim, C. H. Hsich, H. J. Choi, J. Gardener, B. Singh, A. Knapitsch, P. Lecoq, and G. Barbastathis, “Conical photonic crystals for enhancing light extraction efficiency from high refractive index materials,” Opt. Express 23(17), 22730–22739 (2015). [CrossRef]
21. Y. Kashima, N. Maeda, E. Matsuura, M. Jo, T. Iwai, T. Morita, M. Kokubo, T. Tashiro, R. Kamimura, Y. Osada, H. Takagi, and H. Hirayama, “High external quantum efficiency (10%) AlGaN-based deep-ultraviolet light-emitting diodes achieved by using highly reflective photonic crystal on p-AlGaN contact layer,” Appl. Phys. Express 11(1), 012101 (2018). [CrossRef]
22. P. Dong, J. C. Yan, Y. Zhang, J. X. Wang, J. P. Zeng, C. Geng, P. P. Cong, L. L. Sun, T. B. Wei, L. X. Zhao, Q. F. Yan, C. G. He, Z. X. Qin, and J. M. Li, “AlGaN-based deep ultraviolet light-emitting diodes grown on nano-patterned sapphire substrates with significant improvement in internal quantum efficiency,” J. Crystal Growth 395, 9–13 (2014). [CrossRef]
23. P. Dong, J. C. Yan, J. X. Wang, Y. Zhang, C. Geng, T. B. Wei, P. P. Cong, Y. Y. Zhang, J. P. Zeng, Y. D. Tian, L. L. Sun, Q. F. Yan, J. M. Li, S. F. Fan, and Z. X. Qin, “282-nm AlGaN-based deep ultraviolet light-emitting diodes with improved performance on nano-patterned sapphire substrates,” Appl. Phys. Lett. 102(24), 241113 (2013). [CrossRef]
24. T.-Y. Wang, C.-T. Tasi, C.-F. Lin, and D.-S. Wuu, “85% internal quantum efficiency of 280-nm AlGaN multiple quantum wells by defect engineering,” Sci. Rep. 7(1), 14422 (2017). [CrossRef]
25. H. L. Chang, Z. L. Chen, W. J. Li, J. C. Yan, R. Hou, S. Y. Yang, Z. Q. Liu, G. D. Yuan, J. X. Wang, J. M. Li, P. Gao, and T. B. Wei, “Graphene-assisted quasi-van der Waals epitaxy of AlN film for ultraviolet light emitting diodes on nano-patterned sapphire substrate,” Appl. Phys. Lett. 114(9), 091107 (2019). [CrossRef]
26. H.-Y. Wang, Z.-T. Lin, J.-L. Han, L.-Y. Zhong, and G.-Q. Li, “Design of patterned sapphire substrates for GaN-based light-emitting diodes,” Chin. Phys. B 24(6), 067103 (2015). [CrossRef]
27. M. Ohya, K. Naniwae, T. Kondo, A. Suzuki, M. Mori, T. Kitano, A. Usui, and S. Kamiyama, “Improvement of vertical light extraction from GaN-based LEDs on moth-eye patterned sapphire substrates,” Phys. Status Solidi 212(5), 935–940 (2015). [CrossRef]
28. Y. H. Zhang, T. B. Wei, J. X. Wang, D. Lan, Y. Chen, Q. Hu, H. X. Lu, and J. M. Li, “The improvement of GaN-based light-emitting diodes using nanopatterned sapphire substrate with small pattern spacing,” AIP Adv. 4(2), 027123 (2014). [CrossRef]
29. X. H. Huang, J. P. Liu, J. J. Kong, H. Yang, and H. B. Wang, “High-efficiency InGaN-based LEDs grown on patterned sapphire substrates,” Opt. Express 19(S4), A949–A955 (2011). [CrossRef]
30. T. Takano, T. Mino, J. Sakai, N. Noguchi, K. Tsubaki, and H. Hirayama, “Deep-ultraviolet light-emitting diodes with external quantum efficiency higher than 20% at 275 nm achieved by improving light-extraction efficiency,” Appl. Phys. Express 10(3), 031002 (2017). [CrossRef]
31. Y. K. Ooi and J. Zhang, “Light extraction efficiency analysis of flip-chip ultraviolet light-emitting diodes with patterned sapphire substrate,” IEEE Photonics J. 10(4), 8200813 (2018). [CrossRef]
32. L. S. Zhang, F. J. Xu, J. M. Wang, C. G. He, W. W. Guo, M. X. Wang, B. W. Sheng, L. Lu, Z. X. Qin, X. Q. Wang, and B. Shen, “High-quality AlN epitaxy on nano-patterned sapphire substrates prepared by nano-imprint lithography,” Sci. Rep. 6(1), 35934 (2016). [CrossRef]
33. C. Geng, L. Zheng, H. J. Fang, Q. F. Yan, T. B. Wei, Z. B. Hao, X. Q. Wang, and D. Z. Shen, “Fabrication of volcano-shaped nano-patterned sapphire substrates using colloidal self-assembly and wet chemical etching,” Nanotechnology 24(33), 335301 (2013). [CrossRef]
34. T. B. Wei, K. Wu, D. Lan, Q. F. Yan, Y. Chen, C. X. Du, J. X. Wang, Y. P. Zeng, and J. M. Li, “Selectively grown photonic crystal structures for high efficiency InGaN emitting diodes using nanospherical-lens lithography,” Appl. Phys. Lett. 101(21), 211111 (2012). [CrossRef]
35. Y. H. Zhang, J. Zhang, Y. X. Zheng, C. Sun, K. K. Tian, C. S. Chu, Z. H. Zhang, J. G. X. Li, and W. G. Bi, “The effect of sapphire substrates on omni-directional reflector design for flip-chip near-ultraviolet light-emitting diodes,” J. Cryst. Growth 11(1), 8200209 (2019). [CrossRef]
36. H. Y. Ryu, I. G. Choi, H. S. Choi, and J. I. Shim, “Investigation of light extraction efficiency in AlGaN deep-ultraviolet light-emitting diodes,” Appl. Phys. Express 6(6), 062101 (2013). [CrossRef]
37. H. Y. Ryu, “Large enhancement of light extraction efficiency in AlGaN-based nanorod ultraviolet light-emitting diode structures,” Nanoscale Res. Lett. 9(1), 58 (2014). [CrossRef]
38. T. B. Wei, X. L. Ji, K. Wu, H. Y. Zheng, C. X. Du, Y. Chen, Q. F. Yan, L. X. Zhao, Z. C. Zhou, J. X. Wang, and J. M. Li, “Efficiency improvement and droop behavior in nanospherical-lens lithographically patterned bottom and top photonic crystal InGaN/GaN light-emitting diodes,” Opt. Lett. 39(2), 379–382 (2014). [CrossRef]
39. Y. H. Zhang, T. B. Wei, Z. Xiong, L. Shang, Y. D. Tian, Y. Zhao, P. Y. Zhou, J. X. Wang, and J. M. Li, “Enhanced optical power of GaN-based light-emitting diode with compound photonic crystals by multiple-exposure nanosphere-lens lithography,” Appl. Phys. Lett. 105(1), 013108 (2014). [CrossRef]
40. M. R. Krames, O. B. Shchekin, R. Mueller-Mach, G. O. Mueller, L. Zhou, G. Harbers, and M. G. Craford, “Status and future of high-power light-emitting diodes for solid-state lighting,” J. Disp. Technol. 3(2), 160–175 (2007). [CrossRef]
