1. Introduction

With the advantages of longer lifetime, smaller size and higher energy efficiency, LEDs are fast replacing traditional light sources in numerous applications, such as illumination and displays. Furthermore, LEDs can also be used for VLC, which achieves LED illumination and optical data transmission simultaneously [1–3]. Compared with traditional radio frequency wireless communication, VLC is more secure, without electromagnetic interference or license restriction. However, till now the optical modulation bandwidth of conventional commercial LEDs is still quite low [4, 5], which restricts the further development of VLC.

Many techniques have been proposed to improve the optical modulation bandwidth of GaN based LEDs, such as optimization of the epilayer structure [6, 7], improving the current density tolerance to increase the injected carrier concentration [8–10], or using cavity-structure [11, 12]. Utilizing the SP is another effective approach to improve the modulation bandwidth by increasing the spontaneous emission rate, which creates a new energy transition channel of electron-hole pairs in LEDs by the QW-SP coupling [13, 14]. Because the density of states of SP mode is much large, the QW-SP coupling rate will be very fast, and increase the spontaneous emission rate will increase accordingly. An enhancement of spontaneous emission rate (average Purcell factor) exceeding 1000 times has been reported in quantum dot structures [15]. If the scattering efficiency of SPs is high enough compared to the internal quantum efficiency (IQE) of QWs [16–18], the luminous efficacy can also be increased simultaneously. Therefore, the QW-SP coupled LED is an attractive alternative solution for VLC, which will increase both the optical modulation bandwidth and the optical power.

However, till now almost all the reported results for QW-SP coupled LEDs with large average Purcell factor (more than 30) are only based on the QW structures without p-GaN [13, 19], which cannot be used in real LEDs for application. Since the SP is an evanescent wave that exponentially decays with increasing the distance from the metal surface, for QW-SP coupled LEDs, only thin p-GaN layer or nano-structures can guarantee both efficient hole injection and QW-SP coupling, resulting in a significant decrease of average Purcell factor to several folds compared with the bare QW structure [20–22]. In addition, most of studies using QW-SP coupling focused on increasing the luminous efficacy for GaN-based LEDs but few works on the modulation bandwidth. Till now, the enhancement of the modulation bandwidth for GaN-based LEDs by surface plasmon is still less than two times [23–25].

In order to have a larger average Purcell factor to enhance the modulation bandwidth, in this study, we fabricated the QW-SP coupled LEDs based on the nanorod structure [23]. Different Ag nanoparticles were deposited between the nanorods and laterally proximity to the MQWs region to modify the SP resonant wavelength. The influences of the carrier distribution on enhancement of the spontaneous emission rate were carefully investigated by changing the wavelength of the excitation light sources. The photoluminescence (PL) and the time-resolved photoluminescence (TRPL) results show that the QW-SP coupling is more significant for shorter wavelength excitation due to the increased proportion of SP-coupled carriers. For the electrically injected LED, in order to increase the spontaneous emission rate, the active structure and carrier distribution need to be designed to increase the SP-coupling proportion by using the single QW or MQW structure that confines carriers in the topmost QW, which can be an effective way to realize high speed QW-SP coupled LEDs.

2. Experiment

GaN-based LED epilayers were grown on a c-axial sapphire (0001) substrate using metal organic chemical vapor deposition (MOCVD). The epitaxial structure consists of 18 pairs of InGaN/GaN (3nm/14nm) MQWs sandwiched between a 300 nm thick Mg-doped p-type layer and a 3 μm thick layer of Si-doped n-type GaN. Figure 1(a) shows the schematic diagram of the GaN-based SP-LEDs. For the nanorod array structure fabrication, firstly, a 100 nm thick SiO₂ layer was deposited on top of the p-GaN using plasma-enhanced CVD (PECVD) followed by the deposition of 7 nm thick Ni layer using an e-beam evaporator. A rapid thermal annealing (RTA) process was then implemented to form self-assembled Ni metal clusters as etching masks, then the SiO₂ layer, p-GaN and active layer were etched using inductively coupled plasma (ICP), where the etched depth of the active layer was ~20 nm. After that, a 10 nm thick HfO₂ layer was used to cover the nanorod and active layer before coating the Ag film, which can effectively separate the Ag from the p-GaN and the MQWs. After the RTA under N₂ at 550 °C, Ag nanoparticles were obtained. To fabricate different sizes of Ag nanoparticles, 3, 6 and 9 nm thick Ag films were deposited as LED A, B and C, respectively. The nanorod LED without Ag nanoparticles has also been fabricated as the reference sample.

 

Fig. 1 (a) Schematic diagram of the GaN-based SP-LEDs structure. The inset shows the cross section view. (b) SEM image for the SP-LED with Ag nanoparticles obtained from the annealed 9 nm Ag film. (c) The Ag nanoparticles diameter distributions for different designed Ag film thicknesses.

Download Full Size | PDF

During the PL measurements, two excitation light sources were used, including a 405 nm laser diode and a 325 nm He-Cd laser. Both of them have a power of ~40 mW and a pump spot diameter of ~50 μm. The spectral resolution of the PL measurement was set as 0.05 nm. To perform TRPL measurements, two picosecond pulsed sources were used as well. The values for the pulse width, repetition rate and power were chosen as 56.9 ps, 10 MHz and 500 μW for the 405 nm laser source and 849.9 ps, 10 MHz and 40 μW for the 360 nm LED source. A photomultiplier tube (R928P) was used as the detector. Both PL and TRPL were measured for the LED wafers with a configuration of top excitation and detection at room temperature.

3. Results and discussions

3.1 SP characterizations of Ag nanoparticles

Figure 1(b) shows the SEM image of the nanorod LED with Ag nanoparticles annealed from the 9 nm Ag thin film (LED C). The average radius of the nanorods is about 100 nm, comparable to the penetration depth of the SP, which can guarantee a larger proportion of MQWs region coupled to SPs [14, 20]. The etching depth of MQWs is roughly about 20 nm to make Ag nanoparticles laterally proximity to the topmost QW. The 10 nm thick HfO₂ layer covers the nanorods uniformly, which can isolate the Ag nanoparticles from the GaN nanorods. Furthermore, the thickness of the HfO₂ layer also determines the coupling distance between SPs and MQWs at the same time, which cannot only ensure the effective energy coupling but also avoid the quenching effect [26]. Figure 1(c) shows the corresponding Ag nanoparticles diameter distribution for three different designed Ag film thicknesses. As previously reported [22], with the thickness increasing, both the average diameter and density of Ag nanoparticles increases, which is about 29 nm (20.1 μm⁻²), 35 nm (37.6 μm⁻²) and 43 nm (49.8 μm⁻²), respectively.

Using the nanorod LED without Ag nanoparticles as the baseline, the transmission spectra without the influence of the GaN nano-structure can be obtained for Ag nanoparticles annealed from 3, 6 and 9 nm Ag thin film, as shown in Fig. 2(a). With the Ag thin film thicknesses increasing, the strengths of the resonant absorption increase together with peak red-shifts from 505 nm to 546 nm. The red-shifts of the resonant wavelength are mainly due to the increase of the Ag nanoparticles size [27, 28]. However, the resonant wavelength difference (8 nm) between LED A and LED B is much smaller than that (33 nm) between LED B and LED C.

 

Fig. 2 (a) Transmission spectra of Ag nanoparticles annealed from 3, 6 and 9 nm Ag thin film. During the measurements, the baseline is the transmission spectrum of the nanorod LED without Ag nanoparticles. (b) The schematic structure of the 3-D FDTD simulation to calculate the absorption cross section. The Ag nanoparticle is located around the GaN nanorod with a diameter D, and depth at d = 0.3D. (c) Calculated absorption cross section spectra of Ag nanoparticles with different diameter ranging from 10 to 70 nm. (d) The calculated absorption cross section spectra of Ag nanoparticles annealed from 3, 6 and 9 nm thin film.

Download Full Size | PDF

To further investigate the SP resonant behavior of Ag nanoparticles in three different LEDs, 3-D FDTD simulations were carried out to calculate the absorption cross section. Figure 2(b) shows the schematic model structure with a total-field scatter-field (TFSF) source. The shape of Ag nanoparticles was set as a spherical dome with a depth d buried in HfO2/GaN. With the depth d increasing, the resonant peaks of the Ag nanoparticles absorption cross section spectra will have a redshift (as seen in Fig. 7 in Appendix A). When the ratio between the depth and diameter is 0.3, the range of the simulated resonant absorption peaks shown in Fig. 2(c) for Ag nanoparticle with diameter varying from 10 to 70 nm is consistent with those of the measured transmission spectra. The Ag nanoparticle with a larger diameter intends to have a larger absorption intensity.

The absorption cross section spectra concerning the diameter distributions of Ag nanoparticles (Fig. 1(c)) in the three LEDs have also been calculated. Figure 2(d) shows that the peak wavelength difference between LED A and LED B is small because of their similar diameter distributions. Because the size of Ag nanoparticles in LED C is relatively quite large, it results in an obvious red-shift. Although the resonant wavelengths of these LEDs are very different, they are still close to the QW emission peak (~534 nm), indicating that there is an effective energy coupling between QWs and SPs in those LEDs.

3.2 Enhancement of the spontaneous emission rate for optical pumped QW-SP coupled LEDs

Figure 3(a) shows the PL spectra of the nanorod LEDs with and without Ag nanoparticles excited by a 405 nm laser diode. The nanorod LED without Ag nanoparticles has the largest PL intensity. After Ag nanoparticles incorporation, the PL intensities decrease with the Ag film thickness increasing, which is due to the high intrinsic IQE of the MQWs region and the low energy scattering efficiency of Ag Nps [16–18]. In addition, these LEDs show a more obvious wavelength shift with the Ag film thickness increasing, especially for LED C with a blue shift (1.3 nm) from 530.4 nm to 529.1 nm, suggesting that LED C has a more effective QW-SP coupling. Figure 3(b) shows the TRPL measurements for LEDs with different Ag nanoparticles. Compared to the nanorod LED without Ag nanoparticles, the decay rate of LED C after Ag incorporation is much faster than the other two LED samples.

 

Fig. 3 (a) PL spectra and (b) TRPL spectra of nanorod LED without Ag nanoparticles (black), nanorod LED with Ag nanoparticles obtained from annealed 3 nm (blue), 6 nm (red) and 9 nm (green) Ag films. The excitation source is a 405 nm laser diode. (c) PL spectra excited by a 325 nm He–Cd laser. (d) TRPL spectra excited by a 360 nm light sources. The black dashed lines in (a) and (c) correspond to the peak wavelength of the reference nanorod LED without Ag nanoparticles.

Download Full Size | PDF

During the PL and TRPL measurements, because the 405 nm light can only be absorbed by InGaN instead of GaN, the penetration depth of the excitation light will be quite deep, even can cover the whole active region. While SP is an evanescent wave that exponentially decays with increasing the distance, those deep QWs that cannot effectively couple with SPs will also emit photons, which may reduce the QW-SP coupling strength. To decrease the penetration depth, a 325 nm He-Cd laser was used as the excitation source afterwards. Because that the penetration depth by using the 325 nm light is shorter, the PL intensities of all LEDs decrease significantly as shown in Fig. 3(c), where only several QWs at the top of the active region can emit light. Furthermore, the blue shift for the PL spectrum of LED C becomes even larger than that shown in Fig. 3(a), indicating that the QW-SP coupling effect have been enhanced by using the shorter wavelength excitation light source. Similarly, the TRPL spectrum excited by a 360 nm pulsed light source also has a shorter carrier lifetime in Fig. 3(d), suggesting an increased spontaneous emission rate.

Using a bi-exponential decay component, the carrier lifetime can be calculated by fitting the PL decay curves using the following equation:

Table 1 lists the extracted fast decay parameters of our LEDs excited by two different light sources. The rapid carrier lifetimes (fast decay time) decrease dramatically with increasing the Ag thickness for both light excitation sources. Using the fast decay time of the nanorod LED without Ag nanoparticles as the reference, the obtained enhancement factor for τ₁ (average Purcell factor) increases from LED A to LED C for both light sources. Moreover, for the same LED (LED A, B or C), the average Purcell factor further increases when excited using the 360 nm light sources instead of 405 nm.

Table 1. Extracted fast decay parameters of LEDs with and without Ag nanoparticles excited by two different excitation light sources

View Table

For the same light source, the different behaviors for these QW-SP coupled LEDs may be due to their different size and density of Ag nanoparticles. Generally, a smaller metallic sphere will have a larger density state of SP mode, and a large Purcell factor [31, 32]. Considering the shape of the Ag nanoparticles is in spherical dome here, Purcell factor spectra were calculated using 3-D FDTD. The same structure as shown in Fig. 2(b) was used for the simulation with the source changed from a TFSF to a dipole. Figure 4 shows the Purcell factors for the dipole located at the edge of GaN nanorod with the Ag nanoparticle diameter from 10 nm to 70 nm. With the diameter of Ag nanoparticles increasing, obvious red-shifts from 480 nm to 580 nm can be seen for the resonant peaks in the Purcell factor spectra. The emission peak for the reference nanorod LEDs without Ag nanoparticles is 534 nm, which is very similar to the SP resonant wavelength for the 50 nm Ag nanoparticle. The highest enhancement for the spontaneous emission rate is roughly 106 for the 50 nm Ag nanoparticle, where the energies of excited dipole in the InGaN QWs can be transferred very quickly into the SP modes [13, 32]. But with the diameters increasing or decreasing from 50 nm, the enhancements were decreased. Because the average diameter of Ag nanoparticles in LED C is close to 50 nm, it has the largest Purcell factor among these SP-coupled LEDs. The density of Ag nanoparticles also contributes to the average Purcell factor. As mentioned above, the density of Ag nanoparticles increases from LED A to LED C, so more regions in GaN nanorod can be effectively coupled to SPs to increase the proportion of the rapid carriers, as represented by increasing the amplitude A₁ from LED A to LED C. Therefore, both the density and size of Ag nanoparticles can influence the spontaneous emission rate enhancement.

 

Fig. 4 The Purcell factor as a function of wavelength for Ag nanoparticles with a diameter range from 10 nm to 70 nm. The dipole source locates in the first QW and 11 nm horizontal away from the Ag nanoparticle.

Download Full Size | PDF

The influence on the average Purcell factors using two excitation sources were also investigated by 3-D FDTD simulations. Figures 5(a)-5(c) shows the calculated electric field intensity distribution with different excitation light sources. It represents the optical absorption intensity of the active region, which is proportional to the stimulated carrier concentration. With the wavelength of the excitation source increasing, both the light penetration depth and the electric field intensity (carrier concentration) in the active region increase. Due to the absorption by the GaN layer, the electric field is mainly located within the topmost 3-5 QWs for shorter excitation wavelength 325 or 360 nm. But using longer excitation wavelength 405 nm, the electric field at the bottom of the active region becomes even stronger than that close to the surface. With decreasing the excitation light wavelength, the photo-induced carriers intend to be mainly generated near the surface, where SPs locate, and the average Purcell factor will increase accordingly. Therefore, the proportion of SP-coupled carriers will increase by using a shorter wavelength light excitation, which is also consistent with the enhancement of A₁ listed in the second column of Table 1.

 

Fig. 5 The electric field distribution (log|E|²) of the nanorod LED with a 40 nm Ag nanoparticle excited by a light source with a wavelength of (a) 325 nm, (b) 360 nm, and (c) 405 nm. The three dashed lines indicate the topmost three QWs. The white solid lines indicate the interfaces of HfO₂ and GaN. The grey spherical dome at the left side of the GaN nanorod is the Ag nanoparticle.

Download Full Size | PDF

3.3 Design of electrical injected QW-SP coupled LED for VLC

From above discussion, we know that the QW-SP coupling effects are determined by both the Purcell factor spectrum and the proportion of SP-coupled carriers. For electrical injected QW-SP coupled LEDs, to increase the spontaneous emission rate, the size of Ag nanoparticle should be optimized to obtain the largest Purcell factor, where the optimized carrier distribution in the active region can largely increase the SP-coupled carrier proportion.

Using the optimized size for the Ag nanoparticle, the Purcell factor at 534 nm as a function of distances between 50 nm Ag and the dipole in QWs (Purcell factor distribution) can be calculated, in both horizontal and vertical direction, as shown in Fig. 6(a). The Purcell factor exponentially decays with increasing the horizontal distance from the metal location. When the horizontal distance is ~60 nm, the Purcell factor decreases to 1, indicating no enhancement effect for the spontaneous emission rate. In the vertical direction, the Purcell factors also decrease rapidly from 1st QW (z = 10 nm) to 3rd QW (z = −24 nm) because of the evanescent wave properties.

 

Fig. 6 (a) The Purcell factor as a function of horizontal distances between the Ag nanoparticle and the dipole source. The red lines, blue lines and green lines indicate the situation that the dipole locates at the first (z = 10 nm), the second (z = −7 nm) and the third QW (z = −24 nm), respectively. The black dashed line indicates the Purcell factor equals to 1. The inset is the 3-D FDTD simulation model. (b) The enhancement of spontaneous emission rate (average Purcell factor) for the entire nanorod LEDs as a function of the nanorod diameter.

Download Full Size | PDF

Based on the Purcell factor distribution and the spontaneous emission rate reported previously [33, 34], the average Purcell factor as a function of the GaN nanorod diameter for different LEDs is obtained to illustrate the QW-SP coupling effects, as shown in Fig. 6(b) (the calculation details can be found in Appendix B). With the nanorod diameter decreasing, the carrier distribution is different in the horizontal direction, and the average Purcell factors of all LEDs increase. Because strong QW-SP coupling only exists at the edge of GaN nanorod as shown in Fig. 6(a), with the nanorod diameter decreasing, the amount of carriers that cannot be coupled to SPs will decrease, which will enlarge the proportion of SP-coupled carriers. For LEDs with the same nanorod diameter, the QW-SP coupling effects of the LEDs with different carrier distribution in the vertical direction change significantly as well, as shown in Fig. 6(b). Compared to the single QW structure, more carriers transfer deep into the active region for the MQW structure, which will lead to a smaller coupling proportion of carriers and weaker QW-SP coupling effect. In addition, compared to the MQW structure with more uniform carrier distribution as shown in [34], the QW-SP coupling will be more effective to enhance spontaneous emission rate when the carrier recombination mainly occurs in the top two QWs in [33], which is similar to the decreased light penetration depth for the optical pumped LED.

4. Conclusion

In this work, we investigate the influence of metal nanoparticles and excitation sources on the QW-SP coupling process systematically. The results show that the QW-SP coupling becomes more obvious when the SP resonant wavelength of Ag nanoparticle is close to the QW emission wavelength. In addition, the spontaneous emission rate further increases when excited by a shorter wavelength source. Combined with the simulations, we find that the enhancement is due to the decrease of the penetration depth, which will increase the proportion of SP-coupled carriers. The proportion of SP-coupled carriers is determined by the evanescent wave properties of SPs and the carrier distribution. For the electrical QW-SP coupled LEDs, both the Purcell factor and carrier distribution are important factor to enhance the spontaneous emission rate. Our findings can pave a way to design the ultrafast LED light source for the application of visible light communication.

Appendix A Absorption cross section spectra of Ag nanoparticles with different buried depth

 

Fig. 7 Calculated absorption cross section spectra of Ag nanoparticles with different buried depth d range from 0.1D (4 nm) to 0.5D (20 nm).

Download Full Size | PDF

Appendix B Calculation of the average Purcell factor

In this appendix, the calculation of the average Purcell factor of a single GaN nanorod with SPs as a function of the diameter of the nanorod at 534 nm (Fig. 6(b)) is presented, as follows:

(2)$$R_r\prime =F_{{}_r}{R}_{{}_r}$$

The excited carrier with the original radiative decay rate R_r will relax by transferring energy into the SP mode at the rate of F_rR_r, which is enhanced by the Purcell factor F_r. Actually, the Purcell factor varies with the position of carriers as shown in Fig. 6(a). The average original radiative recombination rate is related to A(r), which is the ratio of carrier amount at the position r to the amount of all carriers.

(3)$${\overline{R}}_r={\displaystyle {\int }_0^{{}_{r_{{}_0}}}A}(r)R_r(r)dr$$

Combine Eqs. (2) and (3), the enhanced average radiative recombination rate can be obtained,

(4)$$\overline{R_r\prime }={\displaystyle {\int }_0^{{}_{r_{{}_0}}}A}(r)F_r(r)R_r(r)dr$$

Thus, the average Purcell factor is obtained,

(5)$$\overline{F}=\overline{R_r\prime }/\overline{R_r}=\frac{{\displaystyle {\int }_0^{{}_{r_{{}_0}}}A}(r)F_r(r)R_r(r)dr}{{\displaystyle {\int }_0^{{}_{r_{{}_0}}}A}(r)R_r(r)dr}$$

Funding

National Natural Science Foundation of China (Grants No. 11574306); National Basic Research and High Technology Program of China (Grants No.2015AA03A101, 2014BAK02B08 and 2015AA033303).

References and links

1. H. Haas, L. Yin, Y. Wang, and C. Chen, “What is LiFi?” J. Lightwave Technol. 34(6), 1533–1544 (2016). [CrossRef]  

2. H. D. Chen, C. H. Wu, H. L. Li, X. B. Chen, Z. Y. Gao, S. G. Cui, and Q. Wang, “Advances and prospects in visible light communications,” J. Semicond. 37(1), 011001 (2016). [CrossRef]  

3. L. X. Zhao, S. C. Zhu, C. H. Wu, C. Yang, Z. G. Yu, H. Yang, and L. Liu, “GaN-based LEDs for light communication,” Sci. China Phys. Mech. Astron. 59(10), 107301 (2016). [CrossRef]  

4. M. Hoa Le, D. O’Brien, G. Faulkner, Z. Lubin, L. Kyungwoo, J. Daekwang, O. YunJe, and W. Eun Tae, “100 Mb/s NRZ visible light communications using a postequalized white LED,” IEEE Photonics Technol. Lett. 21(15), 1063–1065 (2009). [CrossRef]  

5. H. Li, X. Chen, J. Guo, Z. Gao, and H. Chen, “An analog modulator for 460 Mb/S visible light data transmission based on OOK-NRS modulation,” IEEE Wirel. Commun. 22(2), 68–73 (2015). [CrossRef]  

6. J. W. Shi, H. Y. Huang, J. K. Sheu, C. H. Chen, Y. S. Wu, and W. C. Lai, “The improvement in modulation speed of GaN-based Green light-emitting diode (LED) by use of n-type barrier doping for plastic optical fiber (POF) communication,” IEEE Photonics Technol. Lett. 18(15), 1636–1638 (2006). [CrossRef]  

7. S. X. Zhu, J. X. Wang, J. C. Yan, Y. Zhang, Y. R. Pei, Z. Si, H. Yang, L. X. Zhao, Z. Liu, and J. M. Li, “Influence of AlGaN electron blocking layer on modulation bandwidth of GaN-based light emitting diodes,” ECS Solid State Lett. 3(3), R11–R13 (2014). [CrossRef]  

8. C. L. Liao, Y. F. Chang, C. L. Ho, and M. C. Wu, “High-speed GaN-based blue light-emitting diodes with gallium-doped ZnO current spreading layer,” IEEE Electron Device Lett. 34(5), 611–613 (2013). [CrossRef]  

9. J. M. Wun, C. W. Lin, W. Chen, J. K. Sheu, C. L. Lin, Y. L. Li, J. E. Bowers, J. W. Shi, J. Vinogradov, R. Kruglov, and O. Ziemann, “GaN-based miniaturized cyan light-emitting diodes on a patterned sapphire substrate with improved fiber coupling for very high-speed plastic optical fiber communication,” IEEE Photonics J. 4(5), 1520–1529 (2012). [CrossRef]  

10. J. J. D. McKendry, D. Massoubre, S. Zhang, B. R. Rae, R. P. Green, E. Gu, R. K. Henderson, A. E. Kelly, and M. D. Dawson, “Visible-light communications using a CMOS-controlled micro-light-emitting-diode array,” J. Lightwave Technol. 30(1), 61–67 (2012). [CrossRef]  

11. H. Shih-Yung, H. Ray-Hua, S. Jin-Wei, K. Hao-Chung, and W. Dong-Sing, “High-performance InGaN-based green resonant-cavity light-emitting diodes for plastic optical fiber applications,” J. Lightwave Technol. 27(18), 4084–4094 (2009). [CrossRef]  

12. G. Shambat, B. Ellis, A. Majumdar, J. Petykiewicz, M. A. Mayer, T. Sarmiento, J. Harris, E. E. Haller, and J. Vučković, “Ultrafast direct modulation of a single-mode photonic crystal nanocavity light-emitting diode,” Nat. Commun. 2, 539 (2011). [CrossRef]   [PubMed]  

13. K. Okamoto, I. Niki, A. Scherer, Y. Narukawa, T. Mukai, and Y. Kawakami, “Surface plasmon enhanced spontaneous emission rate of InGaN/GaN quantum wells probed by time-resolved photoluminescence spectroscopy,” Appl. Phys. Lett. 87(7), 071102 (2005). [CrossRef]  

14. K. Okamoto, I. Niki, A. Shvartser, Y. Narukawa, T. Mukai, and A. Scherer, “Surface-plasmon-enhanced light emitters based on InGaN quantum wells,” Nat. Mater. 3(9), 601–605 (2004). [CrossRef]   [PubMed]  

15. G. M. Akselrod, C. Argyropoulos, T. B. Hoang, C. Ciracì, C. Fang, J. Huang, D. R. Smith, and M. H. Mikkelsen, “Probing the mechanisms of large Purcell enhancement in plasmonic nanoantennas,” Nat. Photonics 8(11), 835–840 (2014). [CrossRef]  

16. W. H. Chuang, J. Y. Wang, C. C. Yang, and Y. W. Kiang, “Numerical study on quantum efficiency enhancement of a light-emitting diode based on surface plasmon coupling with a quantum well,” IEEE Photonics Technol. Lett. 20(16), 1339–1341 (2008). [CrossRef]  

17. A. Fadil, D. Iida, Y. Chen, J. Ma, Y. Ou, P. M. Petersen, and H. Ou, “Surface plasmon coupling dynamics in InGaN/GaN quantum-well structures and radiative efficiency improvement,” Sci. Rep. 4, 6392 (2014). [CrossRef]   [PubMed]  

18. D. Iida, A. Fadil, Y. T. Chen, Y. Y. Ou, O. Kopylov, M. Iwaya, T. Takeuchi, S. Kamiyama, I. Akasaki, and H. Y. Ou, “Internal quantum efficiency enhancement of GaInN/GaN quantum-well structures using Ag nanoparticles,” AIP Adv. 5(9), 097169 (2015). [CrossRef]  

19. I. Gontijo, M. Boroditsky, E. Yablonovitch, S. Keller, U. K. Mishra, and S. P. DenBaars, “Coupling of InGaN quantum-well photoluminescence to silver surface plasmons,” Phys. Rev. B 60(16), 11564–11567 (1999). [CrossRef]  

20. Z. G. Yu, L. X. Zhao, X. C. Wei, X. J. Sun, P. B. An, S. C. Zhu, L. Liu, L. X. Tian, F. Zhang, H. X. Lu, J. X. Wang, Y. P. Zeng, and J. M. Li, “Surface plasmon-enhanced nanoporous GaN-based green light-emitting diodes with Al₂O₃ passivation layer,” Opt. Express 22(S6Suppl 6), A1596–A1603 (2014). [CrossRef]   [PubMed]  

21. D. M. Yeh, C. F. Huang, C. Y. Chen, Y. C. Lu, and C. C. Yang, “Surface plasmon coupling effect in an InGaN/GaN single-quantum-well light-emitting diode,” Appl. Phys. Lett. 91(17), 171103 (2007). [CrossRef]  

22. C. Y. Cho, M. K. Kwon, S. J. Lee, S. H. Han, J. W. Kang, S. E. Kang, D. Y. Lee, and S. J. Park, “Surface plasmon-enhanced light-emitting diodes using silver nanoparticles embedded in p-GaN,” Nanotechnology 21(20), 205201 (2010). [CrossRef]   [PubMed]  

23. S. C. Zhu, Z. G. Yu, L. X. Zhao, J. X. Wang, and J. M. Li, “Enhancement of the modulation bandwidth for GaN-based light-emitting diode by surface plasmons,” Opt. Express 23(11), 13752–13760 (2015). [CrossRef]   [PubMed]  

24. C. Y. Yang, A. A. Bettiol, Y. Shi, M. Bosman, H. R. Tan, W. P. Goh, J. H. Teng, and E. J. Teo, “Fast Electrical Modulation in a Plasmonic-Enhanced, V-Pit-Textured, Light-Emitting Diode,” Adv. Opt. Mater. 3(12), 1703–1709 (2015). [CrossRef]  

25. C. H. Lin, C. Y. Su, E. Zhu, Y. F. Yao, C. Hsieh, C. G. Tu, H. T. Chen, Y. W. Kiang, and C. C. Yang, “Modulation behaviors of surface plasmon coupled light-emitting diode,” Opt. Express 23(6), 8150–8161 (2015). [CrossRef]   [PubMed]  

26. P. Anger, P. Bharadwaj, and L. Novotny, “Enhancement and quenching of single-molecule fluorescence,” Phys. Rev. Lett. 96(11), 113002 (2006). [CrossRef]   [PubMed]  

27. D. M. Yeh, C. F. Huang, C. Y. Chen, Y. C. Lu, and C. C. Yang, “Localized surface plasmon-induced emission enhancement of a green light-emitting diode,” Nanotechnology 19(34), 345201 (2008). [CrossRef]   [PubMed]  

28. S. Jiang, Z. Hu, Z. Chen, X. Fu, X. Jiang, Q. Jiao, T. Yu, and G. Zhang, “Resonant absorption and scattering suppression of localized surface plasmons in Ag particles on green LED,” Opt. Express 21(10), 12100–12110 (2013). [CrossRef]   [PubMed]  

29. M. K. Kwon, J. Y. Kim, B. H. Kim, I. K. Park, C. Y. Cho, C. C. Byeon, and S. J. Park, “Surface-plasmon-enhanced light-emitting diodes,” Adv. Mater. 20(7), 1253–1257 (2008). [CrossRef]  

30. C. K. Choi, Y. H. Kwon, B. D. Little, G. H. Gainer, J. J. Song, Y. C. Chang, S. Keller, U. K. Mishra, and S. P. DenBaars, “Time-resolved photoluminescence of In_xGa_1−xN/GaN multiple quantum well structures: Effect of Si doping in the barriers,” Phys. Rev. B 64(24), 245339 (2001). [CrossRef]  

31. A. F. Koenderink, “On the use of Purcell factors for plasmon antennas,” Opt. Lett. 35(24), 4208–4210 (2010). [CrossRef]   [PubMed]  

32. G. Sun, J. B. Khurgin, and R. A. Soref, “Practical enhancement of photoluminescence by metal nanoparticles,” Appl. Phys. Lett. 94(10), 101103 (2009). [CrossRef]  

33. H. J. Li, J. J. Kang, P. P. Li, J. Ma, H. Wang, M. Liang, Z. C. Li, J. Li, X. Y. Yi, and G. H. Wang, “Enhanced performance of GaN based light-emitting diodes with a low temperature p-GaN hole injection layer,” Appl. Phys. Lett. 102(1), 011105 (2013). [CrossRef]  

34. Y. Guo, M. Liang, J. J. Fu, Z. Liu, X. Y. Yi, J. X. Wang, G. H. Wang, and J. Li, “Enhancing the performance of blue GaN-based light emitting diodes with double electron blocking layers,” AIP Adv. 5(3), 037131 (2015). [CrossRef]