Open Access
Add to BibSonomyAbstract
Phosphor-in-glass (PiG) when used in combination with a blue light emitting diode (LED) chip in a remote phosphor configuration offers precise tuning and yields higher luminous efficiencies at elevated temperatures, compared to the conventional conformal LED packaging. However, drawbacks such as spectral overlapping of the constituent phosphors and resultant reabsorption remain unresolved in multi phosphor color conversion plates. These issues were solved up to a desired extent by arranging different color-emitting PiGs, via cutting and reassembly. However, the interface of the multicolored plates acted as a dissipative layer. In this work, a novel fabrication technique was proposed to overcome this drawback by eliminating the interfacial layer through a one-step process. PiGs were fabricated using glass frits at a low softening temperature of 600 °C. As a result, a higher efficacy of the studied prototypes, i.e., a horizontal 2-layered PiG and a 4-quadrant PiG, was obtained as compared with their counterparts. The angular dependency of the luminescence of the segmented 4-quadrant type PiG was studied, and the results were discussed.
© 2016 Optical Society of America
1. Introduction
The recently developed phosphor-converted white light-emitting diode (WLED) constitutes the most promising alternative to conventional light sources owing to its high efficiency, long life, low power consumption, environmental friendliness, small size, and directionality. The corresponding research performed over the last decade has paved the way for the development of several new phosphors, designs, packaging techniques, encapsulant, etc [1–4 ]. Conventionally, white-light emission has been achieved by applying phosphors (for example, Y3Al5O12:Ce3+) encapsulated in resin paste to a blue LED chip. During operation, the temperature of the chip increases (especially under high forward bias current values), thereby leading to reductions in the permeability and luminous efficacy. This increase also leads to a change in the luminescent color and a shortening of the life of the transparent epoxy or silicone binder [5–7 ]. In order to overcome these drawbacks, certain remote phosphor configurations based on phosphor ceramics, luminescent glasses, phosphor-in-glass (PiG) etc., had been proposed [6–10 ]. Remote type LED configurations were reported to offer enhanced efficiency as much as 20-30% of that it’s conformal counterparts [11–13 ]. Phosphor ceramics, which is a single phase ceramic plate containing microcrystals of some efficient phosphors were not widely preferred owing to the difficulty in fabrication processes involved; however, luminescent glass which is a rare-earth doped glass material could not be used as the sole converting material due to its lower color conversion efficiency. The PiGs are considered a viable alternative to phosphor ceramics and luminescent glasses, and have therefore been extensively investigated.
PiGs fabricated through the mixing of phosphors with glass frit are stable under the extreme operating conditions of the LED. As a result, many new or previously overlooked parameters such as thickness and geometry of the color conversion layers had been defined for fine tuning the emission [6, 14 ]. During the fabrication process, the phosphor powders are all dispersed in a glass. Carefully selected amounts of various green/yellow/red phosphors were mixed in order to improve the color rendering properties. However, the problems of low color purity and emission efficiency persisted owing to the spectral overlapping of the randomly mixed phosphors, which partially reabsorb the emitted light [15, 16 ]. A stacked PiG design, in which alternating green/yellow and red plates are overlaid, and a multicolored 4-quadrant-type PiG with alternating colors was proposed as solutions to eliminate the reabsorption [17, 18 ]. This also brings about a new possibility of emission color tuning by simply changing the geometric parameters of the color conversion layer. Owing to the elimination of reabsorption, these designs resulted in higher efficiencies than those obtained with the use of randomly mixed PiGs. However, the air gap at the interface of the stacked disk and the alternate quadrants in these designs acted as a scattering layer. The actual efficiencies were therefore significantly lower than the expected values. In a stacked PiG, leakage of blue light from the LED chip, owing to the air gap of the middle layer, results in a low efficiency. In the 4-quadrant type PiG, the interfacial air gap between alternate quadrants scatters the blue light and thereby results in a low efficiency. Moreover, the complexity associated with the realization of these designs on a commercial scale has limited the application of the proposed models.
As such, in this study, we report an upgraded PiG design and a fabrication technique for eliminating the interfacial layer, and hence the leakage of light through this layer. PiGs based on commercial Lu3Al5O12:Ce3+ (LuAG:Ce3+) and CaAlSiN3:Eu2+ (CASN:Eu2+) phosphors are fabricated by means of a one-step process that eliminates the stacking or reconfiguration step of the already fabricated disks. The viability of the multi-layered and multi-segmented PiGs fabricated in this work is demonstrated, by using the remote-type configuration of the WLED. Furthermore, the angular dependence of the spectral properties of the PiGs was analyzed. This technique and design proved to be very practical in tuning the optical properties, including the color temperature and color rendering index (CRI) values, of the WLEDs.
2. Materials and methods
The PiGs were fabricated by using glass frits composed of relatively low-melting-point (≤ 600 °C) glass. The glass frits were synthesized from a stoichiometric composition of SiO2, B2O3, ZnO, Al2O3, and K2O that formed the (SiO2, B2O3, ZnO)-Al2O3, K2O system; SiO2, B2O3, and ZnO constituted 25, 25, and 30 wt.%, respectively, and a cumulative fraction of 80 wt.%, while K2O and Al2O3 constituted respective fractions of 15 and 5 wt.% [8]. The PiGs were fabricated from respective mixtures of an appropriate amount of glass frits with commercial LuAG:Ce3+ and CASN:Eu2+ powders [19], which were subsequently fired at 600 °C. The ratios of commercial green-emitting phosphor LuAG:Ce3+ and red-emitting phosphor CASN:Eu2+ PiG were maintained at 2:1 and the corresponding phosphor to glass ratios were 2:10 and 1:10, in all the PiGs discussed in the subsequent sections. Each fabricated green LuAG:Ce3+ and red CASN:Eu2+ PiG samples were arranged in two horizontal layers or four quadrants on a single PiG. Therefore, each PiG consisted of distinct multicolored red and green layers/segments arranged in a single unit. These PiGs were fabricated in a one-step process, which is described in subsequent sections. The assembled 2-layered reference PiGs were fabricated based on the steps prescribed in a patent on stacked phosphor layers [18]. The assembled 4-quadrant PiG reference sample was fabricating by rearranging multicolored segments of the PiGs as proposed in our previous work [17]. This rearrangement was performed by using quarter segments obtained from cutting the PiGs into four equal parts. The red or green colored PiGs were prepared by mixing red-emitting CASN:Eu2+ or green-emitting LuAG:Ce3+ phosphor with the desired quantity of glass frit, and firing at 600 °C. Two pieces of each quadrant of the 250 mm (diameter) × 1 mm (thickness) red and green PiGs were integrated into one circular disc with an optically inert glass substrate as supporting medium.
Figure 1(a) shows the schematic of the one-step fabrication of the horizontal 2-layered PiG, made of green LuAG:Ce3+ and red CASN:Eu2+. In this facile fabrication process, the glass frit was first mixed with either red-emitting CASN:Eu2+ or green-emitting LuAG:Ce3+ phosphors. The mixtures containing the aforementioned composites were subsequently compressed into disks of the desired dimensions. These as-prepared disks were then stacked on top of each other and compressed for 5 min under a pressure of ∼1500 MPa. The resulting 2-layered disk was then heat-treated for 30 min at 600 °C, thereby yielding a 2-layered PiG.
Fig. 1 Schematic one-step design and (a) fabrication of a horizontal 2-piece PiG and (b) lengthwise slicing into four quadrants consisting of green LuAG:Ce3+ and red CASN:Eu2+.
Figure 2(b) shows the schematic of the fabrication of 4-quadrant type PiG with alternating green and red quadrants. This multicolor 4-quadrant type PiG sample was fabricated by employing a film compartment that places the glass frit-phosphor mixtures into alternating quadrants. The film separator was then carefully removed and the glass frit-phosphor mixture was subsequently compressed under the axial load. Subsequent steps of the synthesis process were similar to those of the aforementioned horizontal 2-layered PiG.
Fig. 2 Schematic diagram of the remote phosphor configuration in (a) lateral and (b) top view displaying the arrangement of the blue LED chips.
The overall thickness of all the PiGs was maintained as 1 mm by optical polishing; PiGs of this thickness were used for all the studies. This thickness was selected based on the remote-type configuration of a white LED, where a reasonable mechanical strength is required.
The electroluminescence (EL) of the fabricated PiGs was determined by using a remote-type LED configuration in which a circular PiG disc was mounted on a circular jig. This jig consisted of six LED chips that were arranged symmetrically on a circle with radius of around 7 mm with each LED subtending a central angle of 60° as shown in Fig. 2. The LED configuration was then inserted in an integrating sphere, and arranged in a forward-emitting 2π configuration. The angle-dependent EL spectra of the 4-segmented PiGs were measured by isolating the light flux through a solid angle of ∼0.08 sr. Furthermore, the exposure and the reflections other than those emanating from the desired cone were eliminated by blackening the inner surface of the obstructer. The angular dependence was estimated by measuring the EL spectra at eight geometrically significant points around the PiG plate. These measurements were performed at polar angles of 30, 45, and 60°.
3. Results and discussion
Figure 3 shows the proposed 2-layered and 4-quadrant type PiGs, obtained by rearranging the PiG, and their counterparts. Figure 3(a) and 3(b) show a double-layered PiG and a one-step double-layered PiG, respectively. The respective segmented configuration, of a diced 4-quadrant type PiG, and the proposed 4-quadrant type PiG are shown in Fig. 3(c) and 3(d). Each of these 25-mm-diameter PiGs has a thickness of 1 mm; the disk was mounted on a jig containing six blue LED chips that are symmetrically arranged in a remote-type configuration. The EL analysis was performed under various values of the forward bias current. Previous studies reported that the efficiency of the 2-layered stacking of the red and green phosphor layer increases when the red phosphor is used as the first layer above the chip. This increase occurs since the reabsorption of green emission by the red component (which is typical of other configurations) is eliminated by using this configuration. However if this configuration is employed in a multi phosphor converted WLED (more than two phosphors) the substantial increase in thickness, results in increasing thickness of the color conversion layer. This would essentially reduce the transparency of the overall arrangement and thereby nullifies the advantages due to elimination of reabsorption. Also, the stacked phosphor design is patented and commercialization of this design constitutes a breach of the patent rights [18]. A new phosphor plate design, which ultimately proved superior to its counterparts, was therefore investigated.
Fig. 3 Schematic configuration of the (a) stacked double-layer PiG, (b) one-step double-layer PiG, (c) diced 4-quadrant PiG joined together with a substrate, and (d) one-step 4-quadrant type PiG mounted on a remote-type configuration.
The SEM image of the one-step fabricated 2-layered PiG (Fig. 4 ) indicates that the phosphors (LuAG:Ce3+and CASN:Eu2+) were separated from each other at both ends (S1, S2); S3 is located in the glass matrix. The inset of the Fig. shows a digital image of the boundary between the alternating layers of the one-step double-layered PiG. Furthermore, the energy dispersive spectra of points S1, S2, and S3 reveal distinct compositions that are indicative of alternating layers on a single unit. The spectrum collected from point S1 corresponds to a red phosphor region, as evidenced by the presence of Ca and Eu, which are constituent elements of the red-phosphor CASN:Eu2+. Likewise, the presence of Lu and Ce in the spectrum corresponding to S2 confirms that this point lies in a green-phosphor LuAG:Ce3+ region. The presence of crystalline phosphors in the glass matrix of the PiGs was confirmed via X-ray diffraction (XRD) measurements. The resulting profile (Fig. 4(c)) was indexed to the aggregated patterns of the component phosphors, LuAG:Ce3+ and CASN:Eu2+, indicating that the fabrication process had negligible effect on the composition of the phosphor. This results possibly from the relatively low melting point of the glass frit used in the process. The digital photograph of the one-step fabricated PiGs, along with the envisioned diagrams and the LEDs under operation, using afore mentioned PiGs, are shown in Fig. 5(a) . The EL spectra obtained under an operating condition of 350 mA and 20 V, and the corresponding CIE-1931 chromaticity coordinates along the Planckian locus are shown in Fig. 5(b) and 5(c), respectively.
Fig. 4 (a) SEM image of the PiG, (b) energy dispersive spectra of points S1, S2, and S3, which correspond to the CASN:Eu2+, LuAG:Ce3+, and glass matrix, respectively. (c) XRD pattern of the PiG along with the reference patterns of the component phosphors; the inset of (a) shows the boundary image of the one-step double-layer PiG.
Fig. 5 (a) An actual image of the one-step fabricated PiG, (b) electroluminescence spectra of the four configurations of the PiG-LED system, under operating conditions of 350 mA and 20 V, and (c) the CIE-1931 chromaticity coordinates of the corresponding WLED system.
The EL emission of the 2-layered stacked and one-step fabricated PiGs, are composed of a blue emission at 450 nm, and a broad emission that occurs at wavelengths ranging from 500 to 700 nm; the former arises from the InGaN LED chip and the latter from the green LuAG:Ce3+ and red CASN:Eu2+ phosphors. Furthermore, the total emission intensity of one-step fabricated PiG increased by a factor of ∼1.12, owing to the elimination of the air gap in the horizontal 2-layered PiG and thereby resulting in a high efficiency.
The emission spectra of both the 4-quadrant type PiGs are also composed of a blue emission at 450 nm, and distinguishable emission bands from the component phosphors, the green LuAG:Ce3+ and red CASN:Eu2+. This distinction in emission spectra of the segmented configuration from its layered counterparts arises from the reduction in the overall thickness of the color conversion plate. In this configuration the red component of the emission spectra no longer needs to traverse through the translucent green layer and therefore results in predominant red component. Considering the fact that the red emitting phosphors with large stokes shift are less efficient than its low stokes shift counterparts, the proposed configuration would effectively increase the overall efficiency of the system by reducing the red phosphor content. However, in one-step fabricated 4-quadrant type PiG, which is a single unit rather than the assembly of individual pieces, the scattering loss of blue light from the interfacial layers of the segments is completely eliminated. Also there is no need of employing glass substrate as a supporting medium which could again have an impact on the optical properties of the configuration. The optical properties of all the aforementioned white LED configurations are listed in Table 1 . A comparison of the CIE color coordinates reveals that the proposed technique, may lead to shifts in the targeted CIE coordination (0.41, 0.38) to the required range for LED applications in the lighting industry; this is especially evident as the same parameter values are used for all of the samples. The correlated color temperatures (CCT) of the one-step fabricated 2-layered and 4-quadrant PiGs, evaluated by using an integrated sphere, lies in the warm white-light region (∼3000 K). In fact, the CCT and luminous efficiency of the samples differ only slightly. A further shift in the color coordinates and CCT of the proposed segmented configuration can be simply achieved by appropriately varying the phosphor to glass-frit ratio of the corresponding segments.
Table 1. Electroluminescence data of the 2-layered and 4-quadrant PiGs, prepared via two different processes.
The aggregated properties of the 4-quadrant PiGs were similar to those of their 2-layered counterparts. However, the segmented nature of the 4-quadrant PiGs limits these properties to a few symmetric positions, unless an integrated sphere is used for detection. The angular dependence of the luminous properties was therefore estimated by using a custom-designed arrangement in conjunction with the integrated sphere. In this arrangement, the radiating surface of the 12.5-mm-radius PiG is covered with a hemisphere, in which the inner radius is equal to that of the PiG plate. A small ~12.6 mm2 opening carved out of the blackened hemisphere allows radiation through a small solid angle of ∼0.08 sr. The polar and the azimuthal angles subtended by this opening with respect to a specific point on the PiG (Fig. 6 ) were sequentially varied and the angle-dependent emission property of the combination was studied.
Fig. 6 Azimuthal and polar angles gradated on the 4-segmented PiG illustrating the angle dependent luminescence measurement.
The angular dependence was presented as the dependence of three component emission viz. from the LED chip, from green phosphor, and the red phosphor. Considering the fact that, in color conversion process the individual phosphor particles present in the glass matrix acts as sources of the component colors (green and red), the position of the excitation source is less significant in assessing the angular dependence of the resultant spectra (green and red). The array of 6 LEDs used as excitation source and blue component of the resultant emission spectra is symmetric with respect to the 4 quadrants and would necessarily distribute blue component equally in all the quadrants of the PiGs. The further advancement of the blue component would also be independent of the positioning of the sources as the translucent PiG plate act as a scattering medium and the blue rays would essentially undergoes scattering. The symmetric arrangement of six LED chips could therefore be treated as a point source placed at the center of the jig. Therefore the angle or the position dependency of the blue component was assumed to be dependent on the relative positioning of the secondary source, which in this case is PiG plate. It was also assumed that the positioning of the individual phosphor particles determines the angle dependency of the component colors, green and red.
Figure 7 shows the polar plot of the relative intensity of three distinct components present in the WLED, measured in equal steps at different azimuthal angles. Figure 7(a) shows digital photograph of the PiG with gradation resembling the measurement scale, while the plots obtained at respective polar angles of 30, 45, and 60° are shown in Fig. 7(b), 7(c), and 7(d). The individual blue, green, and red components from the InGaN LED, LuAG:Ce3+ phosphor, and the CASN:Eu2+ phosphor centered at ∼450 nm, 534 nm, and 608 nm, respectively, are shown in each Fig.; the components are plotted using the same angular increments on both axes. These components (especially the blue and green) exhibit significant angular dependence at a polar angle of 30° but only moderate dependence at 60°. However, owing to symmetry, the dependence becomes negligible at a polar angle of 90°.
Fig. 7 (a) Digital photograph of a one-step fabricated 4-quadrant PiG with labeled azimuthal angles. Polar plots of the relative intensities of red, green, and blue component colors at polar angles of (b) 30°, (c) 45°, and (d) 60°.
The trends in the polar plots indicate that the angular dependency of the red component is relatively independent of the arrangement of the red color conversion units in the 4-quadrant PiG. Therefore, although geometrically distinct in the case of the red segment of the PiG, the relative intensity of this component is approximately constant with varying measurement location. This trend was consistently observed for all the other polar angles considered. The green component exhibits more dependence on the azimuthal angle, than its red counterpart, especially for polar angles lower than 60°. In addition, the corresponding intensity distribution varies significantly with the polar angle and, with slight increase in the polar angle, exhibits seemingly opposite trend to those distribution patterns. The blue component of the output exhibits an even sharper angular dependence, than its green counterpart, although the angular distribution of the source blue LED is symmetric. As Fig. 7(b)–7(c) show, the pattern of the relative intensity distribution varies abruptly with azimuthal angles, for polar angles ranging from 30 to 60°. A seemingly identical pattern of the distribution is obtained for all the polar angles considered, although the shape changed from a double dumbbell at 30° to a single dumbbell at 60°. The double dumbbell is, in fact, a broad single dumbbell, in which the tapering decreases with an increase of the polar angle to 45°. In other words, the dumbbell is expanded laterally with increasing polar angle, thereby resulting in a significantly lower angular dependence than that observed at an angle of 30°.
The aforementioned angular dependency of the individual components is indicative of the correlation between the geometric positioning of the detector and the range of wavelengths of the band under consideration. Large-wavelength regions of the spectra are almost uniformly distributed, whereas those with relatively lower wavelengths exhibit greater angular dependence. Furthermore, the alternating distribution pattern of the green component and the abrupt change in the distribution of its blue counterpart, result possibly from several factors. These include the, scattering of the component wavelengths by the particles, relative transparency of the segmented plates, and the geometrical arrangement of the individual segments. The similar size and identical geometric distribution of the green LuAG:Ce3+ and red CASN:Eu2+ phosphors confirm that this is indeed the case. Moreover, light rays are more frequently scattered at high angles than at low ones. The phosphor-to-glass ratio used in the individual segments constitutes, therefore, the major factor that influences the intensity distribution of the component colors.
These findings indicate that the aforementioned sharp angular dependence can be eliminated by performing a rational optimization process during the fabrication of the PiG. During this process, the relative intensity of the blue component would be estimated, rather than controlling the emission color by tuning the ratio of the green to red phosphors in the respective segments. The green LuAG:Ce3+ containing glass which contains more phosphor particles than its red counterpart, provides bright green emission and reduces the transparency of the blue component of the spectrum; this is evident from the relative intensity distribution of the component colors at various polar angles. A low phosphor density in the green segment of the PiG will lead to a reduction in the angular dependence of the blue component. This reduction results from the escape of a larger amount of blue flux at moderate, rather than higher phosphor concentrations. The angular dependence of both the green and blue component is reduced, owing to this escape. In addition, the blue component was rendered almost negligible owing to the complete conversion of the blue flux from the LED by the denser green phosphor layer. However, the highest intensity of the blue component occurs in the vicinity of the red segment of the color conversion plate. This results from the red-to-green phosphor ratio, of 1:2, that leads to the transmission of a large portion of the unconverted blue flux through the red segment. Therefore, the high-intensity region of the green and blue appears in alternating quadrants of the polar diagram.
These results confirm the superiority of the remote-type design in tuning the spectral properties over conventional LED packaging. This demonstrates that the CIE is highly dependent on the configuration and techniques employed during the fabrication, in addition to the fine tuning accomplished by varying geometrical parameters such as the thickness. The angular dependence of the luminescence results primarily from the optical parameters of the PiG material. More importantly, the current design and fabrication of the PiG, which employs a separation and layering strategy, will facilitate a new era of LED design; the tunability of the optical properties will also be clarified. The 4-segmented PiG used in this study can be further increased to eight equal segments, so that a particular colored segment is subtended at an angle of 45° at the center. This configuration can further reduce the angular dependence of the luminescence. Furthermore, if the blue component is angular-dependent rather than homogeneously distributed owing to the radiant blue LED chips, then the relative angular width of the segments with alternating colors can also be varied.
4. Conclusions
We proposed a novel design and fabrication of a PiG by employing a one-step fabrication technique for eliminating the reabsorption of green emitted light by the red component of the high-power remote WLEDs. This fabrication was also aimed at preventing light leakage from the InGaN chip at the interface of the color conversion plates. The proposed strategy reduces the light losses more efficiently than the 2-layered and 4-segmented red CASN:Eu2+ and green LuAG:Ce3+ PiG prepared by stalking and assembling, respectively from individual units. In addition, an analysis of the angular dependence of the luminescence properties shows that the proposed design tends to have spectral uniformity at higher polar angles close to 90°. This dependence can be brought down to a lower value of the polar angle, when the number of segments in the design is increased. These attributes represent the results of a longstanding effort aimed at fabricating highly efficient solid-state lighting devices that are based on high-powered WLEDs.
Acknowledgments
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science, and Technology. This work was also supported by the Strategic Key-Material Development and the Materials and Components Research and Development bodies, under the auspices of the Ministry of Knowledge Economy (MKE, Korea). The authors also gratefully acknowledge the financial support from the Joint Research Project.
References and links
1. Z. Liu, S. Liu, K. Wang, and X. Luo, “Measurement and numerical studies of optical properties of YAG:Ce phosphor for white light-emitting diode packaging,” Appl. Opt. 49(2), 247–257 (2010). [CrossRef] [PubMed]
2. Y. K. Lee, J. S. Lee, J. Heo, W. B. Im, and W. J. Chung, “Phosphor in glasses with Pb-free silicate glass powders as robust color-converting materials for white LED applications,” Opt. Lett. 37(15), 3276–3278 (2012). [CrossRef] [PubMed]
3. W. B. Im, N. George, J. Kurzman, S. Brinkley, A. Mikhailovsky, J. Hu, B. F. Chmelka, S. P. DenBaars, and R. Seshadri, “Efficient and Color-Tunable Oxyfluoride Solid Solution Phosphors for Solid-State White Lighting,” Adv. Mater. 23(20), 2300–2305 (2011). [CrossRef] [PubMed]
4. J. Zhou and Z. Xia, “Luminescence color tuning of Ce3+, Tb3+ and Eu3+ codoped and tri-doped BaY2Si3O10 phosphors via energy transfer,” J. Mater. Chem. C Mater. Opt. Electron. Devices 29, 7552–7560 (2015).
5. H. Rao, W. Wang, X. Wan, L. Zhou, J. Liao, D. Zhou, Q. Lei, and X. Wang, “An improved slurry method of self-adaptive phosphor coating for white pc-LED packaging,” J. Disp. Technol. 9(6), 453–458 (2013). [CrossRef]
6. Y. K. Lee, Y. H. Kim, J. Heo, W. B. Im, and W. J. Chung, “Control of chromaticity by phosphor in glasses with low temperature sintered silicate glasses for LED applications,” Opt. Lett. 39(14), 4084–4087 (2014). [CrossRef] [PubMed]
7. X. Yi, S. Zhou, C. Chen, H. Lin, Y. Feng, K. Wang, and Y. Ni, “Fabrication of Ce:YAG, Ce,Cr:YAG and Ce:YAG/Ce,Cr:YAG dual-layered composite phosphor ceramics for the application of white LEDs,” Ceram. Int. 40(5), 7043–7047 (2014). [CrossRef]
8. J. S. Lee, S. Unithrattil, S. Kim, I. J. Lee, H. Lee, and W. B. Im, “Robust moisture and thermally stable phosphor glass plate for highly unstable sulfide phosphors in high-power white light-emitting diodes,” Opt. Lett. 38(17), 3298–3300 (2013). [CrossRef] [PubMed]
9. Y. H. Song, G. S. Han, E. K. Ji, M.-J. Lee, Y. L. Song, M. K. Jung, B. W. Jeong, H. S. Jung, and D.-H. Yoon, “The novel design of a remote phosphor ceramic plate for white light generation in high power LEDs,” J. Mater. Chem. C Mater. Opt. Electron. Devices 3(24), 6148–6152 (2015). [CrossRef]
10. B. Zhang, R. Mukherjee, H. Fujii, H. Miyagawa, P. Guang, T. Nakamura, and A. Mochizuki, “Garnet-based phosphor ceramic sheets for light emitting device,” (Google Patents, 2011).
11. Y. Shani, “Fundamentals of Planar Remote Phosphor LEDs,” Luger Research e.U. 29, 34–37 (2012).
12. J. K. Kim, H. Luo, E. F. Schubert, J. Cho, C. Sone, and Y. Park, “Strongly enhanced phosphor efficiency in GaInN white light-emitting diodes using remote phosphor configuration and diffuse reflector cup,” Jpn. J. Appl. Phys. 44(21), L649–L651 (2005). [CrossRef]
13. S. C. Allen and A. J. Steckl, “A nearly ideal phosphor-converted white light-emitting diode,” Appl. Phys. Lett. 92, 143309 (2008).
14. C. Kim, H.-A. Park, H. W. Jang, W. J. Chung, and S. Y. Kim, “All-in-one-type organic light-emitting diodes for color tuning using phosphor in glasses with Pb-free silicate powders,” Curr. Appl. Phys. 14(12), 1677–1681 (2014). [CrossRef]
15. L.-Y. Chen, J.-K. Chang, W.-C. Cheng, J.-C. Huang, Y.-C. Huang, and W.-H. Cheng, “Chromaticity tailorable glass-based phosphor-converted white light-emitting diodes with high color rendering index,” Opt. Express 23(15), A1024–A1029 (2015). [CrossRef] [PubMed]
16. H. Lin, B. Wang, J. Xu, R. Zhang, H. Chen, Y. Yu, and Y. Wang, “Phosphor-in-glass for high-powered remote-type white AC-LED,” ACS Appl. Mater. Interfaces 6(23), 21264–21269 (2014). [CrossRef] [PubMed]
17. J. S. Lee, P. Arunkumar, S. Kim, I. J. Lee, H. Lee, and W. B. Im, “Smart design to resolve spectral overlapping of phosphor-in-glass for high-powered remote-type white light-emitting devices,” Opt. Lett. 39(4), 762–765 (2014). [CrossRef] [PubMed]
18. C. G. A. Hoelen, M. P. J. Peeters, and J. G. Boerekamp, “Display device and illumination device,” (W.O. patent application 2009083887 A1, 2009).
19. L.-Y. Chen, W.-C. Cheng, C.-C. Tsai, J.-K. Chang, Y.-C. Huang, J.-C. Huang, and W.-H. Cheng, “Novel broadband glass phosphors for high CRI WLEDs,” Opt. Express 22(S3), A671–A678 (2014). [CrossRef] [PubMed]
