Open Access
Add to BibSonomy
Abstract
Phosphor-converted white light emitting diodes (pc-WLEDs) are used worldwide for an extensive amount of applications. The device is a complex combination of various components that introduce various technical issues: materials, electrical, chemical, thermal, and so on. All of these combined to obtain a targeted optical characteristic. While most of the pc-WLEDs are sufficient for basic illumination performance, there are still many issues to improve the pc-WLED performance. In this work, we deal with the incorporation of micron size particles of titanium oxide (TiO2) in silicone encapsulant that contains yttrium aluminum garnet (YAG) phosphor in remote phosphor pc-WLED. Based on the light output and the scattering spatial distribution measurements of the phosphor plates, we have found that several essential performance indices, like the color uniformity, the efficiency, and the amount of phosphor for the pc-WLEDs, can be adjusted by tuning the amount of TiO2 particles and thus be optimized. With a comprehensive model using a Monte-Carlo ray tracing process combined with the Mie scattering theory, two TiO2 loading conditions are revealed. The first one is the sparse condition that the TiO2 particles act as the scattering particles such as to increase the output flux to improve the efficiency of YAG. The second one is the dense condition that the TiO2 particles act more as barrier particles such so to decrease the output flux.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
The white-light source produced by phosphor-converted White Light Emitting Diode (pc-WLEDs) is nowadays widely developed to replace progressively most of the solid-state light sources [1]. Its process is based on the combination of a blue emitting die (usually InGaN) and phosphor particles (usually Yttrium Aluminum Garnet, YAG) than are incorporated in a specific packaging on the top of the die. The phosphor particles absorb partially the blue light coming from the die and down-convert it into yellow emission [2]. While the use of pc-WLEDs is well established, there is still many lacks regarding the optimization of the combination of all the components inside the lighting devices, especially about the color quality [3]. Lot of improvements can still be performed to optimize perfectly the pc-WLEDs [4]. In this context, the request of adapted modeling tools is a key issue [5].
The properties of pc-WLEDs are various and depend on the targeted use of the light. It can be the luminous intensity, the colorimetry, the color homogeneity, the spatial light distribution, the color rendering ability [6]. In order to optimize a specific property, various characteristics of the pc-WLEDs device can be the adapted like the die materials [7], the phosphor component [8], the electrical source [9], and the combination of each component. Many solutions were proposed, by adjusting the coating [10], using specific lens [11], using an adapted sapphire substrate [12], a graded refractive index multilayer phosphor package [13], or a total internal reflection collimator [14]. The use of organic-inorganic composite in the encapsulation materials is reviewed by Shen et al. [15] In all the cases, improving one property induce a drawback in another property.
Our work is dedicated to the study and optimization of the diffuser particles loading on the top of the die. These phenomena were investigated in previous publications. Two kinds of influence were analyzed as a function of the particle size. Indeed, the particle size get a direct on the Rayleigh and the Mie scattering properties [16,17].
Titanium oxide (TiO2) is known to improve the light extraction of a pc-WLEDs [18,19]. With a specific TiO2 microstructure on the surface of the die, Désieres et al. [20] improve the power of the LED, due to the design and the high refractive index of the compound. Mont et al. [18] loaded surfactant-coated nanoparticles of TiO2 inside an epoxy encapsulant positioned on the top of a pc-WLED. By adjusting the sizes and the number of particles, they optimize the effective refractive index of the encapsulant, so they improve the extraction of light through the device by reducing optical losses from scattering and Fresnel reflection. Huang et al. [19] found similar results. But, in addition, they showed that loading TiO2 particles increase the thermal conductivity, so improve the thermal stability and the reliability of the device. In both cases, they did not improve the color homogeneity. It is due to the fact that for low size particles, there is no scattering effect; the particle is almost transparent to the blue and yellow light. Chou et al. [21] in addition of the luminous flux, enhanced slightly the color uniformity of their LED device, by loading 50nm size of TiO2. Gwang et al. [22] modified the surface of nanometric TiO2 particles before loading into the encapsulant of pc-WLED. Their process improved the dispersibility of TiO2, inducing an enhancement of luminous flux and color uniformity, but lower drastically the CCT. In all these cases, any model were proposed to understand and predict the optical properties of pc-WLED.
Other influences were investigated by adding micron-size particles. In this case, the scattering characteristics of the loaded particles get a drastic influence on the optical properties of the lighting device. The most common self-influence on adding diffuser particles is the improvement on the color uniformity and the drop of the output flux. However, these effects are not the same has a function of the CCT (Correlated Color Temperature) [23]. Various kind of particle were investigated as scattering particles in order to enhance the properties of the pc-WLEDs, like silica [24,25], quantum dots [26], yttrium oxide [27], calcium carbonate [28]. It is shown that adding particles modify the color spatial homogeneity, the efficiency and the CCT [29]. Modifying the CCT is an issue, because modifying pc-WLEDs emission color change the field of application of the device. So, an efficient way of analyzing the influence of the particles loading is the compare the properties for an identical CCT. Huang et al. [30] show that adding anatase TiO2, rutile TiO2 and zirconia ZrO2 reduced considerably the CCT deviation and, in some case, enhanced the luminous flux. TiO2 rutile get the better contribution due to its scattering property [31]. But their analysis does not consider the CCT modification.
Nguyen et al. [29] mentioned that for an 8,500K device, increase the amount of diffuser particles induce the decrease of YAG phosphor to maintain an 8,500 K device. Among various diffuser (SiO2, CaCO3, CaF2, TiO2), TiO2 is the most appropriate to increase the color uniformity. It is confirmed by Lee et al. [31] who estimated the optimization in CCT distribution and the drop of luminous efficacy for various position of TiO2 particles.
All these previous publications correspond to experimental works explained by basic theoretical point of view. Yang et al. [32] did developed a powerful model using a Monte-Carlo ray tracing process combined with the Mie scattering theory to simulate the optical properties (Light scattering, efficiency, CCT, and luminous flux). This model was used in our work.
The aim of this work is to perform systematic analyses of experiments related to the color uniformity, CCT, the efficiency, the amount of phosphor as a function of TiO2 loading, in order to propose a model predicting and optimizing these properties using a Monte-Carlo ray tracing process combined with the Mie scattering theory.
1. Preparation of experimental samples
The technics used in this work were already described in previous publications [32,33]. The materials used are: yellow YAG phosphor (Cool White YAG 00902 from Hung Ta Co.), TiO2 (anatase phase, Aldrich 99.9%), silicone (KER-2600 (A/B) from ShinEtsu Chemical Co).
The YAG particles size is of 8.5 µm for D50, with a refractive index of 1.83 and a density of 4.55 g/cm3. The TiO2 particles are composed of a huge number of small particles (D50 = 1.6 µm), but particles reach a large range of size and can be up to 25 µm. TiO2 particles get a refractive index of 2.609 and a density of 4.17 g/cm3. Silicone is formed by a mixture of polymer and curing agent. It gets a refractive index of 1.41 and a density of 1 g/cm3.
Samples are composed of plates are elaborated by mixing defined amounts of YAG, TiO2 particles and a silicone on a specific mold after removing the air inside the mixture with the use of a vacuum chamber. The mold is then heated at 120°C on a hot plate during 5 min to cure the silicone. Plates get thickness from 0.6 mm to 1.5 mm, and weight concentration of YAG and TiO2 ranging respectively from 3.0% to 18.0% and from 0.00% to 0.20%. The pc-WLED configuration is in Remote Phosphor configuration [34].
Two set-ups are used to performed the experiments. The first one is used to measure the scattering light distribution of the plates [33]. A He-Ne 632.8 nm laser is used as light source. The phosphor plate is positioned in the center of the rotational stage where the power meter is measuring the scattering light. These measurements are performed for all the plates. The second set-up is used to measure the emission of the phosphor plate in a pc-WLED device as a function of the wavelength. A blue chip (EZ-700 from Cree; 449 nm) is connected on a MCPCB board, and a black acrylic tube allows positioning the plate on the top of the device and to remote the plate from blue die to reduce the thermal effect (Fig. 1). The MCPCB board is connected to an efficient home-made heat dissipation to provide a stabilization of the blue light output. Measurements are performed by placing the device on an integrating sphere (Sphere Optics, Ilumnia, injection current 50 mA, voltage from 2.91 to 2.95 V) to characterize the total spectral emission from 350 to 750 nm. The integrating sphere allows measuring the incident blue light (device without plate), then the output blue light passing through the plate and the yellow light emitted by the phosphor.
2. Modelling
The first goal is to assess the key parameters involved in the mathematical model used to fit the measured data. Indeed, the light scattering is monitored by the size particles, the concentration and the refractive index of the plate components.
The light scattering is simulated by a model [35] using a Monte-Carlo ray tracing process combined with the Mie scattering theory. The goal is to compare the experimental data with the theoretical data, as a function of particles concentrations, plate thickness and phosphor/TiO2 ratio.
The weight percentage ${w_t}$ of the particles is defined by ${w_t}(\%)\equiv \frac{{{w_p}}}{{{w_p} + {w_s}}} \times 100\%$, which ${w_p}$.is the weight of the particles, and ${w_s}$ is the weight of the silicone. The total volume of the phosphor particles is equal to
with ${D_p}$ is the density of the phosphor particles, ${D_s}$ is the density of silicone, and ${V_t}$ is the total volume of the phosphor particles and the silicone. Then, the equivalent phosphor particles number N is:
Fig. 2. (a) Photo of the investigated plates containing YAG and (b) the comparison between the simulated and the measured light scattering for plate of 1.5mm thickness and 3% of YAG
Fig. 3. (a) Photo of the investigated plates containing TiO2 and (b) the comparison between the simulated and the measured light scattering for plate of 1.0 mm thickness and 1.5% of TiO2.
The modelling process is described in previous publications [32,33,35]. In order to fit the measurement on plates containing YAG, an effective radius of 2.2 µm and an effective refractive index of 1.83 was used [35]. To fit the measurement on plates containing TiO2, an effective radius of 0.365 µm and an effective refractive index of 2.49 was used. The difference between the calculated value and data from the provider is due to the limit of the Mie theory and the Monte-Carlo ray tracing process used for modelling. Indeed, real powders contain a combination of YAG and TiO2particles, having non-spherical shapes, with a broad size distribution. So, in order for the model to fit perfectly the experimental scattering, the refractive index and the radius are adjusted. Adjusting these parameters induce that these parameters shift from a real value to an effective value.
As seen in the Fig. 2, YAG particles induce a narrow scattering, meaning that the light scattered by the YAG particle is focus in one direction. The backscattering is minor. The result is different for the TiO2 particles as shown in Fig. 3 (a). The scattering is spatially uniform, but with an important backscattering. This phenomenon implies a larger recycling effect of the light. The light hitting the TiO2 particles is then ore dispersed in the plate and get more chance to interact with YAG particles.
As for YAG and zirconium particles ZrO2 [36], measurement and experimental show identical scattering (experimental and modelling) for configurations having identical number of particles, but with different concentration of particles and thickness of plate are performed. Examples are shown in the Fig. 4.
Fig. 4. Comparison between simulated and experimental light scattering for plates (a) of 0.8 mm thickness and 1.5% of TiO2, plates of 1.6 mm thickness and 0.75% of TiO2, and plates (b) of 1.0 mm thickness and 1.5% of TiO2, plates of 2.0 mm thickness and 0.75% of TiO2.
3. Light output dependent on TiO2 concentration
The next step is to analyze the luminous properties of the samples in the integrating sphere to measure their total light outputs. As the particle number parameter defined in Eq. (2) is the key parameter to simulate the scattering. An important amount of measurements was performed in Fig. 5. The particle number replaces both the phosphor particles concentration and plates thickness to effectively analyze the optical properties.
Fig. 5. Luminous fluxes depend on the YAG particles numbers from the samples with various YAG weight concentrations of 3.0%, 5.0%, 7.5%, 14.0%, and 18.0% combining with TiO2 weight concentrations of (a) 0.00%, (b) 0.03%, (c) 0.05%, (d) 0.07%, (e) 0.12%, (f) 0.17%, and (g) 0.20%.
Results were analyzed in term of luminous flux (lumen) as a function of YAG particle number. In principle, only the YAG particles function of the wavelength conversion, while the TiO2 particles do not. So, here we consider the YAG particle numbers only rather than the total particle numbers of YAG and TiO2 both. All the results of the luminous fluxes from the samples with various YAG weight concentrations of 3.0%, 5.0%, 7.5%, 14.0%, and 18.0% combining with several different TiO2 weight concentrations of 0.00%, 0.03%, 0.05%, 0.07%, 0.12%, 0.17%, and 0.20% are shown in Fig. 5.
The luminous flux is the combination of two curves phenomena, that are relevant of cases of few and high amount of particle number. This curve reveals a maximum of luminous flux for an optimal particle number. This optimal particle number corresponds to the most proper condition for the phosphor particles uniformly suspending in the silicone resins. In case of less particles number, the phosphor particles are considered as sparse in the silicone and does not have self-influence. The light beam is scattered by few amounts of phosphor particles. In this case, increasing the particles number increase the probability to convert blue light from the die into yellow light by the phosphor, and its correlated color temperature (CCT) decreases. In case of high amount of particles number, the density of particles increases in the silicone. The phosphor particles are considered as dense in the silicone and get self-influence. The light is scattered by many particles. The probability for the light to escape out of the plate decrease as a function of particles number.
Each phenomenon can be empirically described by $P = {e^{{b_x}}}.{N^{{m_x}}}$, wherein P is the output flux in unit of lumen, N is the YAG particles number, and ${b_x}$ and ${m_x}$ (x can be either s or d) are empirical coefficients. Moreover, ${b_x}$ and ${m_x}$ are different in the case of the sparse condition (${b_s}$ and ${m_s}$) or o the dense condition (${b_d}$ and ${m_d}$). From the observations on the practical cases, ${b_s}$ is negative and ${m_s}$ is positive for the sparse condition, while ${b_d}$ is positive and ${m_d}$ is negative for the dense condition. A factor k is implemented to determine the coupling between the sparse condition and the dense condition. A large k value is relevant of a sharp shifting between the sparse condition and the dense condition, while a low value of k is relevant of a broad range of particle number where the sparse and the dense condition are mixed. So, all the samples in Fig. 5(a)∼5(g) can be fitted to Eq. (3),
With no TiO2 particles in the plate, a genetic algorithm allows to determine the fitting parameters are: $k = 0.39121$; ${b_s} ={-} 3.26385$; ${m_s} = 0.87840$; ${b_d} = 6.64564$; ${m_d} ={-} 1.01908$. The same experiments were performed on all the samples, containing various amount of YAG and TiO2. Results are shown in the Fig. 6, as well as the fitting parameters presented in Table 1.
Fig. 6. Luminous flux measured on all the phosphor plates. All the parameters are determined by fitting to Eq. (3). All the optimal points are evaluated according to Eq. (9) and enlarged in the inset.
Table 1. All the fitting parameters in Eq. (3) with the corresponding TiO2 concentrations. Uncertainties of theses value is 0.00002 (md, ms); 0.00004 (bd), 0.00003 (bs) and 0.00001 (k)
Regarding the Luminous flux shown in Fig. 6, it can be seen that except for low amount of YAG particles, increasing the amount of TiO2 decrease the luminous flux. As for other diffusing particles [19,36], added into the phosphors packaging of LED, these particles increase the scattering probability to the light beam, so decrease its probability to escape out of the device. Comparing these results to the zirconia influence on the output flux [36], the influence of TiO2 seems similar, but with a range of TiO2 amount about 50-fold less compared to the ZrO2. It is due to the high refractive index of TiO2 (2.6) compared to the refractive index of zirconia (2.13).
By fitting the parameters (${b_x}$ and ${m_x}$) of the Fig. 6, it appears that these parameters get a linear relation to the proportion of the TiO2: ${w_t}/({1 - {w_t}} )$. On the other hand, the k parameter gets an exponential influence to the proportion of TiO2.
${b_s}$ increases as the weight percentage of TiO2 increases. This parameter is related to the light recycling effect, which is enhanced by TiO2 particles as mentioned earlier. ${m_s}$ decrease as the weight percentage of TiO2 increases. This parameter is related to the Stokes Shift loss (difference of energy between the incident light and the emitted light). Indeed, enhancing the recycling effect induce an increase of the Stokes Shift loss. ${b_d}$ decreases as the weight percentage of TiO2 increases. This parameter is relevant of the extraction of the light from the plate. As the particle number increases, the light gets more chance to be trapped in the device. ${m_d}$ increases slightly as the weight percentage of TiO2 increases. This parameter would be relevant of the light recycling effect. But the impact of TiO2 particles is quite low on this parameter. k decreases as the weight percentage of TiO2 increases. As previously mentioned, this parameter is relevant of the transition between the sparse condition and the dense condition. Adding TiO2 particles, certainly due to the recycling effect, induces a smooth transition between each condition.Moreover, the optimal YAG particles number ${N_{opt}}$, inducing a maximal luminous flux, can be determined by checking the condition of. $\partial lnP/\partial ln\,N = 0$. Based on Eq. (3), the optimal YAG particles number ${N_{opt}}$ can be found as:
Then, the optimal YAG particles number ${N_{opt}}$ as a function of the weight concentration of TiO2 ${w_t}$ is shown as the curve I in Fig. 7 based on Eqs. (4) ∼ (9). In the same time, the optimal light output ${P_{opt}}$ corresponding to the optimal YAG particles number ${N_{opt}}$ for a certain ${w_t}$ of TiO2 is described in Eq. (3), and all the results of various concentrations of TiO2 are shown in the curve II in Fig. 7. Finally, the optimal output flux ${P_{opt}}$ for various TiO2 concentrations ${w_t}$ is shown as the curve III in Fig. 7. The optimal YAG particles number ${N_{opt}}$ decreases as the concentration of TiO2 increases. Indeed, for the configuration where the output flux is maximum, by increasing the amount of TiO2, the probability for an incident blue beam to be transfer into yellow light increase, because of the TiO2 scattering. The TiO2 particles improve the efficiency of YAG. Therefore, the overall maximal condition comes out as ${w_{t,max}}$ is 0.048%, ${N_{max}}$ is 136,829, and ${P_{max}}$ is 4.908 lm for this experimental pc-WLED system.
Fig. 7. Optimal YAG particles number ${N_{opt}}$ (curve I, Eqs. (4)∼9) and the light output ${P_{opt}}$ (curve II, Eq. (3)) as a function of the weight concentration of TiO2. Then, the maximal light output ${P_{max}}$ and the corresponding maximal YAG particle number ${N_{max}}$ can be found and finally the corresponding TiO2 concentration ${w_{t,max}}$.
4. Conclusion
Various experiments were performed to investigate the influence of TiO2 particles on the optical properties of remote phosphor pc-WLEDs. Experiments consist on measuring the light scattering distribution and the emission of the phosphor plate. Modelling was performed in order to simulate the scattering and the efficiency. The particle number replaces both the phosphor particles concentration and plates thickness. Optical properties were analyzed by using the particle number as a key parameter. So two well defined models were established, the first one related to the output flux P (Eq. (3)), the second one related to the optimal particles number Nopt (Eq. (9))
Micron size TiO2 particles can be used to adjust the targeted properties of a pc-WLEDs to effectively increase the total light output and hence to enhance their energy conversion efficiency. TiO2 particle get various influence because the TiO2 particles increase the pathway of the incident blue light inside the package, so increase the possibility for the blue beam to be absorbed by a phosphor particle. In other words, it increases the efficiency of the YAG phosphor. But adding too much TiO2 particles increase too much the pathway and the scattering, that implies a decrease in the probability of the light to escape out of the device. Empirically, at low amount of TiO2, the condition of the particles inside the plate can be considered as sparse condition, i.e. particles do not have self-influence. The light beam is scattered by few numbers of particles. So, the output flux increases slightly. But adding more TiO2 shift the condition into dense condition, so induce a drastic decrease of the efficiency. Finally, the TiO2 particles improve the efficiency of YAG. Therefore, the overall maximal condition comes out as wt,max is 0.048%, Nmax is 136,829, and Pmax is 4.908 lm for this experimental pc-WLED system.
Funding
Ministry of Science and Technology, Taiwan (105-2911-I-008 -503, 107-2221-E-008-068, 108-2221-E-008 -092 -MY2, 109-2221-E-008-087-MY2, 109-2622-E-008-026, 111-2221-E-008-100); Campus France (PPL, N 26863 QK); Centre National de la Recherche Scientifique (27420).
Acknowledgements
The author would like to thank Breault Research Organization (BRO), Inc. for sponsoring the ASAP software program.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. Pimputkar, J. S. Speck, S. P. DenBaars, and S. Nakamura, “Prospects for LED lighting,” Nat. Photonics 3(4), 180–182 (2009). [CrossRef]
2. Y. H. Kim, N. S. Viswanath, S. Unithrattil, H. J. Kim, and W. B. Im, “Phosphor plates for high-power LED applications: challenges and opportunities toward perfect lighting,” ECS Journal of solid-state science and technology 7(1), R3134–R3147 (2018). [CrossRef]
3. B.-M. Song, B. Han, and J.-H. Lee, “Optimum design domain of LED-based solid-state lighting considering cost, energy consumption and reliability,” Microelectron. Reliab. 53(3), 435–442 (2013). [CrossRef]
4. D. Karunatilaka, F. Zafar, V. Kalavally, and R. Parthiban, “LED based indoor visible light communications: State of the art,” IEEE communications surveys & tutorials 17, 1649–1678 (2015). [CrossRef]
5. V. C. Bender, T. B. Marchesan, and J. M. Alonso, “Solid-state lighting: A concise review of the state of the art on LED and oLED modeling,” IEEE Industrial Electronics Magazine 9(2), 6–16 (2015). [CrossRef]
6. S. Jost-Boissard, P. Avouac, and M. Fontoynont, “Assessing the colour quality of LED sources: Naturalness, attractiveness, colourfulness and colour difference,” Lighting research & technology 47, 769–794 (2015). [CrossRef]
7. P. Roussel and Y. Développement, “How GaN-on-Si could disrupt the current equilibrium of the booming LED industry,” in “Proc. Compound Semicond. Manuf. Technol. (CS ManTech) Conf,” (Citeseer, 2013), pp. 273–274.
8. P. F. Smet, A. B. Parmentier, and D. Poelman, “Selecting conversion phosphors for white light-emitting diodes,” J. Electrochem. Soc. 158(6), R37 (2011). [CrossRef]
9. M.-H. Kim, M. F. Schubert, Q. Dai, J. K. Kim, E. F. Schubert, J. Piprek, and Y. Park, “Origin of efficiency droop in GaN-based light-emitting diodes,” Appl. Phys. Lett. 91(18), 183507 (2007). [CrossRef]
10. C. Sommer, P. Hartmann, P. Pachler, M. Schweighart, S. Tasch, G. Leising, and F. P. Wenzl, “A detailed study on the requirements for angular homogeneity of phosphor converted high power white LED light sources,” Opt. Mater. 31(6), 837–848 (2009). [CrossRef]
11. K. Wang, D. Wu, F. Chen, Z. Liu, X. Luo, and S. Liu, “Angular color uniformity enhancement of white light-emitting diodes integrated with freeform lenses,” Opt. Lett. 35(11), 1860–1862 (2010). [CrossRef]
12. K.-C. Huang, T.-H. Lai, and C.-Y. Chen, “Improved CCT uniformity of white LED using remote phosphor with patterned sapphire substrate,” Appl. Opt. 52(30), 7376–7381 (2013). [CrossRef]
13. M. F. Schubert, D. J. Poxson, F. W. Mont, J. K. Kim, and E. F. Schubert, “Performance of antireflection coatings consisting of multiple discrete layers and comparison with continuously graded antireflection coatings,” Appl. Phys. Express 3(8), 082502 (2010). [CrossRef]
14. C. Prins, C. Schneider, W. IJzerman, T. Tukker, and J. ten Thije Boonkkamp, “Freeform TIR collimators for the removal of angular color variation in white LED spotlights,” SPIE Proc. 8834, 106–115 (2013). [CrossRef]
15. J. Shen and Y. Feng, “Recent advances in encapsulation materials for light emitting diodes: a review,” Silicon 2022, 1 (2022). [CrossRef]
16. N. T. Tran, J. P. You, and F. G. Shi, “Effect of phosphor particle size on luminous efficacy of phosphor-converted white LED,” J. Lightwave Technol. 27(22), 5145–5150 (2009). [CrossRef]
17. N. M. Nursam, J. Hidayat, E. Shobih, L. Rosa, and Pranoto, “A comparative study between Titania and Zirconia as material for scattering layer in dye-sensitized solar cells,” J. Phys.: Conf. Ser. 1011, 012003 (2018). [CrossRef]
18. F. W. Mont, J. K. Kim, M. F. Schubert, E. F. Schubert, and R. W. Siegel, “High-refractive-index TiO2-nanoparticle-loaded encapsulants for light-emitting diodes,” J. Appl. Phys. 103(8), 083120 (2008). [CrossRef]
19. J.-H. Huang, C.-P. Li, C.-W. Chang-Jian, K.-C. Lee, and J.-H. Huang, “Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants,” Journal of the Taiwan Institute of Chemical Engineers 46, 168–175 (2015). [CrossRef]
20. Y. Désières, D. Y. Chen, D. Visser, C. S. Schippers, D. Vaufrey, P. Demars, F. Levy, C. Largeron, Q. Lalauze, and S. Anand, “Nanoparticle-based microstructures for light extraction enhancement in nitride-based LEDs”, Proc. SPIE 10554, Light-Emitting Diodes: Materials, Devices, and Applications for Solid State Lighting XXII (SPIE, 2018).
21. Y. F. Chou, C. F. Chen, S. P. Ying, and Y. Y. Yeh, “The effects of TiO2 diffuser-loaded encapsulation on corrected color temperature uniformity of remote phosphor white LEDs,” Appl. Sci. 9(4), 675–684 (2019). [CrossRef]
22. Y. S. Gwang, J. Inseok, J. Sie-Wook, A. Sun-Hee, K. Ja-Yeon, Y. K. So, and S. Gidong, “Controlling the Surface Properties of TiO2 for Improvement of the Photo-performance and Color Uniformity of the Light-emitting Diode Devices,” J. Ind. Eng. Chem. 94, 180–187 (2021). [CrossRef]
23. Á. Borbély and S. G. Johnson, “Performance of phosphor-coated light-emitting diode optics in ray-trace simulations,” Opt. Eng. 44(11), 111308 (2005). [CrossRef]
24. N. D. Quoc Anh, M.-F. Lai, H.-Y. Ma, and H.-Y. Lee, “Enhancing of correlated color temperature uniformity for multi-chip white-light LEDs by adding SiO2 in phosphor layer,” Journal of the Chinese Institute of Engineers 38(3), 297–303 (2015). [CrossRef]
25. M.-F. Lai, N. D. Quoc Anh, H.-Y. Ma, and H.-Y. Lee, “Scattering effect of SiO2 particles on correlated color temperature uniformity of multi-chip white light LEDs,” Journal of the Chinese Institute of Engineers 39(4), 468–472 (2016). [CrossRef]
26. K.-J. Chen, C.-C. Lin, H.-V. Han, C.-Y. Lee, S.-H. Chien, K.-Y. Wang, S.-H. Chiu, Z.-Y. Tu, J.-R. Li, T.-M. Chen, X. Li, M.-H. Shih, and H.-C. Kuo, “Wide-range correlated color temperature light generation from resonant cavity hybrid quantum dot light-emitting diodes,” IEEE J. Sel. Top. Quantum Electron. 21(4), 23–29 (2015). [CrossRef]
27. M. Zhang, X.-H. Li, Z.-X. Wang, Q.-Y. Hu, and H.-J. Guo, “Synthesis of Y2O3:Eu3+ phosphors by surface diffusion and their photoluminescence properties,” Trans. Nonferrous Met. Soc. China 20(1), 115–118 (2010). [CrossRef]
28. L. Yang, Z. Lv, Y. Jiaojiao, and S. Liu, “Effects of melamine formaldehyde resin and CaCO3 diffuser-loaded encapsulation on correlated color temperature uniformity of phosphor-converted LEDs,” Appl. Opt. 52(22), 5539–5544 (2013). [CrossRef]
29. A. Q. Nguyen and V. H. Nguyen, “Application of CaCO3, CaF2, SiO2, and TiO2 particles to silicone lens for enhancing angular color uniformity of white LED lamp,” in “Journal of Physics: Conference Series,” IOP Publishing885, 012022 (2017).
30. K.-C. Huang, Y.-R. Huang, T.-L. Chuang, S.-Y. Ting, S. H. Tseng, and J.-E. Huang, “Incorporation of anatase TiO2 particles into silicone encapsulant for high-performance white LED,” Mater. Lett. 143, 244–247 (2015). [CrossRef]
31. K.-C. Lee, S.-M. Kim, and J.-H. Moon, “The effects of Titania diffuser on angular color homogeneity in the phosphor conformal coated white LEDs,” in Tenth International Conference on Solid State Lighting, SPIE Proc.7784, 183–188 (2010).
32. T.-H. Yang, C.-C. Chen, C.-Y. Chen, Y.-Y. Chang, and C.-C. Sun, “Essential factor for determining optical output of phosphor-converted LEDs,” IEEE Photonics J. 6, 1–9 (2014). [CrossRef]
33. T.-H. Yang, C.-Y. Chen, Y.-Y. Chang, B. Glorieux, Y.-N. Peng, H.-X. Chen, T.-Y. Chung, T.-X. Lee, and C.-C. Sun, “Precise simulation of spectrum for green emitting phosphors pumped by a blue LED die,” IEEE Photonics J. 6, 1–10 (2014). [CrossRef]
34. B.-C. Li, N.-Z. Zhuo, W.-Q. Li, Q.-Y. He, R.-X. Huang, G.-X. Liu, E.-G. Ye, and H.-B. Wang, “The preparation and properties study of high CRI white LED based on remote phosphor technology,” Guang Pu Xue Yu Guang Pu Fen Xi 37, 728–732 (2017).
35. C.-C. Sun, C.-Y. Chen, H.-Y. He, C.-C. Chen, W.-T. Chien, T.-X. Lee, and T.-H. Yang, “Precise optical modeling for silicate-based white LEDs,” Opt. Express 16(24), 20060–20066 (2008). [CrossRef]
36. Y.-Y. Chang, B. Glorieux, C.-H. Hsu, C.-C. Sun, L.-Z. Yu, T.-H. Yang, and T.-Y. Chung, “Influence of ZrO2 particles on the optical properties of pc-LEDs,” Optical Materials 55, 55–61 (2016). [CrossRef]
