- Email: [email protected]
TiO2 nanocomposites for LED encapsulants
G Model
JTICE-1025; No. of Pages 8 Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants Jui-Hsiung Huang a, Chiu-Ping Li a, Cai-Wan Chang-Jian b,*, Kuen-Chan Lee c,**, Jen-Hsien Huang a,* a b c
Department of Green Material Technology, Green Technology Research Institute, Chinese Petroleum Corporation(CPC Corporation), Kaohsiung, Taiwan Department of Mechanical and Automation Engineering, I-Shou University, Kaohsiung, Taiwan Department of Fragrance and Cosmetic Science, Kaohsiung Medical University, Kaohsiung, Taiwan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 April 2014 Received in revised form 19 August 2014 Accepted 5 September 2014 Available online xxx
In this study, a uniform dispersion of TiO2 nanoparticles (NPs) in silicone for light-emitting diode (LED) encapsulation is demonstrated prepared through high-energy grinding method. Through interaction with surfactant, large clumps of the TiO2 underwent de-aggregation to form a stable dispersed solution, during the grinding process. The silicone/TiO2 composites were prepared by the blending of silicone resin with grinding TiO2 fillers in isopropyl alcohol solvent, which was removed before curing. The refractive index (RI) of surfactant-coated TiO2 NPs loaded silicone is 1.63 at 550 nm, significantly higher than that of conventional silicone (n = 1.50). The barrier properties and thermal conductivity (from 0.9 to 1.32 W/mK) of the silicone/TiO2 composites also can be enhanced, significantly. As a result, a high-power LED encapsulated with this composite showed more than 7.3% increase in the light output and better stability compared to that with the conventional silicone resin. ß 2014 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Keywords: Encapsulant LEDs Silicone Refractive index TiO2 Extraction efficiency
1. Introduction Although significant progress in LEDs has been made, higher light output is necessary to penetrate the general illumination market. In order to increase the light output to meet the demand for general illumination, high power LEDs have been developed. However, the LED efficiency generally is higher at relative low currents and as the injection current increases, the efficiency decreases gradually [1– 3]. Moreover, high power LEDs generally generate much heat leading to higher junction temperatures. The increase in junction temperature has a dramatic effect on the chip’s lifetime. Another method to enhance the light output is to solve the light-trapping issue occurred at the encapsulant (low-n)-LED chip (high-n) interface. The trapped light is originated from the total
E-mail address: [email protected]. *Corresponding authors at: Chinese Petroleum Corporation(CPC Corporation), Green Technology Research Institute, No. 2 Zuonan Rd., Nanzi Dist., Kaohsiung City 81126, Taiwan. Tel.: +886 7 5824141x7331/+886 7 6577711x3231. E-mail addresses: [email protected] (C.-W. Chang-Jian), [email protected] (K.-C. Lee), [email protected], [email protected] (J.-H. Huang). ** Corresponding author at: 100 Shih-Chuan 1st Road, Kaohsiung City 80708, Taiwan. Tel.: +886 7 3121101x2818.
internal reflection (TIR) due to the high-refractive-index contrast between the encapsulant and LED chip. Based on Snell’s law, light incident on a planar semiconductor-encapsulant interface is totally reflected if the angle of incidence is larger than critical angle. The TIR results in significant reduction in light extraction efficiency. In order to extract more light from LEDs, efforts have been made, including wet-chemical texturing of a LED surface [4–6], employing periodic photonic crystals [7,8], planar graded refractiveindex antireflection coatings [9,10], patterning of sapphire substrates [11], and shaping of LED chips [12]. Nevertheless, the fundamental obstacle of light extraction still lies in the large refractive-index contrast between the encapsulant and LED chip. The silicone based resins are the most common encapsulants used in high power LED package due to its excellent thermal stability, good insulation to moisture, strong adhesion to substrate and high transparency. However, silicone based encapsulants reveal low RI (about 1.4–1.5) which limits the light extraction efficiency. Many improvements have been made to address this issue such as adding inorganic NPs to form a composite [13–17] and synthesis of silicone/epoxy hybrid resins [18–21]. The enhancement in RI is not the only benefit for adding NPs. An increase on the thermal conductivity on the introduction of metal oxides is also important. An improvement on the thermal stability and the reduction of the thermal expansion coefficient, both are
http://dx.doi.org/10.1016/j.jtice.2014.09.008 1876-1070/ß 2014 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
2
welcomed for encapsulant, might be achieved at the same time [22]. However, if the added NPs are too large or easily aggregative they would cause scattering effect leading to the reduction of transparency. Consequently, the added NPs with small, uniform particle size and good dispersibility are critical to obtain high transparency and RI silicone/nanoparticle composite. Recently, much works have been demonstrated that the RI of silicon or other LED encapsulants can be increased by adding NPs leading to better light output. Unfortunately, the particle size of the adding NPs is still too large (20–40 nm) [13–17] and the dispersability is poor. This causes the serious decrease in transparency of the NPs/encapsulant composite. According to the Rayleigh equation [23,24], the particle size of adding NPs should be smaller than 15 nm to avoid the decrease in transparency of encapsulant due to the scattering effect. Therefore, an efficient way to produce NPs with excellent dispersability and small diameter (10 nm) is crucial. In this study, we used a wet grinding method to disperse pure TiO2 NPs. During the grinding process, large clumps of the TiO2 NPs underwent de-aggregation to form a stable dispersed solution. The TiO2 NPs with sizes around 10 nm can be blended into a silicone resin to formulate a transparent high RI composite. Moreover, the silicone/TiO2 nanocomposites reveal better water vapor transmission rate (WVTR), oxygen transmission rate (OTR) and thermal conductivity compared with that of pure silicone resin. These enhanced properties are believed to have positive impact on the durability of LEDs. We have further tested the composites as encapsulant for high power LED packages and compared its performance with that using the pure silicone. This highthroughput, simple and cheap technology can offer an approach to be easily applied for industrial application.
2. Experimental 2.1. Materials High-purity anatase TiO2 powder (99.7%, Nanostructure & Amorphous Materials), sodium dodecyl sulfate (SDS) and isopropyl alcohol were used as the starting material, surfactant and solvent, respectively. Transparent silicone resin was purchased from Dow Chemical (OE-6550A and OE-6550AB). 2.2. Preparation of the TiO2 dispersion and silicone/TiO2 composite High-energy ball milling was performed at a speed of 2000 rpm at room temperature using a batch-type grinder (JBM-B035) [25]. The milling duration was typically between 30 and 360 min; the
concentration of TiO2 and SDS were 10 wt% and 0.1 M, respectively. After high-energy grinding, the TiO2 suspensions can be diluted to any concentration without precipitation. The as-prepared TiO2 suspension was added directly into the OE-6550A under stirring for 30 min. Then, a rotary evaporator was used to remove the solvent within OE-6550A. Consequently, the mixture was blended with OE6550B until a homogeneous mixture was obtained. The mixture was vacuum vented until the bubbles exploded and the mixture was clear and transparent. The mixture was poured into a stainless steel mold and heated in an oven for 2 h at 150 8C. After this curing process, the mold was taken out of the oven and the sample was removed from the mold. 2.3. Characterizations Particle sizes and zeta potentials were measured using a particle size analyzer (Brookhaven 90 Plus Sn11408). X-ray diffraction (XRD) studies were performed using a Philips X’Pert/ MPD apparatus. The surface morphologies of the polymer films were investigated using atomic force microscope (AFM, Digital Instrument NS 3a controller equipped with a D3100 stage) and scanning electron microscope (SEM, Hitachi S-4700). The transmittance spectra were obtained using a Jasco-V-670 UV–vis spectrophotometer. The RI of the hybrid materials was determined by an Abbe-refractormeter (model:WAY). The WVTR and OTR were obtained from a commercial PERME-W3/330 instrument. The barrier property was measured under 40 8C and the relative humidity was controlled at 90%.
3. Results and discussion Fig. 1A shows the average particle sizes of the TiO2 powder as a function of grinding time. The average particle size of TiO2 decreased rapidly upon grinding for up to 180 min. With increasing grinding time, the TiO2 particles were deaggregated and broken into smaller particles due to the plastic deformation that consequently led to the reduction of particle size. The images of the TiO2 solutions with various grinding times are also shown in the inset of Fig. 1A. These images were got from the as-prepared solutions after 72 h on standing. It can be found that the TiO2 can be fully dispersed after grinding for 240 min, suggesting that the smaller particle size leads to higher charges and therefore enhances the dispersibility. In order to investigate the correlation between surface charge of the NPs and the grinding time, the zeta potential versus PH for the TiO2 solution with different grinding times was measured as shown in Fig. 1B. The isoelectric point (IEP) of TiO2 solution was found to be a function of particle size (grinding time). When the grinding times increased
Fig. 1. The relationship between particle size and grinding times. Plots of the (A) average particle size in the TiO2 powder with respect to the grinding time (inset: photograph of the TiO2 solutions obtained after different grinding times) and (B) zeta potentials of the TiO2 solutions with respect to pH.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
3
Fig. 2. The microstructure of TiO2 NPs before and after grinding. (A and B) SEM and (C and D) TEM images of TiO2 NPs (A and C) before and (B and D) after grinding for 360 min. After grinding, the particle size decreased significantly, allowing the TiO2 to disperse well without aggregation. (E) XRD patterns of the TiO2 powder before and after grinding. The lower crystallinity of the TiO2 powder after grinding was caused by the decrease in particle size and the lattice strain.
from 30 to 360 min, the IEP decreased from 8.2 to 7.0. Moreover, the surface charge of the TiO2 with longer grinding time also increases significantly, indicating that repulsive force increased leading to more disperse TiO2 solution. The representative SEM images of the TiO2 NPs before and after grinding for 6 h are shown in Fig. 2A and B. Prior to grinding, the
TiO2 powder could not be processed to form a thin film and reveals large aggregation. In contrast, the ground samples show uniform and smaller particles which can form continuous films. This indicates that the grinding process not only reduces the particle size of TiO2 but also enhances its processability. The structure of TiO2 powders was further analyzed by TEM as shown in
Fig. 3. The optical properties of silicone/TiO2 nanocomposites. (A) The transparency and RI of the silicone/TiO2 composites with various TiO2 content. The transmittance of the hybrid materials decreases significantly with TiO2 content greater than 0.5 wt% which is caused by the aggregation of TiO2 NPs.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 4
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
Fig. 4. The optical microscope images of silicone/TiO2 composites. Optical microscope images for silicone/TiO2 composites with various TiO2 content (scale bar is 50 mm); (A) 0 wt%; (B) 0.01 wt%; (C) 0.05 wt%; (D) 0.1 wt%; (E) 0.5 wt%; (F) 1.0 wt%. Large TiO2 aggregation was observed with TiO2 content greater than 0.5 wt%.
Fig. 2. Notably, the average size of the TiO2 NPs decreased significantly, from 35 to 10 nm, and the TiO2 NPs were highly dispersed, without aggregation, after grinding. The crystallinity of TiO2 powders was also examined by XRD as shown in Fig. 2E. The intensity of the XRD peaks corresponding to initial powders decrease with peak broadening after grinding. This reduced intensity and peak broaden is associated with the decrease in the grain size and lattice strains induced by high energy grinding. Based on the XRD results, the crystalline sizes calculated from Scherrer equation is 8.8 nm for the ground sample. The value of crystalline size calculated from XRD result (8.8 nm) is in good agreement with the results obtained from TEM (10 nm), indicating the ground samples are primary particles without aggregation.
Transparent silicone/TiO2 nanocomposites were prepared from transparent silicone (OE-6550) and as-prepared TiO2 nanoparticles via in situ polymerization. The transmittance spectra of silicone nanocomposites containing varying amount of TiO2 are shown in Fig. 3A. The pure silicone resin exhibits highly transparent in the visible light range. However, the transmittance of the hybrid materials decreases with the increase of TiO2 NPs. It can be seen that the transmittance decrease from 90.4 to 71.5% for the samples with 0 and 0.5 wt% TiO2 NPs. The decrease in transmittance is due to the scattering effect caused by the aggregation of TiO2 NPs under high concentration condition. At lower content of TiO2 NPs (less than 0.5 wt%), the transmittance reduce less and still can maintain 80% in the visible light range, which is favor of LED package. Fig. 3B reveals the variation of RI of silicon/TiO2 hybrids. The values of RI
Fig. 5. Representative surface morphology of the silicone/TiO2 composites. AFM images of the silicone/TiO2 composites with various TiO2 content (A) 0 wt%; (B) 0.01 wt%; (C) 0.05 wt%; (D) 0.1 wt%; (E) 0.5 wt%; (F) 1.0 wt%. Although the surface roughness of the composites increases with higher TiO2 content, the TiO2 is believed to be highly dispersed in these areas.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
5
Fig. 6. The effect of TiO2 on the WVTR of composites. (A) The WVTR of silicone/TiO2 composites with different TiO2 content. The inset is the raw data of WVTR. (B–D) Mechanism illustration of the enhancement in barrier property; (B) pure silicone; (C) silicone with dispersed TiO2; (D) silicone with aggregated TiO2. The TiO2 NPs within the silicone matrix can act as a block which impedes the diffusion of H2O and O2 molecules.
can be increased from 1.51 to 1.62 for the samples with 0 and 0.1 wt% TiO2 and maintained around 1.59 for high content of TiO2 NPs. In general, a lower RI of packaging materials make LED light a greater reflection resulted from the TIR. Therefore, the higher level of RI for the silicone/TiO2 composites would have positive effect on the LED brightness. The images of the as-prepared composites with various TiO2 content are also shown in the inset of Fig. 3B. To better understand the mechanism responsible for the variation of transmittance after adding with TiO2 NPs, we used OM and AFM to monitor the morphology of the hybrids as shown in Figs. 4 and 5. It can be seen that the composites are homogeneous without phase separation for the content of TiO2 smaller than 0.5 wt%. However, the morphology of the composites reveals large aggregation, for the content of TiO2 greater than 0.5 wt%. The large aggregation of TiO2 would cause severe light scattering effect which decreases the transparency of the composites dramatically. Based on the results, the adding amount of TiO2 NPs should be controlled within 0.5 wt% to maintain the transparency of composites. To further investigate the morphology of the area without aggregation marked by a yellow square, we also applied AFM to gain more insight. As shown in Fig. 5, with increasing the content of TiO2 NPs, the surface roughness increases gradually from 2.3 nm (0.1 wt% TiO2) to 6.7 nm (1.0 wt% TiO2). Although the roughness increases obviously, the overall
surface morphology is still homogeneous without TiO2 aggregation, indicating the TiO2 NPs are well-dispersed. Durability of LED modules is a significant factor for lowering the cost of applying LED as general illumination. One of the major concerns for module durability is corrosion because it can increase
Table 1 The OTR of silicone/TiO2 composites with various TiO2 content. Sample (content of TiO2) 0.0 wt% 0.01 wt% 0.05 wt% 0.1 wt% 0.5 wt% 1.0 wt% 2.0 wt% 76 ORT (cm3/m2 24 h 0.1 MPa)
73
71
65
64
62
61
Fig. 7. Variation of the thermal conductivity of silicone/TiO2 composites with different TiO2 contents. The inset shows the thermal conductivity of the coposites in the temperature range 25 to 170 8C.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 6
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
the resistance at the electrical interconnects leading to serious damage. Since corrosion is known to be accelerated by the presence of water and oxygen. Therefore, we also investigated the effect of adding TiO2 NPs on moisture and oxygen permeability as shown in Fig. 6 and Table 1. The addition of TiO2 NPs resulted in a significant decrease in WVTR and OTR in relation to the unfilled system. The WVTR decreases from 15.8 to 9.7 g/m2 day for the samples added 0 and 0.5 wt% TiO2, respectively. It is worthy to mention that the WVTR and OTR seems to be unchanged for further adding TiO2 NPs (>0.5 wt%). The enhancement in WVTR and OTR
can be explained by the effective penetration distance as shown in Fig. 6B. For the pure silicone, the water vapor can penetrate through the polymer matrix, directly. For the case of silicone/TiO2 composites, the TiO2 NPs act as diffusion barrier which can block the penetration of H2O and O2 molecules [26,27]. As a result, the penetrating molecules should travel a longer diffusive path as the TiO2 concentration increased. However, the TiO2 NPs have a strong tendency to aggregate at high concentration (>0.5 wt%). In this circumstance, the aggregated particles cannot offer longer diffusive path compared with that of dispersed particles.
Fig. 8. The performance of LED modules with silicone/TiO2 as encapsulant. (A) typical current-voltage characteristic of the LED module (B) efficiency of the LED modules vs current (C) spectrum of the LED module with silicone as encupsulant (D) spectrum of the LED module with silicone/TiO2 composite as encapsulant (E) durability test for the LED module with different packages under 85 8C/85% RH condition.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
7
Table 2 The photoluminescence parameters of the LED modules with different encapsulants. Encapsulant
Color temperature, CT (K)
Color rendering index (Ra)
CIE 1931 (x)
CIE 1931 (y)
0 wt% TiO2 0.03 wt% TiO2 0.05 wt% TiO2 0.10 wt% TiO2 0.50 wt% TiO2
6090 6160 6032 6078 6023
75.28 76.50 75.55 76.20 76.45
0.3207 0.3167 0.3222 0.3133 0.3234
0.3263 0.3214 0.3301 0.3211 0.3198
Therefore, the WVTR and OTR cannot reduce further by adding more TiO2 NPs. High-power LED is well known that the optical output power is degraded with the heat integration. Hence, efficient thermal management is required. Thermal conductivity is regarded as one of the most important properties that directly influence the thermal performance. In this study, we also investigate the variation of the thermal conductivity of silicone/TiO2 composites with TiO2 NPs content. It is seen that the thermal conductivity of the composites increase with increasing the TiO2 NPs. When the content of TiO2 NPs is 2 wt%, the thermal conductivity of the composites can be raised to 46% (up to 1.32 W/mK) as shown in Fig. 7. The enhancement in thermal conductivity is contributed from the high thermal conductivity of TiO2 NPs. Furthermore, the uniform distribution of TiO2 also provides a favorable morphology for heat transfer due to the higher interfacial area between TiO2 and silicon encapsulant. The composites prepared above were further tested as the encapsulant in LED modules as shown in Fig. 8. The high power LED modules (10 W) were offered from Power Opto Co., Ltd. The image of the LED module is shown in the inset of Fig. 8A which were integrated with 20 GaN-based blue-LEDs (B 24T-C2C6-JETA-M9). As depicted in Fig. 8A, the turn on voltage of the LED module is around 26.5 V. Room-temperature photoluminescence measurements with varying encapsulants were also performed to determine the radiative efficiency of LED modules (Fig. 8B). Fig. 8B shows the radiative efficiency of LED module with varying encapsulants as a function of forward current. The efficiency (at 24 mA) gradually increased from 154.5 lm/W for the encapsulant with 0 wt% TiO2 to as high as 165.8 lm/W for 0.05 wt% TiO2, and finally reduced to 126.2 lm/W (0.5 wt% TiO2) as shown in Fig. 8B. The increase of radiative efficiency can be explained by the higher RI of the silicone/TiO2 composites compared with that of pure silicone. Although the RI can be increased further by the increase of TiO2 content, the radiative efficiency decreases to 126.2 lm/W for encapsulant with 0.5 wt% TiO2. The decrease of efficiency is due to the poor transmittance resulted from the aggregation of TiO2 NPs as shown in Fig. 4 which seriously block the emission. Moreover, for all cases, the radiative efficiency decreases with a current larger than 24 mA which is originated form the efficiency droop. The emission spectrum of the LED modules with pure silicone and silicone/TiO2 composite (0.05 wt%) as encapsulant are also shown in Fig. 8C and D, respectively. It can be seen that the spectra are unchanged indicating the TiO2 NPs just can influence the RI of encapsulant but not change its absorbance. Other parameters for the LED module are also summarized in Table 2. It can be seen that the photoluminescence parameters are almost the same in there encapsulants, indicating the TiO2 NPs cannot influence the quality of photoluminescence emission. It is well-known that the TiO2 NPs possess strong photocatalytic activity. Therefore, the long-term stability of the LED module was also performed to investigate the effect of TiO2 NPs on the durability. The durability of the LED modules with different encapsulants was evaluated in a chamber controlled at 85 8C/85% relative humidity (RH) condition and under a continuous bias mode (Fig. 8E). It can be found that the relative intensities of the emission decay from 100 to 94.5% after testing for 300 h and remain a constant for pure
silicone. For the LED module incorporated with silicone/TiO2 (0.05 wt%), the relative intensities of the emission still can maintain as high as 96.2%. This indicates that the incorporation of TiO2 shows no negative impact on the long-term stability of the LED modulus. 4. Conclusion TiO2-nanoparticle-loaded encapsulants for LEDs are prepared. The TiO2 NPs were prepared through wet-grinding method with surfactant. This technology is facile, inexpensive, scalable, clean, and applicable in optoelectronics. A drastic reduction of TiO2 agglomeration is observed for the ground samples. The RI of the TiO2 NPs loaded silicone is n = 1.63, much higher than that of conventional unloaded silicone (n = 1.50). In addition, the barrier properties and thermal conductivity are also enhanced by blending with TiO2 NPs. As a result, a high-power LED encapsulated with this composite showed more than 7.3% increase in the light output and better stability. Acknowledgements The authors are grateful to the CPC Corporation and I-Shou University for financial support. We would also like to thank the research fundings from the Ministry of Science and Technology (NSC 103-2218-E-037-001) and Kaohsiung Medical University (KMU) (KMU-Q103001), Taiwan for financial support.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jtice.2014.09.008. References [1] Kim AY, Go¨tz W, Steigerwald DA, Wierer JJ, Gardner NF, Sun J, et al. Performance of high-power AlInGaN light emitting diodes. Phys Status Solidi A 2001;188:15–21. [2] Akita K, Kyono T, Yoshizumi Y, Kitabayashi H, Katayama K. Improvements of external quantum efficiency of InGaN-based blue light-emitting diodes at high current density using GaN substrates. J Appl Phys 2007;101:033104. [3] Kim MH, Schubert MF, Dai Q, Kim JK, Schubert EF, Piprek J, et al. Origin of efficiency droop in GaN-based light-emitting diodes. Appl Phys Lett 2007;91: 183507. [4] Fujii T, Gao Y, Sharma R, Hu EL, DenBaars SP, Nakamura S. Increase in the extraction efficiency of GaN-based light-emitting diodes via surface roughening. Appl Phys Lett 2004;84:855–7. [5] Han DS, Kim JY, Na SI, Kim SH, Lee KD, Kim B, et al. Improvement of light extraction efficiency of flip-chip light-emitting diode by texturing the bottom side surface of sapphire substrate. IEEE Photonics Technol Lett 2006;18:1406–8. [6] Horng RH, Yang CC, Wu JY, Huang SH, Lee CE, Wuu DS. GaN-based lightemitting diodes with indium tin oxide texturing window layers using natural lithography. Appl Phys Lett 2005;86:221101. [7] Wierer JJ, David A, Megens MM. III-Nitride photonic-crystal light-emitting diodes with high extraction efficiency. Nat Photonics 2009;3:163–9. [8] Matioli E, Rangel E, Iza M, Fleury B, Pfaff N, Speck J, et al. High extraction efficiency light-emitting diodes based on embedded air-gap photonic-crystals. Appl Phys Lett 2010;96:031108. [9] Kim JK, Chhajed S, Schubert MF, Schubert EF, Fischer AJ, Crawford MH, et al. Light-extraction enhancement of GaInN light-emitting diodes by graded-refractive-index indium tin oxide anti-reflection contact. Adv Mater 2008;20:801–4.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 8
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
[10] Chiu CH, Yu P, Chang CH, Yang CS, Hsu MH, Kuo HC, et al. Oblique electronbeam evaporation of distinctive indium-tin-oxide nanorods for enhanced light extraction from InGaN/GaN light emitting diodes. Opt Express 2009;17: 21250–56. [11] Cho J, Kim H, Kim H, Lee JW, Yoon S, Sone C, et al. Simulation and fabrication of highly efficient InGaN-based LEDs with corrugated interface substrate. Phys Status Solidi C 2005;2:2874–7. [12] Wang XH, Fu WY, Lai PT, Choi HW. Evaluation of InGaN/GaN light-emitting diodes of circular geometry. Opt Express 2009;17:22311–19. [13] He Y, Wang JA, Pei CL, Song JZ, Zhu D, Chen J. Novel epoxy-silicone thermolytic transparent packaging adhesives chemical modified by ZnO nanowires for HB LEDs. J Nanopart Res 2010;12:3019–24. [14] Chung PT, Yang CT, Wang SH, Chen CW, Chiang AS, Liu CY. ZrO 2 / epoxy nanocomposite for LED encapsulation. Mater Chem Phys 2012;136: 868–76. [15] Lem KW, Nguyen DH, Kim HN, Lee DS. Preparation and characterization of the transparent hybrids of silicone epoxy resin and titanium dioxide nanoparticles via sol–gel reactions. J Nanosci Nanotechnol 2011;11: 7202–5. [16] Mont FW, Kim JK, Schubert MF, Schubert EF, Siegel RW. High-refractive-index TiO2-nanoparticle-loaded encapsulants for light-emitting diodes. J Appl Phys 2008;103:083120. [17] Wang PC, Lin CL, Su YK. Enhancement of light extraction efficiency in GaNbased blue light-emitting diodes by doping TiO2 nanoparticles in specific region of encapsulation silicone. Jpn J Appl Phys 2013;52:08JG15. [18] Kim JS, Yang S, Bae BS. Thermal stability of sol–gel derived methacrylate oligosiloxane-based hybrids for LED encapsulants. J Sol-Gel Sci Technol 2010;53:434–40.
[19] Yang S, Kim JS, Jin J, Kwak SY, Bae BS. Cycloaliphatic epoxy oligosiloxanederived hybrid materials for a high-refractive index LED encapsulant. J Appl Polym Sci 2011;122:2478–85. [20] Gao N, Liu W, Yan Z, Wang Z. Synthesis and properties of transparent cycloaliphatic epoxy–silicone resins for opto-electronic devices packaging. Opt Mater 2013;35:567–75. [21] Kim JS, Yang S, Bae BS. Thermally stable transparent sol–gel based siloxane hybrid material with high refractive index for light emitting diode (LED) encapsulation. Chem Mater 2010;22:3549–55. [22] Wang JS, Tsai CC, Liou JS, Cheng WC, Huang SY, Chang GH, et al. Mean-time-tofailure evaluations of encapsulation materials for LED package in accelerated thermal tests. Microelectron Reliab 2012;52:813–7. [23] Novak BM. Hybrid nanocomposite materials—between inorganic glasses and organic polymers. Adv Mater 1993;5:422–33. [24] Li YQ, Fu SY, Yang Y, Mai YW. Facile synthesis of highly transparent polymer nanocomposites by introduction of core–shell structured nanoparticles. Chem Mater 2008;20:2637–43. [25] Huang JH, Huang TY, Wei HY, Ho KC, Chu CW. Wet-milled transition metal oxide nanoparticles as buffer layers for bulk heterojunction solar cells. RSC Adv 2012;19:7487–91. [26] Arunvisut S, Phummanee S, Somwangthanaroj A. Effect of clay on mechanical and gas barrier properties of blown film LDPE/clay nanocomposites. J Appl Polym Sci 2007;106:2210–7. [27] Lange J, Wyser Y. Recent innovations in barrier technologies for plastic packaging—a review. Packag Technol Sci 2003;16:149–58.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
JTICE-1025; No. of Pages 8 Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants Jui-Hsiung Huang a, Chiu-Ping Li a, Cai-Wan Chang-Jian b,*, Kuen-Chan Lee c,**, Jen-Hsien Huang a,* a b c
Department of Green Material Technology, Green Technology Research Institute, Chinese Petroleum Corporation(CPC Corporation), Kaohsiung, Taiwan Department of Mechanical and Automation Engineering, I-Shou University, Kaohsiung, Taiwan Department of Fragrance and Cosmetic Science, Kaohsiung Medical University, Kaohsiung, Taiwan
A R T I C L E I N F O
A B S T R A C T
Article history: Received 11 April 2014 Received in revised form 19 August 2014 Accepted 5 September 2014 Available online xxx
In this study, a uniform dispersion of TiO2 nanoparticles (NPs) in silicone for light-emitting diode (LED) encapsulation is demonstrated prepared through high-energy grinding method. Through interaction with surfactant, large clumps of the TiO2 underwent de-aggregation to form a stable dispersed solution, during the grinding process. The silicone/TiO2 composites were prepared by the blending of silicone resin with grinding TiO2 fillers in isopropyl alcohol solvent, which was removed before curing. The refractive index (RI) of surfactant-coated TiO2 NPs loaded silicone is 1.63 at 550 nm, significantly higher than that of conventional silicone (n = 1.50). The barrier properties and thermal conductivity (from 0.9 to 1.32 W/mK) of the silicone/TiO2 composites also can be enhanced, significantly. As a result, a high-power LED encapsulated with this composite showed more than 7.3% increase in the light output and better stability compared to that with the conventional silicone resin. ß 2014 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Keywords: Encapsulant LEDs Silicone Refractive index TiO2 Extraction efficiency
1. Introduction Although significant progress in LEDs has been made, higher light output is necessary to penetrate the general illumination market. In order to increase the light output to meet the demand for general illumination, high power LEDs have been developed. However, the LED efficiency generally is higher at relative low currents and as the injection current increases, the efficiency decreases gradually [1– 3]. Moreover, high power LEDs generally generate much heat leading to higher junction temperatures. The increase in junction temperature has a dramatic effect on the chip’s lifetime. Another method to enhance the light output is to solve the light-trapping issue occurred at the encapsulant (low-n)-LED chip (high-n) interface. The trapped light is originated from the total
E-mail address: [email protected]. *Corresponding authors at: Chinese Petroleum Corporation(CPC Corporation), Green Technology Research Institute, No. 2 Zuonan Rd., Nanzi Dist., Kaohsiung City 81126, Taiwan. Tel.: +886 7 5824141x7331/+886 7 6577711x3231. E-mail addresses: [email protected] (C.-W. Chang-Jian), [email protected] (K.-C. Lee), [email protected], [email protected] (J.-H. Huang). ** Corresponding author at: 100 Shih-Chuan 1st Road, Kaohsiung City 80708, Taiwan. Tel.: +886 7 3121101x2818.
internal reflection (TIR) due to the high-refractive-index contrast between the encapsulant and LED chip. Based on Snell’s law, light incident on a planar semiconductor-encapsulant interface is totally reflected if the angle of incidence is larger than critical angle. The TIR results in significant reduction in light extraction efficiency. In order to extract more light from LEDs, efforts have been made, including wet-chemical texturing of a LED surface [4–6], employing periodic photonic crystals [7,8], planar graded refractiveindex antireflection coatings [9,10], patterning of sapphire substrates [11], and shaping of LED chips [12]. Nevertheless, the fundamental obstacle of light extraction still lies in the large refractive-index contrast between the encapsulant and LED chip. The silicone based resins are the most common encapsulants used in high power LED package due to its excellent thermal stability, good insulation to moisture, strong adhesion to substrate and high transparency. However, silicone based encapsulants reveal low RI (about 1.4–1.5) which limits the light extraction efficiency. Many improvements have been made to address this issue such as adding inorganic NPs to form a composite [13–17] and synthesis of silicone/epoxy hybrid resins [18–21]. The enhancement in RI is not the only benefit for adding NPs. An increase on the thermal conductivity on the introduction of metal oxides is also important. An improvement on the thermal stability and the reduction of the thermal expansion coefficient, both are
http://dx.doi.org/10.1016/j.jtice.2014.09.008 1876-1070/ß 2014 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
2
welcomed for encapsulant, might be achieved at the same time [22]. However, if the added NPs are too large or easily aggregative they would cause scattering effect leading to the reduction of transparency. Consequently, the added NPs with small, uniform particle size and good dispersibility are critical to obtain high transparency and RI silicone/nanoparticle composite. Recently, much works have been demonstrated that the RI of silicon or other LED encapsulants can be increased by adding NPs leading to better light output. Unfortunately, the particle size of the adding NPs is still too large (20–40 nm) [13–17] and the dispersability is poor. This causes the serious decrease in transparency of the NPs/encapsulant composite. According to the Rayleigh equation [23,24], the particle size of adding NPs should be smaller than 15 nm to avoid the decrease in transparency of encapsulant due to the scattering effect. Therefore, an efficient way to produce NPs with excellent dispersability and small diameter (10 nm) is crucial. In this study, we used a wet grinding method to disperse pure TiO2 NPs. During the grinding process, large clumps of the TiO2 NPs underwent de-aggregation to form a stable dispersed solution. The TiO2 NPs with sizes around 10 nm can be blended into a silicone resin to formulate a transparent high RI composite. Moreover, the silicone/TiO2 nanocomposites reveal better water vapor transmission rate (WVTR), oxygen transmission rate (OTR) and thermal conductivity compared with that of pure silicone resin. These enhanced properties are believed to have positive impact on the durability of LEDs. We have further tested the composites as encapsulant for high power LED packages and compared its performance with that using the pure silicone. This highthroughput, simple and cheap technology can offer an approach to be easily applied for industrial application.
2. Experimental 2.1. Materials High-purity anatase TiO2 powder (99.7%, Nanostructure & Amorphous Materials), sodium dodecyl sulfate (SDS) and isopropyl alcohol were used as the starting material, surfactant and solvent, respectively. Transparent silicone resin was purchased from Dow Chemical (OE-6550A and OE-6550AB). 2.2. Preparation of the TiO2 dispersion and silicone/TiO2 composite High-energy ball milling was performed at a speed of 2000 rpm at room temperature using a batch-type grinder (JBM-B035) [25]. The milling duration was typically between 30 and 360 min; the
concentration of TiO2 and SDS were 10 wt% and 0.1 M, respectively. After high-energy grinding, the TiO2 suspensions can be diluted to any concentration without precipitation. The as-prepared TiO2 suspension was added directly into the OE-6550A under stirring for 30 min. Then, a rotary evaporator was used to remove the solvent within OE-6550A. Consequently, the mixture was blended with OE6550B until a homogeneous mixture was obtained. The mixture was vacuum vented until the bubbles exploded and the mixture was clear and transparent. The mixture was poured into a stainless steel mold and heated in an oven for 2 h at 150 8C. After this curing process, the mold was taken out of the oven and the sample was removed from the mold. 2.3. Characterizations Particle sizes and zeta potentials were measured using a particle size analyzer (Brookhaven 90 Plus Sn11408). X-ray diffraction (XRD) studies were performed using a Philips X’Pert/ MPD apparatus. The surface morphologies of the polymer films were investigated using atomic force microscope (AFM, Digital Instrument NS 3a controller equipped with a D3100 stage) and scanning electron microscope (SEM, Hitachi S-4700). The transmittance spectra were obtained using a Jasco-V-670 UV–vis spectrophotometer. The RI of the hybrid materials was determined by an Abbe-refractormeter (model:WAY). The WVTR and OTR were obtained from a commercial PERME-W3/330 instrument. The barrier property was measured under 40 8C and the relative humidity was controlled at 90%.
3. Results and discussion Fig. 1A shows the average particle sizes of the TiO2 powder as a function of grinding time. The average particle size of TiO2 decreased rapidly upon grinding for up to 180 min. With increasing grinding time, the TiO2 particles were deaggregated and broken into smaller particles due to the plastic deformation that consequently led to the reduction of particle size. The images of the TiO2 solutions with various grinding times are also shown in the inset of Fig. 1A. These images were got from the as-prepared solutions after 72 h on standing. It can be found that the TiO2 can be fully dispersed after grinding for 240 min, suggesting that the smaller particle size leads to higher charges and therefore enhances the dispersibility. In order to investigate the correlation between surface charge of the NPs and the grinding time, the zeta potential versus PH for the TiO2 solution with different grinding times was measured as shown in Fig. 1B. The isoelectric point (IEP) of TiO2 solution was found to be a function of particle size (grinding time). When the grinding times increased
Fig. 1. The relationship between particle size and grinding times. Plots of the (A) average particle size in the TiO2 powder with respect to the grinding time (inset: photograph of the TiO2 solutions obtained after different grinding times) and (B) zeta potentials of the TiO2 solutions with respect to pH.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
3
Fig. 2. The microstructure of TiO2 NPs before and after grinding. (A and B) SEM and (C and D) TEM images of TiO2 NPs (A and C) before and (B and D) after grinding for 360 min. After grinding, the particle size decreased significantly, allowing the TiO2 to disperse well without aggregation. (E) XRD patterns of the TiO2 powder before and after grinding. The lower crystallinity of the TiO2 powder after grinding was caused by the decrease in particle size and the lattice strain.
from 30 to 360 min, the IEP decreased from 8.2 to 7.0. Moreover, the surface charge of the TiO2 with longer grinding time also increases significantly, indicating that repulsive force increased leading to more disperse TiO2 solution. The representative SEM images of the TiO2 NPs before and after grinding for 6 h are shown in Fig. 2A and B. Prior to grinding, the
TiO2 powder could not be processed to form a thin film and reveals large aggregation. In contrast, the ground samples show uniform and smaller particles which can form continuous films. This indicates that the grinding process not only reduces the particle size of TiO2 but also enhances its processability. The structure of TiO2 powders was further analyzed by TEM as shown in
Fig. 3. The optical properties of silicone/TiO2 nanocomposites. (A) The transparency and RI of the silicone/TiO2 composites with various TiO2 content. The transmittance of the hybrid materials decreases significantly with TiO2 content greater than 0.5 wt% which is caused by the aggregation of TiO2 NPs.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 4
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
Fig. 4. The optical microscope images of silicone/TiO2 composites. Optical microscope images for silicone/TiO2 composites with various TiO2 content (scale bar is 50 mm); (A) 0 wt%; (B) 0.01 wt%; (C) 0.05 wt%; (D) 0.1 wt%; (E) 0.5 wt%; (F) 1.0 wt%. Large TiO2 aggregation was observed with TiO2 content greater than 0.5 wt%.
Fig. 2. Notably, the average size of the TiO2 NPs decreased significantly, from 35 to 10 nm, and the TiO2 NPs were highly dispersed, without aggregation, after grinding. The crystallinity of TiO2 powders was also examined by XRD as shown in Fig. 2E. The intensity of the XRD peaks corresponding to initial powders decrease with peak broadening after grinding. This reduced intensity and peak broaden is associated with the decrease in the grain size and lattice strains induced by high energy grinding. Based on the XRD results, the crystalline sizes calculated from Scherrer equation is 8.8 nm for the ground sample. The value of crystalline size calculated from XRD result (8.8 nm) is in good agreement with the results obtained from TEM (10 nm), indicating the ground samples are primary particles without aggregation.
Transparent silicone/TiO2 nanocomposites were prepared from transparent silicone (OE-6550) and as-prepared TiO2 nanoparticles via in situ polymerization. The transmittance spectra of silicone nanocomposites containing varying amount of TiO2 are shown in Fig. 3A. The pure silicone resin exhibits highly transparent in the visible light range. However, the transmittance of the hybrid materials decreases with the increase of TiO2 NPs. It can be seen that the transmittance decrease from 90.4 to 71.5% for the samples with 0 and 0.5 wt% TiO2 NPs. The decrease in transmittance is due to the scattering effect caused by the aggregation of TiO2 NPs under high concentration condition. At lower content of TiO2 NPs (less than 0.5 wt%), the transmittance reduce less and still can maintain 80% in the visible light range, which is favor of LED package. Fig. 3B reveals the variation of RI of silicon/TiO2 hybrids. The values of RI
Fig. 5. Representative surface morphology of the silicone/TiO2 composites. AFM images of the silicone/TiO2 composites with various TiO2 content (A) 0 wt%; (B) 0.01 wt%; (C) 0.05 wt%; (D) 0.1 wt%; (E) 0.5 wt%; (F) 1.0 wt%. Although the surface roughness of the composites increases with higher TiO2 content, the TiO2 is believed to be highly dispersed in these areas.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
5
Fig. 6. The effect of TiO2 on the WVTR of composites. (A) The WVTR of silicone/TiO2 composites with different TiO2 content. The inset is the raw data of WVTR. (B–D) Mechanism illustration of the enhancement in barrier property; (B) pure silicone; (C) silicone with dispersed TiO2; (D) silicone with aggregated TiO2. The TiO2 NPs within the silicone matrix can act as a block which impedes the diffusion of H2O and O2 molecules.
can be increased from 1.51 to 1.62 for the samples with 0 and 0.1 wt% TiO2 and maintained around 1.59 for high content of TiO2 NPs. In general, a lower RI of packaging materials make LED light a greater reflection resulted from the TIR. Therefore, the higher level of RI for the silicone/TiO2 composites would have positive effect on the LED brightness. The images of the as-prepared composites with various TiO2 content are also shown in the inset of Fig. 3B. To better understand the mechanism responsible for the variation of transmittance after adding with TiO2 NPs, we used OM and AFM to monitor the morphology of the hybrids as shown in Figs. 4 and 5. It can be seen that the composites are homogeneous without phase separation for the content of TiO2 smaller than 0.5 wt%. However, the morphology of the composites reveals large aggregation, for the content of TiO2 greater than 0.5 wt%. The large aggregation of TiO2 would cause severe light scattering effect which decreases the transparency of the composites dramatically. Based on the results, the adding amount of TiO2 NPs should be controlled within 0.5 wt% to maintain the transparency of composites. To further investigate the morphology of the area without aggregation marked by a yellow square, we also applied AFM to gain more insight. As shown in Fig. 5, with increasing the content of TiO2 NPs, the surface roughness increases gradually from 2.3 nm (0.1 wt% TiO2) to 6.7 nm (1.0 wt% TiO2). Although the roughness increases obviously, the overall
surface morphology is still homogeneous without TiO2 aggregation, indicating the TiO2 NPs are well-dispersed. Durability of LED modules is a significant factor for lowering the cost of applying LED as general illumination. One of the major concerns for module durability is corrosion because it can increase
Table 1 The OTR of silicone/TiO2 composites with various TiO2 content. Sample (content of TiO2) 0.0 wt% 0.01 wt% 0.05 wt% 0.1 wt% 0.5 wt% 1.0 wt% 2.0 wt% 76 ORT (cm3/m2 24 h 0.1 MPa)
73
71
65
64
62
61
Fig. 7. Variation of the thermal conductivity of silicone/TiO2 composites with different TiO2 contents. The inset shows the thermal conductivity of the coposites in the temperature range 25 to 170 8C.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 6
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
the resistance at the electrical interconnects leading to serious damage. Since corrosion is known to be accelerated by the presence of water and oxygen. Therefore, we also investigated the effect of adding TiO2 NPs on moisture and oxygen permeability as shown in Fig. 6 and Table 1. The addition of TiO2 NPs resulted in a significant decrease in WVTR and OTR in relation to the unfilled system. The WVTR decreases from 15.8 to 9.7 g/m2 day for the samples added 0 and 0.5 wt% TiO2, respectively. It is worthy to mention that the WVTR and OTR seems to be unchanged for further adding TiO2 NPs (>0.5 wt%). The enhancement in WVTR and OTR
can be explained by the effective penetration distance as shown in Fig. 6B. For the pure silicone, the water vapor can penetrate through the polymer matrix, directly. For the case of silicone/TiO2 composites, the TiO2 NPs act as diffusion barrier which can block the penetration of H2O and O2 molecules [26,27]. As a result, the penetrating molecules should travel a longer diffusive path as the TiO2 concentration increased. However, the TiO2 NPs have a strong tendency to aggregate at high concentration (>0.5 wt%). In this circumstance, the aggregated particles cannot offer longer diffusive path compared with that of dispersed particles.
Fig. 8. The performance of LED modules with silicone/TiO2 as encapsulant. (A) typical current-voltage characteristic of the LED module (B) efficiency of the LED modules vs current (C) spectrum of the LED module with silicone as encupsulant (D) spectrum of the LED module with silicone/TiO2 composite as encapsulant (E) durability test for the LED module with different packages under 85 8C/85% RH condition.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
7
Table 2 The photoluminescence parameters of the LED modules with different encapsulants. Encapsulant
Color temperature, CT (K)
Color rendering index (Ra)
CIE 1931 (x)
CIE 1931 (y)
0 wt% TiO2 0.03 wt% TiO2 0.05 wt% TiO2 0.10 wt% TiO2 0.50 wt% TiO2
6090 6160 6032 6078 6023
75.28 76.50 75.55 76.20 76.45
0.3207 0.3167 0.3222 0.3133 0.3234
0.3263 0.3214 0.3301 0.3211 0.3198
Therefore, the WVTR and OTR cannot reduce further by adding more TiO2 NPs. High-power LED is well known that the optical output power is degraded with the heat integration. Hence, efficient thermal management is required. Thermal conductivity is regarded as one of the most important properties that directly influence the thermal performance. In this study, we also investigate the variation of the thermal conductivity of silicone/TiO2 composites with TiO2 NPs content. It is seen that the thermal conductivity of the composites increase with increasing the TiO2 NPs. When the content of TiO2 NPs is 2 wt%, the thermal conductivity of the composites can be raised to 46% (up to 1.32 W/mK) as shown in Fig. 7. The enhancement in thermal conductivity is contributed from the high thermal conductivity of TiO2 NPs. Furthermore, the uniform distribution of TiO2 also provides a favorable morphology for heat transfer due to the higher interfacial area between TiO2 and silicon encapsulant. The composites prepared above were further tested as the encapsulant in LED modules as shown in Fig. 8. The high power LED modules (10 W) were offered from Power Opto Co., Ltd. The image of the LED module is shown in the inset of Fig. 8A which were integrated with 20 GaN-based blue-LEDs (B 24T-C2C6-JETA-M9). As depicted in Fig. 8A, the turn on voltage of the LED module is around 26.5 V. Room-temperature photoluminescence measurements with varying encapsulants were also performed to determine the radiative efficiency of LED modules (Fig. 8B). Fig. 8B shows the radiative efficiency of LED module with varying encapsulants as a function of forward current. The efficiency (at 24 mA) gradually increased from 154.5 lm/W for the encapsulant with 0 wt% TiO2 to as high as 165.8 lm/W for 0.05 wt% TiO2, and finally reduced to 126.2 lm/W (0.5 wt% TiO2) as shown in Fig. 8B. The increase of radiative efficiency can be explained by the higher RI of the silicone/TiO2 composites compared with that of pure silicone. Although the RI can be increased further by the increase of TiO2 content, the radiative efficiency decreases to 126.2 lm/W for encapsulant with 0.5 wt% TiO2. The decrease of efficiency is due to the poor transmittance resulted from the aggregation of TiO2 NPs as shown in Fig. 4 which seriously block the emission. Moreover, for all cases, the radiative efficiency decreases with a current larger than 24 mA which is originated form the efficiency droop. The emission spectrum of the LED modules with pure silicone and silicone/TiO2 composite (0.05 wt%) as encapsulant are also shown in Fig. 8C and D, respectively. It can be seen that the spectra are unchanged indicating the TiO2 NPs just can influence the RI of encapsulant but not change its absorbance. Other parameters for the LED module are also summarized in Table 2. It can be seen that the photoluminescence parameters are almost the same in there encapsulants, indicating the TiO2 NPs cannot influence the quality of photoluminescence emission. It is well-known that the TiO2 NPs possess strong photocatalytic activity. Therefore, the long-term stability of the LED module was also performed to investigate the effect of TiO2 NPs on the durability. The durability of the LED modules with different encapsulants was evaluated in a chamber controlled at 85 8C/85% relative humidity (RH) condition and under a continuous bias mode (Fig. 8E). It can be found that the relative intensities of the emission decay from 100 to 94.5% after testing for 300 h and remain a constant for pure
silicone. For the LED module incorporated with silicone/TiO2 (0.05 wt%), the relative intensities of the emission still can maintain as high as 96.2%. This indicates that the incorporation of TiO2 shows no negative impact on the long-term stability of the LED modulus. 4. Conclusion TiO2-nanoparticle-loaded encapsulants for LEDs are prepared. The TiO2 NPs were prepared through wet-grinding method with surfactant. This technology is facile, inexpensive, scalable, clean, and applicable in optoelectronics. A drastic reduction of TiO2 agglomeration is observed for the ground samples. The RI of the TiO2 NPs loaded silicone is n = 1.63, much higher than that of conventional unloaded silicone (n = 1.50). In addition, the barrier properties and thermal conductivity are also enhanced by blending with TiO2 NPs. As a result, a high-power LED encapsulated with this composite showed more than 7.3% increase in the light output and better stability. Acknowledgements The authors are grateful to the CPC Corporation and I-Shou University for financial support. We would also like to thank the research fundings from the Ministry of Science and Technology (NSC 103-2218-E-037-001) and Kaohsiung Medical University (KMU) (KMU-Q103001), Taiwan for financial support.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jtice.2014.09.008. References [1] Kim AY, Go¨tz W, Steigerwald DA, Wierer JJ, Gardner NF, Sun J, et al. Performance of high-power AlInGaN light emitting diodes. Phys Status Solidi A 2001;188:15–21. [2] Akita K, Kyono T, Yoshizumi Y, Kitabayashi H, Katayama K. Improvements of external quantum efficiency of InGaN-based blue light-emitting diodes at high current density using GaN substrates. J Appl Phys 2007;101:033104. [3] Kim MH, Schubert MF, Dai Q, Kim JK, Schubert EF, Piprek J, et al. Origin of efficiency droop in GaN-based light-emitting diodes. Appl Phys Lett 2007;91: 183507. [4] Fujii T, Gao Y, Sharma R, Hu EL, DenBaars SP, Nakamura S. Increase in the extraction efficiency of GaN-based light-emitting diodes via surface roughening. Appl Phys Lett 2004;84:855–7. [5] Han DS, Kim JY, Na SI, Kim SH, Lee KD, Kim B, et al. Improvement of light extraction efficiency of flip-chip light-emitting diode by texturing the bottom side surface of sapphire substrate. IEEE Photonics Technol Lett 2006;18:1406–8. [6] Horng RH, Yang CC, Wu JY, Huang SH, Lee CE, Wuu DS. GaN-based lightemitting diodes with indium tin oxide texturing window layers using natural lithography. Appl Phys Lett 2005;86:221101. [7] Wierer JJ, David A, Megens MM. III-Nitride photonic-crystal light-emitting diodes with high extraction efficiency. Nat Photonics 2009;3:163–9. [8] Matioli E, Rangel E, Iza M, Fleury B, Pfaff N, Speck J, et al. High extraction efficiency light-emitting diodes based on embedded air-gap photonic-crystals. Appl Phys Lett 2010;96:031108. [9] Kim JK, Chhajed S, Schubert MF, Schubert EF, Fischer AJ, Crawford MH, et al. Light-extraction enhancement of GaInN light-emitting diodes by graded-refractive-index indium tin oxide anti-reflection contact. Adv Mater 2008;20:801–4.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008
G Model
JTICE-1025; No. of Pages 8 8
J.-H. Huang et al. / Journal of the Taiwan Institute of Chemical Engineers xxx (2014) xxx–xxx
[10] Chiu CH, Yu P, Chang CH, Yang CS, Hsu MH, Kuo HC, et al. Oblique electronbeam evaporation of distinctive indium-tin-oxide nanorods for enhanced light extraction from InGaN/GaN light emitting diodes. Opt Express 2009;17: 21250–56. [11] Cho J, Kim H, Kim H, Lee JW, Yoon S, Sone C, et al. Simulation and fabrication of highly efficient InGaN-based LEDs with corrugated interface substrate. Phys Status Solidi C 2005;2:2874–7. [12] Wang XH, Fu WY, Lai PT, Choi HW. Evaluation of InGaN/GaN light-emitting diodes of circular geometry. Opt Express 2009;17:22311–19. [13] He Y, Wang JA, Pei CL, Song JZ, Zhu D, Chen J. Novel epoxy-silicone thermolytic transparent packaging adhesives chemical modified by ZnO nanowires for HB LEDs. J Nanopart Res 2010;12:3019–24. [14] Chung PT, Yang CT, Wang SH, Chen CW, Chiang AS, Liu CY. ZrO 2 / epoxy nanocomposite for LED encapsulation. Mater Chem Phys 2012;136: 868–76. [15] Lem KW, Nguyen DH, Kim HN, Lee DS. Preparation and characterization of the transparent hybrids of silicone epoxy resin and titanium dioxide nanoparticles via sol–gel reactions. J Nanosci Nanotechnol 2011;11: 7202–5. [16] Mont FW, Kim JK, Schubert MF, Schubert EF, Siegel RW. High-refractive-index TiO2-nanoparticle-loaded encapsulants for light-emitting diodes. J Appl Phys 2008;103:083120. [17] Wang PC, Lin CL, Su YK. Enhancement of light extraction efficiency in GaNbased blue light-emitting diodes by doping TiO2 nanoparticles in specific region of encapsulation silicone. Jpn J Appl Phys 2013;52:08JG15. [18] Kim JS, Yang S, Bae BS. Thermal stability of sol–gel derived methacrylate oligosiloxane-based hybrids for LED encapsulants. J Sol-Gel Sci Technol 2010;53:434–40.
[19] Yang S, Kim JS, Jin J, Kwak SY, Bae BS. Cycloaliphatic epoxy oligosiloxanederived hybrid materials for a high-refractive index LED encapsulant. J Appl Polym Sci 2011;122:2478–85. [20] Gao N, Liu W, Yan Z, Wang Z. Synthesis and properties of transparent cycloaliphatic epoxy–silicone resins for opto-electronic devices packaging. Opt Mater 2013;35:567–75. [21] Kim JS, Yang S, Bae BS. Thermally stable transparent sol–gel based siloxane hybrid material with high refractive index for light emitting diode (LED) encapsulation. Chem Mater 2010;22:3549–55. [22] Wang JS, Tsai CC, Liou JS, Cheng WC, Huang SY, Chang GH, et al. Mean-time-tofailure evaluations of encapsulation materials for LED package in accelerated thermal tests. Microelectron Reliab 2012;52:813–7. [23] Novak BM. Hybrid nanocomposite materials—between inorganic glasses and organic polymers. Adv Mater 1993;5:422–33. [24] Li YQ, Fu SY, Yang Y, Mai YW. Facile synthesis of highly transparent polymer nanocomposites by introduction of core–shell structured nanoparticles. Chem Mater 2008;20:2637–43. [25] Huang JH, Huang TY, Wei HY, Ho KC, Chu CW. Wet-milled transition metal oxide nanoparticles as buffer layers for bulk heterojunction solar cells. RSC Adv 2012;19:7487–91. [26] Arunvisut S, Phummanee S, Somwangthanaroj A. Effect of clay on mechanical and gas barrier properties of blown film LDPE/clay nanocomposites. J Appl Polym Sci 2007;106:2210–7. [27] Lange J, Wyser Y. Recent innovations in barrier technologies for plastic packaging—a review. Packag Technol Sci 2003;16:149–58.
Please cite this article in press as: Huang J-H, et al. Preparation and characterization of high refractive index silicone/TiO2 nanocomposites for LED encapsulants. J Taiwan Inst Chem Eng (2014), http://dx.doi.org/10.1016/j.jtice.2014.09.008