Open Access
Add to BibSonomyAbstract
We proposed a new device structure to dynamically tune the correlated color temperature (CCT) of a white light-emitting-diode (WLED). The key component is a dynamic color filter, consisting of a liquid crystal (LC) cell sandwiched between two cholesteric LC films whose Bragg reflection band covers the blue wavelength of the WLED. When a voltage is applied to the LC cell, the transmittance of blue light is changed, while the longer wavelength part remains unaffected, resulting in a tunable CCT. Validated by experiment, our design exhibits several advantages, such as reasonably wide tuning range (6916K to 3253K), low operation voltage (~3.2 V), simple device configuration, and low cost. It is a strong contender for next generation smart lighting.
© 2016 Optical Society of America
1. Introduction
Light emitting diodes (LEDs) have found widespread applications, ranging from general lighting, display backlight, automobile front light, public displays, to house decorations [1]. Especially after the invention of high efficiency blue GaN LEDs [2–5], phosphor-converted white LED (pc-WLED) has become an indispensable part of our daily lives, which has revolutionary impact on energy saving globally by replacing the conventional light bulb. Recently, “smart” solid state lighting is emerging and it plays a key role in next generation lighting applications [6,7]. One of the attractive features of smart lighting is tunable correlated color temperature (CCT). As an important characteristic of light source, color temperature has profound effect on human beings, including comfort, productivity, brightness perception and health [8]. Therefore, light source with a proper CCT is of great significance. However, conventionally the color temperature is fixed once the light source is chosen. But in real lives, it is highly desirable to have a real-time tunable CCT light source to meet the varying conditions (e.g. weather, season, daytime, etc.) and multiple functions (working, relaxing, meeting, etc.).
To achieve this goal, several methods have been proposed. The most straightforward one is mixing red, green, blue (RGB) LEDs together and driving them independently [9–11]. Theoretically speaking, all the color temperatures could be obtained by adjusting the RGB intensities separately. However, the required optical and electrical components are relatively complicated, resulting in high cost [9,12]. Moreover, present green LEDs are not efficient enough for general lighting applications. A CCT tunable light source with quantum dot (QD) has been proposed by tuning the absorption and emission spectra of green and red quantum dots [13]. It performs well but the required voltage is fairly high [14–16]. Recently, our group tried to combine liquid crystal (LC) technology with LED to achieve tunable CCT WLEDs [17]. The device configuration is fairly simple and cost is low, but the dichroic ratio of the employed fluorescent dye is inadequate so that the tunable CCT range is limited.
In this paper, we proposed a new design to dynamically tune the CCT of a WLED. The device configuration and working principles will be described in Sec. 2. To demonstrate feasibility, in Sec. 3 we will present our experimental results. To improve performance, in Sec. 4 we analyze the factors affecting our experimental results and propose methods for widening CCT along the blackbody locus.
2. Device configuration and working mechanism
Color temperature is mainly determined by the spectral power distribution (SPD) of a light source. Typically, white light with higher CCT will have more lumens in short wavelength region and it appears more bluish. While, more lumens in longer wavelength (red or yellow light) will lead to lower CCT white light, looking more yellowish, like candle or sunset. Therefore, to tune the CCT of a light source, the basic idea is to change the relative intensity of blue (short wavelength) and yellow light (longer wavelength) [13,18].
Based on this principle, we propose to use a dynamic color filter to control the blue light transmission, while keeping the longer wavelengths unchanged. Figure 1 shows the device configuration, where a dynamic color filter is placed above a conventional WLED. The color filter consists of an LC cell [here we use vertical alignment (VA)] sandwiched between two cholesteric liquid crystal (CLC) films.
Different from commonly used rod-like nematic liquid crystals for flat panel displays, CLC is formed by rod-like molecules whose directors are self-organized in a helical structure [19,20]. In the planes perpendicular to the helical axis, the LC directors are continuously rotated along the helical axis. Due to the helical structure, CLC exhibits Bragg reflection with central wavelength (λ0) and reflection bandwidth (Δλ) as [19]:
where p is the pitch length, HTP the helix twisting power of chiral dopant, c the concentration of chiral dopant, <n> the average refractive index of LC host, ne the extraordinary refractive index, no the ordinary refractive index, and Δn the LC birefringence.From Eqs. (1)–(3), the CLC reflection band can be controlled easily by the pitch length. In addition, to reduce the number of substrates and weight for real applications we can use polymeric CLC films instead of cells [21]. In our design, these two CLC films reflect right-handed circularly polarized (RCP) blue light. Thus, at the voltage-off state (V = 0), only left-handed circularly polarized (LCP) blue light passes through the first CLC film, VA cell, and then the second CLC film. In this case, all the blue light goes out, leading to a high CCT. Please note that the reflected RCP blue light by the first CLC film will be reflected back by the metal cavity and converted to LCP, i.e. the blue light is recycled and high efficiency is obtained. As the voltage exceeds a threshold (Vth), the LC directors (whose dielectric anisotropy Δε < 0) in the VA cell will be reoriented to be parallel to the substrate. In this case, the VA cell functions as a half-wave plate. It converts the transmitted LCP blue light from first CLC film into RCP, which is in turn blocked by the second CLC film. Thus, little blue light transmits through the dynamic filter, resulting in a low CCT. In the whole process, the VA cell serves as a switch to convert LCP blue light into RCP, thus controlling the transmittance of the blue light. The resultant CCT can be controlled by the applied voltage.
3. Experimental results
In experiment, we doped 3.3 wt% chiral compound R5011 (HCCH, China) into LC host BL-038 (Merck, Germany) whose ne = 1.799 and no = 1.527 [22]. Then we filled the mixture into two commercial homogeneous cells with cell gap d = 8 μm. Due to the effect of chiral dopant, LC directors are rotated helically in the CLC cell, resulting in circularly polarized reflection band, as Fig. 2 shows. As expected, the reflection band is centered at λo ≈450 nm and bandwidth is Δλ ≈80 nm. This band covers most of the blue light emitted from commercial blue LEDs. Please note that to prove concept, here we used two CLC cells with glass substrates, but for real applications we should use CLC polymeric films, as demonstrated in [21] in order to reduce device thickness and weight.
For the tunable LC cell, we chose MLC-6882 (Δn = 0.097, and Δε = −3.1) [23] and filled it into a vertical-alignment (VA) cell made of indium tin oxide (ITO) glass substrates. The cell gap was controlled at 5 μm by spacer balls. Combining with two CLC test cells, the dynamic color filter for blue light was fabricated successfully. The transmission spectra of such a color filter were recorded by HR2000 CG-UV-NIR high-resolution spectrometer (Ocean Optics), as depicted in Fig. 3 with respect to different applied voltages. Clearly, the transmittance of blue light decreases gradually as the voltage increases. Meanwhile, the longer wavelength region remains unchanged. In the on-state voltage (Vπ ≈3.2Vrms) where the VA cell works as a half-wave plate, only about 10% of the blue light is transmitted. The reasons for this 10% light leakage will be discussed later. In brief, the VA cell is a half-wave plate only for a laser beam, but not achromatic for the entire blue band.
In our experiment, we adopted a commercial pc-WLED (blue LED with yellow phosphor) as the light source and the measured results are shown in Fig. 4(a). Same as the results shown in Fig. 3, the intensity of blue light (405 nm ~485 nm) decreases as the voltage increases. The remaining part (longer wavelengths) doesn’t change at all. As a result, the corresponding CCT decreases gradually, as illustrated in Fig. 4(b). Initially, the CCT of the WLED is 6916K, which is fairly high and is a cold light source. As the voltage increases, the CCT remains unchangedwhen V< 2 Vrms. This voltage is called Fréedericksz transition threshold [19], below which the LC directors won’t be reoriented. Above 2V, the LCs start to bend. The polarization of the output blue light after the VA cell [Fig. 1] gradually changes from LCP to RCP so that the transmittance after the second CLC cell decreases. During this process, the CCT decreases because of less blue light. When V = Vπ (~3.2 V), the VA cell becomes a half-wave plate, and CCT reaches the lowest (3253K). The tuning range is about 3700K, which is wide enough for most lighting applications. If we keep increasing the voltage, the effective phase of VA cell will increase as well; then CCT will bounce back since VA cell is no longer a half-wave plate.
Fig. 4 (a) Measured spectral power distribution of white light with different voltages, (b) relationship between CCT and applied voltage, and (c) CCT in C.I.E. 1960 chromaticity diagram obtained by different applied voltages.
Next, we plot the chromaticity in CIE 1960 color space to visualize the device performance more clearly. As can be seen from Fig. 4(c), the CCT variation track from 3200K to 6900K is roughly a straight segment. Because the LC directors can be controlled continuously by the applied voltage, the CCT can be tuned precisely. But there is a concern that the CCT variation track is deviated from the blackbody locus, which may cause the light not “white”. This characteristic could be quantified by the CIE chromaticity coordinates deviation (Duv). For real applications, Duv should be less than 0.005 [24,25]. But in our case, this value is larger than 0.02 for some CCTs. To improve this, the spectrum of original WLED should be chosen carefully and some optimizations are needed. In next section, we will give one example of WLED with optimized SPD for achieving better performance.
In addition to CCT, for general lighting, other parameters like color rendering index (CRI) and luminous efficacy of radiation (LER) [26] are equally important. All these properties are summarized and plotted in Fig. 5. Because of the varied spectral power distribution, the CRI and LER are tuned accordingly. Similar to CCT, there is a threshold voltage, above which, CRI and LER start to change. As the voltage increases, the CRI increases first until 2.8 V, and then slightly decreases. This is believed to originate from the SPD variations with applied voltages. As shown in Fig. 4(a), below 2.8 V the whole blue region decreases with the increased voltage, but above 2.8V blue region is split into two parts, where the longer wavelength part (~488 nm) doesn’t change while the shorter wavelength part (~465 nm) keeps decreasing. As a result, the obtained CRI decreases slightly until 3.2 V. If we further increase the voltage (> 3.2 V), the VA cell is no longer a half-wave plate, then CRI will bounce back like CCT does. During the whole process, the CRI is tuned from 72 to 85. To boost CRI to over 90, other light sources should be considered, such as blue LED with green and red phosphors.
To quantitatively evaluate the light spectral efficiency, luminous efficacy of radiation (LER) is typically employed, which is defined as [26]:
where S(λ) is the normalized spectral power distribution, and V(λ) is the photopic eye sensitivity function. The maximum value of LER is 683 lm∕W for a monochromatic light source at λ = 555 nm. However, for a broadband light source (e.g., white light) its LER is usually much lower. Please note that here LER reflects how efficiently the output light can be converted to the brightness perception of human eye, which is an inherent property of the output spectrum. Therefore, the LER increases as the voltage increases due to the varied spectra. But for real applications, the total power consumption will increase undoubtedly with increased operation voltage, or electrical to optical power efficiency (known as luminous efficiency or luminous efficacy of the source) will decrease, since blue light is blocked and sacrificed. As the tuning part, LC cell will consume some electrical power, but it should be really low because LC is a voltage device, and the operation voltage is lower than 5 V.4. Discussion
4.1 Light source
To prove concept, in our experiment we employed a commercial pc-WLED as the light source. The performance is fairly good in terms of CCT tuning range (6916K to 3253K), but the chromaticity coordinate deviation (Duv) is far from satisfaction because of the mismatched SPD of original WLED.
To get better performance, in simulation, we slightly tuned the SPD of light source, as Fig. 6(a) depicts. The central wavelength of the blue band moves from ~465 nm [Fig. 4(a)] to 475 nm. As the voltage increases, the CCT decreases from 7500K to 3100K and CRI is increased from 51 to 71. At the same time, Duv keeps lower than 0.005, which is quite acceptable for real applications. Figure 6(b) depicts the simulated CCT in CIE 1960 chromaticity diagram. The magnified chart shows how closely the CCT follows the blackbody locus.
Fig. 6 (a) Simulated spectral power distribution of white light with different voltages, and (b) CCT in C.I.E. 1960 chromaticity diagram with different voltages. In (b), the magnified chart shows how closely the CCT follows the blackbody locus.
In our experiment, the obtained lowest CCT is 3253K. But for real applications, the required CCT value would be much lower (~2700K). To get warmer white light, more lumens in yellow region should be considered. Meanwhile, the peak wavelength of yellow light should move to a slightly longer wavelength. By choosing a WLED with optimized spectra, it should be fairly easy to achieve low CCT white light, like 2700K. In addition to pc-WLED, our design is also compatible to other existing light sources, including light bulbs, R/G/B WLED, emerging organic LED [27,28], and quantum-dot LED [29–31].
4.2 LC cell
Here we use vertical-alignment LC cell as the tuning part in the dynamic color filter. For practical applications, other LC modes like homogeneous (HG) LC cell can also be considered. The difference is that HG cell can be designed to work as a half-wave plate at V = 0. As V >Vth, the LC directors will tilt up so that the effective phase retardation decreases. In this case, blue light will be blocked initially and warm light is obtained. The CCT of the light source will increase as the voltage increases. Another parameter of LC cell is response time, which is ~5 ms in our VA cell. To get faster response, low viscosity LC materials could be used [32–34]. Also, in our experiment, a single-domain VA cell was employed, which would affect the illuminations at different angles. To widen the acceptance angle, we could use a multi-domain structure, which has been commonly used in LCD TVs.
Next, we investigate the dispersion effect of the VA cell. Theoretically speaking, all the blue light should be blocked if the LC cell can convert LCP to RCP completely. But due to wavelength and birefringence dispersion, our VA cell is not an ideal achromatic half-wave plate. This accounts for the ~10% blue light leakage in our experiment at V = Vπ. We simulated the voltage-transmittance (VT) curves of our VA cell for three blue LED wavelengths (440 nm ~480 nm), and the results are depicted in Fig. 7. Clearly, for different wavelengths, half-wave plate condition is satisfied at different voltages (i.e. voltage with peak transmittance). Overall speaking, a shorter wavelength has a lower Vπ. From Fig. 7, the difference in Vπ for the three blue LED wavelengths is merely ~0.2 V.
4.3 Cholesteric LC films
Based on Eqs. (1)–(3), the reflection band of CLC films could be tuned easily. Here we controlled the band to cover the blue region (405 nm ~485 nm). For different purposes, other reflection band could also be obtained, like green or red. If we combine several dynamic color filters together for different reflection bands (e.g. R/G/B), then the whole SPD of light source can be tuned. In this case, better performance is expected and such devices will have multiple applications other than lighting. Another potential concern for CLC films is the temperature stability, as the working temperature will increase during LED operations. However, this problem can be mitigated by using a LC mixture with high clearing temperature (Tc > 100°C) [35].
5. Conclusion
We have proposed a CCT tunable WLED using a dynamic color filter. This filter is comprised of a voltage-tunable LC cell sandwiched between two CLC films. By applying a voltage to the LC cell, the transmittance of blue light could be tuned precisely, and the resultant CCT changed accordingly. In experiment, we obtained about 3700K CCT tuning range, which is wide enough for most lighting applications. Meanwhile, by optimizing the SPD of a light source, we can achieve better color performances (e.g. Duv). Our design utilizes two mature technologies: LCD and LED. It is a strong candidate for next generation smart lighting.
Acknowledgments
The authors would like to thank Dr. Zhunyue Luo and Dr. James Larimer for helpful discussions, and AFOSR for partial financial support under contract No. FA9550-14-1-0279.
References and links
1. E. F. Schubert, T. Gessmann, and J. K. Kim, Light Emitting Diodes (John Wiley & Sons, Inc., 2005).
2. S. Nakamura, T. Mukai, and M. Senoh, “High-power GaN pn junction blue-light-emitting diodes,” Jpn. J. Appl. Phys. 30(2), L1998–L2001 (1991). [CrossRef]
3. S. Nakamura, T. Mukai, and M. Senoh, “Candela‐class high‐brightness InGaN/AlGaN double‐heterostructure blue‐light‐emitting diodes,” Appl. Phys. Lett. 64(13), 1687–1689 (1994). [CrossRef]
4. S. Nakamura, M. Senoh, N. Iwasa, and S. I. Nagahama, “High-brightness InGaN blue, green and yellow light-emitting diodes with quantum well structures,” Jpn. J. Appl. Phys. 34(2), L797–L799 (1995). [CrossRef]
5. S. Nakamura, “The roles of structural imperfections in InGaN-based blue light-emitting diodes and laser diodes,” Science 281(5379), 956–961 (1998). [CrossRef] [PubMed]
6. E. F. Schubert and J. K. Kim, “Solid-state light sources getting smart,” Science 308(5726), 1274–1278 (2005). [CrossRef] [PubMed]
7. J. Y. Tsao, M. H. Crawford, M. E. Coltrin, A. J. Fischer, D. D. Koleske, G. S. Subramania, G. T. Wang, J. J. Wierer, and R. F. Karlicek, “Toward smart and ultra-efficient solid-state lighting,” Adv. Opt. Mater. 2(9), 809–836 (2014). [CrossRef]
8. P. R. Boyce, Human Factors in Lighting, 2nd ed. (Taylor & Francis, 2003).
9. S. Muthu, F. J. P. Schuurmans, and M. D. Pashley, “Red, green, and blue LEDs for white light illumination,” IEEE J. Sel. Top. Quantum Electron. 8(2), 333–338 (2002). [CrossRef]
10. A. Zukauskas, R. Vaicekauskas, F. Ivanauskas, G. Kurilcik, Z. Bliznikas, K. Breive, J. Krupic, A. Rupsys, A. Novickovas, P. Vitta, A. Navickas, V. Raskauskas, M. S. Shur, and R. Gaska, “Quadrichromatic white solid-state lamp with digital feedback,” Proc. SPIE 5187, 185–198 (2004). [CrossRef]
11. “Osram Opto unveils brilliant-mix LED mixing concept,” LEDs Mag., May 2011. http://www.ledsmagazine.com/articles/2011/05/osram-opto-unveils-brilliant-mix-led-mixing-concept.html.
12. J. Hecht, “Changing the lights: are LEDs ready to become the market standard,” Opt. Photonics News 23(3), 44–50 (2012). [CrossRef]
13. J. Y. Tsao, I. Brener, D. F. Kelley, and S. K. Lyo, “Quantum-dot-based solid-state lighting with electric-field-tunable chromaticity,” J. Disp. Technol. 9(6), 419–426 (2013). [CrossRef]
14. S. A. Empedocles and M. G. Bawendi, “Quantum-confined stark effect in single CdSe nanocrystallite quantum dots,” Science 278(5346), 2114–2117 (1997). [CrossRef] [PubMed]
15. L. I. Gurinovich, A. A. Lyutich, A. P. Stupak, M. V. Artem’ev, and S. V. Gaponenko, “Effect of an electric field on photoluminescence of cadmium selenide nanocrystals,” J. Appl. Spectrosc. 77(1), 120–125 (2010). [CrossRef]
16. L. I. Gurinovich, A. A. Lutich, A. P. Stupak, S. Y. Prislopsky, E. K. Rusakov, M. V. Artemyev, S. V. Gaponenko, and H. V. Demir, “Luminescence in quantum-confined cadmium selenide nanocrystals and nanorods in external electric fields,” Semiconductors 43(8), 1008–1016 (2009). [CrossRef]
17. H. Chen, Z. Luo, R. Zhu, Q. Hong, and S. T. Wu, “Tuning the correlated color temperature of white LED with a guest-host liquid crystal,” Opt. Express 23(10), 13060–13068 (2015). [CrossRef] [PubMed]
18. Z. Luo, H. Chen, Y. Liu, S. Xu, and S.-T. Wu, “Color-tunable light emitting diodes based on quantum dot suspension,” Appl. Opt. 54(10), 2845–2850 (2015). [CrossRef] [PubMed]
19. D. K. Yang and S. T. Wu, Fundamentals of Liquid Crystal Devices (John Wiley & Sons, 2014).
20. N. Tamaoki, “Cholesteric liquid crystals for color information technology,” Adv. Mater. 13(15), 1135–1147 (2001). [CrossRef]
21. Y. Huang, Y. Zhou, and S. T. Wu, “Broadband circular polarizer using stacked chiral polymer films,” Opt. Express 15(10), 6414–6419 (2007). [CrossRef] [PubMed]
22. J. Li, G. Baird, Y. H. Lin, H. Ren, and S. T. Wu, “Refractive index matching between liquid crystals and photopolymers,” J. Soc. Inf. Disp. 13(12), 1017–1026 (2005). [CrossRef]
23. Y. Chen, Z. Luo, F. Peng, and S. T. Wu, “Fringe-field switching with a negative dielectric anisotropy liquid crystal,” J. Disp. Technol. 9(2), 74–77 (2013). [CrossRef]
24. T. W. Murphy Jr., “Maximum spectral luminous efficacy of white light,” J. Appl. Phys. 111(10), 104909 (2012). [CrossRef]
25. D. A. Steigerwald, J. C. Bhat, D. Collins, R. M. Fletcher, M. O. Holcomb, M. J. Ludowise, P. S. Martin, and S. L. Rudaz, “Illumination with solid state lighting technology,” IEEE J. Sel. Top. Quantum Electron. 8(2), 310–320 (2002). [CrossRef]
26. Y. Ohno, “Color rendering and luminous efficacy of white LED spectra,” Proc. SPIE 5530, 88–98 (2004). [CrossRef]
27. B. W. D’Andrade and S. R. Forrest, “White organic light‐emitting devices for solid‐state lighting,” Adv. Mater. 16(18), 1585–1595 (2004). [CrossRef]
28. F. So, J. Kido, and P. Burrows, “Organic light-emitting devices for solid-state lighting,” MRS Bull. 33(7), 663–669 (2008). [CrossRef]
29. J. M. Caruge, J. E. Halpert, V. Wood, V. Bulovic, and M. G. Bawendi, “Colloidal quantum-dot light-emitting diodes with metal-oxide charge transport layers,” Nat. Photonics 2(4), 247–250 (2008). [CrossRef]
30. L. Qian, Y. Zheng, J. Xue, and P. H. Holloway, “Stable and efficient quantum-dot light-emitting diodes based on solution-processed multilayer structures,” Nat. Photonics 5(9), 543–548 (2011). [CrossRef]
31. Y. Shirasaki, G. J. Supran, M. G. Bawendi, and V. Bulović, “Emergence of colloidal quantum-dot light-emitting technologies,” Nat. Photonics 7(1), 13–23 (2012). [CrossRef]
32. H. Chen, F. Peng, Z. Luo, D. Xu, S. T. Wu, M. C. Li, S. L. Lee, and W. C. Tsai, “High performance liquid crystal displays with a low dielectric constant material,” Opt. Mater. Express 4(11), 2262–2273 (2014). [CrossRef]
33. H. Chen, M. Hu, F. Peng, J. Li, Z. An, and S. T. Wu, “Ultra-low viscosity liquid crystals,” Opt. Mater. Express 5(3), 655–660 (2015). [CrossRef]
34. Z. Luo, F. Peng, H. Chen, M. Hu, J. Li, Z. An, and S. T. Wu, “Fast-response liquid crystals for high image quality wearable displays,” Opt. Mater. Express 5(3), 603–610 (2015). [CrossRef]
35. F. Peng, Y. Huang, F. Gou, M. Hu, J. Li, Z. An, and S. T. Wu, “High performance liquid crystals for vehicle displays,” Opt. Mater. Express 6(3), 717–726 (2016). [CrossRef]
