Open Access
Add to BibSonomy
Abstract
One can display images and harvest energy by utilizing ambient light with a stack of an optical shutter array, a luminescent layer and a solar cell. In our experiment, a luminescent layer and a corresponding color filter were attached to a polycrystalline Si solar cell with 13% power conversion efficiency. For each configuration using BBOT, Coumarin 6 and Lumogen F Red 305, the power conversion efficiency was measured to be 6.7%, 8.0% and 8.9%, respectively. The luminance of these configurations was proportional to the illuminance in all cases. Its color gamut was comparable to the National Television System Committee standard.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
A luminous-reflective display (LRD) [1] consists of an array of electro-optical (EO) shutters, a luminescent layer and a reflector stacked in this order. Ambient light passing through the EO shutter is converted to photoluminescent (PL) photons. Those emitted upward pass through the EO shutter and reach an observer. So do the PL photons emitted downward after being reflected by the reflector. For displaying color images, three luminescent materials emitting three primary colors are used. Color filters inserted between the luminescent layers and the reflector absorb the light with unwanted wavelengths. Unlike emissive displays such as an organic light-emitting diode (OLED) display and a transmissive liquid crystal display (LCD) [2], the contrast ratio of the images displayed by an LRD will remain the same irrespective of the illuminance [3]. Like a purely reflective LCD [4], its power consumption will be small because it utilizes ambient light. An LRD will boost the luminance of a purely reflective display by utilizing the light with shorter wavelengths which are otherwise absorbed by the color filters. The color gamut of an LRD is expected to be more stable because the emission spectra of the luminescent materials are fixed irrespective of the excitation light. Although the power consumption of an LRD might be small, it will require some to drive its electronics.
Power generation and storage capability would add flexibility to power management in a display system. By projecting blue images on a luminescent solar concentrator (LSC) [5], one can harvest energy from its light source as well as from ambient light [6]. However, a certain distance is required for projecting images on the LSC screen. A flat-panel configuration is preferred for mobile applications where power management is of critical importance. Such an energy-harvesting display could be installed on the walls of zero-emission buildings.
In this paper, we describe a flat-panel configuration for both displaying images and harvesting energy from ambient light. In experiment, off-the-shelf materials and components are used to measure power conversion efficiency and display characteristics.
2. Energy-harvesting LRD
2.1 Configuration
The configuration proposed here is shown in Fig. 1. Each sub-pixel, consisting of an EO shutter, a luminescent layer, and a color filter, is placed above a light-scattering layer, an infrared-pass filter and a solar cell. Note that there should not be an air gap between each component to avoid light losses by Fresnel reflection. This is equivalent to the structure of an LRD [1] except that the reflector in its each sub-pixel is replaced by an infrared-pass filter and a solar cell.
Fig. 1. Exploded perspective view of the three sub-pixels placed above a solar cell. White ambient light (denoted as W) enters the luminescent layers. Each layer emits PL photons for each primary color. They are denoted as $\textrm{P}{\textrm{L}_\textrm{B}}$, $\textrm{P}{\textrm{L}_\textrm{G}}$ and $\textrm{P}{\textrm{L}_\textrm{R}}$ for blue, green and red, respectively.
The EO shutter controls how much incident light excites the luminescent layer. It should consume no power when not displaying images. For example, a normally-white LCD panel is a good candidate for this component. We will address this issue in Section 4.
Each luminescent layer absorbs the incident light in its characteristic wavelength range and generates PL photons in the range for each primary color. The ambient light with unwanted wavelengths is absorbed by each color filter. These color filters should be as transparent as possible in the near infrared range.
The light-scattering layer can be inserted to out-couple the PL photons which are otherwise waveguided by internal reflection. For a reflective LCD, bumpy pixel-electrodes prevent specular reflection and more sophisticated surface structures are studied to enhance image contrast [7]. These technologies can be adopted in the light-scattering layer in Fig. 1 to improve image quality.
The infrared-pass filter reflects the visible light and transmits the infrared light. The solar cell harvests the downward light. A polycrystalline Si solar cell is suited for this application because of its bandgap energy of 1.1 eV and technological maturity. Note that the use of luminescent materials is well studied in the field of solar cells: down conversion improves spectral matching between sunlight and the responsivity of a solar cell in some cases [8].
2.2 Operation principle
Let us start from the operation of displaying images. Suppose that white ambient light (denoted as “W” in Fig. 1) passes through the EO shutter for a blue sub-pixel. The blue luminescent layer beneath it is excited and blue PL photons are generated. Those emitted upward (denoted as “PLB”) goes through the EO shutter and reach an observer. The rest of the PL photons are emitted downward. After passing through the color filter, they are reflected by the infrared-pass filter. In addition, the incident light not absorbed by the luminescent layer or the color filter is reflected. These two upward light fluxes reach the observer.
Among the three sub-pixels, the red one is expected to have the largest luminance and the blue one the least. This is because the red sub-pixel utilizes blue and green light in the white incident light while the blue sub-pixel can convert only the ultraviolet light to blue PL photons. This issue has been pointed out by the early experiment [9] and modeled by the spectral study [1]. We will address this problem and discuss a possible solution in Section 4.
Obviously, there is a trade-off between luminance and power generation. If the infrared-pass filter in Fig. 1 is removed, the solar cell harvests the downward light in the visible range as well. Hence, more power is generated at the expense of luminance. The light-scattering layer enhances luminance by outcoupling the light trapped in this multi-layer structure. However, it scatters the downward flux and degrades the power generation capability.
2.3 Design considerations
First, luminance and color gamut are important performance indices for displaying color images. These are determined mostly by the characteristics of the luminescent layers and the color filters as described in Ref. 1. Briefly, the luminescent materials should absorb ambient light efficiently to generate PL photons as many as possible for increasing luminance. The thickness of the luminescent layer and the concentration of the material are the design parameters relevant for this objective. The quantum yields of the luminescent materials should be close to unity. For extending a color gamut, the luminescent materials need to emit in the right and narrow wavelength ranges. The color filters should transmit most of the PL photons and absorb ambient light with unwanted wavelengths. To discuss these design issues, a simple model was developed for calculating the spectral fluxes involved. It reproduced the experiment reasonably well [1]. Based on these considerations, in addition to practical problems such as availability and cost, we prepare samples in the experiment to be described below.
Second, the amount of energy harvested is determined by the downward spectral flux and the external quantum efficiency of a solar cell. When the infrared-pass filter and the light-scattering layer are inserted, their characteristics also affect the flux reaching the solar cell as well as the upward flux reaching an observer. The transmittance of an infrared filter changes abruptly at a certain wavelength. It should be noted that this critical wavelength shifts to shorter wavelengths for oblique incidence. This effect is observed in the experiment described below.
For a quantitative discussion on the performances of the configuration in Fig. 1, a new model is required to relate input parameters to performance indices. This is a good research topic and multiple steps might be needed to complete this study in future.
3. Experiment
Because the EO shutter technology in Fig. 1 varies, we focus on the characteristics of the sub-pixel structure without it in this section. First, we describe sample preparation. Second, its power conversion efficiency is measured under the standard condition of Air Mass 1.5 Global (AM1.5G) [11]. Although down conversion by luminescent materials improves it in some cases [8], it is of interest to know how the color filters and the infrared-pass filter degrade it. Third, we measure display characteristics in addition to photocurrent under the illumination by two common light sources.
3.1 Preparation
The drawing in Fig. 2 shows the cross section of the structure prepared in this experiment. The top component is a luminescent layer formed on an acrylic plate. A color filter, a light-scattering film, and an infrared-pass filter are stacked on a solar cell in this order. To eliminate an air gap between each component, an optical adhesive film is used. Its refractive index is 1.48.
Fig. 2. Cross section of the stacked-layer configurations prepared for this experiment and photographs taken under the illumination by white LEDs.
Organic dyes, BBOT and Coumarin 6 purchased from Sigma-Aldridge and Lumogen F Red from BASF, were used for the luminescent materials here. Each material was mixed with ultraviolet-curable resin (NOA81, Norland Products). Each solution was spin-coated on a 1 mm-thick, 50 mm x 50 mm acrylic plate and cured. The photographs in Fig. 2 were taken under the illumination by a desktop lamp. The BBOT sample appears almost transparent because the white LED in the lamp cannot excite this material. This sample is included to assess the problem for the blue sub-pixel quantitatively. Emission spectra were acquired by exciting these samples with a laser emitting at 405 nm. The spectra in Fig. 3(a) are normalized such that the area under each curve is unity. The transmittance for each luminescent layer measured at 405 nm was 84% (BBOT), 73% (Coumarin 6), and 90% (Lumogen Red F 305).
Fig. 3. Spectral characteristic of each component used in this experiment. (a) Emission spectra measured with the luminescent layers. (b) Transmittance of each color filter (CF) and the infrared (IR)-pass filter.
The color filters (CFs) were general purpose stationary. Three were selected based on their measured spectral transmittances. As shown in Fig. 3(b), their peak wavelengths roughly match those of the emission spectra in Fig. 3(a). Note that the transmittance is larger than 90% in the near infrared range. The transmittance of the infrared (IR)-pass filter used in this experiment (Edmund Optics) was measured. As shown in Fig. 3(b), all these filters transmit the infrared light.
Other components used in the experiment are as follows. A polycrystalline Si solar cell was made by Kyocera Corp. It had a $18.6\,\textrm{mm} \times 19.9\,\textrm{mm}$ sensitive area. Its power conversion efficiency was measured to be 13% as shown in Fig. 4 (a). The light-scattering film used here (Tsujiden Co, Ltd., Model D114) was originally developed for applications in backlight units. An optical adhesive film (Mecanusa Inc., refractive index 1.48) was used to eliminate the air gap between each component.
Fig. 4. Current-voltage characteristics of the various stacked configurations mimicking the three sub-pixels. Each layer is abbreviated as follows; PV (solar cell), lum. (luminescent layer), CF (color filter), diff. (light-scattering film), IRpf (infrared-pass filter). The rhomb markers in (a) represent the data for the solar cell only. The colors and the shapes of the markers represent the luminescent materials; red circle (Lumogen F Red 305), green triangle (Coumarin 6) and blue square (BBOT). (a) The effect of inserting the infrared-pass filter. (b) The effect of inserting the light-scattering film.
3.2 Power conversion efficiency
For each of the stacked configurations in Fig. 1, its current-voltage characteristic was measured under AM1.5G. A commercial solar simulator (Optical Associates, Inc., Model TSS-156) illuminated each sample uniformly. The results are shown in Fig. 4. Each component is denoted as follows; lum. (luminescent layer), diff. (light-scattering film), IRpf (infrared-pass filter), PV (solar cell). For example, the abbreviation “lum./CF/IRpf/PV” in Fig. 4(a) corresponds to the structure without the light-scattering film. The color and the shape of the markers in Fig. 4 represent the three luminescent materials as follows: red circle (Lumogen F Red 305), green triangle (Coumarin 6), and blue square (BBOT). The curves marked as “lum./CF/IRpf/PV” are reproduced in Fig. 4(b) to serve as a reference.
Power conversion efficiencies calculated from these curves are summarized in Table 1. The simple structure “lum./CF/ PV” has the largest efficiencies because the downward visible light also reaches the solar cell. Among the three sub-pixel structures, the red sample shows the largest efficiency (8.9%) because its luminescent material converts the blue and green light in the simulated sunlight. The blue sample has the lowest value (6.7%) partly because its luminescent layer does not absorb much light and partly because the transmittance of its color filter is low. Despite the light absorption by the color filters, these values are more than one half of the efficiency of the solar cell used here (13%).
Table 1. Power Conversion Efficiencies of Various Stacked Configurations in %
The infrared-pass filter reflects the visible light. Hence, the photocurrents of the configuration “lum./CF/IRpf/ PV” become smaller as shown in Fig. 4(a). Because only the infrared light reaches the solar cell in this configuration, the blue and green samples behave similarly.
The light-scattering film decreases the photocurrent further as shown in Fig. 4(b). Some of the PL photons and the incident light passing through the color filters are directed upward after scattering. This scattering loss is less apparent for the red sample as shown in Fig. 4(b). This can be explained by considering the property of the infrared-pass filer. As shown in Fig. 3(b), its transmittance changes abruptly around 700 nm for normal incidence. This critical wavelength depends on the incident angle. When the scattered light enters the filter obliquely, it is more likely to pass the filter and reach the solar cell. This gain partially compensates the scattering loss for the red sub-pixel configuration.
3.3 Luminance, color gamut, photocurrent
For characterizing a reflective display, it is customary to illuminate it with ambient light at $\textrm{3}{\textrm{0}^ \circ }$ from its normal and to vary the direction of observation [4,12]. For a quick comparison of the various stacked configurations, the observation direction was fixed in this measurement. As illustrated in Fig. 5(a), an optical fiber bundle relayed the light from a light source to the device under test. It was covered by a piece of black cloth with a 25 mm-square aperture to block stray light if any. Neutral-density filters were inserted to control its intensity. This setup ensured uniform and stable illumination. For every setting, the illuminance at the surface of the device was monitored by an illuminance meter. A luminance and tristimulus colorimeter (Konica Minolta, Model LS-160) was set at the normal direction to record luminance and color coordinates. The region of interest for this luminance measurement was 2mm-diameter circle at the center of the device. A current amplifier (Keithley, Model 428) connected to the solar cell (sensitive area: $18.6\,\textrm{mm} \times 19.9\,\textrm{mm}$) measured the photocurrent.
Fig. 5. Measurement of luminance, color gamut, and current-voltage characteristics: (a) schematic drawing of the experimental setup, and (b) the spectra of the two light sources used in this measurement together with the AM1.5G spectrum for comparison.
A white LED and a halogen lamp were selected for the light source. Their measured spectra are compared to the standard AM1.5G [11] in Fig. 5(b). These are normalized spectra such that the area under each curve is unity. Note that there are a lot of infrared photons in the AM1.5G spectrum for the solar cell to harvest and almost nothing in the white LED spectrum.
The luminance is plotted as a function of the illuminance in Fig. 6. The lines are linear fits to the data. The notations for the configurations and the assignment of the markers are the same with Fig. 4. In all cases, the luminance is proportional to the illuminance. This results in the fact that the ambient contrast ratio of an LRD remains constant even at extremely high illuminance [3].
Fig. 6. Luminance measured under the illumination by (a) the white LED and (b) the halogen lamp. The squares, triangles, and the circles represent BBOT, Coumarin 6, and Lumogen F Red 305, respectively. The empty markers represent the luminance of the structure “lum./CF/diff./IRpf/PV” and the solid markers are for the structure “lum./CF /PV.” The lines are linear fits to the measured data.
The effect of adding the light scattering film and the infrared-pass filter on the luminance is apparent in Fig. 6. For example, the luminance of the green configuration “lum./CF/PV” is $23\,\textrm{cd/}{\textrm{m}^2}$ at the illuminance of 1 klx in Fig. 6(a). It increased to $109\,\textrm{cd/}{\textrm{m}^2}$ by the addition of the two layers. For the case of the halogen lamp in Fig. 6(b), it increased from $16\,\textrm{cd/}{\textrm{m}^2}$ to $80\,\textrm{cd/}{\textrm{m}^2}$. The light-scattering film extracts the waveguided PL photons by changing their propagation directions. The infrared-pass filter reflects the downward flux of visible light. Nevertheless, these luminance values will decrease by the transmittance of an EO shutter.
The luminance for the green configuration is high because the luminous efficiency function peaks at 555 nm. Under the halogen lamp, however, the red configuration “lum./CF/PV” (red solid circles) has slightly larger luminance values than the green one (green solid triangles) as shown in Fig. 6(b). This could be attributed to the spectral matchings between the two light sources and the absorption coefficients of Lumogen F Red 305 [10] and Coumarin 6 [13]. As for the blue configurations, their luminance values are low partly because the luminous efficiency function is small in this wavelength range and partly because neither of the two light sources can excite the material much. A model is desired for further discussions.
The color coordinates recorded by the tristimulus colorimeter are compared to those of the National Television System Committee (NTSC) standard in Fig. 7. The color gamut of the configuration “lum./CF/diff./IRpf/PV” under the white LED illumination is as wide as the NTSC standard as shown in Fig. 7(a). Under the halogen lamp, it becomes substantially smaller. In Fig. 7(b), the coordinates for the blue and green colors are shifted toward those for the red color. This is attributed to the difference in the spectra of these light sources: the halogen lamp emits more in red.
Fig. 7. Color coordinates measured under the illumination by (a) the white LED and (b) the halogen lamp. The markers for the three stacked configurations are as indicated.
As shown in Fig. 7, the infrared-pass filter extends the color gamut of the stacked structure slightly. The light-scattering film extends it further. These observations can be attributed to the longer distance that the light needs to propagate before exiting the top luminescent layer. The infrared-pass filter reflects the downward light. The light-scattering film alters the propagation direction of the downward light. In both cases, the upward flux goes through the color filter and the luminescent layer. In this process, its spectrum becomes narrower due to the absorption by the two layers. A similar mechanism is responsible for the redshift of the PL photons exiting a uniform luminescent layer obliquely [14]. The PL photons with shorter wavelengths are more likely to be absorbed by a luminescent layer itself if they are propagating obliquely.
Therefore, the benefit of adding the infrared-pass filter and the light-scattering layer is twofold: increased luminance and extended color gamut. However, these two layers decrease the amount of light reaching the solar cell, resulting in smaller photocurrents as shown in Fig. 8. The same notations are used for the configurations and the materials.
Fig. 8. Photocurrent measured under the illumination by (a) the white LED and (b) the halogen lamp. The squares, triangles, and the circles represent BBOT, Coumarin 6, and Lumogen F Red 305, respectively. The empty markers represent the photocurrent of the structure “lum./CF/diff./IRpf/PV” and the solid markers are for the structure “lum./CF/PV.” The lines are linear fits to the measured data.
Note that the halogen lamp generates much larger photocurrents than the white LED. For example, the photocurrent of the configuration “lum./CF/diff./IRpf/PV at the illuminance of 1 klx is of the order of 1 mA in Fig. 8 (b) while it is more than one order of magnitude smaller in Fig. 8(a). This is expected from the spectra in Fig. 5(b). In case of AM1.5G in Fig. 4(b), the photocurrent is about 33 mA. The illuminance of AM1.5G is about 100 klx. Thus, at a fixed illuminance value, the halogen lamp is more efficient for energy-harvesting than AM1.5G. This is not surprising because the halogen lamp has more photons available for exciting the green and red luminescent materials as shown in Fig. 5(b). A spectral study is desired for a quantitative discussion.
4. Discussions
The concept of utilizing ambient light with luminescent materials for displaying images was conceived at least two decades ago [16]. Considering the pressing demand for solving the trade-off between readability and power consumption, one might wonder why this concept did not take off long time ago. In this paper, we have proposed to add energy-harvesting capability in a display system. This would ease the trade-off dramatically. However, some practical obstacles are foreseen. In this section, we address these problems and propose new directions for future studies.
4.1 White balance and readability under dark
It has been recognized that not many photons in ambient light are available for exciting the blue-emitting luminescent material [9]. One passive solution is to enlarge the blue sub-pixel area and to remove the blue-emitting luminescent material. In this case, the blue sub-pixel is equivalent to one in a reflective display. One can also control the transmittance of the EO shutter array for balancing white. But light utilization becomes less efficient.
In addition to white balance, one should be able to read a display even in a completely dark place. In this case, an external light source is required. It has been proposed to place a dichroic filter and a blue-emitting backlight unit beneath a pixelated luminescent layer [15]. This design cannot be adopted here because a solar cell is present.
A front light would solve these problems. The OLED technology is suited for this application. One can design a thin light source with apertures, which are required to pass ambient light. The remaining mesh-shaped region emits blue light toward the pixelated luminescent layer. Blocking the ambient light will be kept minimum by placing the mesh pattern on the sub-pixel borders.
4.2 Parallax error
A transmissive LCD panel is a good candidate for the EO shutter in Fig. 1. Its essential optical elements are a thin liquid crystal layer sandwiched between two transparent substrates and the polarizers attached to their external surfaces. The separation between the liquid crystal layer and the luminescent layer would result in parallax error.
An in-cell polarizer can be a solution to this problem. Namely, a luminescent layer is formed on a bottom substrate first. Then a polarizer is formed on it and an LCD panel is assembled with a top substrate. Two decades ago, a reflective LCD was extensively studied for mobile phones in mind. Because of the pixel resolution and the thickness of the glass substrates available then, an in-cell polarizer was required to prevent parallax error [12]. It is also reported that the contrast ratio of a transmissive LCD can be improved by an in-cell polarizer [17]. Hence, there is a continuing interest for developing this technology.
Alternatively, one can make the separation between the liquid crystal layer and the luminescent layer as thin as possible. Recent reports on thin glass substrates [18] and a thin transmissive LCD panel [19] are encouraging.
4.3 Luminescent materials
The three organic dyes were selected by considering their emission and absorption spectra, quantum yields, processability, and some practical issues as described in Sub-section 2.3. Obviously, it is desired to have a blue-emitting material that can be excited at longer wavelengths. Photostability will certainly be an issue in a practical application. The green samples prepared in our experiment degraded in a few months. It is worthwhile to explore other materials such as quantum dots and inorganic phosphors used in backlight units.
5. Conclusions
An LRD will have higher luminance and wider color gamut than a purely reflective display. Its sub-pixel consists of an EO shutter, a luminescent layer, a color filter and a reflector. Using three types of luminescent materials, color images can be displayed. By replacing the reflector with a solar cell, one can harvest energy from ambient light. A light-scattering layer and an infrared-pass filter can be inserted between the color filter and the solar cell. They will enhance luminance at the expense of photocurrent by redirecting the downward light.
In experiment, we formed a luminescent layer on an acrylic plate and placed it on a color filter and a polycrystalline Si solar cell with 13% power conversion efficiency. Color filters were general purpose stationary and three were selected for three primary colors based on their spectral transmittances. For the luminescent materials, organic dyes BBOT, Coumarin 6 and Lumogen F Red 305 were used. Power conversion efficiency measured for each configuration under AM1.5G was 6.7%, 8.0% and 8.9%, respectively. When a light-scattering film and an infrared-pass filter were inserted between each color filter and the solar cell, the efficiency decreased to 3.8%, 3.8%, and 5.0%, respectively. A white LED and a halogen lamp were used to evaluate luminance, color gamut and photocurrent for the three stacked configurations. For example, the two layers increased the luminance of the green configuration from $23\,\textrm{cd/}{\textrm{m}^2}$ to $109\,\textrm{cd/}{\textrm{m}^2}$ under the illuminance of 1 klx by the white LED. The color gamut was roughly equivalent to the NTSC standard and the two layers extended it slightly.
To demonstrate an energy-harvesting LRD, there are some practical obstacles. First, an external light source will be required to ensure white balance and readability under dark. Second, the separation between the EO shutter and the luminescent layer could result in parallax error in high-resolution displays. Either an in-cell polarizer or various technologies for components and fabrication processes would solve this problem. Third, a stable, efficient luminescent material is highly desired.
In future, an energy-harvesting LRD will provide flexibility to power management in smartphones, laptop computers and digital signages. It might be installed on the walls of zero-emission buildings.
Acknowledgments
We would like to thank Prof. Mitsuhiro Shigeta for helpful discussions and Prof. Yu Kawano for the measurements of power conversion efficiencies and spectral transmittances.
Disclosures
The authors declare no conflicts of interest.
Data Availability
We are happy to provide digital data upon request.
References
1. I. Fujieda, Y. Tsutsumi, and S. Matsuda, “Spectral study on utilizing ambient light with luminescent materials for display applications,” Opt. Express 29(5), 6691–6702 (2021). [CrossRef]
2. H. Chen, J. Lee, B. Lin, S. Chen, and S. Wu, “Liquid crystal display and organic light-emitting diode display: present status and future perspectives,” Light Sci Appl 7(3), 17168 (2018). [CrossRef]
3. Y. Yamada, Y. Tsutsumi, and I. Fujieda, “Ambient contrast ratio of a liquid crystal cell stacked on a luminescent layer,” Proc. of 27th International Display Workshops, 76–79 (2020).
4. H. Hakoi, M. Ni, J. Hashimoto, T. Sato, S. Shimada, K. Minoura, A. Itoh, K. Tanaka, H. Matsukizono, and M. Otsubo, “High-performance and low-power full color reflective LCD for new applications,” SID Symp. Dig. Tech. Pap.50, 279–282 (2019).
5. I. Papakonstantinou, M. Portnoi, and M. G. Debije, “The hidden potential of Luminescent Solar Concentrators,” Adv. Energy Mater. 11(3), 2002883 (2021). [CrossRef]
6. I. Fujieda, S. Itaya, M. Ohta, Y. Hirai, and T. Kohmoto, “Energy-harvesting laser phosphor display and its design considerations,” J. Photon. Energy 7(2), 028001 (2017). [CrossRef]
7. Y. Huang, H. D. Shieh, and S. Wu, “Applications of multidirectional asymmetrical microlens-array light-control films on reflective liquid-crystal displays for image quality enhancement,” Appl. Opt. 43(18), 3656–3663 (2004). [CrossRef]
8. E. Klampaftis, D. Ross, K. R. McIntosh, and B. S. Richards, “Enhancing the performance of solar cells via luminescent down-shifting of the incident spectrum: A review,” Solar Energy Materials and Solar Cells 93(8), 1182–1194 (2009). [CrossRef]
9. G. A. Gibson, X. Sheng, P. Manna, and D. Henze, “Luminescent enhancement of reflective displays,” 2010 23rd Annual Meeting of the IEEE Photonics Society, Denver, CO, 2010, pp. 69–70, doi:10.1109/PHOTONICS.2010.5698761.
10. L. R. Wilson and B. S. Richards, “Measurement method for photoluminescent quantum yields of fluorescent organic dyes in polymethyl methacrylate for luminescent solar concentrators,” Appl. Opt. 48(2), 212–220 (2009). [CrossRef]
11. G. P. Smestad, F. C. Krebs, C. M. Lampert, C. G. Granqvist, K.L. Chopra, X. Mathew, and H. Takakura, “Reporting solar cell efficiencies in Solar Energy Materials and Solar Cells,” Solar Energy Materials and Solar Cells 92(4), 371–373 (2008). [CrossRef]
12. Y. Ukai, T. Ohyama, L. Fennell, Y. Kato, M. Paukshto, P. Smith, O. Yamashita, and S. Nakanishi, “Current status and future prospect of in-cell-polarizer technology,” J. Soc. Inf. Disp. 13(1), 17–24 (2005). [CrossRef]
13. E. F. Schubert, Light Emitting Diode, Second edition, (Cambridge University Press, 2006), p. 350.
14. I. Fujieda and M. Ohta, “Angle-resolved photoluminescence spectrum of a uniform phosphor layer,” AIP Adv. 7(10), 105223 (2017). [CrossRef]
15. G. Gibson, X. Sheng, D. Henze, S. Lam, P. Beck, Y. Jeon, Z. Zhou, B. Benson, Q. Liu, G. Combs, Tim Koch, and Kent Biggs, “Fast full-color reflective display via photoluminescent enhancement,” J. Soc. Inf. Disp. 20(10), 552–558 (2012). [CrossRef]
16. Osamu Okumura, patent applicationJP-A-2001-264756 (in Japanese).
17. Haiwei Chen, Guanjun Tan, Ming-Chun Li, Seok-Lyul Lee, and Shin-Tson Wu, “Depolarization effect in liquid crystal displays,” Opt. Express 25(10), 11315–11328 (2017). [CrossRef]
18. D. Chowdhury, K. Woo, B. Kong, J. Lee, H. Yoon, and M. Mlejnek, “Flexible glass substrate for OLED lightning application and efficient internal light extraction for OLED lightning devices,” SID Symp. Dig. Tech. Pap.51, 93–95 (2020).
19. S. Oka, Y. Hyodo, L. Jin, G. Asozu, H. Kaneda, K. Mochizuki, M. Mochizuki, Y. Aoki, and N. Asano, “Ultra-narrow border display with a cover glass using liquid crystal displays with a polyimide substrate,” J. Soc. Inf. Disp. 28(4), 360–367 (2020). [CrossRef]
