Open Access
Add to BibSonomy
Abstract
A luminous reflective display can be constructed by placing an electro-optic shutter on the stack of a luminescent layer, a color filter, and a reflector in this order. The luminescent materials convert a part of the incident light to photoluminescence photons. The reflector redirects the downward photon flux toward an observer. The color filters prevent the photons with unwanted wavelengths from being reflected. The upward spectral flux from this multi-layer structure is formulated. Experiments with off-the-shelf components revealed more than three-fold increase in spectral flux and up to 55% color gamut extension, compared with a control device without luminescent materials.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
To maintain readability in bright environments, one can increase the luminance of an emissive display such as a transmissive liquid-crystal display (LCD) and an organic light-emitting diode display (OLED). Inevitably, its power consumption increases. In contrast, a reflective display utilizes ambient light Because the intensity of the reflected light is proportional to that of the incident light, it is readable under the sun. Ambient contrast ratio (ACR) is often used to quantify this characteristic [1]. For example, a value of 23 was reported for a reflective LCD in 2019 [2]. Like reflection, the intensity of photoluminescence (PL) is proportional to that of the incident light. We coated an acrylic plate with a thin layer of an organic dye (Coumarin 6), stacked a twisted-nematic LC cell on it, and put it on a black sheet of paper. Its ACR remained at around 10 even at 222 klx under illumination by a white LED [3]. There was an air gap between the LC-cell and the luminescent layer in this experiment. Roughly 75% of the PL photons was trapped inside and one half of the leaking PL photons were directed to the opposite side. Out-coupling and redirecting the PL photons toward an observer improved light utilization [4]. Therefore, an LC panel stacked on a luminescent layer and a reflector would make a bright display with a large ACR.
For a color display, we need another performance index, color gamut. A reflective display has a narrow color gamut in general. Filtering unwanted wavelengths extends it at the expense of light utilization [5]. A practical design seems to prioritize brightness over vivid colors. For example, the color gamut of the reflective LCD reported in 2019 covers only 19% of the National Television System Committee (NTSC) standard [2]. Luminescent materials can alleviate this trade-off. They convert the photons with shorter wavelengths, which are otherwise absorbed by color filters, to the photons with the right wavelengths for each color. In addition, a PL spectrum is usually narrow and is fixed irrespective of the excitation wavelength. Hence, a wide and stable color gamut under different illumination conditions is expected. This multi-layer structure was disclosed in a patent application for a reflective LCD in 2001 [6]. A decade later, the same concept was applied for an electrophoretic display with mass-production in mind: luminescent dyes mixed with pigments were printed by ink-jetting [7]. The peak intensity of the spectrum of the light from their red sub-pixel was a little more than two times larger than that from a purely reflective sub-pixel [8]. For fulfilling the potential of this concept, it is desired to understand the interplay of each component and the incident light quantitatively. A simple model will relate the design parameters to the characteristics of such a display and identify some goals for material research.
In this paper, we formulate the spectral flux of the photons emerging from the stack of a luminescent layer, a color filter and a reflector in Section 2. The experiment with off-the-shelf components is described in Section 3. In Section 4, we check the validity of the model and discuss color gamut.
2. Theory
We call a display based on this concept as a luminous reflective display (LRD) because both the PL photons and the reflected light are utilized. The spectral flux of the photons emerging from its sub-pixel is modeled in this section.
2.1 Model
As illustrated in Fig. 1(a), a luminescent layer, a color filter and a reflector are stacked in this order. An electro-optic (EO) shutter is attached on this multi-layer structure. There is no gap between them to prevent Fresnel reflection. When the EO shutter is set to a light-absorbing state, there is no light propagating upward. When it transmits ambient light, PL photons are generated in the luminescent layer as illustrated in Fig. 1(a). The reflector redirects the downward flux and the ambient light passing through the luminescent layer. A diffuse reflector is preferred over a mirror because the latter would trap the PL photons via waveguiding. The configuration without the luminescent layer in Fig. 1(b) represents the sub-pixel of a reflective display. In case of a reflective LCD, the color filter and the reflector are incorporated in the EO shutter: an LC layer is usually sandwiched by a substrate with color filters and another substrate with reflectors. A polarizer and a quarter wave plate are attached on the top substrate [9]. To avoid specular reflection, some microstructures are formed on the reflector. For this purpose, either bumpy reflectors can be distributed randomly as illustrated in Fig. 1 or a film with asymmetrical microlens array can be employed [10]. For both configurations in Fig. 1, the color filter absorbs the photons with unwanted wavelengths.
Fig. 1. Cross-section of a sub-pixel in (a) a luminous reflective display (LRD) and (b) a reflective display. For a reflective LCD, we can interpret that the color filter and the reflector are incorporated in the EO shutter.
For formulating the upward spectral flux from these configurations, we assume that the EO shutter is transparent. Our modeling starts by denoting the spectral flux of the ambient light exiting the EO shutter as ${S_{amb}}(\lambda )$ in the unit of $\textrm{photons/nm} \cdot \textrm{s}$.
First, let us consider the control structure in Fig. 1(b). Suppose that the reflectance of the diffuse reflector is given by a single parameter R. Then, denoting the transmittance of the color filter at wavelength $\lambda$ as ${T_{CF}}(\lambda )$, the upward spectral flux is expressed as follows.
As for the structure in Fig. 1(a), denoting the transmittance of the luminescent layer as ${T_{LL}}(\lambda )$, the flux of the incident ambient light absorbed by the luminescent layer is expressed in the unit of $\textrm{photons/s}$ as,
This flux is converted to PL photons with the probability known as quantum yield $\eta$. Suppose that the emission spectrum of the material ${S_{em}}(\lambda )$ is scaled such that $\int {{S_{em}}(\lambda )} \,d\lambda = 1$. Then, the spectral flux of the PL photons exiting the luminescent layer is expressed in the unit of $\textrm{photons/nm} \cdot \textrm{s}$ as,
Let us assume that the photoluminescence is isotropic. Although the forward spectral flux of the PL photons emerging from a uniform luminescent layer is slightly redshifted due to self-absorption [11], let us neglect this difference for simplicity. Then, one half of ${S_{PL}}(\lambda )$ is directed upward and reach an observer. The other half is directed downward and a part of it reaches the observer. In this process, it passes through the color filter twice and the luminescent layer once. Thus, this component is expressed as $\frac{1}{2}R\,{\{{{T_{CF}}(\lambda )} \}^2}{T_{LL}}(\lambda ){S_{PL}}(\lambda )$. Note that the absorptance term ($1 - {T_{LL}}(\lambda )$) overlaps with the emission spectrum ${S_{PL}}(\lambda )$ for many luminescent materials and this fact results in self-absorption. Hence, it is incorporated in this model at least partially via the term ${T_{LL}}(\lambda )$.
In addition to the PL photons, a part of the incident light can reach an observer after being reflected. This component passes the luminescent layer and the color filter twice. Thus, the total upward spectral flux is expressed as,
So far, we have only considered the PL photons generated by the first passage of the incident light. A part of it enters the luminescent layer after being reflected. The incident flux for this second passage is given by $R\,{\{{{T_{CF}}(\lambda )} \}^2}{T_{LL}}(\lambda ){S_{amb}}(\lambda )$. Therefore, Eq. (2) needs to be modified as follows.
The first term in Eq. (4) represents the contribution from the PL photons. It is proportional to ${F_{abs}}$ via Eq. (5). Decreasing ${T_{LL}}(\lambda )$ in a short wavelength range is effective for increasing ${F_{abs}}$ as apparent from Eq. (2). In fact, ${T_{LL}}$ is a design parameter because we can control it to some extent by changing either the thickness of the luminescent layer with a fixed concentration or the other way around. Because ${T_{LL}}(\lambda )$ can be made close to zero at short wavelengths, red and green sub-pixels have wider wavelength ranges for utilizing ambient light than a blue sub-pixel.
The second term in Eq. (4) represents the contribution from the reflected ambient light. It can be rewritten as ${\{{{T_{LL}}(\lambda )} \}^2}{S_{RD}}(\lambda )$ via Eq. (1). If ${T_{LL}}(\lambda )= 1$ in the wavelength region to be displayed by each sub-pixel, this term is equal to ${S_{RD}}(\lambda )$. This condition is satisfied if its absorption spectrum is well separated from its emission spectrum. Hence, a luminescent material with a large Stokes shift is desired. Interestingly, this demand is shared by luminescent solar concentrators [12]. Search for an ideal material with zero overlap between the two spectra such as nanocrystals [13] continues in this field.
The first term in Eq. (4) generates the following advantages of an LRD over an equivalent reflective display. First, its luminance is larger. Second, its color gamut is wider and more stable.
2.2 Numerical example
A green and a red sub-pixel are considered here because more photons are available in the spectrum of ambient light for them to utilize. We use properties of existing materials and components in this example. First, Coumarin 6 and Lumogen F Red 305 are selected as the luminescent materials emitting in green and red, respectively. Their characteristics are well documented [14,15]. Their emission spectra are scaled such that $\int {{S_{em}}(\lambda )} \,d\lambda = 1$. The results are marked as ${S_{em}}$ in Fig. 2. The transmittance of the luminescent layer is calculated from the absorption coefficients [14,15]. The curves marked as ${T_{LL}}$ in Fig. 2 represent examples of strongly absorbing cases. Second, color filters are chosen. The curves marked as ${T_{CF}}$ are reported to be the characteristics of the color filters used in a commercial transmissive LCD [16]. Third, we assume $\eta \, = R = 1$ and two CIE standard illuminants A and D65 for the ambient light spectrum.
Fig. 2. Input parameters for calculating the spectral flux from the multi-layered structures mimicking (a) a green sub-pixel and (b) a red sub-pixel. The luminescent dyes assumed here are (a) coumarin 6 and (b) Lumogen F 305 Red. For each sub-pixel, the red broken curve represents the emission spectrum and the dotted curve represents the transmittance of the color filter. The blue solid curve is an example for the transmittance of the luminescent layer to start this calculation.
Spectral fluxes calculated for the configurations mimicking green and red sub-pixel are shown in Fig. 3. The legends indicate the multi-layer structures. The abbreviation CF stands for “color filter” and the letter W for “diffuse reflector” which appears white under room light. For example, “C6/CF(G)/W refers to the configuration where a coumarin 6 layer, a color filter transmitting in green and a diffuse reflector are stacked in this order. The solid curves are the spectral fluxes for the structures with the luminescent materials. The broken curves represent the cases without them. The black dotted curve in each graph represents the CIE standard illuminant. In Fig. 3, the spectral fluxes from the structures with luminescent materials are always larger than those from the corresponding control structures. Their peak heights exceed those of the incident light. In contrast, this is not the case for the control structure because reflectance never exceed unity. This enhancement effect by the luminescent materials is more evident for the case of the illuminant D65 and for the structure mimicking a red sub-pixel, where more photons are available for down-conversion.
Fig. 3. Calculated spectral fluxes emerging from the structures mimicking green and red sub-pixels under the illumination by (a) CIE illuminant A, (b) CIE illuminant D65. The legends indicate the layered structures.
These examples are by no means optimum ones. The matching between ${T_{CF}}$ and ${S_{em}}$ in Fig. 2 is not ideal and these color filters absorb some of the PL photons with shorter wavelengths. The overlap between ${S_{em}}$ and ${T_{LL}}$ causes self-absorption. Thus, there is a room for improvement in selecting these input parameters as well as for developing a luminescent material with a large Stokes shift.
3. Experiment
3.1 Device preparation
Three devices were prepared for this experiment. The photograph in Fig. 4 was taken by placing these on a black plastic sheet under room light. On the left side is a control device. It is a 50 mm x 50 mm x 2 mm transparent acrylic plate. Five types of color filters (CFs) were attached on its bottom surface with an optical adhesive film. These CFs are sold as a set to be used with a strobe flash in photography. They have various names such as “rust” and “follies pink” and we adopt them for identification in this paper. Using the same adhesive film, a diffuse reflector (Tsujiden, MTN-W400) was attached to cover these CFs partially. This arrangement allowed us to test multiple regions with a different layered structure quickly. For mimicking a red sub-pixel, we fabricated the device on the right side. It is a pair of 1 mm-thick acrylic plates sandwiching a thin layer of Lumogen F Red 305 (BASF). Its fabrication procedure is described in Ref. 17. Two red CFs and the diffuse reflector were attached on its bottom surface although the two CFs are not clearly visible in this photograph. We repeated this fabrication procedure with Coumarin 6. However, its emission characteristic deteriorated after a few months. So, we purchased a 2 mm-thick acrylic plate in which unknown luminescent material was uniformly distributed. Its emission spectrum resembles that of Coumarin 6. We attached the same CFs and the reflector used for the control device. Its picture is in the middle. From left to right in Fig. 4, we call these samples as “control, green, and red device” in this paper
It is interesting to note that the edges of the control device and the green device have different colors. For example, the regions near the green CFs appear more greenish than the surrounding edge surface for the green device. This is because the CFs allow only the room light with right wavelengths to propagate in the plate. In contrast, the edge of the red device has almost no variation in color. This suggests that these CFs are not required for purifying the color of a red sub-pixel if this luminescent material is adopted. This speculation will be confirmed by the spectrum measurement next.
3.2 Spectrum measurement
As shown in Fig. 5, a black cloth is placed on a laboratory scissor jack and the green device is placed on it. A 40 W halogen lamp irradiates the device from a fixed position. To lead the light from the device to a spectrometer (Brolight, model BM6001), its optical fiber probe (Brolight, model BIM-610-1025) is held above it. The numerical aperture of the optical fiber is 0.5 and its core diameter is 1 mm. The height of the stage is adjusted manually so that the tip of the probe becomes in contact with the first spot on the device. The black cloth provides the cushion between them. A small gap between them would result in detecting the light reflected by the surrounding surface of the device. After recording the spectral from one region of the device, we lower the stage and translate the device with everything else untouched. Then the stage is lifted so that the second point of the device becomes in contact with the tip of the probe. This procedure ensures that the geometrical factor is fixed for collecting the light from every measurement spot on the device.
Fig. 5. The setup for recording a spectrum of the light emerging from a device. The tip of the optical fiber probe is in contact with each region of the device representing a different layered structure.
First, we repeated this measurement for the three devices under the illumination by the halogen lamp. The illuminance at the device surface was measured to be about 850 lx. The integration time of the spectrometer was fixed to 100 ms. The spectra recorded at the regions representing different layered structures are compared in Fig. 6(a). The solid curves are the spectra of the light emerging from the regions on the green and red devices. The CFs for these curves are “moss green” (MG) and “rust.” The broken curves are from the regions on the control device. Each legend indicates the layer structure. The dotted curve marked as “halogen lamp” is the spectrum recorded at the region without a CF on the control device.
Fig. 6. Spectra measured under the illumination by (a) the 40 W halogen lamp and (b) the flashlight equipped with a white LED. The CFs for this data set are “moss green” (MG) and “rust.”
The green solid curve in Fig. 6(a) has a peak around 520 nm. This is 1.2 times larger than that of the dotted curve. As for the peak around 680 nm, the ratio slightly increases to 1.3. This peak exceeds that of the incident light. This is in sharp contrast with the control device where reflection never increase the number of photons. In addition, these peaks in the solid curves are narrower than those of the control device due to the first term in Eq. (4), leading to a wider color gamut as will be shown later. As discussed in Section 2, the mechanism behind these behaviors is photoluminescence: a luminescent material absorbs photons with shorter wavelengths and emits photons with longer wavelengths.
Next, we needed another common light source with a different spectrum to see how the color gamut varies. For this purpose, the halogen lamp was replaced by a flashlight equipped with a white LED (Cree, model XHP99). The illuminance at the device surface was measured to be about 3,500 lx. As shown by the dotted curve, this flashlight has an LED emitting around 450 nm to excite its phosphor. The integration time was reduced to 30 ms and the whole procedure was repeated. As shown in Fig. 6(b), the effects of the luminescent materials are more clearly observed. While the ratio of the peak around 520 nm is 1.2, this factor increases to 3.3 for the peak around 680 nm. The white LED emits more photons in the short wavelength range than the halogen lamp. The luminescent dye (Lumogen F Red 305) utilizes them as shown by the red solid curve in Fig. 6(b).
The spectra recorded at the regions covered by CFs “steel green” (SG) and “follies pink” (FP) are shown in Fig. 7. The spectra of the light sources are reproduced here to serve as references for comparison. The peaks in the solid and broken curves in Fig. 7 are larger, wider and are shifted to the left. These differences are caused by the transmittance characteristics of the CFs. A wider transmission band increases brightness but shrinks color gamut. Note also that the green broken curve in Fig. 7(b) has a large peak around 450 nm. These features in the spectra in Fig. 7 would shrink the color gamut of the reflective display. There is no such a peak in the green solid curve because the luminescent layer absorbs these photons. These examples show that luminescent materials widen the choice of color filters. This design freedom can ease the trade-off between brightness and color gamut.
Fig. 7. Spectra measured under the illumination by (a) the 40 W halogen lamp and (b) the flashlight equipped with a white LED. The CFs for this data set are “steel green” (SG) and “follies pink” (FP).
The role of the CFs becomes clear when we plot some of the data in the previous two figures together with the spectra recorded at the regions without CFs on the two devices. The solid curves in Fig. 8 are the spectra of the PL photons from the luminescent layers and the reflected ambient light. The CF “moss green” absorbs the photons more strongly than the CF “steel green.” The luminescent layer absorbs the photons with shorter wavelengths. In contrast, the effect of the CFs “rust” and “follies pink” on the spectra recorded at the regions without them is small. This is consistent with the fact that these CFs are barely visible in Fig. 2. Hence, if the luminescent layer absorbs the photons with shorter wavelengths completely, CFs are not required for a red-sub-pixel in an LRD. They are needed only to provide a fair comparison in this experiment. In addition, the red dotted curves are clearly above the red solid curves: the CF “follies pink” increases the spectral flux. This unexpected finding suggests that this CF contains some luminescent materials. As pointed out in Ref. 7, a single layer performing both down-conversion and filtering unwanted wavelengths might be advantageous for reducing the manufacturing cost of this layered structure.
Fig. 8. Spectra measured under the illumination by (a) the 40 W halogen lamp and (b) the flashlight equipped with a white LED. The role of the CFs is to shape the spectra.
4. Analysis
4.1 Model prediction
Using the information on the materials and components at hand, we tried to reproduce the spectra in Fig. 6 by the model described in Section 2. The transmittance of each color filter was calculated from the spectrum measured with each CF and the halogen lamp. The spectra of the halogen lamp and the white LED were measured at the region covered by the diffuse reflector on the control device. They were normalized such that $\int {{S_{em}}(\lambda )} \,d\lambda = 1$. As for the emission and absorption spectra of the two luminescent materials, we adopted the data from the literatures [14,15]. The reason behind this decision is because the emission spectra depend on the emission angle and the setup in Fig. 5 accepts PL photons with various emission angles. The absorption coefficients were scaled to adjust the parameter ${T_{LL}}$, which was the only fitting parameter used in this analysis. We also assumed $\eta \, = R = 1$. The spectra were calculated with these input parameters for the two cases of light sources. The results are shown in Fig. 9.
Fig. 9. The spectral fluxes calculated by the model for the case of (a) the halogen lamp and (b) the white LED. The CFs for this data set are “moss green” and “rust.”
At glance, the measured spectra are reproduced by the expression reasonably well. However, there are some discrepancies. They are probably related to the input parameters. First, the two red curves in Fig. 9 have non-zero baselines in the wavelength range smaller than 600 nm. This can be attributed to a measurement error in determining the transmittance characteristics of the color filters. Second, the measured spectra for the configurations with luminescent materials are narrower than the model prediction. One possible mechanism behind this is self-absorption. Namely, the PL photons need to propagate long distances in the layer after diffuse reflection and those with shorter wavelengths are partially absorbed by the luminescent materials due to the overlap between the emission and absorption spectra. The same mechanism is responsible for the angle-dependent PL spectra observed with a uniform luminescent layer [11]. The model described in Section 2 neglects self-absorption. In addition, interaction between a dye molecule and its host material can play a role. In fact, the emission spectra of Coumarin 6 reported in the literatures are not identical [14,18,19]. Self-absorption and possible interactions with a host material lead to the difficulty of measuring quantum yield accurately. Third, the assumption of $\eta \, = 1$ may not hold for the green luminescent material at hand although it is reported this is the case for Lumogen F Red 305 [20].
Despite the difficulties for specifying the input parameters, the model reproduces the following features of the measured spectra. First, a luminescent layer increases the peak height of a spectrum without widening it. This is a clear advantage over the purely reflective structure. Second, this enhancement of the peak heights is larger for the configuration mimicking a red sub-pixel.
4.2 Color gamut
Chromaticity coordinates are calculated from the spectra in Fig. 6 and the CIE1931 color matching functions. The control device represents the sub-pixels in a reflective display. Its color gamut is shown by the broken triangles in Fig. 10. It covers 35% and 53% of the NTSC standard for the cases of the halogen lamp and the white LED, respectively. As for a hypothetical LRD, we assume that its blue sub-pixel is purely reflective. As shown by the solid triangles in Fig. 10, its color gamut is substantially larger than that of a reflective display. Adopting the definition of color gamut area as the ratio of the triangle areas [21], it is 44% and 82% for the cases of the halogen lamp and the white LED, respectively. These values are larger than those for the control device by a factor of from 1.26 to 1.55. The lower left apex is identical for the two triangles in Fig. 10 because the blue broken curves in Fig. 6 are used in this calculation. Its position depends on the light source because the blue sub-pixel is purely reflective. The other apexes of the LRD triangles also move to a lesser extent. This is due to the second term in Eq, (4). Narrowing the transmission bands would decrease this term. Replacing the reflector with a light-absorbing layer would eliminate it. Both measures would result in a stable color gamut but at the expense of brightness.
Fig. 10. Color gamut calculated from the spectra in Fig. 6. The light source is (a) the halogen lamp and (b) the flashlight (white LED).
5. Conclusions
Ambient contrast ratio of an electro-optic shutter stacked on a luminescent layer remains constant even at an extremely high level of illuminance. A reflector below the luminescent layer enhances its brightness. With a color display application in mind, we have expressed the spectral flux of the photons emerging from the stack of a luminescent layer, a color filter and a reflector. There are two terms in this expression, one for the photoluminescence photons and the other for the reflected incident light. For this reason, we call a display based on this configuration a luminous reflective display (LRD). The first term is proportional to the flux of the incident light absorbed by the luminescent layer. The second term becomes equal to the spectral flux from the configuration without the luminescent layer if the transmittance of the luminescent layer is reduced to zero in the wavelength range of the color to be displayed. Hence, the first term presents the following advantages of an LRD over an equivalent reflective display. First, its luminance is larger. Second, its color gamut is wider and more stable.
The experiment with off-the-shelf components have shown that insertion of a luminescent layer increases the spectral fluxes from the layered structures mimicking a green and red sub-pixel. Under the illumination by a halogen lamp, the ratio of the peak height is 1.2 for a green sub-pixel and 1.3 for a red sub-pixel. When the halogen lamp is replaced by a white LED, this factor for the red-sub-pixel increases to 3.3 because more photons are available for the luminescent material to utilize. The measured spectra are reproduced by the expression reasonably well. The color gamut of a hypothetical LRD with red and green sub-pixels and purely reflective blue sub-pixels is calculated from the measured spectra. It covers 44% and 82% of the NTSC standard for the cases of the halogen lamp and the white LED, respectively. These are larger than the values for an equivalent reflective display by a factor of from 1.26 to 1.55.
We expect improvement in these characteristics by tuning the matching between the emission spectra of the luminescent materials and the transmittance characteristics of the color filters. A luminescent material with a large Stokes shift is also desired. This goal for material research is shared by the field of luminescent solar concentrators.
Acknowledgments
We would like to thank Prof. Mitsuhiro Shigeta for helpful discussions.
Disclosures
The authors declare no conflicts of interest.
References
1. 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]
2. 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. - Soc. Inf. Disp. Int. Symp. 50(1), 279–282 (2019). [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. Y. Tsutsumi, S. Matsuda, and I. Fujieda, “Enhanced utilization of ambient light by optical films coupled to a luminescent waveguide,” Proc. of 27th International Display Workshops, 115–118 (2020).
5. C. Liu, T. Yang, and K. Cheng, “Expanding color gamut of reflective liquid crystal displays from filtering undesirable wavelengths of a light source by an embedded etalon,” Appl. Opt. 56(7), 1880–1885 (2017). [CrossRef]
6. O. Okumura, patent application JP-A-2001-264756 (in Japanese).
7. G. Gibson, X. Sheng, D. Henze, S. Lam, P. Beck, Y. Jeon, Z. Zhou, B. Benson, Q. Liu, G. Combs, T. Koch, and K. Biggs, “Fast full-color reflective display via photoluminescent enhancement,” J. Soc. Inf. Disp. 20(10), 552–558 (2012). [CrossRef]
8. 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.
9. S. Wu and C. Wu, “Mixed-mode twisted nematic liquid crystal cells for reflective displays,” Appl. Phys. Lett. 68(11), 1455–1457 (1996). [CrossRef]
10. 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]
11. I. Fujieda and M. Ohta, “Angle-resolved photoluminescence spectrum of a uniform phosphor layer,” AIP Adv. 7(10), 105223 (2017). [CrossRef]
12. W. H. Weber and J. Lambe, “Luminescent greenhouse collector for solar radiation,” Appl. Opt. 15(10), 2299–2300 (1976). [CrossRef]
13. F. Meinardi, A. Colombo, K. A. Velizhanin, R. Simonutti, M. Lorenzon, L. Beverina, R. Viswanatha, V. I. Klimov, and S. Brovelli, “Large-area luminescent solar concentrators based on ‘Stokes-shift-engineered’ nanocrystals in a mass-polymerized PMMA matrix,” Nat. Photonics 8(5), 392–399 (2014). [CrossRef]
14. E. F. Schubert, Light Emitting Diode, 2nd edition (Cambridge University, 2006) pp. 350.
15. O. M. ten Kate, K. M. Hooning, and E. van der Kolk, “Quantifying self-absorption losses in luminescent solar concentrators,” Appl. Opt. 53(23), 5238–5245 (2014). [CrossRef]
16. J. He, H. Chen, H. Chen, Y. Wang, S. Wu, and Y. Dong, “Hybrid downconverters with green perovskite-polymer composite films for wide color gamut displays,” Opt. Express 25(11), 12915–12925 (2017). [CrossRef]
17. R. Matsumura, Y. Tsutsumi, and I. Fujieda, “Concentric re-emission pattern from a planar waveguide with a thin uniform luminescent layer,” Appl. Opt. 59(6), 1703–1710 (2020). [CrossRef]
18. A. S. Kristoffersen, S. R. Erga, B. Hamre, and Ø. Frette, “Testing fluorescence lifetime standards using two-photon excitation and time-domain instrumentation: Rhodamine B, Coumarin 6 and Lucifer Yellow,” J. Fluoresc. 24(4), 1015–1024 (2014). [CrossRef]
19. C. C. Ruiz, J. M. Hierrezuelo, and J. A. Molina-Bolivar, “Analysis of the photophysical behavior and rotational-relaxation dynamics of coumarin 6 in nonionic micellar environments: the effect of temperature,” Molecules 20(10), 19343–19360 (2015). [CrossRef]
20. 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]
21. R. Zhu, Z. Luo, H. Chen, Y. Dong, and S. Wu, “Realizing Rec. 2020 color gamut with quantum dot displays,” Opt. Express 23(18), 23680–23693 (2015). [CrossRef]
