Open Access
Add to BibSonomy
Abstract
The current efficiency and color purity of organic light-emitting diodes (OLEDs) can be easily improved by means of a microcavity structure, but this improvement is typically accompanied by a deterioration in the characteristics of viewing angle. To minimize the angular dependence of the color characteristics exhibited by these strong microcavity devices, we investigated the changes in the optical properties of the green OLED with a bottom resonant structure. This investigation was based on varying the hole transport layer and semitransparent anode thicknesses. The results of optical simulations revealed that the current efficiency and viewing angle characteristics can be simultaneously improved by adjusting the thickness of the two layers. Furthermore, optical simulations predicted that the angular color dependence could be limited to 0.019 in the International Commission on Illumination (CIE) 1976 coordinate system. This optimum condition yielded a current efficiency of ∼134 cd/A. To further reduce this color shift, a nanosized island array (NIA) was introduced through the dewetting process of cesium chloride. By employing NIAs, the color coordinate shift value was reduced to 0.016 in the CIE 1976 coordinate system, and a current efficiency of 130.7 cd/A was achieved.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The development of light and thin products has become quite desirable as industries in various fields have rapidly grown. These market demands apply to all display applications, and as a result, such applications have rapidly shifted from conventional liquid crystal displays (LCDs) to advanced organic light-emitting diodes (OLEDs) [1–8]. Since OLEDs were discovered in 1987, they have overcome the disadvantages (e.g., large thickness, low contrast ratio, and inflexible design) of LCDs owing to the use of thin flexible substrates and spontaneous emitting pixels. These advantages have led to the commercialization of OLED display applications through rapid technological advances, but the short lifetime and low efficiency of these applications have hindered the rapid market expansion of OLED technology [9–11]. In particular, the external quantum efficiency obtainable from OLEDs is only ∼20% due to numerous emission loss mechanisms, including surface plasmon coupling, waveguide modes, and total internal reflection (TIR) [12–14]. Therefore, outcoupling enhancement technologies for overcoming the short lifetime by operating at a low current density have been extensively investigated. These investigations have included suppressing the TIR and, in turn, easily increasing the efficiency by attaching or forming microlens arrays (MLAs) on the substrate surface [14–18]. However, this technology results in blurry pixel problems, where the distinction of pixel boundaries via visual inspection is impossible, thereby preventing its use in commercialized display applications [19,20]. Thus, rather than developing MLA technology to control the surface morphology of OLEDs, OLED display manufacturers are developing strong microcavity devices for application to many display products [21–23]. In this study, we investigated the characteristics of bottom emission type microcavity devices that are applicable to TVs. The bottom emission OLEDs (BEOLEDs) exhibiting a microcavity effect can be described by the Fabry–Perot resonance formulas:
In formula (1), Iext(θ,λ,z) denotes the external spectral emission intensity at the emission angle, θ; the wavelength, λ; and the distance from the highly reflective cathode to the dipole of the emitter, z. Ta corresponds to the transmittance of the semitransparent anode, and Rc and Ra denote the reflectivity of the cathode and anode, respectively. Iint(θ,λ) denotes the internal emitting intensity at λ and θ. In formula (2), ΔφFP and ΔφTBI represent the phase shifts induced by multiple-beam interference and two-beam interference (Fabry–Perot), respectively. φc and φa are phase shifts that are changed by the reflectivity at the cathode and anode, respectively. dn, nn, and θn denote the respective thickness, refractive index, and angle of light progress path in the “n”-th layer. norg corresponds to the refractive index of the organic material. In particular, ΔφFP is correlated with the viewing angle dependency, and the emission spectral value at the angle θ can be calculated from ΔφFP, as presented in formula (3). φ(θ) corresponds to the phase shift due to the cathode and anode at angle θ, and “m” denotes an integer (e.g., 0, 1, 2,…). As the viewing angle is increased, the outcoupled emission spectra generally undergo a hypsochromic wavelength shift, as indicated by formula (3). Thus, control of the angular color uniformity is important when OLEDs with strong microcavity are adopted in large-sized displays, such as in the case of televisions and computer monitors. While the angular chromaticity changes (Δu’v’ in CIE 1976 space) of In-Plane Switching (IPS) LCD applications and commercialized white OLED TVs composed with tandem structure are about 0.015, those of most RGB OLEDs in mobile phones and tablet PCs are about 0.025 to 0.03. Since the angular chromaticity change of RGB OLEDs with strong microcavity structure are about 2 times higher than those of IPS LCD applications and those of white OLED TVs, we have been studying the improvement of viewing angle dependence by nanosized scattering pattern. To improve the angular color uniformity of OLEDs exhibiting strong microcavity behaviors, we introduced a nanosized porous film (NPF) and obtained a stable angular chromaticity change with very weak pixel blurring, as we reported in previous studies [24–27]. Unfortunately, our NPF was fabricated by means of a spin coating process in a humid environment, including water mist. The application of this NPF to large-sized displays is therefore limited due to unstable coating thickness and water mist control. To apply a nanosized pattern on a large-sized OLED display exhibiting strong microcavity, we introduce a considerably more practical fabrication method than spin coating to prepare nanosized island arrays (NIAs) via vacuum deposition. The results revealed that the angular chromaticity change and the luminous efficiency were improved by changing the thickness of the anode and organic layer. In addition, when we formed NIAs on the glass substrate of the devices, the angular chromaticity change was additionally improved with very low pixel blurring effect.
2. Experiment
2.1 Materials
Indium tin oxide (ITO) is extensively used as a transparent electrode for display applications, and therefore, we selected ITO as the anode for a BEOLED device. To realize a strong microcavity effect by increasing the reflectivity of the anode, the thin silver (Ag) layer was deposited under the anode. To improve the luminous efficiency of the BEOLEDs, we applied ITO as a capping layer before depositing the Ag layer. The hole transport in the devices was enhanced by using N,N’-bis(naphthalen-1-yl)-N,N’-bis(phenyl)benzidine (NPB) as a hole transport layer (HTL). We used 1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN) in order to implement a thick HTL by inserting a hole generation layer. Tris(4-carbazol-9-ylphenyl)amine (TCTA) was employed as an electron blocking layer (EBL) as well as HTL. Similarly, to fabricate the green phosphorescent devices, beryllium bis(2-(2′-hydroxyphenyl) pyridine) (Bepp2) and bis(2-phenylpyridine)(acetylacetonato)iridium(III)(Ir(ppy)2(acac)) were applied as a host and dopant material, respectively. Furthermore, 4,7-diphenyl-1,10-phenanthroline (BPhen), lithium fluoride (LiF), and aluminum (Al) were applied as an electron transport layer (ETL), an electron injection layer, and cathode, respectively. We formed NIAs by using cesium chloride (CsCl; Sigma-Aldrich Co., LLC) with a purity of 99.9% and a passivation layer (P/L) by applying SPC-370 (FOSPIA Co., Ltd.).
2.2 Fabrication of devices
We prepared BEOLEDs with an emission area of 4 mm2 (2 × 2 mm). Afterward, 0.5-mm-thick glass substrates with patterned electrodes and a bank layer were successively cleaned in acetone, isopropyl alcohol, and deionized (DI) water. Then, they were exposed to UV and ozone for stable electrical characteristics. Organic layers were deposited (deposition rate: <1 Å/s) under vacuum conditions of 5 × 10−5 Pa, whereas LiF and Al were deposited at a rate of 0.15 and 2.5 Å/s, respectively. After all the evaporation processes, we encapsulated OLED devices by using a glass lid to prevent oxidation and degradation.
2.3 Introduction of nano-island arrays
The schematic presented in Fig. 1 illustrates the fabrication of NIAs on a glass substrate. After inserting the substrates in a vacuum chamber, CsCl powder was thermally evaporated in the chamber at a deposition rate of ∼1 Å/s. The CsCl layer was then exposed to air. The temperature and relative humidity were controlled to 30°C and 60% for 3 min, respectively. During this process, the thin and flat CsCl layer underwent dewetting and was transformed into randomly distributed hemispherical NIAs via the absorption of water vapor [28,29]. Self-assembled CsCl NIAs can be dissolved and volatized in the air, and hence, we coated the SPC-370 over the CsCl NIAs via a spin coating process and then cured the NIAs in the presence of UV radiation [26].
2.4 Measurements
We examined current densities and driving voltages using an electrical sourcemeter (Keithley 238, Keithley). Similarly, we investigated optical properties, such as the electroluminescence (EL) spectra, luminance, and color coordinates, by using a spectroradiometer (CS-2000A). The morphologies of the NIAs were evaluated via FE-SEM (S-4700, Hitachi). Moreover, we used a spectral haze meter (SH 7000, Nippon Denshoku Industries) to measure optical properties such as haze and transmittance induced by various morphologies of the NIAs. To validate the optical properties of strong microcavity OLEDs, simulations were performed using a commercial simulation program (SETFOS, FLUXiM).
3. Results and discussion
3.1 Evaluation of NIAs
To investigate the optical properties based on the diameter and height of the NIAs, we deposited separate CsCl films with thicknesses of 50 and 100 nm. Figure 2 presents SEM images of various NIAs formed from different thicknesses of CsCl. First, NIAs with a diameter of ∼400 nm were formed from a 50-nm-thick film, as presented in Fig. 2(a), and were randomly distributed on the glass. An oblique image of these NIAs [see Fig. 2(b)] reveals the well-formed hemispherical shape of the NIAs (height: 200 nm). These NIAs were obtained by exposing CsCl to 30°C and 60% relative humidity conditions for 10 min. Figure 2(c) shows that NIAs formed from the 50-nm-thick CsCl film were well maintained even after being covered by SPC-370. The NIAs presented in Figs. 2(d) to 2(f) were obtained at a film thickness of 100 nm. As presented in Figs. 2(d) and 2(e), the diameter and height of these hemispherical NIAs were ∼600 and 250 nm, respectively. Furthermore, Fig. 2(f) shows that NIAs formed from the 100 nm film were well maintained even after being overcoated with SPC-370. As previously mentioned, because the CsCl film can absorb water in the air and undergo dewetting, the CsCl NIAs should be protected by overcoating with a passivation layer. The scattering effect of CsCl NIAs with a refractive index of 1.64 in the passivation layer was increased by using SPC-370 with a refractive index of 1.37 as the passivation layer. This usage was based on the fact that the scattering effect increases with increasing difference between the refractive index of the scattering material and the passivation material.
Fig. 2. FE-SEM images showing several morphologies of NIAs. (a) Front image and (b) tilted image of NIAs formed from 50 nm CsCl film. (c) Cross-sectional image of multilayer composed of SPC-370 and NIAs formed from 50 nm CsCl film. (d) Front image and (e) tilt image of NIAs formed from 100 nm CsCl film. (f) Cross-sectional image of multilayer composed of SPC-370 and NIAs formed from 100 nm CsCl film. The inset images in (c) and (f) represent high-magnification views of NIAs in the passivation layer, SPC-370.
As presented in Fig. 2, two NIAs of different diameters and heights were obtained from two different thin film thicknesses (see Fig. 3 and Table 1 for a summary of the corresponding optical properties). Figures 3(a) and 3(b) presents the nondiffusing transmittances and diffusing transmittances of various CsCl NIAs. Nondiffusing transmittance and diffusing transmittance of NIAs prepared from 50 nm CsCl film were 66% and 23%, respectively, at a 550 nm wavelength. For the same wavelength, the corresponding transmittances of NIAs formed from 100-nm-thick CsCl film were 45% and 40%, respectively. After coating of the SPC-370 over the NIAs, the parallel transmittance and diffusing transmittance were improved to 89% and 4% for NIAs prepared from 50-nm-thick CsCl film and 84% and 9% for NIAs prepared from 100-nm-thick CsCl film, respectively. Figure 3(c) presents the total transmittance of various NIAs. The total transmittance can be calculated from the arithmetical sum of the parallel transmittance and diffusing transmittance. Therefore, the total transmittances of NIAs without the passivation layer were relatively lower than that of NIAs, including the layer. Figure 3(d) presents the haze characteristics of various NIAs. The haze value can be calculated from the ratio of the diffusing transmittance to the total transmittance and can be reduced by applying a flat passivation layer. The optical haze values of NIAs prepared from 50-nm-thick CsCl film and 100-nm-thick CsCl film were 26% and 47%, which decreased to 4% and 10%, respectively, when SPC-370 was overcoated on the NIAs. In general, the diameter and height of the CsCl hemisphere increased with increasing initial thickness of the CsCl film, thereby leading to a reduction in the Mie scattering effect and an increase in the diffusing transmittance. Thus, the haze value of NIAs formed from the 100 nm CsCl film was larger than that of NIAs formed from the 50 nm CsCl film.
Fig. 3. (a) Parallel transmittance-wavelength, (b) diffused transmittance-wavelength, (c) total transmittance-wavelength, and (d) haze-wavelength behaviors of various CsCl NIAs.
Table 1. Optical behaviors of NIAs on the glass substratesa.
3.2 Optical simulation of OLED exhibiting strong microcavity behaviors
To investigate the luminous efficiency and the shift in the angular chromaticity of OLEDs exhibiting strong microcavity effects, we performed an optical simulation using the SETFOS software based on the Fabry–Perot resonator model. The reference OLED device for the optical simulation is as follows:
Reference: glass(0.5 mm)/ITO(55 nm)/Ag(10 to 30 nm)/ITO(10 nm)/HTL(150 to 190 nm)/EBL(18 nm)/EML(20 nm)/ETL(35 nm)/LiF(1.5 nm)/Al(100 nm)
The refractive indices of the organic layers were assumed to be 1.75 for the optical simulation. However, the refractive indices of LiF, Al, ITO, and Ag were applied based on the values provided by the SETFOS software. The emission spectrum of the green dopant in EML was taken as the photoluminescence (PL) spectrum of Ir(ppy)2(acac). To verify the optical characteristics of the reference device, the optical simulations considered various thicknesses of the semitransparent anode, Ag, and HTL.
Figure 4 presents a result of an optical simulation capable of predicting the changes in the normalized luminous efficiency and angular chromaticity that accompany variations in the thickness of Ag and HTL. As presented in Fig. 4(a), with changing thickness of Ag, the luminous efficiency changed only slightly in some regions of the plot but changed significantly in other regions. As shown in circles A and B in the figure, for a HTL thickness of 165 nm, the efficiency changed by only ∼2% when the Ag thickness was changed from 16 to 24 nm. However, the efficiency increased by ∼1.3 times when the HTL thickness was changed from 165 nm of circle A to 175 nm of circle C. The efficiency of circle D fabricated with 175-nm-thick HTL and 24-nm-thick Ag was 1.4 times higher than that of circle A fabricated with 165-nm-thick HTL and 24-nm-thick Ag. Figure 4(b) presents the angular chromaticity change accompanying the Ag and HTL thickness changes. The angular chromaticity change was calculated as the maximum chromaticity change occurring when the viewing angle is changed from 0° to 60°.
In the above Eq. (4), u’ and v’ refer to the color coordinates in the CIE 1976 chromaticity space and subscripts, and 0 and x denote the measurement angle. For example, u’0 corresponds to u’ along the perpendicular direction from the emitting area. An angular chromaticity change of ∼0.036 occurred when 24-nm-thick Ag and 165-nm-thick HTL were applied as circle B. This change decreased to ∼0.019 when we applied a 24-nm-thick Ag and 175-nm-thick HTL as circle D. In general, the relationship between the luminous efficiency and angular chromaticity stability in the strong microcavity device is considered a trade-off. However, we obtained a high luminous efficiency and a stable angular chromaticity change simultaneously by setting a certain HTL thickness, which was determined from optical simulations.
Fig. 4. Contour map simulation of (a) normalized luminous efficiency with variations in the thickness of HTL and Ag and (b) angular chromaticity change with variations in the thickness of HTL and Ag. Thicknesses of HTL and Ag in the contour map are 165 nm and 16 nm for circle A, 165 nm and 24 nm for circle B, 175 nm and 16 nm for circle C and 175 nm and 24 nm for circle D, respectively.
3.3 Investigation of OLEDs including NIAs
We fabricated BEOLEDs exhibiting a strong microcavity effect by changing the thickness of Ag and HTL to evaluate the optical characteristics. Bepp2 was doped with 3% Ir(ppy)2(acac) [the thickness conditions for the devices are presented in Table 2 and Fig. 5(a)].
Fig. 5. Optical properties of various BEOLEDs. (a) BEOLED structures used in this verification; Plot showing the (b) current density–voltage–luminance (J-V-L) behaviors, (c) efficacy–luminance behaviors, (d) power efficacy–luminance behaviors, and (e) normalized luminance-viewing angle behaviors.
Table 2. Stack structure of BEOLEDsa.
To confirm the optical properties of the strong microcavity BEOLED, the Ag thickness was varied, with three thicknesses (16, 20 and 24 nm) employed. The first NPB thickness was varied as well, with two values (75 and 85 nm) employed, as indicated by bold letters in Table 2. All green BEOLEDs were fabricated for compliance with the DCI-P3 color gamut; the y color coordinate of green should be >0.690 in the CIE 1931 chromaticity space. The optical properties of various BEOLEDs are presented in Fig. 5. Various luminance values were obtained at the same driving voltage in these BEOLEDs. The driving voltage and current density behaviors differed only slightly among the BEOLEDs, as presented in Fig. 5(b). A luminance of 1000 cd/m2 required driving voltages of 4.20, 4.23, and 4.23 V for BEOLEDs A, B, and C, respectively, which were composed of a 75-nm-thick first NPB layer. However, driving voltages of 4.22, 4.15, and 4.14 V were required for BEOLEDs D, E, and F, which were composed of an 85-nm-thick first NPB layer. These results indicated that the current and power efficiencies [see Figs. 5(c) and 5(d), respectively] of BEOLEDs with an 85-nm-thick first NPB layer are higher than those of BEOLEDs with a 75-nm-thick NPB layer. For 1000 cd/m2 and a 75-nm-thick first NPB layer, current efficiencies of 100.7, 97.9, and 92.1 cd/A were realized for BEOLEDs A, B, and C, respectively. For the 85-nm-thick first NPB layer, the corresponding efficiencies of 113.6, 129.1, and 134.1 cd/A were obtained for BEOLEDs D, E, and F, respectively [see Fig. 5(c)]. The result presented in Fig. 5(c) is quite consistent with the current efficiency behaviors revealed by optical simulations [Fig. 4(a)] of the BEOLEDs. The power efficiencies at 1000 cd/m2 of BEOEDs A (61.3 lm/W), B (52.5 lm/W), and C (44.2 lm/W) were lower than those of BEOLEDs D (74.2 lm/W), E (76.0 lm/W), and F [72.6 lm/W; see Fig. 5(d)]. This result indicated that the current efficiency and power efficiency of BEOLED F were enhanced 1.46 times higher and 1.64 times higher than those of BEOLED C. Furthermore, the power efficiency enhancement rates of BEOLEDs D and E were 1.21 and 1.45 times those of BEOLEDs A and B, whereas the current efficiency enhancement rates were 1.16 and 1.32 times, respectively. The discrepancy between the current efficiency and power efficiency enhancement may have resulted from angular luminance variations, as presented in Fig. 5(e). The power efficiency can be calculated by integrating the angular luminance variation. Therefore, the enhancement rate of this efficiency can be considerably larger than that of the current efficiency when the luminance variation is expanded in the direction of the Lambertian behavior. Figure 5(e) shows that the behavior of BELOEDs with an 85-nm-thick layer was closer to the Lambertian behavior than that of BEOLEDs with a 75-nm-thick layer.
Moreover, the angular chromaticity change of the BEOLEDs was evaluated for angles ranging from 0° to 60° to investigate the viewing angle dependence by changing the thickness of the anode and HTL. When we applied the first NPB thickness of 75 nm, the chromaticity changes increased with increasing thickness of the Ag [see Figs. 6(a)–6(c)]. The angular chromaticity changes and angular EL peak wavelength (Wp) changes for angles ranging from 0° to 60° were 0.017 and 2 nm for BEOLED A, 0.028 and 3 nm for BEOLED B, and 0.037 and 5 nm for BEOLED C, respectively. In addition, we evaluated the angular chromaticity changes of BEOLEDs D, E, and F to determine the effect of the NPB thickness on these changes [Figs. 6(d)–6(f)]. The chromaticity changes and Wp changes of BEOLEDs with a 85-nm-thick NPB were 0.015 and 10 nm for BEOLED D, 0.017 and 13 nm for BEOLED E, and 0.019 and 16 nm for BEOLED F, respectively. The angular chromaticity changes of BEOLEDs D, E, and F were smaller than those of BEOLEDs A, B, and C, although the reverse was true for the angular Wp changes.
Fig. 6. Angular EL spectrum change of (a) BEOLED A, (b) BEOLED B, (c) BEOLED C, (d) BEOLED D, (e) BEOLED E, and (f) BEOLED F. The inset of each plot shows the angular chromaticity changes obtained from calculation in the CIE1976 space.
The small angular chromaticity changes obtained for BEOLEDs D, E, and F can be explained by calculating the length of the angular chromaticity locus presented in Figs. 7(a) and 7(b). Figure 7(a) presents the angular chromaticity loci ranging from 0° to 60° for BEOLEDs A, B, and C. Angular chromaticity locus lengths of 0.019, 0.030, and 0.040 were obtained for BEOLEDs A, B, and C, respectively, and in each case, the angle of maximum chromaticity change was 60°. However, the angular chromaticity locus lengths were 0.023 for BEOLED D, 0.033 for BEOLED E, and 0.041 for BEOLED F, and the angle of maximum chromaticity change in each case was 40° [see Fig. 7(b)]. These results indicated that the length of the angular chromaticity locus was increased by increasing the thickness of the semitransparent electrode or organic layers. Contrarily, if the inflection point was present in the locus, this thickness increase yielded no increase in the value of the chromaticity change. Thus, the angular chromaticity change can be reduced by adjusting the thickness of the device without reducing the current efficiency, as shown in the optical simulation results presented in Fig. 4.
Fig. 7. Angular chromaticity loci of (a) BEOLEDs A, B, and C with a 75-nm-thick first NPB layer and (b) BEOLEDs D, E, and F with an 85-nm-thick first NPB layer. The loci were evaluated in the CIE 1976 chromaticity space.
To further improve the angular chromaticity change, we formed the CsCl NIAs including a SPC-370 passivation layer on the glass substrates of BEOLEDs D, E, and F as shown in Fig. 8(a).
Fig. 8. Optical properties of BEOLEDs including NIAs. (a) BEOLED structures including CsCl NIAs used in this evaluation; Plots showing (b) current J-V-L behaviors, (c) current efficacy–luminance behaviors, (d) power efficacy–luminance behaviors, and (e) normalized luminance-viewing angle behaviors.
BEOLED D’: SPC-370 passivation layer (∼8 µm)/CsCl NIAs (∼250 nm)/BEOLED D
BEOLED E’: SPC-370 passivation layer (∼8 µm)/CsCl NIAs (∼250 nm)/BEOLED E
BEOLED F’: SPC-370 passivation layer (∼8 µm)/CsCl NIAs (∼250 nm)/BEOLED F
The J-V-L behaviors of BEOLEDs D’, E’, and F’ are presented in Fig. 8(b). As shown in the figure, these behaviors differed only slightly between pristine BEOLEDs and BEOLEDs including NIAs. This indicated that CsCl deposition and NIA formation have no effect on the electrical characteristics of the BEOLEDs. Driving voltages of 4.24, 4.21, and 4.19 V were required for BEOLEDs D’, E’, and F’, respectively, to achieve 1000 cd/m2. Through comparison with BEOLEDs D, E, and F, we found that the driving voltages of the BEOLEDs increased slightly when CsCl NIAs were formed on the substrate. This increase can be explained by Figs. 8(c) and 8(d). Efficiencies of the current and power at 1000 cd/m2 were 107.7 cd/A and 73.0 lm/W for BEOLED D’, 125.4 cd/A and 74.2 lm/W for BEOLED E’, and 130.7 cd/A and 71.5 lm/W for BEOLED F’, respectively. Compared with the current and power efficiencies of pristine BELOEDs, the efficiencies of BEOLEDs including NIAs were lower by 5% and 2% for BEOLED D’, 3% and 2% for BEOLED E’, and 3% and 2% for BEOLED F’, respectively. However, the angular luminance variations of the BEOLEDs including NIAs [see Fig. 8(e)] were slightly (∼2%) larger than those of the pristine BEOLEDs.
Thus, the drop rate of power efficiency resulting from the application of NIAs was smaller than that of the current efficiency. Although the efficiencies of the BEOLEDs were reduced by applying NIAs, the angular chromaticity changes of the BEOLEDs including NIAs were improved compared with those of pristine BEOLEDs, as presented in Fig. 9. Figures 9(a), 9(b) and 9(c) present the angular EL spectrum changes of BEOLEDs including NIAs. Angular Wp changes of 9, 12, and 13 nm were determined for BEOLED D’, BEOLED E’, and BEOLED F’, respectively. These results indicated that the Wp change of BEOLEDs exhibiting microcavity characteristics can be reduced by over 1 nm through the application of NIAs. Moreover, the angular chromaticity changes of BEOLEDs were reduced by applying NIAs, as presented in the inset of Figs. 9(a), 9(b), and 9(c). Angular chromaticity changes (in the CIE 1976 system) of 0.013, 0.014, and 0.016 were determined for BEOLED D’, BEOLED E’, and BEOLED F’, respectively. To verify the NIA-induced suppression of angular chromaticity changes, we compared the angular chromaticity loci with those of the pristine BEOLEDs, as presented in Fig. 9(d). Chromaticity locus lengths of 0.021, 0.030, and 0.035 were obtained for BEOLED D’, BEOLED E’, and BEOLED F’, respectively. These results revealed that NIAs had no effect on the shape of the angular chromaticity locus but can reduce the length of the locus. The NIA-induced suppression of the angular chromaticity change was more effective for BEOLEDs exhibiting strong microcavity behaviors (than for BEOLEDs exhibiting weak behaviors). The optical characteristics of the BEOLEDs are summarized in Table 3. (To further investigate angular chromaticity change, we compared various BEOLEDs including NIAs without SPC-370 passivation layer in the Supplemental Document.)
Fig. 9. Angular EL spectrum changes of (a) BEOLED D’, (b) BEOLED E’, and (c) BEOLED F’. (d) Angular chromaticity loci of BEOLEDs in the CIE 1976 chromaticity space. Inset graphs of (a), (b), and (c) show the angular chromaticity changes revealed by calculating Δu’v’.
Table 3. Optical characteristics of BEOLEDs.
3.4 Visibility of pixels for display application
We took pictures of light-emitting pixels to verify whether NIAs could influence the legibility. The level of pixel blur related to legibility was quantified by analyzing the brightness information of the captured pixel images. The pixel blur level was defined as the distance from the pixel edge to the location where 10% brightness of the pixel edge occurs [26,30]. Photographic images of the pixel blur are presented in Fig. 10(a). To compare our NIAs with conventional scattering film, we prepared an additional BEOLED including MLAs (diameter: 80 µm) and referred to this BEOLED as BEOLED D’’. Clear distinction via visual inspection of the pixel boundaries in BEOLED D was difficult. Thus, we converted the photographic images of light-emitting pixels to brightness information to define the pixel blur distance, as presented in Fig. 10(b). The pixel blur distances of BEOLEDs D, D’, and D’’, each with a pixel area of 4 mm2 [see Fig. 10(c)], were 160, 191, and 594 µm, respectively. These results revealed that, compared with conventional MLAs, our NIAs were considerably more effective in suppressing the pixel blur because the NIAs are characterized by low hazy and nanosized scattering patterns.
Fig. 10. (a) Photographic image of light-emitting pixels, (b) brightness information conversion from photographic image of Fig. 10(a), (c) pixel blur distance profiles from the pixel edge position. Inset image of (c) is a microscopic image of MLAs.
4. Conclusions
We could effectively improve the angular chromaticity change of OLEDs with strong microcavity characteristics by optimizing the thickness of HTL and the anode. Moreover, efficiencies of the current and power were enhanced by 1.46 and 1.64 times, respectively, compared with those of the device prior to thickness optimization of the HTL and anode. Furthermore, we suppressed the viewing angle dependence of OLEDs by applying our NIAs, although the OLEDs were optimized by controlling the thickness. More importantly, because our NIAs were prepared via thermal evaporation, they can provide excellent solutions to the unstable angular color shift of various large-sized display applications (e.g., TVs and signage). These NIAs can be selectively formed on each pixel of blue, green, red, and blue by using a fine metal mask, and hence, the viewing angle dependence can be quite accurately and effectively controlled.
Funding
Ministry of Trade, Industry and Energy (20010443, 10079974).
Acknowledgments
This work was supported by the Technology Innovation Program (20010443, Development of mass spectrometry for OLED structure and stability analysis) funded by the Ministry of Trade, Industry & Energy (MOTIE, Korea). Also, we were supported by the MOTIE [Ministry of Trade, Industry & Energy (10079974)] of ‘Development of core technologies on materials, devices, and processes for TFT backplane and light emitting frontplane with enhanced stretchability above 20%, with application to stretchable display’.
Disclosures
The authors declare no conflicts of interest.
See Supplement 1 for supporting content.
References
1. C. W. Tang and S. A. Vanslyke, “Organic electroluminescent diodes,” Appl. Phys. Lett. 51(12), 913–915 (1987). [CrossRef]
2. J. A. E. Wasey, A. Safonov, I. D. W. Samuel, and W. L. Barnes, “Effects of dipole orientation and birefringence on the optical emission from thin films,” Opt. Commun. 183(1-4), 109–121 (2000). [CrossRef]
3. Q. Zhang, B. Li, S. Huang, H. Nomura, H. Tanaka, and C. Adachi, “Efficient blue organic light-emitting diodes employing thermally activated delayed fluorescence,” Nat. Photonics 8(4), 326–332 (2014). [CrossRef]
4. Y. Im and J. Y. Lee, “Recent progress of green thermally activated delayed fluorescent emitters,” J. Inf. Disp. 18(3), 101–117 (2017). [CrossRef]
5. H. S. Kim, S. H. Lee, J. Y. Lee, S. Yoo, and M. C. Suh, “Mitigating the trade-off between triplet harvesting and roll-off by opening a Dexter-type channel in OLEDs,” J. Phys. Chem. C 123(30), 18283–18293 (2019). [CrossRef]
6. S.-R. Park, D. A. Ahn, K.-H. Seol, S. Yang, J. W. Kim, R. Lampande, H. Nam, and M. C. Suh, “Organic interface engineering for stable green PHOLEDs through an ultrathin interface tunneling layer,” Adv. Mater. Interfaces 5(21), 1801034 (2018). [CrossRef]
7. T. Furukawa, H. Nakanotani, M. Inoue, and C. Adachi, “Dual enhancement of electroluminescence efficiency and operational stability by rapid upconversion of triplet excitons in OLEDs,” Sci. Rep. 5(1), 8429 (2015). [CrossRef]
8. H. J. Jang, J. Y. Lee, J. Kim, J. Kwak, and J.-H. Park, “Progress of display performances: AR, VR, QLED, and OLED,” J. Inf. Disp. 21(1), 1–9 (2020). [CrossRef]
9. P. E. Burrows, V. Bulovic, S. R. Forrest, L. S. Sapochak, D. M. McCarty, and M. E. Thompson, “Reliability and degradation of organic light emitting devices,” Appl. Phys. Lett. 65(23), 2922–2924 (1994). [CrossRef]
10. S. Winter, S. Reineke, K. Walzer, and K. Leo, “Photoluminescence degradation of blue OLED emitters,” Proc. SPIE 6999, 69992N (2008). [CrossRef]
11. H. Aziz, Z. D. Popovic, N.-X. Hu, A.-M. Hor, and G. Xu, “Degradation mechanism of small molecule-based organic light-emitting devices,” Science 283(5409), 1900–1902 (1999). [CrossRef]
12. R. Meerheim, M. Furno, S. Hofmann, B. Lüssem, and K. Leo, “Quantification of energy loss mechanisms in organic light-emitting diodes,” Appl. Phys. Lett. 97(25), 253305 (2010). [CrossRef]
13. C. Fuchs, P.-A. Will, M. Wieczorek, M. C. Gather, S. Hofmann, S. Reineke, K. Leo, and R. Scholz, “Enhanced light emission from top-emitting organic light-emitting diodes by optimizing surface plasmon polariton losses,” Phys. Rev. B 92(24), 245306 (2015). [CrossRef]
14. S. Nowy, B. C. Krummacher, J. Frischeisen, N. A. Reinke, and W. Brütting, “Light extraction and optical loss mechanisms in organic light-emitting diodes: Influence of the emitter quantum efficiency,” J. Appl. Phys. 104(12), 123109 (2008). [CrossRef]
15. S. Möller and S. R. Forrest, “Improved light out-coupling in organic light emitting diodes employing ordered microlens arrays,” J. Appl. Phys. 91(5), 3324–3327 (2002). [CrossRef]
16. M.-K. Wei and I.-L. Su, “Method to evaluate the enhancement of luminance efficiency in planar OLED light emitting devices for microlens array,” Opt. Express 12(23), 5777–5782 (2004). [CrossRef]
17. J.-H. Han, J. Moon, D.-H. Cho, J.-W. Shin, H. Y. Chu, J.-I. Lee, N. S. Cho, and J. Lee, “Luminescence enhancement of OLED lighting panels using a microlens array film,” J. Inf. Disp. 19(4), 179–184 (2018). [CrossRef]
18. S. Thomschke, B. Reineke, K. Lüssem, and Leo, “Highly efficient white top-emitting organic light-emitting diodes comprising laminated microlens films,” Nano Lett. 12(1), 424–428 (2012). [CrossRef]
19. J. Y. Kim and K. C. Choi, “Improvement in outcoupling efficiency and image blur of organic light-emitting diodes by using imprinted microlens arrays,” J. Disp. Technol. 7(7), 377–381 (2011). [CrossRef]
20. J.-B. Kim, J.-H. Lee, C.-K. Moon, K.-H. Kim, and J.-J. Kim, “Highly enhanced light extraction from organic light emitting diodes with little image blurring and good color stability,” Org. Electron. 17, 115–120 (2015). [CrossRef]
21. H. Riel, S. Karg, T. Beierlein, B. Ruhstaller, and W. Rieß, “Phosphorescent top-emitting organic light-emitting devices with improved light outcoupling,” Appl. Phys. Lett. 82(3), 466–468 (2003). [CrossRef]
22. T. Shiga, H. Fujikawa, and Y. Taga, “Design of multiwavelength resonant cavities for white organic light-emitting diodes,” J. Appl. Phys. 93(1), 19–22 (2003). [CrossRef]
23. S. Hofmann, M. Thomschke, B. Lüssem, and K. Leo, “Top-emitting organic light-emitting diodes,” Opt. Express 19(S6), A1250–A1264 (2011). [CrossRef]
24. B. Pyo, C. W. Joo, H. S. Kim, B.-H. Kwon, J.-I. Lee, J. Lee, and M. C. Suh, “A nanoporous polymer film as a diffuser as well as a light extraction component for top emitting organic light emitting diodes with a strong microcavity structure,” Nanoscale 8(16), 8575–8582 (2016). [CrossRef]
25. H. S. Kim, C. W. Joo, B. Pyo, J. Lee, and M. C. Suh, “Improvement of viewing angle dependence of the white organic light emitting diodes with tandem structure by introduction of nanoporous polymer films,” Org. Electron. 40, 88–96 (2017). [CrossRef]
26. N. S. Kim, W. Y. Lee, S. H. Jung, and M. C. Suh, “Light modulation of top emission organic light emitting diodes showing strong microcavity effect by applying multilayered scattering film,” Opt. Express 26(2), 1185–1198 (2018). [CrossRef]
27. N. S. Kim, S. H. Jung, and M. C. Suh, “The suppression of viewing angle dependence of top emission organic light emitting diodes having strong microcavity characteristics by applying concave patterned anode,” J. Soc. Inf. Disp. 26(2), 79–84 (2018). [CrossRef]
28. M. Green and S. Tsuchiya, “Mesoscopic hemisphere arrays for use as resist in solid state structure fabrication,” J. Vac. Sci. Technol., B: Microelectron. Process. Phenom. 17(5), 2074–2083 (1999). [CrossRef]
29. S. Tsuchiay, M. Green, and R. R. A. Syms, “Structural fabrication using cesium chloride island arrays as resist in fluorocarbon reactive ion etching plasma,” Electrochem. Solid-State Lett. 3(1), 44–46 (1999). [CrossRef]
30. N. S. Kim, W. Y. Lee, B. Pyo, and M. C. Suh, “Suppression of viewing angle dependence of organic light emitting diodes by introduction of circular polarizer with nanoporous polymer film,” Org. Electron. 44, 232–237 (2017). [CrossRef]
