Top-emission organic light emitting diodes with lower viewing angle dependence

Synthetic Metals 189 (2014) 57–62

Contents lists available at ScienceDirect

Synthetic Metals journal homepage: www.elsevier.com/locate/synmet

Top-emission organic light emitting diodes with lower viewing angle dependence Byung Wan Lim, Hyeon Soo Jeon, Min Chul Suh ∗ Department of Information Display and Advanced Display Research Center, Kyung Hee University, Dongdaemoon-Gu, Seoul 130-701, Republic of Korea

a r t i c l e

i n f o

Article history: Received 17 October 2013 Accepted 23 December 2013 Available online 21 January 2014 Keywords: Top emission View angle dependency Color saturation Microcavity effect Scattering media

a b s t r a c t We obtained nanoporous polymer film which has 20–30% optical haze to suppress the viewing angle dependency of a top emission organic light emitting diode (TOLED) with strong microcavity effect. We controlled the sizes and density of nanopores by changing the spin coating condition such as spinning duration and spin rate in highly humid atmosphere up to 90% relative humidity condition. The resultant nanoporous polymer film has effectively reduced viewing angle dependency of microcavity TOLEDs without any serious decrease in total intensity of out-coupled light observed from the integrating sphere although the luminance observed from the front side was diminished slightly. Despite its negative effect on the efficiency toward the front direction, we could use those nanoporous polymer films as scattering media or diffuser layer on top of the encapsulation glass because it could change the strongly directed emission toward desired emission with Lambertian distribution. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Active matrix organic light emitting diode (AMOLED) display is becoming a main display mode for mobile devices and continues to make progress toward low-power consumption television (TV) applications with high power efficiency, wide viewing angle, wide color gamut, and high-speed video rate [1–6]. However, such achievements have still been faced a lot of challenges of active matrix liquid crystal display (AMLCD) because the technology for AMLCD itself is still growing day by day. Thus, AMOLED should be differentiated by completely different properties in the competition with AMLCD. In that respect, an enormous attention is drawn to a next generation display with flexibility, so-called, flexible AMOLED because of its unquestionably superior bending property to that of thin film transistor-liquid crystal display (TFT-LCD). But, to realize a perfect image quality during bending or flexing, such a display should also have high resolution, high contrast ratio, wide color gamut, and desirable screen viewing angle, etc. Especially, the viewing angle property would be much more important in the flexible AMOLED display because the customer may want to watch the images without any distortion during flexing [6–10]. However, this is not easy issue because most of the current mobile AMOLEDs are prepared by top emission OLED (TOLED) structure with very strong microcavity. From this approach, we could realize more vivid and beautiful displays. In addition, this approach is indispensable

∗ Corresponding author. Tel.: +82 2 961 0694; fax: +82 2 968 6924. E-mail address: [email protected] (M.C. Suh). 0379-6779/$ – see front matter © 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.synthmet.2013.12.020

because the unit pixel is comprised of several transistors and a couple of capacitors and those components and additional metal wirings should be buried under the OLED to realize high resolution display [1–6,11,12]. The TOLED itself is generally constructed with two metallic electrodes, a highly reflective anode and a semi-transparent cathode, and organic layers sandwiched between them [11]. The electrodes are parallel to each other and form a Fabry–Perot resonator, which results in the strongly resonant microcavity effect. Unfortunately, strong microcavity effect of the TOLEDs results in narrowed emission bandwidth with strong angle-dependent emission spectra, which is an obstacle to achieving display application such as large area television sets or flexible AMOLEDs for the future [5–12]. Thus, an angle-independent broadband emission is needed. Employing a low-reflectivity anode to alleviate the undesirable microcavity effect on the viewing angle characteristics has been proposed and demonstrated. Introducing a refractive-index-matching layer on top of the semitransparent cathode has also been demonstrated with optimized viewing characteristics by decreasing the reflectance of the cathode [10–12]. However, all such demonstrations are unable to eliminate the microcavity effect completely, but only minimize the microcavity effect by careful design of the metallic electrodes or its capping layer [13–18]. Despite its negative effect on the viewing angle characteristics, the microcavity resonance gives a positive effect on the emission efficiency through enhancing the spontaneous emission by continuous resonance [16–19]. In this work, we have introduced a nanoporous polymer films as scattering media which could reduce a viewing angle dependency.

58

B.W. Lim et al. / Synthetic Metals 189 (2014) 57–62

2. Experimental 2.1. Fabrication of nanoporous polymer film For the preparation of the nanoporous polymer film, cellulose acetate butyrate (CAB, butyryl content: 50–54%) was purchased from ACROS and processed to prepare a diffuser layer. A 0.8 g of CAB was dissolved in THF (10 ml) and mixed homogeneously by vortex mixer for 10 min. The resultant solution was deposited on the encapsulation glass by using a spin coating method. During spin coating process, the water vapor was directly supplied to spin coater by humidifier. The water vapor was supplied consistently to keep 90% relative humidity (RH) condition. After spinning and evaporation of all the residual solvents, the encapsulation glass with CAB film was completely dried for 10 min at room temperature. After finishing all the drying process, we investigated the morphology of polymer film with numbers of nanopores by using optical microscope (OLYMPUS, MX51) and FE-SEM technology (HITACHI, S-4700) [20]. The hazes of encapsulation glass with nanoporous polymer film were measured using a NDH-5000 haze meter from Nippon Denshoku Industries. 2.2. Device fabrication and characterization 2.2.1. Device fabrication Clean glass substrates precoated with ITO/Ag/ITO layers were used to investigate top emission properties. Line patterns of anode materials were formed on glass by photolithography process. Bank layer was also formed on the anode and glass substrate by photolithography process to define the pixel aperture area by using photoresist. The glass substrates with anode as well as bank layer were cleaned by sonification in an isopropyl alcohol and acetone, rinsed in deionized water, and finally treated in a UV-ozone chamber. All organic materials were deposited by the vacuum evaporation technique under a pressure of ∼1 × 10−7 Torr. The ˚ Then, lithium deposition rate of organic layers was about 0.5 A/s. quinolate (LiQ) and magnesium (Mg): silver (Ag) (9:1) were successively deposited without breaking vacuum by the deposition rates ˚ of 0.15 and 2 A/s, respectively. 2.3. Device characterization The current density–voltage (J–V) and luminance–voltage (L–V) data of OLEDs were measured by Keithley SMU 2635A and Minolta CS-100A, respectively. Electroluminescence (EL) spectra and CIE coordinate were obtained using a Minolta CS-2000A spectroradiometer. The OLED area was 4 mm2 for all the samples studied in this work. An integrating cube (IC2, StellarNet. Inc.) was connected to the spectrometer when integrated spectra for all emission angles were measured. 3. Results and discussion 3.1. Diffuser layer with nanoporous surface morphology Nanoporous polymer films have attracted much attention due to their usefulness as supporting media in tissue engineering, membranes in separation process, templates for inorganic growth, dielectric materials for electronics devices, and optical materials [20–22]. We have chosen nanoporous polymer film as a diffuser (or scattering media) to reduce the viewing angle dependency of TOLED with strong microcavity effect. The glass substrates with various hazes were shown in Fig. 1. We could prepare the various glass substrates with 20–40% of hazes by simple spin coating of CAB solution at 1000–2000 rpm for 30 s in

Fig. 1. Pictures of hazy glasses and SEM images of corresponding nanoporous polymer film of each hazy glass. The glasses with nanoporous polymer film show haze of (a) 40%, (b) 30%, and (c) 20%, respectively. (d) image of bare glass as a reference. Area of glass substrates is 25 mm × 25 mm.

the 90% RH condition. The haze could be controlled by changing of the spinning rate or duration. The CAB films formed by this process were shown in Fig. 1, and they showed transmittance of 87%, 88%, and 90%, respectively, in the order from Fig. 1(a)–(c). However, we selected only 20 and 30% haze conditions to fabricate scattering TOLEDs because immoderate haze could blur pixels and ruin the resolution of the real devices although all the CAB films we prepared show reasonably high transmittance levels. The preparation of glass substrate with haze value less than 10% was difficult due to a lack of process repeatability. Fig. 2 shows the SEM (scanning electron microscope) images of top surfaces and cross-sections of CAB films prepared by spin-coating under highly humid environment (RH = 90%) as aforementioned. We formed those layers on top of the encapsulation glass to investigate the possibility to convert the strongly directed emission into desired emission like Lambertian distribution. Fig. 2(a) shows the randomly formed nanopores on the surface of CAB film. The diameters of those nanopores were easily controlled from 200 to 400 nm while the distances between those nanopores could also be controlled from 50 to 400 nm for the desirable interference of the visible light emitted from the OLED devices [Fig. 2(a) and (b)]. Very interestingly, those nanopores were formed only on the surface of the polymer film as shown in Fig. 2(c). The film formed in this condition showed haze and transmittance up to ∼40% and 87%, respectively. 3.2. Device characteristics Fig. 3 shows the perspective images of the TOLED devices fabricated in this study. We used Indium tin oxide (ITO)/silver (Ag)/ITO as an anode, 2,2 ,7,7 -tetrakis(diphenylamino)9,9 -spirobifluorene (spiro-TAD) as hole transport layer,

B.W. Lim et al. / Synthetic Metals 189 (2014) 57–62

59

Fig. 4. Energy band diagram of fabricated devices.

Fig. 2. SEM images of nanoporous polymer films on glass substrates prepared by spin-coating method. (a) Surface morphology of the film, (b) enlarged image of the nanoporous polymer surface, and (c) cross-section of the nanoporous polymer film.

4,4 ,4 -tris(carbazol-9-yl)-triphenylamine (TCTA) as electron blocking layer, beryllium bis(2-(2 -hydroxyphenyl)pyridine (Bepp2 ) and fac-tris(2-phenylpyridinato)iridium (III) [Ir(ppy)3 ] as host and dopant materials for emission layer, 1,3,5-tri(p-pyrid-3yl-phenyl)benzene (TpPyPB) as electron trasport layer as well as hole blocking layer, LiQ as electron injection layer (EIL), and Mg:Ag as cathode. The detailed configuration of TOLEDs prepared for this study was ITO/Ag/ITO/Spiro-TAD (25 nm)/TCTA (5 nm)/Bepp2 : Ir(ppy)3 (3%, 20 nm)/TpPyPB (25 nm)/LiQ (1.5 nm)/Mg:Ag (10%, 14 nm). We subdivided the devices into Devices A, B, and C according to the existence of diffuser film and haze values of encapsulation glass as shown in Fig. 3. For example, Device A was prepared by using encapsulation glass without any porous film to utilize as a reference TOLED sample. Devices B and C were prepared by using encapsulation glasses with diffuser film which show haze of 20 and 30%, respectively. The energy band diagram of the materials used in those OLED devices was summarized in Fig. 4. Fig. 5(a) shows the current density–voltage (J–V) and luminance–voltage (L–V) characteristics of fabricated green OLEDs. At a given constant voltage of 4.0 V, current density values of 20.1, 22.5, and 22.5 mA/cm2 were observed in the fabricated Devices A, B, and C, respectively. The driving voltage to reach 1000 cd/m2 was 3.2, 3.3, and 3.3 V for the Devices A, B, and C, respectively. The sample without diffuser and with diffuser showed current density with difference within 10%, which means that the significantly decreased current efficiency when we applied haze encapsulation glasses could be originated from the decreased luminance only at the front direction. Indeed, at a given constant luminance of 1000 cd/m2 , the current and power efficiencies were 45.3 cd/A, 43.5 lm/W for the Device B, and 41.0 cd/A, 39.2 lm/W which were significantly reduced values from 55.7 cd/A, 53.4 lm/W for the Device A, respectively (see

Fig. 3. Device structures of TOLEDs with or without nanoporous polymer film.

60

B.W. Lim et al. / Synthetic Metals 189 (2014) 57–62

Table 1 The performances of TOLEDs.a Items

Device A (without diffuser)

Device B (with diffuser)

Device C (with diffuser)

Turn-on voltage (1 cd/m2 ) Operating voltage (1000 cd/m2 ) Efficiency (1000 cd/m2 ) Efficiency (Max) E.Q.E. (1000 cd/m2 )

2.5 V 3.2 V 55.7 cd/A 53.4 lm/W 55.5 cd/A 58.4 lm/W 21.8%

2.5 V 3.3 V 45.3 cd/A 43.5 lm/W 45.9 cd/A 48.3 lm/W 16.2%

2.5 V 3.3 V 41.0 cd/A 39.2 lm/W 41.6 cd/A 43.5 lm/W 15.2%

a

The device performances were measured only in the forward direction.

Fig. 5. Device characteristics of TOLED without nanoporous polymer film as diffuser (Device A: circle), with the 20% haze film (Device B: triangle) and with the 30% haze film (Device C: square); (a) current density (J)–voltage (V) characteristics (solid symbol) and luminance (L)–voltage (V) (open symbol) characteristics, (b) current efficiency–luminance characteristics (solid symbol) and power efficiency–luminance characteristics (open symbol), and (c) The electroluminescent images of the devices without/with diffuser (haze 20%, 30%) (left/center/right).

Fig. 6. (a) Electroluminescence spectra of the fabricated TOLEDs put in and measured from when operating at 2.5 mA/cm2 ; both spectra were normalized by maximum intensity of spectrum of Device A, (b) schematic illustration of TOLEDs showing different pathway of photons through polymer films with or without nanopores, (c) measured 3D far-field distribution of with or without nanopores (left/right, dimensions: 2 ␮m × 2 ␮m).

Fig. 7. Normalized emission intensities with viewing angle of fabricated devices; Device A (solid circle), Device B (solid triangle), Device C (solid square) and Lambertian emission pattern (dashed line). (b) electroluminescence spectra with relative intensities at viewing angles for Devices A and C (Device A: solid, Device C: open circle).

B.W. Lim et al. / Synthetic Metals 189 (2014) 57–62

also Fig. 5(b) and Table 1). The current and power efficiency were decreased quite a lot through a scattering effect of the nanoporous film on the encapsulation glass because the hazes of the film utilized in this study were 20 and 30%. In addition, the luminance was only detected at the front direction of the test device by using spectrophotometer (CS100). The maximum current and power efficiencies were 56.1 cd/A, 58.4 lm/W for the Device A, 45.9 cd/A, 48.3 lm/W for the Device B, and 41.6 cd/A, 43.5 lm/W for the Device C, respectively. In Fig. 5(c), the OLED devices with diffuser film seem to be more diffusive which means that emission from the Device B, and C might be more desirable because they give less directed emission. To identify the origin of the light loss in the front direction, we investigated the total emission of the Devices A and C by using the integrating sphere with CCD spectrometer to measure the spectra of the collected light beams from 450 to 650 nm. And, we found that the relative intensity collected from the spectra obtained from Devices A and C at 2.5 mA/cm2 were almost same without any spectral change as shown in Fig. 6(a). This means that the diffuser film just spread the direction of the emission laterally as shown in Fig. 6(b) without any serious loss of the total light emission. To estimate the scattering effect of those kinds of nanoporous polymer films, we proceeded a finite difference time domain (FDTD) simulation and obtained the power distribution of the light near the surface of encapsulation glass without and with nanoporous polymer films as shown in Fig. 6(c). To simulate the power distribution of the light scattered on the polymer surfaces, we assumed the nanopores which have diameters about 400 nm, and depth about 200 nm. In addition, we assumed the distance between such nanopores about 50–400 nm. The total area we simulated was 2 ␮m × 2 ␮m. From FDTD simulation based on such an assumption, we could visualize a power distribution of the light after interference at the surface without or with randomly dispersed nanopores. As a result, we found that the nanoporous film gives much narrower power distribution than the film without any nanopores. This means that scattering effect could be larger in the device with polymer film. From this interference, the incorporation of diffuser could modify the emission pattern of the TOLED (Devices B and C), making it somehow closer to the desirable Lambertian distribution although it is still far from the reference curve as shown in Fig. 7(a). Fig. 7(b) shows the electroluminescence (EL) spectra which are normalized by the maximum intensities of front direction (0◦ ) without diffuser film, and with 30% haze diffuser film, respectively. The solid line gives the information of the spectral change of Device A (without diffuser film) upon variation of the viewing angle while the line with open circle shows that of Device C (with 30% haze film). We found that the Device A shows the higher viewing angle dependency because the relative intensity decreased more seriously when the viewing angle reaches up to 60◦ . This means that the diffuser film prepared by CAB gives more or less desirable viewing angle property. To investigate the viewing angle dependency more rigorously, we separated the EL spectra of each device. To estimate the viewing angle dependency, we overlaid and normalized the EL spectra according to the viewing angle as shown in Fig. 8(a)–(c). Very interestingly, the TOLED without diffuser film showed serious hypsochromic shift [max : 511 nm (0◦ ) → 507 nm (60◦ )] while the TOLED with diffuser (haze: 30%) showed much less viewing angle dependency [max : 509 nm (0◦ ) → 508 nm (60◦ )] as shown in Fig. 8(a)–(c). A significant hypsochromic shift of the Device A without diffuser film as the viewing angles increase from 0◦ to 60◦ off the surface is mainly due to the strong microcavity effects which are originated from the phase change upon variation of the viewing angle as following the general formula shown below [23–27]:

61

Fig. 8. Normalized electroluminescence spectra in accordance with viewing angle of the TOLEDs (a) Device A, (b) Device B, and (c) Device C. The color shift with viewing angle was significant for device without diffuser (max of each device was given in bracket).

() =

4(˙ni di ) cos() (() − 2m)

(1)

where ni di is the cavity length, ni and di are the refractive index and thickness of ith organic layer, () is total phase shift caused by reflections at two electrodes, and m is the cavity mode number. We utilized m = 0 for this study. As a result, Device A showed larger color coordinate [CIE (1931)] change from (0.193, 0.683) to (0.190, 0.640) when the viewing angle increases from 0◦ to 60◦ while Device C showed negligible color change from (0.186, 0.667) to (0.187, 0.657). This is very interesting because the emission pattern of the TOLED (Devices B and C) was still far from desirable Lambertian distribution in the polar plot shown in Fig. 7(a). Especially, with 30% haze diffuser film [Fig. 8(c)],

62

B.W. Lim et al. / Synthetic Metals 189 (2014) 57–62

the change of EL spectra at different viewing angle is almost negligible. Moreover, the trend of such a viewing angle dependency after introduction of diffuser film is almost same as that previously observed by Liu et al. [7]. Nevertheless, the nanoporous polymer film which has 20–30% optical haze could cause a pixel blur and ruin the resolution of the device. Thus, the preparative condition should be optimized to realize a nanoporous film with optical haze about 10% to minimize such kinds of side effect. 4. Conclusion We obtained nanoporous polymer film which has 20–30% optical haze as a diffuser layer of TOLED with strong microcavity effect. The resultant nanoporous polymer film could effectively reduce a viewing angle dependency of microcavity TOLEDs without any diffuser film although the devices with those haze films decrease the intensity of luminance efficiency at the front direction. Despite its negative effect on the efficiency, we found that the introduction of a nanoporous polymer films as scattering media on top of the encapsulation glass could reduce a viewing angle dependency. Acknowledgements This work was supported by the Industrial Strategic Technology Development Program (No. 10041062) funded by the Ministry of Knowledge Economy (MKE, Korea) and the National Research Foundation of Korea (NRF) (Grant No. NRF-2011-0006847). This work was also supported by the Human Resources Development Program (No. 20134010200490) of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea Government Ministry of Trade, Industry and Energy. The authors are grateful to Mr. Kyung Nam Kang and Prof. Jungho Kim for helpful discussion and collaboration to investigate the viewing angle dependency of the devices.

References [1] J. Yamada, T. Hirano, Y. Iwase, Y. Sato, R. Nakagawa, M. Sekiya, T. Sasaoka, T. Urabe, T. Ishibashi, Jpn. J. Appl. Phys. 45 (2006) 4392. [2] J. Wang, Z. Wu, J. Lumin. 131 (2011) 2419. [3] M.S. Weaver, T.X. Zhou, M. Rothman, R.C. Kwong, M. Hack, J.J. Brown, M.-H. Lu, Appl. Phys. Lett. 81 (2002) 3921. [4] X. Liu, M.A. Khan, W.Q. Zhu, X.Y. Jiang, Z.L. Zhang, S.H. Xu, J. Cao, Curr. Appl. Phys. 7 (2007) 300. [5] C.-J. Yang, S.-H. Liu, H.-H. Hsieh, C.-C. Liu, T.-Y. Cho, C.-C. Wu, Appl. Phys. Lett. 91 (2007) 253508. [6] H. Peng, J. Sun, X. Zhu, X. Yu, M. Wong, H.-S. Kwok, Appl. Phys. Lett. 88 (2006) 073517. [7] C.-C. Liu, S.-H. Liu, K.-C. Tien, M.-H. Hsu, H.-W. Chang, C.-K. Chang, C.-J. Yang, C.-C. Wu, Appl. Phys. Lett. 94 (2009) 103302. [8] M. Thomschke, P. Freitag, M. Furno, B. Lüssem, K. Leo, S. Hofmann, Appl. Phys. Lett. 97 (2010) 253308. [9] C.-L. Lin, P.-Y. Hsieh, H.-H. Chiang, C.-C. Wu, Appl. Phys. Lett. 84 (2004) 3966. [10] Q. Wang, Z. Deng, D. Ma, Appl. Phys. Lett. 94 (2009) 233306. [11] S. Chen, H.-S. Kwok, Org. Electron. 12 (2011) 2065. [12] S.-F. Hsu, C.-C. Lee, S.-W. Hwang, C.H. Chen, Appl. Phys. Lett. 86 (2005) 253508. [13] A. Dodabalapur, L.J. Rothberg, R.H. Jordan, T.M. Miller, R.E. Slusher, J.M. Phillipse, J. Appl. Phys. 80 (1996) 6954. [14] G. Xie, Z. Zhang, Q. Xue, S. Zhang, L. Zhao, Y. Luo, P. Chen, B. Quan, Y. Zhao, S. Liu, Org. Electron. 11 (2010) 2055. [15] C.-W. Chen, P.-Y. Hsieh, H.-H. Chiang, C.-L. Lin, H.-M. Wu, C.-C. Wu, Appl. Phys. Lett. 83 (2003) 5127. [16] Y.-F. Liu, J. Feng, Y.-G. Bi, J.-F. Song, Y. Jin, Y. Bai, Q.-D. Chen, H.-B. Sun, Opt. Lett. 37 (2012) 124. [17] F. Ma, C. Zhang, H. Li, L. Wang, X. Liu, Jpn. J. Appl. Phys. 45 (2006) 9224. [18] M. Thomschke, R. Nitsche, M. Furno, K. Leo, Appl. Phys. Lett. 94 (2009) 083303. [19] S. Ming, G. Xu, F.Q. -Li, C.S. -Fen, H. Wei, Chinese Phys. B 21 (2012) 108507. [20] M.S. Park, W. Joo, J.K. Kim, Langmuir 22 (2006) 4594. [21] N.R. Cameron, Polymer 46 (2005) 1439. [22] M.S. Park, Y.M. Lee, J.K. Kim, Chem. Mater. 17 (2005) 3944. [23] Y.-F. Liu, J. Feng, D. Yin, H.-F. Cui, X.-L. Zhang, Y.-G. Bi, D.-D. Zhang, L.-S. Liu, A.-W. Li, J.-F. Song, Q.-D. Chen, H.-B. Sun, Org. Electron. 14 (2013) 1597. [24] C.-H. Chang, H.-C. Cheng, Y.-J. Lu, K.-C. Tien, H.-W. Lin, C.-L. Lin, C.-J. Yang, C.-C. Wu, Org. Electron. 11 (2010) 247. [25] C.Y. Ho, C.H. Choy, Opt. Express 15 (2007) 13288. [26] S.M. Jeong, Y. Takanishi, K. Ishikawa, S. Nishimura, G. Suzaki, H. Takezoe, Opt. Commun. 273 (2007) 167. [27] C.-L. Lin, H.-C. Chang, K.-C. Tien, C.-C. Wu, Appl. Phys. Lett. 90 (2007) 071111.