Thin Solid Films 547 (2013) Contents lists available at ScienceDirect. Thin Solid Films. journal homepage:

Transcription

1 Thin Solid Films 547 (213) Contents lists available at ScienceDirect Thin Solid Films journal homepage: Al 2 O 3 /Ag/Al 2 O 3 multilayer thin film passivation prepared by plasma damage-free linear facing target sputtering for organic light emitting diodes Jin-A Jeong, Han-Ki Kim Department of Advanced Materials Engineering for Information and Electronics, Kyung Hee University, 1 Seocheon-dong, Yongin-si, Gyeonggi-do , South Korea article info abstract Available online 15 May 213 Keywords: Al 2 O 3 /Ag/Al 2 O 3 Multilayer passivation Plasma damage-free LFTS OLED Lifetime Al 2 O 3 /Ag/Al 2 O 3 multilayer passivation prepared by plasma damage-free linear facing target sputtering (LFTS) was investigated as a function of inserted Ag thickness. Using antireflection effect of the Ag layer that is sandwiched between dielectric Al 2 O 3 layers, we can obtain a transparent Al 2 O 3 /Ag/Al 2 O 3 multilayer passivation for organic light emitting diodes (OLEDs). It was found that insertion of the Ag layer with optimized thickness between Al 2 O 3 layers lead to improvement of the optical transparency and water vapor transmission rate of the Al 2 O 3 /Ag/Al 2 O 3 multilayer. In addition, current density voltage luminescence of an OLED passivated with Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer was similar to that of an OLED with nonpassivated sample, indicating that the performance of an OLED is not affected by high-density plasma during the LFTS process. Moreover, the lifetime to half initial luminance of an OLED passivated with Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer was longer than that of a nonpassivated sample. 213 Elsevier B.V. All rights reserved. 1. Introduction To fabricate high-performance organic light emitting diodes (OLEDs), it is essential to acquire high-performance encapsulation because both lifetime and performance of the OLEDs are critically affected by quality of the encapsulation [1]. For this reasons, several types of encapsulation techniques have been extensively explored to improve long-term stability of OLEDs or flexible OLEDs [1 3]. In particular, thin film passivation has been considered as the most desirable encapsulation for OLEDs, due to its simplicity, thin thickness and flexibility. Although various SiN x,sio x,sio x N y,alo x,andal 2 O 3 :N films have been investigated for transparent thin film passivation for OLEDs, a single inorganic thin film is not sufficiently dense to protect flexible optoelectronic devices from permeation by moisture and oxygen [4 8]. Thus, multilayer passivation, such as Barix coating, which is composed of alternating organic inorganic multilayers, has been proposed as a means to achieve ultra high barrier properties for OLEDs or flexible OLEDs [1,9,1].However, Barix coating still has not been employed in mass production of OLEDs due to its complicate process and long-process time. In addition, most reported deposition techniques for the inorganic films are based on a plasma process which is harmful to organic materials. Therefore, it is imperative to develop simple multilayer passivation prepared by plasma damage free process to substitute Barix coating for high performance OLEDs and flexible OLEDs. However, oxide/metal/oxide (OMO) multilayer has not been investigated for thin film passivation even though it has various advantages such as high transparency and simple process. Corresponding author. address: imdlhkkim@khu.ac.kr (H.-K. Kim). In this work, we report on the characteristics of transparent Al 2 O 3 / Ag/Al 2 O 3 multilayer passivation prepared by a linear facing target sputtering (LFTS) as a function of inserted Ag layer thickness. Due to antireflection effect of the Ag and high density of Al 2 O 3 layer, multilayer showed a high transparency and low water vapor transmission rate (WVTR) despite the very small thickness of the multilayer (~15 nm). Furthermore, the current voltage luminescence (J-V-L) characteristics of the OLED passivated with the Al 2 O 3 /Ag/Al 2 O 3 layer were identical to a nonpassivated sample, indicating no damage by LFTS process. In addition, the lifetime of the OLED passivated by the Al 2 O 3 /Ag/Al 2 O 3 multilayer was compared with that of the nonpassivated OLED. 2. Experimental details Al 2 O 3 /Ag/Al 2 O 3 multilayer passivation was fabricated by a LFTS and thermal evaporation as shown in Fig. 1(a). The 7 nm-thick bottom Al 2 O 3 layer was reactively sputtered on a glass or polyethylene terephthalate (PET) substrate at a constant Ar/O 2 flow ratio of 8/2 sccm, a pulsed direct current (DC) power of 5 W and a working pressure of Pa. Due to effective confinement of high-density plasma between Al targets, we can deposit Al 2 O 3 layer on the substrate without substrate heating caused by bombardments of the energetic particles [11]. Fig. 1(b) shows schematics of linear facing sputter cathodes and picture of the effectively confined plasma between Al targets during LFTS process. Unidirectional magnetic fields between facing Al targets and negative voltage of cathodes lead to the confined plasma, which cannot affect the substrate like Fig. 1(b). In addition, the substrate position out of high-density plasma region makes it possible to deposit Al 2 O 3 layers without severe plasma damage. After the deposition of 4-69/$ see front matter 213 Elsevier B.V. All rights reserved.

2 64 J.-A. Jeong, H.-K. Kim / Thin Solid Films 547 (213) Fig. 1. (a) Schematics Al 2 O 3 /Ag/Al 2 O 3 multilayer deposition process using LFTS and thermal evaporator. (b) Schematic of linear facing target to confine high density plasma between Al targets and picture showing confined plasma during LFTS process. the bottom Al 2 O 3 layer, an Ag layer was thermally evaporated onto the bottom Al 2 O 3 layer at a constant deposition rate of.2 nm/section as a function of thickness (6 ~ 2 nm). Subsequently, the 7 nm-thick top Al 2 O 3 layer was sputtered onto the Ag layer under deposition conditions identical to those for the bottom Al 2 O 3 layer. The optical transmittance of Al 2 O 3 /Ag/Al 2 O 3 multilayer fabricated on PET substrate was measured in a wavelength range from 35 to 8 nm using a UV/visible spectrometer as a function of Ag thickness. The surface morphology of the Ag layer grown on the bottom Al 2 O 3 layer was analyzed as a function of Ag thickness using a field emission scanning electron microscope (FESEM) with operating voltage of 1 kv. To investigate structural properties of the Al 2 O 3 /Ag/Al 2 O 3 multilayer, synchrotron X-ray scattering examination was carried out. The synchrotron X-ray scattering measurements were carried out at the 5C2 GIST beamline at Pohang Light Source. Water vapor transmission rate (WVTR) value for the Al 2 O 3 /Ag/Al 2 O 3 multilayer grown on PET substrate (5 cm 5 cm) was determined at 38 ± 2 C, 1% relative humidity by MOCON (PERMATRAN-W 3/33 MA) for 72 h as shown in the inset picture of Fig. 4. Wehaveused MOCON apparatus, model PERMATRAN-W 3/33 MA from MOCON Inc. To compare the lifetime of OLED with Al 2 O 3 /Ag/Al 2 O 3 multilayer and nonpassivated sample (reference), we prepared conventional OLEDs. α-napthylphenlylbiphenyl (NPB) and tris-(8-hydroxyquinoline) aluminum (Alq 3 ) were employed as the hole transport layer and electron transport layer (Emission layer) layers, respectively. Finally, a 1 nm/ 1 nm-thick LiF/Al cathode was patterned on the Alq 3 layer using a shadow mask for an active device area 3 mm 2. After preparation of the OLED sample, it was passivated with an optimized Al 2 O 3 /Ag (1 nm)/ Al 2 O 3 multilayer. The current density voltage luminescence (J-V-L) characteristics of OLEDs were measured using a Photo Research PR-65 spectrophotometer driven by a programmable dc source. Driven at 2mA/cm 2, the device lifetime of OLED passivated with the Al 2 O 3 /Ag/ Al 2 O 3 multilayer at an initial luminance of 5 cd/m 2 was measured. 3. Results and discussion Fig. 2(a) shows the optical transmittance of the Al 2 O 3 /Ag/Al 2 O 3 multilayer grown on PET substrate as a function of the inserted Ag thickness. The multilayer sample with Ag thickness 6 and 8 nm shows a low optical transmittance due to light absorption and scattering of the Ag islands. However, the Al 2 O 3 /Ag/Al 2 O 3 multilayer with Ag thickness of 1 nm shows a significantly increase in the optical transmittance. The increased optical transmittance of the Al 2 O 3 /Ag/Al 2 O 3 multilayer could be attributed to the antireflection effect which has been observed in dielectric/metal/dielectric structure [12].As discussed by Berning and Tuner, symmetry designs consisting of dielectric layer stacks around a central layer of a Ag film can be produced with very high transmittance in visible spectral range [13]. However, further increases of Ag thickness result in decreases in optical transmittance of Al 2 O 3 /Ag/Al 2 O 3 multilayer passivation due to light absorption by Ag layer. Fig. 2(b) compared the simulated and measured transmittance of the Al 2 O 3 /Ag/Al 2 O 3 multilayer with inset picture showing high transparency of optimized sample (Ag ~ 1 nm). The optical simulation was carried out by means of an emissive thin film optics simulator (ETFOS). The ETFOS is optical engineering software containing a fast numerical algorithm for solving a dipole emission model. The simulation data showed that the insertion of the Ag layer between the Al 2 O 3 layers led to a linear decrease in the optical transmittance of the Al 2 O 3 /Ag/Al 2 O 3 multilayer. However, the measured transmittance of the Al 2 O 3 /Ag/ Al 2 O 3 multilayer shows abruptly increased at Ag thickness of 1 nm unlike simulation data. The increased optical transmittance in a specific Ag thickness region indicates that the optical transmittance of the Al 2 O 3 / Ag/Al 2 O 3 multilayer can be enhanced by controlling the Ag thickness, due to the antireflection and plasmon effect of the Ag layer [14,15]. Fig. 3(a) shows surface FESEM images of the Ag interlayer which is thermally evaporated on the bottom Al 2 O 3 layer as the thickness of Ag layer increases. It was shown that the morphology and shape of the Ag layer mainly depended on the Ag thickness. Both 6 and 8 nm thick Ag layer shows unconnected and randomly distributed islands. However, 1 and 12 nm thick Ag layer shows connected channel shape with some uncovered region. Due to agglomeration and connection of Ag layer, some area of bottom Al 2 O 3 layer is not covered by the Ag layer. Compared to 12 nm thick Ag layer, the 1 nm thick Ag layer shows more corrugated shape. Further increase of the Ag thickness leads to the complete cover of the bottom Al 2 O 3 layer. The transition of the Ag

3 J.-A. Jeong, H.-K. Kim / Thin Solid Films 547 (213) a 1 a Transmittance [%] 8 6 Al 2 O 3 /Ag/Al 2 O 3 4 Ag 6nm Ag 8nm Ag 1nm 2 Ag 12nm Ag 14nm Ag 2nm Wavelength [nm] b b Transmittance [%] Measured data Simulation data Intensity [counts/monitors] 2 nm 14 nm 12 nm 1 nm 8 nm 6 nm Ag thickness [nm] Fig. 2. (a) Optical transmittance spectra and (b) measured optical transmittance at 55 nm and simulation data of the Al 2 O 3 /Ag/Al 2 O 3 multilayer prepared on PET substrate with increasing Ag thickness. The inset picture shows the Al 2 O 3 /Ag (1 nm)/ Al 2 O 3 multilayer with the highest transparency. layer from island to completely covered film is similar to our previously reported works [14,15]. Considering barrier properties of the Al 2 O 3 /Ag/ Al 2 O 3 multilayer, it is desirable to insert thicker Ag layer between Al 2 O 3 layers because the thicker Ag film densely covered the bottom Al 2 O 3 layer as shown in Fig. 3(a). However, due to the transparency of the multilayer, optimization of Ag thickness is necessary in the Al 2 O 3 /Ag/ Al 2 O 3 multilayer. Fig. 3(b) shows the synchrotron X-ray scattering results of the Al 2 O 3 /Ag/Al 2 O 3 multilayer as a function of the Ag thickness. Regardless of the Ag thickness, all of the top and bottom Al 2 O 3 layers shows a very broad peak between Qz =1.5 and Qz = 2. indicating their amorphous structure. As previously reported, the effective confinement of the energetic particles between the facing Al targets enablesustodepositanal 2 O 3 film with a completely amorphous structure at room temperature [11]. However, the microstructure of the Ag layer prepared by thermal evaporation was dependent on its thickness. Up to a thickness of 12 nm, there is no peak related to crystalline Ag at Q z = This indicates that the structure of the islands and channel Ag layer is amorphous under the 12 nm thickness. A crystalline Ag peak of low intensity at Q z = 2.66 starts to appear an Ag thickness of 14 nm. Further increasing the Ag thickness leads to the clear appearance of a crystalline peak of fairly high intensity at Q z =2.66. Fig. 4 shows WVTR value of the Al 2 O 3 /Ag/Al 2 O 3 multilayer samples which were grown on the PET substrate as a function of Ag thickness Qz [A -1 ] Fig. 3. (a) Surface FESEM images of the Ag layer grown on the bottom Al 2 O 3 layer with increasing thickness. (b) Synchrotron X-ray scattering of the Al 2 O 3 /Ag/Al 2 O 3 multilayer as a function of the Ag thickness. shows the Al 2 O 3 /Ag/Al 2 O 3 multilayer sample after MOCON test. Due to the small size of Al 2 O 3 /Ag/Al 2 O 3 multilayer samples, WVTR value for all multilayer passivation were measured by packaging the samples like inset picture. Compared to the WVTR value (1.4 g/m 2 -day) of the single Al 2 O 3 layer, the WVTR value of the Al 2 O 3 /Ag/Al 2 O 3 multilayer is much lower. As the Ag thickness increases in the Al 2 O 3 /Ag/Al 2 O 3 multilayer, the WVTR value monotonically decreases because the thicker Ag layer can effectively prevent the intrusion of water vapor. However, effect of the Ag thickness on the WVRT value is fairly small due to main barrier properties are related to the density of the Al 2 O 3 film. Despite of the very small thickness of the Al 2 O 3 /Ag/Al 2 O 3 multilayer (~15 nm), the WVTR of the multilayer passivation shows the fairly low WVTR value, which is.11 g/m 2 -day for a Ag thickness of 2 nm. Considering transparency and WVTR value, we decided the optimized thickness of Ag layer as 1 nm. Although the WVTR value (~1 2 g/m 2 -day) of the Al 2 O 3 /Ag/Al 2 O 3 multilayer is higher than the previously reported WVTR ( g/m 2 -day) values of several thin film passivation [8 1], high transparency and very thin thickness could be merit of the Al 2 O 3 /Ag/Al 2 O 3 multilayer. Fig. 5 shows X-ray photoelectron spectroscopy (XPS) depth profiles results obtained from the optimized Al 2 O 3 /Ag(1 nm)/al 2 O 3 electrode. It was shown that the individual layers of top Al 2 O 3, Ag, and bottom Al 2 O 3 layer exist without interfacial reaction. The constant atomic percent of Al and O atoms in the bottom and top Al 2 O 3 layers shows

4 66 J.-A. Jeong, H.-K. Kim / Thin Solid Films 547 (213) Water Vapor Transmission Rate [g/day-m 2 ] Ag thickness [nm] (a) Current density [ma/cm 2 ] Luminance [cd/m 2 ] Al 2 O 3 /Ag/Al 2 O 3 passivation W/O Passivation Voltage [V] Voltage [V] Fig. 4. The water vapor transmission rate of the Al 2 O 3 /Ag/Al 2 O 3 multilayer on PET substrate as a function of Ag thickness with an inset picture of the Al 2 O 3 /Ag/Al 2 O 3 multilayer sample after MOCON test. that both bottom and top Al 2 O 3 layer were identically deposited on the glass substrate and Ag layer. However, the Ag layer shows asymmetrical feature due to rough and agglomerated surface morphology as shown in Fig. 3(a). Fig. 6(a) shows J-V-L characteristics of the OLED passivated with a Al 2 O 3 /Ag/Al 2 O 3 multilayer and the reference OLED (Nonpassivated sample). The J-V-L curve of the OLED passivated with a Al 2 O 3 /Ag/ Al 2 O 3 multilayer exhibits an similar turn on voltage, forward bias behavior and luminance to those of the reference OLED. These similar J-V-L characteristics of OLEDs passivated with a Al 2 O 3 /Ag/Al 2 O 3 multilayer indicates that the electrical and optical characteristics of the OLED are not critically affected by exposure to the high density plasma during the Al 2 O 3 deposition. Fig. 6(b) shows the lifetime for an OLED passivated with a Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer and a reference OLED. The operating lifetime of both the OLED passivated with the Al 2 O 3 /Ag/Al 2 O 3 multilayer and the reference OLED was measured at a dc-current drive of 2 ma/cm 2 with an initial luminance of 5 cd/m 2 at room temperature. The lifetime of the reference sample with an initial luminance of 5 cd/m 2 was approximately 4 h. The abrupt decrease in the luminance of the reference OLED is related to severe degradation of organic layers by direct intrusion of moisture and oxygen through the cathode layer. However, the lifetime of the OLED passivated with a Al 2 O 3 /Ag/Al 2 O 3 multilayer is longer than that of the reference sample. The lifetime of the passivated OLED is (b) Normalized EL intensity W/O Passivation Al 2 O 3 /Ag/Al 2 O 3 passivation Time [hour] Fig. 6. (a) J-V characteristics of OLED passivated with Al 2 O 3 /Ag/Al 2 O 3 multilayer and nonpassivated OLED (reference sample) with inset of L-V characteristics for passivated and nonpassivated samples. (b) Typical curves for normalized luminance versus operating time of OLEDs passivated with Al 2 O 3 /Ag/Al 2 O 3 multilayer and a reference OLED. approximately 14 h. It is noteworthy that the lifetime of the OLED could be prolonged by using a Al 2 O 3 /Ag/Al 2 O 3 multilayer unlike other reported thin film passivation procedures, which use a thick film. The prolonged lifetime of the OLED verifies that Al 2 O 3 /Ag/ Al 2 O 3 multilayer grown by LFTS was very a good passivation layer. Therefore, the identical J-V-L data and prolonged lifetime of the OLED passivated with Al 2 O 3 /Ag/Al 2 O 3 multilayer grown by LFTS suggests that the Al 2 O 3 /Ag/Al 2 O 3 multilayer is a promising passivation scheme for high performance OLED and flexible OLEDs. Intensity [arb.unit] 6x1 4 5x1 4 4x1 4 3x1 4 2x1 4 1x1 4 O Top Al 2 O 3 Ag Bottom Al 2 O 3 Ag Al Sputter Time [min] Fig. 5. XPS depth profile of Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer. C 4. Conclusion We reported on the characteristics of the antireflective Al 2 O 3 /Ag/ Al 2 O 3 multilayer passivation prepared by plasma damage-free LFTS. It was found that insertion of the Ag layer with optimized thickness between Al 2 O 3 layers lead to improvement of the optical transparency and WVTR of the Al 2 O 3 /Ag/Al 2 O 3 multilayer. However, effect of the Ag thickness on the WVRT value is fairly small due to main barrier properties are closely related to the density of the Al 2 O 3 film. Using antireflection effect of the Ag layer that is sandwiched between dielectric Al 2 O 3 layers, we can obtain a transparent Al 2 O 3 /Ag/Al 2 O 3 multilayer passivation for OLEDs. The current density-voltage-luminescence of an OLED passivated with Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer was similar to that of an OLED with nonpassivated sample, indicating that the performance of an OLED is not affected by high-density plasma during the LFTS process. Moreover, the lifetime to half initial luminance of an OLED passivated with Al 2 O 3 /Ag (1 nm)/al 2 O 3 multilayer was longer than that of a nonpassivated sample.

5 J.-A. Jeong, H.-K. Kim / Thin Solid Films 547 (213) Acknowledgements This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (212R1A1A241864). References [1] J. Lewis, Mater. today 9 (26) 38. [2] N. Kim, W.J. Potscavage, B. Domercq, B. Kippelen, S. Graham, Appl. Phys. Lett. 94 (29) [3] J. Lewis, M.S. Weaver, Quantum Electron. 1 (24) 45. [4] H. Lin, L. Xu, X. Chen, X. Wang, M. Sheng, F. Stubhan, K.-H. Merkel, J. Wilde, Thin Solid Films 333 (1998) 71. [5] H.-K. Kim, S.-W. Kim, D.-G. Kim, J.-W. Kang, M.S. Kim, W.J. Cho, Thin Solid Films 515 (27) [6] H.-K. Kim, J.-W. Kang, J.-J. Kim, M.-S. Yi, Appl. Phys. Lett. 9 (27) [7] T.N. Chen, D.S. Wuu, C.C. Wu, C.C. Chiang, Y.P. Chen, R.H. Horng, J. Electrochem. Soc. 153 (26) F244. [8] S.J. Yun, Y.-W. Ko, J.W. Lim, Appl. Phys. Lett. 85 (24) [9] M.S. Weaver, L.A. Michalski, K. Rajan, M.A. Rothman, J.A. Silvernail, P.E. Burrows, G.L. Graff, M.E. Gross, P.M. Martin, M. Hall, E. Mast, C. Bonham, W. Bennett, M. Zumhoff, Appl. Phys. Lett. 81 (22) [1] A.B. Chwang, M.A. Rothman, S.Y. Mao, R.H. Hewitt, X. Chu, L. Moro, T. Trajewski, N. Rutherford, Appl. Phys. Lett. 83 (23) 413. [11] J.-M. Moon, H.-K. Kim, J. Electrochem. Soc. 155 (28) J187. [12] J.C. Fan, F.J. Bachner, G.H. Foley, P.M. Zavracky, Appl. Phys. Lett. 25 (1974) 693. [13] P.H. Berning, A.F. Turner, J. Opt. Soc. Am. 47 (1957) 23. [14] J.-A. Jeong, H.-K. Kim, M.S. Yi, Appl. Phys. Lett. 93 (28) [15] J.-A. Jeong, H.-K. Kim, Sol. Energy Mater. Sol. Cells 93 (29) 181.