1. Introduction

Flexible optoelectronic devices have attracted considerable attention for realization of the next generation of displays, solid-state lighting, and energy-generating devices [1–5]. By adopting flexible optoelectronics, these devices can be released from site and design limits, resulting in a new technological regime for the generation and consumption of energy. For the applications, high external quantum efficiency (EQE) is of critical importance for the realization of devices with low power consumption and a long life [6,7]. In recent years, various semiconductors for active layers of optoelectronic devices have been developed having high internal quantum efficiency [8,9]. Each semiconductor had characteristic emission or absorption wavelengths that were determined by its band-gap [8,9]. Therefore, the characteristics of semiconductors can be maximized by pumping or injecting an appropriate wavelength of light. Typical flexible optoelectronic devices on a transparent substrate are composed of a transparent bottom anode, a thin-film semiconductor, and a reflective top cathode [10]. Since the light is coupled-out through the transparent anode, improving the optical transmittance of the anode for the appropriate wavelength is essential to achieve high EQE devices.

Several flexible transparent electrodes with tunable optical transmittance based on nanostructured carbon and metals have been reported [11,12]. The film of single-walled carbon nanotubes (CNTs) could have a wide color variance from 400 to 800 nm by controlling the chirality and diameter of the tubes [11,12]. When the diameter of single-walled CNTs varied from 0.9 to 1.6 nm, the first-order metallic transitions of the CNT shifted from 509 to 778 nm, resulting in a shift of the peak transmittance from 420 nm to 680 nm [11]. However, the transmittance was relatively low (~70%) and, more importantly, the high contact resistance between tubes resulted in high sheet resistance (>1,000 Ω/sq), even in the case of well-sorted metallic CNT (>100 Ω/sq) [13–15]. The electrical properties can be improved by using high conducting metal materials. The metal nanostructure showed good electrical properties (<20 Ω/sq) and could have optical tunability when the size of the nanostructure is sub-wavelength [12,16]. It is reported that the peak transmittance of film could be tuned from 480 nm to 680 nm by controlling the diameter from 120 nm to 220 nm, respectively [16]. However, this showed low transmittance of 50~60% when applied on transparent electrodes, resulting in high cost, a complex patterning process, and poor thermal stability [16,17]. Therefore, it is necessary to develop wavelength-tunable transparent electrodes to achieve flexible optical devices with good optoelectrical performance and simple processability.

One promising electrode is a metal-dielectric multilayer structure. A metal film sandwiched by dielectric materials with a high refractive index such as tungsten trioxide (WO₃) and molybdenum trioxide outperformed other transparent electrodes, including high transmittance (>80%), low sheet resistance (<15 ohm/sq), and good flexibility [18–20]. In addition, it can be fabricated by simple thin-film deposition, even using a roll-to-roll system. Recently, multilayer electrodes have been successfully employed for some of the electrodes in OLEDs and organic photovoltaics [18–20]. Even though various multilayer electrode structures for organic optoelectronics have been reported, only multilayer electrodes that have peak transmittance at the green color region (500~550 nm) have been reported, and work on designing electrodes with wavelength-tunable characteristics has not been conducted.

In order to enhance the optical transmittance of the dielectric-metal multilayer, the reflection of light in the multilayer (which is induced by the phase difference at the interface between the multilayer and the air) needs to be eliminated [21]. Therefore, the optical transmittance of the multilayer can be increased by adjusting the phase of light passing through the multilayer compared to that passing through the air, and the multilayer can have optical tunability by inserting a phase-adjusting layer suited for the desired wavelength. Previously, several results of controlling the optical properties of a dielectric-metal system by tuning the phase of light by the thickness of a dielectric film have been reported [22,23]. However, these methods were accompanied by reduced transmittance with increasing thickness, and the reasons were not explained. In addition, the novel method to tune the phase of light without degradation should be investigated.

We therefore demonstrated a wavelength-controlled flexible transparent electrode by controlling the phase by the refractive index of dielectrics in the metal-dielectric multilayer. The declined transmittance with the increasing thickness of the dielectric layer was accounted for with simulated and experimental results, and we also found the novel method without degradation of transmittance by controlling the refractive index of the dielectric layer. Through controlling the refractive indices of dielectrics between the metal and substrate (wavelength matching layer, WML) from 1.9 to 2.5, the peak wavelength of electrodes could be tuned from 470 nm to 610 nm by adjusting the optical-phase thickness of the entire electrode. To overcome the limitation of the blue region of ITO, we demonstrated the electrodes with WO₃, which has a refractive index of 1.9, as a WML. The WO₃/Ag/WO₃ (WAW) electrode deposited by thermal evaporation showed optical transmittance and sheet resistance comparable to those of ITO. At the λ = 460 nm, the WAW showed higher transmittance (90.5%) than that of ITO (86.4%). The optimal WAW electrode structures were predicted by optical simulation and experimental measurement. By replacing the ITO with WAW electrodes of blue OLEDs (λ = 460 nm), a high optical transmittance and low sheet resistance were achieved, resulting in an enhancement in the luminance of devices from 7020 to 7200 cd/m² at 222 mA/cm².

2. Methods

A glass substrate was used as the starting substrate. The substrate was cleaned with acetone, iso-propyl alcohol, and deionized water. The cleaned glass substrates were loaded into an evaporation chamber. For the deposition of TiO₂, the glass substrates were loaded into a sputter chamber. The TiO₂ targets were 15 cm in diameter with purity higher than 99.99% and were radio-frequency-sputtered in mixed Ar-O₂ discharges in which the concentration of the oxygen gas was at 5% pressure. For the deposition of ZnS and WO₃, the glass substrates were loaded into a thermal evaporation chamber. The films were deposited from a ZnS and WO₃ pallet (99.995%, with a diameter of 10 mm) at a base pressure in the order of 10⁻⁶ torr. The Ag/WO₃ (20 nm) anode was deposited from a WO₃ and Ag pallet (99.99%, with a diameter of 10 mm and 3 mm each) at a base pressure in the order of 10⁻⁶ torr. The thickness of the Ag layer was changed from 10 to 25 nm to optimize the optical transmittance and electrical conductivity. The thickness of the WO₃ layer below the Ag layer was also changed from 2.5 to 50 nm to optimize the optical transmittance. The organic layers and the Al cathode for the OLEDs were also deposited under high vacuum (10⁻⁶ torr) by thermal evaporation onto all samples at the same time to ensure consistent results. The structure of the OLEDs consisted of 4,4’-[N-(1-naphthyl)-N-phenyl-amino]biphenyl (α-NPD, 70 nm) emissive layer; 2,9-dimethyl-4,7-diphenyl-phenanthroline (BCP, 5 nm), Alq₃ (20 nm) LiF (1 nm), and Al (150 nm). Tris(1-phenyl-isoquinolinato-C₂,N)iridium(III)-doped (4,4’-N,N’-dicarbazole)biphenyl (Ir(piq)₃:CBP, 24 wt%, 50 nm); Coumarin 545 tetramethyl-doped Tris(8-hydroxyguinolinato)aluminium (1%-C545T:Alq₃, 40 nm); and iridium (III) bis(4,6-difluorophenylpyridinato)tetrakis(1-pyrazolyl)borate doped TCTA (FIr6:TCTA, 10 wt%, 50 nm) were used as an emissive layer for the green, blue, and deep blue OLEDs. The active area of the device was 3 × 3 mm².

The current density voltage and luminescence-current density characteristics of the devices were measured with an HP-4156A semiconductor parameter analyzer and a Yokogawa 3298F in nitrogen ambient. The refractive index and the transmittance values of the refractive index modulation layer were measured by using ellipsometry (J. A. Woollam Co., Inc M-44), a tungsten-halogen lamp, and a monochromator.

To evaluate the flexibility of film, a two-point bending test was examined with WAW and ITO on polyethylene terephthalate (PET). PET substrates were cleaned with isopropyl alcohol and deionized water, then dried with a high-purity N₂ gas. The PET was loaded into a thermal evaporation chamber for the deposition of WAW. The sheet resistances of WAW and ITO (639303, Sigma-Aldrich) were monitored after bending with the radius of 25 mm.

The finite domain time-dependent (FDTD)-based simulation (Fullwave, LED utility) was performed for analysis of the electric field distribution and far-field emission pattern for OLEDs with DMDs. The actual thicknesses and refractive indices of each layer were used for calculation, while the thickness of the glass reduced to 380 nm for minimization of the calculation time. Point dipole-polarized x and z directions were used as the radiating sources and placed at the interface of an emissive layer with a hole transport layer. The light source wavelength was set to 520 nm.

3. Results and discussion

The admittance (Y), which is the ratio of magnetic field (H) to electric field (E), is an important physical quantity with which to calculate the optical characteristics of film. The admittance is defined as below,

Based on the transfer matrix theory, the admittance of the film with refractive index (n_f) is derived as below,

The δ is optical-phase thickness defined as

Based on the equation, we can calculate the trace of the admittance in a complex plane. In the case of dielectric film, the starting admittance point is (n_sub,0). As the thickness is increased to a quarter wave, a semicircle is traced out clockwise, which re-intersects the real axis in the point (n²_dielectric/n_sub). A further increase of the thickness of the dielectric layer corresponding to a second quarter wave completes the circle. On the other hand, because the metal has an imaginary part of the refractive index, the admittance diagram of the metal film is somewhat distorted from the ideal case, with a loop bowing out along the direction of the real axis and converge to (n,–k). The reflection of film (R) can be calculated by admittance,

 

Fig. 1 Schematic illustrations of OLEDs with (a) tunable dielectric-metal multilayer with control of thickness, (b) the admittance diagram of the system, (c) tunable dielectric-metal multilayer with control of refractive index, and (d) the admittance diagram of the system.

Download Full Size | PDF

Figure 2(a) shows the sheet resistance (R_S) of the WAW multilayer as a function of Ag thickness. Ag is used as a conducting layer because it has the lowest value of n (n_Ag = 0.05 – i2.90 in the visible region), meaning the lowest absorption among metals. It is well known that Au, Ag, and Al have a good electrical conductivity (>3 × 10⁷ S/m at 20°C). However, Ag shows the lowest absorptivity (<5%) in the visible-light region, compared with those of Au (8%) and Al (30%) [24]. Therefore, Ag is the most promising metal layer in designing a DMD transparent electrode. The 5-nm-thick Ag on WO₃ showed high R_S (5100 ohm/sq) due to an island growth of Ag film [25]. As the film thickness increased, the R_S decreased to 11.08 ohm/sq for 12-nm-thick Ag, comparable with ITO (~12 ohm/sq, 120-nm-thick). Thus, we chose the 12 nm as the optimum thickness of Ag. The work function was measured using secondary electron (SE) emission spectra, as shown in Fig. 2(b). The onset of SE is determined by extrapolating two solid lines from the background and a straight onset in the spectra [26]. The cut-off energy of Ag was measured to be 4.49 eV, while that of ITO was measured to be 4.9 eV. The work function of Ag is much lower than that of ITO, which can increase the hole injection barrier between the anodes and the hole transport layer. The thin WO₃ layer could modify the work function to 6.5 eV, which is higher than that of ITO [Fig. 2(b)]. Thus, WO₃ could efficiently inject the hole from the Ag to the hole transport layer, and we chose the WO₃ layer as a hole injection inner dielectric material. The thickness of the WO₃ layer affected the optoelectrical properties [Fig. 2(c)]. The transmittance of Ag(12 nm)/WO₃ was increased from 69% to 87% as the thickness of WO₃ increased from 0 to 30 nm, due to the zero-reflection condition being satisfied. However, the hole transport was limited by its low electrical conductivity (~10⁻⁷ S/cm) [27]. Thus, the electrical performance was degraded as the thickness of WO₃ increased. As a result, the optimized thickness of the WO₃ layer was determined as 20 nm. [Fig. 2(d)].

 

Fig. 2 (a) Sheet resistance of Ag films as a function of Ag thickness. (b) Secondary cut-off spectra of ITO, Ag and Ag/WO₃. (c) Secondary cut-off spectra. (d) Power efficiency of OLEDs as a function of WO₃ thickness.

Download Full Size | PDF

We simulated the optical transmittance of dielectric/Ag/WO₃ structures as a function of the refractive indices [Fig. 3(a)] and thickness [Fig. 3(b)] of the outer dielectric layer. The refractive indices and extinction coefficients were used for simulation. The refractive indices Ag and WO₃ were measured as 0.071 and 1.93, respectively, and extinction coefficients were measured as 2.25 and 0.028, respectively, at a wavelength of 460 nm [Fig. 9], and these values are well matched with those reported in the literature [28,29]. The thickness of Ag and inner WO₃ are fixed as 12 nm and 20 nm, respectively, which are the optimized thicknesses according to the results in Fig. 2. The wavelength of peak transmittance of film shifts from 470 nm to 610 nm as the refractive indices of films increased from 1.8 to 2.5, shown in Fig. 3(a), under the zero-reflection condition, which is defined as the optical-phase thickness to get the peak transmittance, plotted in Fig. 3(c). On the other hand, the wavelength of peak transmittance of film also shifts from 500 nm to 600 nm as the thickness of the outer dielectric layer (WO₃) increased from 30 nm to 50 nm, but the peak transmittance is decreased as the thickness increased [Figs. 3(b) and 3(d)].

 

Fig. 3 Calculated value (line) of transmittance of dielectric/Ag/WO₃ as function of (a) refractive indices and (b) thickness of dielectric. Peak transmittance of dielectric/Ag/WO₃ as a function of (c) refractive indices and (d) thickness of dielectric and the optical-phase thickness of dielectric layer at the peak transmittance.

Download Full Size | PDF

It is well known that emission of OLED comes from isotropically oriented chromophores, so there is no specific angle of incidence. Directions must be taken into account while doing these calculations in Fig. 3. The critical angle at the interface of glass with air is calculated to be 41.8° from the surface normal. Thus, incident photons with a higher angle than 41.8° will be confined in the glass. With the consideration of the incident photons below the critical angle, the transmittance decrement was determined to be only 2% (from 89% to 87%) as the incident angle increased to the critical angle (~40°), shown in Fig. 8. There was only a little degradation (~2%) in transmittance below the critical angle in the wavelength range between 400~700 nm [Fig. 8(b)]. This provides evidence that incident photons to the normal surface is enough to calculate the transmittance in DMD structures.

To compare electrodes with wavelength tunability by changing refractive index and thickness, we simulated the optical properties of the structures. In the case of a varied refractive index system, the highest transmittance (the lowest reflectance) section was maintained from a refractive index of 1.9 to 2.5 [Figs. 4(a) and 4(c)]. On the other hand, the highest transmittance section of a varied thickness system shows an ellipse shape centered at a 40-nm-thick dielectric; the lower transmittance section shows concentric ovals, meaning the peak transmittance is decreased by changing thickness and wavelength [Figs. 4(b) and 4(d)]. The reason for degradation was caused by breaking the condition of absolute zero reflection with a varied wavelength of incident light. As shown in Fig. 4(e), the admittance of a DMD system can cause absolute zero with different incident light by adjusting the refractive index of dielectrics, because the radius of the admittance trace is changed by the refractive index. However, the admittance of the system cannot be zero by changing thickness with light of varied wavelengths, since the only the length of admittance trace can be adjusted by changing thickness of dielectrics [Fig. 4(f)].

 

Fig. 4 Simulated contour plots of transmittance and reflectance for Dielectric/Ag/WO3 multilayers upon variation of the (a) (c) refractive index of the dielectric and (b) (d) thickness of dielectric (n = 1.9). The calculated admittance diagram of the system of (e) variation of refractive index and (f) thickness.

Download Full Size | PDF

To demonstrate an electrode with wavelength tunability, we chose three kinds of materials, WO₃, ZnS, and TiO₂, with different refractive indices of 1.9, 2.3, and 2.5, respectively, as outer dielectrics [Fig. 9]. Based on the transfer matrix method, the transmittance of DMD layers was calculated as a function of metal and dielectric thicknesses [30], Fig. 10. As a result, high transmittance of 90% can be expected for the WAW (λ = 460 nm), ZnS/Ag/WO₃ (ZAW, λ = 520 nm), and TiO₂/Ag/WO₃ (TAW, λ = 620 nm), respectively. The optical properties of designed electrodes with various thicknesses of WO₃ and refractive indices are shown on Fig. 5.The WAW showed high optical transmittance in the blue emission region (90.5% at λ = 460 nm), which is higher than that of ITO. When the thickness of WO₃ increased from 30 nm to 70 nm, the peak wavelength of transmitted light shifted from 470 to 700 nm, but the transmittance decreased from 90.5% to 76.3%. However, the peak wavelength of transmitted light shifted from 470 to 660 nm with ZnS and TiO₂ outer dielectric layers, and the transmittance was maintained as 88%.

 

Fig. 5 Measured transmittance of (a)WO₃/Ag/WO₃ with various thickness of WO₃ and (b) WO₃/Ag/WO₃, ZnS/Ag/WO₃, and TiO₂/Ag/WO₃.

Download Full Size | PDF

We fabricated OLEDs with three different emissive layers; CBP:Ir(piq)₃, Alq₃:C545T, and TCTA:FIr6. The main emission peaks of CBP:Ir(piq)_3, Alq₃:C545T, and TCTA:FIr6 were located at 622 nm, 525 nm, and 460 nm, respectively. Figure 6(a) shows the electroluminescent (EL) spectra of OLEDs with three different emissive layers. The WO₃/Ag/WO₃, ZnS/Ag/WO₃, and TiO₂/Ag/WO₃ were used as electrodes for blue, green, and red devices, respectively. The peak intensities are normalized by peak intensities of devices each with ITO. The device with WO₃/Ag/WO₃ shows 4% higher peak intensity than that of ITO, according to the increasing transmittance. In addition, the device with ZnS/Ag/WO₃ and TiO₂/Ag/WO₃ shows comparable performance with devices with ITO. Figure 6(b) shows the luminance–current density–voltage (L-J-V) characteristics of OLEDs light with 460 nm using ITO and WAW. The operation voltage of the device with the ITO anode was 6.8 V at J = 1 mA/cm². It slightly decreased to 6.5 V as the anode was replaced with WAW. The same results were obtained in red and green devices [Figs. 6(c) and 6(d)]. This indicates that the hole injection barrier and properties between Ag/WO₃/Hole transport layer (HTL) might almost be the same as those between ITO/HTL, since the work function of WO₃ is large enough compared to that of ITO. The luminance value of OLEDs with ITO was 7020 cd/m² (J = 220 mA/cm²) and that of the OLEDs with WAW was increased to 7200 cd/m² due to the higher optical transmittance of WAW. Also, red and green devices with ZnS/Ag/WO₃ and TiO₂/Ag/WO₃ showed similar performance with that of devices using ITO [Figs. 6(c) and 6(d)]. However, the OLEDs with 40-nm-thick and 55-nm-thick WO₃ show decreased luminance caused by low transmittance. Figure 6(e) shows the enhancement factor of power efficiencies at J = 10 mA/cm², defined as the power efficiency of DMD electrodes divided by those of the devices with ITO. This result indicates that the blue OLEDs with the WAW anode have 6% higher power efficiency (1.01 lm/W) than that of the device with the ITO anode (0.95 lm/W). In the case of green and red OLEDs, the enhancement factor is slightly decreased, but still shows enhancement factor value close to 1. Bend experiments were performed for WAW and ITO as a function of the number of cycles at a curvature of 25 mm [Fig. 6(f)]. In the case of ITO, the sheet resistance increased from 65 to 250 ohm/sq after 1,000 cycles, while the values for the WAW exhibited a significantly better performance than those for ITO.

 

Fig. 6 Luminance-current density-voltage characteristics of (b) blue OLEDs, (c) green OLEDs, and (d) red OLEDs with ITO and DMD electrodes. (Three emissive layers: TCTA:FIr6 (460 nm), Alq₃:C545T (525 nm), CBP:Ir(piq)₃ (620 nm) (d) EL spectra of blue, green, and red OLEDs with ITO, and DMD electrodes.. (e) Enhancement factor of devices using DMD compared to ITO with three emissive layers. (f) Sheet resistance after repeated bending as function of the number of cycles for ITO and WO₃/Ag/WO_3.

Download Full Size | PDF

To analyze the effect of DMD electrodes on the light extraction in OLEDs, we used the finite domain time dependent (FDTD) method. Figures 7(a)–7(c) show the simulated electric field distribution of OLEDs with DMD electrodes. It was revealed that the intensity of trapped light in DMD layer increases with the refractive index of dielectric layer. This is due to the optical wave-guided light propagating through dielectric layer in DMD. The larger the difference of refractive indices between glass and dielectric layer, the higher Fresnel reflection at the surface of dielectric layer, leading to the strong electric field at the center of DMD layers, the Ag layer [31]. This means that the amount of photons trapped in the glass substrate reduces as the refractive index of the dielectric layer increases. The critical angle of total internal reflection at the interface of WO₃ with glass (52.2þ) is larger than that of glass with air (41.8þ). Thus, the reflected light between 41.8þ and 52.2þ is reflected at the glass/air interface, resulting in a trap inside of the glass substrate. On the other hand, the critical angle at the TiO₂/glass interface is as small as 36þ, leading to the escape of most photons from the glass substrate. Figure 7(d) shows the calculated angular emission pattern of OLEDs for three kinds of DMD structures. The device with WO₃/Ag/WO₃ shows a Lambertian-like emission pattern, but those with ZnS/Ag/WO₃ and TiO₂/Ag/WO₃ deviate from the Lambertian emission pattern by having extra emission at low angles in the range of (22þ~30þ). This originated from the optical wave-guiding mode in the DMD layers, similar to the results previously reported in other literature [32].

 

Fig. 7 Simulated electric field distribution of OLEDs with (a) WO₃/Ag/WO₃, (b) ZnS/Ag/WO₃, (c) TiO₂/Ag/WO₃ electrodes, and (d) Simulated angular emission pattern of OLEDs with DMD electrodes.

Download Full Size | PDF

4. Conclusions

In conclusion, we demonstrate flexible transparent electrodes with wavelength-tuned Ag-based multilayers as an ITO alternative. Inserting WML can shift the peak wavelength of the transmitted light of the electrodes by adjusting the phase optical thickness. As the refractive indices of the outer dielectric layers increased, the peak wavelength shifted from 470 nm to 610 nm and is experimentally confirmed by adapting WO₃, ZnS, and TiO₂. The inner hole injecting dielectric, WO₃, which has a high work function, can enhance the hole injection properties of the electrode. We employed a WAW multilayer to replace the ITO electrode, which has a low transmittance (86.4% at λ = 460 nm) in the blue emission region. The optimum thicknesses of both Ag and WO₃ to obtain the best transmittance value of WAW were determined by theoretical simulation, and these agreed well with the experimental results. The WAW showed high transmittance in the blue emission region (90.5% at λ = 460 nm) and low sheet resistance (11.08 ohm/sq). Thus, the blue OLEDs with WAW could show a higher luminance value of 7,200 cd/m² (J = 220 mA/cm²) than that of the device with ITO (7,020 cd/m²).

Acknowledgments

This research was financially supported in part by the National Research Foundation of Korea (NRF) Grant funded by the Korean government (MEST) (No. NRF-2013R1A2A2A01069237), Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (2009-0094036), and the Brain Korea 21 PLUS Project for the Center for Creative Industrial Materials.

Appendix

 

Fig. 8 (a) Transmittance at 460 nm and (b) transmittance spectra of Glass/WO₃ (30 nm)/Ag (12 nm)/WO₃ (20 nm) as a function of incident angle.

Download Full Size | PDF

 

Fig. 9 Complex refractive indices of (a) Ag and (b) WO₃, ZnS, TiO₂ (23°C), and TiO₂ (400°C) as a function of wavelength.

Download Full Size | PDF

 

Fig. 10 Simulated contour plots of transmittance for (a) WO₃/Ag/WO₃, (b) ZnS/Ag/WO₃, and (c) TiO₂/Ag/WO₃ multilayers upon variation of the thickness of the WO₃ and Ag layers. Calculated value (line) of transmittance for (a) WO₃/Ag/WO₃ (λ = 460 nm) (b) ZnS/Ag/WO₃ (λ = 520 nm), and (c) TiO₂/Ag/WO₃ (λ = 620 nm) electrode as a function of outer dielectric layer and Ag thicknesses. Corresponding results obtained from experiments (symbols) are also shown for comparison.

Download Full Size | PDF

Acknowledgments

References and links

1. Y. Fu, Z. Lv, S. Hou, H. Wu, D. Wang, C. Zhang, and D. Zou, “TCO-free, flexible, and bifacial dye-sensitized solar cell based on low-cost metal wires,” Adv. Eng. Mater. 2(1), 37–41 (2012). [CrossRef]  

2. C. E. Small, S. Chen, J. Subbiah, C. M. Amb, S.-W. Tsang, T.-H. Lai, J. R. Reynolds, and F. So, “High-efficiency inverted dithienogermole-thienopyrrolodione-based polymer solar cells,” Nat. Photonics 6(2), 115–120 (2011). [CrossRef]  

3. H. Park, P. R. Brown, V. Bulović, and J. Kong, “Graphene as transparent conducting electrodes in organic photovoltaics: studies in graphene morphology, hole transporting layers, and counter electrodes,” Nano Lett. 12(1), 133–140 (2012). [CrossRef]   [PubMed]  

4. M. Vosgueritchian, D. J. Lipomi, and Z. Bao, “Highly conductive and transparent PEDOT:PSS fims with a fluorosurfactant for stretchable and flexible transparent electrodes,” Adv. Funct. Mater. 22(2), 421–428 (2012). [CrossRef]  

5. S. Y. Kim, K. Kim, K. Hong, and J.-L. Lee, “Flexible organic light-emitting diodes using a metal peel-off method,” IEEE Photon. Lett. 20(22), 1836–1838 (2008). [CrossRef]  

6. J. S. Swensen, E. Polikarpov, A. Von Ruden, L. Wang, L. S. Sapochak, and A. B. Padmaperuma, “Improved efficiency in blue phosphorescent organic light-emitting devices using host materials of lower triplet energy than the phosphorescent blue emitter,” Adv. Funct. Mater. 21(17), 3250–3258 (2011). [CrossRef]  

7. L. Dou, J. You, J. Yang, C.-C. Chen, Y. He, S. Murase, T. Moriarty, K. Emery, G. Li, and Y. Yang, “Tandem polymer solar cells featuring a spectrally matched low-bandgap polymer,” Nat. Photonics 6(3), 180–185 (2012). [CrossRef]  

8. H. Sasabe, N. Toyota, H. Nakanishi, T. Ishizaka, Y.-J. Pu, and J. Kido, “3,3′-Bicarbazole-based host materials for high-efficiency blue phosphorescent OLEDs with extremely low driving voltage,” Adv. Mater. 24(24), 3212–3217 (2012). [CrossRef]   [PubMed]  

9. G. Li, R. Zhu, and Y. Yang, “Polymer solar cells,” Nat. Photonics 6(3), 153–161 (2012). [CrossRef]  

10. M. G. Helander, Z. B. Wang, J. Qiu, M. T. Greiner, D. P. Puzzo, Z. W. Liu, and Z. H. Lu, “Chlorinated indium tin oxide electrodes with high work function for organic device compatibility,” Science 332(6032), 944–947 (2011). [CrossRef]   [PubMed]  

11. A. A. Green and M. C. Hersam, “Colored semitransparent conductive coatings consisting of monodisperse metallic single-walled carbon nanotubes,” Nano Lett. 8(5), 1417–1422 (2008). [CrossRef]   [PubMed]  

12. T. Xu, H. Shi, Y.-K. Wu, A. F. Kaplan, J. G. Ok, and L. J. Guo, “Structural colors: from plasmonic to carbon nanostructures,” Small 7(22), 3128–3136 (2011). [CrossRef]   [PubMed]  

13. R. Saito, M. Fujita, G. Dresselhaus, and M. S. Dresselhaus, “Electronic structure of chiral graphene tubules,” Appl. Phys. Lett. 60(18), 2204–2206 (1992). [CrossRef]  

14. N. Hamada, S.-i. Sawada, and A. Oshiyama, “New one-dimensional conductors: graphitic microtubules,” Phys. Rev. Lett. 68(10), 1579–1581 (1992). [CrossRef]   [PubMed]  

15. J. Wu, M. Agrawal, H. A. Becerril, Z. Bao, Z. Liu, Y. Chen, and P. Peumans, “Organic light-emitting diodes on solution-processed graphene transparent electrodes,” ACS Nano 4(1), 43–48 (2010). [CrossRef]   [PubMed]  

16. S. L. Hellstrom, H. W. Lee, and Z. Bao, “Polymer-assisted direct deposition of uniform carbon nanotube bundle networks for high performance transparent electrodes,” ACS Nano 3(6), 1423–1430 (2009). [CrossRef]   [PubMed]  

17. M. Zhang, M. Yu. Efremov, F. Schiettekatte, E. A. Olson, A. T. Kwan, S. L. Lai, T. Wisleder, J. E. Greene, and L. H. Allen, “Size-dependent melting point depression of nanostructure: nanocalorimetric measurements,” Phys. Rev. B 62(15), 10548–10557 (2000). [CrossRef]  

18. Z. B. Wang, M. G. Helander, J. Qiu, D. P. Puzzo, M. T. Greiner, Z. M. Hudson, S. Wang, Z. W. Liu, and Z. H. Lu, “Unlocking the full potential of organic light-emitting diodes on flexible plastic,” Nat. Photonics 5(12), 753–757 (2011). [CrossRef]  

19. N. P. Sergeant, A. Hadipour, B. Niesen, D. Cheyns, P. Heremans, P. Peumans, and B. P. Rand, “Design of transparent anodes for resonant cavity enhanced light harvesting in organic solar cells,” Adv. Mater. 24(6), 728–732 (2012). [CrossRef]   [PubMed]  

20. Y. Jin, J. Feng, X.-L. Zhang, Y.-G. Bi, Y. Bai, L. Chen, T. Lan, Y.-F. Liu, Q.-D. Chen, and H.-B. Sun, “Solving efficiency-stability tradeoff in top-emitting organic light-emitting devices by employing periodically corrugated metallic cathode,” Adv. Mater. 24(9), 1187–1191 (2012). [CrossRef]   [PubMed]  

21. H. A. Macleod, Thin-film optical filter (Taylor & Francis, 2001).

22. X. Liu, X. Cai, J. Qiao, J. Mao, and N. Jian, “The design of ZnS/Ag/ZnS transparent conductive multilayer films,” Thin Solid Films 441(1-2), 200–206 (2003). [CrossRef]  

23. H. Cho, C. Yun, and S. Yoo, “Multilayer transparent electrode for organic light-emitting diodes: tuning its optical characteristics,” Opt. Express 18(4), 3404–3414 (2010). [CrossRef]   [PubMed]  

24. D. Graiffiths, “7. Electrodynamics” in Introduction to Electrodynamics (Prentice Hall, 1999).

25. S. Nigam and C. Majumder, “Growth pattern of Ag(_n) (n = 1-8) clusters on the α-Al₂O₃(0001) surface: a first principles study,” Langmuir 26(24), 18776–18787 (2010). [CrossRef]   [PubMed]  

26. S. Kim, K. Hong, K. Kim, I. Lee, and J.-L. Lee, “Phase-controllable copper oxides for an efficient anode interfacial layer in organic light-emitting diodes,” J. Mater. Chem. 22(5), 2039–2044 (2012). [CrossRef]  

27. K. Aguir, C. Lemire, and D. B. B. Lollman, “Electrical properties of reactively sputtered WO₃ thin films as ozone gas sensor,” Sens. Actuat. B-Chem. 84(1), 1–5 (2002). [CrossRef]  

28. H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell, and K. Kalantar-zadeh, “Nanostructured tungsten oxide – properties, synthesis, and applications,” Adv. Funct. Mater. 21(12), 2175–2196 (2011). [CrossRef]  

29. N. Lavrik and D. Leckband, “Optical and direct force measurements of the interactions between monolayers of aromatic macrocycles on surfactant monolayers,” Langmuir 16(4), 1842–1851 (2000). [CrossRef]  

30. K. Wu, C.-C. Lee, N. J. Brock, and B. Kimbrough, “Multilayer thin-film inspection through measurements of reflection coefficients,” Opt. Lett. 36(16), 3269–3271 (2011). [CrossRef]   [PubMed]  

31. B. G. Bovard, “Rugate filter theory: an overview,” Appl. Opt. 32(28), 5427–5442 (1993). [CrossRef]   [PubMed]  

32. D. Kabra, M. H. Song, B. Wenger, R. H. Friend, and H. J. Snaith, “High efficiency composite metal oxide-polymer electroluminescent devices: a morphological and material based investigation,” Adv. Mater. 20(18), 3447–3452 (2008). [CrossRef]