Open Access
Add to BibSonomyAbstract
This paper describes the luminance uniformity of OLED lighting panels depending on OLED device structures of single emission layer (single-EML), 2-tandem, and 3-tandem. The luminance distribution is evaluated through the circuit simulation and the fabricated panel measurement. In the simulation results with yellow-green color panels of 30 × 80 mm2 emission area, a 3-tandem structure shows the lowest non-uniformity (1.34% at 7.5V), compared to single-EML (5.67% at 2.8V) and 2-tandem (2.78% at 5.3 V) structures at 1,000 cd/m2. The luminance non-uniformity is germane to the OLED conductance showing that the high luminance-current efficiency is of the most importance to achieve the uniform voltage and luminance distribution. In measurement, a 3-tandem structure also achieves the most uniform luminance distribution with non-uniformity of 4.1% while single EML and 2-tandem structures accomplish 9.6%, and 6.4%, respectively, at ~1,000 cd/m2. In addition, the simulation results ensure that a 3-tandem structure panel is allowed to be enlarged the panel size up to about 5,000 mm2 for lower luminance non-uniformity than 10% without any auxiliary metal electrodes.
© 2015 Optical Society of America
1. Introduction
White organic light-emitting diodes (WOLEDs) have attained a lot of interests as a next generation lighting technology because of various promising properties such as high color rendering index (CRI), homogeneous surface emission, thin thickness, high transparency, and flexibility [1–7]. Several kinds of WOLEDs with a variety of emission layer (EML) structures have been developed; a single-EML structure in which multiple emitters are applied on a monolayer, a multi-EML structure which has continuous multiple emitting layers, and tandem structures that have two or three OLED sets stacked. A single-EML structure has the simplest structure with poor efficiency and lifetime. A multiple-EML structure relatively shows higher efficiency and longer lifetime, but requires further improvement of color stability and efficiency. A tandem structure outperforms two above structures in terms of efficiency, color stability, and lifetime. A tandem structure is easier to adjust the spectrum due to the independent multi-photon emission and achieves longer lifetime with the current sharing. As a consequence, a tandem structure has been considered as the promising structure in lighting applications [8,9]. Recently, high efficiency 2- or 3-tandem WOLEDs for lighting applications have been reported with efficiency ranges of 100~140 lm/W at 1000 cd/m2 [10].
One of critical issues in WOLED lighting panels is the luminance non-uniformity. Because the transparent conducting oxide (TCO) anode electrode such as indium tin oxide (ITO) has the limitation on the conductivity, the non-uniform voltage distribution over the large emitting area takes place, which leads to the non-uniform current and luminance distributions on an OLED panel [11–13]. To increase the conductivity of an anode electrode for the high uniformity, conductive pads are implemented on both sides of an anode electrode and some auxiliary high conductive electrodes of low sheet resistance materials such as Al, Cu, Cr, and so on are placed on a TCO layer at the fashion of mesh grids [14]. However, this mesh grid scheme might reduce the overall luminance owing to the decreased aperture ratio [15,16].
There have been two simulation approaches to evaluate the luminance non-uniformity of OLED lighting panels. One is a finite element method (FEM) and the other is a circuit simulation method. FEM makes use of partial differential equations (PDEs) with panel structure and complicated physical properties of materials such as electrical conductivity, characteristic curve between current and voltage, temperature, and heat conductivity. By solving PDEs, the current, temperature, and luminance distributions are obtained [16–19]. Even though precise results are accomplished in FEM, a huge amount of data about structures and materials should be prepared prior to the simulation. However, the circuit simulation method utilizes a simple model in which anode and cathode electrodes are presented with resistor grids and organic layers are described with the array of diodes. Then, the current distribution over diodes is estimated by means of a simulation program with integrated circuit emphasis (SPICE). Despite a bit lower accuracy, this method can provide the simple and fast estimation about the non-uniformity performance.
In this paper, we evaluate the luminance non-uniformity with circuit simulation method and fabricated lighting panels for single-EML, 2-tandem, and 3-tandem yellow-green color OLED structures.
2. Experiment
Three yellow-green color OLED lighting panels of single-EML, 2-tandem, and 3-tandem structures are fabricated as depicted in Figs. 1(a), 1(b), and 1(c) respectively. The thickness of an ITO substrate for an anode electrode is 150 nm and the emitting area of a panel is 30 × 80 mm2. Additionally, OLED unit devices are fabricated with the emitting area of 2 × 2 mm2 to extract the device characteristics. Substrates are ultrasonically cleaned with acetone, isopropyl alcohol, and deionized water and treated with ultraviolet-ozone. 1,4,5,8,9,11-Hexaazatriphenylene hexacarbonitrile (HATCN) and 4,40-bis[N-(1-nathyl)-N-phenylamino]biphenyl (NPB) are used as hole injection layer (HIL) and hole transport layer (HTL) and 6wt% iridium (III) bis(4-phenylthieno[3,2-c]pyridinato-N,C2’)acetylacetonate (PO-01) doped bis[2-(2-hydroxyphenyl)-pyridine]beryllium (Bepp2) is utilized as yellow green EML. 4,7-Diphenyl-1,10-phenanthroline (Bphen) and Li are in use as electron transporting layer (ETL) and n-type dopant, respectively. Organic and cathode materials are deposited using a thermal evaporator system under ~10−7 Torr pressure. Then, 100 nm thick aluminum layer pads are deposited at both sides of an anode electrode. After deposition, all devices and panels are encapsulated by the glass cap.
Fig. 1 Three fabricated OLED devices (a) Single-EML structure (b) 2-tandem structure (c) 3-tandem structure.
Current density-voltage-luminance (J-V-L) relations of unit devices are measured with a luminance and color meter (Konica Minolta CS-100A) and a source meter unit (Keithley 2635A). Electroluminescence (EL) spectra and CIE (Commission Internationale de l’Eclairage) 1931 color coordinators are measured by using a spectroradiometer (Konica Minolta CS-2000). Luminance distributions of panels are obtained by means of a 2D color analyzer (Konica Minolta CA-2500) over 6 and 16 points in vertical and horizontal center at the pitch of 5 mm.
The panel simulation is conducted by SmartSpice with extracted unit device parameters and measured sheet resistances of anode and cathode electrodes. Sheet resistances are measured by a sheet resistance meter (DASOLENG FPP-2000A).
3. Results and discussion
The circuit simulation based evaluation is proceeded as described in Fig. 2. The unit diode is modeled by using electrical properties measured from a unit OLED device. Then, the overall panel circuit is divided into mesh grids of the unit device area that is composed of anode resistors, cathode resistors, and a unit diode [12,14]. Currents and current densities estimated in each mesh area are converted into luminance via the measured luminance-current density (L-J) properties of unit OLED devices. For more accurate results, the heat distribution should be considered owing to the mobility variation with temperature change. According to previous reported papers, the current deviation between electrical and electrothermal simulation is less than 10% because used power density value is low under 50 mW/cm2. Luminescence uniformity variation by this thermal effect is negligible at the low power density devices [16–20]. Therefore, in this paper the thermal effect is allowed to be ignored due to the small emission area of an OLED panel.
Three OLED unit devices are fabricated with 2 × 2 mm2 emitting area and measured device specifications are summarized in Table 1 where driving voltages were 2.7 V, 5.1 V, and 7.6 V for single-EML, 2-tandem, and 3-tandem devices, respectively, at 1,000 cd/m2 according to current-voltage (I-V) and luminance-voltage (L-V) curves depicted in Fig. 3(a). As plotted in Fig. 3(b), the current efficiencies of 2- and 3-tandem devices are almost two and three times higher than that of a single-EML device, which indicates that the fabricated tandem connections are well built without optical losses. The equivalent color coordinators are obtained as (0.39, 0.59) for single-EML devices, (0.39, 0.59) for 2-tandem devices, and (0.37, 0.60) for 3-tandem devices as illustrated in Fig. 3(c).
Table 1. Specification summary of single-EML, 2-tandem, and 3-tandem unit devices
Fig. 3 Performances of three single-EML, 2-tandem, and 3-tandem OLED unit devices (a) Measured current density and luminance versus the applied voltage (b) Measured current efficiency and power efficiency versus luminance (c) Measured electroluminescence (EL) spectra.
Figure 4(a) shows I-V curves of diode models for simulation well matched with measurement results. L-J curve models are established by 3rd order regression curves of measured L-J plots in Fig. 4(b) as Eqs. (1)-(3) with consideration for the efficiency roll-off characteristic. In these models, JS-EML, J2-tandem, and J3-tandem are the current densities at mA/cm2 and LS-EML, L2-tandem, and L3-tandem are luminance values at cd/m2 for single-EML, 2-tandem, and 3-tandem structures. The errors between simulated and measured luminance are estimated within ± 14 cd/m2 where maximum errors are ± 14 cd/m2 at 124 cd/m2, ± 13 cd/m2 at 2,140 cd/m2, and ± 3 cd/m2 at 110 cd/m2 in single-EML, 2-tandem, and 3-tandem devices.
Fig. 4 OLED unit device models (a) Measured and modeled I-V curves (b) Measured and approximated L-J curves.
The panel architecture of a 30 × 80 mm2 emitting area with anode pads and cathode electrode is depicted in Fig. 5(a) and the overall circuit schematic with mesh circuits of resistors and diodes is made up as presented in Fig. 5(b). Sheet resistances of anode and cathode electrodes are measured as 9.86 Ω/sq and 476 mΩ/sq. The area of a mesh circuit is 2 × 2 mm2 which is equivalent to a unit device and the total number of mesh circuits is 600 (15 × 40).
Fig. 5 Luminance distribution evaluation based on circuit simulation (a) Architecture of fabricated panel and position of electrode contacts (b) Overall panel schematic with mesh circuits.
The luminance non-uniformity is calculated with Eq. (4), where Lmax and Lmin are maximum and minimum luminance over the area. Figure 6(a) plots the simulated non-uniformity regarding the maximum luminance showing that a 3-tandem structure shows the lowest non-uniformity. Luminance non-uniformity at ~1,000 cd/m2 is 1.34%, 2.78%, and 5.67% for 3-tandem, 2-tandem, and single-EML devices. As shown in Fig. 6(b), the non-uniformity plot is similar to the OLED conductance plot with respect to luminance. Because the 3-tandem structure leads to a device of the highest current efficiency, the high output luminance can be generated at the small conductance, that is, the low current density, and the lowest non-uniformity can be achieved due to the low voltage variation over the area. The luminance change over the voltage across an OLED also pertains to the luminance non-uniformity, where steeper changes result in higher non-uniformity values.
Fig. 6 The relation between luminance non-uniformity and the conductance of a unit OLED device (a) Simulated luminance non-uniformity versus the maximum luminance (b) Measured conductance of unit devices regarding the output luminance.
To verify our simulation results, three single-EML, 2-tandem, and 3-tandem panels are fabricated and the luminance distribution is measured concerning two cross sections in horizontal and vertical directions as described in Fig. 7(a). Figures 7(b) and 7(c) present the measured normalized luminance distributions of horizontal and vertical directions along with simulation results, respectively. Solid and dash lines represent measurement and simulation results. As expected by simulation results, whereas a single-EML panel makes the largest luminance deviation, a 3-tandem panel accomplishes the best luminance uniformity. However, the measured luminance distributions of fabricated panels depart from simulated ones due to non-uniform deposition of about 10% thickness variation caused by sublimation sources. Since edges of the luminance area are relatively thinner than the center area, the increased currents of edges deteriorate the luminance uniformity further. At 1,000 cd/m2, horizontal non-uniformity values are obtained as 9.6% at 3.2 V, 6.4% at 5.2 V, and 4.1% at 7.6 V for single-EML, 2-tandem, and 3-tandem panels. At 1,000 cd/m2, vertical non-uniformity values are obtained as 12.8% at 3.2 V, 9.7% at 5.2 V, and 3.4% at 7.6 V for single-EML, 2-tandem, and 3-tandem panels.
Fig. 7 The measurement results of a fabricated panel (a) Photograph of a fabricated 3-tandem panel at 1,000 cd/m2 where red and green lines present the horizontal and vertical cross-sections to measure the luminance distribution (b) Normalized horizontal luminance distribution (c) Normalized vertical luminance distribution.
In the end, the luminance uniformity is simulated at ~1,000 cd/m2 concerning the size of a panel which has equal horizontal and vertical lengths with high conduction pads. Without auxiliary metal electrodes, Fig. 8 ensures that a 3-tandem panel is allowed to be enlarged up to 70 × 70 mm2 with the smaller non-uniformity than 10% while a single-EML panel can be increased the size of panel only up to 40 × 40 mm2.
Fig. 8 Estimated non-uniformity plot according to the emission area of square shape panels at ~1,000 cd/m2.
4. Conclusion
In this paper, three single-EML, 2-tandem, and 3-tandem OLED lighting panels are investigated from a perspective of luminance uniformity by circuit model simulation and measurement. A 3-tandem structure achieves the best luminance uniformity compared to two other structures due to high luminance-current efficiency. As the highly efficient device can support high luminance at the low current density, low voltage and luminance variations over the area are established. Consequently, a 3-tandem approach is expected to be a promising way for high performance OLED lighting applications because of outstanding luminance uniformity, high efficiency, and long lifetime. Furthermore, the simulation results ensure that a 3-tandem panel of about 5,000 mm2 emission area can be fabricated with lower non-uniformity than 10% without auxiliary metal electrodes.
Acknowledgment
This work was partly 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.
References and links
1. K. Yamae, H. Tsuji, V. Kittichungchit, N. Ide, and T. Komoda, “Highly efficient white organic light-emitting diodes with over 100 lm/W for next-generation solid-state lighting,” Soc. Inform. Display 21(12), 529–540 (2013). [CrossRef]
2. J. P. Spindler and T. K. Hatwar, “Fluirescent-based tandem white OLEDs designed for display and solid-state-lighting applications,” Soc. Inform. Display 17(10), 861 (2009). [CrossRef]
3. T. Komoda, N. Ide, K. Varutt, K. Yamae, H. Tsuji, and Y. Matsuhisa, “High-performance white OLEDs with high color-rendering index for next-generation solid-state lighting,” J. Soc. Inf. Disp. 19(11), 838 (2011). [CrossRef]
4. G. Gu, P. E. Burrows, S. Venkatesh, S. R. Forrest, and M. E. Thompson, “Vacuum-deposited, nonpolymeric flexible organic light-emitting devices,” Opt. Lett. 22(3), 172–174 (1997). [CrossRef] [PubMed]
5. A. B. Chwang, M. A. Rothman, S. Y. Mao, R. H. Hewitt, M. S. Weaver, J. A. Silvernail, K. Rajan, M. Hack, J. J. Brown, X. Chu, L. Moro, T. Krajewski, and N. Rutherford, “Thin film encapsulated flexible organic electroluminescent displays,” Appl. Phys. Lett. 83(3), 413 (2003). [CrossRef]
6. L. Duan, S. Liu, D. Zhang, J. Qiao, G. Dong, L. Wang, and Y. Qiu, “Improved flexibility of flexible organic light-emitting devices by using a metal/organic multilayer cathode,” J. Phys. D Appl. Phys. 42(7), 075103 (2009). [CrossRef]
7. G. Gu, V. Bulović, P. E. Burrows, S. R. Forrest, and M. E. Thompson, “Transparent organic light emitting devices,” Appl. Phys. Lett. 68(19), 2606 (1996). [CrossRef]
8. B. W. D’Andrade and S. R. Forrest, “White organic light-emitting devices for solid-state lighting,” Adv. Mater. 16, 18 (2004).
9. M. C. Gather, A. Köhnen, and K. Meerholz, “White organic light-emitting diodes,” Adv. Mater. 23(2), 233–248 (2011). [CrossRef] [PubMed]
10. K. Yamae, V. Kittichungchit, N. Ide, M. Ota, and T. Komoda, “Realization of high efficacy white organic light emitting diode by controlling internal light absorption and angular distribution,” J. SID 22, 15 (2015).
11. K. Neyts, M. Marescaux, A. U. Nieto, A. Elschner, W. Lövenich, K. Fehse, Q. Huang, K. Walzer, and K. Leo, “Inhomogeneous luminance in organic light emitting diodes related to electrode resistivity,” J. Appl. Phys. 100(11), 114513 (2006). [CrossRef]
12. R. Ma, P. A. Levermore, H. Pang, P. Mandlik, K. Rajan, J. Silvernail, M. Hack, and J. J. Brown, “66.4: invited paper: challenges and opportunities in scaling up OLED lighting devices,” SID Symp. Dig. Tech. Papers 42, 983 (2011). [CrossRef]
13. J. W. Park, D. C. Shin, and S. H. Park, “Large-area OLED lightings and their applications,” Semicond. Sci. Technol. 26(3), 034002 (2011). [CrossRef]
14. J. Park, J. Lee, and Y. Y. Noh, “Optical and thermal properties of large-area OLED lightings with metallic grids,” Org. Electron. 13(1), 184–194 (2012). [CrossRef]
15. K. Neyts, A. Real, M. Marescaux, S. Mladenovski, and J. Beeckman, “Conductor grid optimization for luminance loss reduction in organic light emitting diodes,” J. Appl. Phys. 103(9), 093113 (2008). [CrossRef]
16. M. Slawinski, M. Weingarten, M. Heuken, A. Vescan, and H. Kalisch, “Investigation of large-area OLED devices with various grid geometries,” Org. Electron. 14(10), 2387–2391 (2013). [CrossRef]
17. M. Slawinski, D. Bertram, M. Heuken, H. Kalisch, and A. Vescan, “Electrothermal characterization of large-area organic light-emitting diodes employing finite-element simulation,” Org. Electron. 12(8), 1399–1405 (2011). [CrossRef]
18. L. Pohl, E. Kollá, A. Poppe, and Z. Kohári, “Nonlinear electro-thermal modeling and field-simulation of OLEDs for lighting applications I: Algorithmic fundamentals,” Microelectronics J. 43(9), 624–632 (2012). [CrossRef]
19. Z. Kohári, E. Kollá, L. Pohl, and A. Poppe, “Nonlinear electro-thermal modeling and field-simulation of OLEDs for lighting applications II: Luminosity and failure analysis,” Microelectronics J. 44(11), 1011–1018 (2013). [CrossRef]
20. L. Li, N. Lu, M. Liu, and H. Bässler, “General Einstein relation model in disordered organic semiconductors under quasiequilibrium,” Phys. Rev. B 90(21), 214107 (2014). [CrossRef]
