In sharp contrast to the rapid progress being made on organic LED (OLED) display technologies, the prospects for OLED lighting are much less optimistic. The slower development of this field can be attributed to the immature technologies and organic materials available for architecture design, especially for light extraction. We simplified the device architecture, reducing the number of thin-film layers and incorporating nanosized particles to improve the overall efficiency and light output of blue phosphorescent OLEDs.
For high-quality illumination, OLEDs must be operated at a high current density to ensure an ample flux of light. This requires driving the OLEDs at a high bias, which limits a device's efficiency and shortens its life span. Because the external quantum efficiency of a lamp depends both on the internal quantum efficiency and the light-extraction efficiency, the challenge to designers of OLEDs for lighting is to find both an architecture and materials that are internally efficient under these conditions and to optimize light extraction.1
The internal efficiency of an OLED depends on charge balance, quantum yield, exciton confinement, and other factors, and can generally be optimized by careful structure design. Phosphorescent OLEDs are attractive because they exploit both singlet and triplet excited states, thus raising the limits of internal efficiency from about 25% to theoretical limits of near 100%. Most phosphorescent OLED designs use many thin-film layers to keep the charge carriers balanced. However, the additional layers reduce the manufacturability of the devices.
We designed and built a phosphorescent OLED with only three layers between the electrodes: a hole-transport layer (HTL), an emissive layer (EML), and an electron-transport layer (ETL).2 We also altered the structure near the anode to help light escape from the device. Extracting light is a congenital problem of OLEDs: Fresnel loss and total internal reflection (TIR) inside the stacked thin-film structures results in out-coupling efficiencies that reach only about 20%. Our device uses a layer incorporating two sizes of nanoparticles (NPs), which smooth changes in the refractive index, reducing reflection and thus boosting the amount of light escaping the device.
Given the design criteria, we selected two materials as host and emitter in the EML: 1,3-bis(9-carbazolyl)benzene (mCP) combined with iridium(III) bis[(4,6-di-fluorophenyl)-pyridinato-N,C2 ′]picolinate (FIrpic). These substances have previously shown a photoluminescent quantum yield of nearly 100%. In addition, for the HTL we chose di-[4-(N,N-ditolyl-amino)-phenyl]cyclohexane (TAPC) and for the ETL 1,3,5-tri[(3-pyridyl)-phen-3-yl]benzene (TmPyPB). We made a simplified tri-layer blue OLED consisting of an indium tin oxide (ITO) anode, a 30nm layer of TAPC to transport holes, a 20nm-thick emissive layer of mCP doped with 8wt% FIrpic, and then the 50nm-thick TmPyPB ETL followed by 0.8nm of lithium fluoride and 150nm of aluminum that act as the electron-injection layer and cathode, respectively. Figure 1 shows the structural drawing of the materials and the modified device.
Figure 1. Molecules (left) used in two simplified organic LED (OLED) device architectures (right). Device layers from the top down: Lithium fluoride is used as an electron-injection material and aluminum is used as the cathode. TmPyPB is the electron-transport material. The light-emitting layer uses mCP as a wide-energy-gap host material doped with 8wt% of blue emitter FIrpic. TAPC is the hole-transport material. PEDOT:PSS is a conductive transparent polymer that sits between the hole-transport layer and the indium tin oxide (ITO) anode. The extra layer in Device A consists of two sizes of nanoparticles (NPs) embedded in a layer of transparent photoresist.
The measured electroluminescence (EL) characteristics show that these devices reached high efficiencies of 20.3% (47.6lm/W) and 19.2% (24.8lm/W) at 100 and 1000cd/m2. Such high external quantum efficiencies imply an internal quantum efficiency near 100%. Given the energy-level relationship, we expect the energy barrier to block the holes at the HTL/EML interface, thereby deterring hole transport and preventing excessive hole injection into the emitting region. Furthermore, the carrier balance was achieved in a range from 0.01 to 1mA/cm2. We believe that the satisfactory performance of the tri-layer architecture is likely due to the stepwise hole injection. The reduced hole accumulation could weaken trapped charge-induced quenching at the anode/EML interface.
To further improve performance, we introduced light-extraction technology. In 2008, Yiru Sun and Stephen R. Forrest showed that internal extraction structures (IESs) can be introduced in OLEDs to decrease the ratio of the light in the waveguide mode and simultaneously increase the ratios of light in the substrate and radiation modes.3 Recently, C.-C. Wu's group developed an IES method using internal scattering layers that resulted in a nearly twofold improvement in efficiency.4 Following this promising technique, we combined 250nm titanium oxide (TiO2) NPs with transparent photoresist (TPR) to form an embedded nanocomposite scattering layer between the ITO and the glass substrate.
On closer inspection, however, we realized that when an additional TPR film was inserted between the electrode and the substrate, the difference in the refractive index caused severe internal reflection. Fortunately, Long-Hua Lee and Wen-Chang Chen showed that the refractive index of the TPR film can be adjusted by mixing in NPs with higher refractive indices.5 Thus, once we considered the possible wavelength ranges of the OLED emissions, we added smaller NPs, only 25nm in diameter, to the TPR to avoid abrupt drops in the forward transmittance. We recorded the transmittance and refractive index of pure TPR at 90.6% and 1.52, respectively. By increasing the concentration of NPs in the TPR films to 18wt%, the refractive index gradually increased to 1.92 with ∼85% transmittance, equal to ITO and thus mitigating the reflection at the ITO/TPR interfaces. Moreover, by comparing scanning electron microscope and atomic force microscope images of single-sized and dual-sized mixed TiO2 films, we saw that adding 25nm NPs in an appropriate concentration has the advantage of flattening the surface of the nanocomposite film.6
In our study we compared substrates without the NP layer (Type-R: glass and ITO) and with the layer (Type-A: glass and layer of TPR containing both 25 and 250nm NPs, then ITO). For both substrates, we used a layer of the conductive polymer PEDOT:PSS, which facilitates planarization of the ITO substrate, especially those with nanocomposite films. Figure 1 shows a structural drawing of the materials and the modified device.
Figure 2. Electroluminescence (EL) characteristics are compared for blue phosphorescent OLEDs with different substrates. (a) Normalized EL spectra. (b) External quantum efficiency versus luminance of Devices R and A. cd/m2: Brightness unit. a.u.: Arbitrary units.
Figure 2 shows the EL characteristics of blue phosphorescent OLEDs with different substrates. We drove the devices hard, increasing the luminescence to 5×103cd/m2 to see whether they could sustain their efficiency. This reduced the efficiency of Device R to 7.5% and 5.7lm/W. But Device A, whose substrate contained dual-sized NPs that provided scattering as well as refractive index matching functions, exhibited efficiencies of up to 23.0% and 24.5lm/W, confirming the superiority of the Type-A substrates over the pristine ITO substrate.
In summary, we developed a simplified architecture and improved light-extraction design that we estimate could improve blue phosphorescent OLED power efficiency by about 4.3 times. Our next project will integrate light-extraction structures with tandem structures, in which individual electroluminescent units connect via charge-generation layers. We expect this will multiply the efficiency and luminance of multiple units. We believe that devices with a tandem structure including a nanocomposite substrate with a graded refractive index will improve efficiency for practical OLED lighting.
The authors gratefully acknowledge financial support from the National Science Council of Taiwan (NSC 99-2221-E-155-035-MY3) and the Ministry of Economic Affairs (100-EC-17-A-08-S1-042).
Chih-Hao Chang, Kuo-Yan Chang, Yu-Jhong Lo, Tzu-Fang Chang
Yuan Ze University
Chih-Hao Chang is an assistant professor in the Department of Photonics Engineering. His research interests include organic optoelectronic and electronic devices, flat-panel displays, and solid-state lighting.
1. Z. Liu, M. G. Helander, Z. Wang, Z. Lu, Efficient single layer RGB phosphorescent organic light-emitting diodes, Org. Electron. 10, p. 1146-1151, 2009.
2. C.-H. Chang, Y.-S. Ding, P.-W. Hsieh, C.-P. Chang, W.-C. Lin, H.-H. Chang, Highly efficient phosphorescent blue and white OLEDs with simplified architectures, Thin Solid Films 519, p. 7992-7997, 2011.
3. Y. Sun, S. R. Forrest, Enhanced light out-coupling of organic light-emitting devices using embedded low-index grids, Nat. Photon. 2, p. 483-487, 2008.
4. H.-W. Chang, K.-C. Tien, M.-H. Hsu, Y.-H. Huang, M.-S. Lin, C.-H. Tsai, Y.-T. Tsai, C.-C. Wu, Organic light-emitting devices integrated with internal scattering layers for enhancing optical out-coupling, J. Soc. Inf. Display 19, p. 196-204, 2011.
5. L.-H. Lee, W.-C. Chen, High-refractive-index thin films prepared from trialkoxysilane-capped poly(methyl methacrylate)-titania materials, Chem. Mater. 13, p. 1137-1142, 2001.
6. C.-H. Chang, K.-Y. Chang, Y.-J. Lo, S.-J. Chang, H.-H. Chang, Four-fold power efficiency improvement in organic light-emitting devices using an embedded nanocomposite scattering layer, Org. Electron. 13, p. 1073-1080, 2012.