Available online at www.sciencedirect.com
Thin Solid Films 516 (2008) 3614 – 3617 www.elsevier.com/locate/tsf
Single dopant white electrophosphorescent light emitting diodes using heteroleptic tris-cyclometalated Iridium(III) complexes Ji Hyun Seo a,b , In Joon Kim a,b , Young Kwan Kim a,b , Young Sik Kim a,b,⁎ a
b
Department of Information Display Engineering, Hongik University, Seoul, 121-791, Republic of Korea Center for Organic Materials and Information devices, Hongik University, Seoul, 121-791, Republic of Korea Available online 23 August 2007
Abstract We report single dopant single emissive layer white organic electroluminescent (EL) device based on the heteroleptic tris-cyclometalated iridium(III) complex, Ir(dfppy)2(pq), as the guest, where dfppy and pq are 2-(2,4-difluorophenyl) pyridine and 2-phenylquinoline, respectively, and 1,4-phenylenesis(triphenylsilane) (UGH2) as the host. The maximum luminous and power efficiencies of the device were 11.00 cd/A (J = 0.05 mA/cm2) and 5.60 lm/W (J = 0.001 mA/cm2), respectively. The CIE coordinates of the device with Ir(dfppy)2(pq) are (0.443, 0.473) and the EL spectrum of device shows emission band at 473 and 544 nm, at the applied voltage of 12 V. The similar phosphorescent decay rate of two ligands can lead to emit luminescence in two ligands at the same time. © 2007 Elsevier B.V. All rights reserved. Keywords: OLED; White phosphorescent material; Heteroleptic ligands; Ir(dfppy)2(pq)
1. Introduction White organic light-emitting devices (WOLEDs) have been attracting much interest. In the last few years, white-light emission from organic compounds has been the subject of increasing interest due to its potential impact on the lighting industry and backlight applications. Various strategies to fabricate WOLEDs include the manufacture of multilayer OLEDs by consecutive evaporation, and that of the single layer polymer blend devices, where all the emitting components are mixed in one layer [1–6]. However, in both devices, the emission color is sensitive to the device structure such as layer thickness and doping concentration. Also, this technique requires complex technological processes and a large amount of wasted organic materials, resulting in relatively high fabrication cost. Spin-coating of a blend of different soluble emitters in a single layer seems to be a more cost-effective technique [7,8]. Though cheaper, this approach has the drawback that customized color combinations are not always possible due to Forster transfer from the high energy emitting material (donor) to the low-energy one (acceptor), which ⁎ Corresponding author. Tel.: +82 2 320 1607; fax: +82 2 3142 0335. E-mail address:
[email protected] (Y.S. Kim). 0040-6090/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2007.08.082
induces emission only from the lower-gap compound [9,10]. An alternative approach which overcomes such a problem is to blend two blue-light-emitting organic molecules of different electron affinities, whose interaction gives rise to exciplex states [11,12]. The combination of the exciplex emission with the blue-light emission of the individual donor molecule results in the generation of white light. However, in both of these approaches the purity of the color emission is strongly dependent on the relative concentration of the different molecular species and, generally, on the applied voltage. This is a problem for lighting applications in which the source intensity (but not the color) has to be varied by changing the applied electrical power. Furthermore, the usage of charge blockers to confine the charges and excitons causes high driving voltage and low efficiencies in multilayer white devices, colors are dependent on the driving voltage in the single layer polymer devices. Most of the problems seem avoidable if a singlecomponent material can be used as the emitting species. Although a few single emitting component WOLEDs have been reported in which white electroluminescence comes from the individual lumophore and excimer (or electromer), no single emitting component WOLED has yet been reported. In this paper, we demonstrated single dopant single emissive layer WOLED based on heteroleptic tris-cyclometalated iridium
J.H. Seo et al. / Thin Solid Films 516 (2008) 3614–3617
3615
Fig. 1. Molecular structure of Ir(dfppy)2(pq) and device structure using the Ir(dfppy)2(pq) as the white dopant.
(III) complex, Ir(dfppy)2(pq), where dfppy and pq are 2-(2,4difluorophenyl) pyridine and 2-phenylquinoline, respectively, and 1,4-phenylenesis (triphenylsilane) (UGH2) as the host. It emits white color from the single dopant in single emissive layer. The excitons formed on pq and dfppy ligands and they emit red and blue color at once, respectively, without intermolecular energy transfer because of their similar decay lifetime and stronger MLCT characteristics of pq ligand than that of dfppy ligand. Emission wavelengths and the lifetime of Ir(dfppy)2(acac) and Ir(pq)2(acac) (acac = acetylacetonate) were reported as 469 nm (1.6 μs) and 597 nm (2.0 μs), respectively [13,14]. We have fabricated OLED using the Ir(dfppy)2(pq) as the single white dopant and UGH2 as the host, then the electrical and chemical characteristics were studied. 2. Experiment 2.1. Theoretical calculation To investigate the suggested luminescent mechanism, we have prepared and theoretically calculated the heteroleptic triscyclometalated Ir(III) complexes having two different ligands. Density functional theory (DFT) of B3LYP with the LANL2DZ and 6–31G(d) basis sets for iridium and the other atoms, respectively, was used for the geometry optimization and the energy level calculation of the ground state of the complex. To obtain the vertical excitation energies of the low-lying singlet and triplet excited states of the complexes, time-dependent DFT (TD-DFT) calculations using the B3LYP functional was performed at the respective ground-state geometry, where the basis set of ligands was changed to 6–31 + G(d).
2.3. Device fabrication OLED using Ir(dfppy)2(pq) as white dopant in emitting layers was fabricated. Other organic materials used as carrier transport, carrier injection and host materials were supplied by Gracel Display Incorporation in Korea. OLEDs were fabricated by high vacuum (5 × 10− 7 Torr) thermal deposition of organic materials onto the surface of an indium tin oxide (ITO, 30 Ω/?, 80 nm) coated glass substrate. The ITO glass was cleaned with acetone, methanol, distillated water and isopropyl alcohol[15]. The organic materials were deposited in the following sequence: 60 nm of 4,4′,4″-tris[2naphthylphenylamino]triphenylamine (2-TNATA) and 20 nm of 4,4′-bis[N-(naphthyl)-N-phenyl-amino]biphenyl (NPB) were applied as a hole injection layer (HIL) and a hole transporting layer (HTL), respectively, followed by a 30 nm thick emissive layer (EML) of the Ir complexes doped in 1,4-phenylenesis (triphenylsilane) (UGH2). The doping rate of the phosphor, an Ir complex, was 10%. 10 nm thick bathocuproine (BCP), 20 nm thick tris-(8-hydroxyquinoline) aluminum (Alq3) and 2 nm
2.2. UV-absorption and photoluminescence (PL) measurement UV–Vis absorption spectra were measured on Hewlett Packard 8425A spectrometer. The PL spectra were obtained on Perkin Elmer LS 50B spectrometer. UV–Vis and PL spectra of Ir(dfppy)2(pq) were measured with a 10− 5 M dilute solution in CH2Cl2.
Fig. 2. Calculated HOMO and LUMO energy levels of Ir(ppy)3, Ir(dfppy)3 and Ir(pq)3.
3616
J.H. Seo et al. / Thin Solid Films 516 (2008) 3614–3617
thick Lithium quinolate (Liq) were deposited as an exciton blocking layer, as an electron transporting layer (ETL) and as an electron injection layer (EIL), respectively. The typical organic deposition rate was 0.1 nm/s. Finally, 100 nm of Al was deposited as a cathode. Configuration of device and molecular structure of Ir(dfppy)2(pq) were shown in Fig. 1. The active area of the OLEDs was 0.09 cm2. After the fabrication, the current density–voltage (J–V) characteristics of the OLEDs were measured with a source measure unit (Kiethley 236). The luminance and CIE chromaticity coordinates of the fabricated devices were measured using a chromameter (MINOLTA CS100A). All measurements were performed in ambient conditions under DC voltage bias. 3. Results and discussion Fig. 4. Luminous efficiency and power efficiency curves as increasing the current density of OLED.
In order to emit white color and to improve the luminescence efficiency by avoiding T–T annihilation, new phosphorescent iridium complexes having a different species of plural ligands are designed for the application in WOLEDs. The iridium complexes prepared herein can be classified into two groups. The first group includes the iridium complexes having two different C^N ligand structures (heteroleptic complex) such as Ir (dfppy)2(pq) and the second one involves the iridium complexes having only one kind of C^N ligand (homoleptic complexes) such as Ir(dfppy)3 and Ir(pq)3. In homoleptic complex cases, luminescence efficiency may decrease because of the saturated quenching effect caused by the energy transfer between the same species of ligands. Thus heteroleptic Ir(III) complex, Ir (dfppy)2(pq), have higher luminescence efficiency than that of homoleptic Ir(III) complexes, Ir(dfppy)3 and Ir(pq)3. The calculated highest occupied molecular orbitals (HOMOs) and the lowest unoccupied molecular orbitals (LUMOs) energy levels of the ground state of Ir(ppy)3, Ir(dfppy)3 and Ir(pq)3 were shown in Fig. 2. Iridium complex with ppy ligands was known as good metal to ligand charge transfer (MLCT) material because the π energy level of ppy ligand and the t2g energy level of central iridium atom are very close to induce the mixing of molecular
orbitals between central metal and ligand. Comparing HOMOs and LUMOs of Ir(dfppy)3 and Ir(pq)3 with those of Ir(ppy)3, we confirmed that the HOMO energy level of Ir(pq)3 are similar to that of Ir(ppy)3. Thus pq ligand has stronger MLCT characteristics than dfppy ligand. When Ir(dfppy)2(pq) was placed in an excited state, the absorbed energy may then be thought to emit each ligand without energy transfer because of their similar phosphorescent lifetime and stronger MLCT characteristics of pq ligand than that of dfppy ligand. Thus, white color combined the blue and orange emission emitted at once can be observed. To realize this idea, we have fabricated the OLED with iridium complex as white dopant. Fig. 3 shows the current densities of the device as a function of the applied voltage, and the figure inserted in Fig. 3 shows the luminance of the device as increase of the current density. The current density of the device was 21.20 mA/cm2 at 14 V when luminance of device was 1000 cd/m2. These results exhibit the low luminance due to low current density, because the host of ultra high band gap suppressed the injected carriers
Fig. 3. Current density curve as increasing the voltage. Inset: luminance curve as increasing the current density of OLED.
Fig. 5. Normalized electroluminescence (EL) spectrum of OLED using Ir complex as white dopant, when applied voltages are 8, 10, 12, and 14 V. Inset: CIE coordinates curves as increasing the current density of OLEDs.
J.H. Seo et al. / Thin Solid Films 516 (2008) 3614–3617
into the emissive layer. However, host having the high band gap such as 1,4-phenylenesis(triphenylsilane) (UGH2) is required to obtain the phosphorescent blue emission of Ir(dfppy)2(pq) in device. The luminous efficiencies and power efficiencies of the devices is shown in Fig. 4, as a function of the applied current densities. The maximum luminous and power efficiencies of the device were 11.00 cd/A (J = 0.05 mA/cm2) and 5.60 lm/W (J = 0.001 mA/cm2), respectively, when device was applied from 0 V to 14 V. Fig. 5 and inset in Fig. 5 show the normalized electroluminescence (EL) spectrum and CIE coordinates. The CIE coordinates of the device with Ir(dfppy)2(pq) are (0.443, 0.473) and the EL spectrum of device shows emission band at 473 and 544 nm, at the applied voltage of 12 V. We note that there is a little current density dependence of the EL spectrum and CIE coordinates from about 1 mA/cm2 (8 V) to 20 mA/cm2 (14 V). The small variations in the EL spectrum and CIE coordinates with the applied voltage is because at the higher applied voltage (i.e. at higher current density) more charge trapping occurs on the dfppy ligand site and direct exciton formation on the site. It is because the triplet energy of the pq ligand site is lower than that of the dfppy ligand site and it is saturated in the low applied voltage. However, the variations of emission color with the applied voltage were subtle as shown in CIE coordinates inserted in Fig. 5. The CIE coordinates result is not good for white color because the emission peak of pq becomes blue shifted. The reason is that the fluorinated dfppy ligands of Ir(dfppy)2(pq) having an electron withdrawing characteristics decrease t2g energy level of central metal atom, it increases the energy gap of Ir(dfppy)2(pq). In result, the pq ligand emitting peak of Ir(dfppy)2(pq) has more blue shifted emission wavelength than one of Ir(pq)3. Thus the problem may solve by replacing with material having more reddish emission peak than pq ligand in order to emit pure white color. 4. Conclusion We report single dopant single emissive layer white organic EL device based on the heteroleptic tris-cyclometalated iridium
3617
(III) complex, Ir(dfppy)2(pq), as the guest and UGH2 as the host. The maximum luminous and power efficiencies of the device were 11.00 cd/A (J = 0.05 mA/cm2) and 5.60 lm/W (J = 0.001 mA/cm2), respectively. The CIE coordinates of the device with Ir(dfppy)2(pq) are (0.443, 0.473) and the EL spectrum of device shows emission band at 473 and 544 nm, at the applied voltage of 12 V. The similar phosphorescent lifetime of two ligands and stronger MLCT characteristics of pq ligand than that of dfppy ligand can lead to emit luminescence in two ligands at the same time, it was found that Ir(dfppy)2(pq) is possible to be good candidates for white phosphorescent materials, by controlling the lifetime of ligand in Ir complex. Acknowledgment This work was supported by Seoul R&D Program (10555). References [1] [2] [3] [4] [5] [6] [7] [8]
[9] [10] [11] [12] [13]
[14] [15]
G. Li, J. Shinar, Appl. Phys. Lett. 83 (2003) 5359. R.S. Deshpande, V. Bulovic, S.R. Forrest, Appl. Phys. Lett. 75 (1999) 888. C.H. Kim, J. Shinar, Appl. Phys. Lett. 80 (2002) 2201. Y.S. Huang, J.H. Jou, W.K. Weng, J.M. Liu, Appl. Phys. Lett. 80 (2002) 2782. K.O. Cheon, J. Shinar, Appl. Phys. Lett. 81 (2002) 1738. C.H. Chuen, Y.T. Tao, Appl. Phys. Lett. 81 (2002) 4499. M. Granstrom, O. Inganas, Appl. Phys. Lett. 68 (1996) 147. S. Tasch, E.J.W. List, O. Ekstrom, W. Graupner, G. Leising, P. Schlichting, U. Rohr, Y. Geerts, U. Scherf, K. Mullen, Appl. Phys. Lett. 71 (1997) 2883. T. Forster, Discuss. Faraday Soc. 27 (1959) 7. M. Anni, G. Gigli, V. Paladini, R. Cingolani, G. Barbarella, L. Favaretto, G. Sotgiu, M. Zambianchi, Appl. Phys. Lett. 77 (2000) 2458. J. Thompson, R.I.R. Blyth, M. Mazzeo, M. Anni, G. Gigli, R. Cingolani, Appl. Phys. Lett. 79 (2001) 560. M. Mazzeo, D. Pisignano, F. Della Sala, J. Thompson, R.I.R. Blyth, G. Gigli, R. Cingolani, L. Favaretto, G. Barbarella, Appl. Phys. Lett. 82 (2003) 334. S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq, R. Kwong, I. Tsyba, M. bortz, B. Mui, R. Bau, M.E. Thompson, Inorg. Chem. 40 (2001) 1704. K. Dedeian, J. Shi, N. Shepherd, E. Forsythe, D. Morton, Inorg. Chem. 44 (2005) 4445. T.P. Nguyen, L.P. Rendu, N.N. Dinh, Synth. Met. 138 (2003) 229.