The OLED structures used in this experiment and energy level diagrams. 

Contexts in source publication

Context 1

... attributed to the balanced charge carrier conduction in R/G EML sequence since a uniform electron–hole recombination zone can be formed over a wide range of driving current density. In contrast, the main recombination zone, which is positioned relatively towards the R layer due to its lower electron mobility, is shifted towards the G layer in the devices with G/R EML sequence at higher driving current density due to the increased electron conduction towards the G layer by increased electric field. Thus, relative intensity of green emission increases at high current density, resulting in large shift of the chromaticity coordinates for devices of W-2 and W-4 as shown in the inset of figure 2. Figure 4 shows the efficiencies of the four white devices as functions of the luminance. The highest efficiency was found to be that of W-3 as expected. In addition, W-1 exhibits a higher efficiency than W-2 (which does not have an interlayer) and a similar efficiency to W-4. This confirms that the phosphorescent R/G EML sequence results in improved white OLED efficiency because of better balance of charge carrier conduction between the EMLs. The DPAVBi molecules in the CBP layer are expected to act mainly as hole traps from the energy level diagram shown in figure 1 but little effect on the electron conduction properties of the CBP host is expected because their LUMO levels are almost the same. Thus the higher efficiency of the white OLEDs with the R/G EML sequence can be explained as discussed above. The characteristics of the four white devices, such as their chromaticity coordinates, maximum EQE and current efficiency, are summarized in detail in table 1. Furthermore, a similar trend is expected when btp 2 Ir(acac) is replaced with other red phosphorescent dopants with higher efficiencies such as pq 2 Ir(acac) [9] and tris[1-phenyliso quinolinato- C2,N] iridium(III) (Ir(piq) 3 ) [18] because of the similarity of their energy levels. Accelerated lifetime tests were conducted for the four white devices capped with a glass lid in an Ar-filled glove box. A constant current bias stress corresponding to an initial luminance of 1000 cd m − 2 was applied. The relative luminances are shown as functions of the operating time in figure 5. The device with a R/G EML sequence was found to have longer half-lifetimes (the time for the luminance to fall by 50%), with W-3 exhibiting the best characteristics. The half-lifetimes of the white devices are proportional to their efficiencies (at the initial luminance are obtained) as generally believed: higher efficiency leads to the longer lifetime of devices when the same EL materials are used. The normalized EL spectrum of the W-3 device after the luminance had reached 40% of its initial value is shown in figure 6 and the inset shows the shifts in the chromaticity coordinates of the four white devices before and after the lifetime tests at a driving current density of 25 mA cm − 2 . The decreases in the blue emission were the most significant and common to all four devices (there were small decreases in green emission relative to red emission) even though we used a fluorescent blue dopant. This kind of change in the EL spectrum is expected from the general development status of organic light emitting materials. The device with the R/G EML sequence was found to exhibit better characteristics, with a smaller decrease in the luminance and a smaller chromaticity coordinate shift despite the rapid decrease in the blue emission. Therefore, the individual lifetime of each primary colour dopant has to be considered together with overall device structure to obtain white devices with truly longer lifetimes, which has a very small chromaticity coordinate shift with the change in driving current density and operating time as well. We have designed white OLEDs in which CBP layers doped with the phosphorescent dopants Ir(ppy) and btp Ir(acac), and the fluorescent dopant DPAVBi, were used as G, R and B EMLs, respectively. Significantly higher (at least 2 orders of magnitude higher) electron conduction was predicted in the G EML than in the R EML, with rather higher hole conduction predicted for the R EML than for the G EML. Thus, white OLEDs with an R/G/B EML sequence were expected to achieve both better balanced charge carrier conduction properties and more balanced white emission. Higher efficiencies, stable chromaticity coordinates and longer lifetimes were obtained for white OLEDs with a R/G/B EML sequence when compared with those with a G/R/B EML sequence. The best performance was obtained with white OLEDs with an interlayer between the G phosphorescent and B fluorescent EMLs, with the highest efficiency of 18.3 cd A − 1 (corresponding to an EQE of 8.5%) at 100 cd m − 2 . This work was supported by the Samsung SDI - Seoul National University Display Innovation ...

Context 2

... organic light emitting diodes (OLEDs) are studied because of their applications in solid state lighting [1], full- colour display panels combined with colour filter arrays [2] and sheet backlight units for liquid crystal displays. Studies of white OLEDs in which both phosphorescent and fluorescent dopants are used have recently been carried out, and have reported very high efficiencies [3–5]. However, their structures are somewhat complicated. In these multi-layer white OLEDs, achieving charge balance [6–8] is more difficult than in single-layer OLEDs because each emitting layer (EML) has a different dopant, which is expected to result in different electrical conduction properties even when the same host material is used, and this is especially true for EMLs doped with phosphorescent dopants, in which a high dopant concentration is usually required for high efficiencies to be achieved. Chin et al [9] demonstrated the presence of biased (non-centred) recombination zones in uniformly doped EMLs by analysing the characteristics of graded doped OLEDs. They showed that the recombination zone in a phosphorescent green OLED device with an EML consisting of CBP doped with 7% Ir(ppy) 3 is formed adjacent to the hole transport layer (HTL)/EML interface whereas that of a phosphorescent red OLED device with CBP doped with iridium (III) bis(1-phenyl[quinolinato- N , C 2 ]) acetyl- acetonate (pq 2 Ir(acac)) is formed adjacent to the electron transport layer (ETL). In addition, the charge carrier mobilities of Ir(ppy) 3 doped and btp 2 Ir(acac)-doped CBP layers have been reported by Matsusue et al [10, 11]: the orders of magnitude of the hole and electron mobilities of non-doped CBP layers are 10 − 3 cm 2 V − 1 s − 1 and 10 − 4 cm 2 V − 1 s − 1 , respectively. The electron mobility of a CBP layer doped with 7% Ir(ppy) 3 was found to be nearly the same as that of the non-doped layer, whereas a decrease in the electron mobility of the CBP layer doped with 7% btp 2 Ir(acac) of at least two orders of magnitude was observed. Because of such difference in the electrical conduction properties between the Ir(ppy) 3 - and btp 2 Ir(acac)- doped CBP layers, their position and doping profile in the EMLs of white OLEDs can significantly affect the spectral shape and the electroluminescence (EL) efficiency. Therefore, optimal design of white OLEDs with stable chromaticity coordinates and high EL efficiency requires careful control of the position and doping concentration of the Ir(ppy) 3 - and btp 2 Ir(acac)-doped CBP layers. In this study, we systematically studied electrical conduction properties of the Ir(ppy) 3 - and btp 2 Ir(acac)-doped CBP layers and, by using these results, optimized the sequence of three primary colour EMLs of the multi-layer white OLEDs in which CBP layers doped with the phosphorescent dopants Ir(ppy) 3 and btp 2 Ir(acac), and the fluorescent dopant DPAVBi, were used as green (G), red (R) and blue (B) EMLs, respectively. Furthermore, we discuss the effect of the R, G and B EML sequences on the accelerated lifetime of devices. We prepared two types of devices: devices for comparison of the hole conduction properties of Ir(ppy) 3 and btp 2 Ir(acac)-doped CBP layers, and devices for comparison of OLED characteristics. The structure of the hole-only devices is as follows: indium-tin- oxide (ITO) anode/4,4 -bis[N-(1-naphthyl)-N-phenyl-amino] biphenyl ( α -NPD, 10 nm)/CBP doped with 6% Ir(ppy) 3 or btp 2 Ir(acac) (100 nm)/Au cathode (100 nm). α -NPD was used as a hole injection layer (HIL) because the energy difference between the ITO work function ( ∼ 4 . 7 eV) and the highest occupied molecular orbital (HOMO) level of CBP ( ∼ 5 . 9 eV) is too high, and the Au cathode ( ∼ 5 . 2 eV) was used to block electron injection through the lowest unoccupied molecular orbital (LUMO) level of CBP ( ∼ 2 . 5 eV). The basic OLED device structure and the energy levels of the materials used in this study are shown in figure 1. The OLED device structure we used is as follows: ITO anode/poly(3,4-ethylenedioxy-thiophene) : poly(styrene sulfonic acid) (PEDOT : PSS, 40 nm) (HIL)/ α -NPD (50 nm) (HTL)/various sequences of the primary colour EMLs/2,9-dimethyl-4,7 diphenyl-1,10-phenanthroline (BCP, 10 nm) hole blocking layer (HBL)/tris(8-hydroxyquinolinato)- aluminium (Alq , 40 nm) (ETL)/LiF (0.5 nm)/Al (100 nm) cathode. The energy levels of these materials were obtained from the literature [9, 11–13]. All layers were prepared with thermal evaporation onto UV-O 3 treated ITO substrates except the PEDOT : PSS layers, which were prepared by spin-coating at 4000 rpm and drying under vacuum of about 10 − 3 Torr for 30 min. The PEDOT : PSS solution was used as purchased (Baytron P, H C Starck GmbH). The substrates were ultrasonically cleaned by dipping them into various solvents such as isopropyl alcohol, acetone and methyl alcohol prior to UV-O 3 treatment. Organic and metal evaporation were conducted under a base pressure of 5 × 10 − 6 Torr without breaking the vacuum, and the evaporation speeds were 1–2 Å s − 1 for the organic materials and 4–5 Å s − 1 for the metals. The doping concentration was adjusted by varying the relative evaporation speeds of the host and dopant materials and the evaporation speeds were monitored with a quartz-oscillator thickness monitor. The current–voltage ( I – V ) characteristics of the hole-only and light emitting devices were measured by using a Keithley-236 source measurement unit, and the luminance and external quantum efficiency (EQE) were calculated from photo-current measurement data obtained with a calibrated Si photo-diode (Hamamatsu S5227-1010BQ). The EL spectra were obtained by using a monochromator (ARC275) combined with a PMT detector. The I – V characteristics of the hole-only devices are shown in figure 2. Significant decreases were found in the current conduction in the CBP layers doped with 6% Ir(ppy) 3 and 6% btp Ir(acac) with respect to that of the pure CBP ...