- Email: [email protected]
Highly efficient tandem organic light-emitting devices adopting a nondoped charge-generation unit and ultrathin emitting layers
Organic Electronics 53 (2018) 353–360
Contents lists available at ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
Highly efficient tandem organic light-emitting devices adopting a nondoped charge-generation unit and ultrathin emitting layers
T
Xinwen Zhanga, Mengke Zhanga, Mengjiao Liua, Yuehua Chena, Jiong Wanga, Xiaolin Zhanga, Junjie Zhanga, Wen-Yong Laia,∗∗, Wei Huanga,b,∗ a
Key Laboratory for Organic Electronics and Information Displays (KLOEID) & Institute of Advanced Materials (IAM), Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing University of Posts & Telecommunications, Nanjing 210023, China b Shaanxi Institute of Flexible Electronics (SIFE), Northwestern Polytechnical University (NPU), 127 West Youyi Road, Xi'an 710072, Shaanxi, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Tandem organic light-emitting devices Charge generation unit Non-doped devices Ultrathin emitting layers
Highly efficient non-doped tandem phosphorescent organic light-emitting devices (PhOLEDs) have been demonstrated by employing LiF/AL/1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN) as a charge generation unit (CGU) and ultrathin phosphorescent dyes as emitting layers (EMLs). It was found that the performance of the tandem PhOLEDs was significantly dependent on the Al thickness in CGU. Analyses regarding the current density-voltage characteristics of the CGU only devices indicate that the charge carriers are generated at the HAT-CN/NPB interface, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent electron transport layer. From the capacitancevoltage and optical transparency characteristics of CGU with different thicknesses of Al interlayer, the thick Al layer is beneficial to charge separation, while it weakens the light output of the tandem device. Under the optimal Al thickness of 5 nm, more than twofold enhancement of current efficiency is achieved for the nondoped tandem blue device. Furthermore, the non-doped tandem white PhOLED was also developed showing a maximum current efficiency of 94.9 cd/A and maximum external quantum efficiency of 31.6%.
1. Introduction Organic light-emitting devices (OLEDs) have attracted widespread attention as next-generation solid state lighting and flat panel displays owing to their excellent properties of flexibility, high efficiency and fast response [1–3]. Up to present, high-performance devices can be achieved by adopting phosphorescent or thermally activated delayed fluorescent emitters due to the full use of both singlet and triplet excitons [4–8]. Nevertheless, the device at high current density usually suffers from the accelerated degradation of lifetime and the roll-off of efficiency [9]. A tandem OLED combines several stacked electroluminescence (EL) units in series through a charge generation unit (CGU), which could achieve both high efficiency and long operating lifetime through reducing the current density in the device [10–13]. Under the applied electrical field, holes and electrons are generated in the CGU, and injected into the adjacent carrier-transporting layers of the individual EL units [12]. Therefore, it is very important for a highperformance tandem OLED to develop an efficient CGU possessing excellent charge generation and separation properties [14].
In recent years, different CGUs with a bilayer structure have been used in tandem OLEDs to enhance the efficiency and stability [15–17], including n-doped electron-transporting layer (n-doped ETL)/metaloxide layer [18], n-doped ETL/organic electron acceptor layer and ndoped ETL/p-doped hole-transporting layer (p-doped HTL) [19]. In these CGUs, the n-doped ETLs composed of Mg, Li, Cs and their compounds doped electron-transporting materials, such as Li:tris(8-hydroxy-quinolinato) aluminium (Alq3) [20], Mg:4,7-diphenyl-1,10-phenanthroline (BPhen) [21], LiNH2:BPhen [22] and Cs2CO3:2,9-dimethyl4,7-diphenyl-1,10-phenanthroline (BCP) [23], and p-doped HTLs composed of 2,3,5,6-tetrafluoro-7,7,8,8-tetracyano-quinodimethane (F4TCNQ), FeCl3, MoO3, WO3 doped electron-transporting materials, such as FeCl3:N,N′-bis(naphthalen-1-yl)- N,N′-bis(phenyl)-benzidine (NPB) [24], F4-TCNQ:4,4′,4″-tris(N-3-methylphenyl-N-phenyl-amino) triphenylamine (m-MTDATA) [25], WO3:NPB [26] and MoO3:4,4′,4″-tris (carbazol-9-yl) triphenylamine (TCTA) [27], are commonly adopted through the complicated co-evaporation technique to achieve effective charge injection from CGUs to adjacent charge-transporting layers of the EL units. According to the recent reports [28,29], diffusion of p-
∗ Corresponding author. Key Laboratory for Organic Electronics and Information Displays (KLOEID) & Institute of Advanced Materials (IAM), Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing University of Posts & Telecommunications, Nanjing 210023, China. ∗∗ Corresponding author. E-mail addresses: [email protected] (W.-Y. Lai), [email protected] (W. Huang).
https://doi.org/10.1016/j.orgel.2017.10.042 Received 10 August 2017; Received in revised form 19 October 2017; Accepted 31 October 2017 Available online 01 November 2017 1566-1199/ © 2017 Published by Elsevier B.V.
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
further purification. In the experiment, the patterned ITO glass substrates were ultrasonically cleaned with detergent, alcohol and acetone, deionized water and then dried at 120 °C in a vacuum oven for more than one hour. After ultraviolet (UV)-ozone treating for 4 min, a 40 nm PEDOT:PSS film was spin coated on the ITO substrate and annealed at 120 °C in for 15 min. The samples were then loaded into a high-vacuum thermal evaporation system. All layers were deposited on top of PEDOT:PSS by thermal evaporation under a pressure of 5 × 10−4 Pa. The thickness of the films was determined in situ by a quartz-crystal sensor. The evaporation rate for the organic layer, LiF, and Al layer in the CGU are 2–3, 0.3–0.5, and 0.5–0.7 Å/s, respectively. The evaporation rate for the phosphorescent dyes and Al cathode are 0.03–0.05 and 3–5 Å/s, respectively. The active emissive area of the devices is 12 mm2. The current-voltage-luminescence characteristics were measured by a Keithley 2602 source meter with a calibrated silicon photodiode. The electroluminescence (EL) spectra of the devices were analyzed with a spectrometer (PR655). Capacitance-voltage (C-V) measurements were performed using an impedance analyzer (Wayne Kerr 6505B) at a constant frequency of 1000 Hz. A 200 mV amplitude AC signal superimposed on a DC bias was used to measure device capacitance as a function of DC bias. The optical transmission and absorption spectra were performed with a Shimadzu UV-3600 spectrophotometer. The ultraviolet–visible–near infrared (UV-VIS-NIR) absorption spectra were measured by a RF-5301PC spectrometer. The photoluminescence (PL) spectra were measured with an excitation at 350 nm using a Shimadzu RF 5301 spectrophotometer. The surface morphology of the films was investigated by atomic force microscopy (AFM, Bruker Dimension Icon). All the devices were characterized without encapsulation, and all the measurements were carried out under ambient condition at room temperature.
dopant and/or n-dopant will cause degradation of the CGU under electrical bias, thus deteriorating device operational stability. Although non-doped organic heterojunctions such as copper hexadecafluorophthalocyanine (F16CuPc)/copper phthalocyanine (CuPc) [30], C60/pentacene [31] and C60/cobalt phthalocyanine (CoPc) [32] are proposed as CGUs in tandem OLEDs, some interfacial modification layers (e.g., p-doped HTLs and n-doped ETLs) have to be introduced to facilitate the charge injection, which will increase the cost and complexity of the devices. Therefore, designing effective non-doped CGU to simplify the fabrication process of tandem OLEDs is very necessary. Recently, the strong electron acceptor materials, such as MoO3 and 1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN), have been reported as the n-type layers to construct non-doped CGU for efficient tandem OLEDs [33–36]. The charge carriers are usually generated at the interface between a strong electron acceptor material and an electron donor material through electron transfer from the donor material to the acceptor material under the application of an external electric field [33–36]. In addition, dye-doped host structures are usually adopted by coevaporation of emitting dyes and host materials in most conventional OLEDs to achieve desirable device performance. However, co-evaporation is a complicated process, especially for white OLEDs, where the doping concentration of dyes must be controlled precisely to obtain balanced white emission. This usually leads to a complicated fabrication process, poor reproducibility and high cost. Recently, non-doped phosphorescent OLEDs (PhOLEDs) by inserting an ultrathin emitting layer (EML) of pure phosphorescent dye between a HTL and an ETL have been introduced to simplify the fabrication process [37–40]. Due to confinement of excitons within ultrathin EML, non-doped PhOLEDs exhibited the maximum external quantum efficiency (EQEmax) about 20%, which was comparable to that of the doped EML devices [37]. Thus, it is a good strategy to develop non-doped PhOLEDs for industrial display and lighting applications. In this paper, we developed highly efficient non-doped tandem PhOLEDs based on ultrathin EMLs and non-doped CGU with the structure of LiF/Al/HAT-CN. We found that the performance of the tandem PhOLEDs was significantly dependent on the Al thickness of CGU. The current density-voltage characteristics of the CGU only devices demonstrate that the charge carriers are generated at the HATCN/NPB interface though disassociation of the charge-transfer complexes formed between HAT-CN and NPB, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent ETL. From the capacitance-voltage and optical transparency characteristics of CGU with different thicknesses of Al interlayer, the thick Al layer is beneficial to charge generation and separation, while it weakens the light output of the tandem device. Under optimal Al thickness of 5 nm, the tandem blue PhOLEDs attain a maximum current efficiency (CEmax) of 53.2 cd/A and a maximum power efficiency (PEmax) of 23.4 lm/W. Furthermore, the tandem white PhOLED presents a CEmax of 94.9 cd/A and EQEmax of 31.6%. The results indicate that ultrathin EML and non-doped CGU can be universally adopted to simplify the fabrication process.
3. Results and discussion To demonstrate efficient tandem OLEDs, it is very important to optimize the structure to efficiently generate charge carriers. We fabricated four tandem devices consisted of LiF (1 nm)/Al (1 nm, 3 nm, 5 nm, 7 nm)/HAT-CN (20 nm) as CGU to investigate the effect of the Al thickness on the device performance, as shown in Fig. 1. The corresponding single unit device with a structure of ITO/PEDOT:PSS (40 nm)/NPB (40 nm)/TCTA (10 nm)/Firpic (0.25 nm)/TmPyPb (40 nm)/LiF (1 nm)/Al (100 nm) was also fabricated for comparison. The single unit device has been optimized through controlling the thickness of ultrathin FIrpic layer. Fig. 2 shows the current densityvoltage, luminance-voltage, current efficiency-luminance, and power efficiency-luminance characteristics of the tandem OLEDs and single unit OLED. The detailed results are summarized in Table 1. It was found that the performance of the tandem devices is strongly dependent on the thickness of Al layer in CGU. As shown in Fig. 2 (a), the current densities of the devices obviously increase with the thickness of Al layer. The driving voltages of the tandem device are 29.6 V (1 nm), 16.2 V (3 nm), 12.1 V (5 nm) and 11.1 V (7 nm), respectively, under a current density of 30 mA/cm2, indicating that the large extra voltage drops in the region of CGU. The strong Al layer thickness dependence of the current density-voltage characteristics indicates that the Al layer plays an important role in charge generation and/or charge injection. From the luminance-voltage characteristics, a similar variation as current density-voltage is also observed. As shown in Fig. 2 (b), the turn-on voltage (the voltage at 1 cd/m2) largely decreases from 12.0 V to 5.7 V as the Al layer thickness increases. Under a luminance of 1000 cd/m2, the driving voltages are 21.0 V (1 nm), 12.3 V (3 nm), 8.5 V (5 nm) and 7.8 V (7 nm), respectively. These results indicate that a thicker Al layer is beneficial to reduction of device driving voltage. Besides, the turn-on voltage of tandem OLED with 7 nm Al layer is around 5.7 V, which is 1.7 times that of the single unit device (3.3 V). At the optimal Al thickness of 5 nm, the tandem device shows a CEmax of 53.1 cd/A at a
2. Experimental The hole-injection material of poly(3,4-ethylenedioxythiophene):poly (styrenesulfonic acid) (PEDOT:PSS, AI4083) was purchased from H. C. Starck Inc. The organic acceptor material of HATCN, the hole transporting materials of N,N′-di(naphth-1-yl)-N,N′-diphenyl-benzidine (NPB) and 4,4′,4″-tris(N-carbazolyl)-triphenylamine (TCTA), the phosphorescent dopants of bis(3,5-difluoro- 2-(2-pyridyl)phenyl-(2-carboxypyridyl)iridium(III) (FIrpic) and bis(4-phenylthieno [3,2-c]pyridine) (acetylacetonate) iridium(III)(PO-01), and the electron transporting material of 1,3,5-tri(m-pyrid-3- yl-phenyl)-benzene (TmPyPB) and Alq3 were purchased from Nichem Fine Technology Co. Ltd. All the above-mentioned materials were used as-received without 354
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 1. (a) Schematic device structures of the single unit and tandem blue PhOLEDs. (b) Energy level diagram of the tandem device. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
luminance of 199 cd/m2, which is higher than two times of that of the single unit device (25.5 cd/A at a luminance of 64 cd/m2), as shown in Fig. 2 (c). Particularly, the tandem device with 5 nm Al layer in CGU shows the improved power efficiency (PEmax = 23.4 lm/W) with
respect to single unit device (PEmax = 20.0 lm/W). These results indicate that the non-doped LiF (1 nm)/Al (5 nm)/HAT-CN (20 nm) CGU has excellent charge generation, injection and transport properties. Furthermore, it should be noted that the efficiencies of the tandem
Fig. 2. Characteristics of the single unit and tandem blue PhOLEDs: (a) current density-voltage, (b) luminance-voltage, (c) current efficiency-luminance, and (d) power efficiencyluminance.
355
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Table 1 EL characteristics of the devices. Device
Von (V)
Lmax (cd/m2)
CEmax (cd/A)
CE1000 (cd/A)
PEmax (lm/W)
PE1000 (lm/W)
EQEmax(%)
CIE (x, y)
Single-blue Tandem: 1 nm Tandem: 3 nm Tandem: 5 nm Tandem: 7 nm Single-white Tandem-white
3.3 12.0 7.8 5.9 5.7 3.2 6.0
12630 6320 7770 13790 14000 22900 35150
25.5 33.0 43.3 53.1 39.7 50.9 94.9
24.0 28.4 38.8 48.8 39.1 46.4 87.4
20.0 6.8 12.6 23.4 17.0 43.3 42.6
15.6 4.2 9.9 18.1 15.8 30.4 33.3
13.3 17.1 22.7 27.8 20.4 17.8 31.6
(0.15, (0.16, (0.13, (0.13, (0.13, (0.35, (0.32,
0.33) 0.38) 0.27) 0.24) 0.34) 0.42) 0.38)
excitation of electrons from NPB HOMO to HAT-CN LUMO [41]. Sun et al. proposed the Zener tunneling model to explain the charge generation process of the HAT-CN/4,4′,4″-tris(3-methylphenylphenylamino)triphenylamine (m-MTDATA) heterojunction under reverse bias [11]. Recent reports have demonstrated that the charge generation is originated from disassociation of the charge-transfer complex which is formed at the HTL/acceptor material interface or in the p-doped HTL by an electron transfer from the HOMO level of HTL to the LUMO level of the acceptor material [44–48]. The charge-transfer complex is Coulomb-bound charge pair, which can be disassociated into free hole and electron rapidly under the applied electrical fields [46–49]. Recently, Dai et al. found that the charge generation efficiency of 1,1-bis-(4-bis(4methyl- phenyl)-amino-phenyl)-cyclohexane (TAPC)/HAT-CN:TAPC/ HAT-CN is higher than that of HAT-CN/TAPC heterojunction due to the formation of the charge transfer complexes in the HAT-CN doped TAPC layer which can generate additional free holes and electrons [10]. Similar results were also achieved in m-MTDATA/HAT-CN:m-MTDATA/ HAT-CN CGU [50]. Fig. 4 (a) shows the UV-VIS-NIR absorption spectra of NPB (10 nm) and NPB:HAT-CN (1:1, 10 nm) films on quartz. The absorption spectrum of NPB film shows only strong absorption peaks at the wavelength less than 400 nm, while HAT-CN doped NPB film reveals additional absorption peak at around 940 nm, which is a strong sign indicating that the charge-transfer complex has been formed in HAT-CN:NPB films [10,51]. The formation of the charge transfer complex at the HAT-CN/NPB interface can be conformed from the PL results of HAT-CN (10 nm)/NPB (30 nm) and NPB (30 nm) layers. As shown in Fig. 4 (b), the PL intensity of HAT-CN/NPB sample significantly decreased compared with that of NPB-only film due to quenching of NPB excitons by the charge-transfer complexes formed at NPB/HAT-CN interface [44,48,52,53]. The charge generation at HATCN/NPB interface could be explained plausibly by the formation and dissociation of the charge transfer complex at HAT-CN/NPB interface. First, electrons are transferred from NPB HOMO to HAT-CN LUMO to form the charge transfer complexes. Second, the charge transfer complexes are separated under an electric field, and then electrons and holes transport in HAT-CN and NPB layers, respectively. In order to get insight into the strong dependence of the tandem device performance on the thickness of Al layer in CGU, the C-V characteristics of ITO/LiF (20 nm)/Alq3 (60 nm)/LiF (1 nm)/Al (1, 3, 5, 7 nm)/HAT-CN (20 nm)/NPB (60 nm)/LiF (20 nm)/Al were explored.
devices gradually increase with increasing the thickness of Al layer, while the efficiency begins to decrease when the thickness of Al layer is more than 5 nm. The detailed discussion will be presented later. To evaluate the role of CGU on charge generation and separation, several special devices with the structures of ITO/Tmpypb (50 nm)/ NPB (50 nm)/Al (device A), ITO/Tmpypb (50 nm)/HAT-CN (20 nm)/ NPB (50 nm)/Al (device B), ITO/Tmpypb (50 nm)/LiF (1 nm)/Al (5 nm)/NPB (50 nm)/Al (device C) and ITO/Tmpypb (50 nm)/LiF (1 nm)/Al (5 nm)/HAT-CN (20 nm)/NPB (50 nm)/Al (device D) were fabricated. The current-voltage characteristics of devices A-D are shown in Fig. 3 (a). According to the energy-level diagram of the CGU only device as shown in Fig. 3 (b), under forward bias, external charge injection can be prevented effectively by Tmpypb and NPB layer because the large injection barriers (about 2.0 eV) exist at the interfaces of ITO/ Tmpypb and NPB/Al, which is verified by the negligible current in device A. Device C with LiF/Al as interlayers exhibits the lowest current density, indicating that the interlayers of LiF/Al cannot generate charge. It is found that device B with 20 nm electron acceptor HAT-CN shows higher current density than devices A and C in the whole range of voltages, demonstrating that the charge carriers are indeed generated at the HAT-CN/NPB interface though electron transfer from the HOMO level of NPB to the LUMO level of HAT-CN [41]. From the energy-level diagram of the CGU only device as shown in Fig. 3 (b), under an external electric field, generated holes can be easily injected into the NPB layer, while electrons extracted from the HAT-CN/NPB interface can be hardly injected into the TmPyPB layer due to a large energy barrier between the LUMO level of TmPyPb (2.8 eV) and the LUMO level of HAT-CN (4.4 eV), resulting in little current flowing in device C. The current is limited by the charge injection at the TmPyPb/HATCN interface rather than the charge generation at the HAT-CN/NPB interface. As shown in Fig. 3 (a), we note that device D with CGU of LiF/Al/HATCN displays much larger current density than device B at the same voltage, which indicates that the interlayer of LiF/Al plays an important role in the charge extraction from CGU. In recent years, the charge generation mechanism of HTL/HAT-CN has been investigated by several researchers. Various models have been introduced to try to explain the charge generation and separation processes [11,41–43]. Some evidences based on ultraviolet photoelectron spectroscopy (UPS) have suggested that the charge carriers are most likely to be generated at the HTL/HAT-CN interface through
Fig. 3. Current density-voltage characteristics of devices A ∼ D and the energy-level diagram of the CGU only device.
356
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
that no charges are generated in these devices. Under forward bias, the capacitance exhibits an abrupt increase for all the devices, and is very sensitive to the thickness of Al interlayer. The increased capacitance indicates that the holes and electrons are generated in the CGUs and injected into the adjacent NPB and Alq3 layer, respectively, because no external charges can be injected into the devices due to the thick LiF (20 nm) insulating layers. It is found that the voltages for the abrupt increase of capacitance gradually decrease from 7.3 V to 3.5 V with increasing the Al thickness. Furthermore, the maximum capacitance also increases gradually with increasing the thickness of Al layer, and the voltages at the maximum capacitance also decrease with the Al thickness. These results indicate that the thick Al interlayer can effectively improve the electron injection and inhibit the accumulation of charges in the CGU, thus to decrease effectively the operation voltages and to improve power efficiency for tandem OLEDs as shown in Fig. 2 (b) and (d). To elucidate the effect of the thickness of Al layer on the electron injection property of LiF/Al, the morphology evolution of Al layer with different thickness was investigated by using AFM. Fig. 6 shows the AFM images of the thermally evaporated Al films with different thickness on the TmPyPb (40 nm) surface. It can be observed that the surface morphology of Al films is quite smooth whatever the film thickness. The average roughness changes from 0.46 to 0.88 nm with increasing Al thickness. No Al nanostructure is evidenced, indicating that continuous Al films are formed on TmPyPb, which may be attributed to a perfect wetting of Al on TmPyPb. The dependence of the electron-injection ability on the thickness of Al layer cannot be attributed to the morphological change. As we know, the sheet resistance is strongly dependent on the thickness of Al films. The thick Al layer would result in improvements in both electrical conductivity and carrier injection. Additionally, it has been found that the surface of Al film could be oxidized rapidly into Al2O3 even in high vacuum [55]. This oxide layer would weaken the electron-injection ability of LiF/Al especially for the thinner Al layer. Therefore, the electron injection property of LiF/Al increased gradually with increasing the thickness of Al layer. Additionally, it is noted that the performance of the device worsens when the thickness of Al layer is above 5 nm, although the thick Al layer is beneficial to charge generation and separation. As we know, the high transmittance of the CGU is very important for high efficiency tandem OLED. Fig. 7 shows the optical transmission spectra of different thickness Al layers deposited on ITO substrates. As shown in Fig. 7, we note that the optical transparency gradually decreases with the thickness of Al layer in the visible spectra region due to reflection and absorption. Compared to ITO substrate, the optical transparency at the main emission peak of FIrpic (472 nm) decreases by 0.6% for 1 nm Al, 3.9% for 3 nm Al, 8.2% for 5 nm Al and 14.0% for 7 nm Al. The thick Al layer weakens the light output of the tandem device. Therefore, under the optimal Al thickness of 5 nm, CGU possesses better capability of charge generation and higher light output, which results in excellent EL performance for the tandem device. Encouraged by the impressive results obtained from the tandem blue device, a non-doped tandem white device with a structure of ITO/ PEDOT:PSS (40 nm)/EL unit/CGU/EL unit/LiF(1 nm)/Al(100 nm) was developed. The EL unit consists of NPB (30 nm)/TCTA (10 nm)/PO-01 (0.1 nm)/mCP (3 nm)/Firpic (0.25 nm)/TmPyPb (40 nm), where 0.1 nm PO-01 and 0.25 nm Firpic are used as non-doped blue and yellow EMLs, respectively. Fig. 8 shows the EL characteristics of the single unit and tandem white devices. The detailed EL performance of the tandem and single unit OLEDs are summarized in Table 1. As shown in Fig. 8, the tandem white device presents a turn-on voltage of 6.0 V, maximum luminance of 35150 cd/m2 at a voltage of 15.0 V and maximum power efficiency of 42.6 lm/W. The CEmax and EQEmax of the tandem white device reach 94.9 cd/A and 31.6% at a luminance of 90 cd/m2, respectively, which are about two times as much as 50.9 cd/ A and 17.8% of the single unit device. More importantly, the CE and PE slightly decrease to 87.4 cd/A and 33.3 lm/W at 1000 cd/m2,
Fig. 4. (a) UV-VIS-NIR absorption spectra of NPB (10 nm) and NPB:HAT-CN (1:1, 10 nm) films deposited on quartz substrate. Inset: relative intensity of charge transfer complex bands. (b) PL intensities of the NPB (20 nm) and HAT-CN (10 nm)/NPB (20 nm) films deposited on ITO substrate.
Fig. 5. Capacitance-voltage characteristics of the capacitance devices. The frequency is set at a constant value of 1 kHz, C0 is the geometric capacitance.
The double 20 nm LiF insulating layers are introduced to prevent charge carrier injection from the anode and cathode. Fig. 5 shows the CV characteristics of the devices with different thickness of Al layer in CGU, where the measured capacitance (C) is normalized by the geometric capacitance (C0) determined by the equation of C0 = ε0εA/d [54]. In this equation, ε is the dielectric constant, ε0 is the vacuum permittivity, A is the contact area, and d is the thickness of the functional layers between the electrodes. It can be seen that the capacitance is almost constant under reverse biases for all the devices, indicating 357
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 6. AFM morphology of Al films deposited on 40 nm-thick TmPyPb for different Al thickness: (a) 1.0 nm, (b) 3.0 nm, (c) 5.0 nm, and (d) 7.0 nm.
device displays lower intensity compared with that of the single unit white device due to absorption of blue light by PO-01. The similar phenomenon was also observed in other tandem white devices [18,26]. As can be seen in Fig. 9 (b), the EL spectra of the white devices are origined from the emission of FIrpic and PO-01. The devices show good white emission with CIE coordinates of (0.35, 0.42) for single unit device and (0.32, 0.38) for tandem device. The excellent performance of the tandem white device indicates that ultrathin EML and non-doped CGU can be used to fabricate high performance white OLEDs for lighting. 4. Conclusion In summary, we present highly efficient non-doped tandem OLEDs using ultrathin phosphorescent dyes as EMLs and LiF/Al/HAT-CN as CGU. It was found that the performance of the tandem OLEDs strongly depends on the thickness of Al layer in CGU. From the current densityvoltage and capacitance-voltage characteristics of the external-carrier excluding devices, the charge carriers are generated at the HAT-CN/ NPB interface though disassociation of the charge-transfer complexes formed between HAT-CN and NPB, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent ETL. We have found that the thick Al layer is beneficial to charge separation, while it weakens the light output of the tandem device. Under optimal Al thickness of 5 nm, CGU possesses better capability of charge generation and higher light output. The tandem blue device reaches a CEmax of 53.1 cd/A, which is about two times higher than that (25.5 cd/A) of the single unit device. Furthermore, based on the CGU of LiF/Al/HAT-CN, the non-doped
Fig. 7. Optical transmission spectra of Al layers with different thickness deposited on ITO substrates.
respectively, showing low efficiency roll-off in the tandem device. Additionally, compared to the tandem blue device with twofold enhancement in the CE, it is noted that the CE of the tandem white device is slightly less than two times of that of the single unit white device. In other words, there is partial energy loss in the tandem white device. The reason may be attributed to the absorption of dye PO-01 in another EL unit. As shown in Fig. 9 (a), the EL peak of FIrPic is approximate 472 nm, which has a partial overlap with the absorption spectrum of PO-01, so the photons radiated from Firpic can be absorbed by PO-01 resulting in partial energy loss. This blue light loss can be confirmed through comparing the EL spectra of the single unit and tandem white devices. As shown in Fig. 9 (b), the blue emission of the tandem white 358
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 8. Characteristics of the single unit and tandem white PhOLEDs: (a) current density-voltage, (b) luminance-voltage, (c) current efficiency-luminance, and (d) EQE-luminance.
application with simplified device structures and high performance tandem OLEDs for display and lighting applications. Acknowledgements This work was supported by the National Key Basic Research Program of China (2014CB648300 and 2017YFB0404500), the NSFC (61774088, 61705112, 21422402 and 21674050), the NSF of Jiangsu Province (BK20161519 and BK20170913), the NUPT “1311 Project” and the Scientific Research Foundation of NUPT (NY214178, NY214093 and NY215076), the National Synergistic Innovation Center for Advanced Materials, the Synergistic Innovation Center for Organic Electronics and Information Displays, and the Project Funded by the PAPD of Jiangsu Higher Education Institutions (YX03001). References [1] X. Zhang, Q. Hu, J. Lin, Z. Lei, X. Guo, L. Xie, W.-Y. Lai, W. Huang, Appl. Phys. Lett. 103 (2013) 153301. [2] Y.-D. Jiu, C.-F. Liu, J.-Y. Wang, W.-Y. Lai, Y. Jiang, W.-D. Xu, X.-W. Zhang, W. Huang, Polym. Chem. 6 (2015) 8019. [3] S. Cao, L. Hao, W.-Y. Lai, H. Zhang, Z. Yu, X. Zhang, X. Liu, W. Huang, J. Mater. Chem. C 4 (2016) 4709. [4] X. Zhang, X. Guo, Y. Chen, J. Wang, Z. Lei, W.-Y. Lai, Q. Fan, W. Huang, J. Lumin. 161 (2015) 300. [5] Y. Tao, X. Guo, L. Hao, R. Chen, H. Li, Y. Chen, X. Zhang, W.-Y. Lai, W. Huang, Adv. Mater 27 (2015) 6939. [6] C. Tang, T. Yang, X. Cao, Y. Tao, F. Wang, C. Zhong, Y. Qian, X. Zhang, W. Huang, Adv. Opt. Mater 3 (2015) 786. [7] B. Liu, F. Dang, Z. Tian, Z. Feng, D. Jin, W. Dang, X. Yang, G. Zhou, Z. Wu, ACS Appl. Mater. Interfaces 9 (2017) 16360. [8] X. Yang, G. Zhou, W.-Y. Wong, Chem. Soc. Rev. 44 (2015) 8484. [9] Q.X. Guo, H.D. Sun, J.X. Wang, D.Z. Yang, J.S. Chen, D.G. Ma, J. Mater. Chem. C 4 (2016) 376. [10] Y.F. Dai, H.M. Zhang, Z.Q. Zhang, Y.P. Liu, J.S. Chen, D.G. Ma, J. Mater. Chem. C 3 (2015) 6809. [11] H.D. Sun, Q.X. Guo, D.Z. Yang, Y.H. Chen, J.S. Chen, D.G. Ma, ACS Photonics 2 (2015) 271. [12] M.-K. Fung, Y.-Q. Li, L.-S. Liao, Adv. Mater 28 (2016) 10381. [13] Y.H. Chen, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, J. Appl. Phys. 110 (2011) 074504. [14] Q. Guo, D. Yang, J. Chen, X. Qiao, T. Ahamad, S.M. Alshehri, D. Ma, J. Appl. Phys. 121 (2017) 115502. [15] H. Sun, Y. Chen, J. Chen, D. Ma, IEE J. Sel. Top. Quantum Electron 22 (2016) 7800110. [16] H.D. Sun, Y.H. Chen, L.P. Zhu, Q.X. Guo, D.Z. Yang, J.S. Chen, D.G. Ma, Adv. Electron. Mater 1 (2015) 1500176. [17] D.Y. Zhou, H.Z. Siboni, Q. Wang, L.S. Liao, H. Aziz, J. Appl. Phys. 116 (2014)
Fig. 9. (a) Absorption spectrum of PO-01 and EL spectrum of FIrPic. (b) EL spectra of the single unit and tandem white PhOLEDs.
tandem white OLED with ultrathin EMLs presents a CEmax of 94.9 cd/A and EQEmax of 31.6%. These results demonstrate that the non-doped CGU structure and ultrathin EML are promising candidates for 359
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
[36] N.A. Talik, K.H. Yeoh, C.Y.B. Ng, B.K. Yap, K.L. Woon, J. Lumin. 154 (2014) 345. [37] Y. Zhao, J. Chen, D. Ma, ACS Appl. Mater. Interfaces 5 (2013) 965. [38] J. Liu, X. Shi, X. Wu, J. Wang, Z. Min, Y. Wang, M. Yang, C.H. Chen, G. He, Org. Electron 15 (2014) 2492. [39] L. Zhu, Y. Zhao, H. Zhang, J. Chen, D. Ma, J. Appl. Phys. 115 (2014) 244512. [40] S. Liu, B. Li, L. Zhang, S. Yue, Appl. Phys. Lett. 98 (2011) 163301. [41] Y.K. Kim, J.W. Kim, Y. Park, Appl. Phys. Lett. 94 (2009) 063305. [42] S.M. Park, Y.H. Kim, Y. Yi, H.-Y. Oh, J.W. Kim, Appl. Phys. Lett. 97 (2010) 063308. [43] S. Lee, J.-H. Lee, J.-H. Lee, J.-J. Kim, Adv. Funct. Mater 22 (2012) 855. [44] T. Matsushima, H. Murata, Appl. Phys. Lett. 95 (2009) 203306. [45] T. Matsushima, G.-H. Jin, Y. Kanai, T. Yokota, S. Kitada, T. Kishi, H. Murata, Org. Electron 12 (2011) 520. [46] A. Mityashin, Y. Olivier, T. Van Regemorter, C. Rolin, S. Verlaak, N.G. Martinelli, D. Beljonne, J. Cornil, J. Genoe, P. Heremans, Adv. Mater 24 (2012) 1535. [47] A.R. Yu, Y. Pei, R.C. Yi, S.B. Liu, M. Chu, H.M. Yu, Y.J. Tang, Z.S. He, X.Y. Hou, Org. Electron 46 (2017) 121. [48] S.-J. Yoo, J.-H. Chang, J.-H. Lee, C.-K. Moon, C.-I. Wu, J.-J. Kim, Sci. Rep. 4 (2014) 3902. [49] S.-J. Yoo, J.-J. Kim, Macromol. Rapid Comm. 36 (2015) 984. [50] L.X. Wang, C.S. Shi, Q.X. Guo, J.S. Chen, X.F. Qiao, D.G. Ma, T. Ahamad, S.M. Alshehri, S.S. Bae, J. Appl. Phys. 122 (2017) 125501. [51] X. Zhang, Z. Wu, D. Wang, D. Wang, R. He, X. Hou, Appl. Surf. Sci. 256 (2010) 4468. [52] T.-W. Ng, M.-F. Lo, Q.-D. Yang, M.-K. Fung, C.-S. Lee, Adv. Funct. Mater 22 (2012) 3035. [53] T. Matsushima, H. Murata, Org. Electron 14 (2013) 1149. [54] K.S. Yook, S.O. Jeon, S.Y. Min, J.Y. Lee, H.J. Yang, T. Noh, S.K. Kang, T.W. Lee, Adv. Funct. Mater 20 (2010) 1797. [55] G. Maidecchi, G. Gonella, R.P. Zaccaria, R. Moroni, L. Anghinolfi, A. Giglia, S. Nannarone, L. Mattera, H.-L. Dai, M. Canepa, F. Bisio, ACS Nano 7 (2013) 5834.
223708. [18] Q. Wang, J.Q. Ding, Z.Q. Zhang, D.G. Ma, Y.X. Cheng, L.X. Wang, F.S. Wang, J. Appl. Phys. 105 (2009) 076101. [19] Y.H. Lee, M.W. Lin, T.C. Wen, T.F. Guo, J. Appl. Phys. 114 (2013) 154512. [20] L.-S. Liao, W.K. Slusarek, T.K. Hatwar, M.L. Ricks, D.L. Comfort, Adv. Mater 20 (2008) 324. [21] Q.D. Ou, L. Zhou, Y.Q. Li, J.D. Chen, C. Li, S. Shen, J.X. Tang, Adv. Opt. Mater 3 (2015) 87. [22] D.Y. Zhou, F.S. Zu, Y.J. Zhang, X.B. Shi, H. Aziz, L.S. Liao, Appl. Phys. Lett. 105 (2014) 083301. [23] T.W. Lee, T. Noh, B.K. Choi, M.S. Kim, D.W. Shin, J. Kido, Appl. Phys. Lett. 92 (2008) 043301. [24] L.S. Liao, K.P. Klubek, C.W. Tang, Appl. Phys. Lett. 84 (2004) 167. [25] F. Yan, R. Chen, H.D. Sun, X.W. Sun, Appl. Phys. Lett. 102 (2013) 203303. [26] M.H. Ho, T.M. Chen, P.C. Yeh, S.W. Hwang, C.H. Chen, Appl. Phys. Lett. 91 (2007) 233507. [27] Y.H. Chen, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, Appl. Phys. Lett. 99 (2011) 103304. [28] Y.H. Deng, Q.D. Ou, Q.K. Wang, H.X. Wei, Y.Q. Li, S.T. Lee, J.X. Tang, J. Mater. Chem. C 2 (2014) 1982. [29] Y.B. Zhao, S.T. Tan, H.V. Demir, X.W. Sun, Org. Electron 23 (2015) 70. [30] S.L. Lai, M.Y. Chan, M.K. Fung, C.S. Lee, S.T. Lee, J. Appl. Phys. 101 (2007) 014509. [31] Y.H. Chen, Q. Wang, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, Org. Electron 13 (2012) 1121. [32] Z. Ma, J. Zhao, X. Wang, J. Yu, Org. Electron 30 (2016) 136. [33] H. Sasabe, K. Minamoto, Y.J. Pu, M. Hirasawa, J. Kido, Org. Electron 13 (2012) 2615. [34] K.A. Knauer, E. Najafabadi, W. Haske, M.P. Gaj, K.C. Davis, C. Fuentes-Hernandez, U. Carrasco, B. Kippelen, Org. Electron 14 (2013) 2418. [35] B. Jiao, Z.X. Wu, Z.Y. Yang, X. Hou, Phys. Status Solidi A 210 (2013) 2583.
360
Contents lists available at ScienceDirect
Organic Electronics journal homepage: www.elsevier.com/locate/orgel
Highly efficient tandem organic light-emitting devices adopting a nondoped charge-generation unit and ultrathin emitting layers
T
Xinwen Zhanga, Mengke Zhanga, Mengjiao Liua, Yuehua Chena, Jiong Wanga, Xiaolin Zhanga, Junjie Zhanga, Wen-Yong Laia,∗∗, Wei Huanga,b,∗ a
Key Laboratory for Organic Electronics and Information Displays (KLOEID) & Institute of Advanced Materials (IAM), Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing University of Posts & Telecommunications, Nanjing 210023, China b Shaanxi Institute of Flexible Electronics (SIFE), Northwestern Polytechnical University (NPU), 127 West Youyi Road, Xi'an 710072, Shaanxi, China
A R T I C L E I N F O
A B S T R A C T
Keywords: Tandem organic light-emitting devices Charge generation unit Non-doped devices Ultrathin emitting layers
Highly efficient non-doped tandem phosphorescent organic light-emitting devices (PhOLEDs) have been demonstrated by employing LiF/AL/1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN) as a charge generation unit (CGU) and ultrathin phosphorescent dyes as emitting layers (EMLs). It was found that the performance of the tandem PhOLEDs was significantly dependent on the Al thickness in CGU. Analyses regarding the current density-voltage characteristics of the CGU only devices indicate that the charge carriers are generated at the HAT-CN/NPB interface, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent electron transport layer. From the capacitancevoltage and optical transparency characteristics of CGU with different thicknesses of Al interlayer, the thick Al layer is beneficial to charge separation, while it weakens the light output of the tandem device. Under the optimal Al thickness of 5 nm, more than twofold enhancement of current efficiency is achieved for the nondoped tandem blue device. Furthermore, the non-doped tandem white PhOLED was also developed showing a maximum current efficiency of 94.9 cd/A and maximum external quantum efficiency of 31.6%.
1. Introduction Organic light-emitting devices (OLEDs) have attracted widespread attention as next-generation solid state lighting and flat panel displays owing to their excellent properties of flexibility, high efficiency and fast response [1–3]. Up to present, high-performance devices can be achieved by adopting phosphorescent or thermally activated delayed fluorescent emitters due to the full use of both singlet and triplet excitons [4–8]. Nevertheless, the device at high current density usually suffers from the accelerated degradation of lifetime and the roll-off of efficiency [9]. A tandem OLED combines several stacked electroluminescence (EL) units in series through a charge generation unit (CGU), which could achieve both high efficiency and long operating lifetime through reducing the current density in the device [10–13]. Under the applied electrical field, holes and electrons are generated in the CGU, and injected into the adjacent carrier-transporting layers of the individual EL units [12]. Therefore, it is very important for a highperformance tandem OLED to develop an efficient CGU possessing excellent charge generation and separation properties [14].
In recent years, different CGUs with a bilayer structure have been used in tandem OLEDs to enhance the efficiency and stability [15–17], including n-doped electron-transporting layer (n-doped ETL)/metaloxide layer [18], n-doped ETL/organic electron acceptor layer and ndoped ETL/p-doped hole-transporting layer (p-doped HTL) [19]. In these CGUs, the n-doped ETLs composed of Mg, Li, Cs and their compounds doped electron-transporting materials, such as Li:tris(8-hydroxy-quinolinato) aluminium (Alq3) [20], Mg:4,7-diphenyl-1,10-phenanthroline (BPhen) [21], LiNH2:BPhen [22] and Cs2CO3:2,9-dimethyl4,7-diphenyl-1,10-phenanthroline (BCP) [23], and p-doped HTLs composed of 2,3,5,6-tetrafluoro-7,7,8,8-tetracyano-quinodimethane (F4TCNQ), FeCl3, MoO3, WO3 doped electron-transporting materials, such as FeCl3:N,N′-bis(naphthalen-1-yl)- N,N′-bis(phenyl)-benzidine (NPB) [24], F4-TCNQ:4,4′,4″-tris(N-3-methylphenyl-N-phenyl-amino) triphenylamine (m-MTDATA) [25], WO3:NPB [26] and MoO3:4,4′,4″-tris (carbazol-9-yl) triphenylamine (TCTA) [27], are commonly adopted through the complicated co-evaporation technique to achieve effective charge injection from CGUs to adjacent charge-transporting layers of the EL units. According to the recent reports [28,29], diffusion of p-
∗ Corresponding author. Key Laboratory for Organic Electronics and Information Displays (KLOEID) & Institute of Advanced Materials (IAM), Jiangsu National Synergetic Innovation Center for Advanced Materials (SICAM), Nanjing University of Posts & Telecommunications, Nanjing 210023, China. ∗∗ Corresponding author. E-mail addresses: [email protected] (W.-Y. Lai), [email protected] (W. Huang).
https://doi.org/10.1016/j.orgel.2017.10.042 Received 10 August 2017; Received in revised form 19 October 2017; Accepted 31 October 2017 Available online 01 November 2017 1566-1199/ © 2017 Published by Elsevier B.V.
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
further purification. In the experiment, the patterned ITO glass substrates were ultrasonically cleaned with detergent, alcohol and acetone, deionized water and then dried at 120 °C in a vacuum oven for more than one hour. After ultraviolet (UV)-ozone treating for 4 min, a 40 nm PEDOT:PSS film was spin coated on the ITO substrate and annealed at 120 °C in for 15 min. The samples were then loaded into a high-vacuum thermal evaporation system. All layers were deposited on top of PEDOT:PSS by thermal evaporation under a pressure of 5 × 10−4 Pa. The thickness of the films was determined in situ by a quartz-crystal sensor. The evaporation rate for the organic layer, LiF, and Al layer in the CGU are 2–3, 0.3–0.5, and 0.5–0.7 Å/s, respectively. The evaporation rate for the phosphorescent dyes and Al cathode are 0.03–0.05 and 3–5 Å/s, respectively. The active emissive area of the devices is 12 mm2. The current-voltage-luminescence characteristics were measured by a Keithley 2602 source meter with a calibrated silicon photodiode. The electroluminescence (EL) spectra of the devices were analyzed with a spectrometer (PR655). Capacitance-voltage (C-V) measurements were performed using an impedance analyzer (Wayne Kerr 6505B) at a constant frequency of 1000 Hz. A 200 mV amplitude AC signal superimposed on a DC bias was used to measure device capacitance as a function of DC bias. The optical transmission and absorption spectra were performed with a Shimadzu UV-3600 spectrophotometer. The ultraviolet–visible–near infrared (UV-VIS-NIR) absorption spectra were measured by a RF-5301PC spectrometer. The photoluminescence (PL) spectra were measured with an excitation at 350 nm using a Shimadzu RF 5301 spectrophotometer. The surface morphology of the films was investigated by atomic force microscopy (AFM, Bruker Dimension Icon). All the devices were characterized without encapsulation, and all the measurements were carried out under ambient condition at room temperature.
dopant and/or n-dopant will cause degradation of the CGU under electrical bias, thus deteriorating device operational stability. Although non-doped organic heterojunctions such as copper hexadecafluorophthalocyanine (F16CuPc)/copper phthalocyanine (CuPc) [30], C60/pentacene [31] and C60/cobalt phthalocyanine (CoPc) [32] are proposed as CGUs in tandem OLEDs, some interfacial modification layers (e.g., p-doped HTLs and n-doped ETLs) have to be introduced to facilitate the charge injection, which will increase the cost and complexity of the devices. Therefore, designing effective non-doped CGU to simplify the fabrication process of tandem OLEDs is very necessary. Recently, the strong electron acceptor materials, such as MoO3 and 1,4,5,8,9,11-hexaazatriphenylene hexacarbonitrile (HAT-CN), have been reported as the n-type layers to construct non-doped CGU for efficient tandem OLEDs [33–36]. The charge carriers are usually generated at the interface between a strong electron acceptor material and an electron donor material through electron transfer from the donor material to the acceptor material under the application of an external electric field [33–36]. In addition, dye-doped host structures are usually adopted by coevaporation of emitting dyes and host materials in most conventional OLEDs to achieve desirable device performance. However, co-evaporation is a complicated process, especially for white OLEDs, where the doping concentration of dyes must be controlled precisely to obtain balanced white emission. This usually leads to a complicated fabrication process, poor reproducibility and high cost. Recently, non-doped phosphorescent OLEDs (PhOLEDs) by inserting an ultrathin emitting layer (EML) of pure phosphorescent dye between a HTL and an ETL have been introduced to simplify the fabrication process [37–40]. Due to confinement of excitons within ultrathin EML, non-doped PhOLEDs exhibited the maximum external quantum efficiency (EQEmax) about 20%, which was comparable to that of the doped EML devices [37]. Thus, it is a good strategy to develop non-doped PhOLEDs for industrial display and lighting applications. In this paper, we developed highly efficient non-doped tandem PhOLEDs based on ultrathin EMLs and non-doped CGU with the structure of LiF/Al/HAT-CN. We found that the performance of the tandem PhOLEDs was significantly dependent on the Al thickness of CGU. The current density-voltage characteristics of the CGU only devices demonstrate that the charge carriers are generated at the HATCN/NPB interface though disassociation of the charge-transfer complexes formed between HAT-CN and NPB, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent ETL. From the capacitance-voltage and optical transparency characteristics of CGU with different thicknesses of Al interlayer, the thick Al layer is beneficial to charge generation and separation, while it weakens the light output of the tandem device. Under optimal Al thickness of 5 nm, the tandem blue PhOLEDs attain a maximum current efficiency (CEmax) of 53.2 cd/A and a maximum power efficiency (PEmax) of 23.4 lm/W. Furthermore, the tandem white PhOLED presents a CEmax of 94.9 cd/A and EQEmax of 31.6%. The results indicate that ultrathin EML and non-doped CGU can be universally adopted to simplify the fabrication process.
3. Results and discussion To demonstrate efficient tandem OLEDs, it is very important to optimize the structure to efficiently generate charge carriers. We fabricated four tandem devices consisted of LiF (1 nm)/Al (1 nm, 3 nm, 5 nm, 7 nm)/HAT-CN (20 nm) as CGU to investigate the effect of the Al thickness on the device performance, as shown in Fig. 1. The corresponding single unit device with a structure of ITO/PEDOT:PSS (40 nm)/NPB (40 nm)/TCTA (10 nm)/Firpic (0.25 nm)/TmPyPb (40 nm)/LiF (1 nm)/Al (100 nm) was also fabricated for comparison. The single unit device has been optimized through controlling the thickness of ultrathin FIrpic layer. Fig. 2 shows the current densityvoltage, luminance-voltage, current efficiency-luminance, and power efficiency-luminance characteristics of the tandem OLEDs and single unit OLED. The detailed results are summarized in Table 1. It was found that the performance of the tandem devices is strongly dependent on the thickness of Al layer in CGU. As shown in Fig. 2 (a), the current densities of the devices obviously increase with the thickness of Al layer. The driving voltages of the tandem device are 29.6 V (1 nm), 16.2 V (3 nm), 12.1 V (5 nm) and 11.1 V (7 nm), respectively, under a current density of 30 mA/cm2, indicating that the large extra voltage drops in the region of CGU. The strong Al layer thickness dependence of the current density-voltage characteristics indicates that the Al layer plays an important role in charge generation and/or charge injection. From the luminance-voltage characteristics, a similar variation as current density-voltage is also observed. As shown in Fig. 2 (b), the turn-on voltage (the voltage at 1 cd/m2) largely decreases from 12.0 V to 5.7 V as the Al layer thickness increases. Under a luminance of 1000 cd/m2, the driving voltages are 21.0 V (1 nm), 12.3 V (3 nm), 8.5 V (5 nm) and 7.8 V (7 nm), respectively. These results indicate that a thicker Al layer is beneficial to reduction of device driving voltage. Besides, the turn-on voltage of tandem OLED with 7 nm Al layer is around 5.7 V, which is 1.7 times that of the single unit device (3.3 V). At the optimal Al thickness of 5 nm, the tandem device shows a CEmax of 53.1 cd/A at a
2. Experimental The hole-injection material of poly(3,4-ethylenedioxythiophene):poly (styrenesulfonic acid) (PEDOT:PSS, AI4083) was purchased from H. C. Starck Inc. The organic acceptor material of HATCN, the hole transporting materials of N,N′-di(naphth-1-yl)-N,N′-diphenyl-benzidine (NPB) and 4,4′,4″-tris(N-carbazolyl)-triphenylamine (TCTA), the phosphorescent dopants of bis(3,5-difluoro- 2-(2-pyridyl)phenyl-(2-carboxypyridyl)iridium(III) (FIrpic) and bis(4-phenylthieno [3,2-c]pyridine) (acetylacetonate) iridium(III)(PO-01), and the electron transporting material of 1,3,5-tri(m-pyrid-3- yl-phenyl)-benzene (TmPyPB) and Alq3 were purchased from Nichem Fine Technology Co. Ltd. All the above-mentioned materials were used as-received without 354
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 1. (a) Schematic device structures of the single unit and tandem blue PhOLEDs. (b) Energy level diagram of the tandem device. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
luminance of 199 cd/m2, which is higher than two times of that of the single unit device (25.5 cd/A at a luminance of 64 cd/m2), as shown in Fig. 2 (c). Particularly, the tandem device with 5 nm Al layer in CGU shows the improved power efficiency (PEmax = 23.4 lm/W) with
respect to single unit device (PEmax = 20.0 lm/W). These results indicate that the non-doped LiF (1 nm)/Al (5 nm)/HAT-CN (20 nm) CGU has excellent charge generation, injection and transport properties. Furthermore, it should be noted that the efficiencies of the tandem
Fig. 2. Characteristics of the single unit and tandem blue PhOLEDs: (a) current density-voltage, (b) luminance-voltage, (c) current efficiency-luminance, and (d) power efficiencyluminance.
355
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Table 1 EL characteristics of the devices. Device
Von (V)
Lmax (cd/m2)
CEmax (cd/A)
CE1000 (cd/A)
PEmax (lm/W)
PE1000 (lm/W)
EQEmax(%)
CIE (x, y)
Single-blue Tandem: 1 nm Tandem: 3 nm Tandem: 5 nm Tandem: 7 nm Single-white Tandem-white
3.3 12.0 7.8 5.9 5.7 3.2 6.0
12630 6320 7770 13790 14000 22900 35150
25.5 33.0 43.3 53.1 39.7 50.9 94.9
24.0 28.4 38.8 48.8 39.1 46.4 87.4
20.0 6.8 12.6 23.4 17.0 43.3 42.6
15.6 4.2 9.9 18.1 15.8 30.4 33.3
13.3 17.1 22.7 27.8 20.4 17.8 31.6
(0.15, (0.16, (0.13, (0.13, (0.13, (0.35, (0.32,
0.33) 0.38) 0.27) 0.24) 0.34) 0.42) 0.38)
excitation of electrons from NPB HOMO to HAT-CN LUMO [41]. Sun et al. proposed the Zener tunneling model to explain the charge generation process of the HAT-CN/4,4′,4″-tris(3-methylphenylphenylamino)triphenylamine (m-MTDATA) heterojunction under reverse bias [11]. Recent reports have demonstrated that the charge generation is originated from disassociation of the charge-transfer complex which is formed at the HTL/acceptor material interface or in the p-doped HTL by an electron transfer from the HOMO level of HTL to the LUMO level of the acceptor material [44–48]. The charge-transfer complex is Coulomb-bound charge pair, which can be disassociated into free hole and electron rapidly under the applied electrical fields [46–49]. Recently, Dai et al. found that the charge generation efficiency of 1,1-bis-(4-bis(4methyl- phenyl)-amino-phenyl)-cyclohexane (TAPC)/HAT-CN:TAPC/ HAT-CN is higher than that of HAT-CN/TAPC heterojunction due to the formation of the charge transfer complexes in the HAT-CN doped TAPC layer which can generate additional free holes and electrons [10]. Similar results were also achieved in m-MTDATA/HAT-CN:m-MTDATA/ HAT-CN CGU [50]. Fig. 4 (a) shows the UV-VIS-NIR absorption spectra of NPB (10 nm) and NPB:HAT-CN (1:1, 10 nm) films on quartz. The absorption spectrum of NPB film shows only strong absorption peaks at the wavelength less than 400 nm, while HAT-CN doped NPB film reveals additional absorption peak at around 940 nm, which is a strong sign indicating that the charge-transfer complex has been formed in HAT-CN:NPB films [10,51]. The formation of the charge transfer complex at the HAT-CN/NPB interface can be conformed from the PL results of HAT-CN (10 nm)/NPB (30 nm) and NPB (30 nm) layers. As shown in Fig. 4 (b), the PL intensity of HAT-CN/NPB sample significantly decreased compared with that of NPB-only film due to quenching of NPB excitons by the charge-transfer complexes formed at NPB/HAT-CN interface [44,48,52,53]. The charge generation at HATCN/NPB interface could be explained plausibly by the formation and dissociation of the charge transfer complex at HAT-CN/NPB interface. First, electrons are transferred from NPB HOMO to HAT-CN LUMO to form the charge transfer complexes. Second, the charge transfer complexes are separated under an electric field, and then electrons and holes transport in HAT-CN and NPB layers, respectively. In order to get insight into the strong dependence of the tandem device performance on the thickness of Al layer in CGU, the C-V characteristics of ITO/LiF (20 nm)/Alq3 (60 nm)/LiF (1 nm)/Al (1, 3, 5, 7 nm)/HAT-CN (20 nm)/NPB (60 nm)/LiF (20 nm)/Al were explored.
devices gradually increase with increasing the thickness of Al layer, while the efficiency begins to decrease when the thickness of Al layer is more than 5 nm. The detailed discussion will be presented later. To evaluate the role of CGU on charge generation and separation, several special devices with the structures of ITO/Tmpypb (50 nm)/ NPB (50 nm)/Al (device A), ITO/Tmpypb (50 nm)/HAT-CN (20 nm)/ NPB (50 nm)/Al (device B), ITO/Tmpypb (50 nm)/LiF (1 nm)/Al (5 nm)/NPB (50 nm)/Al (device C) and ITO/Tmpypb (50 nm)/LiF (1 nm)/Al (5 nm)/HAT-CN (20 nm)/NPB (50 nm)/Al (device D) were fabricated. The current-voltage characteristics of devices A-D are shown in Fig. 3 (a). According to the energy-level diagram of the CGU only device as shown in Fig. 3 (b), under forward bias, external charge injection can be prevented effectively by Tmpypb and NPB layer because the large injection barriers (about 2.0 eV) exist at the interfaces of ITO/ Tmpypb and NPB/Al, which is verified by the negligible current in device A. Device C with LiF/Al as interlayers exhibits the lowest current density, indicating that the interlayers of LiF/Al cannot generate charge. It is found that device B with 20 nm electron acceptor HAT-CN shows higher current density than devices A and C in the whole range of voltages, demonstrating that the charge carriers are indeed generated at the HAT-CN/NPB interface though electron transfer from the HOMO level of NPB to the LUMO level of HAT-CN [41]. From the energy-level diagram of the CGU only device as shown in Fig. 3 (b), under an external electric field, generated holes can be easily injected into the NPB layer, while electrons extracted from the HAT-CN/NPB interface can be hardly injected into the TmPyPB layer due to a large energy barrier between the LUMO level of TmPyPb (2.8 eV) and the LUMO level of HAT-CN (4.4 eV), resulting in little current flowing in device C. The current is limited by the charge injection at the TmPyPb/HATCN interface rather than the charge generation at the HAT-CN/NPB interface. As shown in Fig. 3 (a), we note that device D with CGU of LiF/Al/HATCN displays much larger current density than device B at the same voltage, which indicates that the interlayer of LiF/Al plays an important role in the charge extraction from CGU. In recent years, the charge generation mechanism of HTL/HAT-CN has been investigated by several researchers. Various models have been introduced to try to explain the charge generation and separation processes [11,41–43]. Some evidences based on ultraviolet photoelectron spectroscopy (UPS) have suggested that the charge carriers are most likely to be generated at the HTL/HAT-CN interface through
Fig. 3. Current density-voltage characteristics of devices A ∼ D and the energy-level diagram of the CGU only device.
356
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
that no charges are generated in these devices. Under forward bias, the capacitance exhibits an abrupt increase for all the devices, and is very sensitive to the thickness of Al interlayer. The increased capacitance indicates that the holes and electrons are generated in the CGUs and injected into the adjacent NPB and Alq3 layer, respectively, because no external charges can be injected into the devices due to the thick LiF (20 nm) insulating layers. It is found that the voltages for the abrupt increase of capacitance gradually decrease from 7.3 V to 3.5 V with increasing the Al thickness. Furthermore, the maximum capacitance also increases gradually with increasing the thickness of Al layer, and the voltages at the maximum capacitance also decrease with the Al thickness. These results indicate that the thick Al interlayer can effectively improve the electron injection and inhibit the accumulation of charges in the CGU, thus to decrease effectively the operation voltages and to improve power efficiency for tandem OLEDs as shown in Fig. 2 (b) and (d). To elucidate the effect of the thickness of Al layer on the electron injection property of LiF/Al, the morphology evolution of Al layer with different thickness was investigated by using AFM. Fig. 6 shows the AFM images of the thermally evaporated Al films with different thickness on the TmPyPb (40 nm) surface. It can be observed that the surface morphology of Al films is quite smooth whatever the film thickness. The average roughness changes from 0.46 to 0.88 nm with increasing Al thickness. No Al nanostructure is evidenced, indicating that continuous Al films are formed on TmPyPb, which may be attributed to a perfect wetting of Al on TmPyPb. The dependence of the electron-injection ability on the thickness of Al layer cannot be attributed to the morphological change. As we know, the sheet resistance is strongly dependent on the thickness of Al films. The thick Al layer would result in improvements in both electrical conductivity and carrier injection. Additionally, it has been found that the surface of Al film could be oxidized rapidly into Al2O3 even in high vacuum [55]. This oxide layer would weaken the electron-injection ability of LiF/Al especially for the thinner Al layer. Therefore, the electron injection property of LiF/Al increased gradually with increasing the thickness of Al layer. Additionally, it is noted that the performance of the device worsens when the thickness of Al layer is above 5 nm, although the thick Al layer is beneficial to charge generation and separation. As we know, the high transmittance of the CGU is very important for high efficiency tandem OLED. Fig. 7 shows the optical transmission spectra of different thickness Al layers deposited on ITO substrates. As shown in Fig. 7, we note that the optical transparency gradually decreases with the thickness of Al layer in the visible spectra region due to reflection and absorption. Compared to ITO substrate, the optical transparency at the main emission peak of FIrpic (472 nm) decreases by 0.6% for 1 nm Al, 3.9% for 3 nm Al, 8.2% for 5 nm Al and 14.0% for 7 nm Al. The thick Al layer weakens the light output of the tandem device. Therefore, under the optimal Al thickness of 5 nm, CGU possesses better capability of charge generation and higher light output, which results in excellent EL performance for the tandem device. Encouraged by the impressive results obtained from the tandem blue device, a non-doped tandem white device with a structure of ITO/ PEDOT:PSS (40 nm)/EL unit/CGU/EL unit/LiF(1 nm)/Al(100 nm) was developed. The EL unit consists of NPB (30 nm)/TCTA (10 nm)/PO-01 (0.1 nm)/mCP (3 nm)/Firpic (0.25 nm)/TmPyPb (40 nm), where 0.1 nm PO-01 and 0.25 nm Firpic are used as non-doped blue and yellow EMLs, respectively. Fig. 8 shows the EL characteristics of the single unit and tandem white devices. The detailed EL performance of the tandem and single unit OLEDs are summarized in Table 1. As shown in Fig. 8, the tandem white device presents a turn-on voltage of 6.0 V, maximum luminance of 35150 cd/m2 at a voltage of 15.0 V and maximum power efficiency of 42.6 lm/W. The CEmax and EQEmax of the tandem white device reach 94.9 cd/A and 31.6% at a luminance of 90 cd/m2, respectively, which are about two times as much as 50.9 cd/ A and 17.8% of the single unit device. More importantly, the CE and PE slightly decrease to 87.4 cd/A and 33.3 lm/W at 1000 cd/m2,
Fig. 4. (a) UV-VIS-NIR absorption spectra of NPB (10 nm) and NPB:HAT-CN (1:1, 10 nm) films deposited on quartz substrate. Inset: relative intensity of charge transfer complex bands. (b) PL intensities of the NPB (20 nm) and HAT-CN (10 nm)/NPB (20 nm) films deposited on ITO substrate.
Fig. 5. Capacitance-voltage characteristics of the capacitance devices. The frequency is set at a constant value of 1 kHz, C0 is the geometric capacitance.
The double 20 nm LiF insulating layers are introduced to prevent charge carrier injection from the anode and cathode. Fig. 5 shows the CV characteristics of the devices with different thickness of Al layer in CGU, where the measured capacitance (C) is normalized by the geometric capacitance (C0) determined by the equation of C0 = ε0εA/d [54]. In this equation, ε is the dielectric constant, ε0 is the vacuum permittivity, A is the contact area, and d is the thickness of the functional layers between the electrodes. It can be seen that the capacitance is almost constant under reverse biases for all the devices, indicating 357
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 6. AFM morphology of Al films deposited on 40 nm-thick TmPyPb for different Al thickness: (a) 1.0 nm, (b) 3.0 nm, (c) 5.0 nm, and (d) 7.0 nm.
device displays lower intensity compared with that of the single unit white device due to absorption of blue light by PO-01. The similar phenomenon was also observed in other tandem white devices [18,26]. As can be seen in Fig. 9 (b), the EL spectra of the white devices are origined from the emission of FIrpic and PO-01. The devices show good white emission with CIE coordinates of (0.35, 0.42) for single unit device and (0.32, 0.38) for tandem device. The excellent performance of the tandem white device indicates that ultrathin EML and non-doped CGU can be used to fabricate high performance white OLEDs for lighting. 4. Conclusion In summary, we present highly efficient non-doped tandem OLEDs using ultrathin phosphorescent dyes as EMLs and LiF/Al/HAT-CN as CGU. It was found that the performance of the tandem OLEDs strongly depends on the thickness of Al layer in CGU. From the current densityvoltage and capacitance-voltage characteristics of the external-carrier excluding devices, the charge carriers are generated at the HAT-CN/ NPB interface though disassociation of the charge-transfer complexes formed between HAT-CN and NPB, and Al/LiF layers work as assistant electron injection layers to facilitate electron extraction from CGU and injection into the adjacent ETL. We have found that the thick Al layer is beneficial to charge separation, while it weakens the light output of the tandem device. Under optimal Al thickness of 5 nm, CGU possesses better capability of charge generation and higher light output. The tandem blue device reaches a CEmax of 53.1 cd/A, which is about two times higher than that (25.5 cd/A) of the single unit device. Furthermore, based on the CGU of LiF/Al/HAT-CN, the non-doped
Fig. 7. Optical transmission spectra of Al layers with different thickness deposited on ITO substrates.
respectively, showing low efficiency roll-off in the tandem device. Additionally, compared to the tandem blue device with twofold enhancement in the CE, it is noted that the CE of the tandem white device is slightly less than two times of that of the single unit white device. In other words, there is partial energy loss in the tandem white device. The reason may be attributed to the absorption of dye PO-01 in another EL unit. As shown in Fig. 9 (a), the EL peak of FIrPic is approximate 472 nm, which has a partial overlap with the absorption spectrum of PO-01, so the photons radiated from Firpic can be absorbed by PO-01 resulting in partial energy loss. This blue light loss can be confirmed through comparing the EL spectra of the single unit and tandem white devices. As shown in Fig. 9 (b), the blue emission of the tandem white 358
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
Fig. 8. Characteristics of the single unit and tandem white PhOLEDs: (a) current density-voltage, (b) luminance-voltage, (c) current efficiency-luminance, and (d) EQE-luminance.
application with simplified device structures and high performance tandem OLEDs for display and lighting applications. Acknowledgements This work was supported by the National Key Basic Research Program of China (2014CB648300 and 2017YFB0404500), the NSFC (61774088, 61705112, 21422402 and 21674050), the NSF of Jiangsu Province (BK20161519 and BK20170913), the NUPT “1311 Project” and the Scientific Research Foundation of NUPT (NY214178, NY214093 and NY215076), the National Synergistic Innovation Center for Advanced Materials, the Synergistic Innovation Center for Organic Electronics and Information Displays, and the Project Funded by the PAPD of Jiangsu Higher Education Institutions (YX03001). References [1] X. Zhang, Q. Hu, J. Lin, Z. Lei, X. Guo, L. Xie, W.-Y. Lai, W. Huang, Appl. Phys. Lett. 103 (2013) 153301. [2] Y.-D. Jiu, C.-F. Liu, J.-Y. Wang, W.-Y. Lai, Y. Jiang, W.-D. Xu, X.-W. Zhang, W. Huang, Polym. Chem. 6 (2015) 8019. [3] S. Cao, L. Hao, W.-Y. Lai, H. Zhang, Z. Yu, X. Zhang, X. Liu, W. Huang, J. Mater. Chem. C 4 (2016) 4709. [4] X. Zhang, X. Guo, Y. Chen, J. Wang, Z. Lei, W.-Y. Lai, Q. Fan, W. Huang, J. Lumin. 161 (2015) 300. [5] Y. Tao, X. Guo, L. Hao, R. Chen, H. Li, Y. Chen, X. Zhang, W.-Y. Lai, W. Huang, Adv. Mater 27 (2015) 6939. [6] C. Tang, T. Yang, X. Cao, Y. Tao, F. Wang, C. Zhong, Y. Qian, X. Zhang, W. Huang, Adv. Opt. Mater 3 (2015) 786. [7] B. Liu, F. Dang, Z. Tian, Z. Feng, D. Jin, W. Dang, X. Yang, G. Zhou, Z. Wu, ACS Appl. Mater. Interfaces 9 (2017) 16360. [8] X. Yang, G. Zhou, W.-Y. Wong, Chem. Soc. Rev. 44 (2015) 8484. [9] Q.X. Guo, H.D. Sun, J.X. Wang, D.Z. Yang, J.S. Chen, D.G. Ma, J. Mater. Chem. C 4 (2016) 376. [10] Y.F. Dai, H.M. Zhang, Z.Q. Zhang, Y.P. Liu, J.S. Chen, D.G. Ma, J. Mater. Chem. C 3 (2015) 6809. [11] H.D. Sun, Q.X. Guo, D.Z. Yang, Y.H. Chen, J.S. Chen, D.G. Ma, ACS Photonics 2 (2015) 271. [12] M.-K. Fung, Y.-Q. Li, L.-S. Liao, Adv. Mater 28 (2016) 10381. [13] Y.H. Chen, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, J. Appl. Phys. 110 (2011) 074504. [14] Q. Guo, D. Yang, J. Chen, X. Qiao, T. Ahamad, S.M. Alshehri, D. Ma, J. Appl. Phys. 121 (2017) 115502. [15] H. Sun, Y. Chen, J. Chen, D. Ma, IEE J. Sel. Top. Quantum Electron 22 (2016) 7800110. [16] H.D. Sun, Y.H. Chen, L.P. Zhu, Q.X. Guo, D.Z. Yang, J.S. Chen, D.G. Ma, Adv. Electron. Mater 1 (2015) 1500176. [17] D.Y. Zhou, H.Z. Siboni, Q. Wang, L.S. Liao, H. Aziz, J. Appl. Phys. 116 (2014)
Fig. 9. (a) Absorption spectrum of PO-01 and EL spectrum of FIrPic. (b) EL spectra of the single unit and tandem white PhOLEDs.
tandem white OLED with ultrathin EMLs presents a CEmax of 94.9 cd/A and EQEmax of 31.6%. These results demonstrate that the non-doped CGU structure and ultrathin EML are promising candidates for 359
Organic Electronics 53 (2018) 353–360
X. Zhang et al.
[36] N.A. Talik, K.H. Yeoh, C.Y.B. Ng, B.K. Yap, K.L. Woon, J. Lumin. 154 (2014) 345. [37] Y. Zhao, J. Chen, D. Ma, ACS Appl. Mater. Interfaces 5 (2013) 965. [38] J. Liu, X. Shi, X. Wu, J. Wang, Z. Min, Y. Wang, M. Yang, C.H. Chen, G. He, Org. Electron 15 (2014) 2492. [39] L. Zhu, Y. Zhao, H. Zhang, J. Chen, D. Ma, J. Appl. Phys. 115 (2014) 244512. [40] S. Liu, B. Li, L. Zhang, S. Yue, Appl. Phys. Lett. 98 (2011) 163301. [41] Y.K. Kim, J.W. Kim, Y. Park, Appl. Phys. Lett. 94 (2009) 063305. [42] S.M. Park, Y.H. Kim, Y. Yi, H.-Y. Oh, J.W. Kim, Appl. Phys. Lett. 97 (2010) 063308. [43] S. Lee, J.-H. Lee, J.-H. Lee, J.-J. Kim, Adv. Funct. Mater 22 (2012) 855. [44] T. Matsushima, H. Murata, Appl. Phys. Lett. 95 (2009) 203306. [45] T. Matsushima, G.-H. Jin, Y. Kanai, T. Yokota, S. Kitada, T. Kishi, H. Murata, Org. Electron 12 (2011) 520. [46] A. Mityashin, Y. Olivier, T. Van Regemorter, C. Rolin, S. Verlaak, N.G. Martinelli, D. Beljonne, J. Cornil, J. Genoe, P. Heremans, Adv. Mater 24 (2012) 1535. [47] A.R. Yu, Y. Pei, R.C. Yi, S.B. Liu, M. Chu, H.M. Yu, Y.J. Tang, Z.S. He, X.Y. Hou, Org. Electron 46 (2017) 121. [48] S.-J. Yoo, J.-H. Chang, J.-H. Lee, C.-K. Moon, C.-I. Wu, J.-J. Kim, Sci. Rep. 4 (2014) 3902. [49] S.-J. Yoo, J.-J. Kim, Macromol. Rapid Comm. 36 (2015) 984. [50] L.X. Wang, C.S. Shi, Q.X. Guo, J.S. Chen, X.F. Qiao, D.G. Ma, T. Ahamad, S.M. Alshehri, S.S. Bae, J. Appl. Phys. 122 (2017) 125501. [51] X. Zhang, Z. Wu, D. Wang, D. Wang, R. He, X. Hou, Appl. Surf. Sci. 256 (2010) 4468. [52] T.-W. Ng, M.-F. Lo, Q.-D. Yang, M.-K. Fung, C.-S. Lee, Adv. Funct. Mater 22 (2012) 3035. [53] T. Matsushima, H. Murata, Org. Electron 14 (2013) 1149. [54] K.S. Yook, S.O. Jeon, S.Y. Min, J.Y. Lee, H.J. Yang, T. Noh, S.K. Kang, T.W. Lee, Adv. Funct. Mater 20 (2010) 1797. [55] G. Maidecchi, G. Gonella, R.P. Zaccaria, R. Moroni, L. Anghinolfi, A. Giglia, S. Nannarone, L. Mattera, H.-L. Dai, M. Canepa, F. Bisio, ACS Nano 7 (2013) 5834.
223708. [18] Q. Wang, J.Q. Ding, Z.Q. Zhang, D.G. Ma, Y.X. Cheng, L.X. Wang, F.S. Wang, J. Appl. Phys. 105 (2009) 076101. [19] Y.H. Lee, M.W. Lin, T.C. Wen, T.F. Guo, J. Appl. Phys. 114 (2013) 154512. [20] L.-S. Liao, W.K. Slusarek, T.K. Hatwar, M.L. Ricks, D.L. Comfort, Adv. Mater 20 (2008) 324. [21] Q.D. Ou, L. Zhou, Y.Q. Li, J.D. Chen, C. Li, S. Shen, J.X. Tang, Adv. Opt. Mater 3 (2015) 87. [22] D.Y. Zhou, F.S. Zu, Y.J. Zhang, X.B. Shi, H. Aziz, L.S. Liao, Appl. Phys. Lett. 105 (2014) 083301. [23] T.W. Lee, T. Noh, B.K. Choi, M.S. Kim, D.W. Shin, J. Kido, Appl. Phys. Lett. 92 (2008) 043301. [24] L.S. Liao, K.P. Klubek, C.W. Tang, Appl. Phys. Lett. 84 (2004) 167. [25] F. Yan, R. Chen, H.D. Sun, X.W. Sun, Appl. Phys. Lett. 102 (2013) 203303. [26] M.H. Ho, T.M. Chen, P.C. Yeh, S.W. Hwang, C.H. Chen, Appl. Phys. Lett. 91 (2007) 233507. [27] Y.H. Chen, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, Appl. Phys. Lett. 99 (2011) 103304. [28] Y.H. Deng, Q.D. Ou, Q.K. Wang, H.X. Wei, Y.Q. Li, S.T. Lee, J.X. Tang, J. Mater. Chem. C 2 (2014) 1982. [29] Y.B. Zhao, S.T. Tan, H.V. Demir, X.W. Sun, Org. Electron 23 (2015) 70. [30] S.L. Lai, M.Y. Chan, M.K. Fung, C.S. Lee, S.T. Lee, J. Appl. Phys. 101 (2007) 014509. [31] Y.H. Chen, Q. Wang, J.S. Chen, D.G. Ma, D.H. Yan, L.X. Wang, Org. Electron 13 (2012) 1121. [32] Z. Ma, J. Zhao, X. Wang, J. Yu, Org. Electron 30 (2016) 136. [33] H. Sasabe, K. Minamoto, Y.J. Pu, M. Hirasawa, J. Kido, Org. Electron 13 (2012) 2615. [34] K.A. Knauer, E. Najafabadi, W. Haske, M.P. Gaj, K.C. Davis, C. Fuentes-Hernandez, U. Carrasco, B. Kippelen, Org. Electron 14 (2013) 2418. [35] B. Jiao, Z.X. Wu, Z.Y. Yang, X. Hou, Phys. Status Solidi A 210 (2013) 2583.
360