We report on an investigation of water-processable triple-stacked hole-selective layers for solution-processable organic semiconducting devices using a simple horizontal-dip (H-dip) coating technique. Homogeneous layers were successfully deposited via H-dip-coating using aqueous solutions of graphene oxide (GO), molybdenum oxide (MoO3), and poly(ethylenedioxy thiophene):poly(styrene sulfonate) (PEDOT:PSS). The use of the triple-stacked GO/MoO3/PEDOT:PSS layers as hole-injecting layers (HILs) in solution-processable organic light-emitting diodes (OLEDs) resulted in a considerable improvement of device performance in terms of brightness (maximum brightness: 47,000 cd/m2) as well as efficiency (peak efficiency: 31.5 cd/A), exceeding those of an OLED with a conventional single PEDOT:PSS HIL. Furthermore, polymer solar cells (PSCs) with these triple-stacked layers used as hole-collecting layers (HCLs) showed a considerable improvement in power conversion efficiency (6.62%), which was also higher than that (5.65%) obtained using the single PEDOT:PSS HCL. These results clearly indicate the benefits of using triple-stacked GO/MoO3/PEDOT:PSS layers, which provide better hole-injection/collection, electron-blocking, and improved stability for high performance solution-processable OLEDs and PSCs.
© 2015 Optical Society of America
The recent intense research interest in solution-processable organic and polymeric semiconducting devices continues, including the investigation of organic or polymer light-emitting diodes (OLEDs or PLEDs) and organic photovoltaics (OPVs) [1–16]. In particular, some of the focus has been on the development of new semiconducting materials and device structures to obtain cost-effective, low-temperature devices that are both fast and simple to fabricate and offer as large a processing area as possible [1–14]. In order to achieve these ambitious goals, research efforts have mainly been directed at improving the efficiency, stability, and simplicity of the device fabrication process. The performance of solution-processable organic devices has improved significantly in recent years; for example, following the introduction of phosphorescent guest dopants into their emission layers (EMLs), some solution-processable OLEDs now have a luminosity of about 53 lm/W . In another example, polymer solar cells (PSCs) have been produced with power conversion efficiencies (PCEs) of over 10.6% . Furthermore, important advances have been made in solution-based processes such as roll-to-roll coating using a variety of techniques including the use of new device structures and the development of new organic compounds [7, 9, 10]. However, to date the performance of solution-processed devices in general and OLEDs in particular is still inferior to that of devices prepared using conventional and more complicated vacuum-deposition processes. There is therefore a need to improve the performance and stability of solution-processed devices further, to bring them to the same level as vacuum-deposited devices. One approach that could be used to improve efficiency and stability is to reduce the height of the barriers at the electrode contacts with the organic/polymeric active layers, in order to maximize the efficiency of the charge carrier injection or extraction. For cathode contacts, the barrier height can be optimized by adjusting the work function of the metal cathode to the lowest unoccupied molecular orbital (LUMO) of the active layer [17–19]. For anode contacts, the barrier height can be optimized by applying a hole-selective layer on top of an anode material such as indium-tin-oxide (ITO), which is the most commonly used transparent anode. Because the work function (4.5-5.0 eV) of ITO is significantly lower than the typical highest occupied molecular orbital (HOMO) of the active layer, the hole-selective layer must have an energy level that matches as closely as possible the energy level that lies between the HOMO of the active layer and the work function of the ITO anode [17–19]. There have been several attempts to use a hole-selective layer to match the energy bands in such systems, as well as to reduce the roughness to obtain a uniform surface. Among these, a water-processable p-type transparent conducting polymeric layer of poly(3,4-ethylenedioxythiophene): poly(styrenesulfonate) (PEDOT:PSS) has been used to great effect as the hole-selective layer, either as the hole injection layer (HIL) or as the hole collecting layer (HCL). However, the sulfate ions in PEDOT:PSS make it highly acidic (pH ~1.5), and this acidity can damage the surface of the ITO and dissolve the ions that migrate into the stacked active layer. This results in a gradual reduction in device performance over time [20, 21]. To minimize such undesirable side effects in solution-processable devices, other neutral hole-selective layer materials that are chemically stable, mechanically uniform, and solution-processable, are needed.
Some consideration was recently given to the use of stable, aqueous-processable layers of neutral graphene oxide (GO) sheets as hole-selective layers for ITO anodes [22–25]. Water-soluble GO, which consists of a two-dimensional sheet of graphene functionalized with an oxygen group, has a unique heterogeneous electronic structure due to the presence of mixed sp2 and sp3 hybridizations . These properties, combined with its favorable work function, a fairly large band gap, and the absence of problems associated with ITO erosion, make it a good candidate material for the hole-selective layer in organic devices . The effectiveness of GO films as hole-selective layers has been demonstrated in PSCs consisting of bulk heterojunction (BHJ) PV layers of poly(3-hexylthiophene) (P3HT) and fullerene-derivative [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) blends, which together provide a PCE of 3.25% . It has also been reported that the presence of a layer of GO beneath the PEDOT:PSS to form a GO/PEDOT:PSS stack can protect the ITO from ion diffusion and acid etching, and can extract the holes from the active PV layer. The use of such double-stacked GO/PEDOT:PSS HCLs in OLEDs and PSCs can provide a device performance that is either comparable with or slightly inferior to those of reference devices with a conventional single PEDOT:PSS layer [23, 25].
Due to their favorable energy level alignments, transition metal oxides such as MoO3, V2O5, NiO, and WO3 have recently been investigated as alternatives to PEDOT:PSS in PSCs [27–30]. These materials can be deposited using a variety of methods; for example, an improved device performance has been shown for polymer- and small-molecule-based solar cells with a sol-gel processed MoO3 film, with performances comparable to those obtained in PSCs with PEDOT:PSS layers . However, most conventional hole-selective layers still produce performances comparable with those of single PEDOT:PSS layers. Very few results have been reported showing improvements in performance that have resulted from the use of MoO3 for hole collection in PSCs [31, 32]. Thus, despite recent developments, further development of solution-processable hole-selective layer(s) is therefore clearly required to achieve improved and balanced hole-injection/extraction and superior electron-blocking properties, as well as to develop novel device structures that yield high performance and stability. Such properties would make the performance of these layers superior to that of PEDOT:PSS while avoiding its shortcomings, and enable their use in solution-processable OLEDs, PSCs, and other organic electronic devices.
We herein describe our attempts to overcome the limitations of conventional hole-selective layers. We report on the fabrication and use of thin and homogeneous triple-stacked layers comprised of GO, MoO3, and PEDOT:PSS films (GO/MoO3/PEDOT:PSS) fabricated by horizontal-dip (H-dip) coating [13, 25] on ITO substrates, instead of a conventional single PEDOT:PSS layer or double-stacked GO/PEDOT:PSS layers. We discuss the influence of these water-processable triple-stacked layers on the performance of solution-processable OLEDs and PSCs. In this study, by employing H-dip-coated triple-stacked layers of GO/MoO3/PEDOT:PSS, we obtained a high performance and efficiency of solution-processed OLEDs and PSCs; for the OLED, the maximum luminance and luminance efficiency were 47,000 cd/m2 and 31.5 cd/A, respectively, and the current efficiency was maintained at 19.8 cd/A even at value of luminance as high as 35,000 cd/m2. For our PSC, a maximum PCE of 6.62% was achieved. These values are much higher than those achieved by equivalent devices with conventional single layers of PEDOT:PSS. We attribute such improved device performance to the water-processable triple HILs of GO/MoO3/PEDOT:PSS that allow the formation of a high-quality film and provide good matching of the energy levels between adjacent layers, with improved hole-injecting/extracting and electron-blocking properties, together with device stability. We also demonstrate the feasibility of fabricating large-area, high-performance solution-processable OLEDs and PSCs using H-dip-coated triple-stacked layers of GO/MoO3/PEDOT:PSS.
As shown in Fig. 1(a), the proposed solution-processed semiconducting layers were fabricated on glass substrates pre-coated with an 80 nm thick layer of ITO with a sheet resistance of 30 Ω per square. The ITO substrates were ultrasonically cleaned with detergent, deionized water, acetone and isopropanol, and then dried by blowing nitrogen over them, followed by ultra-violet ozone cleaning for 15 min. On the ITO anode, a GO layer was solution-coated using an aqueous GO dispersion that was synthesized from natural graphite according to a modified version of Hummer’s method . The GO layers coated on the ITO substrates were then annealed at 150°C in a dry N2 environment for 5 min. Another MoO3 layer was solution-coated using an aqueous MoO3 solution prepared using ammonium heptamolybdate ((NH4)6Mo7O24·4H2O, Sigma-Aldrich) dissolved in deionized water at a desired concentration (0.5 wt%) as a precursor . The precursor layers coated on the ITO substrates were then annealed at 150°C in a dry N2 environment for 5 min. The precursor layer was then decomposed into three components, MoO3, NH3, and H2O, of which the NH3 and H2O evaporated, leaving MoO3 as the only component in the layer. The other layer of PEDOT:PSS was also solution-coated on the substrates and then baked at 120°C in a vacuum oven for 20 min to extract the residual water. The PEDOT:PSS solution used was a mixture of PEDOT:PSS solution (CLEVIOS PVP AI 4083, H. C. Starck, 1%) and isopropyl alcohol with a weight ratio of 1:3.
Using simple H-dip solution-coating [13, 25], various types of hole-selective layers were prepared on ITO anodes; single-, double-, and triple-stacked layers were created by combining stacks of GO, MoO3 and PEDOT:PSS layers. A small volume of the solution (~5-10 μl) per unit coating area (1 cm × 1 cm) was fed into the gap of the cylindrical H-dip head using a syringe pump (NE-1000, New Era. Pump Systems, Inc.). The height of the gap h0 was adjusted vertically and the carrying speed U was controlled by means of a computer-controlled translation stage (SGSP26-200, Sigma Koki Co., Ltd.). After a meniscus of the coating solution had formed on the substrate, the substrate was transported horizontally, so that the H-dip head spread the solution evenly on the transporting substrate while maintaining the shape of the downstream meniscus of the solution.
OLEDs were fabricated by coating a blended EL solution to form an EML on the ITO layer, which was pre-coated with the hole-selective layer(s) as the HIL(s). For comparison, several types of HIL(s) were prepared on the ITO layer using the H-dip process. For the blended EL solution, we used a hole-transporting material of N,N'-diphenyl-N,N'-bis(3-methylphenyl)-1,1'biphenyl-4,4'-diamine (TPD, Sigma-Aldrich), an electron-transporting material of 2-(4-Biphenylyl)-5-(4-tert-butylphenyl)-1,3,4-oxadiazole (Bu-PBD, Sigma-Aldrich), a green-emitting phosphor of tris(2-Phenylpyridinato) iridium (Ir(ppy)3, Lumtec), and a host polymer of poly(vinylcarbazole) (PVK, Sigma-Aldrich) without further purification, in mixed solvents (2.5 wt%) of 1,2-dichloroethane and chloroform (3:1) (viscosity ca.1.0 cp) [13, 34]. Here, the HOMO/LUMO energy levels were 5.4/2.4 eV for TPD , 6.0/2.3 eV for Bu-PBD , 5.6/3.0 eV for Ir(ppy)3 , and 5.6/2.2 eV for PVK , while the conduction band (CB)/ valence band (VB) were 4.8-5.5/7.9-9.0 eV for sol-gel-processed MoO3 . The thickness of the EML was fixed at about 85 nm; 2 nm of Cs2CO3 (Sigma-Aldrich) electron-injecting buffer layer and 50 nm of Al cathode were evaporated sequentially on the EML via thermal deposition (0.5 nm/s) at a base pressure below 2.7 × 10−4 Pa. The effective size of the light-emitting area was 3 mm × 3 mm.
The PSCs were fabricated by solution-coating a blended PV solution to form a PV layer on the ITO layer, which was also pre-coated with hole-selective layer(s) as the HCL(s). For comparison, several types of HIL(s) were prepared on the ITO layer using the H-dip process. For the blended PV solution, we used poly[N-9'-heptadecanyl-2,7-carbazole-alt-5,5-(4',7'-di-2-thienyl-2',1',3′-benzothiadiazole)] (PCDTBT, 1-material Chemscitech, Inc.) and a fullerene derivative [6,6]-phenyl-C71- butyric acid methyl ester (PCBM70, Nanostructured Carbon, Inc.) in a weight ratio of 1:4 in a solvent of 1,2-dichlorobenzene . The thickness of the PV layer was fixed at 85 nm. The coated PV layer was then dried at 65°C for 60 min in order to remove the remaining solvents. Then, 1 nm of the Al:Li electron-collecting layer and 50 nm of the Al cathode were evaporated sequentially on the PV layer via thermal deposition (0.5 nm/s) at a base pressure below 2.7 × 10−4 Pa.
To investigate the surface morphology of the functional layer, the surface of the layer was monitored using atomic force microscopy (AFM, Nanosurf easyscan2 AFM, Nanosurf AG Switzerland, Inc.). The optical properties of the fabricated layer were investigated using a UV-visible spectroscopy system (8453, Agilent). For the OLEDs, a Chroma Meter CS-200 (Konica Minolta Sensing, Inc.) and a source meter (2400, Keithley) were used to measure the EL properties and current density-voltage-luminance (J-V-L) characteristics. The performance of the PSCs was measured under an illumination intensity of 100 mW/cm2 generated by an AM 1.5G light source (96000 Solar Simulator, Newport). The photocurrent characteristics were measured using a source meter (2400, Keithley) and calibrated using a reference solar cell (BS-520, Bunkoh-keiki). The characterization of the device was carried out at room temperature under ambient conditions, without encapsulation.
3. Results and discussion
3.1 H-dip-coating of sol-gel processed MoO3 layers
Figure 1(a) shows a schematic illustration of the H-dip-coating process used in this study as solution-coating for the GO, MoO3, and PEDOT:PSS layers [13, 25]. The use of H-dip-coating allows the thickness of the layer to be precisely controlled because the thickness (h) of the coated layer increases with increasing capillary number (Ca = μU/σ) of the coating solution, where μ and σ represent the viscosity and surface tension of the solution, respectively, and U is the coating speed. The thickness h of the H-dip-coated layer can be described by the associated drag-out problem proposed by Landau and Levich for Ca << 1 as h = k· Ca2/3· Rd, where Rd represents the radius of the associated downstream meniscus and k is a constant of proportionality [12, 25, 39]. As a representative example, we showed the thickness of the sol-gel processed MoO3 layer as a function of the coating speed U and gap height h0 (Fig. 1(a)). For an h0 of 0.6 mm, the thickness of the H-dip-coated MoO3 layer showed a continuous increase from about 5 nm to 20 nm when U was increased from 0.3 cm/s to 2.0 cm/s. Furthermore, when h0 was increased to 0.7 mm, the thickness of the H-dip-coated MoO3 layer also showed a further increase with U. These results agree well with the theoretical predictions (solid lines in Fig. 1(a)). It is thus clear that H-dip-coating allows the precise nanoscale thickness control of a sol-gel processed MoO3 film, with U and h0 as the controlling parameters. The optical properties of the representative hole-selective layers including the MoO3 film (thickness: ~7 nm) are shown in Fig. 1(b). In the visible range (400-700 nm), the H-dip-coated MoO3 films on ITO substrates showed high transmittance minima of about 76% at about 400 nm; this is comparable to the transmittance minima of the ITO substrate coated with 4.5 nm of GO and of the bare ITO substrate. Note that the characteristic transmittance spectrum of the ITO substrate is mainly a result of the difference in the refractive index between the ITO layer (~1.90 at 550 nm for a thickness of 80 nm) and the glass (~1.50 at 550 nm). From the optical absorption edge of the H-dip-coated MoO3 films (not shown), we estimate the value of the band gap of the H-dip-coated MoO3 film to be 3.04 eV; this is identical to the band gap (3.1 eV)  of the sol-gel-processed MoO3 film. The optical properties of the H-dip-coated MoO3 film were thus almost the same as those of the spin-coated film.
Next, in order to investigate the film quality of the fabricated MoO3 layers, we monitored the variation in the surface roughness of the H-dip-coated MoO3 films on the flat ITO glass substrates (rms roughness ~2.5 nm) using AFM (not shown). The use of H-dip-coating leads to the formation of high-quality films for the functional GO and MoO3 materials. The AFM results show that the topography was fairly uniform and smooth; the values of the rms surface roughness for MoO3, GO, GO/MoO3, and GO/MoO3/PEDOT:PSS were 0.63, 0.80, 0.99, and 0.92 nm, respectively, indicating good coverage of each coated layer. Moreover, the values of surface roughness were identical at the different positions of the investigated layers. This uniformity was achieved because of the conditions under which the layers were formed, which did not involve the application of any external force, such as the centrifugal force in the conventional spin-coating method, for example [8–11]. The resulting smoothness and uniformity of the MoO3 layers clearly indicate that the functional layers, including the MoO3 layers, produced by H-dip-coating may be suitable for the fabrication of solution-processed OLEDs and PSCs. In comparison, rms surface roughnesses observed for the 40-nm-thick H-dip-coated and spin-coated PEDOT:PSS layers were ~0.8 nm and ~0.9 nm, respectively. In addition, fairly smooth surfaces with rms surface roughnesses of ~0.9 and ~1.0 nm were also observed for both the MoO3/GO and MoO3/GO/PEDOT:PSS HILs, respectively.
3.2 OLEDs fabricated using H-dip-coated multi-stacked HILs
Given the film qualities of the functional layers prepared using the H-dip-coating process, we fabricated solution-processable green OLEDs that included HILs of the H-dip-coated GO, MoO3, and PEDOT:PSS, and investigated the characteristics of these OLEDs in order to assess the hole-injecting properties of the HILs. The following OLED device configuration was used: ITO anode/H-dip-coated HIL(s)/EML(85 nm)/electron-injecting Cs2CO3 buffer layer (2 nm)/Al cathode (50 nm) (Fig. 2(a)). For comparison, we fabricated four representative OLEDs with various HILs: an OLED with H-dip-coated triple-stacked HILs of GO/MoO3/PEDOT:PSS (OLED device 1), an OLED with H-dip-coated triple-stacked HILs of MoO3/GO/PEDOT:PSS (OLED device 2), an OLED with an H-dip-coated single HIL of PEDOT:PSS (reference OLED 1), and an OLED with H-dip-coated double-stacked HILs of GO/PEDOT:PSS (reference OLED 2). We found that OLED performance was maximized when the thicknesses of the functional layers in the HILs were 4.5 nm, 7.3 nm, and 40.0 nm for GO, MoO3, and PEDOT:PSS, respectively. In our case, all the variables were the same except for the structure of the HILs, so the device performance could be directly related to the HIL used. The device performances observed for the studied OLEDs are summarized in Table 1; the J-V-L characteristics of the representative devices are shown in Fig. 2. It is clear from Fig. 2 that for all the representative OLEDs, the slopes of the J-V curves (Fig. 2(b)) indicate excellent diodic behavior and therefore good coverage for the H-dip-coated HILs and EMLs; however, the OLED devices with triple-stacked GO/MoO3/PEDOT:PSS and MoO3/GO/PEDOT:PSS HILs showed clearly different current flow characteristics with a sharper increase in the J-V curve and higher current densities than the reference OLEDs. These J-V curves clearly indicate the noticeably better carrier (hole-) injecting performance of the triple-stacked GO/MoO3/PEDOT:PSS and MoO3/GO/PEDOT:PSS HILs, especially compared to the conventional single PEDOT:PSS HIL. We note that reference OLED 2, fabricated using double-stacked GO/PEDOT:PSS HILs, showed slightly higher current flow characteristics than reference OLED 1 with the single PEDOT:PSS HIL.
Similar to the J-V curves, it is clear from the L-V curves (Fig. 2(c)) that the OLEDs had different electroluminescent (EL) characteristics, with OLED device 1 showing a better performance than the others. For example, for OLED device 1 the EL brightness values are 100 cd/m2, 1,000 cd/m2, and 10,000 cd/m2 at operating voltages of ~4.0, 5.8, and 9.5 V, respectively, with a maximum luminescence (LMAX) of ~47,000 cd/m2 being reached at 15.5 V. It can be also seen from Fig. 2(c) that the onset voltage, defined as the voltage at 1 cd/m2 (VON), is below 2.1 V in OLED device 1, which is lower than that of reference OLED 1 (about 3.0 V), and comparable to that of reference OLED 2 (2.5 V). The lower VON indicates that the hole-injection barrier is greatly reduced by the introduction of multi-stacked HILs. It is also notable that as shown in the inset of Fig. 2(c), single dominant peaks are observed at 510 nm for the sample OLEDs, corresponding to those of the reference OLEDs.
Interestingly, it can also be seen that the efficiencies are significantly greater for the triple-stacked GO/MoO3/PEDOT:PSS HILs, such that the best overall performance was seen for OLED device 1 with the triple-stacked GO/MoO3/PEDOT:PSS HILs. A peak luminance efficiency (LEPEAK) of 31.5 cd/A and a peak power efficiency (PEPEAK) of 20.3 lm/W were achieved (Table 1 and Fig. 2 (d)). Moreover, even for a high luminance of 35,000 cd/m2, the luminance efficiency (LE) of OLED device 1 still reached 19.8 cd/A, which is higher than the values of reference OLED 1 and reference OLED 2 for the same luminance (15.0 and 15.7 cd/A). This clearly indicates that the device performances were significantly better when the triple stacked GO/MoO3/PEDOT:PSS HILs were used.
By contrast, we note that even though the energy levels of the two triple-stacked HILs are similar, the performance of OLED device 2, which contains triple stacked MoO3/GO/PEDOT:PSS HILs, is not that good. To understand the different EL performances of OLED device 1 and OLED device 2, we investigated the surface coverage of the GO sheets in the stacked layers of GO/MoO3 and MoO3/GO using spatially resolved Raman spectra (Raman maps) measured using confocal Raman spectroscopy (WITec alpha300, 632.8 nm, 600 nm spot size, 5 mW, 100 × objective lens). This is a particularly effective means of identifying the structure and degree of disorder in carbon-based materials . The observed Raman spectra of the G (~1590 cm−1) and D (~1320 cm−1) peaks and the Raman maps (100 μm × 100 μm) of the G peak for the H-dip-coated GO sheets in the stacked layers of GO/MoO3 and MoO3/GO on the ITO-coated glass surfaces made by H-dip-coating are shown in Fig. 3. The G peak originates from the sp2 hybridized carbon of the GO sheets. Note that the Raman maps of the D peak corresponding to the breathing modes of the six-membered rings activated by the sp3 defects via the double-resonance process are almost identical to those of the G peak. It can be seen that the Raman intensity ratio (D/G = ~1.6) of our GO sheets shows a similar value to that (~1.6) of conventional GO . Figure 3 also shows that the Raman intensities of the GO in the stacked layers of GO/MoO3 (Fig. 3(a)) are stronger and more homogeneous than those of the GO in the MoO3/GO layers (Fig. 3(b)). By analyzing the intensity distribution in the Raman map, we estimate that more than 91.8% of the mapped area had a Raman intensity over 1,000 counts (bright regions) for the GO/MoO3 stacked layers, implying only a small uncoated or submonolayer-coated area (dark regions: below 1,000 counts, less than 8.2%) among the GO sheets. We note that the 4.5 nm-thick GO film corresponds to about four layers of GO sheets, given that a single-layer GO sheet has a thickness of between 0.8 and 1.2 nm (average 1 nm), and therefore the maximum Raman intensity of 4,000 counts implies four layers of GO sheets, while a Raman intensity of below 1,000 counts indicates a submonolayer of a GO sheet. Interestingly, in contrast with the GO/MoO3 stacked layers, the bright region of the MoO3/GO stacked layers occupied less than 18% of the mapped area. This clearly indicates that H-dip-coating produced a GO film consisting of highly packed GO sheets on ITO, while the H-dip-coated GO film did not fully cover the ITO surface pre-coated with the MoO3 layer; in the latter case, the GO was inhomogeneously and sparsely distributed throughout the layer. It is therefore clear that H-dip-coating yields homogeneous GO/MoO3/PEDOT:PSS HILs that have a high packing density of GO on the ITO anode. This provides a more effective modification of the electrical properties of the ITO anode than the MoO3/GO/PEDOT:PSS HILs, despite the similar energy levels of the two triple-stacked HILs.
3. 3 Hole-only and electron-only devices with triple-stacked HILs of GO/MoO3/PEDOT:PSS
The above results clearly indicate that the triple-stacked GO/MoO3/PEDOT:PSS HILs in OLED device 1 provided the best outcomes among all the HILs studied. The better device performance of OLED device 1 may be attributed to better hole-injecting and/or electron-blocking properties, and the better matching of the energy levels of the GO/MoO3/PEDOT:PSS HILs used between the ITO anode and the EML than in the other devices. To confirm the efficiency of the hole-injecting and/or electron-blocking of the H-dip-coated GO/MoO3/PEDOT:PSS HILs, we also investigated the hole- and electron-current flows versus applied electric field (J–E) characteristics for hole-only and electron-only devices  using [ITO/HILs/EML (150 nm) /Ag (100 nm)] and [ITO/ZnO/HILs/EML (85 nm)/CsF (3 nm)/Al (60 nm)] structures, respectively. During the operation of both the hole-only and the electron-only devices, no light emission was observed from them, confirming the negligible injection of minor charge carriers into the EMLs. As shown in Fig. 4(a), for the hole-only devices, the device with H-dip-coated GO/MoO3/PEDOT:PSS HILs exhibited much higher hole-current flows than the other hole-only devices. Moreover, Fig. 4(b) shows that compared with the other devices, the electron-only device with H-dip-coated GO/MoO3/PEDOT:PSS HILs had much smaller electron-current flows through the HILs. These hole-selecting and electron-blocking abilities of the triple-stacked HILs of GO/MoO3/PEDOT:PSS are largely attributable to the high packing density of the GO sheets and the well matched HOMO energy levels of MoO3 and PEDOT:PSS between the GO and the active layers.
Because the degradation of the ITO anode and the HILs in normal OLEDs is difficult to investigate due to the fast and pronounced degradation of the air-sensitive low-work-function metal cathodes such as Ca or Al, we investigated the hole-injecting characteristics of the pristine and degraded hole-only triple-stacked HILs devices, in order to assess the durability of the device with the triple-stacked HILs. To degrade the hole-only devices, an electrical aging process was used where constant low current flows at J = 50 mA/cm2 were applied to the devices in a nitrogen glovebox at room temperature in the dark [43, 44]. The aging process was carried out for 5 min and the J-V measurements were then performed. The cycle of aging and J-V measurement was then repeated at 5 min intervals; all devices were kept in the dark during the measurements made between the successive electrical agings. The resulting J-V observations for the electrically degraded devices may be explained in terms of the degradation mechanisms generally seen in OLEDs, because the real-time degradation of OLEDs under continuous operation and the variations in J-V characteristics after electrical degradation follow a similar pattern. Figure 5 shows the J-V characteristics of the two representative devices. The hole current of the hole-only device with the single PEDOT:PSS HIL decreased significantly with an increasing number of aging treatments (Fig. 5(a)).
On the other hand, the hole-only device with the triple-stacked GO/MoO3/PEDOT:PSS HILs exhibited relatively small decreases in hole-current flow even after greater numbers of aging cycles (Fig. 5(b)). The decreased hole-current flows for the degraded devices indicate a deterioration in the hole-injecting performance of the ITO and the HILs, which could be due to the generation of trapped charge near the interface between the HILs and the EML [43–45]. After a number of aging cycles, more trapped charges are generated and there is a further increase in the hole-injecting potential, reducing the device performance significantly. The increased hole-injecting potential may also be related to the migration of the indium ions from the ITO anode into the EML layer [20, 21]. To compare the stability of hole-injection in the devices, the changes in the hole-current flows at 24 V for the studied devices are shown in Fig. 6. Inspection of Fig. 6(a) shows that changes in the hole-current flows in the hole-only devices with the multi-stacked HILs are lower than those with the single PEDOT:PSS HIL. Furthermore, for the hole-only devices with the multi-stacked HILs, the device with the triple-stacked GO/MoO3/PEDOT:PSS HILs is more durable than the other devices (Fig. 6(b)). Thus it is clear that the GO/MoO3/PEDOT:PSS HILs are also a durable alternative to the conventional single PEDOT:PSS HIL, although further systematic studies of the long-term stability are needed to understand the fundamental processes underlying the degradation phenomena.
3. 4 Polymer-fullerene BHJ PSCs fabricated with the H-dip-coated triple-stacked HCLs
Our next investigation focused on the hole-collecting ability of the hole-selective layers by measuring the PV performance of PCDTBT:PCBM70 BHJ PSCs . As shown in Fig. 7(a), to investigate the effects of different HCLs on the device performance, we fabricated PSCs using various hole-selective layers as the HCL(s) with a device configuration of ITO anode / H-dip-coated HCL(s) / PV layer of PCDTBT:PCBM70 (85 nm) / electron-collecting Al:Li buffer layer (1 nm) / Al cathode (50 nm). For the PCDTBT:PCBM70 BHJ PSCs fabricated with the different HCL(s), we also investigated the optical properties of the PSCs and found that the PV layers in the PSCs exhibited a strong optical absorption with two dominant peaks at approximately 450 nm and 578 nm; these are mainly due to the π-π* transition in the PCDTBT with the band edge at approximately 660 nm (Fig. 7(a)). The PSCs also show low reflection at wavelengths shorter than the absorption edge (~660 nm) of the PCDTBT, mainly due to the strong optical absorption of the PV layer (Fig. 7(b)). These optical characteristics of the PSCs were found to be similar to those of the PSCs with the different HCLs, despite the differences in the layer structures used for the different HCLs.
Next, we investigated the J-V characteristics of the fabricated PSCs. Figure 8(a) shows the representative J-V curves under dark conditions for four representative examples: a PSC with H-dip-coated triple-stacked GO/MoO3/PEDOT:PSS HCLs (PSC device 1), a PSC with H-dip-coated triple-stacked MoO3/GO/PEDOT:PSS HCLs (PSC device 2), a PSC with a H-dip-coated single PEDOT:PSS HCL (reference PSC 1), and a PSC with H-dip-coated double GO/PEDOT:PSS HCLs (reference PSC 2). The J–V curves in Fig. 8(a) show that the four PSCs clearly behaved as diodes with high rectification ratios of 103-104 at 1.5 V, indicating good coverage of the HCLs and PV layers. However, there were small but clear differences among the current flows; this implies that there were some differences in the flows of the photoexcited charge carriers through the HCLs and the PV layers. The PV characteristics of the devices, measured under AM 1.5G illumination (100 mW/cm2), are summarized in Table 2. The four representative J-V curves are also shown in Fig. 8(b).
For reference PSC 1, a fairly good PV performance was observed under illumination, showing an open-circuit voltage (VOC) of 0.88 V, a short-circuit current density (JSC) of 11.95 mA/cm2, and a fill factor (FF) of 53.5%, corresponding to a PCE of 5.65%. This is similar to the PCEs of the PSCs reported previously . The PV performance of reference PSC 1 is mainly a result of the increased work function (or energy level) of the PEDOT:PSS HCL-coated ITO anode (5.04 eV), which is close to the HOMO energy level of PCDTBT (5.5 eV) , and which raises the built-in potential across the cell. The energy levels of the different HCLs on the ITO anode were estimated by measuring the work functions using an ultraviolet photoelectron spectrometer (Surface Analyzer, Model AC-2, Riken-Keiki. Co.). The observed work functions of the ITO anodes with the different HCLs are summarized in Table 2, showing an insignificant difference between them and the work function of the PEDOT:PSS-coated ITO. Figure 8(b) also shows that the performance of reference PSC 2 is slightly better than that of reference PSC 1, with a VOC of 0.87 V, a JSC of 12.09 mA/cm2, a FF of 56.6%, and a PCE of 5.98%. We mainly attribute this improvement in PV performance not only to the modified energy level but also to the electron-blocking property of the double-stacked GO/PEDOT:PSS HCLs on an ITO anode, which may have been induced by the presence of the GO layer on the ITO electrode (see Table 2 and Fig. 4(b)). By contrast, for PSC device 1 with the triple-stacked GO/MoO3/PEDOT:PSS HILs, an excellent PV performance was observed under illumination, with a VOC of 0.88 V, a JSC of 12.98 mA/cm2, and a FF of 58.2%, corresponding to a PCE of 6.62%. This PCE is much higher than those of reference PSC 1 and reference PSC 2, by 17% and 11%, respectively. Because the work functions of the ITO anode were not significantly different for the different HCLs (Table 2), we attribute the superior PV performance of PSC device 1 not only to the large built-in potential caused by the top PEODT:PSS layer and the close matching of the energy levels of the ITO anode to the bottom GO layer, but also to the electron-blocking properties of the middle MoO3 layer , resulting in the significantly increased FF and JSC. We note that for PSC device 2 with the triple-stacked MoO3/GO/PEDOT:PSS HILs, the observed PV performance was slightly lower than that of PSC device 1, showing a VOC of 0.87 V, a JSC of 12.40 mA/cm2, and a FF of 57.5%, corresponding to a PCE of 6.22%. This may be due to the inhomogeneous distribution of the GO sheets on the MoO3 layer in the triple-stacked HILs (also see Fig. 3). The PV performance of the studied devices is consistent with their incident photon-to-current collection efficiency (IPCE) spectra (Inset in Fig. 8(b)). It is noteworthy that the absorption spectral shapes of the PV layers (Fig. 7(a)) are responsible for the spectral shapes of the IPCE for the studied PSCs (Inset in Fig. 8(b)). It is also notable that the optical absorption, reflection, and IPCE spectra for the studied PSCs (Figs. 7(a), 7(b), and 8(b)) also provide confirmation that the better PCE of the PSC with the triple stacked HCLs is not a result of any optical factor caused by the change in the structure of the stacked HCLs, but is instead due to the electronic properties of the triple-stacked HCLs on the ITO anode (e.g., better hole extraction and electron blocking). It is therefore clear from the above results that the H-dip-coated triple GO/MoO3/PEDOT:PSS HCLs are more effective than the other conventional single PEDOT:PSS or the double GO/PEDOT:PSS HCLs, and we attribute this efficiency not only to the improved hole-selection property of the ITO anode via the modified energy level but also to the electron-blocking property of the stacked layers on the ITO anode, which may have been induced by the presence of the homogeneous GO and MoO3 layer on the ITO electrode.
3. 5 Large-area OLEDs and PSCs with water-processable triple-stacked hole-selective layers
Finally, encouraged by the impressive results obtained from the OLEDs and PSCs fabricated using the triple-stacked hole-selective layers of GO/MoO3/PEDOT:PSS on the ITO anode, we assembled large-area OLEDs and PSCs on 5 cm × 5 cm ITO-coated glass substrates using the H-dip-coating method, in order to assess the processability of these organic semiconducting devices. A photographic image of the devices is shown in Fig. 9. The triple-stacked GO/MoO3/PEDOT:PSS hole-selective layers were deposited on the ITO glass substrates using H-dip-coating to produce OLED lighting devices with a pixel size of about 1.2 cm × 1.2 cm and PSCs with a cell size of about 0.6 cm × 3.5 cm. Although the solution-processed active layers and the triple-stacked hole-selective functional layers of GO/MoO3/PEDOT:PSS in the OLED and PSC devices were fully fabricated in air, Fig. 9 clearly shows that the entire surface of the OLED pixels emits a luminous flux and that the PSC pixels generate electricity directly from visible light. For the OLED, the EL spectra collected from the pixels on the substrate were almost identical to those shown in Fig. 2(c). Moreover, the low variation in intensity in the emission area implies a low variability with the thickness of the solution-coated functional layers. For the PSC, a fairly high voltage output was also observed even under the light conditions of a typical office, as shown in Fig. 9. These results clearly show that the use of triple-stacked GO/MoO3/PEDOT:PSS layers offers the possibility of relatively simple fabrication for OLEDs and PSCs, with easy upscaling to larger sizes at low temperatures using a simple solution-coating process.
All the results described above clearly demonstrate that water-processable triple-stacked GO/MoO3/PEDOT:PSS layers show considerable promise as a potential alternative to the conventional single PEDOT:PSS layer for the fabrication of highly efficient, stable, and large-area OLEDs and PSCs using high-throughput manufacturing such as roll-to-roll production. This could enable the rapid processing of high-performance OLEDs and PSCs. Furthermore, the formation of multi-stacked hole-selective layers by H-dip-coating can also be applied to the design of new organic electronic devices. We also note that due to the easier realization of the desired compositions of the multilayers, the solution-process approach using H-dip-coating is more convenient than the more complicated conventional vacuum evaporation for the fabrication of multi-stacked functional layers.
In summary, we have reported the successful fabrication of highly efficient, stable, and large-area solution-processed OLEDs and PSCs by incorporating water-processable triple-stacked hole-selective layers of GO/MoO3/PEDOT:PSS. Homogeneous, smooth, and thin stacked layers with controllable thicknesses can be successfully deposited from aqueous solutions on an ITO electrode using H-dip-coating. We have shown that the device performances of OLEDs and PSCs with triple-stacked hole-selective GO/MoO3/PEDOT:PSS layers were much better than those observed for a conventional single PEDOT:PSS layer. We ascribe this improvement to their good solution processability, superior hole-injecting/extracting and electron-blocking properties, better matching of the energy level with the active layers, and better durability. Moreover, large-area OLED and PSC devices on 5 cm × 5 cm substrates with high uniformity were successfully fabricated using the GO/MoO3/PEDOT:PSS hole-selective layers. These results clearly demonstrate that the use of homogeneous and smooth hole-selective layers of GO/MoO3/PEDOT:PSS on ITO substrates provides a solid foundation for the further development of high-performance, efficient, stable, and large-area solution-processable OLEDs and PSCs. Furthermore, these innovative triple-stacked layers could be used in the future mass production of highly efficient organic semiconducting devices for use in lighting, displays and photovoltaic and/or optoelectronic applications.
This work was supported by National Research Foundation of Korea (NRF) grant funded by the Korea Government (MEST) (2012R1A2A2A01015654, 2014R1A2A1A10054643); and by Kwangwoon University (2015).
References and links
1. C. W. Tang and S. A. VanSlyke, “Organic electroluminescent diode,” Appl. Phys. Lett. 51(12), 913–915 (1987). [CrossRef]
2. J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Marks, K. Mackay, R. H. Friend, P. L. Burns, and A. B. Holmes, “Light emitting diodes based on conjugated polymers,” Nature 347(6293), 539–541 (1990). [CrossRef]
3. S. R. Forrest, M. A. Baldo, D. F. O’Brien, Y. You, A. Shoustikov, S. Sibley, and M. E. Thompson, “Highly efficient phosphorescent emission from organic electroluminescent devices,” Nature 395(6698), 151–154 (1998). [CrossRef]
4. M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, and S. R. Forrest, “Very high-efficiency green organic light-emitting devices based on electrophosphorescence,” Appl. Phys. Lett. 75(1), 4–6 (1999). [CrossRef]
5. C. Adachi, M. E. Thompson, and S. R. Forrest, “Architectures for efficient electrophosphorescent organic light-emitting devices,” IEEE J. Sel. Top. Quantum Electron. 8(2), 372–377 (2002). [CrossRef]
6. G. He, M. Pfeiffer, K. Leo, M. Hofmann, J. Birnstock, R. Pudzich, and J. Salbeck, “High-efficiency and low-voltage p-i-n electrophosphorescent organic light-emitting diodes with double-emission layers,” Appl. Phys. Lett. 85(17), 3911–3913 (2004). [CrossRef]
7. J. Ouyang, T.-F. Guo, Y. Yang, H. Higuchi, M. Yoshioka, and T. Nagatsuka, “High-performance, flexible polymer light-emitting diodes fabricated by a continuous polymer coating process,” Adv. Mater. 14(12), 915–918 (2002). [CrossRef]
8. F. So, B. Krummacher, M. K. Mathai, D. Poplavskyy, S. A. Choulis, and V.-E. Choong, “Recent progress in solution processable organic light emitting devices,” J. Appl. Phys. 102(9), 091101 (2007). [CrossRef]
9. D. A. Pardo, G. E. Jabbour, and N. Peyghambarian, “Application of screen printing in the fabrication of organic light-emitting devices,” Adv. Mater. 12(17), 1249–1252 (2000). [CrossRef]
10. B.-J. de Gans, P. C. Duineveld, and U. S. Schubert, “Inkjet printing of polymer: State of the art and future developments,” Adv. Mater. 16(3), 203–213 (2004). [CrossRef]
11. S.-R. Tseng, H.-F. Meng, K.-C. Lee, and S.-F. Horng, “Multilayer polymer light-emitting diodes by blade coating method,” Appl. Phys. Lett. 93(15), 153308 (2008). [CrossRef]
12. T. Koyama, S. Naka, and H. Okada, “Investigation of solution-processed organic light-emitting diode fabrication on patterned line structure using bar-coating method,” Jpn. J. Appl. Phys. 51(11R), 112102 (2012). [CrossRef]
14. A. Sandström, H. F. Dam, F. C. Krebs, and L. Edman, “Ambient fabrication of flexible and large-area organic light-emitting devices using slot-die coating,” Nat. Commun. 3, 1002 (2012). [CrossRef] [PubMed]
15. J.-H. Jou, S.-H. Peng, C.-I. Chiang, Y.-L. Chen, Y.-X. Lin, Y.-C. Jou, G.-H. Chen, C.-J. Li, W.-B. Wang, S.-M. Shen, S.-Z. Chen, M.-K. Wei, Y.-S. Sun, H.-W. Hung, M.-C. Liu, Y.-P. Lin, J.-Y. Li, and C.-W. Wang, “Highly efficiency yellow organic light-emitting diodes with a solution-processed molecular host-based emissive layer,” J. Mater. Chem. C 1(8), 1680–1686 (2013). [CrossRef]
16. J. You, L. Dou, K. Yoshimura, T. Kato, K. Ohya, T. Moriarty, K. Emery, C.-C. Chen, J. Gao, G. Li, and Y. Yang, “A polymer tandem solar cell with 10.6% power conversion efficiency,” Nat. Commun. 4, 1446 (2013). [CrossRef] [PubMed]
17. S. Braun, W. R. Salaneck, and M. Fahlman, “Energy-level alignment at organic/metal and organic/organic interfaces,” Adv. Mater. 21(14–15), 1450–1472 (2009). [CrossRef]
18. H. Ma, H.-L. Yip, F. Huang, and A. K.-Y. Jen, “Interface engineering for organic electronics,” Adv. Funct. Mater. 20(9), 1371–1388 (2010). [CrossRef]
19. Z. He, C. Zhong, S. Su, M. Xu, H. Wu, and Y. Cao, “Enhanced power-conversion efficiency in polymer solar cells using an inverted device structure,” Nat. Photonics 6(9), 593–595 (2012). [CrossRef]
20. M. P. de Jong, L. J. van IJzendoorn, and M. J. A. de Voigt, “Stability of the interface between indium-tin-oxide and poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate) in polymer light-emitting diodes,” Appl. Phys. Lett. 77(14), 2255–2257 (2000). [CrossRef]
21. K. W. Wong, H. L. Yip, Y. Luo, K. Y. Wong, W. M. Lau, K. H. Low, H. F. Chow, Z. Q. Gao, W. L. Yeung, and C. C. Chang, “Blocking reactions between indium-tin-oxide and poly (3,4-ethylene dioxythiophene):poly(styrene sulphonate) with a self-assembly monolayer,” Appl. Phys. Lett. 80(15), 2788–2790 (2002). [CrossRef]
22. J.-M. Yun, J.-S. Yeo, J. Kim, H.-G. Jeong, D.-Y. Kim, Y.-J. Noh, S.-S. Kim, B.-C. Ku, and S.-I. Na, “Solution-processable reduced graphene oxide as a novel alternative to PEDOT:PSS hole transport layers for highly efficient and stable polymer solar cells,” Adv. Mater. 23(42), 4923–4928 (2011). [CrossRef] [PubMed]
23. Y. Park, K. S. Choi, and S. Y. Kim, “Graphene oxide/PEDOT:PSS and reduced graphene oxide/PEDOT:PSS hole extraction layer in organic photovoltaic cells,” Phys. Status Solidi A 209(7), 1363–1368 (2012). [CrossRef]
24. J. Liu, Y. Xue, Y. Gao, D. Yu, M. Durstock, and L. Dai, “Hole and electron extraction layers based on graphene oxide derivatives for high-performance bulk heterojunction solar cells,” Adv. Mater. 24(17), 2228–2233 (2012). [CrossRef] [PubMed]
25. H. G. Jeon, Y. H. Huh, S. H. Yun, K. W. Kim, S. S. Lee, J. Lim, K.-S. An, and B. Park, “Improved homogeneity and surface coverage of graphene oxide layers fabricated by horizontal-dip-coating for solution-processable organic semiconducting devices,” J. Mater. Chem. C 2(14), 2622–2634 (2014). [CrossRef]
26. K. A. Mkhoyan, A. W. Contryman, J. Silcox, D. A. Stewart, G. Eda, C. Mattevi, S. Miller, and M. Chhowalla, “Atomic and electronic structure of graphene-oxide,” Nano Lett. 9(3), 1058–1063 (2009). [CrossRef] [PubMed]
28. G. Li, C. W. Chu, V. Shrotriya, J. Huang, and Y. Yang, “Efficient inverted polymer solar cells,” Appl. Phys. Lett. 88(25), 253503 (2006). [CrossRef]
29. J. R. Manders, S.-W. Tsang, M. J. Hartel, T.-H. Lai, S. Chen, C. M. Amb, J. R. Reynolds, and F. So, “Solution-processed nickel oxide hole transporting layer in high efficiency polymer photovoltaic cells,” Adv. Funct. Mater. 23(23), 2993–3001 (2013). [CrossRef]
30. C. Tao, S. Ruan, G. Xie, X. Kong, L. Shen, F. Meng, C. Liu, X. Zhang, W. Dong, and W. Chen, “Role of tungsten oxide in inverted polymer solar cells,” Appl. Phys. Lett. 94(4), 043311 (2009). [CrossRef]
31. S. Shao, J. Liu, J. Bergqvist, S. Shi, C. Veit, U. Würfel, Z. Xie, and F. Zhang, “In situ formation of MoO3 in PEDOT:PSS Matrix: A facile way to produce a smooth and less hygroscopic hole transport layer for highly stable polymer bulk heterojuction solar cells,” Adv. Energy Mater. 3(3), 349–355 (2013). [CrossRef]
32. K. Zilberberg, J. Meyer, and T. Riedl, “Solution processed metal-oxides for organic electronic devices,” J. Mater. Chem. C 1(32), 4796–4815 (2013).
33. W. S. Hummers Jr and R. E. Offeman, “Preparation of graphene oxide,” J. Am. Chem. Soc. 80(6), 1339 (1958). [CrossRef]
34. J. H. Park, S. S. Oh, S. W. Kim, E. H. Choi, B. H. Hong, Y. H. Seo, G. S. Cho, B. Park, J. Lim, S. C. Yoon, and C. Lee, “Double interfacial layers for highly efficient organic light-emitting devices,” Appl. Phys. Lett. 90(15), 153508 (2007). [CrossRef]
36. X. H. Yang, F. Jaiser, B. Stiller, D. Neher, F. Galbrecht, and U. Scherf, “Efficient polymer electrophosphorescent devices with interfacial layers,” Adv. Funct. Mater. 16(16), 2156–2162 (2006). [CrossRef]
37. C. Adachi, R. Kwong, and S. R. Forrest, “Efficient electrophosphorescence using a doped ambipolar conductive molecular organic thin film,” Org. Electron. 2(1), 37–43 (2001). [CrossRef]
38. S. H. Park, A. Roy, S. Beaupré, S. Cho, N. Coates, J. S. Moon, D. Moses, M. Leclerc, K. Lee, and A. J. Heeger, “Bulk heterojunction solar cells with internal quantum efficiency approaching 100%,” Nat. Photonics 3(5), 297–302 (2009). [CrossRef]
39. L. D. Landau and V. G. Levich, “Dragging of a liquid by a moving plate,” Acta Physicochim. URSS 17(1–2), 42–54 (1942).
41. A. C. Ferrari and J. Robertson, “Interpretation of Raman spectra of disordered and amorphous carbon,” Phys. Rev. B 61(20), 14095–14107 (2000). [CrossRef]
42. V. Shrotriya and Y. Yang, “Capacitance-voltage characterization of polymer light-emitting diodes,” J. Appl. Phys. 97(5), 054504 (2005). [CrossRef]
43. S. Nowy, W. Ren, J. Wagner, J. A. Weber, and W. Brütting, “Impedance spectroscopy of organic hetero-layer OLEDs as a probe for charge carrier injection and device degradation,” Proc. SPIE 7415, 74150G (2009). [CrossRef]
44. S. Nowy, W. Ren, A. Elschner, W. Lövenich, and W. Brütting, “Impedance spectroscopy as a probe for the degradation of organic light-emitting diodes,” J. Appl. Phys. 107(5), 054501 (2010). [CrossRef]
45. S. Chen, X. Jiang, and F. So, “Hole injection polymer effect on degradation of organic light-emitting diodes,” Org. Electron. 14(10), 2518–2522 (2013). [CrossRef]