Ran Ji1,2, Ding Zheng3, Chang Zhou4, Jiang Cheng5, Junsheng Yu6, Lu Li7. 1. State Key Laboratory of Electronic Thin Films and Integrated Devices, School of Optoelectronic Information, University of Electronic Science and Technology of China, Chengdu 610054, China. jiran1231@gmail.com. 2. Co-Innovation Center for Micro/Nano Optoelectronic Materials and Devices, Research Institute for New Materials and Technology, Chongqing University of Arts and Sciences, Chongqing 402160, China. jiran1231@gmail.com. 3. State Key Laboratory of Electronic Thin Films and Integrated Devices, School of Optoelectronic Information, University of Electronic Science and Technology of China, Chengdu 610054, China. zd901023@gmail.com. 4. State Key Laboratory of Electronic Thin Films and Integrated Devices, School of Optoelectronic Information, University of Electronic Science and Technology of China, Chengdu 610054, China. 2013000101001@std.uestc.edu.cn. 5. Co-Innovation Center for Micro/Nano Optoelectronic Materials and Devices, Research Institute for New Materials and Technology, Chongqing University of Arts and Sciences, Chongqing 402160, China. caiwu0301@163.com. 6. State Key Laboratory of Electronic Thin Films and Integrated Devices, School of Optoelectronic Information, University of Electronic Science and Technology of China, Chengdu 610054, China. jsyu@uestc.edu.cn. 7. Co-Innovation Center for Micro/Nano Optoelectronic Materials and Devices, Research Institute for New Materials and Technology, Chongqing University of Arts and Sciences, Chongqing 402160, China. lli@cqwu.edu.cn.
Abstract
Tungsten oxide (WO₃) is prepared by a low-temperature ultrasonic spray pyrolysis method in air atmosphere, and it is used as an anode buffer layer (ABL) for organic solar cells (OSCs). The properties of the WO₃ transition metal oxide material as well as the mechanism of ultrasonic spray pyrolysis processes are investigated. The results show that the ultrasonic spray pyrolysized WO₃ ABL exhibits low roughness, matched energy level, and high conductivity, which results in high charge transport efficiency and suppressive recombination in OSCs. As a result, compared to the OSCs based on vacuum thermal evaporated WO₃, a higher power conversion efficiency of 3.63% is reached with low-temperature ultrasonic spray pyrolysized WO₃ ABL. Furthermore, the mostly spray-coated OSCs with large area was fabricated, which has a power conversion efficiency of ~1%. This work significantly enhances our understanding of the preparation and application of low temperature-processed WO₃, and highlights the potential of large area, all spray coated OSCs for sustainable commercial fabrication.
Tungsten oxide (WO₃) is prepared by a low-temperature ultrasonic spray pyrolysis method in air atmosphere, and it is used as an anode buffer layer (ABL) for organic solar cells (OSCs). The properties of the WO₃ transition metal oxide material as well as the mechanism of ultrasonic spray pyrolysis processes are investigated. The results show that the ultrasonic spray pyrolysized WO₃ ABL exhibits low roughness, matched energy level, and high conductivity, which results in high charge transport efficiency and suppressive recombination in OSCs. As a result, compared to the OSCs based on vacuum thermal evaporated WO₃, a higher power conversion efficiency of 3.63% is reached with low-temperature ultrasonic spray pyrolysized WO₃ ABL. Furthermore, the mostly spray-coated OSCs with large area was fabricated, which has a power conversion efficiency of ~1%. This work significantly enhances our understanding of the preparation and application of low temperature-processed WO₃, and highlights the potential of large area, all spray coated OSCs for sustainable commercial fabrication.
Entities:
Keywords:
anode buffer layer; large area; organic solar cells; tungsten oxide; ultrasonic spray pyrolysis method
The steady enhancement in efficiency of organic solar cells (OSCs) has made this class of photovoltaic device a highly promising technology for photo-electric conversion [1,2,3]. Recently, the state-of-the-art single junction OSCs with power conversion efficiency (PCE) approaching 12.4% hasbeen achieved [4]. To obtain OSCs with high efficiency, several approaches have been developed such as material development [5,6,7], morphology control [8,9,10,11], and device engineering [4,12,13,14]. It is well known that the performance of OSCs can be significantly improved by inserting a suitable interfacial buffer layer between active layer and electrode. The interfacial buffer layer is efficient for enhancing charge transfer efficiency, preventing photon-generated carriers from undesired recombination and modifying the electrode [15].In the field of OSCs, an interfacial buffer layer includes cathode buffer layer and anode buffer layer (ABL), and PEDOT:PSS is mainly adopted as the ABL. Nevertheless, PEDOT:PSS is both hygroscopic and acidic, which has an associated reduction in device stability [16]. To circumvent this problem, transition metal oxides such as molybdenum oxide (MoO3) [16,17,18], vanadium oxide (V2O5) [19,20,21], nickel oxide (NiO) [22,23] and tungsten trioxide (WO3) [24,25,26,27,28] have been chosen as promising candidates as ABLs to replace PEDOT:PSS, due to their superb charge transfer capability and stability. However, most of transition metal oxides buffer layers applied in OSCs are fabricated by using vacuum thermal evaporated technique [16,17,18,28] or sol-gel methods with high temperature annealing [29,30,31], leading to increased energy consumption and low production speeds. Besides, it is much more difficult to deposit inorganic buffer materials by solution method especially on active layer, since in the high substrate temperature needed for annealing, even some of the raw materials could not be solubilized.These are limitations for transition metal oxides for further industrial application of OSCs. Therefore, it is important to realize a path towards the commercialization of solution-processed OSCs. The development of fabrication methods should be compatible with high-volume roll-to-roll (R2R) processing techniques [32]. Among them, the ultrasonic spray pyrolysis (USP) method, which is readily scalable and open-air to produce high-quality polycrystalline metal oxide thin films, fully satisfies these demands. In addition, the USP method does not require either extended thermal anneals or additional synthetic steps [33], and several metal oxide thin films as the highly efficient transport layer have been deposited, such as zinc oxide (ZnO) [34], molybdenum trioxide (MoO3) [35] and tungsten trioxide (WO3) [36,37]. However, these WO3 prepared methods require high annealing temperatures (over 250 °C).In this work, we have used a low-decomposition-temperature ammoniumtungstate (AT) ((NH4)10W12O41) solution as a precursor to obtain a transition metal oxide material of WO3 thin films, which was deposited with water-based solution on top of the active layer, directly by USP method, and a mostly spray-coated OSC (without electrode) was fabricated. To illuminate the improvement of charge transport ability of spray-coated WO3 (S-WO3) ABL, the properties of composition, morphology, conductivity, and crystallinity of S-WO3 are investigated. Equivalent circuit model, impedance analysis and mechanism of USP coated films are also discussed. Moreover, the OSCs based on different precursor solution concentrations are optimized. To illustrate how the USP method is useful, and how S-WO3 is a promising candidate as ABL, we choose the OSC based on evaporated WO3 (E-WO3) as a control group, because the evaporation method is a conventional technique to fabricate WO3 films. The performance of S-WO3 ABLs based devices shows an improvement of power conversion efficiency from 3.2% to 3.6%. The result suggests that this solution processed WO3 is a promising anode interfacial layer for the fabrication of high efficiency OSCs. Finally, the fabrication of a large area OSC device based on S-WO3 ABL is performed to test the potential of large area and all spray-coated OSCs.
2. Results and Discussion
2.1.Characterization and Analysis of WO3 Films
Figure 1a shows the configuration of the photovoltaic device in this work. Figure 1b presents the device fabrication apparatus, which comprises an ultrasonic transducer, an atomization chamber, an automatic X-Y table, a heating block, and pipe fittings. Mist droplets are generated in an atomization chamber through an ultrasonic atomization of precursor solution. Then, droplets transfer through the pipe fittings and deposit on the heating block. The sprinkler is passed over the hot plate driven by the drive X-Y table and performed a snake-like curve relative motive with the substrate.
Figure 1
(a) Device architecture of inverted organic solar cell (OSC) and (b) Ultrasonic spray pyrolysis system.
To assay the elementary composition of WO3 films, the surface characteristics of S-WO3 and E-WO3 films on siliconwafer are characterized by X-ray photoelectron spectroscopy (XPS). The full scan spectra are shown in Figure 2a,b. O (1s) and W (4f) of two samples are characterized and shown in Figure 2c–f, respectively. The two full scan spectra show that the chemical states of S-WO3 and E-WO3 are almost identical except for an additional N 1s peak center at 400.0 ± 0.2 eV in the S-WO3 spectrum. It also indicates that AT does not decomposed completely under 80 °C. The thermo gravimetric analysis (TG) and differential scanning calorimeter (DSC) data as shown in Figure 3d exhibit the same result. In Figure 3d, for the dried precursor, the weight losses at 80 °C were ~4.5%, indicating that the dried precursor is the mixture of WO3 and AT.
Figure 2
X-ray photoelectron spectroscopy (XPS) spectra of different process tungsten oxide (WO3) films. (a) Full scan; (c) W (4f) core levels and (e) O (1s) core levels of spray-coated WO3 (S-WO3) film, respectively; (b) Full scan; (d) W (4f) core levels and (f) O (1s) core levels of evaporated WO3) E-WO3 film, respectively.
Figure 3
(a) Thermo gravimetric (TG) and differential scanning calorimeter (DSC) profiles of the AT precursor; (b) ultraviolet photo-electron spectroscopy (UPS) results of S-WO3 film; (c) energy level of the component materials used in the OSCs; (d) XRD pattern of S-WO3 and E-WO3, respectively.
Two main XPS resolved peaks are ascribed to the typical doublet of W6+ with binding energy of W (4f7/2) centered at 36.0 ± 0.2 eV. The spin orbit splitting of the doublet is 2.12 eV and the peak ratio of W (4f7/2) to W (4f5/2) is 4:3. A third broad peak of W (5p3/2) locates at ~42 eV. The high-resolution XPS spectrum of E-WO3 reveals only W6+ oxidation state. W atoms with an oxidation state are ascribed to WO3 [38]. In the S-WO3 film, a second doublet at 34.6 ± 0.1 eV and 36.7 ± 0.1 eV is conventionally used to fitting with the W6+ peak due to the asymmetry of peak at the lower binding energies. It indicates that WO3 readily becomes oxygen deficient to form WO3−, with variable oxygen composition parameter x. This oxygen deficiency greatly influences the bulk of the electronic transport properties by introducing donor electronic states [39,40]. Additionally, the unsaturated valence of metallic W would decrease the hole injection barrier of interface when the S-WO3 film is contact with P3HT:PCBM [41]. Therefore, the S-WO3 might have a better efficiency of hole transport than E-WO3.On the other hand, the O (1s) XPS spectra of two WO3 samples exhibit asymmetric line shapes, and the peaks are fitted with two components. One main peak at 531.0 ± 0.1 eV corresponds to W-O bond, and the other at 532.5 ± 0.1 eV corresponds to -OH groups due to atmospheric contamination or the crystal water [42,43,44]. Obviously, there are less oxide impurities in S-WO3.Then, ultraviolet photo-electron spectroscopy (UPS) is used to probe the electronic properties of S-WO3 which is shown in Figure 3a. We determined the work function of S-WO3 in the ITO/ZnO/P3HT/S-WO3. For the UPS measurement, a 10 nm thick S-WO3 film was formed on the ITO/ZnO/P3HT substrate. The UPS measurement was performed with a He I (21.2 eV) discharge lamp (AXIS–NOVA System, Kratos, Manchester, UK). The sample was kept inside a high-vacuum chamber. The work function (WF) is calculated from the UPS data using the following Equation (1).
where hν (21.2 eV) is the incident photon energy. E (16.1 eV) is the high binding energy cutoff, and E (0.2 eV) is the valence band. Therefore, the WF of S-WO3 is 5.3 eV. The value is much lower than the publicly recognized WF of WO3 (6.7 eV), because of the formation of an interface dipole, induced by an electron transfer from organic film [45]. We think the lower WF value can describe the characteristics of S-WO3 better, which is much more suitable for S-WO3 in an OSC device. Moreover, as shown in Figure 3b, the matched WF value of S-WO3 allows for the formation of Ohmic contacts with a donor [46], thus resulting in the increase of the built-in field. This is beneficial for enhancing charge extraction efficiency and reducing re-combination losses [47,48].To further disclose the physical property of WO3 films, the X-ray diffraction (XRD) pattern of both 20 nm-thick S-WO3 and 20 nm-thick E-WO3 films are characterized and shown in Figure 3c. It indicates that S-WO3 has a good crystallinity. However, no crystalline peaks are observed in E-WO3 film, so it can be regarded as an amorphous film.
2.2. Detailed Analysis of OSC Performances
To further investigate the influence of E-WO3 and S-WO3 with different precursor concentrations on the device performance, the OSCs were fabricated. We optimized the thickness of E-WO3 film, and OSCs based on 20 nm E-WO3which have the best performance. Therefore, we chose 20 nm as the optimal thickness of E-WO3. Figure 4a shows the current density versus voltage (J–V) characteristics of OSCs based on E-WO3 and S-WO3 films made from 100mg/L AT precursor. The J–V characteristics of OSCs based on S-WO3 made from other precursor concentrations are shown in Figure S1, and the detailed parameters are summarized in Table 1. The device using 100 mg/L AT precursor shows the highest PCE of 3.63%, which is also better than that using 20 nm E-WO3. The optimized device has an 11% enhancement in PCE with a simultaneous improvement in V (0.57 V to 0.63 V) and J (9.68 mA·cm−2 to 10.45 mA·cm−2). When the concentration of AT precursor solution increases from 25 to 100 mg/L, the PCE of device increases from 2.03% to 3.63% with enhancement in V (0.55 V to 0.63 V), J (9.58 mA·cm−2 to 10.45 mA·cm−2), FF (38.73% to 52.85%) and R (15.65 Ω·cm2 to 1.18 Ω·cm2). Series resistances are derived from the slope of the J–V characteristic curve and the low R of 1.18 Ω·cm2 is much comparable to the device with E-WO3. Once the concentration of AT precursor increases from 100 to 300 mg/L, the PCE of devices get a decrease from 3.63% to 2.86%.
Figure 4
(a) Current density versus voltage (J–V) characteristics of OSCs with S-WO3 and E-WO3 films; (b) external quantum efficiency (EQE) characteristics of OSCs with S-WO3 and E-WO3 films; Atomic force microscopy (AFM) images of (c) S-WO3; (d) E-WO3.
Table 1
Comparison of device characteristics of OSCs based on E-WO3 and S-WO3 films with different precursor concentrations.
Devices
VOC (V)
JSC (mA·cm−2)
FF (%)
PCE (%)
RS (Ω cm2)
20 nm E-WO3
0.57 ± 0.01
9.68 ± 0.09
58.73 ± 1.02
3.27 ± 0.11
1.06 ± 0.15
S-WO3 with concentration of AT
25 mg/L
0.55 ± 0.03
9.58 ± 0.21
38.74 ± 2.62
2.03 ± 0.22
15.65 ± 1.76
100 mg/L
0.63 ± 0.02
10.45 ± 0.17
55.48 ± 1.27
3.63 ± 0.13
1.18 ± 0.31
200 mg/L
0.61 ± 0.02
10.12 ± 0.19
52.85 ± 1.36
3.27 ± 0.16
2.42 ± 0.73
300 mg/L
0.61 ± 0.02
9.52 ± 0.14
49.36 ± 0.95
2.86 ± 0.12
7.84 ± 0.58
To illustrate the impact on device performance based on different ABLs, the morphology of E-WO3 and S-WO3 films made from different AT precursor concentration is characterized by atomic force microscopy (AFM). The images of E-WO3 and S-WO3 films with 100 mg/L AT precursor are shown in Figure 4. The images of substrate, active layer and S-WO3 films with other precursor concentrations are shown in Figure S2. Due to the surface of spray-coated active layer beingextremely rough, all films are directly fabricated on ITO to characterize the films. We find that S-WO3 and E-WO3 films have similar morphology. For the film of S-WO3 with the best OSC performance, a smooth surface was observed with a root mean square (RMS) of surface roughness of 3.49 nm, and even smoother than the evaporated one with a RMS of surface roughness of 3.61 nm. With the increasing of the concentration of AT precursor, the films get thicker and rougher, and the grain size of WO3 becomes much larger. Compared to the S-WO3 films, E-WO3 film cannot show a uniform and obvious grain. We speculate that it relates to the different crystallinity of S-WO3 and E-WO3, which confirms the result of XRD. The thicknesses of S-WO3 films obtained from 300 mg/L, 200 mg/L and 100 mg/L AT precursors are ~30 nm 20 nm and 15 nm, respectively. However, the S-WO3 films from the 25 mg/L precursor cannot cover the entire substrate. The thickness from 25 mg/L AT precursor is ~5 nm. From the perspective of device performance, it is necessary for S-WO3 ABL to cover the active layer fully and uniformly for reducing the leakage current. Introducing a WO3 layer will generally contribute to the series resistance of the device and the higher thickness of WO3 will block charge transport and result in a lower current [49]. Compared to the device using E-WO3 ABL, the devices with S-WO3 ABL has a much higher V, which is probably in consideration of the high conductivity and matching energy level of S-WO3 [50].Figure 4d shows the results of external quantum efficiency (EQE) measurement for OSCs based on S-WO3 and E-WO3. The device spectrum based on S-WO3 film fabricated by 100 mg/L AT precursor shows a higher EQE than others, and the peak is 67.6% at a wavelength of 510 nm. The EQE curves of the S-WO3 devices with other concentration present similar shape in the entire range of wavelength between 350 and 700 nm, as shown in Figure S3. The relatively high EQE originates from the better Ohmiccontact acquired at the interface. We postulate that the increase of hole transport efficiency is due to the high hole mobility of S-WO3 ABL. The enhancement in EQE is also in good agreement with the improved J. This result indicates that the S-WO3 can bring an improved hole transport efficiency.To verify our postulation, the device with a configuration of ITO/ABLs/Ag wasfabricated to demonstratethe conductivity of ABLs. The I-V curves of device are presented in Figure S4, and details are also provided in supplemental materials. The conductivities of E-WO3 and S-WO3 are 3.05 × 10−4 S∙m−1 and 5.17 × 10−4 S∙m−1, respectively. The conductivity of S-WO3 is much higher than that of E-WO3. These results are in good agreement with the aforementioned phenomena. This high conductivity S-WO3 ABL with suitable energy level can effectively block the electron and provide a well Ohmic contact between anode and active layer to enhance the hole transport efficiency of the whole device.
2.3. Equivalent Circuit Model and Impedance Analysis
To examine the electrical contact of the interfaces in the obtained devices, a circuit model is defined according to the sandwiched device structure of OSCs as shown in Figure 5a. Figure 5b shows the nyquist plots of impedance measurement of devices based on S-WO3 and E-WO3 ABLs for frequencies ranged from 40 Hz to 5 MHz. The parallel circuit of R1 and C1 corresponds to the donor and accepter interface. R2 and C2 represent two electrical contacts of the interfaces between the active layer and electrodes. R3 represents the resistance of electrodes. Parameters employed for the fitting of the impedance spectra by using an equivalent circuit model are shown in Table 2. Compared to the device based on E-WO3, the R2 of the interface junction decreases from 1547 to 850 Ω when using the S-WO3 ABL. Moreover, the C2 of the interface junction increases from 9.60 × 10−9 F to 1.16 × 10−9 F. This indicates that the interface between the S-WO3 and the active layer shows a good Ohmic contact correlated to the efficient charge transport [51].
Figure 5
(a) Equivalent circuit model of the devices. (R1 and CPE1, R2 and C2, R3 represent equivalent of donor and accepter interface, interface between active layer and electrodes, resistance of electrodes, respectively; (b) Cole–Cole plots of the devices based on E-WO3 and S-WO3 films with different precursor concentration.
Table 2
Parameters employed for the fitting of the impedance spectra by using an equivalent circuit model.
Devices
R1 (Ω)
C1 (F)
R2 (Ω)
C2 (F)
R3 (Ω)
20 nm E-WO3
3.32 × 105
1.37 × 10−9
1547.00
9.60 × 10−10
220.20
S-WO3 with concentration of AT
25 mg/L
1.78 × 106
1.25 × 10−9
2049.00
6.983 × 10−10
245.90
100 mg/L
2.09 × 105
2.50 × 10−9
850.80
1.16 × 10−9
107.50
200 mg/L
2.33 × 105
3.49 × 10−9
1024.00
1.01 × 10−9
73.81
300 mg/L
4.08 × 105
1.55 × 10−9
1682.00
8.76 × 10−10
160.30
2.4. Mechanism of USP Coated Films
To realize the uniform films, especially the large area uniform ones, for large area OSCs, the mechanism of USP process should be discussed in more detail. As is well known, the atomization of precursor solution droplets should satisfy with: (1) Small droplets size; (2) Well-proportioned distribution. For USP method, the amount and diameters of droplets are related to the vapor pressure, surface tension and viscosity of precursor solution. The diameter of droplets is given by Equation (2) [52]:
where D is the diameter of droplet, and σ is the surface tension of the solvents, ρ is the density of solvent, and f is the frequency of ultrasonic wave. For a given solvent, the size of droplets is determined by the frequency. When the frequency of ultrasonic wave reached billions of hertz, the droplets with a few micron diameters would be obtained. During the spray process, droplets keep moving. The weight of droplets decreases continuously along with solvent drying, and the velocity of droplets is changed simultaneously. For a single droplet, the rate of mass change induced by evaporation is given by Equation (3) [53]:
where γ and γ are the content of solvent in ambient gas and the saturated concentration of solvent on droplet surface, respectively, and sh is the Sherwood constant. D is the diffusion coefficient of solvent, and R is the radius of droplet. Ultimately, the evaporation degree of droplets can be deduced by the mass change.Viguie et al. considered that the film grown pattern of droplets pyrolytic deposition can be divided into four issues [54], which is shown in Figure 6. In the case of Figure 6a, the temperature of substrate and air atmosphere is relatively low, so solvent does not evaporated completely in air. When the droplets reach to the substrate, the solvent starts to evaporate and the solute decompose. In Figure 6b, the temperature of substrate is relatively high. As a result, before droplets reach substrate, solvent evaporates completely, and solute reaches to substrate and pyrolysis. In the condition of Figure 6c, with the increase of temperature, before droplets reach the substrate, solvent evaporates completely, and then solute melts and vaporizes. Lastly, the gaseous solute is deposited on substrate to form a film. This process is similar to chemical vapor deposition. In Figure 6d, the temperature of substrate is extremely high, the above mentioned processes of a, b, c are all completed. The end products will fall onto the substrate, leading to the forming film with poor adhesion [54].
Figure 6
Dynamics of USP with different substrate temperature.
In this work, the fabrication of S-WO3 was processed at relatively low temperature, so the processing mechanism belongs to the situation of Figure 6a: The droplets of AT precursor are deposited on the substrate, and the excess solvent is evaporated rapidly. Subsequently, partial AT decomposes into WO3, and the mixture grows into film. The low substrate temperature provides only a small driving force for nucleation, so the film has a low nucleation and growth rate. This means that the grain can grow much larger, resulting in smooth and uniform film [34], which is in accordance with the result of AFM image. Hence, the low temperature processing condition can yield thin uniform films with large area for potential device application.
2.5. Large Area OSC
Based on the above discussion of USP mechanism and the analysis of OSCs performance, 100 mg/L solution of precursor is chosen as the optimal concentration to process large area uniform WO3 films. By using the obtained WO3 films, a large area OSC of 5 × 5 cm2 with a same structure as the above devices were successfully prepared. Figure 7a shows the image of practical OSC, and Figure 7b presents the J–V characteristics of the device. The V and the Jsc are 560 mV and 5.94 mA/cm2, respectively, indicating that this low temperature deposition method is applicable for the fabrication of large area OSCs. The PCE of the OSC with a large area of 25 cm2 is ~1%, which is much lower than the former OSC with a small area of 0.02 cm2. Among three factors to determine the PCE of OSC, the low Jsc of large area device is due to the large square resistance caused by the area increase of ITO electrode. The decrease of V and FF is attributed to the high leakage current with the increasing defect along with the area increase of functional films of two buffer layers and the active layer. Therefore, this low temperature USP method with the assistance of ABL can be applied for large area OSCs, but the core issue of large area devices about increasing PCE still needs to be studied.
Figure 7
(a) Image of a 25 cm2 OSC device with as-grown S-WO3 film and (b) J–V characteristic of the OSC.
3. Materials and Methods
3.1. Fabrication Section
The configuration of photovoltaic device is ITO/ZnO (40 nm)/P3HT:PC61BM (300 nm)/ABL/Ag (100 nm). Patterned ITO-coated glass substrates with a sheet resistance of 10 Ù∙sq−1 were consecutively cleaned in ultrasonic bath containing detergent, acetone, deionized water and ethanol for 10 min each step, then dried by nitrogen blow [55]. Prior to the deposition of functional layers, the substrate was treated with UV light for 10 min. A 40 nm ZnO layer was spray-cast on ITO film by ultrasonic spray pyrolysis at 150 °C as depicted in our previous work [34]. Then, a 300 nm active layer of P3HT:PC61BM was casted with a supersonic nozzle (Z95S, Siansonic, Beijing, China) from a solution with P3HT (99.9%, Solarmer, Beijing, China) and (6,6)-phenyl-PCBM (99.9%, Solarmer, Beijing, China) at a weight ratio of 1:0.9 wt % in 1,2-dichlorobenzene (DCB) at a concentration of 5 mg/mL, separately. The spray rate of P3HT:PC61BM solution was 0.075 mL·min−1 and the N2 carrier gas flow rate was held at a rate of 18 L·min−1. An ABL of S-WO3 was prepared by USP on P3HT:PC61BM blend film at 80 °C using N2 as carrier gas with a flow rate of 6 L·min−1, and the solution atomization rate was ~3 mL·min−1. The instrument used in USP is mentioned previously in Section 2.1. The ultrasonic transducer (JR-24, Siansonic) used aФ 20 mm piezoelectric vibrator at an ultrasonic frequency of 1700 KHz. AФ 20 mm nozzle was mounted 10 mm above the heating block. Precursor solutions were obtained by dissolving ammonium metatungstate ((NH4)10W12O41 99.95%, Aladdin, Shanghai, China) powders into deionized water with a desired concentration. An E-WO3 ABL was deposited onto the substrate at a rate of 1–2 Å/S at a pressure of 5 × 10−4 Pa by multifunctional high vacuum film forming equipment (OLED-V Shenyang Vacuum Technology Institute). Subsequently, Ag anodes were finally deposited at a rate of about 10 Å/S under a pressure of 5 × 10−4 Pa without breaking the vacuum.
3.2. Measurement Method
The composition and electron structure of the ABLs were characterized by X-ray photoelectron spectroscopy (XPS, Thermo ESCALAB, Shanghai, China). The UPS measurements were performed with a He I (21.2 eV) discharge lamp (AXIS–NOVA System, Kratos, Manchester, UK). All samples were kept inside a high-vacuum chamber. TGA, DSC were taken using simultaneous thermal analyzer (STA 449 F3, Netzsch, Shanghai, China) with crucible (DSC/TG pan Al2O3), and the flow rate is 60 mL/min in nitrogen condition. The simple mass is ~5.7 mg, and the heating rate is 5 °C/min. The crystal structure was characterized by X-ray diffraction (XRD, X’Pert PRO, PANalytical, Cu Kα radiation λ = 0.154056 nm, 40 kV and 40 mA) in grazing incidence mode. The Surface morphology of the ABLs, ITO and active layer was characterized by atomic force microscope (AFM, AFM 5500, Agilent, Tapping Mode, Chengdu, China). AFM images and RMS roughness are obtained by Gwyddion. Current density–voltage (J–V) curves were measured with Keithley 2400 under a xenon lamp solar simulator (7IS0503A, SOFN, Beijing, China) with an illumination power of 100 mW/cm2 [56]. Conductivity measurement and calculation procedure are shown in supplemental materials in Figure S4. A precision impedance analyzer (4294A, Agilent, Chengdu, China) was employed for impedance spectroscopy measurement. The range of measured frequency is from 40 Hz to 1 MHz, and 50 mV of modulation voltage without DC bias was used to extract the DC bias-dependent AC signal. All the measurements were carried out at ambient circumstance without encapsulation.
4. Conclusions
In summary, the low-temperature S-WO3 film shows low roughness, matched energy level, and high conductivity, which is suitable for the fabrication of OSCs with high charge transport efficiency responsible for the obviously increased J and V, and a higher PCE accordingly. The study of OSC performance processed from different concentrations of AT precursor shows that, when the concentration of AT precursor is 100 mg/L, the OSC has the highest PCE of 3.63%. Then, based on the characterization of S-WO3 film and mechanism analysis of USP method, the almost all-sprayed large area OSC with a PCE of ~1% was realized, which meets the essential prerequisite of roll-to-roll manufacturing and is compatible with large areas of a variety of thin film optoelectronics.
Authors: Longzhen You; Bin Liu; Tao Liu; Bingbing Fan; Yunhao Cai; Lin Guo; Yanming Sun Journal: ACS Appl Mater Interfaces Date: 2017-04-03 Impact factor: 9.229
Authors: Markus C Scharber; Markus Koppe; Jia Gao; Fabrizio Cordella; Maria A Loi; Patrick Denk; Mauro Morana; Hans-Joachim Egelhaaf; Karen Forberich; Gilles Dennler; Russ Gaudiana; Dave Waller; Zhengguo Zhu; Xiaobo Shi; Christoph J Brabec Journal: Adv Mater Date: 2010-01-19 Impact factor: 30.849
Authors: Xiaodan Zhu; Toshiyuki Kawaharamura; Adam Z Stieg; Chandan Biswas; Lu Li; Zhu Ma; Mark A Zurbuchen; Qibing Pei; Kang L Wang Journal: Nano Lett Date: 2015-07-09 Impact factor: 11.189
Authors: Ja-Ryong Koo; Seok Jae Lee; Ho Won Lee; Dong Hyung Lee; Hyung Jin Yang; Woo Young Kim; Young Kwan Kim Journal: Opt Express Date: 2013-05-06 Impact factor: 3.894
Authors: Jens Meyer; Sami Hamwi; Michael Kröger; Wolfgang Kowalsky; Thomas Riedl; Antoine Kahn Journal: Adv Mater Date: 2012-08-01 Impact factor: 30.849
Authors: Evert Jonathan van den Ham; Giulia Maino; Gilles Bonneux; Wouter Marchal; Ken Elen; Sven Gielis; Felix Mattelaer; Christophe Detavernier; Peter H L Notten; Marlies K Van Bael; An Hardy Journal: Materials (Basel) Date: 2017-09-12 Impact factor: 3.623