In recent years, polymer solar cells (PSCs) gained a lot of attention because of their superior characteristics such as mechanical flexibility, low cost, low environmental load, and their potential to be produced in large scale.^[1–3] However, PSCs technology is not mature yet. PSCs, especially those with normal configuration, suffer from instability and low power conversion efficiency (PCE).^[4] These photovoltaic cells have common problems in the PEDOTS:PSS layer. This layer is sensitive to ambient conditions and its acidic nature increases the degradation rate of the layer. Oxidation of low work function top metal oxide with the degradation of the PEDOT:PSS layer under UV illumination is another problem of the normal configuration of PSCs.^[5–9] An inverted configuration is a common approach seeking to overcome the mentioned problems of PSCs.^[3,9] In this structure, a buffer layer is inserted between the active layers and electrodes. Different n-type metal oxides such as ZnO^[9–12] and TiO₂,^[13–15] alkali metal oxides like LiF^[16] and Cs₂CO₃,^[17,18] and in some cases ultra thin layers of Ca^[19,20] are used as the electron-selective and hole-blocking layer in PSCs.

Special qualities of TiO₂ cause it to be widely used in semiconductor industries such as OLEDs and organic photovoltaic cells. The TiO₂ thin film has high electron mobility, great optical transparency, good physical and chemical stability; and as a nontoxic material, it can be produced in large scale inexpensively. Also, TiO₂ has substantial oxygen and water protection because of photocatalysis combination and inherent oxygen deficiency.^[21–23] Atomic layer deposition,^[14,23] magnetron sputtering,^[13,24] doctor blade method,^[25] and spin coating^[26] are some of the techniques used to fabricate TiO₂ thin films. Some reports show that doping TiO₂ can improve the performance of solar cells. For example, a Cs-doped TiO₂ interlayer exhibited a power conversion efficiency of 4.2%, while pure thin films of TiO₂ showed PCE of ∼ 2.4%.^[9] Materials such as Al, Ga, and In were used as dopants in ZnO (ETLs) to enhance the electrical properties of the thin films and improve the PCE of cells.^[27–30] Materials such as vanadium and niobium could be used as a dopant to improve the electrical properties of TiO₂ as an electron transport layer. Vanadium has a resonant level within the conduction band of Ti and it can be used as a promising dopant to create a high density of free carriers.^[31–34] In this study, we investigate cells using 1% vanadium-doped TiO₂ thin films with different thicknesses and compare them with those using pure TiO₂ thin films.

All chemicals were used as received without further purification. Titanium (IV) butoxide (TB, 97%), oleic acid (OA, 90%), and oleylamine (OM, 70%) were purchased from Sigma-Aldrich. Absolute ethanol (anhydrous, ACS, 94%–96%) and vanadium (V) oxytriethoxide (VO(OPr)₃, 95%) were also purchased from Sigma-Aldrich. In this work, pure and vanadium-doped TiO₂ nanoparticles (NPs) were prepared by the solvothermal method.^[31] Typically, TB (2.27 ml, 6.7 mmol) and VO(OPr)₃ (0.018 g, 1% V/Ti mole ratio) were added to a mixture of 10.57 ml (33.3 mmol) OA, 10.97 ml (33.3 mmol) OM, 5.84 ml (100 mmol) absolute ethanol, and 0.67 ml (36.7 mmol) deionized (DI) water. The obtained mixture was stirred under argon atmosphere for 10 min before being transferred into a 50 mL Teflon-lined stainless steel autoclave. The system was then heated at 180 °C for 24 h. The obtained precipitates were separated by centrifugation (5000 rpm, 15 min) and washed several times with ethanol and then dried at room temperature. The solvothermal synthesized vanadium-doped TiO₂ NPs' products were dispersed in ethanol with different ratios. The procedure of this synthesis is summarized in Fig. 1.

Fig. 1.

	Figure Option ViewDownloadNew Window
	Fig. 1. (color online) Procedure of the synthesis of TiO2 NPs.

Fabrication of solar cells began with ultrasonic cleansing of ITO-coated glass in acetone and isopropanol for 30 min. Afterward for electron transport layer deposition, 10% and 40% stable solutions of pure and vanadium-doped TiO₂ NPs were dissolved in 2-methoxy ethanol to be deposited using doctor blade method (4 samples with the thicknesses of 30 and 60 nm, respectively).^[12] In the next step, 100 μL of the mentioned solutions was injected into the substrate. Finally, deposition started with a speed of 40 mm/s on the heated substrate (65 °C). In order to prepare the active layer of the cells, we dissolved P3HT and PCBM in chlorobenzene separately. A 60 °C hot plate was used for stirring the dissolved solutions overnight. At a weight ratio of 1:0.9 we blended the solutions. Afterward we used the 60 °C hot plate to stir the solutions again. The active layer was deposited on the electron transport layer in air via doctor blade method at a speed of 30 mm/s and the temperature of 60 °C. Then the thin layer of PEDOT: PSS was deposited on the active layer with a speed of 15 mm/s and the temperature of 60 °C as the hole transport layer (HTL). Inert atmosphere annealing on a hot plate at 140 °C was employed for 5 min. Thermal evaporation at ∼ 510⁻⁶ mbar and a shadow mask were used to deposit the Ag electrode. The structure of the cells is shown in Fig. 2.

Fig. 2.

	Figure Option ViewDownloadNew Window
	Fig. 2. (color online) An inverted configuration of polymer solar cells.

The structures of pure and vanadium-doped TiO₂ thin film NPs were analyzed using x-ray diffraction (XRD, Model Philips MPD PW 3040) over the 2θ range of 20°–80° at the scan rate of 0.02°/s. The surface morphology of pure and vanadium-doped TiO₂ thin film NPs was examined by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). The optical properties of the thin films were investigated using a UV–VIS–IR (Perkin-Elmer-950 lambda) spectrometer. The current density–voltage characteristics of the cells were measured under AM 1.5 illumination provided by Oriel Sol 1A solar simulator.

The crystalline structure of pure and vanadium-doped TiO₂ thin films NPs was investigated by XRD. Figure 3(a) shows the XRD pattern of pure TiO₂ thin film NPs. The diffraction pattern indicates the presence of eight peaks which are attributed to the anatase TiO₂ crystalline (JCPDS card No. 89-492). The eight peaks each are presented in the fields (101), (004), (200), (105), (204), (116), (220), and (215). The XRD pattern of vanadium-doped TiO₂ thin film NPs is depicted in Fig. 3(b). As can be seen, the diffraction peaks move gradually to lower angles due to the vanadium doping of the TiO₂ thin film. This slight shift towards a lower angle demonstrates doping of TiO₂ thin film NPs. Also, the eight peaks are attributed to the anatase phase of vanadium-doped TiO₂ thin films. There are no rutile peaks in both patterns of pure and vanadium-doped TiO₂ thin films. The peak assigned to the (101) plane is the most intense throughout the pattern, which indicates a preferential orientation of the vanadium-doped TiO₂ thin film NPs along the (101) direction. The degree of orientation for the (11) plane is calculated by the relative texture coefficient (χ_c)^[35]

Fig. 3.

	Figure Option ViewDownloadNew Window
	Fig. 3. XRD patterns of (a) pure anatase TiO2 and (b) vanadium-doped TiO2 NPs prepared by the solvothermal method.

The surface morphology of TiO₂ thin film NPs is depicted in Figs. 4–6. Figure 4 shows (a) the SEM image and (b) the size distribution histogram of TiO₂ thin films NPs. These images show that TiO₂ thin film NPs have a size about 10 nm. Figure 5 shows (a) the TEM image and (b) the size distribution histogram of TiO₂ thin film NPs. The particles are round in shape with good uniform size distribution. Most of the particles have the diameter of about 8 nm. We fabricated our inverted polymer solar cell with the configuration of ITO/pure and vanadium-doped TiO₂ NPs/P3HT:PCBM/PEDOT: PSS/Ag. The cross-sectional SEM image of the solar cell employed TiO₂ thin film NPs is shown in Fig. 6. This image confirms that the polymer sufficiently infiltrates the interval between nanoparticles. The thickness of the TiO₂ thin film is 30 nm.

Fig. 4.

	Figure Option ViewDownloadNew Window
	Fig. 4. (color online) (a) SEM image and (b) size distribution histogram of TiO2 thin film NPs fabricated via the solvothermal method.

Fig. 5.

	Figure Option ViewDownloadNew Window
	Fig. 5. (color online) (a) TEM image and (b) size distribution histogram of TiO2 thin film NPs fabricated via the solvothermal method.

Fig. 6.

	Figure Option ViewDownloadNew Window
	Fig. 6. Cross-sectional SEM image of the inverted polymer solar cell with TiO2 NPs.

Optical properties of pure and vanadium-doped TiO₂ thin films with the thickness of 30 nm were measured by a UV–Vis–IR spectrometer, as shown in Fig. 7. According to this spectrum, the vanadium-doped TiO₂ thin film has a high optical transmittance (>70%) in the visible spectrum.

Fig. 7.

	Figure Option ViewDownloadNew Window
	Fig. 7. (color online) Transmittance spectra of pure and vanadium-doped TiO2 thin films with the thickness of 30 nm.

The visible light transmittance decreases for pure TiO₂ thin film. Thus, the pure TiO₂ thin film shows lower transmittance than the doped thin film. According to the SEM images, better transmittance might be due to the surface morphology and crystallinity of the thin film.^[36–38] The energy band gap of pure and vanadium-doped TiO₂ thin film NPs is determined using the Tauc formula

Fig. 8.

	Figure Option ViewDownloadNew Window
	Fig. 8. (color online) Optical band gap of (a) pure and (b) vanadium-doped TiO2 thin film NPs.

Figure 9 shows the J–V curves of the inverted polymer solar cells employing pure and vanadium-doped TiO₂ thin film NPs under illumination. It shows that the cell using vanadium-doped TiO₂ thin film NPs has the best J–V performance. The solar cells using vanadium-doped thin film NPs show better J–V performance compared to their thickness counterparts, and this confirms the effectiveness of using vanadium as a dopant in cells. V_oc of the polymer fullerene is heavily dependent on the energy difference between the HOMO level of the donor and the LUMO level of the acceptor, so all of the cells use the same active layer, their V_oc are the same.^[40] Table 1 shows the extracted data of the fabricated cells. As predicted, the cells with vanadium-doped TiO₂ thin film NPs show higher efficiencies. The cell using vanadium-doped thin film NPs with thickness 60 nm has an efficiency ∼ 0.42% higher than its pure counterpart. The cell with 30 nm vanadium-doped TiO₂ thin film NPs has the highest efficiency of 3.68%, but the pure TiO₂ thin film with the same thickness shows an efficiency of 2.95% and this is because of the lower conductivity of the pure TiO₂ thin film. It is remarkable to see that even with J_sc of 10.70 mA/cm², the cell could not reach a higher efficiency. As the fill factor depends on the series resistance, for achieving higher PCEs even at this thickness, higher fill factors must be achieved. This somehow shows that at higher thicknesses, increasing the series resistance causes a loss in power conversion efficiency and in lower thicknesses while the series resistance is low, the lack of good interfacial area with active layer prevents the cell from having higher performance and this puts a higher power conversion efficiency limit in the thickness of 30 nm thin films. Also lower fill factor is another criterion affecting the performance of the cell. External quantum efficiency (EQE) of cells plays a great role in confirming the efficiency of cells and is a key parameter for characterizing solar cell performance under different conditions.^[9] Figure 10 shows the EQE curves of the cells using pure and vanadium-doped TiO₂ thin films NPs. Doped TiO₂ thin film NPs are superior to their pure counterpart. Considering that the integration of EQE should be proportional to J_sc, there is consistency between data obtained from the EQE curve and the solar simulator. Curves also show the decreases of the efficiency of these cells for wavelengths longer than 600 nm. So by enhancing the cell with materials that can absorb longer wavelengths, the cell performance can be improved.

Fig. 9.

	Figure Option ViewDownloadNew Window
	Fig. 9. (color online) J–V curves of devices using pure and vanadium-doped TiO2 thin films with different thicknesses.

Fig. 10.

	Figure Option ViewDownloadNew Window
	Fig. 10. (color online) External quantum efficiency of the cells with the best efficiency.

Table 1.

Table 1.

Table 1. Key parameters of inverted polymer solar cells under illumination. . Sample Voc/V Jsc/mA⋅cm−2 FF/% Efficiency/% 30 nm TiO2 0.552 9.2 58.21 2.95 60 nm TiO2 0.553 8.4 62.08 2.88 30 nm V doped TiO2 0.553 10.7 62.36 3.68 60 nm V doped TiO2 0.554 10 59.74 3.30

Table 1. Key parameters of inverted polymer solar cells under illumination. .

Employing vanadium-doped TiO₂ thin film NPs as the electron transport layer was effective and it improved the optical and electrical performance of the cells. Best thickness of our pure and vanadium-doped thin films was 30 nm. The cell using 30 nm vanadium-doped TiO₂ thin films achieved 3.68% power conversion efficiency with J_sc of 10.7 mA/cm² and V_oc of 0.553. The fill factor of this cell was 62.36%. This somehow shows that at higher thicknesses, increase of series resistance causes a loss in power conversion efficiency and in lower thicknesses while the series resistance is low, but the lack of good interfacial area with active layer prevents the cell from having higher performances and this puts a higher power conversion efficiency limit in the thickness of 30 nm thin films; also lower fill factor is another criterion affecting the performance of cell. So we suggest using other forms of vanadium-doped TiO₂, like nanorods and nanotubes.