Background

Dye-sensitized solar cells (DSSCs) have attracted considerable interests due to their simpler fabrication and low production costs compared with conventional silicon-based solar cells [1, 2]. A traditional DSSC consists of a transparent photoanode with dye-sensitized mesoporous thin-film-like TiO2 or ZnO, I−/I3− redox electrolyte, and a counter electrode (CE) with a catalytic layer deposited on FTO substrate. As one of the most crucial components of DSSC, the CE works as a catalyst for the reduction of I3− to I−, and the materials used in catalytic layer and conductive substrates significantly affect the performance and costs of the DSSCs. Platinized FTO is the most common material for CE as it has good conductivity and high catalytic activity. However, noble metal platinum is expensive, scarce, and easy to be eroded by the I−/I3− electrolyte [3, 4]. Moreover, the Pt catalytic layer is usually prepared by thermal annealing or electrodeposition method, and both methods require high temperature (450°C), which is beyond the sustaining ability of plastic substrates to realize the flexible DSSCs. The common FTO substrates are very expensive and hard, also preventing the production of flexible DSSCs. Therefore, it is imperative to develop Pt- and FTO-free CEs with low cost and good catalytic activity for DSSCs.

Many reported materials have been used as the substitute for Pt-based CEs like conductive polymers (polyaniline [5], ploypyrrole [6], poly(3,4-ethylenedioxy-thiophene) (PEDOT) [7], carbon materials (graphene [8], carbon black [9], carbon nanotube [10], etc.), and most of them have lower catalytic activity than Pt [11]. In order to achieve a cost-effective Pt-free CE, PEDOT:PSS has attracted much attention because of good catalytic activity, better film-forming property, low cost, and easy coating [12–14]. Modified PEDOT:PSS has potential to replace TCO in organic electronics for its high conductivity [15]. Though with many of strengths, the catalytic ability of DSSC with PEDOT:PSS/FTO CE still exists a distance from Pt/FTO CE and needs to be further improved.

Consequently, in this work, a hierarchical TiO2-PEDOT:PSS/PEDOT:PSS/glass CE was used in the fabrication of DSSC. The TiO2-PEDOT:PSS layer was fabricated utilizing the mixture of PEDOT:PSS and TiO2 nanoparticles. The neat PEDOT:PSS layer acts as a high conductive electrode in order to develop charge passageway. This hierarchical TiO2-PEDOT:PSS/PEDOT:PSS/glass CE performed better catalytic activity than the PEDOT:PSS/FTO CE, and as a result, the DSSC using TiO2-PEDOT:PSS/PEDOT:PSS/glass CE also performs good photovoltaic properties.

Methods

Preparation of TiO2 photoanodes

TiO2 paste was blade-coated on FTO substrates and subsequently sintered at 450°C for 30 min. After cooling down to room temperature, the samples were put into 40 mmol/L TiCl4 solution at 70°C for 30 min and then sintered at 450°C for 30 min. Finally, after cooling down to 80°C, the as-prepared TiO2 photoanodes were soaked in the ethanol solution of N719 dye for 24 h.

Preparation of the counter electrodes

In total, we have prepared four kinds of CEs, including Pt/FTO, PEDOT:PSS/FTO, TiO2-PEDOT:PSS/FTO, and TiO2-PEDOT:PSS/PEDOT:PSS/glass. The Pt/FTO CE was prepared by spraying H2PtCl6 solution on the pre-cleaned FTO substrate and subsequently sintered at 450°C for 15 min. The PEDOT:PSS/FTO and TiO2-PEDOT:PSS/FTO CEs were fabricated by spin coating PEDOT:PSS (Clevios PH 1000, purchased from Heraeus, Hanau, Germany) solution and TiO2-PEDOT:PSS solution onto FTO substrates, respectively. The TiO2-PEDOT:PSS/PEDOT:PSS/glass was obtained by spin coating PEDOT:PSS mixed with 6% volume of ethylene glycol (EG) on glass substrate (5,000 rpm/s for 30 s) and sintered at 120 °C for 15 min. This process was repeated four times. Then, the TiO2-PEDOT:PSS (40 mg P25 powder added in 1 ml PEDOT:PSS solution) solution was spin-coated on top of the PEDOT:PSS layer at 1,000 rpm/s for 40 s and sintered at 120°C for 15 min. Finally, the resultant substrates were immediately put into EG for 30 min and then dried in the oven at 120°C for 15 min.

Fabrication and characterization of DSSCs

The processed TiO2 photoanodes have an active area of 0.16 cm2, and these prepared CEs were assembled together with 60-μm surlyn film, respectively. The I−/I3− electrolyte was injected through the interspace and sealed with paraffin.

The sheet resistance of the catalytic layers was measured using a four-probe tester (model RTS-8, Four Probe TECH, Guangzhou, China). The surface morphologies of CEs were scanned by field emission scanning electron microscope (quanta 200 F, FEI, OR, USA). Electrochemical impedance spectroscopy (EIS) and Tafel polarization curves were measured using an electrochemical workstation (model CHI600, CH Instruments, Inc., Austin, TX, USA) at room temperature. The current density-voltage characteristics of photocurrent density-photovoltage were simulated at AM 1.5G illumination (100 mV cm−2, XES-301S, SAN EI, Osaka, Japan) and recorded by a Keithley source meter (Keithley, Cleveland, OH, USA).

Results and discussion

The sheet resistance of different CEs, PEDOT:PSS/FTO CE, TiO2-PEDOT:PSS/FTO CE, TiO2-PEDOT:PSS/PEDOT:PSS/glass CE, and Pt/FTO CE, is 6.3, 7.5, 35, and 7.2 Ω sq−1, respectively. Though the sheet resistance of TiO2-PEDOT:PSS/PEDOT:PSS/glass CE is larger than that of TiO2-PEDOT:PSS/FTO CE and Pt/FTO CE, it is still qualified, i.e., the sheet resistance below 100 Ω sq−1 can be used as electrode [16, 17].

The surface morphologies of pristine PEDOT:PSS film and TiO2-PEDOT:PSS composite film are depicted in Figure 1a,b, respectively. As is shown in the two images, the surface of modified PEDOT:PSS film is almost smooth, while the TiO2-PEDOT:PSS composite film is rough and has a large surface area which is good for catalytic reduction of I3−. In TiO2-PEDOT:PSS composite film, as shown in Figure 1b, the thin catalytic layer is composed of TiO2 nanoparticles, and their diameter ranges from 20 to 50 nm. These nanoparticles are uniformly dispersed in PEDOT:PSS, forming a network structure, beneficial for electron conduction. Therefore, the performance of DSSCs with TiO2-PEDOT:PSS/PEDOT:PSS/glass CEs could be greatly improved by the addition of TiO2 nanoparticles.

Figure 1
figure 1

SEM images of PEDOT:PSS film (a) and TiO 2 -PEDOT:PSS composite film (b).

Full size image

A typical EIS spectrum for a DSSC exhibits three semicircles in the Nyquist plot, as is shown in Figure 2a. Traditionally, the first semicircle in high-frequency region corresponds to charge transfer resistance (Rct) of the CE/electrolyte interface, while the second semicircle in the middle-frequency region represents charge transfer and recombination resistance in the TiO2/dye network [18, 19]. The low-frequency semicircle is attributed to the Nernst diffusion impedance of the I−/I3− redox couple. From Figure 2a, we can obviously see that the spectra of TiO2-PEDO:PSS/PEDO:PSS/glass CE has a smaller semicircle than that of the POEDT:PSS/FTO CE, which indicates that TiO2-PEDO:PSS/PEDO:PSS/glass CE has a better catalytic activity than POEDT:PSS/FTO CE. The simulated values of series resistance (Rs), charge tansfer resistance (Rct), and diffusion element (Zw1) of corresponding cells calculated by Zview software are shown in Table 1. The simulated Rct and Zw1 of TiO2-PEDO:PSS/PEDO:PSS/glass CE (1.51 and 4.02 Ω cm2, respectively) are lower than those of PEDOT:PSS/FTO CE (4.47 and 11.28 Ω cm2, respectively), indicating that the addition of TiO2 nanoparticles greatly improves the catalytic activity for the redox reaction. The Rs value of TiO2-PEDOT:PSS/PEDOT:PSS/glass CE is higher than that of PEODT:PSS/FTO CE due to a lower conductivity of PEDOT:PSS layer than that of FTO substrate, and the result is in accordance with the conclusion from the sheet resistance. However, the Rct of TiO2-PEDOT:PSS/PEDOT:PSS/glass composite CE is lower than that of Pt/FTO CE (5.73 Ω cm2) which is opposite to the traditional standpoint that a smaller Rct may lead to a higher fill factor (FF) and η in photovoltaic performance. However, for TiO2-PEDOT:PSS/PEDOT:PSS/glass CE, the charge transfer of the CE/electrolyte interface is mainly illustrated by the second semicircle of the spectra. Similar findings have been reported by He et al. [20] and Roy-Mayhew et al. [8], and they contend that the second semicircle is due to the electrolyte/CE interface.

Figure 2
figure 2

Electrochemical characters. Nyquist plots (a) and Tafel polarization curves (b) of DSSCs based on PEDOT/FTO CE, TiO2-PEDOT:PSS/PEDOT:PSS/glass CE, and Pt/FTO CE.

Full size image
Table 1 Electrochemical impedance spectra (EIS) parameters of PEDOT/FTO CE, TiO 2 -PEDOT:PSS/PEDOT:PSS/glass CE, and Pt/FTO CE
Full size table

Furthermore, Tafel polarization curves were carried out on the same dummy cells used in EIS measurement to investigate the interfacial charge transfer properties of CE/electrolyte, and the corresponding results are shown in Figure 2b. The exchange current (J0) = 0.58 mA, calculated from the intersection of the linear cathodic and anodic Tafel polarization curves [16, 21], was derived from the TiO2-PEDOT:PSS/PEDOT:PSS/glass composite film and higher than that of PEDOT:PSS/FTO film (0.14 mA). Correspondingly, the catalytic activity of TiO2-PEDOT:PSS/PEDOT:PSS/glass composite CE is much higher than that of PEDOT:PSS/glass CE, which demonstrates that the big surface area of TiO2 nanoparticles enhances the reduction of I3− to I− remarkably. Though the J0 of TiO2-PEDOT:PSS/PEDOT:PSS/glass composite CE is smaller than that of Pt/FTO CE (1.2 mA), the former still exhibits superior catalytic activity and has great potential to act as CE for DSSC.

Figure 3 presents the photocurrent density-voltage (J-V) curves of DSSCs using PEDOT:PSS/FTO CE, TiO2-PEDOT:PSS/PEDOT:PSS/glass CE, and Pt/FTO CE, respectively, and the related photovolatic parameters are shown in Table 2. There is little difference in Voc values of these three cells. The FF of the DSSC with PEDOT:PSS/FTO CE is just 0.43 because of the poor catalytic activity of PEDOT:PSS solution. After modified by the TiO2 nanoparticles, the DSSC with TiO2-PEDOT:PSS/PEDOT:PSS/glass CE has obtained higher FF of 0.51 and thus higher η = 4.67% (increasing 22% compared with 3.64% for the DSSC with PEDOT:PSS/FTO CE). This is mainly due to the reduced charge transfer resistance and porous diffusion impedance because of the large electrochemical surface area in the porous TiO2-PEDOT:PSS layer. Compared with DSSC based on Pt/FTO CE, the one with TiO2-PEDOT:PSS/PEDOT:PSS/glass CE has lower FF, but its overall efficiency has already reached 91.39% of the one with Pt/FTO CE. It is noticeable that the performance of TiO2-PEDOT:PSS/PEDOT:PSS layers can befurther enhanced by optimazation of their weight ratio and the film thicknesses, referring to the previous studies using TiO2-PEDOT:PSS/FTO CE [22]. With such an excellent performance, the TiO2-PEDOT:PSS/PEDOT:PSS/glass CE has great potential to be a substitute for Pt- and FTO-based CEs which are very expensive and account for a large part of the cost. At the same time, due to the low preparation temperature, the TiO2-PEDOT:PSS/PEDOT:PSS composite film can be applied in flexible cells and make them more functional and lightweight.

Figure 3
figure 3

Current density-voltage ( J - V ) characteristics of DSSCs based on PEDOT/FTO, TiO 2 -PEDOT:PSS/PEDOT:PSS/glass, and Pt/FTO CEs.

Full size image
Table 2 The performances of dye-sensitized solar cells with different CEs measured under an AM 1.5G illumination
Full size table

Conclusions

In summary, we utilize a facile wet method to fabricate a novel hierarchical Pt- and FTO-free CE for the dye-sensitized solar cell. It is found that the TiO2 doped PEDOT:PSS catalytic activity layer will dramatically affect the electrochemical properties of the final device. By adjusting the composition of TiO2, the properties of CE have been optimized preliminarily. Because of the large active area of TiO2 nanoparticles, the proposed composite CE shows excellent enhancement in the conductivity and the superior catalytic activity for the reduction of I3− to I−. The conversion efficiency is increased by 22% than that of the DSSC with PEDOT:PSS/FTO CE and is comparable to that of the DSSC with traditional Pt/FTO CE. After further optimization, the TiO2-PEDOT:PSS/PEDOT:PSS/glass CE can be more cost-effective, high efficient, and flexible to replace Pt and FTO CEs and more broadly used for future commercial applications.