A chitosan binder-based TiO2 photoelectrode is used in dye-sensitized solar cells (DSSCs). Field-emission scanning electron microscope (FE-SEM) images revealed that the grain size, thickness, and distribution of TiO2 films are affected by the chitosan content. With addition of 2.0 wt% chitosan to the TiO2 film (D2), the surface pore size became the smallest, and the pores were fairly evenly distributed. The electron transit time, electron recombination lifetime, diffusion coefficient, and diffusion length were analyzed by IMVS and IMPS. The best DSSC, with 2.0 wt% chitosan addition to the TiO2 film, had a shorter electron transit time, longer electron recombination lifetime, and larger diffusion coefficient and diffusion length than the other samples. The results of 2.0 wt% chitosan-added TiO2 DSSCs are an electron transit time of  s, electron recombination lifetime of  s, diffusion coefficient of  cm2 s−1, diffusion length of 14.81 μm, and a solar conversion efficiency of 4.18%.

1. Introduction

Since the Grätzel group discovered dye-sensitized solar cells (DSSCs), many researchers have become interested in them [15]. The low-cost, high-solar conversion efficiency of DSSCs is considered as a possible alternative to contemporary silicon solar cells [6, 7]. DSSCs are basically a thin layer solar cell formed by the sandwich arrangement of two electrodes, that is, a photoelectrode which is a few micron thick mesoporous TiO2 layer coated with photosensitizer, and a platinum (Pt) coated counter electrode. The interlayer space is filled with a soluble redox couple electrolyte, such as / [811]. TiO2 is chemically stable, nontoxic, and readily available in vast quantities. DSSCs have many components that have to be optimized; the mesoporous TiO2 layer is very important for increasing the solar conversion efficiency, among others [1218]. So far, the TiO2-based DSSCs fabricated using multilayer (about 14 μm thickness) approaches have shown a solar conversion efficiency of 11.3%, which is lower than the theoretical maximum (33%) [19, 20]. Many researchers, in order to increase the solar conversion efficiency in DSSCs, have studied improvements to the photoelectrodes, such as the synthesis of wide band-gap TiO2, small particle size of 10 to approximately 20 nm, large surface area of TiO2, and increased porosity. These can increase the adsorption of the dye, and by extension, the solar conversion efficiency could be increased [21, 22].

In this present research, chitosan was adopted as a new electrode binder for DSSCs. Chitosan is a polysaccharide composed mainly of -(1,4)-linked 2-deoxy-2-amino-D-glucopyranose units. Chitosan can be considered the most environment-friendly binder, and chitosan-based aqueous slurries possesses good viscosity and so can be considered as an effective electrode binder. We investigate natural chitosan binder-based DSSCs, describing the effect of different contents of chitosan on the activity of DSSCs. We selected for our research four amounts of 1.5 wt%, 2.0 wt%, 2.5 wt%, and 3.0 wt% of chitosan content in the TiO2 electrode, named as D1, D2, D3, and D4 respectively.

2. Experiment

2.1. Preparation of Chitosan Sol and TiO2 Paste

The chitosan used in this study was kindly supplied by Sehwa Co., Korea. The degree of deacetylation and molecular weight of the chitosan were 85% and 5.2 × 105 g/mol, respectively. Chitosan sol with different concentrations (1.5, 2.0, 2.5, and 3.0 wt%) were prepared by dissolving the proper amount of chitosan in 2 mL of 3% (v/v) aqueous acetic acid solution. The solution was mixed using a shaking incubator for 24 h at 250 rpm, and then the solution was left to stand for 24 h at room temperature, for complete hydration of the polymer and removal of bubbles.

The chitosan-based TiO2 paste was prepared in the following way: nitric acid treated TiO2 (P-25, Deagesa) power was added to 2 mL of prepared chitosan colloidal solution, and then stirring was maintained until the TiO2 colloid was well mixed with the chitosan solution slurry. For increased dispersal of the TiO2 paste, a three-roll mill (DEA WHA TECH., EXAKT50i) was used for about 7 h. The three-roll mill makes it possible to simultaneously mill and mix the paste. Table 1 shows the composition of chitosan sol-based TiO2 pastes of D1, D2, D3, and D4 used in this study.

2.2. Preparation of the Photo and Counter Electrode

The prepared TiO2 paste was cast on precleaned FTO (Pilkington FTO glass, 8 Ω/cm2), using the squeeze printing method [3, 2325]. The coated TiO2 films were sintered at 150°C for 3 h. The active area of the TiO2 film was 0.25 cm2. The TiO2 film was immersed into a 5 × 10−4 mol/L ethanol solution of Ru(dcbpy)2(NCS)2 (N719, Solaronix Co., Switzerland) overnight, then rinsed with anhydrous ethanol, and finally dried. The counter electrode was prepared using the squeeze printing technique and subsequently sintered at 450°C for 30 min. The counter electrode material was a Pt catalyst (Solaronix Co., Switzerland).

2.3. Assembly of the Testing Cells

The Pt electrode was placed over the dye-adsorbed TiO2 electrode, and the edges of the cell were sealed. The sealing was accomplished by hot-pressing two electrodes together at 110°C. The redox electrolyte was injected into the cell through two small holes drilled in the counter electrode. The / redox electrolyte was composed of 0.3 mol/L 1,2-dimethyl-3-propylimidazolium iodide (Sigma-Aldrich Co., USA), 0.5 mol/L 4-tert-butylpyridine (Sigma-Aldrich Co., USA), and 3-metoxypropionitrile. The holes were then covered and sealed, with a small square of sealing material and a microscope objective glass.

2.4. Measurements

The thermogravimetric (TGA) analyses of the chitosan sol were performed with an RIS diamond TG-DTA (PerkinElmer) analyzer. The chitosan sol was loaded into alumina pans and heated from room temperature to 530°C.

Field-emission scanning electron microscopy (FE-SEM) (Hitachi S-4700, Japan) was used to examine the film morphology, such as the surface of TiO2 films and thickness.

The electron transit time and electron recombination lifetime were measured by intensity modulated photocurrent spectroscopy (IMPS) and intensity modulated photovoltage spectroscopy (IMVS). Blue light-emitting diodes (LEDs, 475 nm) were used as the light source. The light intensities were modulated (10%), by modulating the bias applied to the LED with sine waves, in a frequency range typically from 0.1 Hz to 1000 Hz.

The photovoltaic properties were investigated by measuring the photocurrent-voltage characteristics under illumination, with an air mass (AM) of 1.5 (100 mW/cm2) simulated sunlight. The charge transport characteristics were investigated by intensity-modulated photovoltage spectroscopy (IMVS). The IMVS was measured using red light-emitting diodes (LED, 635 nm). The light intensities were modulated by 10%, in a frequency range typically from 0.01 to 100 Hz.

3. Results and Discussion

The thermal weight loss curves for the pure chitosan sol (2.0 wt%) and chitosan sol-based TiO2 paste (D2) are shown in Figure 1. This shows that there is a large weight loss in the range from room temperature to 80°C in the TG curve, which is due to the dehydration of the pure chitosan sol and chitosan sol-based TiO2 paste. This weight loss can be assigned to the thermal decomposition of aqueous acetic acid in the pure chitosan sol. As shown in the TGA curve of D2, we can verify that about 20 wt% residue is accurate, and the residue can be ascribed to the TiO2 powder. So the heating temperature of the TiO2 photoelectrode (after TiO2 casting on the FTO substrate) can be set to be very low.

The surface morphologies of the pure TiO2 photoelectrode and TiO2 photoelectrode containing 1.5 wt% (D1), 2.0 wt% (D2), 2.5 wt% (D3), and 3.0 wt% (D4) of chitosan were obtained by FE-SEM and are depicted in Figure 2. As can be seen in Figure 2, the TiO2 pastes containing different weight of chitosan could result in the TiO2 films having different distributions of pore diameters. It was found that the TiO2 film prepared with the D2 paste shows a smaller pore structure and that prepared with D1, D3, and D4 pastes has a larger pore diameter. The penetration of electrolyte depends on the pore size and porosity of the TiO2 film [26]. We will investigate the influence on lifetime of the TiO2 electrode, as well as the performance of the DSSC, by varying the amounts of chitosan content in the TiO2 electrode.

Figure 3 shows FE-SEM images of the cross-section of chitosan sol-based TiO2 films. In this figure, the thickness of TiO2 of D1, D2, D3, and D4 is 6.8, 6.9, 6.8, and 6.6 μm, respectively. The solar conversion efficiency of DSSCs depends upon several factors, including the porosity, surface area, and amount of dye loading [26].

Figure 4 shows the photocurrent-voltage characteristics of different chitosan-based DSSCs. The solar cell was irradiated with a 1,000 W xenon lamp with a light intensity of 100 mA cm−2 as a light source. As shown in Figure 4, the D2-based DSSC has a higher solar current density than the D1, D3, and D4-based DSSCs. Table 2 shows the data of the photocurrent-voltage characteristics. The open circuit voltage , short-circuit current density , fill factor , and the solar energy conversion efficiency of the D2-based DSSC are 0.69 V, 10.15 mA cm−2, 59.41%, and 4.18%, respectively. The of D1, D3, and D4-based DSSCs are 3.8%, 3.95%, 3.27%, respectively. Grätzel and coworkers have reported on a highly efficient (7–11%) photoelectrochemical cell based on sensitized oxidation of iodide on a nanocrystalline TiO2 electrode using an artificial binder [27, 28]. However, we used natural binder (chitosan) for TiO2 paste. In comparison to the artificial binder, the efficiency of DSSCs using natural binder is not high. However, there is little data available on the use of natural binder for TiO2 paste in the literature. Yang et al. [29] reported on a DSSC with a novel Pt counter electrodes using pulsed electroplating techniques: the efficiency of DSSC was 3.85%. Deepa et al. [30] reported on efficiency enhancement in DSSCs using metal nanoparticles: a size-dependent study; the efficiency of DSSCs range from 0.43 to 1.52. Compared to the research of Yang et al. and Deepa et al., the efficiency of DSSCs using natural binder is high. The TiO2 photoelectrode morphology can also influence photocurrent generation. The higher efficiency of photocurrent generation has been explained by a reduction in grain boundaries, higher dye loadings, and slower recombination rates [3135]. The electron recombination lifetimes within DSSCs are determined primarily by the recombination of electrons with iodine, electrolyte, and oxidized sensitizers. To understand the charge transport characteristics, the films were investigated by IMPS and IMVS. The electron transit time ( ) and electron recombination time can be determined using the equations and . In the equations, and are the frequencies giving the largest imaginary components in IMPS and IMVS, respectively. The IMPS and IMVS plots are shown in Figure 5. Figure 6 displays the diffusion coefficient of chitosan-based DSSCs, and Table 3 shows the electron transport characteristic data. The electron transit time of D1, D2, D3, and D4-based DSSCs is 3.121 × 10−3, 2.189 × 10−3, 2.295 × 10−3, and 4.056 × 10−3 s−1, respectively. The electron recombination lifetime of the D2-based DSSC was lower than those of the other samples. For the D2-based DSSC, the recombination lifetime is 6.336 × 10−2 s. This result clearly indicates that electron recombination with the oxidized species is reduced in the D2 TiO2 film. This can be understood by either looking at improvements in the network of TiO2 nanoparticles, or at the electron behavior during transition in the D2-based TiO2 film. From the FE-SEM, I-V, and electron transport characteristic results, the D2-based DSSCs have a very good film surface uniformity and high porosity. Consequently, the D2-based DSSCs have increased surface absorption and enhanced solar energy conversion efficiency.

4. Conclusions

DSSCs were fabricated using different chitosan binder sol (1.5~3.0 wt%)-based photoelectrodes. The chitosan binder sol had low calcination temperature (~150°C) and successfully prepared low temperature TiO2 photoelectrodes. The D2 (~2.0 wt%)-based DSSCs had faster electron transit time and slower electron recombination time than the other samples (D1, D3, and D4). The D2-based DSSC exhibited higher solar conversion efficiencies than D1, D3, and D4-based DSSCs, because the electron recombination is more than 1.5~3 times slower in the D2-based DSSC. The electron diffusion coefficient of the D2-based DSSC was 3.463 × 10−5 cm2 s−1, and the diffusion length was 14.81 μm. The efficiency of the D2 based DSSC was optimized to achieve 4.16% under illumination of simulated solar light, AM 1.5 global ( V,  mA cm−2; fill factor is 59.41%).


This research was supported by Specialized Local Industry Development Program through Korea Institute for Advancement of Technology (KIAT) funded by the Ministry of Trade, Industry and Energy (1415127923-R0002040). And the Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, Science and Technology (2012010655).