Fast sensitization process of ZnO-nanorod-array electrodes by electrophoresis for dye-sensitized solar cells

Abstract

A novel electrophoresis based dye adsorption technique is developed for the process of ZnO-nanorod-array dye-sensitized solar cells (DSSC). By applying a constant current electrophoresis procedure, ZnO electrodes were successfully sensitized by N719 dye, and the effects of varying the electrophoretic current and duration time on the performance of DSSC were investigated. It is found that abundant dye adsorption can be obtained within significantly reduced sensitization time by electrophoresis compared to standard immersion method. Thicker ZnO-nanorod-array film and lower electrophoretic current requires longer electrophoresis time to reach the optimal sensitization effect, while very long sensitization time will result in dye agglomeration, which is harmful to the performance of DSSC. The fast dye adsorption process is more advantageous as the ZnO-nanorod-array film becomes thicker because dye molecules can readily penetrate into the depth of ZnO film in a very short time, during which the formation of Zn²⁺/dye complex can be suppressed.

---

1. Introduction

Dye-sensitized solar cells (DSSC) are promising photovoltaic devices because of their competitive power conversion efficiency, as well as their low cost and simple fabrication process, which can meet the requirements of new energy in the future.^1,2 TiO₂ nanocrystalline is the most commonly used material for DSSC but other wide-band-gap semiconductors, including ZnO and SnO₂, are also showing promising properties in terms of electron mobility^3,4 and nanostructure diversity.^5–7 ZnO nanorods demonstrate better electron transport properties than nanoparticles because of high crystallinity and low recombination rate because of the internal electric field existing in the nanorod, which can inhibit electrons escaping back to the electrolyte.⁸ However, ZnO-nanorod-array based DSSC suffer from low photovoltaic performance not only because of poor dye adsorption within limited surface area but also because of obstructions from the formed Zn²⁺/dye complex during prolonged sensitization process, which will hinder electron injection from dye to ZnO.^9–11

In the fabrication of DSSC, sensitization is of significant importance because it involves the formation of interface between dye molecules and the ZnO semiconductor, which is the key agency for electrons and holes to separate. Traditional dye adsorption is implemented by immersing ZnO electrodes into dye solutions for a period of time which can last for several hours or even days. For ZnO, an amphoteric oxide which can be corroded by most of the commercial dyes, it is especially harmful to have the electrodes soaking in acidic dye solutions for a long time. Enormous surface destruction and dramatic performance decline will occur when the ZnO electrodes are immersed in N719 dye for as short as 20 min.¹² Few studies aiming to overcome this drawback have been reported, including developing new types of sensitizers^13,14 and modifying the solvent used for dye adsorption.¹⁵ Recently, we have used ZnO nanoparticles¹⁶ and nanorod-array¹⁷ electrodes modified by Al₂O₃ or SiO₂ layers to improve their stability in dye loading process. Herein, we developed a new sensitization technique with the use of electrophoresis to enhance dye adsorption. With the help of electric field, sufficient dye loading can be achieved within a couple of minutes. The effects of varying electrophoretic current and duration on the performance of DSSC are investigated. The electrophoresis method is advantageous compared to the traditional immersing method because the reduced sensitization time is helpful for reducing the corrosion of ZnO.

2. Experimental

2.1 Fabrication of ZnO-nanorod-array electrodes

ZnO-nanorod-array electrodes were synthesized by the hydrothermal method.¹⁰ Briefly, a 0.5 M colloid solution with equivalent zinc acetate (Zn(CH₃COO)₂·2H₂O) and ethanolamine (NH₂CH₂CH₂OH) dissolved in 2-methoxyethanol (CH₃OCH₂CH₂OH) was spin-coated on FTO glass (8 Ω sq⁻¹, 80% transmittance) substrates followed by heat treatment at 350 °C for 0.5 h to form a ZnO seed layer. Afterward, the coated substrates were immersed into the precursor solution containing 0.05 M Zn(NO₃)₂ and 0.05 M methenamine for a hydrothermal reaction at 95 °C. After reaction for 12 h, a layer of ZnO-nanorod-array film was synthesized on the FTO glass substrate. To produce longer ZnO nanorods and a thicker nanorod-array film, hydrothermal reactions were repeated several times. Meanwhile, polyetherimide (PEI) was added to the precursor solution to keep the c-axis growth of ZnO nanowires.

2.2 Assembly of DSSC

A N719 dye [cis-bis-(isothiocyanato)-bis-(2,2′-bipyridyl-4,4′-dicarboxylato) ruthenium(II) bis-tetrabutyl ammonium, Dalian Heptachroma, China] dispersed in absolute ethanol solution with the concentration of 30 mM was used in the sensitization process. FTO glass substrate coated with platinum on the conducting surface was utilized as a counter electrode. The ZnO electrode with an active area of 0.25 cm² was fixed with the counter electrode by thermal-plastic spacers (Suryln 1702, 25 μm thickness). Liquid electrolyte containing 0.1 M lithium iodide, 0.1 M iodine, 0.5 M tetrabutylammonium iodide, and 0.5 M 4-tert-butylpyridine in acetonitrile was injected into the gap of the two electrodes by capillary action.

2.3 Characterization and measurements

The amount of adsorbed dye was determined by desorbing the dye from the ZnO surface into a solution of 0.1 mM NaOH in ethanol and water and measuring its absorption spectrum by a UV-vis-NIR spectrophotometer (Varian, Cary 5000). Field emission scanning electron microscopy (FE-SEM) (Zeiss, SUPRA-55) and energy dispersive X-ray spectroscopy (EDS) (Link-Inca) were used to characterize the morphology and composition of the electrodes, respectively. J–V Curves of DSSC were measured by electrochemical interface instruments (Solartron SI 1287/SI 1260) under AM1.5G illumination of the solar simulator (Oriel, 91159A, 100 mW cm⁻²).

3. Results and discussion

A commercial electrochemical cell was employed for the electrophoresis process. As schematically shown in Fig. 1a, it is composed of two electrodes immersed in the 30 mM N719 ethanol solution. One of the electrodes is the ZnO-nanorod-array coated FTO glass with an active area of 0.7 cm², and the opposite one is a piece of FTO glass deposited with platinum. An electrochemical interface instrument was used to provide constant current and record voltage changes. As control experiments, ZnO electrodes were also sensitized by a standard method, which was performed by immersing ZnO electrodes into identical dye solution for different durations.

Fig. 1 (a) Schematic diagram of the electrophoresis system. (b) Absorption spectroscopy of ZnO electrode sensitized by different methods. The inset table shows the calculated dye adsorption amounts. The right inset shows the photograph of ZnO electrodes sensitized by different methods – PC-ZnO: positively connected ZnO electrode at 25 °C for 60 s, NC-ZnO: negatively connected ZnO electrode at 25 °C for 60 s, I-ZnO (60): immersing method at 60 °C for 60 min, I-ZnO: immersing method at 25 °C for 60 s, B-ZnO: bear ZnO. (c) Photograph of ZnO electrodes sensitized with an electrophoresis current of 0.1 mA for different durations.

To determine the charge condition of dye molecules in ethanol, ZnO electrodes were connected to the positive or negative terminal of the constant current source. After electrophoresis with a current of 0.1 mA for 60 seconds, an interesting observation occurred when comparing the color changes of the electrodes. The right inset of Fig. 1b illustrates the photograph of ZnO electrodes sensitized under different conditions. As can be seen, the positively connected ZnO electrode (PC-ZnO) shows a dark red color, while the negatively connected ZnO electrode (NC-ZnO) shows practically no change compared with the bare ZnO (B-ZnO) electrode. When the ZnO electrode is immersed into the identical dye solution directly for 60 seconds (referred as I-ZnO), only a faint pink color is observed. Fig. 1b illustrates the absorption spectroscopy of the ZnO electrode sensitized by different methods. The PC-ZnO electrode shows the strongest absorption among the four samples, followed by the I-ZnO (60) and I-ZnO electrodes, which represent the ZnO electrode directly immersed into dye solutions at 60 °C for 60 min and 25 °C for 60 s, respectively. The inset table illustrates the dye adsorption amounts, which were calculated by the intensity of the absorption peak. The dye uptake amount of the PC-ZnO electrode was 2.04 × 10⁻⁷ mol cm⁻², which was more than two folds than that of the I-ZnO (60) electrode. This unambiguously confirmed that an enhanced dye adsorption can be obtained on the positively charged electrode. In addition, with the electrophoresis current of 0.1 mA, the color of the electrodes gradually become darker as the electrophoresis time is prolonged, as presented in Fig. 1c. All these results indicate that N719 ethanol solution behaves like a colloid and the molecules are probably negatively charged because they tend to adsorb on the positive electrode. Consequently, all ZnO electrodes were set as positive electrodes during the electrophoresis process in our following study.

To further investigate the effect of electrophoresis on ZnO electrodes, scanning electron microscopy (SEM) was used to observe the surface morphology of the electrodes. As shown in Fig. 2a blank ZnO electrode without any sensitization typically consists of pristine ZnO nanorods, whose composition is detected by the energy dispersive X-ray spectroscopy (EDS) in the inset. After immersing in dye solution for 60 seconds, the surface morphology shows no changes (Fig. 2b) but trace Ru is evidenced by EDS, which reveals the presence of the N719 sensitizer. When the electrode is sensitized by electrophoresis with a current of 0.1 mA for 60 seconds, as shown in Fig. 2c, a thin layer of foreign matter is found in the spaces of nanorod arrays. Local EDS analysis demonstrates that these foreign matters contain high levels of Ru with an atom percentage of about 2.74%. The penetrated dense layer can also be observed in the cross-sectional view of the electrode. The formation of this layer may be attributed to the accumulation of dye molecules induced by the strong electric field, which forces numerous charged molecules to migrate towards ZnO electrode. In this circumstance, the adsorption sites on ZnO nanorods can saturate in a very short time. As electrophoresis time is prolonged, excess dye molecules will agglomerate in the pores of the electrode in response to the constant current. As the electrophoresis current is decreased to 0.01 mA, although Ru can still be detected on the surface of the ZnO electrode, no obvious foreign layers can be found both on the surface and the cross section of the electrode, which may indicate that the agglomeration of dye molecules is effectively diminished. Notably, as presented in Fig. 2b and d, no evident surface destruction of ZnO nanorods can be seen, as we previously observed,¹⁰ which implies that the time scale used for sensitization here is sufficiently short to avoid the corrosion of ZnO.

Fig. 2 SEM and EDS results (left bottom insets) of: (a) ZnO electrode without sensitization treatment, (b) ZnO electrode sensitized by immersing method for 60 seconds, (c) ZnO electrode sensitized by electrophoresis with a current of 0.1 mA for 60 seconds, top right inset is the cross-sectional view, (d) ZnO electrode sensitized by electrophoresis with the current of 0.01 mA for 60 seconds, top right inset is the cross-sectional view.

To compare the effect of different sensitization conditions on the performance of solar cells, J–V curves of the DSSC are plotted and photoelectric conversion efficiencies (η) are calculated. As can be seen in Fig. 3a and d, for the ZnO electrode with a thickness of ∼6 μm and an electrophoretic current of 0.1 mA, a decreasing η from 0.20% to 0.10% is presented with the sensitization time increasing from 15 to 90 seconds. This can be attributed to the agglomeration of dye molecules on the surface of ZnO-nanorod-array film, as identified by the SEM results, which is induced by the very large electrophoresis current. Although the electrode is equipped with abundant sensitizers, most of them are not directly anchored to ZnO, which will instead retard the electron injection efficiency. When the current is set at 0.01 mA, as shown in panel b and e of Fig. 3, the photovoltaic performance of the DSSC first increases and then reduces as the electrophoresis time is prolonged. The peak value achieved is 0.37% with the sensitization time of 90 seconds. Shorter and longer duration time will result in inadequate adsorption and a surplus pile of sensitizers, respectively, both of which can cause declined photocurrent. Moreover, better performance is achieved when the electrophoretic current is reduced from 0.1 mA to 0.01 mA. This again confirms that 0.1 mA is very large for the electrophoresis process, and a smaller current is desirable to avoid dye agglomeration.

Fig. 3 Performance of DSSCs based on ZnO-nanorod-array with a thickness of ∼6 μm. (a) and (d) Electrophoresis with a current of 0.1 mA. (b) and (e) Electrophoresis with the current of 0.01 mA. (c) and (f) Immersing method with 60 °C.

For optimized DSSC performance, electrophoretic sensitization with an electrophoretic current of 0.001 mA was conducted, and the results are presented in the ESI.† As shown in Fig. S1,† photoelectric conversion efficiency as high as 0.42% has been obtained with a significantly prolonged electrophoresis time of 480 seconds. We further decrease the concentration of the dye solution to 10 mM and performed electrophoretic sensitization. Fig. S2† shows the J–V curves and photoelectric conversion efficiencies of DSSC. No discrepancy in results is observed between the situation where the dye concentrations of 30 mM and 10 mM were utilized. This is reasonable because we applied a constant current in the electrophoresis process, and the dye adsorption should be proportional to the current density and duration time. The influence of dye concentration can be neglected.

Traditional sensitization method in which ZnO electrodes are immersed in identical dye solution at 60 °C for a period of time has also been performed to further assess the effect of electrophoresis. As shown in Fig. 3c and f, photoelectric conversion efficiency as high as 0.48% can be achieved when the sensitization time is 30 minutes. The results of electrophoresis method are inferior to that of the traditional method. This is probably because the electrophoretic current applied here is still unsuitable for the ZnO-nanorod-array film with a thickness of ∼6 μm.

We then adopted ZnO-nanorod-array electrodes with a thickness of ∼10 μm for the sensitization process. With the electrophoretic current of 0.1 mA and 0.01 mA, the highest η obtained are 0.57% and 0.88%, respectively, as shown in Fig. 4d and e. A thicker ZnO-nanorod-array possesses larger surface area for dye molecules to attach, thus larger electrophoresis current and longer sensitization time is required to allow dye molecules to penetrate into the depth of the film. Fig. 4c presents the J–V behaviour of DSSC sensitized by the immersing method, where a short-circuit current density (J_sc) of 2.44 mA cm⁻² and η of 0.86% are demonstrated (see Table 1) with an immersing time of 40 minutes.

Fig. 4 Performance of DSSCs based on ZnO-nanorod-array with a thickness of ∼10 μm. (a) and (d) Electrophoresis with a current of 0.1 mA. (b) and (e) Electrophoresis with a current of 0.01 mA. (c) and (f) Immersing method with 60 °C.

Table 1 Peak photovoltaic parameters obtained under different conditions

Thickness of ZnO film (μm)	Sensitization time	Jsc (mA cm^−2)	Voc (V)	Fill factor	η (%)
a Electrophoresis method with a current of 0.1 mA.b Electrophoresis method with a current of 0.01 mA.c Immersing method with 60 °C.
6	15^a s	0.76	0.53	0.41	0.20
6	90^b s	1.32	0.58	0.47	0.37
6	30^c min	1.67	0.60	0.48	0.48
10	60^a s	1.84	0.59	0.45	0.57
10	120^b s	2.78	0.62	0.51	0.88
10	40^c min	2.44	0.60	0.49	0.86

---

Unlike the case of 6 μm thick ZnO electrodes, 10 μm thick ZnO electrodes sensitized by electrophoresis method show comparable photovoltaic performance with the standard immersing method. This improvement can be attributed to two reasons. As the thickness of ZnO-nanorod-array film increases to ∼10 μm, the expanded surface area allows more dye molecules to adsorb on ZnO and thereby eliminates the agglomeration of dye molecules with the same electrophoresis current. In addition, it has been reported⁹ that acidic dye solution can dissolve ZnO into Zn²⁺ followed by the formation of Zn²⁺/dye complex on the surface of ZnO electrode. After a prolonged immersing time, although dye molecules can approach the depth of the ZnO-nanorod-array film and ensure sufficient dye adsorption, the outer layer of the ZnO film is inevitably corroded by dye solution. It is reasonable to infer that after immersing for 40 minutes, Zn²⁺/dye complex formed on the surface of the ZnO nanorods. Fortunately, for the electrophoresis process, the entire sensitization process is completed within hundreds of seconds, during which ZnO corrosion can be effectively reduced. From the SEM image presented in Fig. 2, no surface destruction can be observed for ZnO nanorods sensitized in N719 dye for as short as 60 seconds, which agrees with this speculation. As a result, the formation of Zn²⁺/dye complex is suppressed, and the performance of DSSC is improved.

4. Conclusions

To summarize, electrophoresis technique is introduced to the sensitization process of ZnO-nanorod-array based DSSC. The adsorption of dye has been dramatically accelerated and enhanced with the assistance of constant current electrophoresis. By adjusting the thickness of ZnO-nanorod-array film and the electrophoresis current and duration time, an optimal photoelectric conversion efficiency of 0.88% can be achieved, which is comparable with the DSSC sensitized by classical immersing method. Electrophoresis sensitization method still remains important for further improving the performance of DSSC when all the electrophoresis parameters are precisely controlled. The fast dye adsorption process appears to be more advantageous as the ZnO-nanorod-array film becomes thicker because dye molecules can readily penetrate into the depth of the ZnO film in a very short time, during which the formation of the Zn²⁺/dye complex is suppressed.

Acknowledgements

This work was supported by the National Major Research Program of China (2013CB932602), the Major Project of International Cooperation and Exchanges (2012DFA50990), NSFC (51172022, 51232001, and 51372020), the Fundamental Research Funds for Central Universities, the Program for New Century Excellent Talents in University, Beijing Higher Education Young Elite Teacher Project, the Programme of Introducing Talents of Discipline to Universities, and Program for Changjiang Scholars and Innovative Research Team in University.

Notes and references

1. B. O'regan and M. Grtzeli, Nature, 1991, 353, 24.
2. A. Yella, H.-W. Lee, H. N. Tsao, C. Yi, A. K. Chandiran, M. K. Nazeeruddin, E. W.-G. Diau, C.-Y. Yeh, S. M. Zakeeruddin and M. Grätzel, Science, 2011, 334, 629–634.
3. I. Gonzalez-Valls and M. Lira-Cantu, Energy Environ. Sci., 2009, 2, 19–34.
4. C. Bauer, G. Boschloo, E. Mukhtar and A. Hagfeldt, J. Phys. Chem. B, 2001, 105, 5585–5588.
5. S. H. Ko, D. Lee, H. W. Kang, K. H. Nam, J. Y. Yeo, S. J. Hong, C. P. Grigoropoulos and H. J. Sung, Nano Lett., 2011, 11, 666–671.
6. Z. Dong, X. Lai, J. E. Halpert, N. Yang, L. Yi, J. Zhai, D. Wang, Z. Tang and L. Jiang, Adv. Mater., 2012, 24, 1046–1049.
7. M. McCune, W. Zhang and Y. Deng, Nano Lett., 2012, 12, 3656–3662.
8. M. Law, L. E. Greene, J. C. Johnson, R. Saykally and P. Yang, Nat. Mater., 2005, 4, 455–459.
9. K. Keis, J. Lindgren, S.-E. Lindquist and A. Hagfeldt, Langmuir, 2000, 16, 4688–4694.
10. Z. Qin, Y. Huang, J. Qi, Q. Liao, W. Wang and Y. Zhang, Mater. Lett., 2011, 65, 3506–3508.
11. H. Horiuchi, R. Katoh, K. Hara, M. Yanagida, S. Murata, H. Arakawa and M. Tachiya, J. Phys. Chem. B, 2003, 107, 2570–2574.
12. T. P. Chou, Q. Zhang and G. Cao, J. Phys. Chem. C, 2007, 111, 18804–18811.
13. M. Quintana, T. Marinado, K. Nonomura, G. Boschloo and A. Hagfeldt, J. Photochem. Photobiol., A, 2009, 202, 159–163.
14. H.-M. Nguyen, R. S. Mane, T. Ganesh, S.-H. Han and N. Kim, J. Phys. Chem. C, 2009, 113, 9206–9209.
15. R. Schölin, M. A. Quintana, E. M. J. Johansson, M. Hahlin, T. Marinado, A. Hagfeldt and H. k. Rensmo, J. Phys. Chem. C, 2011, 115, 19274–19279.
16. Z. Qin, Y. Huang, J. Qi, L. Qu and Y. Zhang, Colloids Surf., A, 2011, 386, 179–184.
17. Z. Qin, Y. Huang, Q. Liao, Z. Zhang, X. Zhang and Y. Zhang, Mater. Lett., 2012, 70, 177–180.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra05644b

---