TiO₂ nanoparticles synthesized by the molten salt method as a dual functional material for dye-sensitized solar cells

Abstract

We report the simple and versatile molten salt fabrication of mesoporous anatase TiO₂ nanoparticles with a high surface area of ∼200 m² g⁻¹, which show an impressive efficiency of ∼7.5% in dye-sensitized solar cells.

---

Titanium dioxide (TiO₂) is one of the most studied metal oxide semiconductors of the last two decades. TiO₂ is used for applications relating to environmental remediation,¹ solar cells,² lithium ion batteries,³ gas sensors,⁴ photonic crystals,⁵ and self-cleaning coatings,⁶ owing to its chemical stability, environmental friendliness, and low cost. The performance of TiO₂ in such applications (particularly for solar cells) depends on its crystallinity, crystal phase, surface area, and morphology.

Dye-sensitized solar cells (DSCs) have received prime consideration in renewable energy research owing to factors that include the simple non-vacuum processing conditions used for their fabrication, their low cost as a consequence of the relatively high natural abundance of the raw materials used in manufacturing, and properties that include being light-weight, having high tolerance to impurities, conversion efficiencies comparable to that of amorphous Si solar cells, and suitability for building-integrated photovoltaics.⁷ High surface area TiO₂ is central to the fabrication of DSCs as it supports the sensitizers and transports the photoexcited electrons from the sensitizer to the transparent conducting oxide. Despite different nanostructures of TiO₂ being studied for use in DSCs, including fibers,⁸ tubes,⁹ rice-like arrangements,¹⁰ beads,¹¹ and other morphologies,¹² traditional nanoparticles have the best performance.¹³ However, most synthetic methods to produce anisotropic TiO₂ nanostructures involve multi-step reactions, long reaction times, and post-annealing treatments, which often introduce defects and impurities into the TiO₂ nanostructures, consequently decreasing the overall device performance. Thus, the search for TiO₂ with enhanced surface area and crystallinity for application in high-performance DSCs has become an active research area.

Herein, we report the fabrication of high surface area anatase TiO₂ nanoparticles for dual functionality (both as an active and scattering layer) in high performance DSCs. We use the molten salt method (MSM) of synthesis due to its simplicity and versatility in the fabrication of high purity oxide materials, as commercially the MSM has proven effective in the synthesis of LiCoO₂,¹⁴ ZnCo₂O₄,¹⁵ and Li(Ni_1/3Co_1/3Mn_1/3)O₂¹⁶ materials for lithium ion battery applications. To the best of our knowledge, this is the first application of MSM produced TiO₂ nanoparticles in the field of dye-sensitized solar cells. We fabricated TiO₂ nanoparticles using the MSM and characterized these using UV-visible spectroscopy, scanning and transmission electron microscopy (SEM and TEM, respectively), X-ray and neutron powder diffraction (XRPD and NPD, respectively), Brunauer–Emmett–Teller (BET) surface area measurements, and their performance as electrodes within DSCs. The dual functional material displays a higher efficiency, 7.5%, than that of commercially available materials (e.g. P25 which shows an efficiency of 6.54% with a scattering layer).

The TiO₂ nanoparticles were synthesized as follows: A 1 : 8.8 : 1.2 molar mixtures of TiOSO₄·xH₂SO₄ : LiNO₃ : LiCl was heated at 3 °C min⁻¹ to 280 °C for 1 h in an Alumina crucible in air. The mixture was cooled at the same rate as the mixture was heated and the excess Li-salts were washed, filtered, and dried under vacuum at 70 °C for 12 h. The as-obtained TiO₂ nanoparticles were mixed with ethyl cellulose and α-terpineol¹⁷ to form a paste, and fabricated into electrodes by screen printing/doctor-blading on clean, TiCl₄-treated conductive fluorine-doped tin oxide (FTO) substrates. The electrodes were then sintered at 450 °C for 30 min, treated with TiCl₄, and sintered again. Dye loading was undertaken by immersing these electrodes in N3 dye for 24 h. The electrodes were then sealed with counter electrodes for performance tests. For comparison, electrodes fabricated from commercial P25 TiO₂ nanoparticles (with and without a scattering layer) were also investigated (see the ESI†-1 for more fabrication details).

The TiO₂ nanoparticles were characterized by SEM and TEM (Fig. 1). Fig. 1A shows a typical SEM image of the MSM produced TiO₂ and reveals the presence of agglomerated nanostructures with diameters in the range 100–300 nm. The sizes of the spherical aggregates determined using TEM (Fig. 1B) were 40–60 nm in diameter and this can be attributed to the sonication-assisted dispersion of the MSM produced TiO₂ in methanol for casting onto the TEM grid. The higher magnification TEM image (Fig. 1C) illustrates that the larger, 40–60 nm diameter, particles are composed of smaller nanoparticles with an average diameter of 5 nm. It is these ∼5 nm nanoparticles that are responsible for the very high surface area of the synthesized TiO₂ (∼200 m² g⁻¹, from BET surface area measurements). The high-resolution TEM image (Fig. 1D) indicates that the nanoparticles in this projection have a lattice spacing of 0.35 nm, corresponding to the (101) crystal face indices of anatase.

Fig. 1 SEM image (a), TEM image in low (b) and high magnification (c), and high resolution TEM image (d) of MSM TiO₂ nanoparticles.

Phase-pure anatase was confirmed by XRPD data (ESI†-2, JCPDS No. 04–0477) and shown to be relatively crystalline. The crystal structure of the MSM produced TiO₂ was further investigated by Rietveld analysis using neutron powder diffraction (NPD, Fig. 2 and ESI†-1).¹⁸ The final Rietveld-refined model had figures of merit that include profile factor (R_p) of 2.01%, the weighted-profile factor, wR_p, of 2.59%, and the goodness-of-fit term, χ², of 1.53 for 16 refinable parameters.

Fig. 2 The Rietveld-refined fit for anatase TiO₂ using NPD data. The red crosses are the data, the black line through the crosses is the calculated model, the purple line below is the difference between the data and calculated model, and the vertical black lines are reflection markers for anatase TiO₂.

The lattice parameters obtained from the NPD data were a = 3.7823(9) Å and c = 9.475(4) Å in the tetragonal I4₁/amd space-group and these values agree with our XRPD studies (ESI†-2) and literature reports.¹⁹ The O atom positional parameter was refined to z = 0.1639(5) and the atomic displacement parameters of Ti and O atoms were found to be 0.0144(26) and 0.0142(12) Ų, respectively. No statistically significant evidence was found for long-range ordered O atom vacancies.

Our synthesis of pure anatase TiO₂ with high surface area prompted the investigation of the material's performance as an electrode in DSCs, for comparison with the photovoltaic (PV) parameters of commercially available P25 TiO₂. The photocurrent density (J_sc) vs. voltage characteristics of the DSCs are shown in Fig. 3A (see ESI†-3 for details on the PV parameters based on the test results on 5 devices each from MSM and P25 TiO₂ nanoparticles). The working area and electrode thickness of both the electrodes were 0.25 cm² and 13 μm, respectively (see ESI†-4 for thickness-dependent PV parameters). The PV parameters were an open-circuit voltage (OCV) of 0.765 V for MSM TiO₂ compared with 0.75 V for P25, a fill factor of 67.5% for MSM TiO₂ compared with 65% for P25, and a J_sc of 14.48 mA cm⁻² for MSM TiO₂ compared with 12.36 mA cm⁻² for P25. The consistently higher PV parameters of MSM TiO₂ result in a higher efficiency (η) of 7.48% for MSM TiO₂ compared with 6.03% for P25 under the same conditions.

Fig. 3 Photocurrent–photovoltage characteristic curves (A), incident photon conversion efficiency (IPCE) spectra (B), Nyquist plots (C), and Bode phase plots (D) of MSM produced and P25 TiO₂.

Our MSM produced TiO₂ enables both a higher J_sc²⁰ and OCV. This is in part because anatase TiO₂ has a larger band-gap than that of rutile TiO₂²¹ and phase pure anatase TiO₂ eliminates the recombination that can occur at grain boundaries of different TiO₂ crystalline forms (P25 is a mixture of ∼75% anatase and ∼25% rutile TiO₂). The higher J_sc of the MSM produced TiO₂, relative to P25 TiO₂, is enabled by its relatively high surface area (200 m² g⁻¹ in comparison to 55 m² g⁻¹ for P25 TiO₂) that allows a higher dye-loading, 2.57 × 10⁻⁷ mol cm⁻² for MSM TiO₂ compared with 1.28 × 10⁻⁷ mol cm⁻² for P25 (see ESI†-5 for the UV-Visible spectra of the desorbed dye solutions), in turn enhancing the light harvesting efficiency (η) of devices that contain this material.²² Further evidence of the effect of dye-loading is provided by the incident photon-to-current conversion efficiency (IPCE) spectra from the devices (Fig. 3B). The peak in the IPCE spectrum from MSM produced TiO₂ is 70%, which is 13% higher than that from P25 (62%). It can also be observed that the IPCE of the MSM TiO₂ is higher that of the P25 TiO₂ in all the wavelengths up to ∼750 nm. The enhanced IPCE in the small wavelength region (<600 nm) is due to the high dye-loading (indicating an efficient light harvesting efficiency of the device, see the dye-loading data above) and that in the long wavelength region (>600 nm) is due to the strong light scattering effect as a consequence of the presence of clustering-type aggregates of ∼100–500 nm diameters (see SEM images in ESI†-6) in the MSM TiO₂. These aggregates can function as secondary scattering centers that reflect photons back into the electrode and improve the light harvesting efficiency of the electrodes.^23,24 This is further evident from the plot of IPCE_(MSM TiO2)/IPCE_(P25)vs. wavelength given in ESI†-7. The enhanced light scattering effect is further confirmed from a comparison of the UV-visible spectra of the MSM produced- and P25 TiO₂ electrodes (ESI†-8). To get further insights on the better performance of MSM TiO₂, electrochemical impedance spectra (EIS) were also measured (under 1 Sun illumination at an applied bias of OCV). From the Nyquist plots presented in Fig. 3C, it is evident that the second semi-circle related to the charge transport resistance of TiO₂/dye/electrolyte interface was lower in the case of the MSM TiO₂ electrode than that of the P25 electrode. The characteristic peak of the Bode phase plots (Fig. 3D) shifted to lower frequency in the case of MSM TiO₂, indicating suppressed charge transport resistance and higher electron lifetime in the MSM TiO₂.^23,25 These come as a direct consequence of the phase purity (pure anatase with fewer grain boundaries) and the good connectivity between the particles by the clustering-type aggregations in MSM TiO₂.

Recently, scientists have been focussing on the fabrication of mesoporous microspheres with large diameters ∼500 nm with high surface area for high performance DSCs.^26,27. The intention is to avoid the use of double layer configurations (high-performance DSCs usually have a double layer structure consisting of an active and a scattering layer), as they are complex and cost-intensive. Furthermore, the packing densities of these nanostructures are known to be relatively low²⁸ owing to their large diameters, which in turn decreases the total mass of active TiO₂ and the surface area of the film, lowering the overall photocurrent in a device.^22,28 Our approach uses the MSM to produce a dual functional electrode. We compare the functionality of an electrode of the MSM produced TiO₂ in a single-layer DSC to a double-layer electrode DSC constructed using a 2 μm thick scattering layer of 200 nm diameter TiO₂ nanoparticles on the surface of the P25 TiO₂ electrode. The double layer electrode DSC shows an efficiency of 6.54% (ESI†-9). We note that the efficiency of the MSM produced TiO₂ DSC is higher (7.48%) than that of the double-layer DSC. Thus, we postulate that TiO₂ nanostructures with aggregated particles perform dual functions, that of an active and a scattering layer, to produce high-performance DSCs.¹¹ Such materials also facilitate the fabrication of DSCs by allowing a simpler method to be used that avoids the need for additional scattering layers (normally comprising ∼400 nm diameter particles of TiO₂/ZrO₂ to achieve high-performance) and the double layer configuration.

In summary, we have fabricated a high surface area, porous, and nanostructured anatase TiO₂ by the MSM. The MSM produced TiO₂ nanoparticles show better PV performance than that of commercially available P25 TiO₂ in DSC applications. The high performance is evident from J_scvs. V and IPCE traces, and the EIS data. We believe that the high-performance MSM TiO₂ will have wide applications in other fields such as photocatalysis and self-cleaning materials. The simple and versatile MSM can also be extended for the fabrication of other metal oxides, such as ZnO and SnO₂, with wide ranging applications.

Acknowledgements

We thank the Clean Energy Program Office (CEPO) of National Research Foundation (NRF, Grant No. NRF 2007 EWT-CERP 01-0531) and M3TC (Economic Development Board, grant No. R-261-501-018-414), Singapore for financial assistance.

References

1. S. Horikoshi and H. Hidaka, Environ. Sci. Technol., 2002, 36, 1357.
2. B. O'regan and M. Grätzel, Nature, 1991, 353, 737.
3. P. Zhu, Y. Wu, M. V. Reddy, A. S. Nair, B. V. R. Chowdari and S. Ramakrishna, RSC Adv., 2012, 2, 531.
4. A. M. Ruiz, G. Sakai, A. Cornet, K. Shimanoe, J. R. Morante and N. Yamazoe, Sens. Actuators, B, 2003, 93, 509.
5. S. Kanehira, S. Kirihara and Y. Miyamoto, J. Am. Ceram. Soc., 2005, 88, 1461.
6. A. Bozzi, T. Yuranova and J. Kiwi, J. Photochem. Photobiol., A, 2005, 172, 27.
7. L. M. GonÇalves, V. de Zea Bermudez, H. A. Ribeiro and A. M. Mendes, Energy Environ. Sci., 2008, 1, 655.
8. P. Zhu, A. S. Nair, S. Yang, S. Peng and S. Ramakrishna, J. Mater. Chem., 2011, 21, 12210.
9. G. K. Mor, K. Shankar, M. Paulose, O. K. Varghese and C. A. Grimes, Nano Lett., 2006, 6, 215.
10. (a) A. S. Nair, Y. Shengyuan, Z. Peining and S. Ramakrishna, Chem. Commun., 2010, 46, 7421; (b) Y. Shengyuan, Z. Peining, A. S. Nair and S. Ramakrishna, J. Mater. Chem., 2011, 21, 6541.
11. F. Sauvage, D. Chen, P. Comte, F. Huang, L. P. Heiniger, Y. B. Cheng, R. A. Caruso and M. Grätzel, ACS Nano, 2010, 4, 4420.
12. W. J. Zhou, H. Liu, R. I. Boughton, G. J. Du, J. J. Lin, J. Y. Wang and D. Liu, J. Mater. Chem., 2010, 20, 5993.
13. C. Y. Chen, M. Wang, J. Y. Li, N. Pootrakulchote, L. Alibabaei, C. H. Ngoc-Le, J. D. Decoppet, J. H. Tsai, C. Grätzel and C. G. Wu, ACS Nano, 2009, 3, 3103.
14. K. S. Tan, M. V. Reddy, G. V. S. Rao and B. V. R. Chowdari, J. Power Sources, 2005, 147, 241.
15. M. V. Reddy, K. Kenrick, T. Y. Wei, G. Y. Chong, G. H. Leong and B. V. R. Chowdari, J. Electrochem. Soc., 2011, 158, A1423.
16. M. V. Reddy, G. V. S. Rao and B. V. R. Chowdari, J. Power Sources, 2006, 159, 263.
17. S. Peng, Y. Wu, P. Zhu, V. Thavasi, S. Ramakrishna and S. G. Mhaisalkar, J. Mater. Chem., 2011, 21, 15718.
18. A. J. Studer, M. E. Hagen and T. J. Noakes, Phys. B, 2006, 385–386, 1013.
19. M. V. Reddy, R. Jose, T. H. Teng, B. V. R. Chowdari and S. Ramakrishna, Electrochim. Acta, 2010, 55, 3109.
20. A. Wahl and J. Augustynski, J. Phys. Chem. B, 1998, 102, 7820.
21. C. Di Valentin, G. Pacchioni and A. Selloni, Phys. Rev. B: Condens. Matter Mater. Phys., 2004, 70, 085116.
22. C. J. Barbe, F. Arendse, P. Comte, M. Jirousek, F. Lenzmann, V. Shklover and M. Grätzel, J. Am. Ceram. Soc., 1997, 80, 3157.
23. K. Parmar, E. Ramasamy, J. Lee and J. S. Lee, Chem. Commun., 2011, 47, 8572.
24. S. Hore, P. Nitz, C. Vetter, C. Prahl, M. Niggemann and R. Kern, Chem. Commun., 2005, 2011.
25. C. P. Hsu, K. M. Lee, J. T. W. Huang, C. Y. Lin, C. H. Lee, L. P. Wang, S. Y. Tsai and K. C. Ho, Electrochim. Acta, 2008, 53, 7514.
26. D. Chen, F. Huang, Y. B. Cheng and R. A. Caruso, Adv. Mater., 2009, 21, 2206.
27. H. J. Koo, Y. J. Kim, Y. H. Lee, W. I. Lee, K. Kim and N. G. Park, Adv. Mater., 2008, 20, 195.
28. H.-E. Wang, L.-X. Zheng, C.-P. Liu, Y.-K. Liu, C.-Y. Luan, H. Cheng, Y. Y. Li, L. Martinu, J. A. Zapien and I. Bello, J. Phys. Chem. C, 2011, 115, 10419.

---

Footnote

† Electronic Supplementary Information (ESI) available: Details of the synthesis and characterization of the MSM TiO₂ nanoparticles, powder XRD data, SEM and UV-Vis comparison of MSM and P25 TiO₂ electrodes and current density–voltage trace of P25 DSC with a scattering layer. See DOI: 10.1039/c2ra00041e/

---