Fabrication of a dye-sensitized solar cell module using spray pyrolysis deposition of a TiO₂ colloid

Abstract

We provide a complete illustration of dye-sensitized solar module fabrication by spray pyrolysis deposition (SPD) of a TiO₂ colloid having ∼10 nm sized TiO₂ nanoparticles. The process was first optimized for cell level fabrication, and the parameters (mainly the thickness) obtained from the study were subsequently used for module level fabrication. The best efficiency obtained at the cell level (area 0.2 cm² and thickness of 12 μm) was 7.79% and that for the (12 cm × 12 cm) module was 3.2%.

---

Introduction

Solar energy will play an important role in satisfying future energy needs of mankind since coal and natural reserves are being consumed at faster rates.¹ However, the processing and manufacturing costs of widely used Si solar panels are high.^2,3 Further innovations in material science and technology are required to make solar energy affordable to the common man. Dye-sensitized solar cells (DSCs) developed by Grätzel and O'Regan using nanocrystalline semiconductor oxide material sensitized by a ruthenium (Ru) dye enabled a paradigm shift in the field of solar energy conversion technology.⁴ This pioneering work provided a new outlook to solar energy conversion technology by integrating different fields of science. The best efficiency achieved for DSCs has been 12% using a conventional Ru dye and a liquid electrolyte system; however, recently, a record power efficiency of 15% was obtained at the cell level for a perovskite-sensitized solar cell using a cobalt (II/III)-based electrolyte.⁵ DSC solar panels are potential candidates for indoor lighting purposes because they slightly outperform conventional Si solar panels under low sunlight.^6,7 Simple fabrication methods at low temperatures make DSCs more appealing than Si solar cells in solar energy markets.

Doctor-blading is the technique widely used for thin film fabrication on a transparent conductive oxide (TCO) substrate at the R&D level.⁸ However, it has major disadvantages such as (1) doctor-blading is not a scalable method, and (2) lack of reproducibility on the thickness and uniformity of the thin film. In this work, we have adopted a scalable spray pyrolysis deposition (SPD) technique using an automated machine (SPD Laboratory Inc., Japan), where the thickness of the film can be easily and accurately controlled. An optimized TiO₂ film thickness is essential for reproducible efficiency since it controls both the dye loading and the extent of electron transport before collection. Generally in cell fabrication, 10 μm has been adopted for efficient charge collection.⁹ In the present case, we used an optimized thickness of the film by spray pyrolysis deposition at the cell level, and the same thickness was also adopted for module fabrication using spray pyrolysis deposition.

Experimental methods

For DSCs fabrication, one side of the glass substrate (Corning Eagle XG, USA) was made conductive by spray pyrolysis deposition of fluorine and tin oxide precursors using a spray pyrolysis deposition machine (KM-150, SPD Laboratory, Inc. Japan). Fig. 1 shows the important parts of the SPD machine, KM-150. Dibutyltin diacetate (DBTDA) in 2-propanol (0.2 M) and ammonium fluoride (NH₄F) in water (9 M) were mixed and stirred for preparing the precursors of the FTO.¹⁰ By controlling the number of spraying cycles, an optimum thickness of 800 nm was deposited on the glass substrate. The sheet resistance of the FTO plates was ∼8–10 Ω □⁻¹.

Fig. 1 (A) An overview of the spray pyrolysis deposition machine (KM-150). Inset shows the chamber where solution for spraying is loaded. (B) Three nozzles of the machine by which the flow-rate of the solution can be controlled for controlled deposition. (C) A high temperature furnace inside the machine by which glass substrate can be pre-heated for thin film deposition.

The TiO₂ colloidal solution for spraying was prepared as follows: 2.5 mL of acetic acid (NICE Chemicals, India) and 20 mL of titanium isopropoxide (Sigma Aldrich, 99.9%) were added in 25 mL isopropyl alcohol (IPA, Sigma Aldrich, 99.9%). To obtain a TiO₂ colloidal solution, steam was passed through the prepared solution which resulted in the expulsion of IPA and the precipitation of the TiO₂ colloidal mass. Then the TiO₂ colloidal mass was ground in a mortar with 50 mL water and autoclaved at 180 °C for 3 h. The TiO₂ film was deposited on the FTO plates by the spray pyrolysis deposition method using the KM-150 machine. The 2 × 2 cm² TiO₂ films with different thicknesses were fabricated on FTO plates by varying the number of spraying cycles in the KM-150 machine. The thickness of the film was varied between 6 μm to 18 μm for optimizing the desired thickness of the TiO₂ film. The films were sintered at 450 °C for 30 min. After sintering, an active area of 0.2 cm² was carved out from the film for dye deposition. The TiO₂ films were soaked in 3 mM solution of the dye (N719) in acetonitrile and tert-butyl alcohol (1 : 1) for 12 h. The photoanode and counter electrode were sealed together with a UV sealant and filled with iodide/triodide (I⁻/I₃⁻) electrolyte by a vacuum back-filling process. The current density–voltage (J–V) characteristics of DSCs were measured using a Keithley 2400 source meter (Newport Oriel class A-Solar simulator, USA). The internal quantum efficiency of the DSCs was measured as a function of wavelength using an IPCE measurement instrument (Oriel Newport (QE-PV-SI/QE) IPCE Measurement kit, USA). A 230 W xenon (Oriel) lamp was used as the light source for generating monochromatic light. The J–V characteristics of the dye solar module were measured using a large area solar simulator (XES-200S1, SAN-EI Electric, Japan). The intelligent design and precise fabrication technique adopted for the dye solar module are shown in Fig. 2 that ensure the maximum usage of FTO as the active area. The large area 12 × 12 cm² dye solar module was fabricated by employing the optimized TiO₂ film thickness. The KM-150 machine was used for depositing the FTO on the 12 × 12 cm² glass substrate, as documented in Fig. 2. The transparency of the FTO glass substrate is evident from Fig. 3. Subsequently the TiO₂ colloidal solution was sprayed on the pre-heated (at 90 °C) FTO glass substrate. The FTO was partially masked during the spraying of TiO₂ colloidal solution using a metallic mask (Fig. 4). This masking also facilitated the drawing of silver (Ag) grid lines through the masked areas for charge collection. The TiO₂ was deposited on the FTO glass substrate on the unmasked regions. The TiO₂ deposition resulted in rectangular-shaped cells or compartments in parallel. The silver grid was drawn on unmasked regions using an Ag grid drawing robot (SPD Laboratory, Inc. Japan) for proper electron collection to the external circuit. The TiO₂-coated FTO glass was sintered at 530 °C for 1 h. Then, it was mounted in a dye coating bath machine, which provided temperature controlled dye-sensitization of the TiO₂ film. The temperature was maintained at 26 °C. The washings of DSC by acetonitrile and drying were also carried out by the dye coating bath machine. Holes were drilled on the Pt-coated glass substrate by a hole-drilling robot to fill the electrolyte by a vacuum back-filling process. The UV curable sealant was drawn on the Ag grid pattern, which acted both as a sealant and a protective cover for the Ag (since I⁻/I₃⁻ electrolyte is corrosive and can react with the Ag grid lines).

Fig. 2 Schematic showing the different steps of fabrication of the dye-sensitized solar cell module. In step 1: Corning Eagle XG plane glass with ∼100% transparency is heated at 450 °C. Step 2: precursors of FTO are sprayed onto the glass plate to make it a transparent conductive oxide substrate, where the transparency of the glass is almost retained. Step 3: TiO₂ colloidal solution is sprayed at 90 °C on masked FTO (masking is done by a metal grid as shown in Fig. 4) for 12 μm thickness. Step 4: Ag grid was drawn on unmasked areas of FTO and sintered at 530 °C for 1 h. Step 5: TiO₂ film was sensitized with N719 dye in a dye coating bath for 20 h.

Fig. 3 (A) A 12 × 12 cm² Corning glass (bare) substrate. (B) Shows the transparency of the FTO-coated glass.

Fig. 4 The metallic (Ti) mask used to spray-deposit TiO₂ as compartments on the FTO substrate.

The dye-sensitized photoanode and the hole-drilled counter electrode were sandwiched using an electrode pile-up machine (SPD Laboratory Inc., Japan). This was further exposed to UV irradiation for hardening the sealant. Electrolyte was filled through the holes in the counter electrode using the vacuum back-filling process. After electrolyte filling, the holes were sealed by a glass slide using a UV sealant.

The TiO₂ colloid used for spraying was characterized by transmission electron microscopy (TEM). The sample solution was drop-casted on carbon-coated copper grid and dried for TEM analysis. Powder XRD was carried out by X'pert pro PAN Analytical with a data interval of 0.03° at a current and voltage of 30 mA and 40 kV, respectively. XPS (Kratos Analytical, UK) was used for the elemental characterization of the material and for assessing the oxidation state of the elements involved. Carbon correction was done using the standard software from the manufacturer.

Results and discussion

Characterization of the TiO₂ colloidal solution

Fig. 5A shows the powder XRD pattern of the TiO₂ used for spraying, which indicates its anatase nature and crystallinity. Major peaks are indexed in the spectrum itself (JCPDS file no. 21-1272). The particle size estimated using Debye–Scherrer equation [from the full width-half maximum of the (101) peak] was ∼12 nm. This was further supported by the TEM studies (Fig. 6A and C, respectively).

Fig. 5 (A) Powder XRD spectrum and (B) wide XPS spectrum of TiO₂ in the TiO₂ colloidal solution. C&D show the high-resolution XPS spectra of (C) Ti and (D) O, respectively.

Fig. 6 (A) TEM image showing the morphology of TiO₂ in the colloidal solution (B) its SAED pattern showing polycrystallinity and (C) HRTEM image showing the 0.35 nm lattice spacing corresponding to the anatase TiO₂.

The TEM image indicated nearly monodisperse spherical particles with an average diameter of 10 nm. The selected-area electron diffraction (SAED) pattern shown in Fig. 6B further confirms the polycrystalline nature of TiO₂. Fig. 5C shows a lattice-resolved image, showing the most prominent (101) lattice orientation with a spacing of 0.35 nm.

XPS was used to determine the elemental composition and oxidation state of TiO₂. The Fig. 5B shows a wide scan spectrum which gives the elemental composition (Ti and O) of the material. The peaks at the binding energies of 455.71 eV and 461.48 eV correspond to that of Ti 2p_3/2 and Ti 2p_1/2, respectively (Fig. 5C). The O1s showed a single peak at 526.90 eV (Fig. 5D).¹¹

Photovoltaic (PV) performance of DSCs

Since TiO₂ film has lower electron drift mobility (10⁻⁴ to 10⁻⁷ cm V⁻¹ s⁻¹), optimization of film thickness was inevitable in fabricating efficient dye-sensitized solar cell modules. At the cell level, different TiO₂ thicknesses (10 μm, 12 μm, 15 μm, respectively) were fabricated and the PV parameters were compared to obtain the optimum thickness for the best performance. It was found that 12 μm thick TiO₂ film gives the best overall efficiency at the cell level. The J–V performances of cells with different TiO₂ thicknesses are shown in Fig. 7A. An overall efficiency of 7.79% was obtained for the 12 μm thick TiO₂ film. High V_oc (0.75 V) and current density (13.94 mA cm⁻²) make the 12 μm thick film a better performing DSC compared with the other TiO₂ films (Table 1).

Fig. 7 (A) Current density–voltage (J–V) characteristics (B) IPCE (%) of DSCs with different TiO₂ film thicknesses.

Table 1 Variations of V_oc, current density, fill factor, IPCE (%) and efficiency with respect to change in thickness of the film

Thickness of the film	Voc (V)	Current density (mA cm^−2)	Fill factor (%)	Efficiency (%)
10 μm	0.759	11.94	73.88	6.70
12 μm	0.755	13.94	74	7.79
15 μm	0.75	13.86	73.18	7.61
18 μm	0.67	10.70	66.40	4.78

---

As the TiO₂ thickness was increased, the internal resistance of the cell increased considerably, and this contributed to a decrease in FF (Fig. 8A). An IPCE (%) of 70.5% was obtained for the 12 μm thick TiO₂ film (Fig. 7B). This shows the spectral response of the current flow to the external circuit, where the highest current was shown to be at 525 nm, which corresponds to the absorption maximum of the dye. Lower dye loading at thinner films and poor electron collection at thicker films contribute to a lower J_sc above and below the optimum film thickness. The V_oc was almost stable at lower thicknesses, but as thickness is increased, it decreased considerably.

Fig. 8 (A) The plot showing the dependence of current density and fill factor on film thickness. (B) Nyquist plots obtained from EIS showing interfacial recombination resistance of different DSCs with different film thicknesses (for a proper comparison, all the impedance traces have been overlapped under the same series resistance of 25.2 Ohm).

An increase in film thickness increases the number of trapping surface states and enhances the back electron transfer to the I₃⁻, which eventually results in a lowering of V_oc and J_sc. The increase in thickness increases the resistance of the cell, which could reduce the power loss in the system (i.e. lowering of the FF). Electrochemical impedance spectroscopy (EIS) was used for confirming the dependence of film thickness on the interfacial charge recombination.¹² The impedance measurements were carried out by applying a bias voltage equal to the open-circuit voltage (V_oc) at dark. The frequency varied from 10⁵ Hz to 0.005 Hz with an ac amplitude of 12 mV. According to literature, at higher potentials, impedance plots show three arcs where the middle arc corresponds to the recombination resistance at the TiO₂/dye/electrolyte interface.^13–15

The Fig. 8b shows the Nyquist plots of DSCs with the different TiO₂ film thicknesses. The width of the middle arc corresponds to recombination resistance at the interfacial region (TiO₂/dye/electrolyte interface). From the impedance plots, we can observe that the width of the second semicircle is the largest for the 12 μm thick DSC. However, as the thickness is gradually increased, the recombination resistance decreased. The EIS studies reported in the past also suggest that 10–12 μm is the optimum thickness for DSCs with spherical TiO₂ particles.^16–18 Thus, from EIS studies we can infer that a 12 μm thick film offers the minimum interfacial recombination in the device.

The 12 × 12 cm² DSC module (Fig. 9) with a 12 μm thick TiO₂ film has shown an efficiency of 3.28% (Fig. 10). The solar module consisted of 9 parallel rectangular compartments of 111 × 10.3 mm dimensions. It can be connected to an external circuit from the edges of the photoanode and the counter electrode. The open-circuit voltage of 0.674 V was obtained for the module with a current density of 10.71 mA cm⁻². The fill-factor (%) of the module was about 45.4%, which indicates power losses in the system due to series resistance. The resistance at the Pt/electrolyte interface, the TCO substrate resistance and the diffusion resistance of I₃⁻ ions in electrolyte contribute to the series resistance in DSCs.¹⁹ In the scaling up of DSCs, the contact area of TCO with the electrolyte and the amount of effective electrolyte are increased considerably, which contributes to the increase in series resistance of the cell. This would essentially lower the FF and hence the efficiency.²⁰

Fig. 9 A dye-sensitized solar module (12 × 12 cm²) fabricated at the optimized film thickness of 12 μm.

Fig. 10 Current density–voltage (J–V) characteristics of the 12 × 12 cm² module.

Conclusions

We adopted the spray pyrolysis deposition technique as one of the simple and scalable approaches to fabricate thin films of TiO₂ for the fabrication of DSC cells and modules. The TiO₂ thickness for best performing DSCs by spray pyrolysis deposition technique has been optimized at 12 μm. At the cell level (with 0.2 cm² active area and thickness of 12 μm), a power conversion efficiency of about 7.79% has been obtained. Using this optimized thickness, a 12 × 12 cm² DSC module has been created using the precise fabrication methods as outlined. A power conversion efficiency of about 3.28% was obtained for the DSC module.

Acknowledgements

The authors thank the Ministry of New and Renewable Energy (MNRE) and Department of Science and Technology, Govt. of India for financial assistance.

Notes and references

1. (a) K. G. Reddy, T. G. Deepak, G. S. Anjusree, S. Thomas, S. Vadukumpully, K. R. V. Subramanian, S. V. Nair and A. S. Nair, Phys. Chem. Chem. Phys., 2014, 16, 6838; (b) P. V. Kamat, J. Phys. Chem. C, 2007, 111, 2834–2860; (c) L. M. Gonçalves, V. d. Z. Bermudez, H. A. Ribeiro and A. M. Mendes, Energy Environ. Sci., 2008, 1, 655–667.
2. P. D. Moskowitz and V. M. Fthenakis, Solar Cells, 1990, 29, 63.
3. P. D. Moskowitz and V. M. Fthenakis, Solar Cells, 1991, 31, 513.
4. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737.
5. http://www.fujikura.co.jp/eng/rd/gihou/backnumber/pages/__icsFiles/aieldfile/2013/05/23/42e_30.pdf.
6. J. Burschka, N. Pellet, S. J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316.
7. http://www.swissphotonics.net/libraries.files/TobyMeyerSolaronix.pdf.
8. C. J. Brabec and J. R. Durrant, MRS Bull., 2008, 33, 670.
9. M. Gratzel, Inorg. Chem., 2005, 44, 6841.
10. G. R. A. Kumara, S. Kawasaki, P. V. V. Jayaweera, E. V. A. Premalal and S. Kaneko, Thin Solid Films, 2012, 520, 4119.
11. H. C. Choi, Y. M. Jung and S. B. Kim, Vib. Spectrosc., 2005, 37, 33.
12. Q. Wang, J. E. Moser and M. Grätzel, J. Phys. Chem. B, 2005, 109, 1494.
13. J. Bisquert, J. Phys. Chem. B, 2002, 106, 325.
14. M. Adachi, M. Sakamoto, J. Jiu, Y. Ogata and S. Isoda, J. Phys. Chem. B, 2006, 110, 13872.
15. F. Fabregat-Santiago, J. Bisquert, E. Palomares, L. Otero, D. Kuang, S. M. Zakeeruddin and M. Grätzel, J. Phys. Chem. C, 2007, 111, 6550.
16. V. Baglio, M. Girolamo, V. Antonucci and A. S. Aricò, Int. J. Electrochem. Sci., 2011, 6, 3375.
17. C.-Y. Huanga, Y.-C. Hsua, J.-G. Chena, V. Suryanarayanana, K.-M. Leeb and K.-C. Ho, Sol. Energy Mater. Sol. Cells, 2006, 90, 2391.
18. S. Ito, T. N. Murakami, P. Comte, P. Liska, C. Grätzel, M. K. Nazeeruddin and M. Grätzel, Thin Solid Films, 2008, 516, 4613.
19. L. Han, N. Koide, Y. Chiba, A. Islam, R. Komiya, N. Fuke, A. Fukui and R. Yamanaka, Appl. Phys. Lett., 2005, 86, 213501.
20. M. G. Kang, N. G. Park, Y. J. Park, K. S. Ryu and S. H. Chang, Sol. Energy Mater. Sol. Cells, 2003, 75, 475.

---