Facile formation of a hierarchical TiO₂–SnO₂ nanocomposite architecture for efficient dye-sensitized solar cells

Abstract

In this paper, a hierarchical nanocomposite architecture consisting of TiO₂ nanoflakes on SnO₂ nanofibers was successfully synthesized using a simple and cost-effective hydrothermal method. The correlation between the structural changes in nanoflake size and density in accordance with the prolongation of hydrothermal growth duration was observed and discussed using high-resolution scanning and transmission electron microscopies. The dye-sensitized solar cells (DSSCs) utilizing these double-layered TiO₂–SnO₂ hierarchical nanostructures as photoelectrodes exhibited significant improvement in their photovoltaic properties. In particular, the DSSCs made from 24 h hydrothermally grown TiO₂ nanoflakes on the surface of SnO₂ nanofibers stood out to be the best performing cells, delivered a promising efficiency of 3.73%, which was more than a 4-fold increment over the DSSC with SnO₂ nanofibers alone. This enhancement was attributed to the superiority in dye-loading and light scattering properties of the hierarchical architecture. Benefiting from the constituent advantages of both materials, the photoelectrodes with hierarchical nanostructures also inhibited back electron transfer to reduce electron recombination, leading to improved electron lifetime. These results suggest that a simple strategy to synthesize hierarchical nanostructures for efficient DSSCs was achieved.

---

1. Introduction

Since the introduction of low-cost, high-efficiency dye-sensitized solar cells (DSSCs) in 1991 by O'Regan and Grätzel, they have emerged as a potential alternative to conventional thin film solar cells.¹ A typical DSSC consists of a few micrometer-thick nanocrystalline oxide covered with a monolayer of Ru dye, a redox electrolyte and a platinum counter electrode. So far, the conventional DSSCs using ruthenium- and porphyrin-based sensitizer have achieved respectable efficiencies of ∼11.0% and 13.0%, respectively.^2–4 Recent breakthrough in the development of DSSC has been demonstrated using organic halide perovskite (e.g. CH₃NH₃PbX₃, X = Br, I, Cl or their combinations) as photosensitizer whereby the power conversion efficiency of over 15.0% was achieved.⁵ Over the past two decades, significant research efforts have been devoted on the various components of DSSC to further improve the energy conversion efficiency. While metal complexes and organic dyes are widely developed, natural and vegetable dyes as sensitizer for DSSC are also gaining more attentions due to their “environmental friendly” characteristics while producing promising photovoltaic efficiency ∼2.7%.^6–8 Extensive research activities have been carried out as well to replace the liquid electrolytes in DSSC into solid-state or quasi solid-state electrolytes in order to enhance the stability and efficiency of DSSC.^9–11 New counter electrode materials utilizing carbon-based nanomaterials have also been investigated and applied in DSSC as alternative to Pt-coated FTO glasses to improve the charge transfer properties.^4,12,13 The choice and the architecture of wide band gap oxide semiconductor is also an interesting research topic that has been extensively studied to improve the conversion efficiency of DSSC.^14,15 The mesoporous semiconductor oxide layer deposited on the transparent conducting oxide (TCO) glass substrate as the photoelectrode, is an essential constituent structure of the DSSC in separating and transferring the photo-generated electrons in the dye sensitizer towards the collection electrode. Currently, TiO₂ is the favored metal oxide for DSSCs. Several alternative oxide semiconductor materials such as ZnO,^16,17 SnO₂,^18,19 SrTiO₃,²⁰ Nb₂O₅ ²¹ and BaSnO₃ ²² have been investigated for their potential in substituting the TiO₂ as photoelectrode. Among them, SnO₂ stands out to be a potential candidate for DSSCs.

One of the key advantages of SnO₂ over TiO₂ is that SnO₂ offers higher electron mobility (100 to 200 cm² V⁻¹ s⁻¹) which can be translated into faster transport of photo-injected electrons for collection.²³ Other than that, SnO₂ also possesses wider band gap (3.6 eV) than anatase TiO₂ (3.2 eV), which creates less oxidative holes in the valence band under UV illumination to minimize the dye degradation rate and improve the long-term stability of DSSCs.²⁴ However, the use of SnO₂ is also limited by a few drawbacks. The inferior performance of the DSSCs with SnO₂ photoelectrode has been attributed to higher electron recombination rate due to the 300 mV positive shift of the conduction-band edge of SnO₂ with respect to TiO₂.²⁵ Other than that, the lower isoelectric point also leads to poorer dye loading.²⁶ The rate of electron injection from the sensitizer to the conduction band of SnO₂ was also found to be also much slower as compared to TiO₂.²⁷ As proposed by Goodenough and co-workers, the dehydrative coupling of carboxylic acid groups with surface TiO₂ formed ester-type linkages.²⁸ The π* orbitals of the dcb ligand would promote rapid excited-state electron injection into the conduction band of TiO₂, which mainly composed of unfilled d orbitals. On the other hand, the SnO₂ possesses predominantly s-orbital character. The Sn s orbitals are orthogonal to the carboxylate π system resulting in less favorable orbital mixing.²⁹ Hence, the electron injection rates into conduction band of SnO₂ are not as efficient as TiO₂. These issues have been addressed by incorporating a core–shell structure with interfacial layers of wide band gap oxides, such as TiO₂,^30–32 MgO^33,34 or Al₂O₃ ³⁵ in between the SnO₂ and dye sensitizer. By using this core–shell system, a better charge separation has been achieved by suppressing charge recombination in the coupled oxide semiconductors system with different energy levels.

The development of one-dimensional (1D) metal–oxide architectures have also been explored for DSSCs to overcome the poor interconnectivity in between the nanoparticles that resulted in high charge recombination losses.^36,37 These ordered nanomaterials have contributed to faster electron transport and slower charge recombination in DSSCs. One simple and versatile technique for producing continuous ultrafine nanofibers in nanometer scale diameter is through electrospinning.^38–40 Electrospinning of 1D polycrystalline SnO₂ nanomaterials with extremely high aspect ratios and specific surface areas have been reported and investigated in many functional applications such as solar cells,⁴¹ photocatalysis,^42,43 batteries⁴⁴ and gas sensor.^45,46

In this work, we propose a simple approach to revamp 1D SnO₂ nanofibers to three-dimensional (3D) hierarchical TiO₂–SnO₂ nano-architecture. This redefined core–shell nanostructure was achieved by hydrothermally growth of TiO₂ nanoflakes for different durations on the SnO₂ nanofibers. Then, the resulting TiO₂–SnO₂ hierarchical heterostructures were characterized and fabricated into the photoelectrode for DSSCs. To study the effects of these double-layered TiO₂–SnO₂ hierarchical heterostructures as photoelectrode, the photovoltaic properties of these DSSCs were investigated in details.

2. Experimental

2.1 Electrospinning of SnO₂ nanofibers

A solvent containing 22 g of ethanol and 22 g of N,N-dimethyl formamide (DMF) were mixed together using magnetic stirring for 1 h at room temperature. Then, 2.67 g of tin(II) chloride dihydrate (SnCl₂·2H₂O) was added into the mixed solvent under stirring condition until it was completely dissolved. Subsequently, 4 g of polyvinyl pyrrolidone (PVP, M_w = 1 300 000 g mol⁻¹) was added into the resulting solution under vigorous stirring for 2 h at room temperature to form a homogenized viscous precursor. The as-prepared precursor solution was then transferred into a plastic syringe for electrospinning. The electrospinning solution was fed at a speed of 1.5 ml h⁻¹. The voltage applied to the tip was 20 kV and the distance between the needle tip and the collector plate was 10 cm. During the electrospinning process, the solution on the tip of the needle was ejected under strong electric field to form a dense fiber membrane on a collector of aluminium foil. Then, the fiber membrane was manually peeled off from the aluminium foil and calcinated at 500 °C for 2 h with a heating rate of 1.0 °C min⁻¹ to remove the PVP content. Finally, the electrospun SnO₂ nanofibers were obtained.

2.2 Preparation of SnO₂ nanofiber photoelectrodes

The electrospun SnO₂ nanofibers were firstly dispersed in ethanol using stirring and ultrasonication to form a homogenized colloidal. Next, this homogenized colloidal was poured into the beaker containing 10 wt% ethyl cellulose (Aldrich) in ethanol prepared earlier. Finally, alpha-terpineol (97%, Aldrich) was added into the colloidal and stirred. The colloidal was then heated up to 80 °C under stirring on a hotplate to evaporate the ethanol content. Finally, a thick homogenized paste was formed and it was deposited on the FTO glass using doctor blade method. The as prepared photoelectrodes made of SnO₂ nanofibers were subjected to annealing at 450 °C in air for 30 minutes.

2.3 Synthesis of TiO₂–SnO₂ hierarchical nanostructures

The TiO₂–SnO₂ hierarchical nanostructures were grown using hydrothermal method. First, 0.1 ml of titanium isopropoxide (TTIP, 97%, Aldrich) was added to 50 ml of a 0.45 M hydrochloric acid solution under vigorous stirring at room temperature for 2 h until a clear solution was obtained. Then, the solution was transferred to a Teflon vessel. The SnO₂ nanofibers and photoelectrodes prepared earlier were immersed into solution in the Teflon vessel. Then, the vessel was placed into the stainless steel autoclave for hydrothermal growth at temperature 95 °C for 3, 7, 15 and 24 hours respectively. After the hydrothermal process, the nanofibers and photoelectrodes were taken out and rinsed thoroughly with DI water. The photoelectrodes with mesoporous TiO₂–SnO₂ hierarchical nanostructures from different growth durations were ready for use.

2.4 Fabrication of DSSC

The photoelectrodes with mesoporous TiO₂–SnO₂ hierarchical nanostructures were soaked in 40 mM TiCl₄ (99%, Merck) solution at 70 °C for 30 minutes, followed by annealing at 450 °C for 30 minutes in air. Next, the freshly-annealed photoelectrodes were dipped into the 0.25 mM N719 dye solution (Solaronix) for overnight. The platinum coated counter electrode was then clamped onto the photoelectrode, with the Surlyn spacer (30 μm) in between. Finally, the iodide electrolyte (Solaronix, Iodolyte AN50) was filled through capillary action in vacuum condition to allow the complete infiltration of electrolyte into the mesoporous layer. The DSSCs prepared were ready for photovoltaic testing. A total of five different types of photoelectrodes made from pristine SnO₂ nanofibers (sample SS) and composites TiO₂–SnO₂ hierarchical nanostructures grown at different hydrothermal durations (3 h for sample TS03, 7 h for sample TS07, 15 h for sample TS15 and 24 h for sample TS24) were prepared for investigation.

2.5 Characterizations

The crystallography of the electrospun SnO₂ fiber and hierarchical nanostructures TiO₂ on SnO₂ nanofibers were studied using X-ray diffraction (XRD) measurements. The XRD patterns were recorded at room temperature using Pan Analytical X'Pert Pro MPD X-ray diffractometer with Cu K-alpha radiation (λ = 1.5406 Å) operating at 40 kV and 40 mA. The surface morphologies of the nanofibers were observed using field emission scanning electron microscope (FESEM, JEOL-JSM-6701F) and Field Emission Transmission Electron Microscopy (TEM, JEOL-JEM 2010F). The PerkinElmer Lambda 950 UV/Vis/NIR spectrophotometer system was used to determine the amount of dye adsorbed onto the photoelectrode. The dye desorption was carried out in a mixed solution of water and ethanol containing 0.1 M NaOH. The current–voltage (I–V) curves of the solar cells were measured under simulated solar light irradiation with intensity of 100 mW cm⁻². Electrochemical Impedance measurements were carried out using Metrohm Autolab system in the frequency range of 1 mHz to 1000 kHz at room temperature, illuminated under solar light with intensity of 100 mW cm⁻². The applied bias was set to be the open circuit potential of the DSSC and the ac amplitude was 10 mV.

3. Results and discussion

3.1 Material characterizations

The phase purity and crystal structure of the as-electrospun SnO₂ nanofibers and the hydrothermal growth of TiO₂–SnO₂ hierarchical heterostructures for different durations were determined by XRD, as shown in Fig. 1. After the annealing at 500 °C for 2 h, all the peaks in the XRD patterns for SnO₂ nanofibers were indexed accordingly for the cassiterite phase (space group P4₂/mnm) of SnO₂ (JCPDS file 41-1445). No impurity or secondary phase could be detected in the XRD pattern of SnO₂ nanofibers. The as-synthesized SnO₂ nanofibers were polycrystalline and the crystalline particle size was estimated by using the Scherrer equation, D = 0.9λ/β(cos θ), where λ is the wavelength of the X-ray (1.54 Å), β is the full-width at half-maximum (FWHM) in radians, and θ is the Bragg angle in the diffraction pattern. Using this equation, the mean crystalline particle size was estimated to be 13.9 nm.

Fig. 1 XRD patterns of SnO₂ nanofibers and TiO₂–SnO₂ hierarchical heterostructures hydrothermally grown at different durations (3, 7, 15 and 24 h).

Compared with the pristine SnO₂ nanofibers, additional XRD characteristic peaks of rutile TiO₂ could be observed for the case of composite TiO₂–SnO₂ hierarchical heterostructures. The co-existence of the tetragonal SnO₂ and TiO₂ phases has confirmed that the formation of bilayered hierarchical structure with TiO₂ nanocrystallites grown on the surface of SnO₂ nanofibers through the facile single-step hydrothermal process. There was no other diffraction peak indicating the formation of any new composite. With the prolonged growth duration from 3 h to 24 h, the overall peak intensity of the TiO₂ became more prominent, resulted in the peak intensity of SnO₂ to be partially screened off. This suggests that the presence and growing amount of TiO₂ nanostructures formed on top of the surfaces of SnO₂ nanofibers. Increasing the hydrothermal growth duration has also contributed to larger average sizes of TiO₂ nanocrystallites sizes as calculated based on the Scherrer equation (ESI S1†).

The morphologies of the electrospun SnO₂ nanofibers and the corresponding TiO₂–SnO₂ hierarchical heterostructures were investigated by FESEM. Fig. 2a presents the FESEM images of electrospun SnO₂ nanofibers after annealed at 500 °C in air for 2 h. The SnO₂ nanofibers were randomly oriented with rough porous structure after the PVP was completely decomposed (ESI S2†). The size of SnO₂ nanofibers was around 150–200 nm in diameter and lengths could reach up to a few tens of micrometers. Observation at higher magnification as shown in Fig. 2a(ii) reveals that the nanofibers were formed by SnO₂ nanoparticles that were densely packed together to create a long and compact fibrous structure. The rough surface of SnO₂ nanofibers is beneficial for high amount of adsorption of dye molecules.

Fig. 2 SEM images of TiO₂–SnO₂ hierarchical heterostructures hydrothermally grown at different durations. (a-i/ii) Pristine SnO₂ nanofibers at different magnifications; (b) 3 h; (c) 7 h; (d) 15 h; (e) 24 h.

The FESEM images in Fig. 2b–e reveal that the time dependent study (3, 7, 15, 24 h) of hierarchical nano-architecture formed by TiO₂ nanoflakes hydrothermally grown on the SnO₂ nanofibers. The length of these hierarchical nano-architectures maintained at several tens of micrometers (ESI S3†). This shows that the incorporation of TiO₂ nanoflakes using hydrothermal method did not destroy the tubular nanofibers structure. In this case, the TiO₂ nanoflakes were evenly distributed to form a uniform coating along the length of nanofibers. Hence, the SnO₂ nanofibers can be regarded as the basic matrix for the TiO₂ nanoflakes to grow and form the composite hierarchical nanostructures. In the initial 3 and 7 h growth as shown in Fig. 2b and c, tiny TiO₂ nanoflakes nucleated on the surfaces of SnO₂ nanofibers. Further prolongation of the hydrothermal growth to 15 h (Fig. 2d) resulted in the enhancement of filling rate between the adjacent SnO₂ nanoparticles and the TiO₂ nanoflakes to form a densely populated nanoflakes network. Finally, when the hydrothermal growth duration was extended to 24 h, numerous TiO₂ nanoflakes started to evolve into this prominent hierarchical nano-architecture as revealed in Fig. 2e. The average diameter of these nanofibers has also increased noticeably. In short, a hierarchical heterostructures composed of large amount of TiO₂ nanoflakes arrays on SnO₂ nanofibers were successfully achieved using this facile one-step hydrothermal process.

The morphology and crystallinity of SnO₂ nanofibers and TiO₂ nanoflakes were further investigated in details using TEM. Fig. 3a depicts a TEM of an individual SnO₂ nanofiber with the tube-shaped structure with hollow core and a wall thickness of about 20 nm. The nanotubes consisted of bonded nanoparticles with an average size of around 20 nm. Many nanopores were observed within the spaces between adjacent SnO₂ nanoparticles which could be due to the decomposition of PVP. The high resolution TEM (HRTEM) image in Fig. 3a(iii) shows well resolved lattice fringes, signifying that the SnO₂ nanoparticle in the nanotube possessed excellent crystallinity. The lattice fringes with the d spacing of 0.264 nm corresponds to the (101) plane of tetragonal SnO₂. The ring-shaped selected area electron diffraction (SAED) pattern (inset of Fig. 3a(iii)) indicates that the SnO₂ nanofibers were polycrystalline, in good accordance with the XRD data described above.

Fig. 3 (a) TEM and HRTEM images of SnO₂ nanofibers. The inset in (a-iii) is the SAED patterns reveal the crystallinity of SnO₂. Selected TiO₂–SnO₂ hierarchical heterostructures hydrothermally grown at different durations; (b) 7 h; (c) 24 h.

Fig. 3b and c further illustrates the hierarchical TiO₂ nano-architecture arrays constituted by an assembly of multi-scale nanoflakes based on 7 and 24 h of growth duration. The TEM observations coincide well with the FESEM images shown earlier. These sequential images reveal a morphological and structural transformation from pristine SnO₂ nanofibers to TiO₂–SnO₂ hierarchical nanostructures. With SnO₂ nanofibers as the building block, the TiO₂ nanoflakes started to germinate and form the tiny branch-like nanostructures surrounding the entire nanofibers after 7 h of growth duration. With increased hydrothermal growth duration, higher surface coverage density was achieved and the structure of the nanoflakes becomes more prominent. Extending the hydrothermal growth duration to 24 h resulted in prominent TiO₂ nanoflakes that were compactly developed on the surfaces of SnO₂ nanofibers. The length of the nanoflakes could reach up to around 50 nm (Fig. 3c(ii)). The clear lattice fringes of the TiO₂ nanoflakes were observed from the HRTEM images. The interplanar spacing obtained from the lattice fringes along d (110) was 0.325 nm, which could be assigned to the rutile TiO₂ (110). The lattice constant spacing with 0.250 nm pertained to the d-spacing of rutile TiO₂ (101) crystal planes. These results confirmed that the hierarchical heterostructures with TiO₂ nanoflakes on SnO₂ nanofibers were well created.

The successfully growth of TiO₂ nanoflakes onto the surface of SnO₂ nanofibers can be explained by the recrystallization process which involved the dissolve and grow method.^32,33 During the hydrothermal growth process, The Ti species from TTIP precursor reacts with concentrated acidic solution to form TiCl₃. The Ti³⁺ species formed are not stable. Hence, it will immediately react with water to form TiOH²⁺. Finally, the TiOH²⁺ is oxidized into Ti(IV) from the reaction with dissolved oxygen. These Ti(IV) complex ions act as the growth units to form nanostructures. The overall growth process can be described as follow:

2Ti + 6HCl → 2TiCl₃ + H₂O (1)

Ti³⁺ + H₂O → TiOH²⁺ + H⁺ (2)

TiOH²⁺ + O₂ → Ti(IV)-oxo species + O₂⁻ → TiO₂ (3)

3.2 Photovoltaic performance

The hierarchical TiO₂–SnO₂ nanostructures were fabricated into the photoelectrode for DSSCs. A total of five DSSCs based on SnO₂ nanofibers and TiO₂–SnO₂ hierarchical nanostructures as photoelectrode were fabricated for study. Fig. 4 depicts photovoltaic performance of DSSC with hydrothermal growth of TiO₂ nanoflakes/SnO₂ nanofibers. The corresponding photovoltaic parameters such as short-circuit current density (J_SC), open-circuit voltage (V_OC), fill factor (FF), and conversion efficiency (η) derived from the photocurrent density–voltage (J–V) curves were summarized in Table 1.

Fig. 4 (a) Photovoltaic J–V curves for DSSCs with TiO₂–SnO₂ hierarchical heterostructures grown at different durations. (b) Dark current measurements.

Table 1 Photocurrent density–voltage (J–V) characteristic of DSSC with photoelectrode made from hierarchical TiO₂–SnO₂ nanostructures hydrothermally grown at different durations

Samples	VOC (V)	JSC (mA cm^−2)	FF (%)	η (%)
SS	0.38	6.50	34.0	0.84
TS03	0.51	8.35	40.1	1.70
TS07	0.59	8.30	42.6	2.08
TS15	0.60	11.65	46.9	3.28
TS24	0.61	12.68	48.5	3.73

---

The DSSC with pristine SnO₂ nanofibers as photoelectrode (sample SS) showed very poor photovoltaic performance, with the overall power conversion efficiency of less than 1.0%. As mentioned previously, this result can be attributed to the lower electron trapping density of SnO₂, which has been reported to be the key in retarding interfacial charge recombination dynamics in DSSC.¹³ Consequently, the high interfacial charge recombination resulted in low V_OC by decreasing the concentration of electrons in the conduction band of the semiconductor. With reduced forward injection current, the J_SC value was also limited.^47,48

By utilizing the TiO₂–SnO₂ hierarchical nanostructures as photoelectrode, the conversion efficiencies of the DSSCs improved drastically. The best cell (sample TS24) delivered a conversion efficiency of 3.73%, demonstrated more than 4-fold improvement over the DSSC with pristine SnO₂ nanofibers as photoelectrode. Noticeable increments in the photovoltaic parameters were also observed. The improvement in the V_OC value can be explained from the formation of this double-layered TiO₂–SnO₂ structure that creates a surface dipole layer that is dependent on the isoelectric point of the materials involved.^49,50 The isoelectric point of TiO₂ is higher than SnO₂ and hence it yields a surface dipole layer towards the SnO₂. This dipole layer might shift the conduction band of SnO₂ to a more negative value, causing an increase in V_OC. The prominent increases in J_SC and FF also indicate that the interfacial charge recombination on SnO₂ electrodes were effectively suppressed by applying the TiO₂ nanoflakes on the SnO₂ nanofibers. The dark current measurement as shown in Fig. 4b was also carried out to investigate effectiveness of TiO₂–SnO₂ system for achieving high efficiency DSSC. The dark current can be used to represent the charge recombination between the electrons in photoelectrode and I₃⁻ ions in electrolyte.⁵¹ Owing to lower conduction band energy level for SnO₂ than TiO₂, the SnO₂ establishes an expressway between TiO₂ and FTO glass to promote the transport of the photogenerated electrons. At the same time, the TiO₂ also acts as a barrier to effectively reduce the recombination of photogenerated electrons in anode and I₃⁻ ions in. As such, the dark currents for the TS-based solar cells were generally lower than the sample SS, which in turn improved the overall photovoltaic performance.

The sequential improvement of photovoltaic performance according to the growth duration of TiO₂ nanoflakes is explained as follows. In the first 3 and 7 h of growth, some back electron transfer from uncovered areas of SnO₂ nanofibers still existed. Hence, the improvement in the photovoltaic performances for sample TS03 and TS07 were relatively minor. When the hydrothermal growth duration was extended to 24 h, prominent TiO₂ nanoflakes formed densely on the surface of SnO₂ nanofiber to produce the hierarchical nano-architecture. This has greatly enhanced the coverage density to inhibit the back electron transfer and increased the surface area for dye loading. The dye-desorptions on these photoelectrodes were performed and the amount of dye absorbed onto the surface of TiO₂–SnO₂ hierarchical heterostructures (sample TS24, 2.9 × 10⁻⁷ mol cm⁻²) was found to be 61.1% greater than the pristine SnO₂ nanofibers (sample SS, 1.8 × 10⁻⁷ mol cm⁻²) (ESI S5†). The enhancement in the dye-loading as a result of increased specific surface area of the photoelectrode made from TiO₂–SnO₂ hierarchical nanostructures has directly contributed to higher J_SC.

In addition to higher dye loading capability, the hierarchical TiO₂–SnO₂ photoelectrode also appeared to possess enhanced light scattering ability. Enhanced light scattering ability of a photoelectrode is crucial for high light harvesting efficiency that contributes to higher photocurrent density of a DSSC.^52–54 The difference in the architecture of photoelectrode has a direct influence on the light scattering property of a DSSC.^55,56 As clearly perceived from the diffuse reflectance spectra of sample TS24 and SS in Fig. 5, the hierarchical TiO₂–SnO₂ film has higher diffuse reflection ability in the wavelength range from 430 nm to 800 nm. This suggests that the film possessed supreme light scattering capability, which allowed multiple light traveling paths to improve the light interactions with the dye-molecules anchored on the surface of the metal oxide. In contrast, the SnO₂ nanofiber film demonstrated comparatively poorer light scattering capability. This has resulted in the unabsorbed light to penetrate through the film with limited back-scattered light for multiple interactions. Hence, the light harvesting efficiency and photocurrent were relatively poor.

Fig. 5 Comparison of diffuse reflectance spectra for SnO₂ nanofiber film and hierarchical TiO₂–SnO₂ nanocomposite film.

EIS has been widely employed to investigate the kinetics of electrochemical and photoelectrochemical system. It has been used to study the prominent electronic and ionic processes in DSSCs.^57,58 To distinguish the difference in the interfacial characteristics of these photoelectrodes, the EIS spectra of the DSSCs were measured at an applied bias of V_OC and a frequency range from 0.1 Hz to 100 kHz, with ac amplitude of 10 mV in a simulated solar light. Fig. 6 shows EIS spectra of the DSSCs with the photoelectrode made from SnO₂ nanofiber and hierarchical TiO₂–SnO₂, respectively. In general, 3 semicircles in the order of increasing frequency can be observed from the Nyquist plot. The low frequency region is attributed to the Nernst diffusion within the electrolyte, while the intermediate frequency features the electron transport in the oxide semiconductor/dye/electrolyte interface. Lastly, the high frequency peak corresponds to the charge transfer at the platinum coated counter electrode. The equivalent circuit diagrams were fitted based on the general transmission line model for DSSCs as shown in the inset of Fig. 6a.

Fig. 6 (a) Nyquist plot of DSSCs with SnO₂ nanofibers and hierarchical TiO₂–SnO₂ heterostructures as photoelectrode under simulated sunlight illumination. Inset shows the equivalent circuit diagram. (b) Corresponding Bode phase plot.

Since the EIS measurements were carried out at open-circuit potential bias, there was no current passing through the external circuit. The photogenerated electrons injected from the dye molecules into the oxide semiconductor must be recombined by I₃⁻ ions at this interface. Hence, the R_μ resistance shown in the Nyquist plot represents a net charge-transfer resistance at the oxide semiconductor/dye/electrolyte interface. With higher dye-loading for sample TS24, more photogenerated electrons were injected into the oxide semiconductor under illumination. Consequently, the net charge-transfer resistance at this interface was much lower for sample TS24 as compared to sample SS.

The lifetime of photogenerated electrons in the oxide film (t_r) can be estimated from the maximum angular frequency (ω_max) of the impedance semicircle arc at middle frequencies, according to the relation t_r = 1/ω_max = 1/2πf_max, where f_max is the maximum frequency of the mid-frequency peak.⁵⁹ As shown in the inset of Fig. 6b, the f_max value for sample TS24 is 7.14 Hz (t_r-TS24 = 22.3 ms), much lower than the values for sample SS at 10.99 Hz (t_r-SS = 14.5 ms). These data suggest that photogenerated electrons in the hierarchical TiO₂–SnO₂ film have a longer lifetime than in SnO₂ nanofiber film. This increment in the t_r value corresponds to the effective retardation of charge recombination reaction between photoinjected electrons with I₃⁻ during the electron transport through the film made with hierarchical TiO₂–SnO₂ composite.

Although SnO₂ is always considered as a potential material for photoelectrode, the DSSC with SnO₂ photoelectrodes generally shows lower conversion efficiency as compared to its TiO₂ counterpart due to its inherently larger charge recombination, slower charge injection and poorer dye-loading. This study has demonstrated that proper design of SnO₂ photoelectrodes can be used to alleviate these issues. The hierarchical nanostructure of TiO₂–SnO₂ employed for DSSC has led to improved photovoltaic performance with respect to those based on pristine SnO₂ nanofiber. The proposed architecture is efficient for DSSC with SnO₂ photoelectrode due to the larger surface area for dye-loading, inhibition of back-electron transfer, better carrier lifetime and enhanced light scattering effects.

However, the improved conversion efficiency is still inferior with reference to the state of the art DSSC. The overall low conversion efficiency might be attributed to lower dye-loading in these photoelectrodes deposited in this work, in contrast to a typical DSSC with densely packed nanoparticles film. The thicknesses of the SnO₂ layer used in this study were controlled to be around ∼6 μm for better consistency. It is believed that increasing the thickness of SnO₂ layer can improve the overall dye-loading to further enhance the short circuit current density. Besides that, large pores between the nanofibers could be observed from the SEM images. This has caused a less compact structure and affected the dye-loading and thus the overall efficiency.

4. Conclusions

In summary, the hierarchical hybrid architecture with TiO₂ nanoflakes on SnO₂ nanofibers has been successfully developed by combining the electrospinning and single-step hydrothermal growth method. The high-surface to volume ratio of this hierarchical hybrid nanostructure, along with the appropriate combination of materials, makes it an excellent candidate for photoelectrode in DSSC. In general, the DSSC made using hierarchical hybrid nanostructure of TiO₂–SnO₂ demonstrated improved photovoltaic performance compared to those based on pristine SnO₂ nanofiber. As the hydrothermal growth duration increases, the makeover of tubular structure of SnO₂ nanofiber to hierarchical hybrid architecture of TiO₂–SnO₂ was successfully achieved, leading to higher specific surface area. The back electron transfer for recombination in this hierarchical hybrid structure has been successfully retarded. Consequently, the lifetime of the photogenerated electron transported through the hierarchical matrix was improved. The synergistic effects from all these factors, together with the excellent light scattering ability have contributed to extraordinary improvement of photocurrent density, open circuit potential and the fill factor of the DSSC. These improvements have in turn transformed into a remarkable 4-fold improvement in the photovoltaic power conversion efficiency of the DSSC made from the well-defined TiO₂ nanoflakes on SnO₂ nanofibers hydrothermally grown for 24 h. This work suggests that this novel hierarchical nanostructure synthesized through this facile method may serve as a new generation photoelectrode for high efficiency DSSC. On top of that, its potential applications are foreseeable to be extended and implemented in other fields such as photocatalysis and charge storage.

References

1. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
2. M. K. Nazeeruddin, S. M. Zakeeruddin, R. Humphry-Baker, M. Jirousek, P. Liska, N. Vlachopoulos, V. Shklover, C.-H. Fischer and M. Grätzel, Inorg. Chem., 1999, 38, 6298–6305.
3. C.-Y. Chen, M. Wang, J.-Y. Li, N. Pootrakulchote, L. Alibabaei, C.-h. Ngoc-le, J.-D. Decoppet, J.-H. Tsai, C. Grätzel, C.-G. Wu, S. M. Zakeeruddin and M. Grätzel, ACS Nano, 2009, 3, 3103–3109.
4. S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, F. E. CurchodBasile, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, K. NazeeruddinMd and M. Grätzel, Nat. Chem., 2014, 6, 242–247.
5. M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
6. G. Calogero, A. Bartolotta, G. di Marco, A. di Carlo and F. Bonaccorso, Chem. Soc. Rev., 2015, 44, 3244–3294.
7. G. Calogero, J. H. Yum, A. Sinopoli, G. di Marco, M. Grätzel and M. K. Nazeeruddin, Sol. Energy, 2012, 86, 1563–1575.
8. C. Sandquist and J. L. McHale, J. Photochem. Photobiol., A, 2011, 221, 90–97.
9. L. Y. Chang, C. P. Lee, C. T. Li, M. H. Yeh, K. C. Ho and J. J. Lin, J. Mater. Chem. A, 2014, 2, 20814–20822.
10. J. Burschka, A. Dualeh, F. Kessler, E. Baranoff, N. L. Cevey-Ha, C. Yi, M. K. Nazeeruddin and M. Grätzel, J. Am. Chem. Soc., 2011, 133, 18042–18045.
11. I. Chung, B. Lee, J. He, R. P. H. Chang and M. G. Kanatzidis, Nature, 2012, 485, 486–489.
12. M. J. Ju, I. Y. Jeon, J. C. Kim, K. Lim, H. J. Choi, S. M. Jung, I. T. Choi, Y. K. Eom, Y. J. Kwon, J. Ko, J. J. Lee, H. K. Kim and J. B. Baek, Adv. Mater., 2014, 26, 3055–3062.
13. G. Calogero, F. Bonaccorso, O. M. Marago, P. G. Gucciardi and G. di Marco, Dalton Trans., 2010, 39, 2903–2909.
14. J. Liu, T. Luo, S. Mouli, F. Meng, B. Sun and M. Li, Chem. Commun., 2010, 46, 472–474.
15. H. Sun, J. Deng, L. Qiu, X. Fang and H. Peng, Energy Environ. Sci., 2015, 8, 1139–1159.
16. S. Zhu, L. Shan, X. Chen, L. He, J. Chen, M. Jiang, X. Xie and Z. Zhou, RSC Adv., 2013, 3, 2910–2916.
17. L.-Y. Chen and Y.-T. Yin, Nanoscale, 2013, 5, 1777–1780.
18. H. J. Snaith and C. Ducati, Nano Lett., 2010, 10, 1259–1265.
19. A. Birkel, Y. G. Lee, D. Koll, X. V. Meerbeek, S. Frank, M. J. Choi, Y. S. Kang, K. Char and W. Tremel, Energy Environ. Sci., 2012, 5, 5392–5400.
20. S. Yang, H. Kou, J. Wang, H. Xue and H. Han, J. Phys. Chem. C, 2010, 114, 4245–4249.
21. X. Jin, C. Liu, J. Xu, Q. Wang and D. Chen, RSC Adv., 2014, 4, 35546–35553.
22. S. S. Shin, J. S. Kim, J. H. Suk, K. D. Lee, D. W. Kim, J. H. Park, I. S. Cho, K. S. Hong and J. Y. Kim, ACS Nano, 2013, 7, 1027–1035.
23. S. Gubbala, V. Chakrapani, V. Kumar and M. K. Sunkara, Adv. Funct. Mater., 2008, 18, 2411–2418.
24. N. G. Park, M. G. Kang, K. S. Ryu, K. M. Kim and S. H. Chang, J. Photochem. Photobiol., A, 2004, 161, 105–110.
25. A. N. M. Green, E. Palomares, S. A. Haque, J. M. Kroon and J. R. Durrant, J. Phys. Chem. B, 2005, 109, 12525–12533.
26. A. Kay and M. Grätzel, Chem. Mater., 2002, 14, 2930–2935.
27. J. B. Asbury, E. Hao, Y. Wang, H. N. Ghosh and T. Lian, J. Phys. Chem. B, 2001, 105, 4545–4557.
28. S. Anderson, E. C. Constable, M. P. Dare-Edwards, J. B. Goodenough, A. Hamnett, K. R. Seddon and R. D. Wright, Nature, 1979, 280, 571–573.
29. S. Ardo and G. J. Meyer, Chem. Soc. Rev., 2009, 38, 115–164.
30. S. H. Ahn, D. J. Kim, W. S. Chi and J. H. Kim, Adv. Mater., 2013, 25, 4893–4897.
31. C. Gao, X. Li, B. Lu, L. Chen, Y. Wang, F. Teng, J. Wang, Z. Zhang, X. Pan and E. Xie, Nanoscale, 2012, 4, 3475–3481.
32. J. Huo, Y. Hu, H. Jiang, W. Huang and C. Li, J. Mater. Chem. A, 2014, 2, 8266–8272.
33. P. Docampo, P. Tiwana, N. Sakai, H. Miura, L. Herz, T. Murakami and H. J. Snaith, J. Phys. Chem. C, 2012, 116, 22840–22846.
34. M. K. I. Senevirathna, P. K. D. D. P. Pitigala, E. V. A. Premalal, K. Tennakone, G. R. A. Kumara and A. Konno, Sol. Energy Mater. Sol. Cells, 2007, 91, 544–547.
35. C. Prasittichai and J. T. Hupp, J. Phys. Chem. Lett., 2010, 1, 1611–1615.
36. K. Zhu, N. R. Neale, A. Miedaner and A. J. Frank, Nano Lett., 2007, 7, 69–74.
37. Z. Sun, J. H. Kim, Y. Zhao, D. Attard and S. X. Dou, Chem. Commun., 2013, 49, 966–968.
38. S. S. Mali, H. Kim, W. Y. Jang, H. S. Park, P. S. Patil and C. K. Hong, ACS Sustainable Chem. Eng., 2013, 1, 1207–1213.
39. T. Krishnamoorthy, V. Thavasi, G. M. Subodh and S. Ramakrishna, Energy Environ. Sci., 2011, 4, 2807–2812.
40. S. Cavaliere, S. Subianto, I. Savych, D. J. Jones and J. Rozière, Energy Environ. Sci., 2011, 4, 4761–4785.
41. E. N. Kumar, R. Jose, P. S. Archana, C. Vijila, M. M. Yusoff and S. Ramakrishna, Energy Environ. Sci., 2012, 5, 5401–5407.
42. Z. Liu, D. D. Sun, P. Guo and J. O. Leckie, Nano Lett., 2007, 7, 1081–1085.
43. X. Peng, A. C. Santulli, E. Sutter and S. S. Wong, Chem. Sci., 2012, 3, 1262–1272.
44. S. Jiang, B. Zhao, R. Ran, R. Cai, M. O. Tade and Z. Shao, RSC Adv., 2014, 4, 9367–9371.
45. L. Wang, X. Luo, X. Zheng, R. Wang and T. Zhang, RSC Adv., 2013, 3, 9723–9728.
46. L. Xu, R. Xing, J. Song, W. Xu and H. Song, J. Mater. Chem. C, 2013, 1, 2174–2182.
47. D. Cahen, G. Hodes, M. Grätzel, J. F. Guillemoles and I. Riess, J. Phys. Chem. B, 2000, 104, 2053–2059.
48. M. K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Müller, P. Liska, N. Vlachopoulos and M. Grätzel, J. Am. Chem. Soc., 1993, 115, 6382–6390.
49. Y. Diamant, S. Chappel, S. G. Chen, O. Melamed and A. Zaban, Coord. Chem. Rev., 2004, 248, 1271–1276.
50. Q. Zhang and G. Cao, Nano Today, 2011, 6, 91–109.
51. S. Ito, P. Liska, P. Comte, R. Charvet, P. Pechy, U. Bach, L. Schmidt-Mende, S. M. Zakeeruddin, A. Kay, M. K. Nazeeruddin and M. Grätzel, Chem. Commun., 2005, 4351–4353,  10.1039/b505718c.
52. G. Shang, J. Wu, S. Tang, M. Huang, Z. Lan, Y. Li, J. Zhao and X. Zhang, J. Mater. Chem., 2012, 22, 25335–25339.
53. E. Ramasamy and J. Lee, J. Phys. Chem. C, 2010, 114, 22032–22037.
54. Q. Zhang, D. Myers, J. Lan, S. A. Jenekhe and G. Cao, Phys. Chem. Chem. Phys., 2012, 14, 14982–14998.
55. Y. F. Wang, J. W. Li, Y. F. Hou, X. N. Yu, C. Y. Su and D. B. Kuang, Chem.–Eur. J., 2010, 16, 8620–8625.
56. J. Qian, P. Liu, Y. Xiao, Y. Jiang, Y. Cao, X. Ai and H. Yang, Adv. Mater., 2009, 21, 3663–3667.
57. Q. Wang, J. E. Moser and M. Grätzel, J. Phys. Chem. B, 2005, 109, 14945–14953.
58. F. Fabregat-Santiago, J. Bisquert, G. Garcia-Belmonte, G. Boschloo and A. Hagfeldt, Sol. Energy Mater. Sol. Cells, 2005, 87, 117–131.
59. R. Kern, R. Sastrawan, J. Ferber, R. Stangl and J. Luther, Electrochim. Acta, 2002, 47, 4213–4225.

---

Footnote

† Electronic supplementary information (ESI) available: Calculated grain size from XRD pattern; thermogravimetric analysis (TGA) of electrospun SnO₂ nanofibers; SEM images of hierarchical TiO₂–SnO₂ nanostructures after annealing; Energy Dispersive X-ray (EDX) measurement of hierarchical TiO₂–SnO₂ nanostructures; dye absorbance measurements. See DOI: 10.1039/c5ra25772g

---