Introduction

Nanoclusters of metals and semiconductors are more and more considered as building blocks of future technology. This is due to the size of these particles. Nano crystalline TiO2 has attracted continuous attention due to its versatile applications in optical devices, sensors, catalysis and photocatalysis etc. (Rufen and Huating 2011). In particular, nanosized TiO2 has many advantages in the dye-sensitized solar cells. With regards to nanocrystalline TiO2, the optical properties have been tentatively studied in recent years and some interesting results obtained. The use of large surface area semiconductor for materials in dye-sensitized solar cells (DSSC) is necessary to provide sufficient light absorption and charge separation which are the two critical stages in the solar–electric energy conversion. The mesoporous nano TiO2/SiO2 composite is a promising area due to optimum porous size. Nanosized TiO2 has been fabricated using sol–gel, sputtering, combustion flame, and thermal plasma (Zhang and Xu 2004; San Vicente et al. 2001). Although the sol–gel method is considered as a suitable method to synthesize ultra-fine particles, this method needs a large quantity of solution, longer processing time and heat treatment for crystallization since amorphous TiO2 has a very little photocatalytic activity. The photo catalytic efficiency of Titania (TiO2) depends highly on particle size and surface area of the material (Hakki et al. 2009; Ohno et al. 2009). The efficiency of TiO2 for dye-sensitized solar cells (DSSC) is highly depending on particle size and surface area. Commercially, TiO2 is available with surface area of around 60 m2/g. The surface area has further more increased by forming composite with SiO2.

In this work, TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposites were prepared by a novel and simple route. The TiO2, SiO2 nanoparticles were synthesized by sol–gel method using titanium isopropoxide and tetra methyl orthosilicate as starting material. The present work aims at studying the structural, optical and electrical conductivity of TiO2/SiO2 nanocomposites in comparison with synthesized pure TiO2 and SiO2 nanoparticles.

Experimental

Synthesis of TiO2 nanoparticles

The solution of titanium (IV) isopropoxide Ti(OC3H7)4 was added dropwise in isopropyl alcohol and stirred for 30 min. The metal oxide gel was produced by increasing the pH by dropwise addition of 1 N NH3 solution. The resultant solution was stirred for 24 h and kept for 1 day aging. The solution was filtered after 1 day of aging in order to remove any particulates. The precipitate was washed several times with distilled water and dried in oven for 24 h to remove the solvent. Removal of residual organics and the stabilization of the materials were carried out by calcination for 3 h at 400°C (Aguado et al. 2006).

Synthesis of SiO2 nanoparticles

The solution of tetra-ethylorthosilicate Si(OC2H5)4 was added dropwise in isopropyl alcohol and stirred for 30 min. The metal oxide gel was produced by increasing the pH by dropwise addition of 1 N NH3 solution. The resultant solution was stirred for 24 h and kept for 1 day aging. The solution was filtered after 1 day of aging in order to remove any particulates. The precipitate was washed several times with distilled water and dried in oven for 24 h to remove the solvent. Removal of residual organics and the stabilization of the materials were carried out by calcination for 3 h at 400°C (Aguado et al. 2006).

Synthesis of TiO2/SiO2 nanocomposites

The solution of titanium (IV) isopropoxide Ti(OC3H7)4 was added dropwise in isopropyl alcohol and stirred. A solution of tetra-ethylorthosilicate Si(OC2H5)4 in isopropyl alcohol was added to the reaction medium and stirred for 30 min. The mixed oxide gel was produced by increasing the pH by dropwise addition of 1 N NH3 solution. The resultant solution was stirred for 24 h and kept for 1 day aging. The solution was filtered after 1 day of aging in order to remove any particulates. The precipitate was washed several times with distilled water and dried in oven for 24 h to remove the solvent. Removal of residual organics and the stabilization of the materials were carried out by calcination for 3 h at 400°C (Aguado et al. 2006).

Characterization

The crystal structure of the powder was studied by powder X-ray diffraction with Rigaku II Cu K(α) using a Cu Kα radiation (λ = 0.154 nm). The diffraction patterns were taken over the 2θ range 20°–80° by step scanning with a step size of 0.02°. The average crystallite sizes of TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposites were determined according to Scherrer’s equation D = 0.94λ/β cosθ (Castro et al. 2008; Bartram 1967), where D is crystallite size, λ is wavelength, β is full width half maximum, and θ is angle of diffraction. The Fourier transform infrared spectra of the samples were studied using Perkin–Elmer infrared spectrophotometer. The spectrum is recorded in the range of wavenumber 500–4,000 cm−1. Nitrogen adsorption–desorption isotherm at 74 K was obtained from ASAP 2020 Micrometrics. Surface areas were determined according to the Brunauett–Emmett–Teller (BET) method. The pore size distributions were calculated by applying the Barrett–Joyner–Halenda (BJH) model. The BET surface area, micropore area, macropore volume, mesopore volume and total pore volume were calculated. The UV–Vis spectra were obtained using UV–Vis–NIR spectrophotometer. The spectra were recorded at room temperature in the range 200–1,000 nm. The field-dependent dark and photoconductivity studies were carried out using Keithley picoammeter. The experimental setup for the measurement of field-dependent dark and photoconductivity is as used by Ponniah and Xavier (2007). The samples in the form of pellets are attached to the microscopic glass slide and two electrodes of thin copper wire (0.14-mm diameter) were fixed by the use of silver paint. The ends of the copper wire were connected to DC power supply through picoammeter (Keithley picoammeter 6485) as shown in Fig. 1. The applied field was varied and the corresponding current in the circuit was measured. To measure the photocurrent, light from 100 W halogen light was illuminated onto the sample (Ponniah and Xavier 2007).

Fig. 1
figure 1

Experimental set up for the measurement of field-dependent conductivity

Full size image

Results and discussion

Figure 2 shows the XRD pattern of the TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite obtained by the sol–gel method. The XRD pattern of the as-prepared TiO2 shows the presence of broad peak. All the diffraction lines are assigned well to the crystallite phase of TiO2 (Zhao et al. 2007). The XRD pattern is in excellent agreement with the reference pattern (JCPDS 21-1272) of TiO2. It should be noted that only anatase TiO2 is detected and no rutile phase can be found in the sample (Khanna et al. 2007). The XRD pattern of prepared SiO2 nanoparticle and TiO2/SiO2 nanocomposites shows the presence of very broad peak. The broad peak indicates that either the particles are of very small crystallite size, or particles are semi crystalline in nature (Yeh et al. 2004; Zhou et al. 2006). The average crystalline sizes of the as-synthesized nanoparticles were estimated from XRD line broadening using Scherrer’s equation (Castro et al. 2008; Bartram 1967) by considering the full width and half maximum (FWHM) value (shown in Table 1). The crystallite size of TiO2/SiO2 nanocomposite is found to be 0.50 nm. This relatively low crystallite size is due to the low growth rate (Jian et al. 1991).

Fig. 2
figure 2

XRD pattern of the as-synthesized mesoporous a TiO2 nanoparticles, b SiO2 nanoparticles and c TiO2/SiO2 nanocomposite

Full size image
Table 1 Particle size and optical band gap of the as-synthesized samples
Full size table

UV–Vis absorption study was carried out in order to characterize the optical absorbance of the sample. The absorption spectra of the as-synthesized TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite are shown in Figs. 3, 4 and 5, respectively. The optical band gap can be estimated by plotting (αhν)2 versus photon energy (hν) based on the relation αhν = A (hν − Eg)n/2 where α is the absorption coefficient, A is a constant, Eg is the band gap and n is the exponent depending on quantum selection rule for a particular material, n = 1 for a direct transition (Khanna et al. 2007; Oral et al. 2004). According to the above relation, the intercept of the tangent on the photon energy axis corresponds to optical band gap (Oral et al. 2004). From the (αhν)2 versus photon energy (hν) plots, the optical band gaps Eg for all the three synthesized samples were estimated and tabulated in Table 1. Figure 6 shows the UV–Vis spectrum of the as-synthesized samples. All the three as-synthesized samples show similar absorption pattern in which the TiO2/SiO2 nanocomposite shows better absorbance in visible range compared to TiO2 and SiO2 nanoparticles.

Fig. 3
figure 3

UV-Vis spectrum of the as-synthesized TiO2 nanoparticles inset shows the corresponding plot of photon energy versus (αhν)2

Full size image
Fig. 4
figure 4

UV-Vis spectrum of the as-synthesized SiO2 nanoparticles inset shows the corresponding plot of photon energy versus (αhν)2

Full size image
Fig. 5
figure 5

UV-Vis spectrum of the as-synthesized TiO2/SiO2 nanocomposite inset shows the corresponding plot of photon energy versus (αhν)2

Full size image
Fig. 6
figure 6

UV-Vis spectrum of the as-synthesized TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite

Full size image

Nitrogen adsorption isotherm pattern of the as-synthesized mesoporous TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposites is shown in Figs. 7, 8 and 9, respectively. Some characteristics of the samples, such as BET surface area, micropore area, average pore diameter, micropore volume, mesopore volume and total pore volume are listed in Table 2. According to IUPAC nomenclature, the absorbent pores are classified into three groups: micropore (diameter <2 nm), mesopore (diameter 2–50 nm) and macropore (diameter >50 nm) (Wang et al. 2010). The BET measurements confirmed the absence of macropores in all nanoparticles. The pore diameter for TiO2, SiO2 nanoparticle and TiO2/SiO2 nanocomposite was found to be 8.39, 9.32 and 9.82 nm. The mesopore volume was obtained by subtracting micropore volume from the corresponding total volume (Sayilkan et al. 2007). The mesoporosities (percentage of mesopore to total pore volume Vmes/Vtot) were calculated and found to be 75.45, 73.8 and 50.53% for TiO2, SiO2 nanoparticle and TiO2/SiO2 nanocomposite, respectively. For TiO2/SiO2 nanocomposites, the microporosity increased and the mesoporosity decreased when compared to TiO2 and SiO2 nanoparticles. This may be due to shrinking of mesopores or due to the SiO2 being absorbed into the mesopores of TiO2 (Sayilkan et al. 2007). The BET surface area was found to be 62 to 303 m2/g in the case of TiO2 nanoparticles and TiO2/SiO2 nanocomposite, respectively. This sizeable increase in surface area of TiO2/SiO2 nanocomposite may be due to the SiO2 limiting the agglomeration of TiO2 particles (Nilchi et al. 2011; Sirimahachai and Ndiege 2010).

Fig. 7
figure 7

Nitrogen adsorption isotherm pattern of TiO2 nanoparticle (inset pore size distribution)

Full size image
Fig. 8
figure 8

Nitrogen adsorption isotherm pattern of SiO2 nanoparticle inset pore size distribution

Full size image
Fig. 9
figure 9

Nitrogen adsorption isotherm pattern of TiO2/SiO2 nanocomposite inset pore size distribution

Full size image
Table 2 The physical properties of the as-synthesized samples
Full size table

Fourier transform infrared (FTIR) spectrum of as-synthesized mesoporous TiO2 nanoparticles is shown in Fig. 10. It was absorbed that the strong band in the range of 900–500 cm−1 is associated with the characteristic vibrational modes of TiO2 (Khanna et al. 2007). This confirms that the TiO2 phase has been formed. The absorption in the range from 3,640 to 2,500 cm−1 may be related to the presence of O–H stretching vibration (Monomer, intermolecular, intramolecular and polymeric). The absorption band at 1,629 cm−1 was due to the presence of O–H bending vibration which is probably because the reabsorption of water from the atmosphere has occurred (Mohan 2009). Fourier transform infrared spectrum of as-synthesized mesoporous SiO2 nanoparticles is shown in Fig. 11. The two strong bands 1,118 and 804 cm−1 observed are associated with asymmetric and symmetric Si–O–Si stretching vibrations (Aziz and Sopyan 2009), respectively. This confirms that the SiO2 phase is formed. FTIR also showed that the band at 1,076 cm−1 was slightly shifted towards lower wavenumber as the particle size is reduced (Singh et al. 2011). The absorption bands at 3,428 and 1,635 cm−1 were due to the presence of O–H stretching and bending vibrations (Mohan 2009), respectively. Fourier transform infrared (FTIR) spectrum of as-synthesized mesoporous TiO2/SiO2 nanocomposite is shown in Fig. 12. The band observed at 923 cm−1 is associated with Si–O–Ti vibration (Aziz and Sopyan 2009). The two strong bands at 1,050 and 803 cm−1 observed are associated with asymmetric and symmetric Si–O–Si stretching vibration (Aziz and Sopyan 2009), respectively. The strong bands in the range 900–500 cm−1 are associated with vibrational modes of TiO2. The absorption bands at 3,419 and 1,631 cm−1 were due to the presence of O–H stretching and bending vibrations (Aziz and Sopyan 2009), respectively.

Fig. 10
figure 10

FTIR spectrum of as-synthesized TiO2 nanoparticles

Full size image
Fig. 11
figure 11

FTIR spectrum of as-synthesized SiO2 nanoparticles

Full size image
Fig. 12
figure 12

FTIR spectrum of as-synthesized TiO2/SiO2 nanocomposite

Full size image

The field-dependent dark and photoconductivity of TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite are shown in Figs. 13 and 14, respectively. The plot indicates a linear increase of current in the dark and visible light-illuminated samples in all the three cases with increase in applied field depicting the ohmic nature of the contacts (Dhar and Chakrabarti 1996). TiO2/SiO2 nanocomposite showed better dark and photo currents compared to TiO2 and SiO2 nanoparticles. For example, for a fixed field of 800 V/cm, the TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite showed dark current of 0.0178, 0.4 and 4.607 μA, respectively. It is 258 fold more than the value for TiO2 nanoparticle. For a fixed field of 800 V/cm, the TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite showed dark current of 0.0191, 0.5 and 5.6375 μA, respectively. It is nearly 300 times more the value of TiO2 nanoparticle. This may be due to the enhanced surface area available for conduction or attributed to the increase in charge carrier concentration and drift mobility in the composite (Xavier and Goldsmith 1995).

Fig. 13
figure 13

The field-dependent dark conductivity of as-synthesized mesoporous TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite

Full size image
Fig. 14
figure 14

The field-dependent photoconductivity of as-synthesized mesoporous TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposite

Full size image

Conclusion

A novel, easy and reproductive method was followed for the synthesis of TiO2, SiO2 nanoparticles and TiO2/SiO2 nanocomposites. TiO2/SiO2 nanocomposite showed enhanced surface area of 303 m2/g in comparison with TiO2 nanoparticle which increases the photocurrent in the field-dependent photoconductivity. The pore size distribution shows that the as-synthesized nanocomposite is mesoporous. The material with large surface area and mesoporous nature would increase the adsorption of dye on it, which in turn will improve photosensitivity to solar radiation. Thus, TiO2/SiO2 nanocomposites can be dye sensitized and used as working electrode in dye-sensitized solar cells.