Quantitative studies on the size induced anatase to rutile phase transformation in TiO₂–SiO₂ binary oxides during heat treatments

Abstract

The effect of SiO₂ content on the phase behavior of TiO₂ during heat treatment was investigated in the present study. Sol–gel synthesis was employed to form five different TiO₂–SiO₂ binary oxides. The characterization results confirmed the vital role of SiO₂ in reducing the crystallite growth of TiO₂, which was instrumental in restricting the anatase (a-TiO₂) to rutile (r-TiO₂) phase transformation. It was revealed from the characterization techniques that crystallite size exceeding ∼50 nm induced a-TiO₂ to r-TiO₂ phase transition. Heat treatment beyond 1100 °C resulted in the gradual conversion from a-TiO₂ to r-TiO₂ and moreover the crystallization of cristobalite (c-SiO₂) could not be avoided at 1300 °C. The results from the photoemission spectra ensured a blue shift for the a-TiO₂ stabilized by SiO₂ from the characteristic emissions of pure r-TiO₂. Antibacterial tests also indicated a good response for TiO₂–SiO₂ binary oxides in encountering the microbes.

---

1. Introduction

The application of titanium dioxide (TiO₂) in biomedical applications is profound owing to its outstanding biocompatibility, high chemical stability, excellent corrosion resistance, antimicrobial property and low toxicity.^1–4 The ability of the hydrated TiO₂ to form an apatite layer during its immersion in simulated body fluids (SBF) is considered an additional advantage while employing TiO₂ as a replacement for bone and tooth enamel. Biocompatibility testing with osteoblasts have indicated the TiO₂ ability to induce the attachment and growth of cells, making it an excellent osteoblast compatible material.^5,6 Consequently, implants with TiO₂ could promote osseointegration between an implant and hard tissue and also reduce the risk of infection. Generally, TiO₂ is available in the form of four different polymorphs namely rutile TiO₂ (r-TiO₂), anatase TiO₂ (a-TiO₂), brookite TiO₂ (b-TiO₂) and monoclinic TiO₂ (m-TiO₂). Among these available polymorphs, r-TiO₂ and a-TiO₂ are the most investigated forms because of their favorable features.⁷ a-TiO₂ is stable at low temperature whereas r-TiO₂ is stable at high temperature.⁸ Since the surface Gibbs free energy of a-TiO₂ is lower than that of r-TiO₂, TiO₂ initially prefers to nucleate into anatase form rather than in its rutile form.⁹ Moreover, the applications of a-TiO₂ and r-TiO₂ differ because of the significant differences observed in their respective bonding arrangement and positioning of the individual atoms in the crystal structures. In terms of biological applications, a-TiO₂ is proven to show better activity in countering the attack of microbes rather than r-TiO₂.^10,11 The major shortcoming of a-TiO₂ is its low thermal stability thus implying a severe restriction in the attainment of high density which is generally obtained during high temperature heat treatment.^12,13 It has been prescribed that a-TiO₂ to r-TiO₂ phase transformation are governed by annealing temperature, compactness of the anatase crystallites, particle arrangement and grain boundary defects.^14,15 Literature survey have evidenced the fact that a-TiO₂ to r-TiO₂ phase transformations could be retarded by the usage of several additives in the form of Al₂O₃, SiO₂, ZrO₂, F and CuO in TiO₂ matrix.^16–20 However, stabilization of a-TiO₂ without appropriate inorganic oxide dopant possesses difficultly in the attainment of quick bonding with natural bone during the early stage of implantation procedures. Hence, the affirmations on the quality of ideal biomaterial could not be ensured if a-TiO₂ is stabilized by any dopant.²¹ It is quite a logical phenomenon that reinforcement of TiO₂ with a suitable bioactive inorganic oxide is a viable option to explore the biocompatibility characteristics of TiO₂.

In this context the utilization of SiO₂ for the reinforcement of TiO₂ is expected to be a suitable alternative due to the favorable features of SiO₂. The bioactivity features of SiO₂ are documented from previous reports, which states that the SiO₂ content plays a major role on the faster bony adaptation and the biocompatibility of a biomaterial.²² Moreover, the addition of SiO₂ in the form of its amorphous content has played a significant role in the stabilization of a-TiO₂.^17,23,24 Some of the investigations have reported on the thermal stability of a-TiO₂ by alloying it with the SiO₂, however, such reports have limited their investigation only in the temperature range around 900–1000 °C.²⁵ Recently, He et al.,²⁶ have reported the stabilization of a-TiO₂ by SiO₂ additions till 1100 °C and heat treatment at 1200 °C had resulted in the complete transformation to r-TiO₂. However, that particular study had targeted for the photo-catalytic applications and the maximum range of the investigated SiO₂ concentrations were found to be 20 mol%. In terms of applications in biomedicine, the content of SiO₂ is determined to have played a major role in the bioactivity of a biomaterial as determined by Hench.²² Even the commercially successful bio-glass 45S5 possesses the SiO₂ content in the range of 40–45 wt%.

Thus, the present investigation is aimed at the objective of synthesizing TiO₂–SiO₂ binary oxides with varying SiO₂ to TiO₂ precursor concentrations through a simple sol–gel technique. The effect of SiO₂ content on the thermal stability of a-TiO₂ was systematically investigated by employing the techniques involving X-ray diffraction, optical studies and quantitative analysis of the powder diffraction data by Rietveld refinement. Further, the efficiency of TiO₂–SiO₂ binary oxides in encountering the microbes were determined from the antibacterial tests.

2. Materials and methods

2.1 Synthesis of TiO₂–SiO₂ binary oxides

The synthesis of TiO₂–SiO₂ binary oxides were carried out through sol–gel technique. For this purpose, titanium(IV) isopropoxide [C₁₂H₂₈O₄Ti, Sigma-Aldrich, India] and tetraethyl orthosilicate [(C₂H₅)₄OSi, TEOS, Sigma-Aldrich, India] were taken as precursors for TiO₂ and SiO₂ respectively. Five different concentrations were attempted in the present study just by varying the molar concentrations of (C₂H₅)₄OSi while maintaining a constant molar concentration of C₁₂H₂₈O₄Ti. The molar ratios of the five different concentrations and their respective sample codes are presented in the Table 1. The procedure for the formation of TiO₂–SiO₂ gels from C₁₂H₂₈O₄Ti and (C₂H₅)₄OSi are explained as follows.

Table 1 Molar concentrations of the precursors used in the formation of TiO₂–SiO₂ binary oxides

Sample code	Molar concentrations of the precursors
	C12H28O4Ti	(C2H5)4OSi
TS 0.50	1.00 M	0.50 M
TS 0.75	1.00 M	0.75 M
TS 1.00	1.00 M	1.00 M
TS 1.25	1.00 M	1.25 M
TS 1.50	1.00 M	1.50 M

---

A known amount of C₁₂H₂₈O₄Ti was dissolved in 150 ml of ethanol to make 1 M C₁₂H₂₈O₄Ti solution and was transferred to a 500 ml beaker. An appropriate molar concentration of (C₂H₅)₄OSi was separately prepared by dissolving in 150 ml of ethanol and this solution was stirred (TARSONS, INDIA) at 50 °C and 300 rpm. Then, the separately prepared 1 M C₁₂H₂₈O₄Ti solution was added drop wise to the continuously stirring precursor solution of (C₂H₅)₄OSi. After 10 minutes of the completion of addition, 0.1 M of HNO₃ as a catalyst was added to the stirring mixture. This homogeneous solution was stirred for a few hours to yield a gel. The formed gel was transferred to a hot air oven (TECHNICO OVEN, INDIA) and dried at 120 °C overnight. The obtained dried samples were grounded well to fine powders with the help of mortar and pestle and this powder was considered to be as prepared powder. This was followed by the heat treatment at different temperatures for a soaking time of 4 hours and was investigated for their phase behavior by employing different characterization techniques. All the heat treatments were made in air atmosphere at ambient conditions.

2.2 Powder characterization

X-ray diffraction studies for all the powders after heat treatment at different temperatures were carried out using a high resolution X-ray diffractometer (RIGAKU, ULTIMA IV, JAPAN) with Cu Kα radiation (λ = 1.5406 nm) produced at 40 kV and 30 mA to scan the diffraction angles (2θ) between 10 and 70° with a step size of 0.02° 2θ per second. The slow scan for the analysis through Rietveld refinement was done with a step size of 0.01° 2θ per second. Phase determination was made using standard ICDD (International Centre for Diffraction Data) card no. 01-071-1166 for a-TiO₂ and 01-076-0317 for r-TiO₂. Rietveld analysis was performed using GSAS-EXPGUI software package.^27,28 Quantitative refinement was carried out for all the powder samples after heat treatment at various temperatures. All the standard crystallographic information file was obtained from American mineralogist crystal structure database. The standard crystallographic data for the refinement of both a-TiO₂ and r-TiO₂ were obtained from Howard et al.,²⁹ and cristobalite SiO₂ (c-SiO₂) was obtained from Dera et al.³⁰ In structural refinement, numerous cycles were run to determine the quantitative analysis, weight fraction and structural parameters of the powders. In a first step of the refinement, all the structural parameters were fixed to the literature values. Then during the successive refinement cycles, numerous parameters were allowed to vary accordingly to the relative weight amount of the observed phases. The following refinement sequence was used as a standard for all the structures: scale factor, zero shift, background as Chebyshev polynomial of fifth grade, peak profile, lattice parameters. Fittings were performed using pseudo-Voigt peak profile functions and a preferred orientation along [001] was taken into account with the Marsh model. The fractional coordinates, isotropic temperature and atomic parameters were employed during refinement.

2.3 Determination of crystallite size from refinement

In the diffraction pattern, position of the peaks indicates the crystal structure and symmetry of the phase, whereas the intensity of the peak is mainly related to the structure as well as composition of the phase.

The diffraction intensity eqn (1)³¹ is described as follows:

where, I_(hkl)α = intensity of reflection of hkl in phase α, I₀ = incident beam intensity, r = distance from specimen to detector, λ = X-ray wavelength, = square of classical electron radius, μ_s = linear absorption coefficient of the specimen, V_α = volume fraction of phase α, M_(hkl) = multiplicity of reflection hkl of phase α, = Lorentz-polarization, v_α = volume of the unit cell of phase α, 2θ_m = diffraction angle of the monochromator, and F_(hkl)α = structure factor for reflection hkl of phase α (i.e., the vector sum of scattering intensities of all atoms contributing to that reflection).

The crystallite size is determined from the Rietveld refinement analysis by using the following eqn (2), with the conversion of centi-degree to radians;

where K is the Scherrer constant. X is the Lorentzian broadening factor which is equal to and is obtained from Braggs law as . The units are in Å.

2.4 Optical studies

UV-Visible (UV-Vis) absorption spectra were recorded on a UV-Vis spectrophotometer (PERKIN ELMER LAMBDA 650 S) and Photoluminescence (PL) spectra were measured on a spectrofluorometer (PERKIN-ELMER LS 50B) with a xenon discharge lamp excitation. The photoexcitation was made at different excitation wavelengths at a 45° angle of the pellet plane with the excitation beam.

2.5 Tests on antibacterial activity

The evaluation of antimicrobial activity against Staphylococcus aureus (S. aureus, Gram-positive) and Escherichia coli (E. coli, Gram-negative) was performed by plate counting method. Prior to antimicrobial analysis, all the samples (Pure TiO₂, TS 0.5, TS 1.0, TS 1.5) were subjected to sterilization in autoclave at 121 °C for 20 min. Fresh inoculums of S. aureus and E. coli were prepared in broth (Müller-Hinton broth, Hi-Media, India) and were kept at 37 °C for overnight incubation before proceeding with further analysis.

Quantitative tests were performed to determine the antimicrobial activity for pure TiO₂ and different concentrations of TiO₂–SiO₂ binary oxides. All the experiments were performed with the minor modifications of the work reported by Stanic et al.,³² The procedure for the antibacterial tests is described as follows. 0.5 g of each sterile sample (Pure TiO₂, TS 0.5, TS 1.0, TS 1.5) was added to the test tube containing 9.9 ml of sterile phosphate buffer saline (PBS) solution (pH 7.4) and this mixture was shaken for 5 minutes using a vortex. This was followed by the addition of 0.1 ml of overnight incubated culture to the test tube. All the tubes were kept in water bath shaker at 37 °C for incubation. For viable cell determination, 1 ml aliquots were taken from each test tube after a time interval of 1, 2 and 4 h and sterile PBS solution was used for required dilution. Müller-Hinton agar (Hi-Media, India) was used to prepare a solid media on Petri dishes. After the appropriate dilution, 0.1 ml was spread on Petri dishes and kept it for 24 h incubation at 37 °C. Viable microorganism counts were made after different time intervals namely 0, 1, 2, and 4 h. Average value of results were taken after performing the experiment three times.

The percentage of microorganism reduction (R) was calculated by eqn (3):

where C₀ is the mean number of bacteria on the control sample tube (CFU per sample) without the sample, and C is the mean number of bacterial colonies with the samples.

3. Results and discussion

3.1 Phase analysis

The phase behaviors of the heat treated samples at various temperatures were determined through X-ray diffraction. Fig. 1 presents the powder XRD patterns of pure TiO₂ after heat treatment at different temperatures along with the standard powder XRD patterns of r-TiO₂ and a-TiO₂ [ICDD card no. 01-076-0317 (r-TiO₂) and 01-071-1166 (a-TiO₂)]. Pure TiO₂ has yielded a-TiO₂ at 400 °C and its progressive heat treatment at 500 and 600 °C has indicated the combined presence of a-TiO₂ and r-TiO₂. The complete transformation of pure TiO₂ to r-TiO₂ was confirmed at 700 °C. The XRD patterns of the TiO₂–SiO₂ binary oxides that contain varied SiO₂ content confirmed the presence of single phase a-TiO₂ after heat treatment at 1000 °C (Fig. 2). The presence of any additional crystalline phases in the form of r-TiO₂ or SiO₂ was not detected in the XRD patterns recorded at 1000 °C. Moreover, with the increased content of SiO₂ in the TiO₂–SiO₂ binary oxides, the XRD patterns recorded at 1000 °C showed a simultaneous reduction in the crystallinity by the virtue of observed sharp diffraction peaks to broad diffraction peaks. This infers the fact that SiO₂ content played a major role in restricting the a-TiO₂ to r-TiO₂ phase transformation till the heat treatment temperature of 1000 °C. Heat treatment at 1200 °C (figure not presented) and 1300 °C (Fig. 3) resulted in the significant change in the phase behavior of TiO₂–SiO₂ binary oxides. At 1200 °C, the gradual transformation from a-TiO₂ to r-TiO₂ phase was observed, which were mainly dependent on the SiO₂ content in binary oxides. Except the TS 1.50 sample that possesses the maximum SiO₂ content, all the other compositions indicated a-TiO₂ to r-TiO₂ phase transformations at 1200 °C. However, the amount of transformed r-TiO₂ was dependent on the SiO₂ content in which the sample that possess low SiO₂ content (TS 0.50) indicated the high intensity XRD peaks respective of r-TiO₂ whereas the sample that possess high SiO₂ content (TS 1.25) indicated the low intensity XRD peaks respective of r-TiO₂. The degradation of TiO₂–SiO₂ binary oxides was evident from the XRD patterns recorded at 1300 °C. The entire transformation of a-TiO₂ to r-TiO₂ phase was evident for TS 0.50 sample and the sample containing high SiO₂ content (TS 1.25) resulted in the partial transformation of a-TiO₂ to r-TiO₂ phase at 1300 °C. In addition to the presence of a-TiO₂ and r-TiO₂ phase at 1300 °C, the evolution of new crystalline phase in the form of cristobalite (c-SiO₂) was demonstrated in all the samples. The delayed crystallization of c-SiO₂ in the TiO₂–SiO₂ binary oxides at higher temperature is mainly related to the thermodynamic factors. The standard enthalpy of formation (Δ_fH°) and standard free energy change (Δ_fG°) of individual TiO₂ and SiO₂ are reported before. The Δ_iH° values of individual TiO₂ and SiO₂ are correspondingly reported as −519.7 and −910.7 kJ mol⁻¹ and the Δ_fG° values of individual TiO₂ and SiO₂ are correspondingly reported as −495.0 and −856.3 kJ mol⁻¹.³³ Based on the above mentioned thermodynamic considerations, crystallisation of SiO₂ requires very high amount of energy which is possibly provided at high temperature treatment.

Fig. 1 XRD patterns of pure TiO₂ sample after heat treatment at four different temperatures.

Fig. 2 XRD patterns of five different TiO₂–SiO₂ binary oxides after heat treatment at 1000 °C.

Fig. 3 XRD patterns of five different TiO₂–SiO₂ binary oxides after heat treatment at 1300 °C.

3.2 Quantitative analysis

The quantitative analysis on the phase behavior of TiO₂–SiO₂ binary oxides was determined from Rietveld refinement. The refined diffraction patterns of the TS 0.5, TS 0.5 and TS 1.00 samples correspondingly presented in Fig. 4–6 . Fig. 4 confirms the presence of single phase a-TiO₂ at 1000 °C whereas Fig. 5 confirms the presence of mixture of a-TiO₂ and r-TiO₂ phases at 1100 °C. The refined diffraction patterns presented in Fig. 6 for TS 1.00 confirms the presence of mixture of three different phases namely a-TiO₂, r-TiO₂ and c-SiO₂ phases at 1300 °C and thus the results from the refinement are found consistent with the results obtained from the qualitative XRD analysis. The refined structural parameters of a-TiO₂ (Tables 2a and b) for all the investigated compositions confirmed their tetragonal setting that crystallizes in I4₁/amd(141) space group. In a similar manner, the refined structural parameters of r-TiO₂ (Tables 2a and b) that formed above the heat treatment temperature of 1100 °C for the investigated compositions confirmed their tetragonal setting that crystallizes in P4₂/mnm(136) space group. The emergence of c-SiO₂ at 1300 °C also confirmed their tetragonal setting that crystallizes in P4₁2₁2(92) space group. The structural parameters presented in Tables 2a and b have not indicated any significant trend in the a-TiO₂ and r-TiO₂ phases with respect to the combined factors of either with the varied amount of SiO₂ content or with the influence of heat treatment. Any shift in the diffraction patterns of a-TiO₂ due to the addition of varied amount of SiO₂ was not noticed in any of the diffraction patterns. This inference shows that SiO₂ remained as an individual amorphous phase in the composites till 1200 °C and have not occupied the lattice positions of crystalline TiO₂. Moreover, the crystallization of SiO₂ as c-SiO₂ at 1300 °C is mainly influenced by their thermodynamic factors as described in the previous section.

Fig. 4 Refined powder diffraction patterns for the TS 0.50 sample after heat treatment at 1000 °C.

Fig. 5 Refined powder diffraction patterns for the TS 0.50 sample after heat treatment at 1100 °C.

Fig. 6 Refined powder diffraction patterns for the TS 1.00 sample after heat treatment at 1300 °C.

Table 2 Refined structural parameters determined for the different TiO₂–SiO₂ binary oxides after heat treatment at different temperatures

(a) At 1000 and 1100 °C
Sample code	Structural parameters
	1000 °C		1100 °C
	a-TiO2		a-TiO2		r-TiO2
	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)
TS 0.50	3.7825 (2)	9.5185 (6)	3.7833 (1)	9.5247 (2)	4.5943 (2)	2.9612 (3)
TS 0.75	3.7821 (3)	9.5147 (7)	3.7839 (1)	9.5251 (3)	4.5955 (6)	2.9614 (7)
TS 1.00	3.7823 (3)	9.5178 (8)	3.7840 (1)	9.5248 (3)	—	—
TS 1.25	3.7833 (4)	9.5165 (9)	3.7841 (1)	9.5258 (4)	—	—
TS 1.50	3.7837 (5)	9.5163 (9)	3.7839 (2)	9.5252 (5)	—	—

---

(b) At 1200 and 1300 °C
Sample code	Structural parameters
	1200 °C				1300 °C
	a-TiO2		r-TiO2		a-TiO2		r-TiO2		c-SiO2
	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)	a-axis (Å)	c-axis (Å)
TS 0.50	3.7821 (3)	9.521 (1)	4.5934 (1)	2.9612 (2)	—	—	4.5930 (1)	2.9593 (1)	5.0179 (5)	7.028 (1)
TS 0.75	3.7840 (2)	9.5290 (3)	4.5949 (1)	2.9616 (1)	—	—	4.5930 (1)	2.9593 (1)	5.0147 (6)	7.024 (1)
TS 1.00	3.7842 (7)	9.5307 (3)	4.5937 (2)	2.9601 (2)	3.7823 (2)	9.520 (1)	4.5909 (1)	2.9579 (1)	5.0093 (4)	7.007 (1)
TS 1.25	3.7843 (7)	9.5338 (3)	4.5980 (8)	2.9628 (9)	3.7838 (1)	9.5282 (4)	4.5929 (1)	2.9592 (1)	5.0104 (5)	7.013 (1)
TS 1.50	3.7860 (1)	9.5382 (4)	—	—	3.7853 (1)	9.5391 (3)	4.5961 (2)	2.9609 (2)	5.019 (2)	7.012 (6)

---

The weight fractions of the phase compositions determined from Rietveld refinement of the powder XRD patterns are presented in Table 3. The formation of single phase a-TiO₂ was confirmed for all the TiO₂–SiO₂ binary oxides at both the heat treatment temperatures of 900 and 1000 °C. At 1100 °C, the initiation of r-TiO₂ phase formation was confirmed for the lower SiO₂ containing samples (TS 0.50 and TS 0.75) whereas the other compositions have confirmed the presence of single phase a-TiO₂. At 1200 °C, all the samples have indicated the formation of r-TiO₂ phase irrespective of the SiO₂ content, however in a decreasing trend of its weight fractions with respect to the SiO₂ content. A significant variation in the phase fractions at 1300 °C were witnessed for all the TiO₂–SiO₂ binary oxides in which the formation of considerable amount of c-SiO₂ was notified. Not even a trace of a-TiO₂ was detected for the lower SiO₂ containing samples (TS 0.50 and TS 0.75) whereas all the samples demonstrated the presence of r-TiO₂, however, its weight fractions were determined in an decreasing order with a simultaneous enhancement in the SiO₂ content. It is also important to note here that the weight fractions of c-SiO₂ have not indicated a uniform trend with respect to the SiO₂ content in the samples. However, the formed c-SiO₂ could be correlated to the detected a-TiO₂ content in the samples that could be explained as follows. The rate of a-TiO₂ to r-TiO₂ conversion was mainly dependent on the amorphous SiO₂ present in the samples. The onset of SiO₂ crystallization in the form of c-SiO₂ has left with negligible amount of amorphous SiO₂ to resist a-TiO₂ to r-TiO₂ conversion. In case of low SiO₂ containing samples (TS 0.50 and TS 0.75) that did not contain any trace a-TiO₂, the weight fractions of c-SiO₂ indicated an increasing trend with respect to the increasing SiO₂ content. The TS 1.00 sample, which detected a trace a-TiO₂ in the range of 5 wt% had shown a maximum c-SiO₂ content of 42 wt%. Further, with the increasing presence of a-TiO₂ content in the samples (TS 1.25 and TS 1.50), the formation of c-SiO₂ had shown a reducing trend. Thus the presence of a-TiO₂ had played a major role in the crystallization of c-SiO₂ at 1300 °C.

Table 3 Weight fractions of the a-TiO₂, r-TiO₂ and c-SiO₂ for the TiO₂–SiO₂ binary oxides determined from Rietveld refinement at different heat treatment temperatures

Sample code	Weight fraction (%)
	900 °C	1000 °C	1100 °C		1200 °C		1300 °C
	a-TiO2	a-TiO2	a-TiO2	r-TiO2	a-TiO2	r-TiO2	a-TiO2	r-TiO2	c-SiO2
TS 0.50	100	100	94	6	5	95	0	73	27
TS 0.75	100	100	97	3	64	36	0	67	33
TS 1.00	100	100	100	0	78	22	5	53	42
TS 1.25	100	100	100	0	97	3	29	33	38
TS 1.50	100	100	100	0	100	0	72	14	14

---

3.3 Size dependent phase transformation

As it was confirmed from the data obtained from structural parameters that SiO₂ has not entered the structural lattice of TiO₂, it was decided to investigate further about the reason for the a-TiO₂ to r-TiO₂ phase transformation with respect to the SiO₂ content in the samples. Literature reports have evinced the investigation on the influence of heat treatment on SiO₂–TiO₂ binary oxides.^34–36 However, majority of the studies have restricted their investigation up to 900 °C. A particular report states that the SiO₂ content plays a major role in restricting the crystallite growth, which has been the major cause for a-TiO₂ to r-TiO₂ phase conversion.³⁷ The crystallite sizes of all the investigated TiO₂–SiO₂ binary oxides with respect to the different heat treatment temperatures were determined from refinement and the values are presented in the form of graphical representation in Fig. 7. It is quite obvious from the Fig. 7 that the gradual increase in the heat treatment temperature resulted in the steady enhancement in the crystallite size of all the investigated TiO₂–SiO₂ binary oxides. Similarly, the increase in SiO₂ content revealed a steady decline in the crystallite size of all the investigated TiO₂–SiO₂ binary oxides. It is quite important to mention here that the calculated crystallite size for all the TiO₂–SiO₂ binary oxides at 900 and 1000 °C, which yielded a single phase a-TiO₂ were measured in the range less than ∼50 nm. The single phase a-TiO₂ detected for the all the TiO₂–SiO₂ binary oxides irrespective of the both SiO₂ content and heat treatment temperatures were yielding the crystallite size value less than 50 nm. The crystallite size exceeding the critical value of ∼50 nm resulted in the formation of r-TiO₂ for all the TiO₂–SiO₂ binary oxides. Moreover, the crystallite size values determined for the pure TiO₂ at all the temperatures that yielded single phase r-TiO₂ was determined in the range well above 250 nm. Thus the presence of excess amorphous SiO₂ has restricted the crystallite growth of TiO₂ thereby delaying the a-TiO₂ to r-TiO₂ phase conversion in TiO₂–SiO₂ binary oxides. Based on the above observation, it could be stated that the crystal growth and rearrangement of Ti–O to form its octahedral position were ceased by the additions of SiO₂ content and it was found consistent with the availing literatures (Fig. 8).

Fig. 7 Crystallite size behavior of TiO₂–SiO₂ binary oxides with respect to the varied heat treatment temperatures.

Fig. 8 UV-Vis spectra recorded for Pure TiO₂ and five different TiO₂–SiO₂ binary oxides after heat treatments.

3.4 UV-Vis spectroscopy

The band gaps for all the TiO₂–SiO₂ binary oxides were determined from the tangent of onset absorbance edge (Table 4). The TS 0.5 sample heat treated at 900 °C showed a band gap of 3.12 eV, which got lowered to 2.94 eV on heat treatment at 1000 °C. Similarly, the reduction in band gap of TS 0.75, TS 1.0, TS 1.25 and TS 1.5 samples were observed at 1000 °C, which were determined to be initially on the higher range at 900 °C. This reason for the reduction could be attributed to the condensation mechanism started to occur at 900 °C and thereby leading to formation of Ti–O–Si linkage at 1000 °C. Further, heat treatment at 1100 °C reflected an increase in the band gap for all the composites, which has been mainly attributed to the enhancement in the crystallinity with increasing temperature as revealed by XRD analysis. Heat treatment beyond 1100 °C indicated a gradual decline in their respective band gap values at 1200 °C and 1300 °C and this inference is mainly attributed to the emergence of r-TiO₂.³⁸ An increasing trend in the band gap was also noticed on increasing SiO₂ content in the TiO₂–SiO₂ binary oxides. This trend in band gap on increasing SiO₂ content is also supported from XRD analysis which revealed the increase in highly crystalline a-TiO₂ on increasing SiO₂ content.³⁹ There was also increasing absorbance in UV-Visible region on increasing SiO₂ content for calcined at 900 °C, 1000 °C, 1100 °C and 1200 °C. Nevertheless, the composites heat treated at 1300 °C demonstrated decrease in absorbance on increasing SiO₂ content.

Table 4 Calculated band gap for the TiO₂–SiO₂ binary oxides after heat treatment at different temperatures

Sample code	Band gap (Eg, eV)
	900 °C	1000 °C	1100 °C	1200 °C	1300 °C
0.50 TS	3.12	2.94	3.03	2.93	2.93
0.75 TS	3.14	2.96	3.08	2.95	2.94
1.00 TS	3.16	3.05	3.16	3.02	2.95
1.25 TS	3.17	3.11	3.17	3.11	2.94
1.50 TS	3.18	3.13	3.18	3.13	3.01

---

3.5 PL spectra

PL spectra (Fig. 9) were recorded to examine the photoelectric properties of the pure TiO₂ and five different TiO₂–SiO₂ binary oxides after excited at 425 nm at room temperature. The PL spectra of pure TiO₂ demonstrate three major peaks about 595, 718 and 835 nm for all the heat treatment temperatures above 400 °C, which reveal the characteristic photoemission of r-TiO₂. The preservation of a-TiO₂ due to the added SiO₂ had reflected in the PL spectra recorded for all the TiO₂–SiO₂ binary oxides after heat treatment at higher temperatures (Fig. 9). The intensity of the emission peaks had become more intense with the increasing calcination temperatures. PL spectra of TiO₂–SiO₂ binary complex heated at 900 °C consist of three major peaks at 560, 674 and 782 nm and three minor peaks at 541, 727 and 754 nm and these observed peaks had revealed a blue shift from pure TiO₂ heated at same temperature. The photoemission recorded for 1000 °C also revealed blue shift in emission from pure TiO₂ with three major peaks identified at 584, 704 and 816 nm and three minor peaks observed at 527, 759 and 787 nm. A similar trend of blue shift also occurred at 1200 °C with three major peaks determined at 583, 703 and 815 nm and three minor peaks noticed at 528, 757 and 785 nm. A noticeable change in the range of blue shift was observed for the heat treated samples at 1300 °C, however, with the observed shift was not the same for all the samples. As the content of SiO₂ had increased, the shift away from the photoemission peaks of pure TiO₂ was found more prominent at 1300 °C. The observed inferences from the PL spectra could be interpreted on the following ways. The results from the quantitative XRD analysis reveals the stabilization of single phase a-TiO₂ till 1100 °C and its heat treatment beyond 1100 °C had resulted in the gradual conversion of a-TiO₂ to r-TiO₂ which were solely dependent on the SiO₂ content and heat treatment temperatures. Comparison of the quantitative data between 1200 and 1300 °C reveals the presence of major content of r-TiO₂ at 1300 °C for all the samples with no traces of a-TiO₂ detected specially for TS 0.50 and TS 0.75 which had resulted in the complete r-TiO₂ formation. These observed results reflected in the PL spectra for TS 0.50 and TS 0.75 samples recorded at 1300 °C, which had initiated their red shift on comparison with their corresponding results observed at 1200 °C.

Fig. 9 PL spectra recorded for Pure TiO₂ and five different TiO₂–SiO₂ binary oxides after heat treatments.

3.6 Antibacterial tests

The efficiency of pure TiO₂ and three different TiO₂–SiO₂ binary oxides in countering the microbes were analyzed in a quantitative manner. Initial testing done by using 0.1 g and 0.2 g of the powder samples has not indicated any effective reduction in microorganism colonies. Increasing the concentration to 0.5 g has started to show some effective reduction in microorganism colonies. The performance of experiments by increasing the concentration of the powder to 1.0 g indicated the reduction of some viable cells of both microorganism species (Table 5). The overall reduction in cell number was confirmed from the Table 5 values and this activity is mainly attributed to the adhesion of microorganism cell to the particles of tested samples. The adhesion of TiO₂ particles to the bacterial cell wall facilitated the damaging effect to the cell either by cell wall plasmolysis or the separation of cytoplasm from their cell wall. The data from Table 5 signifies the fact that S. aureus strain is less susceptible to the entire powder samples than E. coli and antimicrobial efficacy of the materials is function of TiO₂ concentration. It was also confirmed the Table 5 values that as there is an enhancement in the concentration of SiO₂ in TiO₂–SiO₂ binary oxides, their corresponding efficiency in countering the microbes started to get reduced. Further, the antibacterial efficiency started to get reduced with time and the presence of higher content of SiO₂ indicated less efficiency as the time proceeds. These inferences confirmed the negligible role or inefficiency of SiO₂ against antibacterial action. The mechanism of antibacterial action of TiO₂ has been reported before in which some researchers had suggested the photo-catalytic degradation mechanism for killing of bacterial cells by using TiO₂.^40,41 Blake et al., reported that TiO₂ had the capacity to damage the cell membrane by photo-catalytic degradation of endotoxin from E. coli.⁴² They also suggested that there is bacterial cell lysis by leaking out of cytoplasm from E. coli after UV irradiation of TiO₂. However, the experiments in the present study were not done with the assistance of UV irradiation and hence the E. coli bacteria have shown less antimicrobial activity.

Table 5 Antimicrobial efficiency of pure TiO₂ and different concentration of TiO₂–SiO₂ binary oxides

Sample	Microorganism	Time (h)				R% (h)
		0	1	2	4	1	2	4
	Staph. aureus	2.0 × 10^6
Pure TiO2			7.2 × 10^3	4.7 × 10^4	3.3 × 10^5	99.6	97.6	83.5
TS 0.5			3.3 × 10^4	8.4 × 10^4	4.9 × 10^5	98.4	95.8	75.5
TS 1.0			5.8 × 10^4	3.7 × 10^5	8.3 × 10^5	97.1	81.5	58.5
TS 1.5			6.4 × 10^4	5.2 × 10^5	8.7 × 10^5	96.8	71.0	56.5
	E. coli	2.3 × 10^6
Pure TiO2			8.7 × 10^3	5.1 × 10^4	3.7 × 10^5	99.6	97.8	83.9
TS 0.5			3.5 × 10^4	9.1 × 10^4	6.1 × 10^5	98.5	96.0	73.5
TS 1.0			7.3 × 10^4	3.4 × 10^5	8.7 × 10^5	96.8	85.2	62.2
TS 1.5			7.5 × 10^4	6.7 × 10^5	9.8 × 10^5	96.7	70.9	53.4

---

4. Summary

The syntheses of five different TiO₂–SiO₂ binary oxides along with the pure TiO₂ without the added SiO₂ have been done using sol–gel technique. Pure TiO₂ without the addition of any SiO₂ precursor led to the formation of a-TiO₂ at 430 °C whereas its complete transformation from a-TiO₂ to r-TiO₂ occurred at 700 °C. The characterization results confirmed the major influence of added SiO₂ on the phase behavior of TiO₂ in which pure a-TiO₂ phase was preserved till the heat treatment temperature of 1100 °C. Heat treatment beyond 1100 °C resulted in the gradual phase transition of a-TiO₂ to r-TiO₂ and their quantitative yield were mainly dependent on the three major factors such as the content of SiO₂, heat treatment temperatures and crystallite size effect. The SiO₂ content remained as an amorphous phase till 1200 °C and its crystallization in the form of c-SiO₂ occurred at 1300 °C. However, the quantitative yield of c-SiO₂ was dependent on the amount of a-TiO₂ in the composites but not found to be consistent with respect to the added SiO₂ precursor. The results obtained from optical studies were also found consistent with the quantitative analysis in which the SiO₂ stabilized a-TiO₂ induced a blue shift in emission spectra. However, a less pronounced blue shift was observed at 1300 °C, which was mainly attributed to the considerable presence of r-TiO₂ in the samples. The results from the antibacterial tests indicated less efficiency of TiO₂–SiO₂ binary oxides on comparison with the pure TiO₂.

Acknowledgements

The authors are thankful to the financial support received from University Grants Commission, India [F. no. 42-901/2013 (SR) dated 25.03.2013]. The Instrumentation facility used from the Central Instrumentation Facility (CIF) of Pondicherry University is greatly acknowledged.

References

1. H. J. Rack and J. I. Qazi, Titanium alloys for biomedical applications, Mater. Sci. Eng., C, 1989, 26, 1269–1277.
2. C. J. Chung, H. I. Lin, P. Y. Hsieh, K. C. Chen, J. L. He, A. Leyland and A. Matthews, Growth behavior and microstructure of arc ion plated titanium dioxide, Surf. Coat. Technol., 2009, 204, 915–922.
3. J. Park, S. Bauer, M. K. Vonder and P. Schmuki, Nanosize and vitality: TiO₂ nanotube diameter directs cell fate, Nano Lett., 2007, 7, 1686–1691.
4. K. S. Brammer, S. Oh, J. O. Gallagher and S. Jin, Enhanced cellular mobility guided by TiO₂ nanotube surfaces, Nano Lett., 2008, 8, 786–793.
5. V. K. Balla, P. D. De Vas Con Cellos, W. Xue, S. Bose and A. Bandyopadhyay, Fabrication of compositionally and structurally graded Ti–TiO₂ structures using laser engineered net shaping (LENS), Acta Biomater., 2009, 5, 1831–1839.
6. C. F. Huang, H. C. Cheng, C. M. Liu, C. C. Chen and K. L. Ou, Microstructure and phase transition of biocompatible titanium oxide film on titanium by plasma discharging, J. Alloys Compd., 2009, 476, 683–688.
7. J. Shi, J. Chen, Z. Feng, T. Chen, Y. Lian, X. Wang and C. Li, Photoluminescence characteristics of TiO₂ and their relationship to the photoassisted reaction of water–methanol mixture, J. Phys. Chem. C, 2007, 111, 693–699.
8. G. Madras, B. J. McCoy and A. Navrotsky, Kinetic model for transformation from nano-sized amorphous TiO₂ to anatase, J. Am. Ceram. Soc., 2007, 90, 250–253.
9. K. Ding, Z. Miao, B. Hu, G. An, Z. Sun, B. Han and Z. Liu, Study on the anatase to rutile phase transformation and controlled synthesis of rutile nanocrystals with the assistance of ionic liquid, Langmuir, 2010, 26, 10294–10302.
10. J. T. Chang, C. W. Su and J. L. He, Photocatalytic TiO₂ film prepared using arc ion plating, Surf. Coat. Technol., 2006, 200, 3027–3034.
11. C. J. Chung, H. I. Lin and J. L. He, Microstructural effect on the antimicrobial efficacy of arc ion plated TiO₂, J. Mater. Res., 2007, 22, 3137–3143.
12. X. Ding and X. Liu, Correlation between anatase-to-rutile transformation and grain growth in nanocrystalline titania powders, J. Mater. Res., 1998, 139, 2556–2559.
13. J. Ovenstone and K. Yanagisawa, Effect of hydrothermal treatment of amorphous titania on the phase change from anatase to rutile during calcination, Chem. Mater., 1999, 11, 2770–2774.
14. W. Li, C. Ni, H. Lin, C. P. Huang and S. I. Shah, Size dependence of thermal stability of TiO₂ nanoparticles, J. Appl. Phys., 2004, 96, 6663–6669.
15. C. C. Wang and J. Y. Ying, Sol–gel synthesis and hydrothermal processing of anatase and rutile titania nanocrystals, Chem. Mater., 1999, 11, 3113–3112.
16. G. Calleja, D. P. Serrano, R. Sanz and P. Pizarro, Mesostructured SiO₂-doped TiO₂ with enhanced thermal stability prepared by a soft-templating sol–gel route, Microporous Mesoporous Mater., 2008, 11, 429–436.
17. C. Kang, L. Jing, T. Guo, H. Cui, J. Zhou and H. Fu, Mesoporous SiO₂-modified nanocrystalline TiO₂ with high anatase thermal stability and large surface area as efficient photocatalyst, J. Phys. Chem. C, 2009, 113, 1006–1013.
18. J. Kim, K. C. Song, S. Foncillas and S. E. Pratsinis, Dopants for synthesis of stable bimodally porous titania, J. Eur. Ceram. Soc., 2001, 21, 2863–2872.
19. J. H. Pan and W. I. Lee, Preparation of highly ordered cubic mesoporous WO₃/TiO₂ films and their photocatalytic properties, Chem. Mater., 2006, 18, 847–853.
20. Y. Lv, L. Yu, H. Huang, H. Liu and Y. Fen, Preparation of F-doped titania nanoparticles with a highly thermally stable anatase phase by alcoholysis of TiCl₄, Appl. Surf. Sci., 2009, 255, 9548–9552.
21. L. Xuanyong, K. C. Paul and D. Chuanxian, Surface modification of titanium, titanium alloys, and related materials for biomedical applications, Mater. Sci. Eng., R, 2004, 47, 49–121.
22. L. L. Hench and J. K. West, Biological Applications of Bioactive Glasses, Life Chem. Rep., 1996, 13, 187–241.
23. W. Y. Dong, Y. J. Sun, C. W. Lee, W. M. Hua, X. C. Lu, Y. F. Shi, S. C. Zhang, J. M. Chen and D. Y. Zhao, Controllable and repeatable synthesis of thermally stable anatase nanocrystal-silica composites with highly ordered hexagonal mesostructures, J. Am. Chem. Soc., 2007, 129, 13894–13904.
24. G. Calleja, D. P. Serrano, R. Sanz and P. Pizarro, Mesostructured SiO₂-doped TiO₂ with enhanced thermal stability prepared by a soft-templating sol–gel route, Microporous Mesoporous Mater., 2008, 11, 429–440.
25. Q. H. Zhang, L. Gao and S. Zheng, Preparation of mesoporous TiO₂ photocatalyst by selective dissolving of titania–silica binary oxides, Chem. Lett., 2001, 11, 1124–1125.
26. C. He, B. Tian and J. Zhang, Thermally stable SiO₂-doped mesoporous anatase TiO₂ with large surface area and excellent photocatalytic activity, J. Colloid Interface Sci., 2010, 344, 382–389.
27. A. C. Larson, R. Von Dreele and R. Blanl, General Structure Analysis System Report LAUR Los Alamos National Laboratory, 2004, 86, p. 748.
28. B. H. Toby, EXPGUI, a graphical user interface for GSAS, J. Appl. Crystallogr., 2001, 34, 210.
29. C. J. Howard, Z. M. Sabine and F. Dickson, Structural and thermal parameters for rutile and anatase, Acta Crystallogr., Sect. B: Struct. Sci., 1991, 47, 462.
30. P. Dera, J. D. Lazarz, V. B. Prakapenka, M. Barkley and R. T. Downs, New insights into the high-pressure polymorphism of SiO₂ cristobalite, Phys. Chem. Miner., 2011, 38, 517–529.
31. J. R. Connolly, Introduction to X-ray Powder Diffraction for EPS 400, Springer, 2012, pp. 1–12, http://epswww.unm.edu/xrd/xrdclass/09-Quant-intro.pdf.
32. V. Stanic, S. Dimitrijevi, J. Anti-Stankovi, M. Mitri, B. Joki and S. Raicevi, Synthesis, characterization and antimicrobial activity of copper and zinc-doped hydroxyapatite nanopowders, Appl. Surf. Sci., 2010, 256, 6083–6090.
33. CRC Handbook of Chemistry and Physics, ed. Lide D. R., Taylor and Francis, Boca Raton, FL, 2006, Internet Version, http://wwwhbcpnetbasecom.
34. J. Lu, K. M. Kosuda, R. P. Van Duyne and P. C. Stair, Surface Acidity and Properties of TiO₂/SiO₂ Catalysts Prepared by Atomic Layer Deposition: UV-Visible Diffuse Reflectance, DRIFTS, and Visible Raman Spectroscopy Studies, J. Phys. Chem. C, 2009, 113, 12412–12418.
35. M. Jin, X. Zhang, A. V. Emeline, Z. Liu, D. A. Tryk, T. Murakami and A. Fujishima, Fibrous TiO₂–SiO₂ nanocomposite photocatalyst, Chem. Commun., 2006, 4483–4485.
36. R. J. Davis and Z. Liu, Titania–Silica:  A Model Binary Oxide Catalyst System, Chem. Mater., 1997, 9, 2311–2324.
37. K. Joseph Antony Raj, Y. R. Smith, V. Subramanian and B. Viswanathan, Structural studies of silica modified titania and its photocatalytic activity of 4-chlorophenol oxidation in aqueous medium, Indian J. Chem., Sect. A: Inorg., Bio-inorg., Phys., Theor. Anal. Chem., 2010, 49, 867–875.
38. A. Lukowiak, R. Dylewicz, S. Patela, W. Strek and K. Maruszewski, Optical properties of SiO₂–TiO₂ thin film waveguides obtained by the sol–gel method and their applications for sensing purposes, Opt. Mater., 2005, 27, 1501–1505.
39. I. V. Krivtsov, M. V. Ilkaeva, V. D. Samokhina, V. V. Avdin, S. A. Khainakov, D. A. Uchaev and J. R. Garcia, Synthesis of silica–titania composite oxide via “green” aqueous peroxo-route, J. Sol-Gel Sci. Technol., 2013, 67, 665–669.
40. Z. Lu, L. Zhou, Z. Zhang, W. Shi, Z. Xie, H. Xie, D. Pang and P. Shen, Cell damage induced by photocatalysis of TiO₂ thin films, Langmuir, 2003, 19, 8765–8768.
41. K. Sunada, Y. Kikuchi, K. Hashimoto and A. Fuhishima, Bactericidal and detoxification effects of TiO₂ thin film photocatalysts, Environ. Sci. Technol., 1998, 32, 726–728.
42. M. D. Blake, P. Manesss, Z. Huang, J. E. Wolfrun and J. Huang, Application of the photocatalytic chemistry of titanium dioxide to disinfection and the killing of cancer cells, Sep. Purif. Methods, 1999, 28, 1–50.

---