Preparation of the anatase phase TiO₂ nanocrystallites using subcritical water as the solvent and evaluation of their photocatalytic properties under visible light irradiation

2.1 Synthesis method

The commercial P25 used in this work was produced by Evonik, Germany. Other chemical reagents were purchased from Sinopharm Chemical Reagent Co., Ltd., China. They are of analytical grade and used without further purification. The TiO₂ nanocrystallites were prepared by the hydrolysis method of the titanium sulfate using subcritical water as the solvent. First, 20 g Ti(SO₄)₂ were dissolved in 200 ml of deionized water at room temperature, after which the solution was placed inside an autoclave and heated to experimental temperatures (573–623 K). The operating pressure ranged from 12 to 16 MPa. Figure 1 shows the schematic diagram of the hydrolysis experiment. Afterwards, the mixture was centrifuged at 12,000 rad/min to separate the particles from the reaction medium. The centrifuge (Avanti Model JXN-30) was produced by Backman Kurt, USA. The obtained powders were washed by absolute ethyl alcohol thrice to remove the sulfate ions on the surface of the powders and then dried in an oven at 343 K for 2 h. The obtained powders were calcined at various temperatures for 2 h in ambient temperature to obtain anatase phase TiO₂ nanocrystallites with high crystallinity.

Figure 1:

The schematic diagram of the hydrolysis experiment.

2.2 Design of the experiments

Several factors can influence the crystallite size of the products. To optimize the growth process of the TiO₂ nanocrystallites and to evaluate the interactions between the factors, surface response methodology was used in the present work. The pressure of the solution, calcination temperatures, calcination time and hydrolysis temperature varied from 12 to 16 MPa, 873–1193 K, 40–120 min and 573–613 K, respectively.

2.3 Sample characterization

The morphology of the prepared powders was determined using high-resolution transmission electron microscopy (HRTEM; KEM-ARM200F, JEOL Ltd. Tokyo, Japan). For this, a 2.5 wt.% suspension of the powder was dispersed in methanol via 3 min of ultrasound (FS-450, Shenxi Ultrasound Instruments, Shanghai, China) and was deposited on a carbon-coated copper grid (Zhongjing Scientific Instrument, Beijing, China). The sample was then dried in an oven at 323 K for 20 min before loading into a single tilt holder. The HRTEM, which was also used to obtain the particle lattice fringes, was operated at 200 kV. Field emission scanning electron microscope (FESEM, ULTRA55, Carl Zeiss Corporation), operated at 20 kV, as also used to observe the morphologies of samples. The sizes of the products were determined by a laser particle analyzer (Zetasizer Nano S90, Malvern Instruments Ltd., UK). The particles were dispersed by ultrasonic dispersion technique in order to avoid particle aggregation before determination.

The samples were characterized using a Philip X pert machine (model: MPDDY2094, the Netherlands) with copper Kα irradiation (λ=1.5406 nm), the operating voltage was 30 kV and the scanning angle ranged from 10 to 90 degrees. The crystallite sizes of the prepared TiO₂ nanocrystallites are calculated by using the equation

where K is the Scherrer constant, β is the width of the half height of the diffraction peak, λ is the wavelength of incident light, θ is the Bragg angle and D is the crystallite size of TiO₂ particles. The uncertainty value is ±1 nm.

2.4 Photocatalytic activity test

The photocatalytic degradation of RB using the TiO₂ nanocrystallites was carried out in a glass reactor. A 40-W lamp was used as the visible light source. The pH values of the RB solution were adjusted by the HCl (5 wt.%) or NH₃×H₂O (3%) solution to the desired values, after which the TiO₂ photocatalysts were added into the solution. Prior to degradation, the suspension was stirred in the presence of air. With the photocatalytic reaction, about 1-ml aliquots were withdrawn from the solutions at intervals of 5–10 min and subsequently centrifuged at 14,000 rpm for 8 min. A 721 spectrophotometer (Shanghai Precision Scientific Instrument Co., Ltd., China) was used to test the absorbance of the degraded solution. The degradation of the RB can be calculated by using the equation

where η is the degradation rate of RB, A₀ is the initial absorbance of the RB solution and A_t is the absorbance of RB after degradation.

In the photodegradation process, 200 ml of suspension was placed in a glass reactor with a volume of 350 ml. The pH range varied from 2 to 12 and the TiO₂ concentrations ranged from 0.01 to 0.025 g/l, corresponding to RB concentrations ranging from 50 to 150 mg/l. The degradation time and temperature were varied from 10 to 90 min and from 298 to 338 K, respectively.

3.1 Formation and growth of the anatase phase TiO₂ nanocrystallites

Figure 2A and B present the XRD analysis results of the obtained TiO₂ nanocrystllites before and after calcination, respectively. All the samples are obtained at different hydrolysis temperatures from 308 to 338 K, whereas the pressure of the solution was kept at 15 MPa. As for the calcined sample in Figure 2B, the calcination temperature was 1193 K, and the calcination time was 2.5 h. As shown in Figure 2A and B, all the products are composed of anatase, indicating the formation of the TiO₂ nanocrystallites during the hydrolysis process, and that the crystallinity of the TiO₂ nanocrystallites increases with the increase of hydrolysis temperatures. Ramaujam et al. [21] have investigated the formation of YAG crystals and found that the crystals are generated in the supercritical water, which is in agreement with our results. Hertz et al. [24] investigated the preparation of the nanocrystalline TiO₂ powders in supercritical carbon and reported that anatase powders with high specific surface areas are obtained directly in the CO₂ solvent under the supercritical conditions at temperatures as low as 523 K. However, compared with the calcined sample, there are some low-intensity peaks in the non-calcined products, indicating that the crystallinity of the TiO₂ nanocrystallites can be enhanced effectively by the calcination process. Figure 3 shows the FESEM, HRTEM micro-morphologies and laser particle size analysis results of the product analyzed in Figure 2A. The particle size of the product ranges from 40 nm to 60 nm. It is interesting to note that the obtained particle size using XRD technology is different from that determined using FESEM and TEM. This difference can be attributed to the XRD as the only suitable method to measure the particles with a high degree of crystallinity; hence, similar results can be seen in the reference [21]. The laser particle size analysis results show that the average particle size is 63.78 nm. Compared with the FESEM and HRTEM results, we can confirm that the particles do not agglomerate. According to the XRD results in Figure 2A, the images clearly show the structure of the TiO₂ nanocrystallites, with the fringes corresponding to anatase (101) exhibiting a distance of 0.335 nm. As summarized by the above analysis results, it can be confirmed that the TiO₂ nanocrystallites are generated in the hydrolysis process but grow in the calcianation process. Furthermore, the crystallinity of the products is significantly increased after calcination.

Figure 2:

The X-ray diffraction results of the TiO₂ nanocrystallites before and after calcination. (A) Before calcination, (B) after calcination.

Figure 3:

The FESEM and HRTEM micro-morphologies of the product shown in Figure 2A. (A) FESEM image, (B and C) HRTEM images, (D) laser particle size analysis result.

3.2 Investigations on the crystallite size of the TiO₂ nanocrystallites

The crystallite size of the TiO₂ nanostructured materials can be determined by two steps: (i) growth in the hydrolysis process of the starting materials and (ii) growth in the calcination process of the intermediate products. Several factors can affect the crystallite size of the TiO₂ materials, such as hydrolysis temperature, calcination temperature, pressure of the solution and calcination time. In addition, the interactions among the abovementioned factors are crucial to the growth of the TiO₂ nanocrystallites. The response surface methodology is considered an ideal method of investigation in evaluating the interactions among these factors. In the present work, the central composite design (CCD) method was used to predict the crystallite size of TiO₂ materials. The experimental times can be represented as [25], [26], [27], [28]

where N is the experimental times, n is the number of variables and c₀ is the central point of the design.

Table 1 shows the four-factor design table for the CCD experiments. The experimental data were analyzed by Design-experiment version 8.0 software. Table 2 presents the experimental data and running results.

According to the above fitting results, the regression equation is given by

Figure 4 shows the relationship between the predicted and experimental values. The R square is 0.987, indicating that the predicted results are in good agreement with experimental results.

Figure 4:

The relationship between the predicted and experimental values.

In order to further determine the significance of each factor, according to the CCD experimental design method, we use p to evaluate the significance of each factor. According to the theory of surface response methodology, if p<0.05, this indicated that the factor has significant influence on the growth of the TiO₂ nanocrystallites [25], [26], [27]. The results are shown in Table 3.

From Table 3, we know that the growth of the TiO₂ nanocrystallites is mainly determined by hydrolysis temperature, calcination time and pressure of the solution. However, the interactions among the factors are quite strong, including those of the calcination temperature and time, hydrolysis temperature and pressure of the solution as well as the calcination temperature and pressure of the solution. For the degradation process of the organic compounds, a smaller crystallite size means higher photocatalytic activities. Thus, the optimal conditions for the preparation of TiO₂ nanocrystallites obtained are as follows: hydrolysis temperature of 603 K, calcination temperature of 953 K, calcination time of 60 min and pressure of the solution 13 MPa. Under this condition, the predicted result is 21.32 nm. The practical experiments with the optimal conditions are carried out in triplicate, and the average particles size is 21.38 nm.

To further visualize the interactions among the abovementioned factors, we draw the contour line maps and response surfaces using the Design expert software. Figure 5A and B present the contour line maps and corresponding response surfaces for the growth of TiO₂ particles, respectively. Figure 5A and B also present the interaction effects of the hydrolysis temperature and the calcination temperature on the crystallite size of TiO₂, respectively. The crystallite size is very sensitive to calcination temperature, that is, the crystallite size of TiO₂ material increased with the decreasing calcination temperature. The hydrolysis products contain moisture, which can be divided into three forms: adsorbed water, crystalline water and deep crystallization water in the crystal. When the product is heated, the volume of the deep crystallization water in the crystals expand. The results presented by Li et al. [29] show that the crystal strcture of the nanomaterials can be easily ruptured by the volume expantion, which makes the crystallite size smaller. This result is in agreement with the results of the current work.

Figure 5:

The contour line map and response surface for the growth of the TiO₂ nanocrystallites. (A) Contour line map, (B) response surface.

3.3 Degradation of Rhodamine B

The synthetic products under the above optimal condition were used to evaluate the effects of the important parameters (photocatalysts dosage, pH, concentration of RB and temperature) on the RB photodegradation. Figure 6 shows the RB degradation using different catalysts. As can be seen, the prepared photocatalysts exhibit better photocatalytic properties than commercial P25. The commercial P25 catalyst is composed of both the anatase and rutile phases TiO₂. Silva and Faria [30] indicated that the photocatalytic activity of rutile is lower than that of the anatase phase TiO₂ materials, which is in agreement with the results shown in Figure 6. In addition, the degradation rate of RB is relatively lower in the dark, indicating that the absorption of the powders is limited and that the visible light plays a key role in the degradation process.

Figure 6:

The degradation and fitting results of the RB using different catalysts.

Figure 7A shows the effect of photocatalyst dosage on the RB degradation and the fitting results of the reaction rate constant (other parameters are set as follows: pH: 7, RB concentration: 100 mg/l, degradation temperature: 328 K). The fitting results of the reaction rate constant indicate that the photocatalytic reactions are enhanced effectively by increasing the TiO₂ dosage, that is, the reaction rate is proportional to the dosage of the TiO₂ powders. However, the reaction rate decreases when the dosage of TiO₂ is more than 0.021 g/l, simply because the excessive dosage of the TiO₂ powder can lead to the poor mass transfer efficiency of the visible light. In Figure 7B, the experimental pH varies from 3 to 11 (other parameters are set as follows: dosage of photocatalysts: 0.021 g/l, RB concentration: 100 mg/l, degradation temperature: 328 K). The degradation reaction rates of RB increased when the pH increases from 5 to 9. As known to all, there is a carboxy in the RB molecular; thus, the pH of the RB solution is 3.92. A larger pH means higher hydroxide ion concentration in the solution, which can break the carboxy easily, which is in favor of the RB degradation. In addition, our experimental results indicate that RB would be degraded into methanoic acid and then neutralized when the pH is larger than 7.0, which in turn, promotes the reaction rate effectively. However, the isoelectric point of TiO₂ is 6.3; when the pH deviates from this value too large, the surface of the TiO₂ nanoparticles exhibits a positive charge. This phenomenon leads to a repulsive interaction between the RB molecular and TiO₂ nanoparticles, which in turn, prohibits the smooth completion of the RB photodegradation process. This is why the reaction rate decreases when the pH value varies from 9 to 11. In addition, the RB is degraded more completely under higher pH values. The degradation rate of RB is 77.36% with the pH of 3, whereas it reached up to 88.73% with the pH of 9, indicating that the optimal degradation pH for RB is 9. As for Figure 7C (other parameters are set as follows: dosage of photocatalysts: 0.021 g/l, pH: 9, degradation temperature: 328 K), the degradation rates decrease with the increased concentration of RB. In addition, a low concentration (80 mg/l) of the RB solution can be degraded by more than 90% when the degradation time is 90 min, whereas it is only 69.23% with the concentration of 120 mg/l. In Figure 7D (other parameters are set as follows: dosage of photocatalysts: 0.021 g/l, pH: 9, RB concentration: 100 mg/l), the degradation rate of RB increases remarkably with the reaction temperature. The reaction rate increased from 0.0082 to 0.00985 when the temperature ranges from 298 to 328 K, indicating that a high temperature induces degradation. The effect of temperature on the reaction rate constant can be described by the Arrhenius equation [11]. The fitting results according to the Arrhenius equation are shown in Figure 6D, in which the R square is 0.9913. According to the results, the activation energy is 4.076 kJ/mol, indicating that the photocatalytic reactions of RB can proceed easily under visible light.

Figure 7:

The effects of parameters on the degradation of RB and the fitting results of the reaction rate constant. (A) Photo-catalyst dosage, (B) pH, (C) concentration of the RB, (D) temperature.

The L-H model has been widely used to describe the kinetics of the degradation process of organic compounds. The kinetics rate expression in the L-H model can be presented as [31], [32], [33]:

where r₀ is the reaction rate of RB, c₀ is the initial concentration of the RB solution, k is the reaction rate constant and K_S is the adsorption equilibrium constant between RB and photocatalysts.

The adsorption between the RB molecular and photocatalysts is much weaker than the photocatalytic reactions, implying that K_Sc₀=1. Consequently, Eq. (5) can be written as

where r is the reaction rate, k₁ is the reaction rate constant for first order reaction and c is the concentration of RB. The following equation can be deduced from Eqs. (5) and (6):

According to the L-H model, we determine the relationship between −ln(c/c₀) and reaction time using the linear method (shown in Figure 7D). Here, the linear dependent coefficient is 0.99445, which indicates that the L-H model is feasible in describing the RB the degradation. The kinetic equation is given by

In addition, according to the obtained kinetic equation, the reaction rate constant is 0.2248.