Synthesis of Nanostructured TiO₂ Microparticles with High Surface Area

Abstract

Hydrothermal reactions represent a simple and efficient method for the preparation of nanostructured TiO₂ particles that could be of interest as photocatalysts or catalytic supports. Although the particle size is in the range of 2–5 µm, the nanostructures composing the particles ensure a large specific surface area with values above 100 m²/g. The effects of the different synthesis parameters on the morphology, photocatalytic activity, and stability of the prepared material were studied. The surface morphology of the prepared TiO₂ powders was studied by scanning electron microscopy (SEM). To further characterize the samples, the specific surface area for different morphologies was measured and the photocatalytic activity of the prepared powders was tested by degrading model pollutants under UV irradiation. The results show that the initial morphology had little effect on the photocatalytic properties. On the other hand, the final calcination temperature significantly increased the degradation rates, making it comparable to that of P25 TiO₂ (particle size 20–30 nm).

1. Introduction

Although TiO₂ photocatalysis for water treatment has been extensively researched in recent decades [1,2,3,4,5,6,7,8,9], its use in practice is still limited. This is partially due to the lack of efficient reactor systems that can treat large volumes of water economically and in a timely manner [1,10,11].

An important parameter for photocatalysts is their specific surface area. To increase it, the particle size is usually reduced, with the most active TiO₂ photocatalysts having particle sizes in the range of 10 to 30 nm. When used as a photocatalyst, the best results are obtained when the particles are suspended in a solution. However, an additional step is then required to separate the particles from the solution after the reaction. Since filtration of nanoparticles is difficult and expensive, larger particles with a morphology that promotes a high specific surface area are desirable. This would theoretically allow the construction of a flow through slurry reactor that could combine the efficiency of suspended particles with the ease of use of an immobilized photocatalyst [12,13,14,15,16,17].

Hydrothermal reactions represent a simple and efficient method for the preparation of nanostructured TiO₂ particles that could be of interest as photocatalysts or catalyst supports. Although there are many papers on the hydrothermal synthesis of TiO₂, few of them investigate the effects of a strongly alkaline solution on the morphology of the prepared materials. Some research papers addressed the modification of the surface of already prepared TiO₂ particles, where they successfully prepared nanotubes and nanoparticles as powders or deposited them on various supports. Preparation of larger particles has also been explored from already prepared TiO₂ or metallic titanium [18,19,20,21,22]. However, to our knowledge, the effect of synthesis parameters and calcination temperature on the morphology and photocatalytic activity of TiO₂ materials prepared from fully dissolved Ti ions in an alkaline solution by hydrothermal methods has not been fully explored.

In this work, we present a simple method to prepare nanostructured TiO₂ particles. Although the particle size is in the range of 2–5 µm, the nanostructures composing the particles ensure a large specific surface area with values above 100 m²/g. The powders were prepared by a hydrothermal method in a strongly alkaline solution. The nanomorphology of the larger particles can be easily controlled by adjusting the reaction parameters. By optimizing key parameters, we were able to increase the yield of the reaction compared to previously reported results. The morphology of the prepared samples is stable and does not change even after calcination at up to 500 °C. The short reaction times, scalability, and high titanium concentrations in the reaction solution also allow the production of larger quantities of the powders.

2. Results

2.1. Optimizing Synthesis

Although methods to prepare flower-like TiO₂ particles by a hydrothermal process in an alkaline solution have been described previously, they were complicated and had low yields while often requiring days of mixing [21]. By optimizing the preparation of the starting solution, we were able to increase the amount of titanium dissolved while reducing the time required to prepare the precursor solution to less than 30 min.

The goal in preparing the initial solution was to obtain the maximum concentration of titanium ions that could still be completely dissolved. The actual mechanism of dissolution is still unknown, but from the experiments, it appeared that the addition of H₂O₂ at the beginning was crucial. However, the addition of more H₂O₂ to the initial solution had no effect on the morphology or photocatalytic activity though.

The use of titanium alkoxides as precursors is also possible. It does not affect the final result, but it significantly increases the time required to prepare the starting solution.

The yield of the reaction was between 80% and 90% after washing and calcination. The scale up of the synthesis was tested by increasing the volume of the starting solution to 250 mL in a 300 mL autoclave, with no difference observed in the samples prepared. This suggests that larger quantities can be readily prepared.

A possible pathway for the formation of TiO₂ is shown in Figure 1. First, sodium metatitanate (Na₂TiO₃) was obtained from the hydrothermal reaction. Washing with 0.1 M HCl replaced the sodium ions with hydrogen, giving metatitanic acid (H₂TiO₃). During the final calcination, metatitanic acid decomposed to TiO₂ and H₂O.

2.1.1. Effect of Preparation Conditions on Morphology and Photocatalytic Activity of TiO₂ Particles

The morphology depends mainly on the parameters of the hydrothermal reaction. It is also important that all of the titanium precursor is dissolved in the solution before the reaction, otherwise completely different morphologies are formed. The size of the particles is 2–5 µm for the larger ones and about 0.5–1 µm for the smaller ones (Figure 2A) and is not significantly affected by changing the parameters, which allows the optimization of other properties while maintaining the ability to easily filter the particles after use.

Figure 2 shows SEM images of different samples (Supplementary Materials). From the results, it is evident that increasing the time or temperature of hydrothermal synthesis is interchangeable, while the amount of NaOH has a much more profound effect on the morphology. At low concentrations of NaOH, the particles consist of thicker fibres. When the concentration is increased to about 20 wt.%, the fibres are replaced by leaf-like structures, originating from the centre of the hemispherical particles. Further increase to 30 wt.%, retains the leaf-like structure, but the surface is covered with a thick mesh of very thin (−6 nm) fibres, resembling a spider web. This structure appears only on the surface of the particles and binds them together into larger clusters (Figure 2G).

The same thin mesh also appears at longer reaction times or higher temperatures. It can be assumed that the large particles are formed first during the reaction, but that these are partially dissolved during the course of the reaction, and the resulting material is deposited on the surface in the form of the thin fibres. This is confirmed when comparing the particles prepared at shorter times/lower temperatures, where the leaf-like structure is more densely packed compared to the others.

The thin mesh on the surface greatly increases the surface area of the material (

Table 1. Specific surface area values for the prepared samples. The sample designations are the same as in Section 3.1.

Sample	Surface Area [m2/g]
A	133 ± 0.7
B	135 ± 0.9
C	140 ± 0.9
D	121 ± 0.7
E	147 ± 0.8
F	114 ± 0.8
G	160 ± 1.4

) and the amount of adsorbed compounds on the surface. However, it does not increase the photocatalytic activity. This is probably due to the diameter of the fibres being too small, which results in an increased recombination rate, as the charge carriers cannot be separated efficiently [23].

Table 1 shows the results of the specific surface area measurements, which are in good agreement with the morphologies seen on the SEM images. Longer reaction times and higher reaction temperatures increase the surface area. The same effect is observed when the amount of NaOH is increased.

Due to the high specific surface area of the prepared samples, the adsorption of the pollutant on the surface has a great influence on the photocatalytic reaction. Therefore, the photocatalytic activity was tested in two steps by separating the adsorption and degradation reactions. All samples were first stirred in the dark to saturate the surface of the photocatalyst. Then, the suspension was exposed to UV illumination. Interestingly, adsorption was faster than photocatalytic degradation, which is why the reaction appears to be 0-order after the light is turned on, rather than 1st-order, as would be typical for photocatalysis.

Figure 3 shows the results of the photocatalytic experiments for different samples treated at 500 °C. For the first 35 min, the samples were stirred in the dark to determine the effect of adsorption on the concentration of the dye. After 35 min, the UV lights were turned on and photocatalytic degradation began. During the first 10 min of illumination, a steeper drop in concentration is observed, which is due to the presence of oxygen in the initial solution. After the oxygen is used up, a linear decrease in dye concentration follows.

It can be seen that the photocatalytic activity is comparable between the samples calcinated at the same temperature, as the slope of the curves in the linear part is very similar (

Table 2. Reaction constants from the linear part of photocatalytic degradation (γ(TiO₂) = 0.25 g/L).

Sample	k [µg L−1 min−1]
A	76 ± 5
B	75 ± 2
C	71 ± 2
D	75 ± 3
E	76 ± 6
F	82 ± 2
G	69 ± 2

), although the initial morphologies and surface areas are quite different. However, there is a clear difference in the amount of dye adsorbed, which correlates to some extent with the different surface areas of the samples.

It is also worth noting that the concentration of the photocatalyst (γ(TiO₂) = 0.25 g/L) is lower than usual, to obtain better measurements and to limit the adsorption on the surface. When the amount is increased to 1 g/L, most of the dye is adsorbed on the surface and the photocatalyst changes its color from white to gray. After 15 min of UV illumination, the samples return to their original color, as the adsorbed dye is degraded.

2.1.2. Effect of Thermal Treatment on Morphology and Photocatalytic Activity of TiO₂ Particles

To optimize the photocatalytic activity of the material, different calcination temperatures were tested and the effects on activity and morphology were determined. Figure 4 shows the results of XRD measurements for different treatment temperatures. At higher temperatures, the diffraction peaks become more intense and defined, indicating a higher degree of crystallinity in the samples. This in turn also affects the photocatalytic activity. All peaks visible on the diffraction patterns correspond to the anatase modification of TiO₂.

The morphology of the material remains unchanged up to 500 °C, after which some “melting” of the surface can be seen (Figure 5). Even though the specific surface area of the samples decreases with increasing calcination temperature (

Table 3. Specific surface areas and size of anatase crystallites of sample A, calcinated at different temperatures.

Sample	Surface Area [m2/g]	Crystalite Diameters [nm]
A-450	185 ± 1.2	10
A-500	133 ± 0.7	14
A-550	113 ± 0.8	18
A-600	93 ± 0.7	23
A-650	55 ± 0.3	31
A-700	39 ± 0.2	44

), the increase in photocatalytic activity is greater due to the higher crystallinity. The XRD measurements were also used to calculate the average size of nanocrystallites for the material treated at different temperatures (Table 3). As expected, higher temperatures increase the size of the crystallites.

Unlike the parameters of the hydrothermal synthesis, the temperature of the final thermal treatment has a great influence on the photocatalytic activity of the prepared material. Figure 6 shows the results of photocatalytic activity testing for sample A treated at different temperatures, starting from the end of the adsorption phase. The activity of the samples increases up to 650 °C (

Table 4. Reaction constants from the linear part of photocatalytic degradation for samples calcinated at different temperatures (γ(TiO₂) = 0.25 g/L).

Sample	k [µg L−1 min−1]
P25	96 ± 5
A-450	49 ± 5
A-500	76 ± 5
A-550	80 ± 9
A-600	90 ± 7
A-650	99 ± 4
A-700	95 ± 7

), which correlates with the increased crystallinity of the samples treated at higher temperatures.

Comparison of the slopes of the linear part for samples treated at 500 °C (k = 76 µg L⁻¹ min⁻¹) and the samples treated at 650 °C (k = 99 µg L⁻¹ min⁻¹) shows an increase of about 30% in photocatalytic activity. The amount of dye adsorbed on the surface decreases at higher treatment temperatures, which correlates with the lower surface area of the samples. To compare the prepared materials with other TiO₂ photocatalysts, the same photocatalytic activity tests were also performed with Aeroxide TiO₂ P25 from Evonik, which showed comparable activity to the samples treated at 650 °C or higher.

2.2. Investigation of Catalyst Stability

The reusability of the photocatalyst was tested for sample A. The results for the linear part of the photocatalytic degradation are presented in Figure 7. The amount of dye adsorbed on the surface decreases with each use, which could be due to insufficient removal of the dye between experiments. However, the degradation rate constants remain unchanged, with values ranging from 75 to 80 µg L⁻¹ min⁻¹, from which it can be concluded that the photocatalytic activity of the sample does not decrease with repeated use.

SEM image of sample A after repeated use is shown in Figure 8. While there is some damage to the particles, it is minor and likely the result of drying and scraping the sample after each use rather than the photocatalytic experiments themselves.

2.3. Investigation of the Degradation of Ibuprofen and Diclofenac Using the Prepared TiO₂

Since pharmaceuticals are also a critical problem in wastewater, we also tested the degradation of ibuprofen and diclofenac using sample A calcined at 500 °C. Figure 9 shows the results of the experiments. It can be seen that adsorption is minimal in these two cases, unlike the tests with the dye. This is due to the absence of electrostatic interactions which are present between plasmocorinth B (negative charge) and TiO₂ (positively charged surface at pH 4). The degradation itself is also faster compared to the dye and is completed in ~20 min for diclofenac and ~40 min for ibuprofen.

3. Materials and Methods

3.1. Synthesis

To determine and evaluate the effect of the synthesis parameters on the morphology and photocatalytic activity of the prepared powders, several samples were synthesized by modifying the basic synthesis as follows: 0.6 mL of H₂O₂ (30% solution) was added to 30 mL of dH₂O, then 0.24 mL of TiCl₄ was added dropwise while stirring to obtain a red solution; then 7.5 g of NaOH was added stepwise (~20 wt.%), after which we obtained a clear colorless solution. The addition of NaOH facilitates the dissolution of titanium ions. The solution was transferred to a Teflon-lined autoclave and heated at 150 °C for 3 h to precipitate the TiO₂ particles from the solution. After the reaction, the product was washed thoroughly with dH₂O and then with 0.1 M HCl to remove sodium ions from the product. After washing again with dH₂O, the product was dried at 100 °C and then calcinated in air at 500 °C for 30 min to promote crystallization of TiO₂.

By changing the amount of NaOH (10, 20, 30%w), the temperature (120, 150, 180 °C) and the time (1.5, 3, 4.5 h) of the hydrothermal reaction, it is possible to strongly influence the morphology of the prepared material. However, this did not affect their photocatalytic properties. On the other hand, increasing the calcination temperature had a great influence on the photocatalytic properties of the prepared samples. The specific synthesis parameters for each sample are listed in

Table 5. Specific synthesis parameters of different prepared samples (time and temperature of hydrothermal reaction, NaOH content).

Sample	Parameters
A	t = 3 h; T = 150 °C; w(NaOH) = 20%
B	t = 1.5 h; T = 150 °C; w(NaOH) = 20%
C	t = 4.5 h; T = 150 °C; w(NaOH) = 20%
D	t = 3 h; T = 120 °C; w(NaOH) = 20%
E	t = 3 h; T = 180 °C; w(NaOH) = 20%
F	t = 3 h; T = 150 °C; w(NaOH) = 10%
G	t = 3 h; T = 150 °C; w(NaOH) = 30%

.

3.2. Characterization

The XRD pattern of the powder was acquired using a PANalytical X’Pert PRO MPD instrument in the 2θ range of 20–80° with a step size of 0.034°. The average diameter of crystallites, coherently diffracting crystalline domains, was calculated using the Scherrer equation.

The morphology and size of the particles of the prepared samples were examined using a scanning electron microscope (SEM, Ultra Plus Zeiss) at 2 kV.

The Brunauer–Emmett–Teller (BET) specific surface area analyses were performed using the ‘ASAP 2020 Micromeritics’ instrument. Samples were pre-prepared using degas procedure with setpoint vacuum (50 mmHg), heated up to 300 °C and held at final conditions for one hour. Samples were analyzed under nitrogen atmosphere (adsorption-desorption isotherms at 77 K) in a volumetric working device. Specific surface areas were calculated using ten point values for relative pressures between 0.01 and 0.3.

3.3. Photocatalytic Tests

Photocatalytic activity was determined and compared by observing the degradation of plasmocorinth B in deionized water (initial concentration: γ = 12 mg/L; pH = 4). The concentration of the photocatalyst was 0.25 g/L in all experiments. The suspension was continuously stirred and illuminated with two UV Black Light Blue (UV BLB) lamps and one Actinic Blue lamp (λ = 365 nm) with a total light intensity of 20 W/m². The activity was determined by measuring the absorption spectra of the dye at different time intervals.

UV–Vis absorption spectra were recorded using an Agilent Cary 60 UV–Vis spectrophotometer in the wavelength range λ = 400–800 nm. The absorbance value at 550 nm was used to plot the change in concentration of the dye. Prior to measurement, TiO₂ particles were removed from the suspension by filtration through a syringe filter with a pore size of 0.45 µm after the filter surface was saturated with the dye, to prevent additional removal during filtration. For reusability experiments the photocatalyst was removed by centrifugation and stored for the next experiment.

Reaction rate constants were calculated by multiplying the slope of the linear part of the graph by the initial mass concentration of the dye solution. The margin of error was calculated from at least 3 experiments.

Experiments with ibuprofen (γ = 10 mg/L) and diclofenac (γ = 4 mg/L) were performed in the same way, but the determination of concentration was done by using HPLC. The specific parameters were: mobile phase—50% acetonitrile, 50% 0.1% acetic acid; flow rate—0.7 mL/min; column—Poroshell C-18 (Agilent); injected volume: ibuprofen—5 µL; diclofenac—20 µL; UV detector: ibuprofen—222 nm; diclofenac—275 nm.

In order to separate the adsorption and degradation reactions, all tests were initially performed in the dark and illumination started once the maximum amount of dye or pharmaceutical was adsorbed on the surface.

4. Conclusions

The results of the study show that the parameters of hydrothermal synthesis strongly influence the morphology of the prepared samples, while calcination has a dominant effect on the photocatalytic activity of the samples. Reusability experiments show that the photocatalyst is stable and the photocatalytic activity does not decrease with repeated use. The prepared powders can efficiently degrade organic pollutants in photocatalytic reactions at rates comparable to nanoparticles of TiO₂, but their removal is much easier since classical filtration is possible. In combination with tangential flow filtration, a flow-through slurry reactor could potentially be developed.