Amorphous titania as a precursor to brookite-based materials obtained via hydrothermal treatment†

A. O. Revenko^a, D. A. Kozlov^b, I. V. Kolesnik^a, A. S. Poluboiarinov^a, S. Yu. Kottsov^b and A. V. Garshev*^ac
aFaculty of Materials Science, Lomonosov Moscow State University, Moscow 119991, Russia. E-mail: garshev@inorg.chem.msu.ru
^bKurnakov Institute of General and Inorganic Chemistry of RAS, Moscow 119991, Russia
^cFaculty of Materials Science, Shenzhen MSU-BIT University, Shenzhen 518172, China

First published on 4th September 2024

---

---

Introduction

Amorphous phases are commonly found in different materials and are often neglected during analysis, although they can drastically decrease the functional properties of the materials, such as photocatalytic activity,¹ electrical conductivity² and thermal³ conductivity. Therefore, the control of their content is a major consideration in the design of materials. Crystallization from amorphous phase might be promising as one of the ways to decrease its percentage. Having no long-range order, the structure of newly developed crystalline phases can be manipulated via treatment conditions. Titania composites can be considered as examples of materials in which the phase content is crucial, and their photocatalytic activity depends dramatically on the amorphous phase content. Titanium dioxide and materials based on it are one of the well-studied nanomaterials because of their potential applications in heterogeneous photocatalysis,^4–7 photovoltaics,^8,9 biomedicine,¹⁰ and cosmetics^11,12 owing to their low cost, high chemical and thermal stability and non-toxicity.^13–17 The phase content in titania materials determines their properties; therefore, its control during the synthesis of titania materials is important in terms of further applications.¹⁸ The amorphous phase is often difficult to determine in these materials, and hence, it is neglected frequently. Single-phase titania formation is often confirmed using diffraction methods, which are unable to prove the absence of amorphous titania. However, its presence dramatically decreases functional properties such as photocatalytic activity (PCA), which has been proved in anatase and rutile-containing samples.¹⁹ In order to minimize the content of amorphous titania, samples are often calcined at high temperatures. However, in that case, the surface area of the samples shrinks and their morphology is affected.^20,21 Such results are unacceptable in the design of mesoporous and hierarchical nanostructures and lead to poor photocatalytic performance.^22–24

Among the methods used for the formation of crystalline titania particles, hydrothermal methods are proved to be promising owing to the good control of the product phase and morphology.⁹ Furthermore, it has been reported that brookite as modified titania can be obtained via hydrothermal techniques. As one of the metastable modifications of titania, brookite has only been studied actively in recent times,^25–29 and, consequently, the number of works devoted to brookite-based materials is significantly less than those devoted to rutile and anatase phases.³⁰ By comparing the functional properties of various titania modifications, the authors observed a lower photocatalytic activity of brookite than that of rutile and anatase.³¹ Nevertheless, anatase may show poor photocatalytic performance compared to brookite or brookite-based composites depending on the synthesis conditions.³² Moreover, brookite might show better PCA in electron-donor reactions such as CO₂ reduction into solar fuels³³ potentially due to the lower level of the brookite conduction band than those of anatase and rutile.³⁴ Thus, both the development of the synthesis techniques of brookite-based materials and their investigation as photocatalysts are still current scientific tasks.^35–38

Moreover, for other titania modifications, two main approaches are dominant among the hydrothermal methods of single-phase brookite synthesis: hydrothermal treatment of either amorphous titanium dioxide³⁹ or water-soluble titanium compounds.⁴⁰ As a classical method, brookite can be obtained by hydrothermal treatment in the presence of sodium hydroxide. Although it is possible to obtain brookite from amorphous titania in a certain pH range, in a majority of reports on the synthesis of amorphous titania via hydrothermal treatment in a high-pH solution, only multiphase samples with a high brookite content can be formed, whereas brookite can be obtained via purification using peptization.⁴¹ However, when trying to synthesise single-phase brookite, anatase crystallizes at neutral pH values,⁴² and titanates and β-TiO₂ can be found in highly alkaline solutions.^43,44

An alternative method for brookite synthesis is the hydrothermal treatment of water-soluble titanium complexes with acetylacetone⁴⁵ or α-hydroxy acids such as lactic acid. In several studies, single-phase samples of different TiO₂ polymorphs (rutile, anatase, brookite, and β-TiO₂) were obtained by varying the pH of the solution.⁴⁰ At first, single-phase titania formation was explained by the topochemical similarity of the organic complex structure, which distorts at different pH values and forms fragments of the structure of various titania polymorphs containing attached [TiO₆] octahedra.⁴⁶ However, this theory does not explain not only the formation of some titania modifications, such as monoclinic,⁴⁷ but also huge amorphous titania formation at the initial stages of synthesis⁴⁸ and phase transformations of crystalline phases during hydrothermal treatment.⁴⁹ Phase transitions of titania under hydrothermal conditions lead to an explanation of the brookite growth mechanism. According to the proposed mechanism,⁵⁰ titanium complex hydrolysis and anatase particle growth take place at the initial stages, mainly because of the oriented attachment, and when the critical size is reached, a phase transition to brookite occurs with the attachment to larger brookite particles. The formation of the amorphous phase in large amounts at the initial stages of synthesis can generalize the proposed mechanism to cases of brookite formation using pre-obtained amorphous titania. Therefore, amorphous titania treatment in the presence of lactic acid may lead to selective crystallization of brookite.

By comparing these approaches to brookite synthesis, the main advantages of each of them are worth noting. Although the use of titanium-containing complexes requires an additional synthetic stage, it allows obtaining single-phase high-crystalline brookite with an amorphous content of no more than 10%. However, pre-obtained amorphous titania as a precursor allows such a brookite synthesis technique to be considered a post-treatment method, while amorphous titania materials can be obtained in various ways with different morphologies of the particles.

Thus, the present work demonstrates the differences in the phase content (including the residual amorphous phase) of brookite-containing samples synthesized via amorphous titania treatment with NaOH and lactic acid under hydrothermal conditions. Moreover, the effect of micromorphology and the initial titania crystallinity was investigated, and the photocatalytic activity of the obtained brookite samples was characterized. Special attention was paid to the amorphous phase in the samples. The photocatalytic activity of the obtained brookite–anatase samples was shown to be better in the case of brookite-rich composites than in individual anatase. This contradicts the popular opinion of brookite being the least photoactive modification of titania.

Materials and methods

Materials

Titania (TiO₂) Evonik Aeroxide P25 (80% anatase and 20% rutile) was used without modification. Tetra-n-butyl titanate (TNBT) Ti(OⁿBu)₄ (97%, Aldrich 244112) was used for amorphous titania synthesis. Lactic acid (85%, Aldrich), sodium hydroxide (analytical grade, Rushim), and urea (analytical grade, Rushim) were used as additives in hydrothermal syntheses without any purification.

Synthesis

At the first stage, amorphous titania was synthesized by TNBT hydrolysis under vigorous stirring at room temperature. The obtained precipitate was centrifuged, rinsed with water and ethanol, and then used for further treatment in sodium hydroxide or lactic acid mixed with urea solutions.

Treatment in alkali media was performed with either adjustment of amorphous titania suspension to a certain pH level (equilibrium method) or a fixed sodium hydroxide concentration (non-equilibrium method).

For the first set of samples (equilibrium method), the pH level of amorphous titania suspension was adjusted by pH control with a NaOH solution and the equilibrium was reached. The obtained samples were labeled “am_TiO₂(pH_n)”, where “n” is the equilibrium pH before hydrothermal treatment. For the other set of samples (nonequilibrium method), a NaOH solution of exact concentration was added to the amorphous titania suspension and further equilibrium in the mixture was not reached. The samples obtained by this approach were labeled “am_TiO₂(NaOH_c)”, where “c” is juxtaposed with a base concentration such as c = −log[c(NaOH)]. Then, 24 ml of obtained suspensions containing 100 mg of amorphous titania were placed in a 60 mL Teflon-lined stainless steel autoclave and heated in a muffle oven at 180 °C for 48 h. After that, the white precipitate was centrifuged, rinsed with water and ethanol, and then dried at 60 °C overnight.

Other amorphous titania treatments under hydrothermal conditions were carried out using a solution containing lactic acid (Hla) and urea. After 100 mg of amorphous titania and 8.64 g of urea were placed in an autoclave, different amounts of Hla were added to achieve a Hla-to-TiO₂ molar ratio in the range from 1/5 to 5/1. The total suspension volume was therefore appended with deionized water up to 24 mL. The samples obtained using this method were named am_TiO₂(Hla_x:y), where “x:y” is the Hla-to-TiO₂ ratio.

For comparison, high-purity brookite samples were synthesized according to a previous work⁵⁰ by the hydrothermal treatment of a lactate complex of titania. A titanium lactate complex was prepared by the hydrolysis of 48 mmol TNBT in 50 mL of lactic acid water solution (144 mmol) under stirring at 60 °C. Then, the solution was heated for 4 hours in order to remove the redundant butanol. The obtained titanium complex solution (4.8 mmol) was mixed with 8.64 g urea, and then the total volume was adjusted to 24 ml with deionized water. The resulting solution was placed in a 60 mL Teflon-lined stainless steel autoclave and, as in the previous experiment, heated at 180 °C for 48 h. The white precipitate obtained after the treatment was centrifuged, rinsed with water and ethanol, and then dried at 80 °C.

In order to evaluate the titania morphology and crystallinity effect on brookite crystallization, a series of samples were synthesized using either amorphous titania microspheres prepared according to a previous work¹¹ or commercial Aeroxide P25 as titania precursors using similar aforementioned methods. Amorphous titania microsphere synthesis was performed in a 100 ml polypropylene flask by the following procedure. A certain volume of an aqueous sodium hydroxide solution was added to 30 ml of absolute ethanol, and then the obtained mixture was cooled down to −5 °C. Afterwards, 640 μL of TNBT was added dropwise under vigorous stirring, and then the solution was kept for two hours. The obtained titania colloid solution was left to dry in a propylene beaker in air at room temperature for a week.

Material characterization

X-ray diffraction (XRD) experiments were conducted using a Rigaku D/MAX 2500 diffractometer with a rotating anode (Rigaku, Japan) in the reflection mode (Bragg–Brentano geometry) using CuKα_1,2 radiation and a graphite monochromator. The pattern accumulation for a phase analysis was performed in the 2θ scanning mode with a step of 0.02° on the 2θ scale and a spectrum accumulation time of 1 s for an interval of 5–80°. Diffraction maxima were identified using the JCPDS Data Bank. Full-profile analysis of diffractograms using the Rietveld method was performed using the free software package Jana2006 [57]. In order to estimate the amorphous phase fraction, XRD patterns were registered in the presence of an external standard, which was corundum. The process of amorphous phase content calculation used the obtained patterns described in the ESI† as well as the calculations of particle average sizes.

Scanning electron microscopy (SEM) was performed using a LEO SUPRA 50 VP and NVision 40 devices (Zeiss, Germany). The analysis was performed at an accelerating voltage of 20 keV, using a 30 μm condenser aperture.

Transmission electron microscopy (TEM) and selected area electron diffraction (SAED) was performed using a Libra 200 MC (Zeiss, Germany) equipped with an AZTEC energy-dispersion microanalysis system (Oxford Instruments Inc., UK) and an X-MAX 80T detector (Oxford Instruments Inc., UK). TEM microphotographs were acquired using a CCD camera (Gatan, USA) with a matrix size of 4096 × 4096 pixels. Electron diffraction patterns were radially integrated in the CrysTBox software.⁵¹

The specific surface area of the samples was measured by the low-temperature nitrogen adsorption method using an ATKh-O6 analyzer (Katakon, Russia). Samples were degassed in a nitrogen flow (1 atm) at 200 °C for 1 hour prior to analysis. Based on the data obtained, the specific surface area of the samples was calculated using the Brunauer–Emmett–Teller (BET) model, and the five-point method in the partial pressure range of 0.05–0.25.

The photocatalytic activity of the obtained samples was measured using a previously constructed measuring device.⁵² The structure of the unit includes a quartz glass with a volume of 50 mL, in which the suspension of the specimen and dye are placed. The reaction mixture was stirred during the reaction and the temperature was maintained at 35 °C. The working area of the reactor was irradiated using a high-pressure mercury lamp with a power of 5.5 W. During the experiment, aliquots were continuously collected using a peristaltic pump and the extinction spectra are recorded using a QE65000 spectrophotometer (Ocean Insight, USA) with a 50 W HPX-2000 xenon lamp.

During the measurement, a sample weighing 2–5 mg was placed in 30 mL of phosphate buffer solution. Then, the sample was dispersed using ultrasound for 5 minutes to remove large aggregates. Then, the resulting suspension was placed in the working area of the reactor and kept for about 5 minutes. After the suspension was stabilized, 10 ml of a 100 mg L⁻¹ methylene blue solution was added to the reaction mixture. The mercury lamp was turned on 5–10 minutes after the suspension mixing and dye adsorption processes were completed. The measurements were performed for 1 hour, with a periodicity of extinction spectra accumulation equaled to 3 s. The decomposition constant was calculated as the slope of the line, constructing the kinetic curve in semi-logarithmic coordinates and normalized to the mass or surface area of the photocatalyst.

Results and discussion

In the course of the work, experiments were conducted to determine the influence of hydrothermal synthesis conditions on the phase composition of the obtained samples. The synthesis methods were divided according to the precursors used: 1) classical synthesis from amorphous titania in the presence of sodium hydroxide; 2) synthesis from amorphous titania in the presence of lactic acid and urea; 3) synthesis from the titanium lactate complex in the presence of urea; and 4) synthesis from crystalline titania Aeroxide P25 and amorphous TiO₂ microspheres. Classical synthesis using sodium hydroxide was divided into two sub-methods: equilibrium method (am_TiO₂(pH_n)) and non-equilibrium method (am_TiO₂(NaOH_c)). In the first case, the pH value of the solution was registered after the processes of sorption were completed and the basicity of the mixture was not changed. In the other sub-methods, a NaOH solution was added to the amorphous titania without reaching an equilibrium, and the mixture was treated under hydrothermal conditions.

Methods using sodium hydroxide

The XRD patterns of the samples crystallized in the NaOH solution under hydrothermal condition are shown in Fig. S1.† Different approaches to sodium hydroxide presence control lead to a similar phase content of the samples – in both cases, a mixture of brookite and anatase is observed (Fig. 1), and this result is in good agreement with other works.³⁹ The brookite content is low with low pH values of the solution in both methods; however, the am_TiO₂(NaOH_c) samples show a slightly larger amount of brookite phase with the same amount of amorphous phase. In both cases, the brookite content increases when the pH grows, which, besides the pH influence, can be explained by the mechanism of brookite stabilization by Na⁺ ions.⁵³ Nonetheless, after the critical pH value, initial amorphous titania transforms into titanate phases, and the brookite content with large NaOH concentrations plummets to 10–20% (Fig. S2†). At the critical pH value, the brookite content reaches its maximum, which is 70% for the am_TiO₂(NaOH_c) samples with a NaOH concentration of 0.1 M. The general brookite content amount does not depend significantly on the method chosen. However, it should be noted that the critical pH values in the series differ with pH = 11.5 in the equilibrium method and c(NaOH) = 0.1 M in the non-equilibrium method, which initially corresponds to pH = 13. The observed effect may be caused by a decrease in the sodium hydroxide concentration in the reaction mixture due to sorption by hydrated titania when titration is performed, which leads to a consequent decrease in the pH of the solution. Following the results, the sub-methods seem to have a similar mechanism of brookite formation; however, there might be differences. In the equilibrium method, the fully titrated surface of amorphous titania can influence the formation of brookite particles, whereas in the non-equilibrium method, the surface is not titrated fully. Such difference leads to different phase compositions of the resulting samples. Thus, the maximum content of brookite in the non-equilibrium series is more than 70 per cent, which is higher than this number for the equilibrium method. Therefore, samples obtained by equilibrium and non-equilibrium methods may vary not only in phase composition but also in functional properties due to the modification of titania surface in equilibrium.

While increasing the pH level of the solution, the content of amorphous titania is declining monotonously, but even at the critical pH value, the brookite fraction remains significant and accounts for at least 15%, and crystallization is not complete. The anatase content in the samples shows a similar dependence, and anatase is found in all the samples below the critical pH value. This effect can be explained by the simultaneous crystallization of anatase and brookite particles during hydrothermal treatment, and therefore even at pH values close to the critical, the anatase content in the samples remains high. Instead of anatase and brookite, titanate phases are formed in large percentages after the critical pH point. Overall, the samples obtained by the method using sodium hydroxide consist of a large amount of amorphous phases and anatase. Thus, this method is considered non-optimal for obtaining brookite-rich titania photocatalysts.

The average brookite and anatase particle sizes calculated using the Scherrer formula are presented in Fig. 1c and d. When the anisotropy of particles is considered, the XRD patterns fitting quality increases; therefore, the brookite particles sizes were evaluated in the (001) and (210) directions, which are perpendicular to each other. The similar growth tendencies of brookite and anatase particle sizes were observed with the rise in the pH value. Thus, the brookite particles reach a maximum of about 150 × 90 nm in a 0.1 M NaOH solution (am_TiO₂(NaOH_c) samples) and 160 × 60 nm at pH = 11.5 (am_TiO₂(pH_n) samples). The size of the anatase particles is shown to be approximately similar in both of the series, fluctuating in the range of 15–35 nm; however, in the equilibrium method (am_TiO₂(pH_n) samples) anatase particles appear to be slightly larger.

The TEM images also prove the transformation of amorphous titania to the brookite–anatase mixture (Fig. 2a–c). The brookite particles are represented on the images as anisotropic rods, whereas the anatase particles have an isotropic shape. The size of the brookite particles appears to be similar to the numbers estimated from the XRD analysis. The average size of the particles on the images of the sample am_TiO₂(NaOH_3) is approximately 17–20 nm, which is close to the number estimated from the XRD patterns. However, large particles with sizes up to 120 nm × 20 nm in minor amounts can be found. On the images of the sample am_TiO₂(NaOH_1) with the highest brookite content, the average particle size is indeed about 150 nm; however, a large dispersion in the sizes can be noted with particles' sizes being in the range from 40 to 300 nm.

By comparing the TEM images, the brookite content is found to be rising, when increasing the basicity of the solution, as the ratio of anisotropic and isotropic particles increases, which confirms the results of X-ray diffraction. The same result can be obtained from the radial average of SAED data (Fig. 2c). In this case, it is impossible to estimate the phase fractions in the samples using the most intense peak since the peaks of anatase [101] and brookite [120], [111] overlap (the peak at 2.85 nm⁻¹). Therefore, the second most intense peak of brookite [121] is usually chosen, which is at 3.45 nm⁻¹, and the proportion of the brookite and anatase is estimated using the peak intensity ratio, which gives similar results in terms of the phase content of the samples.

Lactate-involved method

According to X-ray diffraction, the samples synthesized in the presence of lactic acid and urea represent a mixture of brookite and anatase (Fig. 3a). The brookite content in the samples is already high even with a low amount of lactic acid added to the system and accounts for about 35%. When increasing the lactic acid concentration, the brookite content rises and achieves 83% at a Hla-to-TiO₂ ratio of 5/1. The amorphous phase fraction in the samples does not depend significantly on the lactic acid concentration and remains at the level of approximately 10%, which is slightly less than this amount in the syntheses with sodium hydroxide. Despite minor anatase amounts in the samples, the lower residual amorphous phase and high brookite content make this synthetic approach more suitable as a post-treatment method to obtain brookite-rich samples.

The sizes of the titania particles estimated by X-ray diffraction (Fig. 3b) depends on the lactic acid concentration. The brookite particles are anisotropic in the entire range of concentrations with larger anisotropy observed in the samples obtained with large differences in titania and lactic acid amounts. The size of brookite particles is growing constantly, and the maximum is achieved at the largest Hla-to-TiO₂ ratio, with the brookite size being 45 nm × 26 nm. The anatase particle size also increases with the increase in lactic acid concentration, although being less than the brookite size. The maximum of 28 nm for anatase particles size is achieved when the most amount of Hla is used. It is worth noting that the anatase size is similar to the size of brookite particles in the (210) direction.

The TEM images confirmed the increase in the brookite content of samples, when the lactic acid concentration increases (Fig. 4a–c). Brookite and anatase represent prolonged anisotropic rods and isotropic particles, correspondently, as well as in the methods using sodium hydroxide. For the sample am_TiO₂(Hla_1:5) anatase and brookite particle sizes are predominantly about 20 nm and less, though few large brookite rods with lengths of more than 50 nm can be noticed. The content of isotropic particles of anatase is significantly less for the sample am_TiO₂(Hla_1:1); here, the sizes of anisotropic brookite particles demonstrate good agreement with the XRD data. Though in the sample with the largest amount of lactic acid, brookite particles represent a majority of the particles, and their sizes strongly fluctuate from 30 nm to 70 nm, the average sizes correspond to the XRD results. By the increase in the number of rods on the microphotographs, one can judge about growth in the brookite content in the samples, which agrees with the XRD analysis results. Based on the radial averages of SAED data (Fig. 4d), the growth of peak intensities at 3.45 nm⁻¹ and 4.0–4.5 nm⁻¹ indicates an increase in brookite fraction in the samples and correlates with the enlargement of titania particles.

Based on the obtained results, we can strengthen the idea of brookite formation mechanism, which has already been mentioned.⁵⁰ In the beginning of the hydrothermal process, anatase particles form, and then they might undergo the recrystallization process. During such process, they can transform into brookite particles but only if they have a size below some critical point; otherwise, anatase might continue growing. The parameter responsible for the possibility of anatase to transform into brookite is the amount of the amorphous phase in the solution. With large amounts of amorphous titania in the reaction area, a large amount of anatase particles emerge, which accelerates its growth beyond the critical size with no possible further transformation into brookite. Another situation is observed when titanium lactate is used as a precursor during the hydrothermal treatment. In this case, gradual hydrolysis of the complex occurs with amorphous titania in low concentrations being the product of the reaction. With low concentrations of amorphous titania, a small amount of anatase is formed, and therefore its growth decelerates, which makes the phase transition to brookite occur more likely. In such case, samples with no anatase phase can be obtained. Besides the amorphous phase, the lactate acid concentration seems to increase the possibility of anatase-to-brookite transformation and accelerates this process since its larger amount leads to brookite content growth.

This mechanism of brookite particle growth appears to differ from the one observed in the method using sodium hydroxide. In the last case, simultaneous formation of both anatase and brookite particles is observed, where brookite fraction is controlled by the pH value of the solution and possibly by the sodium cation concentration, which has been reported previously.^53,54 Therefore, the brookite crystallization process has a higher probability at high pH values with large Na⁺ concentrations than this process for anatase. In the case of brookite, the particle formation rate remains constant regardless of the pH value, whereas the growth rate increases gradually with the increase in NaOH concentration. This leads to the size of brookite particles being proportional to its content in the sample, in contrary to the mechanism for the lactate-involved method. Although the brookite content grows in both of the methods, the method using sodium hydroxide has a limit in this growth, and after a critical pH point, the brookite content can no longer be exceeded since under these conditions, titanates are more stabilized. However, in the lactate-involved method, there is no such critical point, and even at high lactic acid concentrations, brookite forms in large amounts.

Synthesis using other precursors

To investigate the transformation of crystalline titanium dioxide to brookite during the hydrothermal process, syntheses using crystalline Aeroxide P25 containing anatase and rutile were performed. The samples obtained by the lactate-involved method and the method using sodium hydroxide were examined by XRD analysis (Fig. S3†). The phase content of Aeroxide P25 does not change during the hydrothermal treatment, and the characteristic peak of brookite [121] does not appear. According to the earlier proposed mechanism, anatase can recrystallize into brookite as an intermediate during the hydrothermal treatment of amorphous titania. The observed difference is explained by the size effect: as an intermediate in the process, anatase particles should not achieve the critical size to transform into brookite. Therefore, crystalline titania particles, which are already bigger than the critical size, such as 20 nm for Aeroxide P25, do not transform into brookite particles.

When studying whether the morphology of amorphous titania affects the brookite content in the final product, lactate-involved synthesis was performed using microspheres of amorphous TiO₂ with 2 μm diameter as precursors. The resulting samples contain a mixture of anatase and brookite. However, comparing the phase compositions of the samples obtained and those obtained using amorphous titanium dioxide, an increased percentage of the amorphous phase and a lower content of brookite can be noticed (Fig. 5c). Nonetheless, the amount of amorphous phase remains similar at the level of 15%. The sizes of brookite and anatase particles do not depend on the concentration of lactic acid and constitute 25 nm × 10 nm for brookite and 18 nm for anatase (Table S1†). According to the SEM images (Fig. 5a and b), hydrothermal treatment of microspheres appears not to affect the form of the particles after the treatment. However, the surface of the particles is altered significantly, and crystalized particles can be observed there. Therefore, one can assume that the crystallization process of brookite occurs predominantly on the surface of the particles. It also shows the effect of lactate ions in the reaction, since there is a higher concentration near the surface of microspheres, where brookite formation occurs, and anatase forming in the center of the microspheres does not transform into brookite without lactate ion presence.

The single-phase brookite was obtained by the hydrothermal treatment of a titanium lactate complex in a urea solution and analyzed by XRD analysis. The full-profile analysis of the diffraction pattern shows (Fig. S4†) that the sample represents high-purity brookite with an amorphous phase fraction of less than 10%. Brookite particles (Fig. 6) grow anisotropically along the [001] direction and reach an average size of 100 nm in length and 30 nm in width.

Photocatalytic properties

The photocatalytic activity (PCA) was measured using the model reaction of methylene blue decomposition. For the following tests, the photocatalytic activity of Aeroxide P25 and the previously obtained single-phase brookite sample was measured, which turned out to be equal to 1.2 g⁻¹ s⁻¹ and 0.4 g⁻¹ s⁻¹ respectively. These values were considered reference when studying the photocatalytic activity of other samples.

Since the specific surface area (SSA) of the sample is a significant characteristic of heterogeneous catalysts, this property was measured for all the samples by low-temperature nitrogen adsorption analysis. It can be noted that with the increase in the brookite content in the samples, the SSA decreases. This tendency is seen most vividly for the samples obtained using the lactate-involved method (Fig. 7c), where the SSA continuously declines from about 160 m² g⁻¹ to 90 m² g⁻¹, while the brookite particles become not only prevalent but also larger. In terms of the non-equilibrium method (am_TiO₂(NaOH_c)) (Fig. 7b), the samples also show similar dependency. For the am_TiO₂(pH_n) samples obtained by the equilibrium method, this trend is seen for pH values from 10 to 11.5, whereas at pH = 9 there is a lower value of the specific surface area. Comparing the equilibrium and non-equilibrium methods, the second method shows a larger SSA, probably because of smaller anatase particles. The hydrothermal treatment of amorphous TiO₂ microspheres leads to a significant decline in specific surface area; meanwhile, the particle sizes of crystalline particles do not change considerably. The SSA of the single-phase brookite turns out to be 34 m² g⁻¹, which is lower than this number for the samples with mixed phases, whereas the SSA of Aeroxide P25 is 55 m² g⁻¹, which is explained by smaller anatase and rutile particles in the sample.

The photocatalytic activity of the obtained samples was analyzed using two values (Fig. 8): the photocatalytic activity measured by the model reaction of methylene blue decomposition (PCA) and this number normalized by the surface area (specific surface area-normalized photocatalytic activity (SSA-normalized PCA)) to evaluate the surface capability in photocatalysis. For the samples obtained by the equilibrium method (am_TiO₂(pH_n)), PCA turns out to be relatively the same, fluctuating from 0.12 to 0.16 g⁻¹ s⁻¹. However, when normalized by SSA, the growth can be seen clearly. Analyzing the am_TiO₂(NaOH_c) samples obtained by the non-equilibrium method, brookite and anatase – crystalline phases in the samples – show a similar photocatalytic performance since SSA-normalized PCA of the samples treated at a NaOH concentration from 10⁻⁵ to 0.1 M remains at a constant value. At the same time, PCA is decreasing continuously in this region of concentration, which happens because of the decline in surface area.

The photocatalytic properties of the samples of the lactate-involved method set depend significantly on the Hla-to-TiO₂ ratio used in the synthesis: both PCA and SSA-normalized PCA values increase when the ratio increases. However, this growth occurs only for samples with a high Hla-to-TiO₂ ratio. Anatase happens to have extremely low photocatalytic properties in this set since even samples with a considerably high amount of this phase show low values of both PCA and SSA-normalized PCA. When treating amorphous titania microspheres, the samples start showing photocatalytic properties, which are more likely due to the rise in brookite content, whereas the presence of anatase and amorphous phases does not affect the PCA or SSA-normalized PCA considerably. Therefore, when treated with lactic acid and urea under hydrothermal conditions, amorphous titania can be crystallized without changing its shape; however, with improvement in the photocatalytic properties of the material. It is worth mentioning that despite the large improvement in the photocatalytic properties of titania under such conditions, the PCA of the samples is significantly lower than the PCA of single-phase brookite or Aeroxide P25 (maximum of 70 or 23 per cent respectfully). Therefore, such synthesis should not be used for achieving large PCA values of the samples, but rather to improve the PCA of the samples with the desired morphology without its alteration.

When analyzing the role of brookite phase in the photocatalytic activity of the samples, the dependence on the SSA-normalized PCA was reviewed (Fig. 9). Overall, the PCA of the samples increases monotonously when the brookite content increases. This indicates brookite as a better photocatalyst under such synthetic conditions. Moreover, there is no extremum on the dependency, which proves that brookite and anatase do not form composites, and the PCA of these phases is additive. Higher values of SSA-normalized PCA can be seen in the equilibrium NaOH method, where the number reaches 4.5·10⁻³ m⁻² s⁻¹. The difference in equilibrium and nonequilibrium methods is worth mentioning. Sodium hydroxide seems to be an important factor in the PCA of brookite. It has been discussed previously in this article that NaOH does not modify the surface of brookite particles in the nonequilibrium method, which might lead to different functional properties of the samples. This is seen in the SSA-normalized PCA dependence, where it seems to be constant in the non-equilibrium method, whereas sodium hydroxide modifies brookite's surface significantly in the equilibrium method, which leads to a higher PCA. Nevertheless, for NaOH-involved methods of synthesis, anatase seems to have lower photocatalytic properties than those of brookite, so when the brookite content increases, the SSA-normalized PCA increases and grows threefold with pH values from 10 to 11.5. A similar picture can be seen for lactate-involved synthesis. Although the brookite content is observed to be high in all of the samples, the PCA increase starts to be noticeable only when higher amounts of lactic acid are used, which also tells about the positive modification of the brookite surface by lactic acid. It is worth mentioning that although there are different amounts of the phases in the samples, the mechanism of organic dye decomposition seems to be consistent for all of the samples, which includes molecules' oxidation by hole formatting in the valence band of the semiconductors. The differences in PCA values are dictated only by the phase composition of the samples. Overall, in all of the methods discussed, brookite, obtained by the amorphous titania post-treatment, shows a photocatalytic activity much larger than that of the amorphous precursor. Thus, the PCA of amorphous titania with various morphology can be increased by applying the approach to brookite crystallization. Long-term UV irradiation of the samples showed that the phase composition of titania, including the metastable phases of brookite and anatase, as well as the amorphous phase, does not change (Fig. S5†).

It is worth noting that anatase shows various photocatalytic properties in different sets of samples. With the photocatalytic properties at the level of brookite in the non-equilibrium method set, the use of the equilibrium method leads to lower photocatalytic properties, becoming even lower in the lactate set. This tendency can be explained by different anatase surface modifications that occur during synthesis, which might affect the photocatalytic properties tremendously. This effect has been shown for alkaline modification,³² and, moreover, lactic acid shrinks the photocatalytic properties of anatase dramatically, according to the data obtained in this work. Therefore, brookite has SSA-normalized PCA values the same or more than those of anatase depending on the synthesis conditions, despite the generally believed assumption that brookite has much less photocatalytic properties than those of anatase.⁵⁵ We suggest that brookite can show poorer photocatalytic performance due to the lower surface area; however, when normalized by surface area, brookite might show even better photocatalytic abilities. This fact is also proved by the single-phase brookite SSA-normalized PCA value, 12 m⁻² s⁻¹, which significantly exceeds this number in the other samples. For comparison, the SSA-normalized PCA value of Aeroxide P25 was calculated, which is 22 m⁻² s⁻¹. The higher number can be explained by the presence of rutile phase in the sample. Overall, all of the samples show much lower PCA than that of the single-phase brookite, reaching only 50% of this number for single-phase brookite. This behavior can be explained by the influence of other phases, especially the amorphous phase, which has been shown to decrease the photocatalytic performance greatly.¹⁹

Conclusion

Various methods of amorphous titania post-treatment under hydrothermal conditions resulting in brookite-rich titania have been examined. The classical hydrothermal treatment of amorphous titania in a sodium hydroxide solution leads to brookite–anatase mixture crystallization. The amorphous titania and anatase contents decrease, when the reaction mixture pH increases and is less than the titanate formation pH value. Synthesis using lactic acid and urea demonstrates that a high Hla-to-TiO₂ ratio leads to an increase in brookite content, with the fraction of anatase and amorphous phases being less than 17%. Various dependencies of particle sizes and phase contents in both methods point to differences in brookite formation mechanisms. In the method involving lactic acid, brookite seems to form via the anatase phase, whereas in the sodium hydroxide method, anatase and brookite crystallizations are parallel processes. Such phenomenon predominantly occurred due to the difference in the particle size observed during the synthesis, which shows the important role of the thermodynamically stable sizes for various titania phases in the reaction. Hence, crystalline titania is not able to transform into another crystalline phase during hydrothermal synthesis. These results suggest amorphous titania being favorable to obtain materials with high control of its crystalline phase content. We suppose amorphous precursors can also be preferable when obtaining a wide range of crystalline materials.

However, amorphous titania transforms into brookite without any significant changes in its morphology. The measurements of the samples' photocatalytic properties prove that the amorphous phase decreases the photocatalytic activity, whereas these activities of single-phase brookite are relatively high. It is shown that a higher brookite content in samples leads to a higher PCA; moreover, NaOH and lactic acid modify the surface of brookite particles, which increases the PCA considerably. Brookite demonstrates not worse photocatalytic properties than those of anatase, when normalized by the specific surface area. Thus, the methods suggested in this work can be seen as amorphous titania post-treatment to improve its photocatalytic properties via selective brookite crystallization without changing the form of the particles. Such method can be used to enhance the photocatalytic performance of the already obtained titania materials to minimize the negative impact of amorphous titania.

Data availability

The data supporting this article have been included as part of the ESI.†

Author contributions

Revenko A. O. – software, investigation, writing – original draft, writing – review & editing, Visualization; Kozlov D. A. – methodology, software, writing – original draft, writing – review & editing; Kolesnik I. V. – investigation, validation; Poluboiarinov A. S. – investigation, validation; Kottsov S. Yu. – investigation, validation; Garshev A. V. – conceptualization, resources, data curation, writing – review & editing, supervision.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The research was carried out using the equipment of MSU Shared Research Equipment Centre “Technologies for obtaining new nanostructured materials and their complex study” and purchased by MSU in the frame of the Equipment Renovation Program (National Project “Science and Universities”) and in the frame of the MSU Program of Development. SEM and BET measurements were performed using the equipment of the Joint Research Centre (JRC PMR) IGIC RAS and supported by IGIC RAS state assignment (Ministry of Higher Education and Science of the Russian Federation).

References

1. Y. Zheng, et al., Plasma-induced hierarchical amorphous carbon nitride nanostructure with two N₂ C-site vacancies for photocatalytic H₂O₂ production, Appl. Catal., B, 2022, 311, 121372 CrossRef CAS .
2. B. Qin, et al., Crystalline molybdenum carbide−amorphous molybdenum oxide heterostructures: In situ surface reconfiguration and electronic states modulation for Li−S batteries, Energy Storage Mater., 2022, 47, 345–353 CrossRef .
3. L. Guo, et al., Extremely high thermal conductivity of carbon fiber/epoxy with synergistic effect of MXenes by freeze-drying, Compos. Commun., 2020, 19, 134–141 CrossRef .
4. J. Jin, S. Chen, J. Wang, C. Chen and T. Peng, One-pot hydrothermal preparation of PbO-decorated brookite/anatase TiO₂ composites with remarkably enhanced CO₂ photoreduction activity, Appl. Catal., B, 2020, 263, 118353 CrossRef CAS .
5. Y. V. Kolen'ko, et al., Hydrothermal synthesis of nanocrystalline and mesoporous titania from aqueous complex titanyl oxalate acid solutions, Chem. Phys. Lett., 2004, 388, 411–415 CrossRef .
6. I. V. Kolesnik, et al., Preparation of nanocrystalline nitrogen-doped mesoporous titanium dioxide, Mendeleev Commun., 2013, 23, 11–13 CrossRef CAS .
7. T. T. Nguyen and K. Edalati, Brookite TiO₂ as an active photocatalyst for photoconversion of plastic wastes to acetic acid and simultaneous hydrogen production: Comparison with anatase and rutile, Chemosphere, 2024, 355, 141785 CrossRef CAS PubMed .
8. K. Chen, et al., High Efficiency Mesoscopic Solar Cells Using CsPbI₃ Perovskite Quantum Dots Enabled by Chemical Interface Engineering, J. Am. Chem. Soc., 2020, 142, 3775–3783 CrossRef CAS PubMed .
9. M. Monai, T. Montini and P. Fornasiero, Brookite: Nothing New under the Sun?, Catalysts, 2017, 7, 304 CrossRef .
10. O. L. Galkina, V. K. Ivanov, A. V. Agafonov, G. A. Seisenbaeva and V. G. Kessler, Cellulose nanofiber-titania nanocomposites as potential drug delivery systems for dermal applications, J. Mater. Chem. B, 2015, 3, 1688–1698 RSC .
11. A. S. Poluboyarinov, et al., Titanium oxide microspheres with tunable size and phase composition, Materials, 2019, 12, 1472 CrossRef CAS PubMed .
12. H. J. Leong and S. G. Oh, Preparation of antibacterial TiO₂ particles by hybridization with azelaic acid for applications in cosmetics, J. Ind. Eng. Chem., 2018, 66, 242–247 CrossRef CAS .
13. M. Shakeel, et al., Toxicity of Nano-Titanium Dioxide (TiO₂-NP) Through Various Routes of Exposure: a Review, Biol. Trace Elem. Res., 2016, 172, 1–36 CrossRef CAS PubMed .
14. Y. Wang, et al., Biomolecule-controlled hydrothermal synthesis of C-N-S-tridoped TiO₂ nanocrystalline photocatalysts for NO removal under simulated solar light irradiation, J. Hazard. Mater., 2009, 169, 77–87 CrossRef CAS PubMed .
15. C. Su, et al., Sol-hydrothermal preparation and photocatalysis of titanium dioxide, Thin Solid Films, 2006, 498, 259–265 CrossRef CAS .
16. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Environmental Applications of Semiconductor Photocatalysis, Chem. Rev., 1995, 95, 69–96 CrossRef CAS .
17. N. N. Gracheva, et al., Am(III) sorption onto TiO₂ samples with different crystallinity and varying pore size distributions, Appl. Geochem., 2014, 42, 69–76 CrossRef CAS .
18. D. R. Eddy, et al., Heterophase Polymorph of TiO₂ (Anatase, Rutile, Brookite, TiO₂ (B)) for Efficient Photocatalyst: Fabrication and Activity, Nanomaterials, 2023, 13, 704 CrossRef CAS PubMed .
19. V. A. Lebedev, et al., The amorphous phase in titania and its influence on photocatalytic properties, Appl. Catal., B, 2016, 195, 39–47 CrossRef CAS .
20. M. Trochowski, et al., How insignificant modifications of photocatalysts can significantly change their photocatalytic activity, J. Mater. Chem. A, 2019, 7, 25142–25154 RSC .
21. A. Rose, A. Hofmann, P. Voepel, B. Milow and R. Marschall, Photocatalytic Activity and Electron Storage Capability of TiO₂ Aerogels with an Adjustable Surface Area, ACS Appl. Energy Mater., 2022, 5, 14966–14978 CrossRef CAS .
22. Z. Li, H. Li, S. Wang, F. Yang and W. Zhou, Mesoporous black TiO₂/MoS₂/Cu₂S hierarchical tandem heterojunctions toward optimized photothermal-photocatalytic fuel production, Chem. Eng. J., 2022, 427, 131830 CrossRef CAS .
23. J. Lyu, et al., Platinum-enhanced amorphous TiO₂-filled mesoporous TiO₂ crystals for the photocatalytic mineralization of tetracycline hydrochloride, J. Hazard. Mater., 2019, 373, 278–284 CrossRef CAS .
24. W. Wang, H. Chen, J. Fang and M. Lai, Large-scale preparation of rice-husk-derived mesoporous SiO₂@TiO₂ as efficient and promising photocatalysts for organic contaminants degradation, Appl. Surf. Sci., 2019, 467–468, 1187–1194 CrossRef CAS .
25. A. Di Paola, M. Bellardita and L. Palmisano, Brookite, the Least Known TiO₂ Photocatalyst, Catalysts, 2013, 3(1), 36–73 CrossRef CAS .
26. M. Kamran, T. A. Kandiel, S. Abdel-Azeim, M. A. Morsy and D. W. Bahnemann, Mechanistic Insights into the High Selectivity and Photocatalytic Activity of Brookite TiO₂ toward NO_X Abatement, J. Phys. Chem. C, 2023, 127, 7707–7717 CrossRef CAS .
27. Y. Zou, et al., Hydrothermal synthesis of Zn-doped brookite TiO₂ for enhanced visible-light-responsive photocatalytic performance, Mater. Res. Express, 2023, 10, 085005 CrossRef .
28. G. Ren, M. Zhou and H. Wang, Weakened Interfacial Hydrogen Bond Connectivity Drives Selective Photocatalytic Water Oxidation toward H₂O₂ at Water/Brookite-TiO₂ Interface, J. Am. Chem. Soc., 2024, 146, 6084–6093 CrossRef CAS PubMed .
29. S. Gervasi, N. Blangetti, F. S. Freyria, S. Guastella and B. Bonelli, Undoped and Fe-Doped Anatase/Brookite TiO₂ Mixed Phases, Obtained by a Simple Template-Free Synthesis Method: Physico-Chemical Characterization and Photocatalytic Activity towards Simazine Degradation, Catalysts, 2023, 13, 667 CrossRef CAS .
30. T. T. H. Tran, T. T. H. Bui, T. L. Nguyen, H. N. Man and T. K. C. Tran, Phase-Pure Brookite TiO₂ as a Highly Active Photocatalyst for the Degradation of Pharmaceutical Pollutants, J. Electron. Mater., 2019, 48, 7846–7861 CrossRef CAS .
31. J. Zhang, P. Zhou, J. Liu and J. Yu, New understanding of the difference of photocatalytic activity among anatase, rutile and brookite TiO₂, Phys. Chem. Chem. Phys., 2014, 16, 20382–20386 RSC .
32. S. Yurdakal, et al., Alkaline treatment as a means to boost the activity of TiO₂ in selective photocatalytic processes, Catal. Sci. Technol., 2020, 10, 5000–5012 RSC .
33. L. Liu, H. Zhao, J. M. Andino and Y. Li, Photocatalytic CO₂ reduction with H₂O on TiO₂ nanocrystals: Comparison of anatase, rutile, and brookite polymorphs and exploration of surface chemistry, ACS Catal., 2012, 2, 1817–1828 CrossRef CAS .
34. S. Mo and W. Y. Ching, Electronic and optical properties of three phases of titanium dioxide: Rutile, anatase, and brookite, Phys. Rev. B: Condens. Matter Mater. Phys., 1995, 51, 13023–13032 CrossRef CAS PubMed .
35. M. Bellardita, A. Di Paola, B. Megna and L. Palmisano, Absolute crystallinity and photocatalytic activity of brookite TiO₂ samples, Appl. Catal., B, 2017, 201, 150–158 CrossRef CAS .
36. J. J. M. Vequizo, et al., Trapping-Induced Enhancement of Photocatalytic Activity on Brookite TiO₂ Powders: Comparison with Anatase and Rutile TiO₂ Powders, ACS Catal., 2017, 7, 2644–2651 CrossRef CAS .
37. S. M. El-Sheikh, et al., Tailored synthesis of anatase–brookite heterojunction photocatalysts for degradation of cylindrospermopsin under UV–Vis light, Chem. Eng. J., 2017, 310, 428–436 CrossRef CAS .
38. H. T. T. Tran, et al., Photocatalytic Performance of Highly Active Brookite in the Degradation of Hazardous Organic Compounds Compared to Anatase and Rutile, Appl. Catal., B, 2017, 200, 647–658 CrossRef CAS .
39. S. Pigeot-Rémy, et al., Size and shape effect on the photocatalytic efficiency of TiO₂ brookite, J. Mater. Sci., 2019, 54, 1213–1225 CrossRef .
40. K. Tomita, et al., A water-soluble titanium complex for the selective synthesis of nanocrystalline brookite, rutile, and anatase by a hydrothermal method, Angew. Chem., Int. Ed., 2006, 45, 2378–2381 CrossRef CAS PubMed .
41. A. Pottier, C. Chanéac, E. Tronc, L. Mazerolles and J.-P. Jolivet, Synthesis of brookite TiO₂ nanoparticles by thermolysis of TiCl4 in strongly acidic aqueous media, J. Mater. Chem., 2001, 11, 1116–1121 RSC .
42. F. Wu, et al., A novel method to synthesize anatase TiO₂ nanowires as an anode material for lithium-ion batteries, J. Alloys Compd., 2011, 509, 3711–3715 CrossRef CAS .
43. M. Kakihana, M. Kobayashi, K. Tomita and V. Petrykin, Application of water-soluble titanium complexes as precursors for synthesis of titanium-containing oxides via aqueous solution processes, Bull. Chem. Soc. Jpn., 2010, 83, 1285–1308 CrossRef CAS .
44. Y. V. Kolen'ko, et al., Hydrothermal synthesis and characterization of nanorods of various titanates and titanium dioxide, J. Phys. Chem. B, 2006, 110, 4030–4038 CrossRef PubMed .
45. H. Kominami, M. Kohno and Y. Kera, Synthesis of brookite-type titanium oxide nano-crystals in organic media, J. Mater. Chem., 2000, 10, 1151–1156 RSC .
46. T. A. Kandiel, A. Feldhoff, L. Robben, R. Dillert and D. W. Bahnemann, Tailored titanium dioxide nanomaterials: anatase nanoparticles and brookite nanorods as highly active photocatalysts, Chem. Mater., 2010, 22, 2050–2060 CrossRef CAS .
47. K. K. Paul, R. Ghosh and P. K. Giri, Mechanism of strong visible light photocatalysis by Ag₂O-nanoparticle-decorated monoclinic TiO₂(B) porous nanorods, Nanotechnology, 2016, 27, 315703 CrossRef PubMed .
48. Y. Morishima, et al., Hydrothermal synthesis of brookite type TiO₂ photocatalysts using a water-soluble Ti-complex coordinated by ethylenediaminetetraacetic acid, J. Ceram. Soc. Jpn., 2009, 117, 320–325 CrossRef CAS .
49. M. Kobayashi, V. Petrykin, K. Tomita and M. Kakihana, Hydrothermal synthesis of brookite-type titanium dioxide with snowflake-like nanostructures using a water-soluble citratoperoxotitanate complex, J. Cryst. Growth, 2011, 337, 30–37 CrossRef CAS .
50. I. V. Kolesnik, D. A. Kozlov, A. S. Poluboyarinov, A. V. Garshev and V. K. Ivanov, Non-classical growth of brookite nanorods, CrystEngComm, 2019, 21, 5673–5681 RSC .
51. M. Klinger, More features, more tools, more CrysTBox, J. Appl. Crystallogr., 2017, 50, 1226–1234 CrossRef CAS .
52. V. A. Lebedev, V. V. Sudin, D. A. Kozlov and A. V. Garshev, Photocatalytic properties of nanocrystalline TiO₂ modified with CuO and WO₃, Nanotechnol. Russ., 2016, 11, 20–28 CrossRef CAS .
53. Y. Zheng, E. Shi, S. Cui, W. Li and X. Hu, Hydrothermal preparation of nanosized brookite powders, J. Am. Ceram. Soc., 2000, 83, 2634–2636 CrossRef CAS .
54. T. Nagase, et al., Hydrothermal synthesis of brookite, Chem. Lett., 1999, 911–912,  DOI:10.1246/cl.1999.911 .
55. M. J. López-Muñoz, A. Revilla and G. Alcalde, Brookite TiO₂-based materials: Synthesis and photocatalytic performance in oxidation of methyl orange and As(III) in aqueous suspensions, Catal. Today, 2015, 240, 138–145 CrossRef .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ce00606b

This journal is © The Royal Society of Chemistry 2024