Stable synthesis mechanism and photocatalytic properties of TiO₂ nanocrystals with different morphologies derived from potassium titanate nanowires

Abstract

This study explored key factors influencing TiO₂ morphology using potassium titanate nanowires (KTNWs) as precursors. The pH of the hydrothermal solution was identified as critical in controlling TiO₂ morphology. Different washing methods for precursors synthesized under alkaline conditions lead to varying pH environments in the subsequent hydrothermal solutions, thereby influencing the growth direction of TiO₂ nuclei. Even a very slight pH change could cause a huge difference in the morphology of TiO₂, by adjusting precursor washing methods, rod-like, cuboidal, and octahedral bipyramidal TiO₂ nanocrystals were synthesized, the octahedral bipyramidal TiO₂ exhibiting the smallest particle size. Acid treatment could stabilize the octahedral bipyramidal morphology of the TiO₂ nanocrystal, and reduced particle size by nearly 86% than that of rod-like TiO₂-R nanocrystal. Acid treated sample TiO₂-R-H7 achieved the best photocatalytic activity, which was nearly 3 times than that of original TiO₂. Adjusting (NH₄)₂CO₃ morphology-controlling agent concentrations further regulated {001} facet exposure ratio to improve the photocatalytic activity, the TiO₂-R-0.14 synthesized at 0.14 mmol per L (NH₄)₂CO₃ showing about 2.6 times than that of original TiO₂. However, excessive (NH₄)₂CO₃ concentrations would reduce the photocatalytic activity due to increased particle size and fewer oxygen vacancies. This study provides insights into the growth mechanisms of TiO₂ morphologies and highlights acid treatment as a strategy to reduce particle size and enhance photocatalytic activity, offering guidance for designing high-performance TiO₂ based photocatalyst.

---

1. Introduction

Photocatalytic technology is extensively utilized in air purification, water treatment, and sterilization, offering an effective approach to mitigating environmental pollution. TiO₂ has become one of the most widely used photocatalysts, due to its non-toxic, chemical stability, low cost, and strong photocatalytic activity. However, the wide bandgap of TiO₂ and the rapid recombination of photogenerated electron–hole pairs restrict it to UV light excitation and result in low quantum efficiency, limiting its industrial applications.¹ To address these limitations, TiO₂ is commonly modified by methods such as ion doping, noble metal loading, semiconductor coupling, and facet engineering to enhance its photocatalytic efficiency.^2,3 Among them, facet engineering normally using morphology-controlling agents to expose the high-energy crystal face of TiO₂ to improve its photocatalytic activity. Furthermore, the heterojunction between the {001} and {101} facets can promote the migration and separation of photogenerated electron–hole pairs, further improving the photocatalytic activity of TiO₂.⁴ However, following the principle of the lowest surface energy, TiO₂ crystals will spontaneously expose the low-energy {101} crystal face with an energy of 0.44 J m⁻² under natural conditions.⁵ To expose the high-energy {001} crystal face (0.90 J m⁻²), facet-controlling agents are needed to reduce its surface energy.^6,7 Due to the slow hydrolysis rate of titanate nanowires, which is beneficial to the selective adsorption of morphology-controlling agents onto the {001} crystal face, many researchers choose titanate nanowires (K₂Ti₃O₇ or Na₂Ti₃O₇) as the precursor for morphology-controlled TiO₂ synthesis.^8,9 (NH₄)₂CO₃ is inexpensive and readily available, can reduce the surface energy of the {001} facet, and the NH₄⁺ can act as a pH buffer, providing a stable environment for TiO₂ crystal growth.¹⁰ Therefore, this study employs (NH₄)₂CO₃ as the morphology-controlling agent.

The photocatalytic activity of the photocatalyst is not only related to the exposure of the crystal facet, but also related to its specific surface area. Previous studies have shown that acid treatment of titanates can reduce the particle size of hydrothermally synthesized TiO₂ nanocrystals, increase their specific surface area, and thereby enhance their photocatalytic activity. Chacko et al. employed an alkaline hydrothermal method to convert TiO₂ into potassium titanate nanowires, which was then subjected to acid treatment and ion exchange to form a titanate precursor (H₂Ti₃O₇). Subsequent hydrothermal treatment yielded ultrafine TiO₂ nanofibers. BET analysis revealed that the modified TiO₂ had a specific surface area 2.5 times that of the pristine TiO₂.¹¹ Viet et al. conducted sodium titanate to water washing and acid treatment for 30 minutes, followed by hydrothermal treatment at 135 °C for 24 hours. The results showed that water washing produced irregular TiO₂ nanoblocks with a particle size of approximately 38 nm, while acid treatment yielded TiO₂ nanotubes with a diameter of about 10 nm and a specific surface area 2.03 times that of the original TiO₂.¹² Furthermore, acid treatment can reduce alkali metal ions in titanates and thereby modifying the morphology and particle size of hydrothermally synthesized TiO₂. Xiong et al. treated sodium titanate nanowires with HNO₃ at various concentrations for 6 hours. After acid treatment, the precursor was subjected to hydrothermal processing at 180 °C for 24 hours, the specific surface area of the produced TiO₂ was 1.9 times that of the TiO₂ synthesized without precursor acid treatment. The photocatalytic degradation efficiency of this TiO₂ for MO was twice that of commercial P25. This acid treatment reduced the Na⁺ content in the precursor from 7.37% to zero, resulting in a morphological transition of the synthesized TiO₂ from rod-shaped to spherical, indicating that its morphology is significantly influenced by alkali metal ions of the precursor.¹³ Wei et al. utilized potassium titanate nanowires as a precursor, conducting microwave-assisted hydrothermal treatment at 160 °C for 6 hours, resulting in octahedral bipyramidal TiO₂ nanocrystals with a particle size of 17.2 nm. When using acid-treated potassium titanate nanowires as the precursor, hydrothermal synthesis similarly yielded octahedral bipyramidal TiO₂ with a particle size of approximately 16.24 nm.^14,15 It has been observed that, after acid treatment of titanate nanowires, the morphology of TiO₂ obtained through hydrothermal synthesis varies among different studies. Some researchers claimed that alkali metal ion content influences TiO₂ morphology, while other experimental results show no significant effect. Therefore, it is necessary to investigate whether the alkali metal ion content in the precursor is a primary factor affecting the morphology of synthesized TiO₂.

Studies have also indicated that the pH of the hydrothermal solution can influence the morphology of TiO₂. Du et al. treated potassium titanate nanowires with HCl, subsequently adjusting the pH of the hydrothermal solution with HCl and tetramethylammonium hydroxide at a hydrothermal temperature of 180 °C. When pH ≤ 3, rod-like TiO₂ crystals with {111} facets, approximately 90 nm in length and 20 nm in width, were synthesized. At pH 5, octahedral bipyramidal TiO₂ crystals with {010} facets and a particle size of 50 nm were obtained. When pH was between 7 and 14, spindle-shaped TiO₂ nanocrystals, approximately 200 nm in length and 30 nm in width, with exposed {001} facets were produced.¹⁶ Wei et al. used titanates as precursors, performing microwave-assisted hydrothermal synthesis of TiO₂ at 160 °C for 6 hours under pH values of 3, 6, and 11. The results showed that in both acidic and neutral environments, bipyramidal TiO₂ nanocrystals of approximately 30 nm were synthesized, the particle size in neutral condition was slightly larger than acidic condition. When KOH was added to adjust the pH to 11, rod-like TiO₂ nanocrystals approximately 80 nm in length were obtained.¹⁴ Chen et al. performed hydrothermal synthesis of TiO₂ nanocrystals at 175 °C after acid treatment of titanates. At pH 3.5, they obtained tetragonal TiO₂ nanocrystals with {101} facets and a particle size of 20 nm; at pH 11.5, they synthesized prismatic TiO₂ nanocrystals with {100} facets, 150 nm in length and 30 nm in width; when the pH was raised to 13.5, irregular sheet-like TiO₂ nanocrystals with {101} facets were obtained.¹⁷ The above studies indicate that the pH of the solution significantly impacts the morphology of TiO₂ nanocrystals synthesized hydrothermally from titanate precursors. However, TiO₂ nanocrystals synthesized by different researchers under similar hydrothermal temperatures and comparable pH conditions exhibit notable differences in morphology and particle size. Our previous work also revealed significant morphological differences in TiO₂ nanocrystals synthesized in different batches under identical conditions. Therefore, when synthesizing TiO₂ nanocrystals from K₂Ti₃O₇ nanowires as a precursor, it is evident that factors such as hydrothermal pH, temperature and time, acid treatment of the precursor or not play crucial roles. Thus, it is essential to investigate the key factors affecting the morphology of TiO₂ nanocrystals under controlled reactant concentration, hydrothermal temperature, and reaction time. This approach will enable the controlled synthesis of TiO₂ nanocrystals, ensure batch-to-batch stability of photocatalysts, and allow facet engineering to further enhance the photocatalytic activity of TiO₂ nanocrystals.

In summary, this study investigates the effect of the pH of the hydrothermal solution on the morphology of TiO₂ nanocrystals by controlling the washing method of K₂Ti₃O₇ nanowire precursors. The K⁺ content in K₂Ti₃O₇ is regulated by controlling the acid treatment duration of K₂Ti₃O₇ nanowires to explore the primary factors influencing the morphology and particle size of the synthesized TiO₂. Based on stable TiO₂ morphology synthesis, K₂Ti₃O₇ nanowires treated under optimal acid conditions are used as precursors, and a series of TiO₂ nanocrystals with varied {001} facet exposure ratios are synthesized by altering the concentration of (NH₄)₂CO₃ as a morphology-controlling agent. The photocatalytic degradation performance of these nanocrystals is evaluated using methylene blue as the target pollutant, aiming to produce TiO₂ nanocrystals with controlled morphology and enhanced photocatalytic activity and to provide insights for the preparation of high-activity photocatalysts.

2. Experimental

2.1 Preparation of potassium titanate nanowires

A total of 1.0 g of anatase TiO₂ nanoparticles with a particle size of 20 nm was weighed and dispersed in 70 mL of 10 mol per L KOH solution. The mixture was subjected to ultrasonic treatment for 30 minutes to form a homogeneous white suspension. Subsequently, the suspension was transferred to a Teflon-lined stainless-steel autoclave for hydrothermal treatment at 180 °C for 16 hours. After the reaction, the autoclave was allowed to cool naturally to room temperature inside the oven. The reaction products were washed thoroughly using deionized water or a combination of acid washing followed by water washing to remove impurities. The washed products were separated via centrifugation and then dried to obtain the potassium titanate nanowire precursor (KTNWs). The synthesized potassium titanate nanowires were designated as R-KTNWs.

2.2 Potassium titanate nanowires acid treatment

1.0 g R-KTNWs precursor was added to 60 mL of 1 mol per L HCl solution. The mixture was stirred for varying durations, followed by centrifugation to separate the solid product. The separated product was washed repeatedly with deionized water until the washing solution reached neutral pH. After the final centrifugation, the product was dried and collected for further use.

2.3 Preparation of TiO₂ nanocrystals

A total of 0.1 g of R-KTNWs was dispersed in 70 mL of deionized water and subjected to ultrasonic treatment to ensure uniform dispersion. The resulting suspension was transferred into a Teflon-lined stainless-steel autoclave and subjected to hydrothermal treatment at 180 °C for 16 hours. After the reaction, the autoclave was allowed to cool naturally to room temperature in the oven. The products were washed repeatedly with deionized water until the supernatant reached a neutral pH. Finally, the products were dried to obtain the series of TiO₂ nanocrystals.

The preparation conditions are summarized in Table 1. Specifically, the sample labeled as TiO₂-R refers to TiO₂ obtained from R-KTNWs without acid treatment, while TiO₂-R-H0 denotes TiO₂ nanocrystals prepared via hydrothermal treatment following HCl solution washing. For samples labeled as TiO₂-R-HX, the value of X represents the duration of acid treatment in hours.

Table 1 Synthesis conditions of acid-treated TiO₂ nanocrystals

R-KTNWs/g	H2O/mL	Acid treatment time/h	Named
0.1	70	—	TiO2-R
0.1	70	Washing	TiO2-R-H0
0.1	70	7	TiO2-R-H7
0.1	70	24	TiO2-R-H24
0.1	70	48	TiO2-R-H48

---

2.4 Preparation of TiO₂ nanocrystals with exposed {001} facets

0.1 g R-KTNWs precursor was added to (NH₄)₂CO₃ solutions of varying concentrations. The mixture was ultrasonically dispersed to form a homogeneous suspension, which was then transferred to a Teflon-lined stainless-steel autoclave for hydrothermal treatment at 180 °C for 16 hours. After the reaction, the product was separated via centrifugation and washed thoroughly with distilled water until the supernatant became neutral. The washed product was subsequently dried for further analysis. The detailed preparation conditions are summarized in Table 2. In the sample notation TiO₂-R-H7-Y, Y denotes the concentration of (NH₄)₂CO₃ in the solution.

Table 2 Synthesis conditions of TiO₂ nanocrystals with exposed {001} facets

R-KTNWs/g	H2O/L	(NH4)2CO3/mmol L^−1	Acid treatment time/h	Named
0.1	70	0	0	TiO2-R
0.1	70	0.07	0	TiO2-R-0.07
0.1	70	0.14	0	TiO2-R-0.14
0.1	70	0.28	0	TiO2-R-0.28
0.1	70	0	7	TiO2-R-H7
0.1	70	0.07	7	TiO2-R-H7-0.07
0.1	70	0.14	7	TiO2-R-H7-0.14

---

2.5 Characterization

Scanning electron microscopy (SEM, SIGMA500, Germany) and transmission electron microscopy (TEM, Talos F200S, G2, America) were used to observe the microscopic morphology of the samples. The light absorption capacity of the samples was analyzed and measured using a UV-visible spectrometer (Agilent Carry5000, America) with BaSO₄ as the reference and a scanning wavelength range of 200–800 nm. The steady-state photoluminescence (PL) spectra were recorded using an FL fluorescence spectrophotometer (F-7000, Japan) with a stimulating wavelength range of 260 nm. The paramagnetic resonance spectrometer (EPR, Bruker MS 5000X, Germany) was used to test the active substances of the samples. The crystal structure of TiO₂ catalyst was determined by X-ray diffractometer (XRD, D8 Advance). The scanning range was 2θ = 20–80°, the scanning speed was 10° min⁻¹, the step size was 0.02°, the X-ray source was Cu Kα, the tube voltage was 40 kV, and the tube current was 100 mA.

2.6 Photocatalytic evaluation

The photocatalytic degradation of methylene blue (MB) dye was investigated under a 15 W, 365 nm UV lamp to evaluate the photocatalytic activity of the series of photocatalysts. A total of 30 mg of TiO₂ catalyst was added to 30 mL of 10 mg per L MB solution. The mixture was stirred in the dark for 30 min to achieve adsorption–desorption equilibrium. After centrifugation, the supernatant was collected, and its absorbance at 664 nm was measured and recorded as A₀. Subsequently, samples were taken at regular intervals during UV irradiation, and the corresponding absorbance values were measured and recorded as A_i. The degradation rate of the MB dye under UV irradiation was calculated using eqn (1).

At low dye concentrations, the TiO₂ photocatalytic reaction follows pseudo-first-order kinetics. The degradation kinetics were fitted using eqn (2), expressed as:

ln(c_t/c₀) = −k_appt (2)

where t represents the reaction time, k_app is the apparent rate constant, and c₀ and c_t are the pollutant concentrations at t = 0 and t = t, respectively.

3. Results and discussion

3.1 Effect of pH on the morphology of TiO₂ nanocrystals

3.1.1 Morphology of TiO₂ nanocrystals under different synthesis conditions. Fig. 1(a) presents the SEM image of potassium titanate nanowires (R-KTNWs) synthesized via hydrothermal treatment of anatase TiO₂ nanoparticles in KOH solution. During the reaction, TiO₂ nanoparticles undergo hydrolysis in KOH solution to form amorphous Ti(OH)₄, whose octahedral units connect via edge-sharing to establish a framework. K⁺ associate with the Ti(OH)₄ structure through electrostatic interactions. Over time, under high-temperature and high-pressure conditions, a fibrous network structure emerges, eventually forming R-KTNWs, as shown in Fig. 1(a).

Fig. 1 SEM images of KTNWs precursor and TiO₂ nanocrystals with different morphologies (a) R-KTNWs, (b) TiO₂-R-H0, (c) TiO₂-R, (d) TiO₂-R-alk.

Using the R-KTNWs as precursors, further hydrothermal treatment can produce a series of TiO₂ nanocrystals. Analysis of experimental results from different batches indicates that, under constant reactant concentration, hydrothermal temperature, and reaction time, the surface properties of the potassium titanate precursor are likely the critical factors determining the final morphology of the TiO₂ nanocrystals. The transformation of commercial TiO₂ to R-KTNWs occurs in a strongly alkaline environment. Achieving a completely neutral product through simple deionized water washing is challenging. Although washing steps were standardized, variations in water volume and washing duration affected the thoroughness of precursor washing. Differences in the washing degree of R-KTNWs result in varying pH of the subsequent hydrothermal solution, potentially influencing the morphology of the synthesized TiO₂ nanocrystals.

To validate this hypothesis, a series of experiments were conducted. R-KTNWs were prepared hydrothermally using anatase TiO₂ nanoparticles as the raw material. The nanowires were then subjected to different washing treatments, including acid washing followed by deionized water washing to subacidity, repeated deionized water washing to slight alkalinity, and deionized water washing resulting in weakly alkaline conditions. The treated precursors were hydrothermally processed at 180 °C for 16 h to produce TiO₂ nanocrystals, with their morphologies characterized via SEM, as shown in Fig. 1(b–d). Fig. 1(b) demonstrates that TiO₂-R-H0 nanocrystals with a bipyramidal morphology and an average particle size of approximately 52 nm were obtained when the R-KTNWs were acid-washed and then washed with water to subacidity. In contrast, washing the nanowires only with deionized water to slight alkalinity yielded rod-like TiO₂ nanocrystals under identical conditions, named TiO₂-R (Fig. 1(c)), with an average width of ∼103 nm and length of ∼303 nm. When the R-KTNWs were insufficiently washed and remained weakly alkaline, the hydrothermal process resulted in cuboidal TiO₂ nanocrystals, named TiO₂-R-alk, as shown in Fig. 1(d), with dimensions of ∼384 nm in length and ∼228 nm in width. As show in Fig. 2, the high-resolution transmission electron microscopy (HRTEM) revealed that the lattice fringes of the TiO₂ with the three different morphologies exhibited a spacing of 0.35 nm, corresponding to the {101} crystal plane, which is the most stable facet under natural conditions.

Fig. 2 TEM images of TiO₂ nanocrystals with different morphologies (a) TiO₂-R-H0, (b) TiO₂-R, (c) TiO₂-R-alk.

The results demonstrate that the washing method of R-KTNWs precursors indirectly regulates the pH environment of the subsequent hydrothermal solution. Acid washing followed by water rinsing produces a slightly acidic solution, extensive repeated water washing results in a slightly alkaline solution, and insufficient washing leads to a more alkaline solution. These variations in the hydrothermal pH environment ultimately result in TiO₂ nanocrystals with different morphologies. Although the pH of the final solution through the above three washing methods is close to 7, subtle changes in the pH of the hydrothermal solution played a critical role in controlling the morphology of TiO₂ nanocrystals. Additionally, as the pH of the solution increases, the particle size of TiO₂ nanocrystals also increases. TiO₂ nanocrystals obtained from acid-washed R-KTNWs precursors exhibit smaller particle sizes and larger specific surface areas, which are more advantageous for enhancing photocatalytic activity.

3.1.2 The growth mechanism of different TiO₂ morphologies. Based on SEM morphology and reaction mechanisms, the possible growth mechanisms of TiO₂ nanocrystals with the three distinct morphologies under different pH environments were analyze and illustrated in Fig. 3. The transformation of R-KTNWs into TiO₂ nanocrystals follows a “dissolution–recrystallization” nucleation mechanism. Under hydrothermal conditions, the R-KTNWs gradually decompose into Ti(OH)₄ fragments, –OH groups, and K⁺ ions, as described in eqn (3), this stage represents the hydrolysis process. Subsequently, Ti(OH)₄ fragments undergo rearrangement via dehydration reactions between Ti–OH and HO–Ti groups, forming octahedral TiO₂ crystal nucleus through edge-sharing, as shown in eqn (4), this stage corresponds to the condensation process.¹⁸ Later on, the TiO₂ crystal nucleus gradually grow up to form nanocrystalline TiO₂ when Ti(OH)₄ fragments diffused to their surface.

K_2−xH_xTi₃O₇ + (7 − 0.5x)H₂O → 3Ti(OH)₄ + (2 − x)KOH (3)

Ti(OH)₄ → TiO₂ + 2H₂O (4)

Fig. 3 The growth mechanism of TiO₂ nanocrystals with different morphologies synthesis during different hydrothermal condition.

Because of the anisotropy in adsorption stability of the capping agents, the adsorbates adsorbed onto specific crystallographic faces more strongly than others. This reduces the surface energy of the bound facet and hinders the growth of crystals or some crystal facets. Barnard et al. investigated the surface structures and energies of selected low-index surfaces for both rutile and anatase forms of TiO₂. They discovered that under basic conditions, the oxygen-terminated {100} facets exhibit lower surface energy compared to the oxygen-terminated {101} and {001} facets.^19,20 Yu et al. also believed that the OH⁻, which are gradually released from Na-titanate nanotubes, would preferentially attach to the anatase {100} facets. This selective adsorption lowers the surface energy of the {100} facets, thereby restricting the crystal growth along the a- and b-axes, the concentration of OH⁻ would gradually decrease with the reaction, resulting in spindle or round-rod shapes.¹⁸

As illustrated in Fig. 3, in the present system, R-KTNWs with deionized water washing resulting in weakly alkaline conditions, we consider that the following hydrothermal reaction at a fixed basic condition, the selective adsorption of OH⁻ would lower the surface energy of the {100} facets, thereby restricting the crystal growth along the a- and b-axes, the octahedral Ti(OH)₄ fragments connect edge-sharing both longitudinally and laterally, with surface-edge adjustments leading to the formation of cuboidal TiO₂ nanocrystals. With repeated deionized water washing to slight alkalinity close to neutral hydrothermal environment, the OH⁻ gradually released from R-KTNWs, would preferentially attach to the anatase {100} facets and lowers the surface energy of the {100} facets, thereby restricting the crystal growth along the a- and b-axes. In this case no extra OH⁻, the concentration of OH⁻ would gradually decrease with the reaction, the octahedral Ti(OH)₄ fragments connect edge-sharing along the equatorial plane and point-sharing longitudinally, eventually forming rod-like TiO₂ nanocrystals. R-KTNWs with acid washing followed by deionized water washing to subacidity close to neutral, part of K₂Ti₃O₇ nanowires were transformed into H₂Ti₃O₇ nanowires. The splitting of the nanowires maybe along the {101} and {−101} facets and the growth of nanocrystals are accompanied by Ostwald ripening process, ultimately forming octahedral bipyramidal TiO₂ nanocrystals.²¹

Studies have shown that the morphology of TiO₂ nanocrystals is primarily governed by the growth rates along the [001] and [101] directions.²² Under natural conditions, the surface energy of the {001} facet is higher (0.90 J m⁻²), and the growth rate along the [001] direction is faster. According to the principle of minimum surface energy, the {101} crystal plane with lower surface energy (0.44 J m⁻²) will eventually be exposed to minimize the surface energy.²³ By adjusting the pH of the solution through different washing methods, we obtained TiO₂ nanocrystals with varied morphologies, all contain {101} crystal plane, as shown in Fig. 2. This observation indicated that, the change of the pH of the hydrothermal solution will change the crystal growth mechanism to obtain TiO₂ with different morphologies, but will not change the characteristics of tending to expose low-energy {101} crystal facet to minimize the total surface energy of TiO₂ and make the system the most stable.

3.2 Effect of acid treatment of R-KTNWs on the morphology of TiO₂ nanocrystals

3.2.1 Morphological changes of R-KTNWs induced by acid treatment. The above findings indicate that the properties of R-KTNWs play a crucial role in determining the morphology of subsequently synthesized TiO₂ nanocrystals. Acid washing of R-KTNWs results in the formation of bipyramidal TiO₂ nanocrystals with smaller particle sizes. Additionally, previous studies have reported that TiO₂ nanocrystals synthesized after exposing potassium titanate nanowires to an acidic environment for a certain period exhibit smaller particle sizes, which contributes to an increased specific surface area of TiO₂.²⁴ To further enhance the photocatalytic activity of TiO₂, we immersed the R-KTNWs precursor in HCl solution and stirred the mixture continuously at room temperature. This study aims to investigate the influence of acid treatment duration on the morphology, particle size, and photocatalytic activity of the resulting TiO₂ nanocrystals.

Fig. 4(a–c) show SEM images of R-KTNWs precursors subjected to different acid treatment durations. Fig. 4(a) and (b) represent the SEM images of untreated R-KTNWs and those treated with acid for 7 hours, respectively. In both images, the R-KTNWs exhibit a uniform size distribution and clear, well-defined structures, with the nanowires intertwined. There are no significant morphological changes before and after acid treatment, indicating that a certain duration of acid treatment does not alter the macrostructure of the R-KTNWs. Liu et al. treated sodium titanate nanowires with 0.6 mol per L HCl for 1 h, HRTEM also showed that the morphology of the nanowires did not change before and after acid treatment, but the lattice fringe spacing changed from 0.81 nm to 0.78 nm after Na⁺ were replaced by H⁺.²⁵ As shown in Fig. 4(c), after further extending the acid treatment time to 48 h, although the appearance morphology still maintains the nanowire structure, the nanowires are significantly shortened and messy compared with the non-acid treatment, and some nanowires also appear to adhere to each other. If the acid treatment is prolonged to 72 h, the R-KTNWs nanowires are almost completely dissolved in the acid solution, indicating that the acid treatment duration should not be excessively long. Nakahira et al. also claimed that the symmetry of the Ti environment in TiO₂-derived titanate nanotubes has hardly changed with the post HCl treatment, local structures around Ti such as the coordination geometry in TiO₂-derived titanate nanotubes have changed with the acid treatment using HCl, and the excess acid treatment led to the disordering, and resulted in the change of the morphology.²⁶

Fig. 4 SEM images of R-KTNWs nanowires treated with acid for different time, (a) 0 h, (b) 7 h, (c) 48 h, (d) EDS spectra.

Therefore, it is essential to strictly control the preparation and washing processes of R-KTNWs in experiments to ensure that the surface properties of the nanowires remain consistent across batches and to minimize their impact on subsequent experiments. Fig. 4(d) shows the EDS spectra of R-KTNWs before and after 7 hours of acid treatment. The spectra reveal a significant reduction in the relative content of K⁺ after acid treatment, with most K⁺ ions being replaced by H⁺ ions. The proportion of K⁺ decreases from approximately 10% to about 3% after 7 hours of acid treatment. As discussed previously, after K⁺ were replaced by H⁺, both of the interlayer spacing and the thermal stability of the nanowires decreased.²⁶ The influence of the acid treatment duration of R-KTNWs precursors on the particle size, morphology, and photocatalytic activity of the TiO₂ nanocrystals were further investigated by TEM and photocatalytic activity test.

3.2.2 The size and morphology changes of the synthesized TiO₂. Fig. 5 presents the TEM images of TiO₂ nanocrystals synthesized from R-KTNWs precursors subjected to different acid treatment durations. The results indicate that, similar to acid washing, acid treatment also could produce octahedral bipyramidal TiO₂ nanocrystals. The average particle size of different TiO₂ samples in TEM images was measured by Nano Measure software, the particle size distribution of TiO₂ nanocrystals synthesized after different acid treatment time for R-KTNWs precursors is shown in Fig. S1.† As shown in Fig. 5(a) and S1(a),† TiO₂ nanocrystals with a dominant octahedral bipyramidal morphology and an average particle size of approximately 42 nm were obtained after hydrothermal treatment of R-KTNWs acid-treated for 7 h. Extending the acid treatment to 24 h resulted in octahedral bipyramidal TiO₂ nanocrystals with a reduced average particle size of ∼37.5 nm, accompanied by a small proportion of plate-like TiO₂ nanocrystals, as shown in Fig. 5(b) and S1(b).† For R-KTNWs acid-treated for 48 h, the resulting TiO₂ nanocrystals included bipyramidal structures along with a higher proportion of irregular plate-like morphologies, likely due to damage to the precursor structure. The average particle size of TiO₂ nanocrystals at this stage was approximately 37.2 nm, as shown in Fig. 5(c) and S1(c).†

Fig. 5 TEM images of TiO₂ nanocrystals synthesized after acid treatment for different time, (a) TiO₂-R-H7, (b) TiO₂-R-H24, (c) TiO₂-R-H48.

3.2.3 Mechanism analysis of acid treatment of R-KTNWs to change the morphology of TiO₂. The acid treatment of the precursor transforms the morphology of the TiO₂ from rod-like to the typical bipyramidal structure, with particle size decreasing as the acid treatment duration increases. This indicates that acid treatment can effectively control the morphology and particle size of TiO₂ nanocrystals, promoting the formation of bipyramidal TiO₂ nanocrystals. However, excessively long acid treatment durations are undesirable, as the proportion of bipyramidal TiO₂ structures decreases with prolonged treatment, leading to the emergence of other irregular morphologies. This indicates that acid treatment effectively controls the morphology and particle size of TiO₂ nanocrystals, promoting the formation of bipyramidal TiO₂ nanocrystals.

It can be concluded that the variation in K⁺ content within potassium titanate nanowires is not the primary factor driving changes in TiO₂ morphology. Instead, the pH of the hydrothermal solution is the main determinant of the formation of bipyramidal TiO₂ nanocrystals. In acidic environments, TiO₂ nuclei grow into bipyramidal structures through point-sharing and edge-sharing mechanisms.²⁷ The reduction in particle size is attributed to the decrease in K⁺ content caused by acid treatment, which reduces the interlayer spacing of potassium titanate and ultimately results in smaller TiO₂ nanocrystals.²⁸ Smaller particle sizes correspond to larger specific surface areas, which are beneficial for enhancing photocatalytic activity.²⁹ However, excessively long acid treatment durations can overly reduce K⁺ content, compromise the stability of potassium titanate nanowires, and decrease the crystallinity of TiO₂ nanocrystals, leading to the formation of irregular morphologies.³⁰

The mechanism of morphology and particle size changes in TiO₂ nanocrystal after acid treatment of R-KTNWs was illustrated in Fig. 6. During acid treatment, K⁺ ions in R-KTNWs are gradually replaced by H⁺ ions. With prolonged acid treatment, the structurally stable R-KTNWs undergoes protonation to form H₂Ti₃O₇. H₂Ti₃O₇ exhibits weak acidity and a negative zeta potential. During hydrothermal conversion, its negatively charged surface attracts H⁺ ions from the reaction medium, resulting in the protonation of Ti–OH groups to form Ti–OH₂⁺.³¹ Two Ti(OH)₄ fragments then undergo rearrangement and dehydration to polymerize into TiO₂ crystal nucleus, which grow along the {101} plane.

Fig. 6 The growth mechanism of TiO₂ nanocrystals after precursor acid treatment.

This transition from H₂Ti₃O₇ to anatase TiO₂ nanocrystal involves the in situ dehydration and rearrangement of Ti(OH)₄ structural units and is thus considered a topochemical reaction, ultimately yielding octahedral bipyramidal TiO₂ nanocrystals.³¹ The replacement of K⁺ ions in R-KTNWs during this process reduces the interlayer spacing, which in turn decreases the particle size of the TiO₂ nanocrystals.

3.3 Photocatalytic activity of acid-treated TiO₂ nanocrystals

To explore the relationship between precursor acid treatment conditions and the photocatalytic activity of the synthesized TiO₂ nanocrystals, a series of TiO₂ nanocrystals were subjected to methylene blue (MB) photocatalytic degradation experiments. As illustrated in Fig. 7, TiO₂ nanocrystals prepared from acid-treated R-KTNWs precursors exhibited better photocatalytic activity than those prepared without acid treatment (TiO₂-R). However, as the acid treatment duration increased, the photocatalytic activity of the TiO₂ nanocrystals initially increased and then decreased. Among the samples, TiO₂-R-H7, synthesized from a precursor acid-treated for 7 hours, showed the fastest degradation rate of MB, with photocatalytic activity nearly 3 times and 5 times higher than that of the original anatase TiO₂ and TiO₂-R, respectively.

Fig. 7 Photocatalytic activity diagram of TiO₂ nanocrystals synthesized after precursor acid treatment (a) photocatalytic degradation diagram of MB, (b) photodegradation rate diagram.

The enhanced photocatalytic activity can be attributed to the morphological and particle size changes of the TiO₂ nanocrystals after acid treatment. As previously analyzed, the acid treatment of R-KTNWs precursors transforms the morphology of the synthesized TiO₂ from nanorods to bipyramidal structures, with particle size decreasing from 303 nm to 42 nm or smaller. The improvement in TiO₂ crystallinity and the increase in specific surface area significantly enhance the photocatalytic activity of the TiO₂ nanocrystals. As show in Fig. 5, with the extension of acid treatment time to 24 h or even 48 h, although the particle size was further reduced to 37.2 nm, the proportion of octahedral bipyramidal TiO₂ structures decreases, and irregular morphologies emerge due to the excess acid treatment, resulting in a decline in photocatalytic activity. Additionally, acid treatment converts most of the Ti–O–K bonds on the surface of R-KTNWs into Ti–OH bonds, as H⁺ replaces the larger K⁺ ions. This substitution increases the number of surface oxygen vacancies, the hydroxyl group content, thereby enhancing the photocatalytic activity. However, with the excess acid treatment, the content of hydroxyl groups and oxygen vacancies decreases due to the continued action of the acid, leading to a reduction in photocatalytic activity.^32,33

The oxygen vacancies of acid-treated TiO₂ nanocrystals were analyzed using EPR, and the results are shown in Fig. 8. Theoretically, within a certain range, a higher content of oxygen vacancies is beneficial for enhancing photocatalytic activity. However, excessive oxygen vacancies can act as recombination centers for photogenerated charge carriers, leading to a decline in photocatalytic activity of TiO₂.³⁴

Fig. 8 Oxygen vacancy diagram of TiO₂ nanocrystals synthesized by acid treatment for different time.

As shown in Fig. 8, no oxygen vacancies were detected in the original anatase TiO₂. The rod-like TiO₂-R, synthesized from untreated precursors, exhibited the highest oxygen vacancy content at g = 2.003. The bipyramidal TiO₂-R-H7, obtained after 7 hours of acid treatment, displayed a slightly lower oxygen vacancy peak intensity. The oxygen vacancy content further decreased in the TiO₂-R-H24 sample synthesized after 24 hours of acid treatment. These findings indicate that the synthesis method employed in this study effectively promotes the formation of oxygen vacancies in TiO₂. However, prolonged acid treatment of KTNWs will reduce the oxygen vacancy content in TiO₂.

From the photocatalytic activity results in Fig. 7, it can be observed that, despite the high oxygen vacancy content, the rod-like TiO₂-R nanocrystals exhibit relatively low photocatalytic activity due to their larger particle size. As the precursor acid treatment time increases, the oxygen vacancy content on the TiO₂ surface gradually decreases, and the crystallinity deteriorates, ultimately resulting in reduced photocatalytic activity. This demonstrates that the photocatalytic activity of TiO₂ is influenced by multiple factors, including particle size, morphology, and oxygen vacancy content. Moderate acid treatment of KTNWs precursors provides a stable approach for synthesizing bipyramidal TiO₂ nanocrystals with reduced particle size and increased surface oxygen vacancy content. This is one of the effective methods to enhance the photocatalytic activity of TiO₂.

3.4 Morphology of TiO₂ nanocrystals controlled by (NH₄)₂CO₃

Based on the previous work, this study employed (NH₄)₂CO₃ as a morphology-controlling agent to investigate whether exposing highly active {001} crystal facets while maintaining small particle sizes can further enhance the photocatalytic activity of TiO₂. Fully washed and acid-treated 7 h R-KTNWs precursors were selected for subsequent crystal facet control experiments. By varying the concentration of the (NH₄)₂CO₃ solution during the hydrothermal process, the exposure ratio of the {001} crystal facets was controlled. The morphologies of the two groups of TiO₂ samples are shown in Fig. 9.

Fig. 9 SEM images of TiO₂ nanocrystals prepared under different conditions, (a) TiO₂-R, (b) TiO₂-R-0.07, (c) TiO₂-R-0.14, (d) TiO₂-R-0.28, (e) TiO₂-R-H7, (f) TiO₂-R-H7-0.07, (g) TiO₂-R-H7-0.14, (h) TiO₂-R-H7-0.28.

As shown in Fig. 9(a), using fully washed R-KTNWs as precursors, TiO₂-R samples without (NH₄)₂CO₃ were hydrothermally synthesized into rod-like TiO₂ nanocrystals with a length of 303 nm and a width of 103 nm. When varying concentrations of (NH₄)₂CO₃ morphology-controlling agents were introduced during hydrothermal synthesis using the same R-KTNWs precursors, truncated decahedral TiO₂ nanocrystals were obtained. Fig. 9(b–d) illustrate the morphologies of TiO₂-R-0.07, TiO₂-R-0.14, and TiO₂-R-0.28, corresponding to (NH₄)₂CO₃ concentrations of 0.07 mmol L⁻¹, 0.14 mmol L⁻¹, and 0.28 mmol L⁻¹, respectively. The average particle sizes of these samples were 68 nm, 98 nm, and 105 nm, respectively. Calculations showed that the proportion of exposed {001} facets was 7%, 11%, and 15%, respectively. As the proportion of {001} facets increased, the particle size of the TiO₂ nanocrystals also gradually increased. The study of Han et al. also shown the phenomenon that the particle size of TiO₂ nanocrystals increases with the increase of urea concentration.³⁵ In this work, the adsorption of NH₄⁺ on the {001} facet reduced the surface energy, changed the crystal growth direction and finally exposed the {001} facet. It is speculated that the higher the concentration of (NH₄)₂CO₃, the more NH₄⁺ adsorbed on the {001} plane, and the proportion of {001} plane also increased, which ultimately increased the particle size.

As shown in Fig. 9(e), when R-KTNWs acid-treated for 7 hours were used as precursors without the addition of (NH₄)₂CO₃, bipyramidal TiO₂ nanocrystals with an average particle size of approximately 42 nm were obtained. Compared to the rod-like TiO₂ shown in Fig. 9(a), the particle size was reduced by approximately 86%. When the acid-treated 7 h R-KTNWs were hydrothermally synthesized in (NH₄)₂CO₃ solutions with concentrations of 0.07 mmol L⁻¹, 0.14 mmol L⁻¹, and 0.28 mmol L⁻¹, truncated decahedral TiO₂ nanocrystals exposing {001} facets were similarly obtained, as shown in Fig. 9(f–h). The proportions of exposed {001} facets were approximately 6%, 10%, and 12%, respectively, and the average particle sizes of the TiO₂ nanocrystals were about 71 nm, 85 nm, and 93 nm. The small particles observed on the TiO₂ particle surfaces are gold clusters for the SEM test.

These results demonstrate that the (NH₄)₂CO₃ morphology-controlling agent promotes the formation of {001} crystal facets, while the acid treatment of nanowire precursors significantly reduces the particle size of TiO₂ nanocrystals and increases their specific surface area.

To further confirm the presence of {001} facets in TiO₂, HRTEM characterization was performed on the TiO₂-R-0.14 sample. The results, shown in Fig. 10, reveal lattice fringes with a spacing of 0.35 nm, corresponding to the {101} crystal plane of TiO₂ nanocrystals, and a spacing of 0.24 nm, corresponding to the {001} crystal plane. These findings indicate that (NH₄)₂CO₃ effectively reduces the surface energy of the TiO₂ {001} facets, facilitating their exposure.¹⁰

Fig. 10 TEM and HRTEM images of TiO₂-R-0.14 nanocrystals, (a) TEM image of TiO₂-R-0.14 nanocrystals, (b) {001} crystal plane partial enlarged detail, (c) {101} crystal plane partial enlarged detail.

We selected several typical samples of original TiO₂, TiO₂-R-0.14, TiO₂-R-H-0.07 for XRD detection, the results are shown in Fig. S2.† It can be seen from the figure that the diffraction peaks of TiO₂-R-0.14 and TiO₂-R-H-0.07 correspond to those of anatase TiO₂. It can be seen that pure anatase TiO₂ nanocrystals were prepared by crystal plane control in this work.

3.5 Photocatalytic activity of TiO₂ nanocrystals with exposed {001} facets

3.5.1 Effect of (NH₄)₂CO₃ on photocatalytic activity of non-acid treated series TiO₂ nanocrystals. To investigate the effect of crystal facet engineering on the photocatalytic activity of TiO₂ nanocrystals, photocatalytic degradation experiments of MB were conducted under UV light irradiation. As shown in Fig. 11(a and b), TiO₂-R, synthesized by direct hydrothermal treatment of fully washed R-KTNWs without facet engineering, exhibited the weakest photocatalytic activity, followed by original anatase TiO₂. The photocatalytic activity of TiO₂ synthesized from fully washed R-KTNWs improved with increasing concentrations of (NH₄)₂CO₃ but then declined beyond a certain concentration. The TiO₂-R-0.14 sample, synthesized at an (NH₄)₂CO₃ concentration of 0.14 mmol L⁻¹, showed the highest photocatalytic activity, with a degradation rate about 3.6 times that of the TiO₂-R sample and 2.6 times that of the original anatase TiO₂.

Fig. 11 Photocatalytic activity diagram of TiO₂ nanocrystals with exposed {001} crystal plane, (a) photocatalytic degradation of MB by synthesized TiO₂ without acid treatment, (b) photodegradation rate diagram.

The low activity of the rod-like TiO₂-R sample can be attributed to its large particle size, with a length of 303 nm, 15.2 times larger than the particle size of the original anatase TiO₂. The improvement in photocatalytic activity upon introducing the (NH₄)₂CO₃ morphology-controlling agent is primarily due to the exposure of the high-energy {001} facets. Additionally, the formation of heterojunctions between the {101} and {001} facets also enhanced the separation efficiency of electron–hole pairs. This is because of preferential flow of photogenerated holes and electrons to the oxidative {001} facets and reductive {101} facets, on which oxidation and reduction occur respectively.³⁶ In particular, Maitani et al. demonstrated that the magnitude of electron transfer from anthracene derivatives to the {001} facet is 10 times larger than that to the {101} facets of TiO₂ nanocrystals.³⁷ This finding unambiguously suggests that the {001} facets are significantly more enriched with positively charged holes compared to the {101} facets, rendering the {001} facets more reactive for photooxidation processes. Thus, the coexistence of both the {001} facets (oxidation site) and {101} facets (reduction site) facilitate the prolonged separation of photogenerated electron–hole pairs. This spatial separation enhances the efficiency of TiO₂ nanocrystals in the photocatalytic degradation of MB. However, as shown in Fig. 9(d), when the concentration of (NH₄)₂CO₃ reached 0.28 mmol L⁻¹, the average particle size of TiO₂-R-0.14 sample increased to 105 nm and significant defects appeared on the surfaces of the TiO₂ nanocrystals, along with the emergence of many irregularly shaped particles. The increase of the particle size would decrease the specific surface area of the TiO₂ crystals, and the defects would act as recombination centers for electron–hole pairs, ultimately reducing the photocatalytic activity of the TiO₂ nanocrystals. This indicates that although the (NH₄)₂CO₃ morphology-controlling agent can promote the formation of high-energy {001} crystal facet of TiO₂, considering the comprehensive factors affecting the photocatalytic activity, the concentration of (NH₄)₂CO₃ in the synthesis process should not be excessively high.

The photocatalytic activity of samples with crystal facet engineering was higher than that of the original TiO₂ nanoparticles. Although the photocatalytic activity of the TiO₂-R-0.14 nanocrystals was about 2.6 times that of the original TiO₂ nanoparticles, their particle size was approximately 98 nm, which is about 4.9 times larger than the particle size of the original anatase TiO₂. This indicates that crystal facet engineering effectively enhances the photocatalytic activity of TiO₂, but excessively large particle size remains a limiting factor for further improvement. Therefore, it is necessary to further reduce the particle size while maintaining the benefits of crystal facet engineering.

3.5.2 Effect of (NH₄)₂CO₃ on photocatalytic activity of acid-treated series TiO₂ nanocrystals. As shown in Fig. 9, compared with non-acid treatment, acid treatment of R-KTNWs followed by crystal facet engineering effectively reduces particle size. The photocatalytic activity of the resulting TiO₂ nanocrystals is shown in Fig. 12(a and b). From Fig. 12(b), it can be seen that TiO₂ nanocrystals synthesized after a 7 h acid treatment of R-KTNWs exhibit better photocatalytic activity compared to the original TiO₂ nanoparticles. The photocatalytic activities of the samples TiO₂-R-H7, TiO₂-R-H7-0.07, and TiO₂-R-H7-0.14, synthesized at different (NH₄)₂CO₃ concentrations, were about 2.9, 1.9, and 1.4 times that of anatase TiO₂, respectively. Among these, the TiO₂-R-H7 sample, synthesized without (NH₄)₂CO₃, exhibited the highest photocatalytic activity, with nearly complete degradation of MB after 120 min of UV irradiation, indicating that acid treatment enhances the photocatalytic activity of TiO₂ nanocrystals.

Fig. 12 Photocatalytic activity diagram of TiO₂ nanocrystals with exposed {001} crystal plane, (a) photocatalytic degradation of MB by synthesized TiO₂ with acid treatment, (b) photodegradation rate diagram.

However, different from the series of samples without acid treatment, introducing (NH₄)₂CO₃ to synthesize TiO₂ nanocrystals from 7 h acid-treated R-KTNWs did not further enhance the photocatalytic activity. Instead, as the (NH₄)₂CO₃ concentration increased, the photocatalytic activity of the synthesized TiO₂ nanocrystals gradually decreased. Based on Fig. 9 SEM morphology and particle size analysis, this reduction in photocatalytic activity may be attributed to the lower proportion of {001} facets in the acid-treated samples, which limits the formation of crystal facet heterojunctions. Additionally, the increased particle size negatively impacts photocatalytic activity. The particle sizes of TiO₂-R-H7-0.07 and TiO₂-R-H7-0.14 were approximately 71 nm and 85 nm, respectively, which are about 1.7 and 2 times larger than that of the TiO₂-R-H7 sample.

3.5.3 The relationship between the concentration of (NH₄)₂CO₃ and the photocatalytic degradation rate constant of a series of TiO₂ nanocrystals. To visually demonstrate the MB degradation activity of the TiO₂ samples, the relationship between photocatalytic degradation rate constants and (NH₄)₂CO₃ concentration was plotted. As shown in Fig. 13, the degradation rate constant of the original anatase TiO₂ was 0.0074. The rate constants for TiO₂-R, TiO₂-R-0.07, TiO₂-R-0.14, and TiO₂-R-0.28, prepared from fully washed R-KTNWs with crystal facet control, were 0.0054, 0.0144, 0.01959, and 0.01031, respectively. The most active sample, TiO₂-R-0.14, exhibited a rate constant about 2.6 times that of the original TiO₂.

Fig. 13 The relationship between the concentration of (NH₄)₂CO₃ and the photocatalytic degradation rate constant of a series of TiO₂ nanocrystals.

For samples synthesized from R-KTNWs acid-treated for 7 hours, the rate constants of TiO₂-R-H7, TiO₂-R-H7-0.07, and TiO₂-R-H7-0.14 were 0.0213, 0.0137, and 0.0105, respectively, all of which were higher than those of the non-acid-treated series. Among these, TiO₂-R-H7 had the highest rate constant, approximately 2.9 times that of the original anatase TiO₂. These results indicate that TiO₂-R-H7, synthesized from acid-treated potassium titanate without the addition of (NH₄)₂CO₃, exhibited the highest photocatalytic activity. Adding (NH₄)₂CO₃ did not further enhance the activity.

From the SEM morphology images, it can be observed that the acid-treated samples exhibit a relatively low proportion of exposed {001} facets than no acid-treated samples at the same (NH₄)₂CO₃ concentration. Therefore, at this scale, changes in particle size have a greater impact on their photocatalytic activity. This highlights that acid treatment of potassium titanate precursors could reduce the particle size of TiO₂ during crystal facet engineering, effectively enhancing the photocatalytic activity of TiO₂ nanocrystals. The acid-treated precursors exhibit surface acidity, which may influence subsequent crystal facet engineering. To further improve the photocatalytic activity of TiO₂ through crystal facet control, it is necessary to explore the optimal concentration of (NH₄)₂CO₃ as a morphology-controlling agent and investigate the structure–activity relationship between acid treatment and crystal facet engineering.

3.6 Spectral analysis of the synthesized TiO₂ nanocrystals

3.6.1 PL analysis. The separation and recombination of photogenerated electron–hole pairs in two series of TiO₂ nanocrystals, synthesized from precursors treated with 7 hours acid washing and fully water washing with crystal facet control, were investigated using photoluminescence (PL) spectroscopy. The results are shown in Fig. 14. The characteristic peak at 400 nm corresponds to the intrinsic bandgap emission of TiO₂ due to electron–hole recombination, while other peaks may result from oxygen vacancies or other radical excitations.³⁸ Theoretically, lower peak intensities indicate higher electron–hole separation efficiency.³⁹

Fig. 14 Photoluminescence spectra of facet-controlled TiO₂ nanocrystals (a) synthesized TiO₂ without acid treatment, (b) synthesized TiO₂ with acid treatment.

As shown in Fig. 14(a), the PL peak intensity of fully washed, non-acid-treated TiO₂ nanocrystals after crystal facet control initially decreases and then increases with increasing (NH₄)₂CO₃ concentration. The TiO₂-R-0.14 sample, synthesized with 0.14 mmol per L (NH₄)₂CO₃, exhibited the lowest PL peak intensity, indicating the lowest electron–hole recombination rate and the highest photocatalytic activity. This trend is consistent with the photocatalytic activity trend shown in Fig. 11(a).

Fig. 14(b) shows that when R-KTNWs acid-treated for 7 hours are subjected to crystal facet control, the PL intensities of TiO₂-R-H7, TiO₂-R-H7-0.07, and TiO₂-R-H7-0.14 are all lower than those of the original TiO₂ nanoparticles. However, as the (NH₄)₂CO₃ concentration increases, the PL peak intensity gradually increases, corresponding to higher electron–hole recombination rates and weaker photocatalytic activity.

A comparison of the results for the 7 h acid-treated and non-acid-treated series reveals that all acid-treated samples exhibit lower PL peak intensities than their non-acid-treated counterparts, indicating that acid treatment effectively promotes carrier separation. Among these, the bipyramidal TiO₂-R-H7 sample, synthesized from 7 h acid-treated potassium titanate without (NH₄)₂CO₃, demonstrated the highest electron–hole separation efficiency and the strongest photocatalytic activity. This PL result aligns with the photocatalytic activity tests.

3.6.2 UV-vis analysis. The optical response range of different series of facet-controlled TiO₂ nanocrystals was analyzed using UV-visible diffuse reflectance spectroscopy, as shown in Fig. 15. It indicates that, compared to the original anatase TiO₂, the synthesized TiO₂ samples in this study exhibit stronger absorption intensity in the range of 400–650 nm, suggesting an expanded light absorption range.

Fig. 15 UV-vis diffuse reflectance spectra of TiO₂ nanocrystals synthesized under different conditions, (a) TiO₂ synthesized without acid treatment, (b) TiO₂ synthesized after 7 h acid treatment.

The bandgap values were determined by drawing tangents at the steepest descent points of the absorption curves,⁴⁰ as shown in Fig. S3.† The original TiO₂ had a bandgap of 2.8 eV, while the bandgap values for the fully washed facet-controlled TiO₂-R series (TiO₂-R, TiO₂-R-0.07, TiO₂-R-0.14, TiO₂-R-0.28) were 2.75 eV, 2.68 eV, 2.6 eV, and 2.7 eV, respectively. After 7 hours of acid treatment, the bandgap values for the crystal-facet-engineered TiO₂–H7 series (TiO₂-R-H7, TiO₂-R-H7-0.07, TiO₂-R-H7-0.14) were 2.52 eV, 2.77 eV, and 2.74 eV, respectively.

For the fully washed R-KTNWs precursor, the bandgap of the crystal-facet-engineered TiO₂ decreased and then increased with the rise in (NH₄)₂CO₃ concentration, with TiO₂-R-0.14 exhibiting the smallest bandgap of 2.6 eV. Similarly, for the 7 h acid-treated R-KTNWs precursor, the bandgap of the TiO₂–H7 series also decreased and then increased with (NH₄)₂CO₃ concentration. The TiO₂-R-H7 sample, synthesized without (NH₄)₂CO₃, had the smallest bandgap of 2.52 eV.

The UV-visible diffuse reflectance trends align with the photocatalytic activity trends shown in Fig. 11 and 12. Both acid treatment and crystal facet engineering caused a redshift in the light absorption edge of TiO₂, with the TiO₂-R-H7 sample exhibiting the largest redshift. This sample also demonstrated the highest electron–hole separation efficiency and the strongest corresponding photocatalytic activity.

3.6.3 EPR analysis. EPR analysis was conducted to investigate oxygen vacancies and further explain the trends in photocatalytic activity for the TiO₂ samples. An appropriate amount of oxygen vacancies can enhance the photocatalytic activity of TiO₂.⁴¹ As illustrated in Fig. 16(a), no significant oxygen vacancy characteristic peaks were detected for the original TiO₂. However, distinct oxygen vacancy peaks were observed for the TiO₂-R and TiO₂-R-0.14 samples synthesized from fully washed R-KTNWs precursors. Among these, TiO₂-R-0.14, which exhibited the strongest photocatalytic activity in this series, also showed higher oxygen vacancy peak intensity.

Fig. 16 Oxygen vacancy detection of facet-controlled TiO₂ nanocrystals synthesized under different conditions, (a) no acid treatment, (b) acid treatment for 7 h.

The oxygen vacancy intensities for the acid-treated precursor series are shown in Fig. 16(b). The TiO₂-R-H7 sample, synthesized without (NH₄)₂CO₃, displayed relatively high oxygen vacancy peak intensity. In contrast, the TiO₂-R-H7-0.07 and TiO₂-R-H7-0.14 samples, synthesized with the addition of (NH₄)₂CO₃, showed significantly reduced oxygen vacancy intensities. It is speculated that the decline in photocatalytic activity for this series with increasing (NH₄)₂CO₃ concentration is not only related to the increase in TiO₂ nanocrystal particle size but also potentially to the decrease in oxygen vacancy content.

Based on the oxygen vacancy analysis, it can be concluded that moderate acid treatment of potassium titanate nanowire precursors enables the stable synthesis of small-sized bipyramidal TiO₂ nanocrystals while maintaining a sufficient level of oxygen vacancies. This effectively enhances the photocatalytic activity of TiO₂.

3.7 Stable synthesis route of TiO₂ nanocrystal photocatalysts with different morphologies

Highly active TiO₂ nanocrystal photocatalyst can be synthesized by controlling the crystal facet derived from KTNWs nanowire precursor. But discrepancies in the size and morphology of TiO₂ nanocrystals synthesized using similar methods by different researchers highlight the need to explore the key factors influencing the morphology of TiO₂ nanocrystal. As show in Fig. 17, this study explored key factors influencing TiO₂ nanocrystal morphology using KTNWs as precursors.

Fig. 17 Stable synthesis route of TiO₂ nanocrystal photocatalysts with different morphologies.

We found that subtle changes in the pH of the hydrothermal solution played a critical role in controlling the morphology of TiO₂ nanocrystals. The transformation of commercial TiO₂ to R-KTNWs occurs in a strongly alkaline hydrothermal environment. It is difficult to achieve a completely neutral product through simple deionized water washing. Although washing steps were standardized, the difference in washing time and water dosage would also cause the difference in the surface acidity and alkalinity of the R-KTNWs precursor, which would affect the pH environment of the subsequent hydrothermal process, thereby influencing the growth direction of TiO₂ nuclei. We have realized the controllable synthesis of TiO₂ nanocrystals by simply adjusting the R-KTNWs precursor washing method, including deionized water washing resulting in weakly alkaline conditions, repeated deionized water washing to slight alkalinity close to neutral, and acid washing followed by deionized water washing to subacidity close to neutral, and cuboidal, rod-like, and octahedral bipyramidal TiO₂ nanocrystals could be obtained, respectively. In the above several synthesis methods, acid treatment could stabilize the octahedral bipyramidal morphology of the TiO₂ nanocrystal, and with the extension of acid treatment time, the particle size can be reduced. In this work, the TiO₂-R-H7 sample synthesized after acid treatment of R-KTNWs for 7 h can stably obtain a uniform octahedral bipyramid structure with an average particle size of 42 nm, showing the best photocatalytic degradation of MB activity. By varying the concentration of the (NH₄)₂CO₃ morphology control agent during the hydrothermal process, the exposure ratio of the TiO₂ {001} crystal facets could be controlled. Exposing the highly active {001} facets of TiO₂ could form a heterojunction of {001} and {101} facets. The electrons tend to transfer to the {101} facets, and the holes tend to transfer to the {001} facets, which promotes the spatial separation of electrons and holes, thereby improving the photocatalytic activity of TiO₂ nanocrystals.⁴² However, in this work, after the combination of acid treatment and crystal facet regulation, the exposure of {001} facet did not further improve the photocatalytic activity of TiO₂ due to the increase of particle size and insufficient exposure of {001} facet. The synergistic effect between acid treatment and crystal facet regulation needs to be further explored.

4. Conclusions

This study demonstrates that the morphology and photocatalytic activity of TiO₂ nanocrystals can be effectively controlled by adjusting the washing and acid treatment of R-KTNWs. Different washing methods for precursors synthesized under alkaline conditions lead to varying pH environments in the subsequent hydrothermal solutions, thereby influencing the growth direction of TiO₂ nuclei. A weakly alkaline hydrothermal solution produced cuboidal TiO₂ nanocrystals, a slight alkalinity near-neutral solution yielded rod-like TiO₂ nanocrystals, and a subacidity near-neutral solution resulted in octahedral bipyramidal TiO₂ nanocrystals. Acid treatment could reduce particle size and stabilize octahedral bipyramidal TiO₂ synthesis was considered as the best synthesis method in this work. By controlling the acid treatment duration, it was found that the reduction in K⁺ content would decrease the interlayer spacing of R-KTNWs, leading to smaller TiO₂ particle sizes. However, prolonged acid treatment would reduce structural stability, crystallinity, and photocatalytic performance.

TiO₂-R-H7, synthesized from precursors acid-treated for 7 hours shown the best photocatalytic activity, exhibited a particle size reduction of 86% that of TiO₂-R and a photocatalytic activity nearly5 times and 3 times that of TiO₂-R and original TiO₂, respectively. Introducing (NH₄)₂CO₃ morphology-controlling agent to synthesis {001} facet exposed TiO₂, but did not surpass the photocatalytic activity of TiO₂-R-H7 due to increased particle size, reduced oxygen vacancies, and insufficient {001} facet exposure. Further improvement of the photocatalytic activity of TiO₂ nanocrystals requires in-depth exploration of the structure–activity relationship between acid treatment and crystal facet engineering. This study provides insights into the growth mechanisms of TiO₂ morphologies and highlights acid treatment as a strategy to reduce particle size and enhance photocatalytic activity, offering guidance for designing high-performance photocatalytic materials.

Data availability

The data supporting the findings of this study are available within the article and its ESI.† Additional data are available from the corresponding author and the data have not been submitted to any public repository.

Author contributions

Meng Chai: writing – original draft, writing – review & editing, methodology, investigation, data curation. Bo Yuan: writing – review & editing, conceptualization. Yao Ban: investigation, methodology, data curation. Jing Guo: writing – original draft, writing – review & editing, resources, project administration, funding acquisition, conceptualization. Huifang Lou: investigation. Hongyuan Kang: investigation. Qiaoling Zhang: supervision. Zhiwei Liu: funding acquisition, writing – review & editing. Dongming Zhang: funding acquisition, writing – review & editing.

Conflicts of interest

All the authors have approved the manuscript and agree with submission to your esteemed journal and there are no conflict of interest exits in the submission of this manuscript.

Acknowledgements

This research was supported by the “National Natural Science Foundation of China (21808214)”, “Research Project Supported by Shanxi Scholarship Council of China (2023-126)”, “Fund Program for the Scientific Activities of Selected Returned Overseas Professionals in Shanxi Province (20220013)”, “Fundamental Research Program of Shanxi Province (202203021211085)”, “Fundamental Research Program of Shanxi Province (202403021211196)”.

References

1. Y. Du, X. Niu, J. He, L. Liu, Y. Liu, C. Chen, X. Yang and Q. Feng, ACS Omega, 2020, 5, 14147–14156.
2. N. Parsafard, R. Abedi and H. Moodi, RSC Adv., 2024, 14, 19984–19995.
3. L.-Y. Zhang, Y.-L. Han, M. Liu and S.-L. Deng, RSC Adv., 2023, 13, 16797–16814.
4. R. Katal, S. Masudy-Panah, M. Tanhaei, M. H. D. A. Farahani and H. Jiangyong, Chem. Eng. J., 2020, 384, 123384.
5. L. Liu, Y. Du, X. Niu, W. Li, J. Li, X. Yang and Q. Feng, ChemistrySelect, 2018, 3, 9953–9959.
6. J. Wang, B. Liu and K. Nakata, Chin. J. Catal., 2019, 40, 403–412.
7. C. Yao, X. Wang, W. Zhao, T. Li, Y. He, X. Ran and L. Guo, J. Alloys Compd., 2020, 846, 156335.
8. K. Zhu, H. Gao, G. Hu and Z. Shi, J. Supercrit. Fluids, 2013, 83, 28–34.
9. X. Han, X. Wang, S. Xie, Q. Kuang, J. Ouyang, Z. Xie and L. Zheng, RSC Adv., 2012, 2, 3251.
10. Y. Du, X. Niu, K. Hou, X. He and C. Zhang, Catalysts, 2022, 12, 232.
11. D. K. Chacko, A. A. Madhavan, T. A. Arun, S. Thomas, G. S. Anjusree, T. G. Deepak, A. Balakrishnan, K. R. V. Subramanian, N. Sivakumar, S. V. Nair and A. S. Nair, RSC Adv., 2013, 3, 24858.
12. P. V. Viet, B. T. Phan, L. V. Hieu and C. M. Thi, J. Nanosci. Nanotechnol., 2015, 15, 5202–5206.
13. J. Xiong and L. He, J. Exp. Nanosci., 2017, 12, 384–393.
14. Z. Wei, E. Kowalska, K. Wang, C. Colbeau-Justin and B. Ohtani, Catal. Today, 2017, 280, 29–36.
15. Z. Wei, E. Kowalska, J. Verrett, C. Colbeau-Justin, H. Remita and B. Ohtani, Nanoscale, 2015, 7, 12392–12404.
16. Y. Du, Q. Feng, C. Chen, Y. Tanaka and X. Yang, ACS Appl. Mater. Interfaces, 2014, 6, 16007–16019.
17. C. Chen, L. Xu, G. A. Sewvandi, T. Kusunose, Y. Tanaka, S. Nakanishi and Q. Feng, Cryst. Growth Des., 2014, 14, 5801–5811.
18. J. Li and D. Xu, Chem. Commun., 2010, 46, 2301.
19. A. S. Barnard and L. A. Curtiss, Nano Lett., 2005, 5, 1261–1266.
20. A. S. Barnard, P. Zapol and L. A. Curtiss, Surf. Sci., 2005, 582, 173–188.
21. L. Liu, Y. Du, X. Niu, W. Li, J. Li, X. Yang and Q. Feng, ChemistrySelect, 2018, 3, 9953–9959.
22. C. H. Cho, M. H. Han, D. H. Kim and D. K. Kim, Mater. Chem. Phys., 2005, 92, 104–111.
23. S. Li, L. Ruan, S. Wang, Z. Wang, Z. Ren and G. Han, J. Mater. Sci. Technol., 2020, 46, 139–144.
24. J. Wang, H. Li, H. Wang, K. Huang, G. Sun, S. Yin and T. Sato, Res. Chem. Intermed., 2011, 37, 165–175.
25. B. Liu, J. E. Boercker and E. S. Aydil, Nanotechnology, 2008, 19, 505604.
26. A. Nakahira, T. Kubo and C. Numako, ACS Appl. Mater. Interfaces, 2010, 2, 2611–2616.
27. Y. Yu and D. Xu, Appl. Catal. B Environ., 2007, 73, 166–171.
28. T. Gupta, Samriti, J. Cho and J. Prakash, Mater. Today Chem., 2021, 20, 100428.
29. M. Humayun, F. Raziq, A. Khan and W. Luo, Green Chem. Lett. Rev., 2018, 11, 86–102.
30. J. Xiong and L. He, J. Exp. Nanosci., 2017, 12, 384–393.
31. M. Sluban and P. Umek, J. Phys. Chem. C, 2019, 123, 23747–23757.
32. L. K. Dhandole, M. A. Mahadik, S.-G. Kim, H.-S. Chung, Y.-S. Seo, M. Cho, J. H. Ryu and J. S. Jang, ACS Appl. Mater. Interfaces, 2017, 9, 23602–23613.
33. C. Lv, X. Lan, F. Li, L. Wang, L. Xiao, C. Wang, J. Shi and S. Yu, Catal. Sci. Technol., 2020, 10(3), 690–699.
34. W. Wang, C.-H. Lu, Y.-R. Ni, J.-B. Song, M.-X. Su and Z.-Z. Xu, Catal. Commun., 2012, 22, 19–23.
35. X. Han, X. Wang, S. Xie, Q. Kuang, J. Ouyang, Z. Xie and L. Zheng, RSC Adv., 2012, 2, 3251.
36. N. Roy, Y. Sohn and D. Pradhan, ACS Nano, 2013, 7, 2532–2540.
37. M. M. Maitani, K. Tanaka, D. Mochizuki and Y. Wada, J. Phys. Chem. Lett., 2011, 2, 2655–2659.
38. E. P. Estévez Ruiz, J. L. Lago and S. P. Thirumuruganandham, Materials, 2023, 16, 3076.
39. J. Zhu, S. Zhu, X. Kong, Y. Liang, Z. Li, S. Wu, S. Luo, C. Chang and Z. Cui, ACS Appl. Nano Mater., 2020, 3, 10349–10359.
40. X. Zhang, H. Zhang, H. Zhang, L. Peng and S. Huang, New J. Chem., 2024, 48, 17458–17464.
41. F. Li, G. Liu, F. Liu, J. Wu and S. Yang, J. Hazard. Mater., 2023, 452, 131237.
42. A. K. R. Police, S. V. P. Vattikuti, Y.-J. Baik and B. Chan, Ceram. Int., 2019, 45, 2178–2184.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ra08805k

---