In situ hydrothermal synthesis of visible light active sulfur doped TiO₂ from industrial TiOSO₄ solution

Abstract

A cost-effective industrial TiOSO₄ solution was employed to fabricate visible light active sulfur-doped titanium dioxide (S-TiO₂) via a facile hydrothermal method. The effect of calcination temperature on morphology, particle size, crystallinity, and photocatalytic property of S-TiO₂ was systematically investigated. Successful incorporation of sulfur into TiO₂ was confirmed by carbon-sulfur analysis, X-ray photoelectron spectroscopy (XPS), and Energy dispersive spectrometer (EDS). The research results demonstrated that calcination temperature significantly impacted the crystallinity, specific surface area, sulfur content, and light absorption properties of S-TiO₂. The catalyst calcined at 400 °C revealed the highest photocatalytic activity, with a rate constant of 0.02408 min⁻¹, approximately 25 times higher than commercial P25 catalyst. The higher activity was attributed to the synergistic effect of well-crystallized anatase phase, specific surface area, and red shift of spectral absorption.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-82640-z.

Introduction

In recent decades, semiconductor-based heterogeneous photocatalysis has attracted lot of attention as an emerging alternative technology, exhibiting unique advantages in pollutant degradation^1–4, hydrogen production^5–8, sterilization⁹, water oxidation¹⁰ and other applications. Among the many catalysts, such as SnO₂, ZnO, TiO₂, V₂O₅, WO₃, and PbS, TiO₂ stands out as a superior one due to its strong oxidation ability, good stability, chemical inertness, biocompatibility, non-toxicity, and environmental friendliness^11–13. However, two major drawbacks hinder the practical application of TiO₂: (i) the high recombination rate of electrons and holes, and (ii)wide bandgap energy, which can only be activated by ultraviolet. Various strategies have used to overcome these limitations, including metal doping^14,15, non-metal doping^16,17, physical modification^18–20, and heterostructure construction^21,22.

Numerous studies have demonstrated that sulfur doping is a most effective approach to improve photocatalytic activity due to it can enhance thermal stability and reduce bandgap energy of TiO₂. Ramacharyulu et al. synthesized S-TiO₂and employed them for the degradation of sulfur mustard under sunlight²³. They found that S-doping could efficiently reduced bandgap energy and enhanced visible light absorption of TiO₂, thus exhibited superior activity in photocatalytic degradation of sulfur mustard compared to undoped and commercial TiO₂ catalysts. Gomathi Devi et al. prepared S-TiO₂and investigated its catalytic performance for phenol degradation²⁴. They discovered S⁶⁺ ions doped into the TiO₂ lattice after calcination under atmospheric conditions, and the surface sulfur was modified into SO₄²⁻. Geetha et al.²⁵ prepared sulfated TiO₂ by sol-gel method. The sulfated TiO₂ (5wt% SO₄²⁻) showed excellent catalytic activity in the multicomponent synthesis of highly functionalized piperidines due to the increased Bronsted acidity. Xu et al.²⁶ investigated the activity of pristine TiO₂ and sulfated TiO₂. They found that bidentate chelating SO₄²⁻ and monodentate SO₄²⁻ coexisted in sulfated TiO₂. The presence of SO₄²⁻ enhanced the adsorption of dye molecules and visible light absorption. Li et al. prepared SO₄²⁻-TiO₂ by and found the sample with higher activity and longer lifetime in phenol degradation compared to undoped TiO₂²⁷. However, synthesis routes of catalyst in these studies are often complex, and the employed titanium precursors (e.g., titanium alkoxides or titanium tetrachloride) are relatively expensive. This renders the overall photocatalytic process less economical.

This study utilized a low-cost industrial TiOSO₄ solution as titanium precursor to synthesize S-TiO₂ via a simple hydrothermal method. The effect of heat treatment temperature on morphology, particle size, crystallinity, and catalytic property of S-TiO₂ was systematically investigated. The prepared S-TiO₂ with better light absorption and exhibits good photocatalytic behaviors for degradation Rhodamine B (Rh.B). In addition, the degradation mechanism of S-TiO₂ was discussed.

Experimental

Chemicals and materials

The industrial TiOSO₄ solution used in this work was obtained from titanium dioxide pigment factory, and its primary compositions are: TiO₂ = 189 g/L, F = [effective H₂SO₄/TiO₂] = 1.87. Commercial P25 was purchased from Degussa. Commercial anatase TiO₂ (A-TiO₂) and Rh.B was bought from Fuchen Chemical Reagent Co., Ltd (Tianjin China). Disodium ethylenediamine tetraacetate (EDTA-2Na), isopropyl alcohol (IPA), and benzoquinone (BQ) were obtained from Guanghua Sci-Tech Co., Ltd (Guangdong China). Silver nitrate (AgNO₃) were obtained from Dongjulong Chemical Technology Development Co., Ltd (Tianjin China).

Preparation of S-TiO₂

The S-TiO₂ was synthesized by using hydrothermal route. In a typical prepare process, 92 ml industrial TiOSO₄ solution and 50 ml water were heated to 96 ± 1 °C, respectively. Afterwards, the heated TiOSO₄ solution was added into the heated water in 20 min (using Peristaltic pump) under constant stirring. After adding off, the mixed was transferred into a autoclave (200 ml) and aged at 110 °C for 3 h. After the reaction, the precipitate was centrifuged, washed, and dried at 60 °C overnight. Finally, the solid was calcined in air at 300 °C, 400 °C, 500 °C, 600 °C and 700 °C for 2 h ( heating rate 10 °C/min). The resulting samples were labeled as ST-x, where x represents the calcination temperature (300 °C, 400 °C, 500 °C, 600 °C, or 700 °C).The synthetic process of S-TiO₂ is depicted in Fig. S1.

Characterization

The X-ray powder diffraction(XRD) were obtained on X’ Pert3 Powder diffractometer (Malvern Panalytical), with Cu Kα1 irradiation. Raman spectra were obtained using inVia (Renishaw). The TG-DSC test was obtained in thermal analyzer (STA449C, NETZSCH) under nitrogen atmosphere. Fourier transform infrared (FTIR) tests were carried out using NICOLET 380 (Thermo Electron Corporation). XPS experiments were obtained on Thermo Scientific K-Alpha (PerkinElmer) using Al Kα radiation (hv = 1253.6 eV) with an operating voltage and heater current of 12.0 kV and 6 mA, respectively. Scanning electron microscopy (SEM) was performed on a ZEISS MERLIN Compact (SU8010). Transmission electron microscopy (TEM) tests carried out by FEI Tecnai F20 electron microscope operated at 80 keV. Aomic force microscope (AFM, MFP3D Infinity, America) was employed to characterize the nanoparticle size. UV-visible diffuse reflection spectra (UV-Vis DRS) were conducted on a UV-vis-NIR spectrophotometer (UV-3600, Shimadzu). Nitrogen adsorption apparatus (ASAP 2460, Micromeritics) is used for research specific surface area and pore size distribution. The content of S was determined by Carbon Sulfur Analyzer (CS230, LECO).

Photocatalytic activity

The activity of each ST-x and P25 was evaluated by degradation Rh.B (10 mg/L). A 300 W xenon lamp (λ ≥ 420 nm) was used as light source. For each photocalytic experiment, 100 mg catalyst was added to 100 mL Rh.B. Before irradiation, the mixed solution was avoided light for 30 min. At constant intervals, a specified amount of the sample was taken out and centrifuged. The supernatant was detected on spectrophotometer (Model 721, Shanghai Xinmao Instrument Co., Ltd. China) at λ = 554 nm to investigate the degradation extent of the model compound Rh.B.

Electrochemical performance test

Electrochemical performance were tested by an electro-chemical workstation (CHI760E instruments, Shanghai Chenhua Instrument Co., Ltd. China) in a standard three-electrode with the Pt plate as the counter electrode, Ag/AgCl as the reference electrode, and FTO glass as the working electrode. The sample areas in the electrolyte for 0.2826 cm². The electrolyte 0.2 M Na₂SO₃. A 300 W Xe lamp as the light source (> 420 nm) was used for the photocurrent tests.

Results and discussion

Characterization of composition

Figure 1a shows a comparison of XRD analysis of different samples. Unlike commercial P25 (normally displays both anatase and rutile phases), all S-TiO₂ samples shows only anatase phase of TiO₂ (JCPDS 99 − 0008), even after calcination at 700  °C, which clearly demonstrates the excellent thermal stability of S-TiO₂. The incorporation of sulfur is generally favorable for the formation of anatase TiO₂crystals²⁸. No diffraction peaks of single S elemental is detected, indicating that S probably doped into the TiO₂ crystals. The average crystallite size is listed in Table 1. In brief, the crystallite size of anatase TiO₂grows from 5.6 nm to 24.1 nm with the increase of heat treatment temperature. In general, crystallinity is an important factor for photocatalytic²⁹. The carbon-sulfur analysis results (see Table 1) shows that the sulfur content decreases from 2.13 to 0.17% with the calcination temperature increases from 300 °C to 700 °C. Therefore, a lower heat treatment temperature is beneficial for obtaining a higher sulfur content.

Fig. 1.

Comparison of XRD analysis (a), TG and DSC analysis profiles of un-calcined catalyst (b), Raman (c) and FTIR (d) spectra of the photocatalysts.

Table 1.

Comparison summary of average crystallite size (L) and sulfur content (S) of the photocatalysts.

Photocatalyst	L (nm)	S (wt%)
ST-300	5.6	2.13
ST-400	6.8	2.11
ST-500	10.3	2.05
ST-600	14.8	0.93
ST-700	24.1	0.17

TG and DSC analysis profiles of the un-calcined catalyst are shown in Fig. 1b. The TG analysis reveals two main weight loss stages: (i) from 100  °C to 500  °C and (ii) from 500  °C to 800  °C. The first stage accounts for approximately 27.5% weight loss, primarily the loss of free and bound water. The second stage, with a weight loss of 6.3%, is likely due to the decomposition of strongly bonded sulphate species. The DSC profile exhibits a small endothermic effect around 550  °C, which is associated with the decomposition of sulphates³⁰. Consequently, the sulfate species will gradually decrease with the increase of heat treatment temperature. A higher heat treatment temperature, such as above 700  °C, ensures the almost complete removal of sulfur species. This result is consistent with the carbon-sulfur analysis.

Raman spectra was further used to study the structure of TiO₂ (Fig. 1c). Anatase always exhibits six Raman peaks at about 144 cm⁻¹, 197 cm⁻¹, 399 cm⁻¹, 513 cm⁻¹, 519 cm⁻¹, and 639 cm⁻¹, while rutile typically displays four Raman bands at 143 cm⁻¹, 447 cm⁻¹, 612 cm⁻¹, and 826 cm^−131,32. It can be seen that all samples shows similar peaks (Fig. 1c). According to the literature comparison, all S-TiO₂ samples belong to anatase phase which is consistent with the corresponding XRD patterns. Meanwhile, comparison with P25, the peaks at 144 cm⁻¹ have a slight shift for S-TiO₂, which may be due to the S doped into the TiO₂crystals³³. With increasing calcination temperature, this shift decreasing due to the decrease in sulfur content (see Table 1).

FTIR measurements were performed to study the functional groups and bonding characteristics of the catalysts. Figure 1d shows a comparison of the FTIR spectra of catalysts. The absorption peak at 3424 cm⁻¹ is recognized as the OH groups, and the peak at 1630 cm⁻¹is recognized as the bending vibration of adsorbed water on the surface³⁴. Compared with P25, prepared S-TiO₂ samples has two distinct peaks at about 1141 cm⁻¹ and 1049 cm⁻¹. The former is recognized as the stretching vibrations of S-O bond, and the latter represent Ti-O-S bond, confirm the S is successful doped into TiO₂³⁵. The peak of Ti-O-S gradually weaken and almost disappears in the ST-700 sample. This result indicates that the sulfur content decreases with the increase of heat treatment temperature, which is consistent with the TG analysis and carbon-sulfur analysis. A broad peaks between 700 and 1000 cm⁻¹ is the librational band of adsorbed water³⁶. In addition, the absorption band between 400 and 800 cm⁻¹is belonged to Ti-O stretching vibration³⁷.

XPS was used to investigate the surface chemical composition and element valence states, and the results are revealed in Fig. 2. According to Fig. 2a, the XPS survey spectrum of the ST-400 sample shows clear peaks of Ti, O, and C elements, but only a very small S peak is noticed at 169 eV due to its low content. The high-resolution spectra of Ti 2p, O 1s, and S 2p are presented in Fig. 2b–d. Figure 2b shows Ti 2p3/2 and Ti 2p1/2 at 458.8 eV and 464.5 eV, respectively, indicating a Ti⁴⁺oxidation state³⁸. The high-resolution S peak (Fig. 2c) exhibits two peaks at 168.4 eV and 169.6 eV, which belong to S⁶⁺ such as sulfur in SO₄²⁻. In general, the peaks of elemental S usually appears between 163 eV and 164 eV³⁹. Therefore, no elemental sulfur peak is present in the prepared ST-400 catalyst. The deconvolution of O 1s peak (Fig. 2d) reveals three oxygen peaks, corresponding to 530.0 eV (lattice oxygen, such as in Ti-O bonds), 531.1 eV (oxygen in SO₄²⁻⁴⁰), and 532.0 eV (oxygen in surface hydroxyl groups and absorbed water molecules⁴¹).

Fig. 2.

XPS analysis of ST-400: (a) complete spectra of sample, (b) deconvoluted Ti 2p peak, (c) deconvoluted 2p peak of sulfur, and (d) deconvoluted peak of O 1s.

Characterization of structure and morphology

N₂ adsorption-desorption measurements was used to test surface areas and pore size distributions of catalysts. As depicted from Fig. 3a, all S-doped TiO₂ samples exhibit a distinct hysteresis loop with a typical IV Langmuir adsorption-desorption isotherm shape, indicating that all samples are typical mesoporous materials. With increase of calcination temperature from 300 °C to 700 °C, the hysteresis loop appears at high relative pressures and become cragginess due to the increase of pore diameter. In addition, pore size distribution (Fig. 3b) becomes wider with increasing calcination temperature. The pore structure parameters were shown in Table 2. The specific surface area gradually decreases with increasing calcination temperature because of the growth of TiO₂crystallite. Extraordinary, the specific surface areas was significantly decreased for ST-500, probaly because of the crystal growth lead to collapse of the pore structure. So, a lower calcination temperature facilitates to forming a higher specific surface area, which is beneficial for providing abundant active sites and enhancing the adsorption of reactants⁴². The pore volume generally decreases with increasing calcination temperature. Higher calcination temperatures also lead to an increase in average pore diameter, especially when the calcination temperature is higher than 400 °C. In summary, higher calcination temperatures results in smaller surface area and pore volume, but larger pore diameter.

Fig. 3.

N₂ adsorption and desorption isotherms (a), pore-size distribution (b) of S-TiO₂.

Table 2.

Specific surface area (S_BET), pore volume (V_P), and average pore diameter (D_BJH) of the photocatalysts.

Photocatalyst	SBET (m2/g)	Vp (cm³/g)	DBJH (nm)
ST-300	280.9	0.214409	4.2
ST-400	186.8	0.258151	5.2
ST-500	73.7	0.233120	10.1
ST-600	67.6	0.228768	12.6
ST-700	39.1	0.164780	13.9

The morphologies of the prepared S-TiO₂ were analyzed using SEM. Figure 4a–e reveals the SEM images of prepared samples. All samples exhibit a spherical morphology and high degree of agglomeration. The XRD results indicate that the crystallite size of the samples is 5.6 ~ 24.1 nm, which can lead to agglomerate and reduce the surface energy of the system. Additionally, it can be seen that the particle size become larger with the increase of calcination temperature. Further analysis of ST-400 was carried out by TEM and HR-TEM. Figure 4f shows that the morphology of ST-400 is spherical and the diameter is about 7 nm. The HR-TEM image (Fig. 4g) shows the lattice fringes of the ST-400 sample. The interplanar spacing (~ 0.35 nm) corresponds to the (101) plane of anatase TiO₂⁴³. AFM measurements are performed to further investigate the microscopic size of the ST-400 sample (Fig. 5). According to Fig. 5a, b, the sample particles are evenly distributed on the mica sheet. Figure 5c shows the particles size is about 2 ~ 10 nm, which is similar to the TEM test.

Fig. 4.

SEM images of (a) ST-300, (b) ST-400, (c) ST-500, (d) ST-600, (e) ST-700; (f) TEM and (g) HR-TEM images of ST-400.

Fig. 5.

AFM images of ST-400: (a) 3D morphology, (b) top view morphology, and (c) corresponding profile map of the particles.

The elemental composition of the ST-400 sample was analyzed using EDS mapping (Fig. 6). As exhibited in Fig. 6a–c, O, S and Ti elements are observed and uniformly distribution in the ST-400 sample, which confirmed that S is doped into TiO₂. A detailed elemental line scan (Fig. 6d) reveals a weight% of 2.03% for S, which is similar to the carbon-sulfur analysis.

Fig. 6.

EDS mapping of elements (a) O, (b) S, (c) Ti and (d) EDS spectrum of ST-400.

Optical properties

The optical properties was analyzed using UV-vis spectra. Considering that laboratory-prepared S-TiO₂ samples were anatase TiO₂, we added the test of commercial titanium dioxide (A-TiO₂) for better comparison. As shown in Fig. 7a, all samples exhibit effective UV-light absorption. As depicted in the enlarged image (inset), the absorption spectra of each S-TiO₂ shows increased absorption intensity in compared to P25 and A-TiO₂. This is duo to the incorporation of S elements into the TiO₂crystal structure⁴⁴. High visible light absorption is a premise for high visible light photocatalysis. Additionally, the bandgap energy was estimated by the equation: (Ahν)² = hν − E_g^45,46 (see Fig. 7b). As displayed in Fig. 7b, the bandgap energies of ST-400 and A-TiO₂are 3.25 eV and 3.35 eV, respectively. The ST-400 sample exhibits the lowest optical bandgap, which favors visible light response. This finding is attributed to the probable formation of hybridized states near the valence band due to S doping⁴⁷. Figure 7c is the XPS valence band spectrum of ST-400, and the obtained valence band (VB) values are 2.41 eV. The corresponding band structures are illustrated in Fig. 7d.

Fig. 7.

(a) UV-vis spectra of catalyts and enlarged image (inset). (b) Their corresponding estimated band gaps. (c) XPS valence band spectrum of ST-400. (d) band structures of ST-400.

Photocatalytic properties and mechanism

The photocatalytic degradation of Rh.B of different samples were manifested in Fig. 8. Apparently, only negligible Rh.B degradation could be observed without a catalyst (Fig. 8a). Compared to P25, all S-TiO₂ exhibited better activity for degradation Rh.B. Meanwhile, the activity of S-TiO₂ is significant effected by calcination temperature. Except for ST-400, the photocatalytic activity of S-TiO₂ decreased gradually with the increase of calcination temperature. This is attributed to the synergetic effect of sulfur content, specific surface area, visible light absorption, et al. As seen from Fig. 8b, the linear dynamics curve of ln(C₀/C) vs. time is consistent with the first-order dynamics of Langmuir-Hinshelwood (L-H) model⁴⁸. The calculated rate constants for all catalysts are shown in Fig. 8c. Obviously, the ST-400 sample exhibits the highest rate constant (0.02408 min⁻¹), much higher than the commercial P25 (0.00097 min⁻¹). In general, the excellent activity of the ST-400 sample is attributed to its good crystallinity, larger specific surface area, and good light absorption. The photocatalytic activity in this work were comparable with other photocatalysts and shows relatively good activity (Table S1). The stability of the ST-400 photocatalyst was investigated by cycle experiments, and the results were revealed in Fig. 8d. The sample was recycled from the solution by centrifugation, washed and dried at 80 °C for 10 h between tests during cyclic stability. The photodegradation efficiency of Rh.B was 93.2%, 84.6%, and 77.1% in three cycles, respectively. The decrease in activity may be caused by partial inactivation of the catalyst. The efficiency decreases by only 16.1% after three cycles, which reveals the ST-400 sample has good stability.

Fig. 8.

(a) Comparison of photocatalytic activity for Rh.B (λ ≥ 420 nm, repeated three times). (b) Kinetic curves of different TiO₂ samples. (c) Corresponding degradation rate constants. (d) Cycling stability of ST-400.

The chemical stability of ST-400 was also investigated by XRD and FT-IR. The XRD patterns (Fig. S2a) of the fresh and used ST-400 nearly remain the same, indicating the relatively higher stability of the prepared ST-400. As shown in Fig. S2b, the fresh ST-400 shows similar FT-IR spectra as those of the used ST-400, also indicating the higher stability of the prepared ST-400⁴⁹.

Electrochemical performance of ST-400 and P25 were measured, and the results is revealed in Fig. S3. The transient photocurrent response curves are shown in Fig. S3a. It can be see that the ST-400 produced the higher photocurrent than P25, suggesting that the sample of ST-400 could efficiently transfer of photoexited charge carriers⁵⁰. Fig. S3b illustrates the electrochemical impedance spectroscopy profiles (EIS) of ST-400 and P25. Compared with P25, ST-400 exhibited smaller radius of EIS, suggesting that the S-doping beneficial to the internal carrier transfer resistance and further speed up electron transport.

For identify the main reactive species of photocatalytic reaction, radical trapping experiments were conducted on the ST-400 photocatalyst using different scavengers (Fig. 9a). EDTA-2Na (1mM), IPA (1mM), BQ (1mM) and AgNO₃ (1mM) were employed as scavengers for holes (h⁺), hydroxyl radical (·OH), superoxide radical (·O₂⁻), and electrons (e⁻), respectively⁵¹. The degradation efficiency has almost no change after the incorporation of AgNO_3, indicating that e⁻ is not the reactive species. When BQ is added, the degradation efficiency decreases by only 20.1% (from 93.2 to 73.1%), suggesting that ·O₂⁻ is not the main reactive species. However, the degradation efficiency decreased significantly after the addition of EDTA-2Na and IPA, showing that h⁺ and ·OH are the main reactive species in the photocatalytic reaction.

Fig. 9.

Photocatalytic degradation of Rh.B over ST-400 without and with addition of different scavengers (a) and photocatalytic mechanism of ST-400 (b).

The possible photocatalytic mechanism of S-TiO₂ is displayed in Fig. 9b. With incorporation of S into the TiO₂lattice, new S 2p impurity energy level can be formed above the VB. This results in a decrease of bandgap and increased visible light response^52,53. Thus, S-TiO₂ can be easy stimulated by visible light to generate the electron/hole pairs. Electrons on the conduction band of S-TiO₂ are donated to oxygen molecules to generating ·O₂⁻⁵⁴ and the holes generated in the VB are captured by OH⁻to form ·OH⁵⁵. Meanwhile, the h⁺ on S-TiO₂ has strong oxidizing ability and can directly degrade Rh.B. Therefore, Rh.B can be degraded by strong oxidants such as h⁺,·OH, and ·O₂⁻, which were confirmed radical trapping experiments.

Conclusion

In this study, S-TiO₂ was synthesized via a simple and straightforward hydrothermal route using low cost industrial TiOSO₄ solution, and the influence of calcination temperature was investigated. The presence of the S element was strongly confirmed by instrument analysis. The calcination temperature had a significant impact on crystallinity, specific surface area, S content, and light absorption. XRD and Raman spectra were confirmed all prepared S-TiO₂ exhibited only the anatase phase of TiO₂ even after calcination at 700 °C, clearly certifying the excellent thermal stability of S-TiO₂. The sample calcined at 400 °C revealed the highest activity, which is attributed to its good crystallinity, large specific surface area, and good light absorption. This study provides a simple method for synthesize of visible-light active S-TiO₂ from low-cost raw material.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

HongPu wrote the main manuscript text, Congxue Tian guide the experimental analysis and Hui Zhang conduct guidance and review of article writing norms. All authors reviewed the manuscript.

Data availability

If someone wants to request the data from this study, please feel free to contact Hong Pu.

Declarations

Competing interests

The authors declare no competing interests.