Cost-effective Bi₂WO₆ for efficient degradation of rhodamine B and tetracycline

Abstract

The morphology-controlled synthesis of nanostructured photocatalysts by an environmentally friendly and low-cost method provides a feasible way to realize practical applications of photocatalysts. Herein, Bi₂WO₆ (BWO) nanophotocatalysts with mulberry shape, sheet-like, and round-cake morphologies have been successfully synthesized through a highly facile solvothermal process by simply adjusting the solvothermal temperature or utilizing selective addition of ethylene glycol as an orientation agent without using strong acids and bases and/or hazardous chemicals. The ratio of ethylene glycol and glacial acetic acid can affect the morphology and oxygen vacancy content of BWO, thereby influencing the photocatalytic performance. The photocatalytic activity of the as-prepared samples was evaluated by degradation of rhodamine B (RhB) and tetracycline under visible-light irradiation. The results indicated that all the BWO samples exhibited morphology-associated photocatalytic activity, and the sheet-like structure of BWO obtained via solvothermal treatment at 120 °C with ethylene glycol and glacial acetic acid ratio of 1:3 achieved the maximum specific surface area and possessed abundant oxygen vacancies, exhibiting outstanding photocatalytic activity for degradation of RhB and tetracycline. The degradation rate of RhB reached 100% within 20 min. To the best of our knowledge, this value is one of the most remarkable values for pristine BWO photocatalysts. Radical capture experiments demonstrated that hydroxyl radicals (·OH) play major roles compared with electrons (e⁻) and holes (h⁺) in the photocatalytic degradation process. A possible mechanism for the photocatalytic degradation of pollutants was proposed to better understand the reaction process. We believe that the more economical, efficient and greener methodology can provide guidance to develop highly efficient photocatalysts with favourable morphology and structure.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10854-022-09654-z.

Introduction

While social development brings convenience to people’s lives, it also inevitably causes damage to the ecological environment. A large number of pollutants, such as industrial wastewater, waste gas and waste residue, are discharged into the environment. Most of these pollutants do not easily naturally degrade and thus widely exist in water, the atmosphere and soil, causing serious pollution and damage to the ecological environment. For example, organic dyes commonly used in textile and leather factories are not only difficult to degrade but also pose a risk of harming aquatic ecosystems and inducing cancer [1, 2]. In medical treatment, antibiotics are widely used to treat various diseases caused by bacterial infection, but once refractory antibiotics enter the ecological circulation system, they not only cause water pollution but also are conducive to the reproduction of drug-resistant bacteria, thus affecting the ecosystem and human health and safety [3, 4]. The outbreak of COVID-19 worldwide has attracted increasing attention to environmental protection and physical health. Therefore, methods for treating industrial and medical pollutants have become one of the hotspots of current research.

Semiconductor photocatalysis technology provides a new method for the green, safe and effective degradation of pollutants. Compared with the traditional physical, chemical and biological methods for purifying environmental pollutants, this technology has the advantages of energy savings, good chemical stability, non-selectivity, environmental friendliness and complete degradation [5–7]. However, traditional photocatalysts, such as TiO₂ [8], ZnO [9] and CdSe [10], have problems in visible-light utilization and photocatalytic efficiency, which restrict their large-scale applications. To solve these problems, researchers have paid much attention to modifying photocatalysts, including optimizing the structural morphology, depositing precious metals, doping ions, and forming multicomponent material composites [11–13]. However, these modifications increase the complexity of the chemical composition of the photocatalyst, the difficulty of preparation, and the production cost. Therefore, studying how to prepare photocatalyst materials with a simple process flow, a stable structure and high performance is an important direction of photocatalytic research.

As a typical aurivillius oxide, Bi₂WO₆ (BWO) has the ability to respond to visible light because it possesses an appropriate energy band potential. At the same time, the built-in electric field between the layered structures promotes the separation of photogenerated carriers. BWO is considered a promising photocatalyst [14]. Nevertheless, the visible-light catalytic ability of BWO still has much room for improvement for practical applications. An efficient photocatalyst should have not only a small bandgap to absorb more photons within the visible region but also excellent crystal properties and controllable particle size to facilitate effective separation and migration of photogenerated carriers in the photocatalyst [15]. The solvothermal method is one of the most common and effective synthetic routes to fabricate nanomaterials with a variety of morphologies. This method can facilitate and accelerate reactions among the reactants and promote hydrolysis, followed by crystal growth, resulting in self-assembly of nanomaterials in the solution. Moreover, the properties, morphology, size, and structure of nanomaterials can be easily tailored by varying the different reaction parameters, such as the reaction time, temperature, reaction medium, pressure, pH, concentration of the reactants and filled volume of the autoclave. This method can be suitable for the preparation of nanomaterials with a variety of shapes compared to other methodologies. Mi [16] and coworkers prepared tubular, flower-like and cake-like BWO nanostructures through self-assembly. The performance of these structures in the photocatalytic decomposition of pollutants depended on the appearance, size and structure. The rate of RhB degradation by flower-like BWO after 20 min of visible-light irradiation reached 80%. Zhang [17] and coworkers used cetyltrimethylammonium bromide (CTAB) as a template, and BWO particles with layered structures were synthesized using an assisted solvothermal method to produce a larger specific surface area than spherical structures and enhance the photocatalytic activity.

Nevertheless, in the process of preparing BWO catalysts, researchers often employ strong acids such as nitric acid and strong bases such as sodium hydroxide to adjust the pH of the solution and add some organic substances to improve the properties of the solution [18–20]. The addition of strong acids and bases may cause environmental pollution in the process of preparing materials. In addition, during solvothermal or coprecipitation reactions, most researchers use 160–200 °C as the reaction temperature and then calcine the obtained powder. Higher reaction temperatures and subsequent heat treatment processes require more complex production equipment and consume a large amount of energy, which not only increases the production expenses but also does not meet the concept of green development. However, there have been very few reports on the synthetic preparation of BWO using low-temperature, low-cost, environmentally friendly methods, which is important for researching control of the BWO morphology in the future.

Herein, this is the first study to prepare BWO with various morphologies only by adjusting the solvothermal temperature or utilizing selective addition of ethylene glycol (EG) as an orientation agent without using strong acids and bases and/or hazardous chemicals. This is the first report that the EG and glacial acetic acid ratio can affect the morphology and oxygen vacancy content of BWO, thereby influencing the photocatalytic performance. In addition, photocatalytic degradation of RhB based on BWO nanocrystals was carried out under visible-light irradiation. The results indicated that the round cake-shaped structure of BWO with a large specific surface area exhibited excellent photocatalytic activity.

Experimental procedures

Preparation of BWO using the solvothermal method

The chemicals and reagents used included Bi(NO₃)₃, NaWO₄, EG and glacial acetic acid, which were purchased from Aladdin with analytical purity and did not need further treatment. Solution A was formed by dissolving 1 mmol Na₂WO₄ in 10 mL of pure water, and solution B was formed by dissolving 2 mmol Bi(NO₃)₃ in a certain amount of EG and glacial acetic acid. After stirring with a magnetic mixer to dissolve the chemicals and generate a transparent solution, solution A was slowly added dropwise to solution B and vigorously stirred for 20 min. The suspension was poured into a 50 mL polytetrafluoroethylene lining, placed in a stainless steel reactor, and then placed in an oven at a preprepared temperature for insulation for several hours. After heat maintenance and natural cooling, the precipitate was obtained by filtration, washed several times with ethanol and pure water, and then transferred to a 60 °C oven for drying for 12 h to obtain BWO powder. Using the process for preparing a BWO powder at 120 °C as an example, the flowsheet of the process is shown in Fig. 1.

Fig. 1.

Schematic diagram of the process used to prepare the 120 °C BWO.

Characterization of materials

The crystal structures of the samples were determined using a Drudy D8 Advance type X-ray diffractometer (XRD) from Germany. The morphological and microstructural characteristics were determined with a Hitachi SU8010 scanning electron microscope (SEM). N₂ adsorption-desorption isotherms were measured using a JW-BK112 specific surface and pore size analyser, and the test temperature was 77 K. UV–Vis absorption spectra of the samples were recorded using a Hitachi U-4100 spectrophotometer at wavelengths ranging from 200 to 700 nm. Photoluminescence (PL) spectra were obtained on a Horiba Flurolog-3 fluorescence spectrometer at an excitation wavelength of 300 nm.

Photocatalytic test

Standard photocatalytic tests were performed in a glass reaction vessel. A 300 W Xe lamp (CEL-HXF300-T3) was used as a solar light source for the simulation, and an AM1.5G filter was used to isolate ultraviolet light. One hundred milligrams of BWO samples obtained under different preparation conditions was poured into 100 mL of a rhodamine B (RhB) solution (10 mg/L) or a tetracycline (TC) solution (20 mg/L) and magnetically stirred in the dark for 20 min for even dispersion and to attain adsorption equilibrium.

The degradation system was then exposed to visible light at room temperature, and the suspension was periodically extracted with a 5 mL syringe and filtered with a 0.22 μm needle filter. The dye concentration remaining after the reaction was measured using a UV–Vis spectrophotometer. The stability of the photocatalyst was tested as described below. After the photodegradation reaction, the used photocatalyst was collected again by centrifugal methods, washed with ethanol and pure water, and dried, and the test was repeated under the conditions described above.

To detect the reactive groups generated during the photocatalytic process, reactive species capture experiments were performed by adding free radical quenchers to the degradation system. AgNO₃, KI, isopropyl alcohol (IPA) and p-benzoquinone (BQ) were used as scavengers of photogenerated electrons (e⁻), holes (h⁺), hydroxyl radicals (·OH) and superoxide radicals (·O₂⁻), respectively. The concentration of the above quenchers added to the degradation system was 0.5 mmol/L. To avoid the interference of the dark adsorption of the catalyst on the degradation rate values, we calculated the degradation rate using the concentration after 20 min of stirring in the dark as the initial concentration (C₀). The other conditions were consistent with those in the photocatalytic degradation reaction process.

Results and discussion

Sample characteristics

BWO samples were prepared at 100 °C, 120 °C, 140 °C, 160 °C and 180 °C by controlling the solvothermal reaction temperature. The XRD patterns of BWO prepared at different solvothermal temperatures (Fig. 2a) showed poor crystallinity of the BWO powder obtained by reaction at 100 °C and no obvious diffraction peaks. A possible explanation for the amorphous structure observed here is that the reaction temperature was too low to provide sufficient energy for crystal growth. Obvious characteristic diffraction peaks were observed at 2θ = 28°, 32°, 47° and 55° starting at the temperature of 120 °C, matching the B2ab space group structure of BWO (JCPDS card 73-1126), which indicated the successful preparation of BWO powder and the absence of an impurity phase. The diffraction peak intensities of BWO powder gradually increased and the half peak widths gradually narrowed with increasing reaction temperature, indicating improved crystallinity of the BWO powder, and the crystallization degree of BWO prepared at 180 °C was the highest. SEM images and the particle size distributions of BWO powder prepared at different solvothermal temperatures are shown in Fig. 2b–f. The BWO powder prepared at a reaction temperature of 100 °C was mainly formed of fine particles with a diameter of approximately 26 nm, without an obvious nano-stacking structure. With increasing reaction temperature, the particle size of the BWO powder gradually increased, and an agglomeration structure with an obvious morphology appeared. The average particle sizes of BWO prepared at 120–180 °C were 26 nm, 27.85 nm, 69.24 nm, 102.97 and 128.83 nm (HRSEM images are shown in Fig. S1). The BWO powder began to change from a particle to flake appearance at 120 °C and 140 °C and then existed in the form of round cake and mulberry nanostructures (Fig. 2c and d). The accumulation of fine particles provides a large exposed surface area and conditions for effective adsorption of pollutants. Some researchers postulate that pie-like nanostructures are usually unstable and often develop into flower-like structures to reduce the surface energy [21]. As shown in Fig. 2e, the round cake-shaped BWO powder was transformed into a nanoflake structure with a side length of 100–200 nm, and the nanoflakes had a smooth surface and self-assembled to form a flower shape, consistent with the research report mentioned above. The prepared BWO grains grew further, the flower shape was destroyed, and the grains developed into irregular columnar structures (Fig. 2f) due to the high temperature and high pressure environment at 180 °C.

Fig. 2.

a XRD patterns of BWO prepared via solvothermal treatment under different temperature. SEM images and particle size distributions of BWO prepared under solvothermal treatment at b 100 °C, c 120 °C, d 140 °C, e 160 °C, and f 180 °C

A decrease in the Bi³⁺ concentration in the solution slows the reaction speed because Bi³⁺ easily reacts with ·OH; therefore, different BWO powder morphologies can be obtained by controlling the ·OH content in the solution. The EG content in the solvent exerted a substantial effect on the morphology regulation of BWO powder. One molecule of EG (molecular formula: CH₂OHCH₂OH) can provide two molecules of OH to complex with Bi³⁺. The surface tension of EG is 46.49 mN/m (20 °C), greater than that of glacial acetic acid (29.58 mN/m), which increases the cohesion and adsorption of liquid and improves the solution viscosity [22]. Therefore, EG was used as a solvent and chelating agent to improve the agglomerated form of BWO in this experiment.

The samples denoted 0 EG BWO, 0.3 EG BWO, 1 EG BWO, and 3 EG BWO were prepared at 120 °C with EG-to-glacial acetic acid solvent ratios of 0, 1:3, 1:1, and 3:1 by volume, respectively. The measured XRD patterns and SEM images are shown in Fig. 3. The XRD patterns (HRSEM images are shown in Fig. S2) show that the crystallinities of the samples prepared by adding different EG contents were approximately the same, and no other impurity phase was produced, indicating that the addition of EG generally does not alter the crystal structure of BWO. However, the surface morphologies showed a substantial difference under the electron microscope: the BWO powder prepared without EG addition did not interact to form aggregated BWO particles due to the low viscosity of the glacial acetic acid solvent and was only stacked in granular form (Fig. 3b). After the addition of EG, ·OH in the molecule reacted with metal ions to form an alkoxide, resulting in a gradual decrease in the concentration of free Bi³⁺ in the solution, slowing the dynamic process of crystal growth, and gradually increasing the viscosity of the solution to enable gradual aggregation of particles. The 0.3 EG BWO sample formed a round cake shape with a rough surface (Fig. 3c), and 1 EG BWO presented a round cake morphology with a smooth surface and a diameter of approximately 1 μm (Fig. 3d). When the proportion of EG was relatively large, the reaction process became slower, and the lamellar BWO was not easy to separate. The 3 EG BWO sample was stacked and self-assembled into a larger double concave cake shape (Fig. 3e) with a diameter of approximately 2 μm and a thickness of approximately 800 nm.

Fig. 3.

a XRD patterns of BWO prepared via solvothermal treatment with different EG content. SEM images of BWO obtained via solvothermal treatment with EG content at b 0 EG, c 0.3 EG, d 1 EG, and e 3EG.

The temperature was maintained at 120 °C for 8 h, 12 h, 16 h, 20 and 24 h with an EG/glacial acetic acid ratio of 1:3 to obtain BWO grains at different growth stages to obtain a better understanding of the changes in the structure, morphology and agglomeration form of BWO nanosheets during the growth process, and the samples were denoted 8 h, 12 h, 16 h, 20 and 24 h BWO, respectively. Figure 4a shows that the XRD diffraction peak positions and intensities of the BWO samples obtained at different reaction times showed little difference at the same reaction temperature, as the powder particles were stacked in a circular cake shape and the average particle size was maintained at approximately 20 nm (Fig. 3b–f, HRSEM images are shown in Fig. S3). These results showed that stable BWO samples were obtained within 8–12 h at 120 °C, and extension of the holding time had no significant effect on the structure, morphology or agglomeration pattern of BWO.

Fig. 4.

a XRD patterns of BWO obtained after solvothermal treatment at 120 °C for different time. SEM images and particle size distributions of BWO prepared after solvothermal treatment at 120 °C for b 8 h, c 12 h, d 16 h, e 20 h, and f 24 h

The BWO powder obtained from the solvothermal reaction for 12 h was calcined in a muffle furnace to further study the effect of the calcination temperature on BWO. The corresponding BWO samples were obtained at calcination temperatures of 300 °C, 400 °C and 500 °C, which were denoted BWO 300 °C, 400 °C and 500 °C, respectively. As shown by the XRD patterns in Fig. 5a, the positions of the diffraction peaks of BWO after calcination still corresponded well to the BWO standard card with the B2ab space group structure, indicating that the calcination treatment did not change the crystal structure of BWO. Compared with BWO prepared without annealing, the diffraction peak intensities of the three samples increased, and the half peak widths decreased, indicating that the crystallinity was significantly improved. SEM images show that the average particle size of the calcined BWO powder also gradually increased with increasing temperature from 27.85 nm without calcination treatment to 50.41 nm for BWO 500 °C (HRSEM images are shown in Fig. S4). The round cake agglomeration morphology of BWO was destroyed, and the material existed in flocculent and dendritic forms (Fig. 5c, d); BWO 500 °C formed columnar crystals (Fig. 5e), potentially due to the collapse of the nanostructure of BWO powder in the heat treatment environment resulting in a change in morphology.

Fig. 5.

a XRD patterns of BWO obtained after solvothermal treatment at 120 °C followed by different annealing temperature. SEM images and particle size distributions of BWO prepared after solvothermal treatment at 120 °C (b), followed by annealing at c 300 °C, d 400 °C, and e 500 °C

In order to better understand the photocatalytic performance of BWO prepared under different conditions, We selected a typical set of BWOs for the subsequent experimental study. The 120 °C BWO, 180 °C BWO, 3 EG BWO, and BWO 500 °C were obtained via solvothermal treatment under 120 and 180 °C with EG content at 0.3, solvothermal treatment under 120 °C with EG content at 3, solvothermal treatment under 120 °C with EG content at 0.3 followed by annealing at 500 °C, respectively. The surface elements and chemical states of the samples were tested by X-ray photoelectron spectroscopy (XPS). Figure 6a shows the complete XPS spectra of the obtained samples. The appearance of the C1s peak was due to the storage of the samples and the CO₂ adsorbed in the test environment, and no impurity peak was detected. Figure 6b shows the XPS spectra of the Bi 4f orbital of the samples. Two different sets of peaks were detected near the binding energies (BEs) of 164.6 eV and 159.3 eV for the samples, which were attributed to Bi 4f_5/2 and Bi 4f_7/2 [23]. The orbital dipole splitting energy was BE(Bi 4f_5/2)-BE(Bi 4f_7/2) = 5.3 eV, which corresponded to the + 3 valence oxidation state of the Bi ion and demonstrated that Bi³⁺ was the dominant Bi form in the BWO samples [24]. Figure 6c shows two peaks near 37.7 eV and 35.6 eV, corresponding to 4 f_5/2 and 4 f_7/2 of W, characteristic of W⁶⁺ in the WO₄ octahedron [25]. The spectrum of the O 1s region (Fig. 6d) could be fitted by a superposition of two peaks near 530.2 eV and 531.1 eV, which corresponded to the lattice oxygen Bi–O and W–O of the [Bi₂O₂]²⁺ and [WO₄]²⁻ layers in Bi₂WO₆, respectively [26]. The fitted area ratio of the Bi–O bonds to the W–O bonds was calculated, and the minimum value for the 3 EG BWO sample was 1.88, while the values for the 120 °C BWO, 180 °C BWO and BWO 500 °C samples were approximately 2.07, 2.05 and 2.01, respectively, which were consistent with the theoretical ratio of Bi₂WO₆. This indicated the formation of oxygen vacancies due to the excess EG providing more reducing ·OH, which reduced the free Bi³⁺ in the solution and inhibited binding to the W–O bonds [27].

Fig. 6.

a XPS full spectra of 120 °C, 180 °C, 3 EG and 500 °C BWO samples; b Bi 4f orbital spectrum; c W 4f orbital spectrum; d O 1s orbital spectrum

A larger specific surface area of photocatalytic materials may expose more reaction active centres, potentially increasing the collision probability between pollutants and the photocatalyst surface, improving the adsorption performance of photocatalytic materials, allowing more photogenerated electrons and/or holes to be captured in a timely manner by pollutants, avoiding recombination of photogenerated electrons and holes, and improving the photocatalytic reaction rate and catalytic activity [28]. The structural attributes of the BWO samples were analysed by recording N₂ adsorption-desorption isotherms at 77 K (Fig. 7). According to the IUPAC classification, the four groups of representative samples showed type II adsorption isotherms. Their absorption gradually increased with increasing relative pressure, and the adsorption and desorption curves coincided without hysteresis, indicating that multilayer adsorption occurred with increasing relative pressure [29]. The pore size distribution showed that the samples mainly formed a mesoporous structure, which was considered to be the result of accumulation of fine BWO particles. The specific surface area and pore data of the four groups of samples computed using the Brunauer–Emmett–Teller (BET) method are recorded in Table 1. Compared with the samples prepared under other reaction conditions, 120 °C BWO exhibited a larger specific surface area (53.321 m²/g) and a larger pore structure. On the one hand, the specific surface area is conducive to increasing the adsorption of pollutants and providing more reaction active sites [30]. On the other hand, the accumulated mesoporous structure is also conducive to the propagation of photons into the interior of the catalyst and stimulates the participation of more grains in the photocatalytic reaction, contributing to excellent photocatalytic performance.

Fig. 7.

N₂ absorption–desorption isotherms and pore size distribution plot (inset) of the 120 °C, 180 °C, 3 EG and BWO 500 °C samples

Table 1.

BET specific surface area, average pore diameter and pore volume of the 120 °C, 180 °C, 3 EG and BWO 500 °C samples

Sample	SBET (m2/g)	Average pore (nm)	Pore volume (m3/g)
120 °C BWO	53.32	6.14	0.08
180 °C BWO	14.88	20.72	0.08
3 EG BWO	39.79	4.64	0.05
BWO 500 °C	12.43	9.59	0.03

Photocatalytic activity

Photocatalytic degradation efficiency

In the experiment, RhB, a widely applied organic dye, was used as the target degradation product. The performance of BWO powders obtained under different conditions in the degradation of RhB under visible-light irradiation was evaluated to determine the effect of the preparation process on the photocatalytic performance of BWO. The RhB solution was prepared in pure water as the solvent, and the main light absorption peak was located at 554 nm. First, the self-photolysis of RhB under 300-W xenon lamp illumination without adding any catalyst was studied. As shown in Fig. S5, the RhB solution concentration did not change under sunlight illumination for 40 min (the black curve). This result demonstrated that the molecular structure of RhB is stable under sunlight illumination. A dark adsorption experiment of the catalysts was also carried out to prove the limited adsorption capacity of the BWO samples. Figure S5 shows that the RhB solution concentration tended to be stable after adsorption for 20 min without sunlight illumination, indicating that all the photocatalysts reached adsorption saturation. The results indicated that all the photocatalysts exhibited limited adsorption of RhB and that the catalysts would reach adsorption equilibrium within 20 min under dark stirring. Figure 8a shows that the BWO sample obtained by controlling the solvothermal reaction temperature at a low temperature of 100 °C produced a slow degradation rate of RhB due to the insufficient formation of crystals, whereas the well-crystallized BWO samples showed excellent photocatalytic ability: 120 °C BWO showed a good degradation rate, and the degradation rate of RhB reached 100% within 20 min. However, the photocatalytic degradation rate improvement with increasing reaction temperature was restricted, indicating that the continuous improvement in the crystallinity did not improve the degradation ability of BWO. Therefore, the photocatalytic efficiency of the BWO powder was directly affected by the specific surface area and stacking morphology rather than the crystallinity. The regulatory effect of EG on the BWO morphology also altered the photocatalytic efficiency. The round cake BWO powder tended to adsorb pollutants to decrease the surface energy, which was conducive to the capture and degradation of pollutants. Therefore, as shown in Fig. 8b, the round cake BWO prepared with EG-to-glacial acetic acid ratios of 1:3 and 1:1 had the best degradation ability and completely degraded RhB in approximately 20 min. The reaction processes of both scattered 0 EG BWO and regular and larger 3 EG BWO were slow. According to the SEM images shown in Fig. 3b, the 0 EG BWO powder particles were fine, and the tight accumulation of nanoparticles reduced the specific surface area and pores such that the internal BWO particles were relatively closed and the number of catalytic active sites was reduced. At the same time, more interface defects became recombination centres of photogenerated electrons and holes, which would also reduce the photocatalytic efficiency. Figure 3e and Table 1 show that the photocatalytic performance of the 3 EG BWO powder that formed a large double concave filter cake structure was reduced due to the reductions in the specific surface area, pore diameter and pore volume. Because BWO crystals with complete crystal growth and a stable structure were obtained after holding for 8–12 h, the crystallinities and morphological characteristics of BWO samples obtained with different solvothermal reaction times were approximately the same, and thus, the solvothermal reaction time had no significant effect on the photocatalytic activity of BWO (Fig. 8c). Calcination treatment inhibited the photocatalytic activity of BWO, and the RhB degradation rate was only 45% within 20 min in the presence of BWO 500 °C (Fig. 8d). The change law of the degradation curve was consistent with the solvothermal reaction temperature, which further proved that the improvement in the crystallinity is not a direct factor that improves the photocatalytic degradation efficiency. As shown in the data presented in Fig. 5c–e and Table 1, the degradation efficiency of the BWO powder gradually slowed, mainly because the morphology of the BWO powder was destroyed and the nanostructure collapsed after calcination, resulting in a reduction in the exposed area.

Fig. 8.

Photocatalytic degradation curves of RhB solution (10 mg/L) by BWO prepared under different conditions

The photodegradation of a drug follows the pseudo-first-order kinetics model:

$$-ln\left(\frac{C}{C_0}\right)=Kt$$ 1

where K and t are the first-order rate constant (min⁻¹) and the reaction time (min), respectively. Figure 9 shows the calculated reaction rate constants of the BWO samples degrading RhB, which intuitively shows the photocatalytic activity of BWO obtained under different conditions. The K of 120 °C BWO was 0.31471 min⁻¹, which is the highest value recorded for all samples and 1.1–14 times higher than that of the other samples. Combined with the material characterization results, we speculated that the difference in the photocatalytic performance of BWO samples prepared using the four groups of processes is closely related to their morphology and size: the round cake agglomerated form and a large exposed area may be more conducive to the photocatalytic reaction.

Fig. 9.

Reaction rate constants in the degradation of an RhB solution (10 mg/L) by BWO prepared under different conditions

In contrast to the preparation of BWO photocatalysts by other researchers (Table 2), we did not introduce environmental substances such as strong acids and bases to adjust the pH or other chelating agents to improve the solution properties in the preparation of BWO. At the same time, in the solvothermal reaction process, BWO with stable structure and performance can be obtained by simply applying a lower temperature of 120 °C for 12 h without additional calcination or other subsequent processes. The process of preparing BWO is simple and has the advantages of environmental friendliness and low cost. The obtained BWO is the most effective BWO photocatalyst known to degrade RhB pollutants. The photocatalytic effect is obvious and can meet the requirements of practical applications in the future.

Table 2.

Comparison of the 120 °C BWO photocatalytic performance with that of other previously reported Bi₂WO₆ photocatalysts for the degradation of pollutants

Photocatalyst	Target pollutant	Concentration	Removal rate	Ref.
CoO/Bi2WO6	TC	40 mL/L	90.7% (90 min)	Lu et al. [31]
Bi2O2CO3/Bi2WO6	RhB	10 mL/L	99.6% (60 min)	Qiang et al. [32]
	TC	20 mg/L	80% (60 min)
BiOBr/Bi/Bi2WO6	RhB	10 mL/L	98.02% (60 min)	Chen et al. [33]
LaCoO3/Bi2WO6	TC	20 mL/L	85.5% (150 min)	Guo et al. [34]
MOF-5/Bi2WO6	RhB	10 mL/L	99.31% (30 min)	Yao et al. [35]
Carbon modified Bi2WO6	RhB	10 mL/L	98.8% (100 min)	Lei et al. [36]
Bi–Bi2WO6	RhB	10 mg/L	93.7% (24 min)	Jin et al. [37]
	TC	40 mg/L	60.98% (28 min)
Bi2WO6/g-C3N4/BPQDs	RhB	20 mg/L	95.6% (120 min)	Du et al. [38]

Possible mechanisms of RhB degradation

Academic researchers generally propose two mechanisms for the degradation of RhB. One is that the strong oxidation and reduction of free radicals mediated by the photocatalytic process destroys the conjugated structure. In this process, RhB is decomposed into small organic molecules and CO₂, which decreases the absorbance, but the maximum absorption wavelength remains unchanged at 554 nm. The second mechanism involves a photosensitization reaction and gradual removal of ethyl groups. In this process, individual ethyl groups are successively removed from tetraethylated RhB to form tri-, di-, mono-, and non-ethylated molecules, and the corresponding absorption peak positions are 539, 522, 510 and 498 nm, respectively. Therefore, the absorption peak appears to undergo a blueshift during the degradation process [39], and the solution will gradually change from red to yellow and finally colourless.

Figure 10 exhibits the evolution of the optical absorption spectrum in RhB degradation by the 120 °C BWO photocatalyst to further clarify the change in RhB during degradation. The amplitude of the main optical absorption peak of RhB at 554 nm gradually decreased during the visible-light degradation process and almost completely disappeared in approximately 20 min, indicating that the conjugate structure of the RhB molecule was continuously destroyed and decomposed into small-molecule organic substances and CO₂. With the extension of the illumination time, the optical absorption peak of RhB exhibited a progressive blueshift to near 498 nm and continued to decrease, indicating that RhB underwent an ethyl removal process and finally became non-ethylated. We concluded that damage to the conjugated structure of RhB and separation of ethyl groups simultaneously occurred in the BWO photocatalytic system, and RhB was completely degraded and transformed into non-ethylated organic small-molecule substances and CO₂. The removal of RhB is a chemical reaction rather than a simple adsorption procedure, which may achieve harmless treatment of pollutants.

Fig. 10.

Temporal evolution of the absorption spectrum of an RhB solution (10 mg/L) degraded by 120 °C BWO under visible light

Evaluation of universality and stability

General applicability and stability of photocatalysts are the basic requirements for practical applications. Meanwhile, the massive use and disorderly discharge of antibiotics have led to continuous detection of antibiotic residues in water environmental samples, which constitute a great threat to the environment and human health because their structures are relatively stable and difficult to degrade. We subsequently performed a photocatalytic degradation test with the common antibiotic TC as the target pollutant to test whether the prepared BWO photocatalyst has wide application and stable cyclic degradation abilities.

The results of the photocatalytic test using the TC solution (20 mg/L) were in complete agreement with those of the test using the RhB solution (10 mg/L). The 120 °C BWO sample that performed well in the RhB degradation experiment was selected as the photocatalyst. The degradation curve in Fig. 11a shows that the BWO sample also maintained an excellent ability to degrade TC with a more stable structure, and the degradation rate of pollutants reached 75% within 40 min of visible-light illumination. The BWO sample maintained a stable degradation effect, and the photocatalytic activity did not significantly decrease during four consecutive degradation cycles. Comparing the XRD patterns of unused and recycled BWO (Fig. 11b), the crystalline phase of the 120 °C BWO sample did not change after the photocatalytic degradation reaction. The SEM images (Fig. 11c, d) also showed that the particle surface of the BWO sample was not covered by other substances, and the agglomeration form was still a thin round cake after the reaction. Based on these results, the 120 °C BWO photocatalyst was not corroded by light when reused and still maintained activity and stability.

Fig. 11.

a Cyclic degradation curve of 120 °C BWO in a TC solution (20 mg/L); b XRD patterns before and after illumination, c, d SEM images before and after illumination

Photocatalytic mechanism

The 120 °C, 180 °C, 3 EG and 500 °C BWO samples were selected as representatives to test the optical and electrical properties and further explain the change in photocatalytic properties. UV–Vis diffuse reflectance spectroscopy (DRS) data (Fig. 12a) showed that the prepared BWO samples displayed good light absorption in both the ultraviolet and visible regions, and steep light absorption edges were detected, indicating that the absorption of light by BWO was caused by an electronic transition at the energy band level instead of an electronic transition at the impurity band level [40]. The extension line of the light absorption edge is located in the wavelength range of 420-500 nm, indicating that BWO has the ability to respond to visible light and that photogenerated electrons can be excited under visible light. Compared with other samples, the redshift of the absorption edge of 180 °C BWO was the most obvious, indicating that good crystallinity was conducive to optical excitation of electrons. The bandgap width of the semiconductor was determined using Eq. 2.

$$\alpha hv={K(hv-Eg)}^{n/2}$$ 2

where α, h and v are the absorption coefficient, Planck constant and incident light frequency, respectively; n is set to 1 for a direct semiconductor and to 4 for a BWO-type indirect semiconductor [41]. The Tauc diagram was drawn to obtain the correlation between (αhv)² and hv, which showed that the minimum bandgap width of the 3 EG BWO sample was 2.90 eV, while the bandgap widths of BWO obtained by other preparation processes were approximately the same and ranged from 2.93 to 3.02 eV. This is due to the presence of oxygen vacancies in the 3 EG BWO sample that introduced impurity energy levels and reduced its band width [42]. The electrons excited by light have a certain energy, which is emitted in the form of fluorescence when the electrons in the excited state combine with defects and holes in the process of migration. Figure 12b illustrates the PL ability of the prepared BWO samples. The 120 °C BWO sample had the lowest amplitude peaks, suggesting that the recombination probability of photogenerated carriers was the lowest and that transfer and separation were the most efficient. Meanwhile, the variation in the PL intensity of the BWO samples matched the RhB degradation results, indicating that photogenerated electrons and holes with strong reducing and oxidizing abilities play a direct role in the photocatalytic degradation of pollutants.

Fig. 12.

a UV–Vis DRS spectra and (inset) bandgaps; b PL spectra of the 120 °C, 180 °C, 3 EG and 500 °C BWO samples

Photoelectrochemical tests were carried out in a conventional three-electrode electrochemical cell. A working electrode was prepared by dispersing 0.1 g of the BWO sample to be measured in an appropriate amount of ethanol, spin-coating the sample on FTO using a homogenizer and drying it. A platinum electrode and a saturated calomel electrode were used as the counter and reference electrodes, respectively. These three electrodes were dipped in a 0.5 M NaSO₄ electrolyte solution. Upon irradiation with visible light, a CHI660E electrochemical workstation was used to document electrochemical impedance spectroscopy (EIS) data, and the transfer efficiency of photogenerated carriers was further confirmed. Figure 13 shows that the arc radius of the 120 °C BWO/FTO film in the EIS Nyquist diagram was the smallest, indicating that it had the smallest photogenerated carrier migration resistance, and the rapid electron transfer effectively repressed recombination of photogenerated electron-hole pairs [43]. The rapid transfer of electrons effectively inhibits their recombination with defects or holes, which further proves that the main explanation for the improvement in the photocatalytic performance is the large specific surface area and small size of 120 °C BWO. The electrons generated by photoexcitation quickly reach the catalyst surface through the short migration path to participate in the degradation reaction rather than prematurely recombining with holes or other defects.

Fig. 13.

EIS spectra of the 120 °C, 180 °C, 3 EG and 500 °C BWO samples

Photocatalytic mechanism

To reveal the active substances of the catalyst in the photodegradation process, free radical trapping experiments were carried out on the 120 °C BWO sample. The results (Fig. 14a) show that after adding AgNO₃, KI, IPA, and BQ free radical quenching agents to the degradation system, the rate of RhB solution degradation by the 120 °C BWO catalyst decreased to 71.96%, 78.38%, 76.25% and 98.16%, respectively. Therefore, e⁻, h⁺, and ·OH can be proven to be the reactive species of the 120 °C BWO sample in the photocatalytic degradation system.

Fig. 14.

a Degradation rate of the 120 °C BWO photocatalyst for RhB solution after the addition of sacrificial agents; b schematic illustration of the photocatalytic mechanism based on 120 °C BWO under visible light

The conduction band (CB) potential E_CB and valence band (VB) potential E_VB of a semiconductor can be calculated according to empirical Eqs. 3 and 4 [44].

$$E_{\text{CB}}=X-E_{\text{e}}-0.5E_{\text{g}}$$ 3

$$E_{\text{VB}}=E_{\text{CB}}+E_{\text{g}}$$ 4

where X is the electronegativity of the semiconductor (BWO has a value of 6.36 eV) [45] and E_e is the energy of the free electron of the hydrogen atom (approximately 4.5 eV). The E_CB and E_VB values of the 120 °C BWO sample were calculated to be 0.35 eV and 3.37 eV, respectively. Comparing all the prepared samples, 120 °C BWO exhibited the most negative CB and the most positive VB, resulting in the 120 °C BWO sample possessing the best redox capacity. The photogenerated holes accumulated at the VB position are sufficient to oxidize H₂O and produce ·OH (E_H2O/·OH = + 2.72 eV vs. NHE), and ·OH will further participate in the redox reaction of pollutants [46]. The theoretical reduction potential of O₂ reduction to ·O₂⁻ is − 0.046 eV (vs. NHE), which is more negative than the E_CB of 120 °C BWO, resulting in ·O₂⁻ not being produced [47]. However, the photogenerated electrons located at the CB position can reduce O₂ to H₂O₂ because the redox potential of O₂/H₂O₂ is 0.695 eV. The generated H₂O₂ is further converted to ·OH by capturing electrons [48]. This conclusion corresponds well to the active radical capture experiments. On the basis of the aforementioned analyses, a reasonable mechanism process for photocatalytic degradation of RhB by 120 °C BWO is proposed, as shown in Fig. 14b. Under light irradiation, electrons are excited from the VB of BWO to the CB, producing a large number of photogenerated electrons that accumulate in the CB. These photogenerated electrons oxidize H₂O to produce ·OH, and the photogenerated holes reduce O₂ to H₂O₂ and are further converted to ·OH by capturing electrons. The generated ·OH will further participate in the photocatalytic degradation process of RhB. Due to the 2D round cake-shaped structure that had a large surface/volume ratio, 120 °C BWO exhibited the maximum specific surface area, which effectively enhanced the active sites involved in photocatalytic reactions. In addition, abundant oxygen vacancies on the surface of 120 °C BWO acted as traps for electrons, which could maintain the photogenerated holes in the VB and enhance the separation of carriers in space. More importantly, the fluorescence spectra (PL) showed that 120 °C BWO had the lowest PL intensity among all the as-prepared samples, suggesting that 120 °C BWO contributes to facilitating charge migration and suppressing recombination of photogenerated electron-hole pairs, thereby enhancing the photocatalytic performance.

Conclusions

In summary, BWO with a stable structure and high photocatalytic performance was obtained using a simple solvothermal method. The changes in the BWO morphology during the reaction were studied. BWO with different morphologies was obtained by adding EG, controlling the reaction rate and adjusting the properties of the solution. The specific surface area, grain size and agglomerated morphology of the catalyst exert important effects on the photocatalytic performance. The round cake-shaped 120 °C BWO sample has a large specific surface area that is conducive to the adsorption of pollutants and provides more reaction active sites for photocatalytic reactions. The accumulation of simple and small round cake particles results in a large pore volume, enabling photons to reach the depths of the catalyst to stimulate the production of photogenerated carriers in more particles and facilitating migration of electrons to the catalyst surface along a short path to participate in the degradation reaction. Excess EG can provide more ·OH with a reducing capacity, and the generated oxygen vacancies provide impurity energy levels, thus reducing the bandgap of BWO. However, because EG regulates the morphology of the catalyst by changing the chelation of the solution, when EG is excessive, the sample exists as a larger double concave cake structure, which limits the catalytic activity. The 120 °C BWO photocatalyst has the highest RhB photocatalytic degradation rate and has good degradation ability and recyclability for TC, which is difficult to degrade. Compared with other studies, the temperature used in the preparation process is lower, the holding time is shorter, and chemical substances such as strong acids or strong bases are not needed to modify and improve the properties of the liquid. The preparation method is simple, and the performance is excellent, properties that are very important to solve the problem of environmental water pollution.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. All authors read and approved the final manuscript. BDY contributed to the data curation, conceptualization, formal analysis, writing—original draft, writing—review and editing, and visualization; GWZ contributed to the formal analysis, fund acquisition, and writing—review and editing; YNL was involved in the instrumentation. LXW assisted in the methodology and investigation; XMX contributed to the instrument. YYW, YL and JXH were involved in the verification; FQZ contributed to the funding acquisition, project management, writing—review and editing. All authors read and approved the final manuscript.

Funding

This work was supported by Shandong Provincial Natural Science Fund (ZR2022ME071 and ZR2020KE020), the Research Fund for the Doctoral Program of Shandong Jianzhu University (Grant No. X21045Z), the Program of the Housing and Urban-Rural Construction Department of Shandong Province (2019-K7-10) and State Key Laboratory of Silicate Materials for Architectures (Wuhan University of Technology) (SYSJJ2020-05).

Data availability

The datasets generated and analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of  interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.