Remarkably enhanced catalytic performance in CoO_x/Bi₄Ti₃O₁₂ heterostructures for methyl orange degradation via piezocatalysis and piezo-photocatalysis

Graphical abstract

Highlights

• •CoO_x nanoparticles were photodeposited on Bi₄Ti₃O₁₂ nanosheets to form a novel piezocatalyst.
• •CoO_x/Bi₄Ti₃O₁₂ realized piezocatalytic organic degradation by harvesting ultrasonic vibration energy.
• •CoO_x acts as hole trappers to improve the charge separation of Bi₄Ti₃O₁₂.
• •CoO_x/Bi₄Ti₃O₁₂ showed superior piezo and piezo-photocatalytic performance than Bi₄Ti₃O₁₂.

Abstract

A novel heterojunction composite of CoO_x/Bi₄Ti₃O₁₂ was synthesized through a combination of molten salt and photodeposition methods. The optimal sample exhibited superior performance in the piezocatalytic degradation of methyl orange (MO) dye with a degradation rate of 1.09 h⁻¹, which was 2.4 times higher than that of pristine Bi₄Ti₃O₁₂. Various characterizations were conducted to reveal the fundamental nature accountable for the outstanding piezocatalytic performance of CoO_x/Bi₄Ti₃O₁₂. The investigation of the band structure indicated that the CoO_x/Bi₄Ti₃O₁₂ composite formed a type-I p-n heterojunction structure, with CoO_x acting as a hole trapper to effectively separate and transfer piezogenerated carriers. Significantly, the MO degradation rate of the best CoO_x/Bi₄Ti₃O₁₂ sample further increased to 2.96 h⁻¹ under combined ultrasonic vibration and simulated sunlight. The synergy between piezocatalysis and photocatalysis can be ascribed to the following factors. The photoexcitation process ensures the sufficient generation of charge carriers in the CoO_x/Bi₄Ti₃O₁₂, while the piezoelectric field within Bi₄Ti₃O₁₂ promotes the separation of electron-hole pairs in the bulk phase. Furthermore, the heterojunction structure between Bi₄Ti₃O₁₂ and CoO_x significantly facilitates the surface separation of charge carriers. This increased involvement of free electrons and holes in the reaction leads to a remarkable enhancement in catalytic MO degradation. This work contributes to the understanding of the coupling mechanism between the piezoelectric effect and photocatalysis, and also provides a promising strategy for the development of efficient catalysts for wastewater treatment.

1. Introduction

Low-frequency mechanical energy, abundantly present in nature without temporal or spatial constraints [1], has garnered increasing interest in studying its conversion into electrical energy within natural systems since the report of ZnO nanogenerators [2]. Particularly, the field of piezocatalysis, which involves utilizing piezoelectric polarization induced by ultrasound to drive reactions, has received significant attention. This technology not only enables efficient utilization of mechanical energy but also has potential applications in water purification and energy conversion [3], [4]. Piezoelectricity exhibits an energy conversion efficiency of approximately 78 %, surpassing the energy conversion efficiency of photovoltaic technology (around 20 %), demonstrating the great potential of piezocatalysis for energy conversion and environmental remediation purposes [5], [6]. However, research on piezocatalysis is still in its early stages, and the development of new piezocatalytic materials is a crucial requirement to propel the advancement of this technology.

Bi₄Ti₃O₁₂ is a layered perovskite oxide composed of alternating (Bi₂O₂)²⁺ layers and (Bi₂Ti₃O₁₀)^2- pseudo-perovskite layers [7]. This layered structure provides an appropriate architecture for the growth of ultrathin nanosheets, resulting in shorter charge migration distances and increased exposure of active sites [8]. Additionally, due to the coherent off-center displacement of Ti ions within the TiO₆ octahedral units, Bi₄Ti₃O₁₂ exhibits a large spontaneous polarization (P_s = 50 μC cm⁻²) [9] and a strong piezoelectric response (d₁₁ = 39 pC N⁻¹) [10]. However, due to the aggregation of nanoplates, the spontaneous polarization in random directions can cancel each other out, leading to a reduction in total piezoelectric susceptibility [11]. Therefore, the piezocatalytic performance of Bi₄Ti₃O₁₂ in current reports is not sufficiently high [12], [13], [14]. As a common candidate for efficient hole trapping and constructing hole transport channels, CoO_x has been utilized as a cocatalyst for heterostructural engineering in numerous semiconductor photocatalysts, including CoO_x/TiO₂ [15], CoO_x/ZnO [16], CoO_x/Ta₃N₅ [17], and CoO_x/Ag₃PO₄ [18]. CoO_x has also been proven to be effective in the separation of piezoelectric-induced carriers. Wang et al. successfully achieved efficient capture and transport of piezoelectric-induced holes by anchoring CoO_x onto the surface of BiFeO₃ nanodisks, resulting in a significant enhancement of the piezocatalytic performance of BiFeO₃ [19]. Additionally, Liu et al. selectively deposited CoO_x on the {1 1 0} plane of BiVO₄ to trap holes together with methanol, substantially increasing the efficiency of piezoelectric hydrogen production [20]. To date, piezocatalytic performance of Bi₄Ti₃O₁₂ has primarily been promoted through morphology regulation [12] and the introduction of oxygen vacancies [21], [22]. However, there have been no reports on the investigation of the CoO_x/Bi₄Ti₃O₁₂ heterojunction for piezocatalytic applications.

Apart from regulating the composition and properties of the catalyst, a key factor in enhancing catalytic efficiency is the synergistic interaction between light fields and piezoelectric fields. Several studies have explored the piezo-photocatalytic performance of semiconducting materials with piezoelectric properties, such as BiOCl [23], BaTiO₃ [24], and ZnO [25]. Some studies [13], [21], [26] have also reported on the piezo-photocatalytic performance and catalytic mechanism of Bi₄Ti₃O₁₂. Generally speaking, photodriven processes have a higher propensity for carrier injection. When driven by the piezoelectric polarized electric field, photoinduced charge carriers exhibit reduced recombination efficiency and migrate efficiently to the surface to participate in redox reactions. Therefore, coupling the effects of piezoelectricity and photoexcitation could be an effective approach to further improve the catalytic performance of CoO_x/Bi₄Ti₃O₁₂.

In this study, a novel catalyst, CoO_x/Bi₄Ti₃O₁₂, was synthesized via photodeposition and utilized for the piezocatalytic degradation of methyl orange (MO). The piezocatalytic activities of unmodified and CoO_x-deposited Bi₄Ti₃O₁₂ nanosheets were compared based on the monitoring of the degradation rate constant of MO. Moreover, CoO_x/Bi₄Ti₃O₁₂ exhibited enhanced catalytic performance through the synergistic effect of piezocatalysis and photocatalysis. The structures, morphologies, and electrochemical properties of the CoO_x/Bi₄Ti₃O₁₂ composite were systematically characterized to understand the underlying mechanism behind its superior catalytic properties. Furthermore, the piezocatalytic potential of CoO_x/Bi₄Ti₃O₁₂ was investigated for the degradation of diverse organic pollutants.

2. Experimental

2.1. Preparation of CoO_x/Bi₄Ti₃O₁₂ catalysts

Bi₄Ti₃O₁₂ was synthesized using a one-step molten salt synthesis method as described in previous literature [27]. The stoichiometric ratios of Bi₂O₃ (2 mmol), TiO₂ (3 mmol), NaCl (60 mmol), and KCl (60 mmol) were thoroughly mixed and ground for 30 min. Subsequently, the mixed powder was subjected to calcination in an air atmosphere at 800 °C for a duration of two hours. After cooling, the resulting product was submerged in a 120 mL solution of 0.5 M HNO₃ for 10 min. The Bi₄Ti₃O₁₂ powders were then obtained by rinsing multiple times with deionized water and alcohol, followed by drying at 60 °C for 12 h. The CoO_x/Bi₄Ti₃O₁₂ catalyst was prepared using a photodeposition method. A specific amount of Co(OAc)₂·4H₂O and 0.5 g of Bi₄Ti₃O₁₂ were dispersed in a mixture of 10 mL methanol and 40 mL deionized water. After stirring for 30 min in the dark, the homogeneous suspension was exposed to a Xenon lamp (model PLS-SXE300C, 300 W) for 20 min. Ultimately, the suspension was filtered and rinsed multiple times with deionized water and ethanol, then dried at 60 °C for 12 h to obtain the CoO_x/Bi₄Ti₃O₁₂ composite. CoO_x/Bi₄Ti₃O₁₂ composites were prepared with different molar ratios of Co(OAc)₂·4H₂O to Bi₄Ti₃O₁₂ (20 %, 30 %, 50 %, 70 %). The resulting composites were labeled as CoO_x/BiTO-n (n = 1, 2, 3, and 4).

2.2. Piezocatalytic test and characterizations of CoO_x/Bi₄Ti₃O₁₂ catalysts

The degradation of MO (5 ppm) under mechanical pressure was employed to assess the piezocatalytic activity of the CoO_x/Bi₄Ti₃O₁₂ composite. The mechanical pressure was generated by a digital ultrasonic generator (JP010T, 40 kHz, 120 W). In the case of the piezo-photocatalytic reaction, a 300 W Xe lamp was utilized as the light source. Further details regarding the piezocatalytic test and the characterizations of the CoO_x/Bi₄Ti₃O₁₂ catalyst can be found in the supplementary materials.

3. Result and discussion

3.1. Characterization of CoO_x/Bi₄Ti₃O₁₂ composite

The actual content of CoO_x in the CoO_x/Bi₄Ti₃O₁₂ was investigated through X-ray fluorescence spectroscopy (XRF) analysis. The results revealed that the molar ratios of CoO_x to Bi₄Ti₃O₁₂ in the CoO_x/BiTO-n (n = 1, 2, 3, and 4) samples were approximately 0.27 %, 0.43 %, 0.55 %, and 0.47 %, respectively, which were significantly lower than the theoretical ratios of 20 %, 30 %, 50 %, and 70 %, respectively. In fact, the substantial difference between the actual and theoretical content of photodeposited metal oxides is common and usually attributed to the complex photodeposition process [28], [29]. CoO_x is formed through the direct oxidation of Co²⁺ by photogenerated holes [30]. Consequently, the poor adsorption of Co²⁺ on the surface of Bi₄Ti₁₃O₁₂ is considered the primary reason for the low CoO_x content. Additionally, cobalt acetate is a salt with weak acidity. Increasing the content of Co(OAc)₂ lowers the solution pH, which weakens the adsorption of Co²⁺ on the surface of Bi₄Ti₁₃O₁₂, thereby hindering the photodeposition of CoO_x. Hence, increasing the concentration of Co(OAc)₂ does not always enhance the CoO_x content in the synthesized CoO_x/Bi₄Ti₃O₁₂.

X-ray diffraction (XRD) was employed to examine the crystal structures of Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂, and the results are presented in Fig. 1a. No impurity signals, such as Bi₂O₃ and TiO₂, were observed, indicating a high level of purity. The strong diffraction signals observed for the peaks corresponding to the (0n0) lattice plane (i.e. (0 4 0), (0 6 0), (0 8 0), and (0140) planes) can be attributed to the lower bonding energy generated by the Bi₄Ti₃O₁₂ layered crystal along the b-axis, promoting the growth of the (0n0) plane [21]. No signals corresponding to CoO_x species were detected, which may be due to the low content, high dispersion, and low crystallinity of CoO_x on the surface of Bi₄Ti₃O₁₂ [31], [32]. Fig. 1b illustrates the Raman spectra of pristine Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂. The pristine Bi₄Ti₃O₁₂ exhibits eight Raman signals in the measured range, with Raman active modes at 228, 268, 330, 566, 562, and 850 cm⁻¹ assigned to the stretching and bending vibrations of the TiO₆ octahedra [33], [34]. The two-dimensional morphological characteristics of Bi₄Ti₃O₁₂ may enhance the distortion of the TiO₆ octahedra, leading to the observation of the Raman signal at 228 cm⁻¹. No other modes were observed in the Raman spectra of CoO_x/Bi₄Ti₃O₁₂, which can be attributed to the same reason identified by the XRD analysis. However, a slight shielding effect is observed in the Raman spectra when CoO_x species are added to Bi₄Ti₃O₁₂, providing circumstantial evidence for the presence of CoO_x.

Fig. 1.

XRD and Raman spectra of Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂ composites.

To investigate the sample morphology and gather detailed information about the microstructure, scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of CoO_x/BiTO-3 were obtained. Bi₄Ti₃O₁₂ exhibits a square nanosheet morphology with varying sizes and an average thickness of approximately 100 nm (Fig. 2a and b). The flat sheet shape facilitates efficient migration of charge carriers from the bulk to the surface, thereby minimizing recombination within internal defects [26]. Due to the low concentration and high dispersion of CoO_x, observing the simultaneous presence of both CoO_x and Bi₄Ti₃O₁₂ in the SEM and TEM images of CoO_x/BiTO-3 is challenging. However, high-resolution TEM provides more detailed structural information. Pure Bi₄Ti₃O₁₂ reveals a series of lattice stripes with an average spacing of 0.407 nm, corresponding to the (0 8 0) crystal surface of Bi₄Ti₃O₁₂ (Figure S1). Additionally, the surface appears clean. In the case of CoO_x/Bi₄Ti₃O₁₂, an amorphous nanoparticle was observed at the edge of this crystal plane, indicating the presence of CoO_x species (Fig. 2c). To further confirm the successful loading of CoO_x onto the Bi₄Ti₃O₁₂ surface, energy-dispersive spectroscopy (EDS) elemental mapping was conducted (Fig. 2d). The EDS map clearly shows the presence of Co elements, providing additional evidence of the successful incorporation of CoO_x onto the Bi₄Ti₃O₁₂ surface.

Fig. 2.

SEM (a), TEM (b, c), EDS and element mapping (d) images CoO_x/BiTO-3 sample.

X-ray photoelectron spectroscopy (XPS) analysis was employed to assess the elemental composition and valence states on the surface of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 samples. The Co 2p XPS spectrum of CoO_x/BiTO-3 is depicted in Fig. 3a, showing a Co 2p_3/2 peak at 780.9 eV, indicating the presence of Co species in the CoO_x/Bi₄Ti₃O₁₂ system. Given that the binding energy (BE) of Co²⁺ and Co³⁺ is very close [31], [35], this peak can be attributed to the mixed valence states of + 2 and + 3. Combining the analysis of XRD, Raman spectra, and high-resolution TEM images, the presence of CoO_x in the CoO_x/Bi₄Ti₃O₁₂ sample can be confidently confirmed. The Ti 2p XPS spectrum of Bi₄Ti₃O₁₂ (Fig. 3b) exhibits a peak at 457.6 eV assigned to Ti 2p_3/2. The broad peak can be deconvoluted into two peaks at 463.3 and 465.5 eV, corresponding to Ti 2p_1/2 and Bi 4d_3/2, respectively [36]. The energy difference between Ti 2p_3/2 (457.6 eV) and Ti 2p_1/2 (463.3 eV) corresponds to 5.7 eV, indicating the presence of Ti⁴⁺ [37]. The Bi 4d_3/2 peak of the CoO_x/BiTO-3 sample exhibits a slight shift towards higher binding energy compared to the Bi₄Ti₃O₁₂ sample. A similar shift is also observed in the Bi 4f spectra. The Bi4f_7/2 and 4f_5/2 peaks of Bi₄Ti₃O₁₂ are identified at 158.6 and 163.8 eV, respectively, corresponding to Bi³⁺ species [38], [39] (Fig. 3c). For CoO_x/BiTO-3 sample, the Bi³⁺ signals shift to 158.8 and 164.0 eV, suggesting a decrease in electron cloud density around Bi in the CoO_x/BiTO-3. Conversely, the Ti 2p peak of CoO_x/BiTO-3 does not exhibit an offset, possibly due to the stronger Coulomb binding effect between the electrons in the Ti 2p layers and the nucleus, as compared to the Bi 4f layers. Fig. 3d shows the O 1 s XPS spectra of the two catalysts. The O1s peak of Bi₄Ti₃O₁₂ is observed at 529.2 and 531.8 eV, corresponding to lattice oxygen and surface OH species [38], [39]. In contrast, the lattice oxygen peak of CoO_x/BiTO-3 exhibits an offset behavior similar to that of Bi, suggesting that the surrounding electron cloud density of O is significantly influenced by the electron transfer between CoO_x and Bi₄Ti₃O₁₂.

Fig. 3.

XPS spectra of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 sample: (a) Co 2p; (b) Ti 3d; (c) Bi 4f; (d) O1s.

Fig. 4a illustrates the hysteresis loops acquired from the CoO_x/BiTO-3 sample using a piezoelectric force microscope (PFM). Despite the imperfect preparation of the working film (Figure S2), the presence of typical amplitude-voltage curves and hysteresis loops can still be observed, indicating the piezoelectric properties of the CoO_x/Bi₄Ti₃O₁₂ samples and their potential applications in the field of piezoelectric catalysis. The ultraviolet–visible (UV–vis) diffuse reflectance spectra (DRS) of Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂ composite are presented in Figure S3. It is evident that the presence of CoO_x increases the visible light absorption of Bi₄Ti₃O₁₂ without affecting its absorption edge position. The absorption edge of pure Bi₄Ti₃O₁₂ is approximately 394 nm, corresponding to a band gap of 3.15 eV (inset of Fig. 4b), consistent with previous reports [40], [41]. Mott-Schottky (MS) analysis was conducted to determine the band structure of Bi₄Ti₃O₁₂, as shown in Fig. 4b. The positive slope in the MS plot confirms that Bi₄Ti₃O₁₂ is an n-type semiconductor, with a flat-band potential (V_fb) of −1.59 V (vs Ag/AgCl). The corrected V_fb, taking into account the pH level, is estimated to be −1.01 V (vs NHE). In many studies, V_fb is commonly used to replace the Fermi level (E_f) for band calculations [42], [43]. And for non-degenerate n-type semiconductors, the E_f is typically 0.1 ∼ 0.2 V lower than the potential of conduction band (CB) [44]. Consequently, the CB position of Bi₄Ti₃O₁₂ is determined to be −1.21 V, in excellent agreement with previous findings [22]. The valence band (VB) potential is calculated to be 1.94 V using the equation E_g = E_VB − E_CB.

Fig. 4.

Hysteresis loop of CoO_x/BiTO-3 composite (a), MS plots of Bi₄Ti₃O₁₂ (b)_, and band diagram of CoO_x/Bi₄Ti₃O₁₂ composite (c).

As a p-type semiconductor, CoO_x shares a band structure resembling that of CoO and Co₃O₄ (Figure S4), with a band gap ranging from 2.00 to 2.15 eV [45], [46]. According to reports, the work function of CoO_x is 5.60 eV [46], which implies that the Fermi energy of CoO_x is located at 1.10 eV. In the case of p-type semiconductors, the VB top typically lies 0.1–0.2 V below the E_f. Consequently, the VB of CoO_x can be estimated to be 1.30 eV, exceeding that of Bi₄Ti₃O₁₂, as depicted in Fig. 4c. Upon close contact between the two semiconductors, electrons transfer from regions of high concentration (Bi₄Ti₃O₁₂ side) to regions of low concentration (CoO_x side), propelled by the Fermi energy difference between Bi₄Ti₃O₁₂ and CoO_x. As a result of electron redistribution, the energy bands at the interfaces of Bi₄Ti₃O₁₂ and CoO_x exhibit upward and downward bending, respectively, leading to the formation of a potential barrier with a specific height. The opposite bending of the VB facilitates the migration of holes from Bi₄Ti₃O₁₂ to CoO_x, while the potential barrier formed at the CB impedes the migration of electrons at the interface of Bi₄Ti₃O₁₂. Significantly, electron migration results in the formation of an intrinsic electric field at the interface of the two phases (Fig. 4c), enhancing the hole collection by CoO_x. This distinctive type-I heterojunction structure effectively mitigates charge carrier recombination and enhances the efficiency of charge separation and migration.

A series of electrochemical tests were conducted to assess the efficiency of charge separation and transport in samples of Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂, providing additional confirmation of the previously stated hypothesis. Fig. 5a illustrates the electrochemical impedance spectroscopy (EIS) curves of Bi₄Ti₃O₁₂ and CoO_x/Bi₄Ti₃O₁₂ composites with different CoO_x contents. With an increase in the theoretical molar ratio of CoO_x to Bi₄Ti₃O₁₂, the arc radius of the as-synthesized CoO_x/Bi₄Ti₃O₁₂ composite initially decreases and subsequently increases. The CoO_x/BiTO-3 sample exhibits the minimum radius of curvature, indicating the lowest interface charge transfer resistance and the highest charge separation efficiency [47], [48], [49]. Given that the CoO_x/BiTO-3 sample has the highest CoO_x content (XRF analysis), the increased number of heterojunctions may contribute to its superior charge carrier separation capability. Fig. 5b presents the photoluminescence (PL) emission spectrum of both Bi₄Ti₃O₁₂ and a representative sample, CoO_x/Bi₄Ti₃O₁₂. The lower PL intensity of CoO_x/BiTO-3 compared to pure Bi₄Ti₃O₁₂ suggests a decrease in the electron-hole recombination rate within the composite catalyst [50], [51]. Similar results were obtained from the linear sweep voltammetry (LSV) analysis, as shown in Fig. 5c. At a current density of −1.0 mA cm⁻², the reduction potentials of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 were measured as −1.428 and −1.418 V, respectively. The lower reduction potential indicates that the CoO_x/Bi₄Ti₃O₁₂ sample possesses stronger reduction capability due to its superior charge separation [52]. Notably, an additional reductive peak is observed in the LSV curve of CoO_x/BiTO-3, indicating the reduction of CoO_x and confirming its presence. The transient piezoelectric current responses of both Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 samples were assessed under ultrasonic vibration, and the results are depicted in Fig. 5d. The piezoelectric current generated by CoO_x/BiTO-3 is significantly higher than that of Bi₄Ti₃O₁₂ under periodic ultrasound vibration, indicating enhanced charge transport performance resulting from the inclusion of CoO_x [53], [54]. Integrating the analysis of EIS, PL, and LSV results, it can be inferred that the formation of a CoO_x/Bi₄Ti₃O₁₂ heterojunction structure promotes the separation and transport of charge carriers.

Fig. 5.

EIS (a), PL (b), LSV (c), and piezoinduced current response (d) of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 samples.

3.2. Piezocatalytic and piezo-photocatalytic activity of CoO_x/Bi₄Ti₃O₁₂

The piezocatalytic activity of Bi₄Ti₃O₁₂ and CoO_x/ Bi₄Ti₃O₁₂ composites was assessed for the removal of MO under ultrasonic vibration, as depicted in Fig. 6. As a control experiment (blank test, sonolysis), the concentration of MO remained nearly constant in the degradation system without catalysts (Fig. 6a), suggesting that the self-decomposition characteristics of MO under ultrasonic vibration can be ignored. However, after being subjected to 1.5 h of ultrasonic vibration treatment, Bi₄Ti₃O₁₂ achieved approximately 45 % degradation of MO, with an estimated degradation rate constant of 0.46 h⁻¹ (Fig. 6b). This is attributed to the sheet-like structure of Bi₄Ti₃O₁₂, which is susceptible to deformation under mechanical stress, thus exhibiting excellent piezoelectric response characteristics and piezoelectric catalytic activities. The adsorption of MO onto Bi₄Ti₃O₁₂ is relatively weak. After 1 h of adsorption, the MO content decreases by 1.3 %, possibly due to the electrostatic repulsion between the negatively charged surface of Bi₄Ti₃O₁₂ and the anionic MO molecule [26]. The introduction of CoO_x slightly increases MO adsorption, possibly due to its role in elevating the Zeta potential of Bi₄Ti₃O₁₂ (Figure S5). The increased repulsive forces between the catalyst particles render CoO_x/Bi₄Ti₃O₁₂ more readily dispersible in the solution, thereby facilitating the adsorption of MO. Additionally, the CoO_x nanoparticles can serve as active sites for MO adsorption due to their substantial surface area. Nevertheless, the limited quantity of CoO_x curtails its impact on the surface property of Bi₄Ti₃O₁₂, thereby preventing a significant enhancement in MO adsorption upon CoO_x addition. Conversely, the incorporation of CoO_x leads to noteworthy enhancement in the piezocatalytic activity of Bi₄Ti₃O₁₂. Moreover, the catalytic performance of CoO_x/Bi₄Ti₃O₁₂ is closely correlated with the CoO_x content. As shown in Fig. 6a, the degradation rate of MO by the CoO_x/Bi₄Ti₃O₁₂ catalyst initially increases and then decreases with an increasing theoretical molar ratio of CoO_x to Bi₄Ti₃O₁₂. The sample of CoO_x/BiTO-3 exhibits the highest MO degradation rate of 76 % under 1.5 h of sonication. The degradation rate constant is estimated to be 1.09 h⁻¹ (Figure S6), which is 2.4 times greater than that of the pristine Bi₄Ti₃O₁₂. Total organic carbon (TOC) analysis was conducted to further evaluate the piezocatalytic activity of CoO_x/Bi₄Ti₃O₁₂ (Fig. 6b). The enhanced TOC removal efficiency is still observed in the CoO_x/Bi₄Ti₃O₁₂ composite. CoO_x/BiTO-3 sample exhibits the highest TOC removal efficiency of 72.1 %, indicating its best capability in piezocatalytic degradation of MO. Generally, the changes in catalytic activity resulting from variations in the concentration of co-catalysts are explained by employing an optimal co-catalyst concentration. This is due to the fact that high concentrations of additives can result in surface agglomeration of composites and the loss of active sites [55], [56], which negatively impacts the piezocatalytic reaction. XRF analysis in this study indicates that the photodeposition method is not effective in loading CoO_x onto Bi₄Ti₃O₁₂. The CoO_x content in the CoO_x/Bi₄Ti₃O₁₂ composite is extremely low, while the CoO_x/BiTO-3 composite exhibits the highest CoO_x content of 0.55 %. Therefore, it is believed that the CoO_x content does not exceed the optimal balance point of the CoO_x/Bi₄Ti₃O₁₂ composites. The CoO_x/Bi₄Ti₃O₁₂ composite, with the highest CoO_x content, demonstrates superior hole trapping capability as confirmed by electrochemical analysis, thereby exhibiting the highest piezocatalytic performance.

Fig. 6.

Piezoelectric catalytic activity of CoO_x/Bi₄Ti₃O₁₂ composites (a), the corresponding TOC removal efficiency (b). Cycling test of CoO_x/BiTO-3 sample. Influence of solution pH (d) and co-existing ions (e), and scavengers (f) on the piezocatalytic activity (d) of CoO_x/BiTO-3.

The stability of catalysts is a crucial factor in piezocatalytic applications. Therefore, the stability of CoO_x/BiTO-3 was assessed through cyclic degradation tests, as depicted in Fig. 6c. Even after five cycles, CoO_x/BiTO-3 consistently maintains an efficiency of approximately 75 % in degrading MO, indicating the establishment of a stable heterojunction structure between Bi₄Ti₃O₁₂ and CoO_x. This stability is also verified using the XRD and Raman analysis of the used CoO_x/BiTO-3 sample (Figure S7).

The effect of solution pH and the influence of co-existing ions on the piezocatalytic activity of CoO_x/Bi₄Ti₃O₁₂ are presented in Fig. 6d and 6e. Fig. 6d indicates that the CoO_x/Bi₄Ti₃O₁₂ catalyst exhibits superior performance under acidic conditions. As the solution pH gradually decreased from 6 to 3, there is a corresponding increase in the degradation rate constant, rising from 1.09 h⁻¹ to 1.67 h⁻¹ (Fig. S8a). Conversely, elevating the pH of the solution diminishes the efficiency of piezocatalytic activity. This phenomenon can be attributed to the varying adsorption capacity of the anionic dye MO onto the catalyst's surface. Under acidic conditions, the catalyst's surface carries a positive charge, facilitating the adsorption of MO and enhancing the piezoelectric catalytic reaction. Fig. 6e illustrates the impact of coexisting ions (SO₄^2-, CO₃^2–, NO₃^–, and Cl^-) on the piezocatalytic performance of CoO_x/BiTO-3. Evidently, the introduction of these ions hampers the piezocatalytic degradation of MO (Fig. S8b), primarily due to their competitive adsorption with MO on the catalyst surface [57]. Furthermore, these ions can react with the reactive species, thereby diminishing the degradation efficiency of CoO_x/BiTO-3. The interference of Cl^- with MO removal is minimal, likely because the addition of Cl^- may result in the formation of hypochlorite in the solution [58], consequently mitigating the adverse effects of Cl^-.

Fig. 8.

Piezocatalytic activity of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 composite for different organics.

Piezocatalytic reactions are usually accompanied by the generation of the active species. To identify the dominant active species in the piezocatalytic system, radical trapping experiments were conducted on the CoO_x/BiTO-3 sample. Benzoquinone (BQ), KI, and isopropanol (IPA) were employed as sacrificial agents to capture superoxide radicals (·O₂^–), holes (h⁺), and hydroxyl radicals (·OH), respectively. The results are presented in Fig. 9, Fig. 6f. As the concentration of the sacrificial agent increased, there was a substantial decrease in the piezocatalytic activity of the CoO_x/BiTO-3 catalyst, indicating the simultaneous presence of hydroxyl radicals, holes, and superoxide radicals as three active species in the reaction. However, the inhibitory effect of sacrificial agents on the piezocatalytic activity varied. When the concentration of BQ was 0.1 mmol/L, the decolorization rate of the CoO_x/BiTO-3 catalyst dropped to 40 %, while to achieve the same effect, KI and IPA concentrations needed to be raised to 0.6 and 2.6 mmol/L, respectively. This result suggests that, compared to hydroxyl radicals, superoxide radicals and holes are the primary active species [32]. Electron paramagnetic resonance (ESR) analysis further confirms the active species identified in the capture experiment. As shown in Fig. 7, after 5 min of sonication, strong signals of ·O₂^– and h⁺ are observed in the reaction solution of CoO_x/BiTO-3. Notably, the two signals are stronger than those of Bi₄Ti₃O₁₂, which indicates that CoO_x/Bi₄Ti₃O₁₂ catalyst exhibits better capability in generating reactive species. This finding is consistent with the piezocatalytic test and confirms the elucidated charge transfer mechanism at the interface between Bi₄Ti₃O₁₂ and CoO_x. Although IPA exhibits a slight inhibitory effect in the capture experiment, no signals of ·OH ∙are detected in the ESR results, possibly due to its low concentration. Considering that the standard redox potential of H₂O/·OH (2.27 eV vs. NHE) [59], [60] is significantly higher than the VB potential of Bi₄Ti₃O₁₂ (1.94 eV), the oxidation of H₂O by Bi₄Ti₃O₁₂ is challenging. However, the piezoelectric field induced by ultrasound-generated polarization can alter the energy band [21], causing the VB position of Bi₄Ti₃O₁₂ to approach the potential of H₂O/·OH, thereby resulting in the generation of a certain amount of ·OH.

Fig. 9.

UV–visible absorption spectra of MO solution during piezocatalysis (a), photocatalysis (b), and piezo-photocatalysis (c) of CoO_x/BiTO-3 (a) and the corresponding degradation kinetics (d).

Fig. 7.

ESR spectra of DMPO-O₂^– (a), TEMPO-h⁺ (b), and DMPO-·OH (c) for CoO_x/BiTO-3 sample.

The CoO_x/Bi₄Ti₃O₁₂ composites also show great potential for the degradation of various organic pollutants. Fig. 8 demonstrates the piezocatalytic degradation activity of different organic pollutants by original Bi₄Ti₃O₁₂ and CoO_x/BiTO-3. Although CoO_x/BiTO-3 exhibits limited piezocatalytic properties in the degradation of tetracycline hydrochloride and p-nitrophenol, the introduction of CoO_x significantly increases their degradation rate constants by nearly threefold. This enhancement is also observed in the degradation of malachite green, rhodamine B, and methylene blue, with rate constants increased by 1.3, 1.8, and 1.5 times, respectively. Importantly, CoO_x/BiTO-3 exhibits strong adsorption capacity for various contaminants, which can be ascribed to its large exposed surface area.

Excitingly, the CoO_x/Bi₄Ti₃O₁₂ piezocatalyst also exhibits exceptional ability in piezo-photocatalysis. Fig. 9a-c present the ultraviolet–visible (UV–vis) absorption spectra of an MO solution catalyzed by CoO_x/BiTO-3 under different catalytic conditions. The absorption intensity of the MO solution gradually decreases with the passage of ultrasound time, indicating the progress of the catalytic reaction. Compared to piezocatalysis and photocatalysis alone, the absorbance of the MO solution corresponding to piezo-photocatalysis shows a characteristic “cliff-type” decline. Importantly, the piezo-photocatalytic rate constant of CoO_x/BiTO-3 for MO is 2.96 h⁻¹, which is 2.7 times higher than the piezocatalytic rate constant and 1.8 times higher than the photocatalytic rate constant, respectively (Fig. 9d). This significant enhancement of the rate constant demonstrates a strong coupling effect between the piezoelectric effect and photocatalysis. The above conclusion is further supported by transient current response and electrochemical impedance spectroscopy (EIS), as shown in Fig. 10. Under the combined action of vibration and simulated sunlight, CoO_x/BiTO-3 exhibits a pronounced transient current response, with a much higher current value than that under vibration or light alone. This finding aligns with the results in Fig. 9d, indicating the role of the piezoelectric-induced polarized electric field in facilitating the separation of photo-generated carriers. Similarly, CoO_x/BiTO-3 under piezo-photo synergy shows a smaller radius of curvature, indicating lower charge transfer resistance at the interface. In other words, the piezoelectric field drives the migration of photo-generated electrons, thereby facilitating subsequent cooperative catalytic reactions.

Fig. 10.

EIS (a) and current response (b) of Bi₄Ti₃O₁₂ and CoO_x/BiTO-3 under light, vibration and light-vibration.

3.3. The mechanism for enhanced catalytic activity of CoO_x/Bi₄Ti₃O₁₂

Based on the obtained characterization results and catalytic test, the objectivity and effectiveness of the CoO_x/Bi₄Ti₃O₁₂ heterojunction structures are successfully demonstrated. TEM and XPS analysis reveal the deposition of CoO_x nanoparticles onto the surface of square Bi₄Ti₃O₁₂ nanosheets. XRF results demonstrate that CoO_x exists in low content and is highly dispersed on the surface of Bi₄Ti₃O₁₂. Moreover, Bi₄Ti₃O₁₂ possesses a certain piezoelectric foundation due to its unique layered structure. Therefore, there is no doubt that Bi₄Ti₃O₁₂ serves as the active phase of the composite. In the presence of ultrasound, the cavitation-induced mechanical wave acts on Bi₄Ti₃O₁₂. Due to its unique planar plate morphology, Bi₄Ti₃O₁₂ is highly susceptible to deformation under mechanical stress, resulting in superior piezoelectric polarization. Nonetheless, due to the compound effect and the limited polarization potential, free charges truly enriched to the surface are minimal. Although the inner recombination effect of charge carriers is reduced by the axial size of Bi₄Ti₃O₁₂, surface recombination still occurs relatively frequently. Consequently, this imposes additional limitations on the piezocatalytic efficiency of pristine Bi₄Ti₃O₁₂.

To overcome this limitation, a type-I heterojunction structure is introduced through in situ photodeposition loading of CoO_x. The introduction of CoO_x does not damage the structure of Bi₄Ti₃O₁₂ itself (Fig. 6a and 6b), which provides the foundation for the modification of piezoelectric materials. CoO_x enhances the absorption of visible light due to its narrower band gap, but its contribution to piezoelectric catalysis in the absence of light is negligible. Meanwhile, due to the low content of the introduced CoO_x, the morphology of Bi₄Ti₃O₁₂ undergoes minimal alteration. Consequently, one may deduce that the deposition of CoO_x has an insignificant impact on the surface area of the Bi₄Ti₃O₁₂. This inference can be verified by the pre-adsorption processes of the catalyst before the reaction. The introduction of CoO_x does not effectively promote the adsorption of MO on the Bi₄Ti₃O₁₂ catalyst (Fig. 6a). Therefore, the enhancement of piezocatalytic activity is primarily attributed to the improved efficiency of carrier separation. Analyses of the band structure indicate that CoO_x and Bi₄Ti₃O₁₂ have matching VB positions, allowing for the formation of hole transmission channels. Driven by the built-in electric field in the p-n junction, piezoelectric-induced holes can migrate from Bi₄Ti₃O₁₂ to CoO_x. The barrier formed by band bending causes the accumulation of electrons at the interface of Bi₄Ti₃O₁₂. This process greatly promotes the surface transport of thermally excited charge carriers and accelerates the piezocatalytic reaction. Hence, compared to pure Bi₄Ti₃O₁₂, CoO_x/Bi₄Ti₃O₁₂ can generate a higher yield of ·O₂^– and h⁺ (Fig. 7) under ultrasonic vibration, which aligns with the piezocatalytic activity. However, it is noteworthy that piezoelectric catalysis primarily relies on charge carriers generated from thermal-excited free electrons and holes [53], [61], with their concentration being notably low at room temperature, thereby constraining the efficiency of piezoelectric catalysis. Introducing light can mitigate this limitation. Simulated sunlight exposure results in the generation of a substantial quantity of electron-hole pairs within Bi₄Ti₃O₁₂. Ultrasound vibrations induce an internal piezoelectric field in Bi₄Ti₃O₁₂, facilitating the directed migration of photogenerated electrons and holes towards the surface, effectively suppressing bulk charge carrier recombination. Additionally, the heterojunction structure at the Bi₄Ti₃O₁₂/CoO_x interface enhances the separation of surface charge carriers, with holes migrating to the CoO_x surface and electrons accumulating on the Bi₄Ti₃O₁₂ surface, thereby actively engaging in the catalytic reaction. The results of EIS and transient current response tests (Fig. 10) suggest that light injection under vibration can provide more free charge carriers for the catalytic degradation process, thereby corroborating the proposed inference. Piezo-photocatalysis experiments conducted on CoO_x/BiTO-3 demonstrate clear synergistic effects between piezocatalysis and photocatalysis, further providing additional confirmation of the aforementioned hypothesis. Under the joint drive of piezoelectric polarization and the heterojunction structure, photo-generated charge carriers can be effectively separated and rapidly migrate to the surface, significantly reducing the recombination probability and leading to high efficiency in degrading organic pollutants (Fig. 11). Certainly, CoO_x is capable of generating electron-hole pairs under light exposure and contributing to the catalytic degradation of MO [31], [32]. Nonetheless, due to its low concentration and inherent limitations in charge carrier separation, its impact in this context may be negligible. The principal function of CoO_x should be to act as a hole scavenger, enhancing the charge carrier separation efficiency of the composite catalyst.

Fig. 11.

Possible piezo-photocatalytic mechanism of CoO_x/Bi₄Ti₃O₁₂.

4. Conclusion

In summary, piezoelectric Bi₄Ti₃O₁₂ nanosheets were fabricated using a modified molten salt method, and CoO_x/Bi₄Ti₃O₁₂ composites were subsequently prepared through a photodeposition mechanism, further enhancing the piezocatalytic activities of Bi₄Ti₃O₁₂. The deposition of CoO_x accelerated surface charge separation through a type I mechanism, which was considered the source of the enhanced piezocatalytic activity in CoO_x/Bi₄Ti₃O₁₂. Moreover, external field-coupled regulation provides an additional driving force to the catalyst, promoting the participation of charge carriers in catalytic reactions. When combined with light irradiation and ultrasonic vibration, CoO_x/Bi₄Ti₃O₁₂ exhibited an exceptional piezo-photocatalytic degradation rate of 2.96 h⁻¹, which was approximately 2.7 and 1.8 times higher than that of photocatalysis and piezocatalysis alone, respectively. This research provides valuable insights into the development of highly efficient catalytic reactions based on composite heterojunction structures and external field coupling.

CRediT authorship contribution statement

Kaiqi Wang: Investigation, Formal analysis, Writing – original draft, Data curation. Ziying Guan: Validation, Investigation. Xiaoya Liang: Investigation. Shuyue Song: Investigation. Pengyu Lu: Validation. Chunran Zhao: Investigation. Lin Yue: Validation. Zhihao Zeng: Validation. Ying Wu: Conceptualization, Methodology, Resources. Yiming He: Conceptualization, Writing – review & editing, Supervision, Project administration.

Declaration of Competing 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.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

The authors do not have permission to share data.