N-Doped Modified MoS₂ for Piezoelectric–Photocatalytic Removal of Tetracycline: Simultaneous Improvement of Photocatalytic and Piezoelectric Properties

Abstract

Piezoelectric and photocatalytic technologies use mechanical and light energy to decompose environmental contaminants, demonstrating a beneficial synergistic impact. This investigation employs a two-step hydrothermal-calcination technique to synthesize N-doped MoS₂ photocatalytic materials. The ideal catalyst, N-MoS₂-3, utilizing the synergistic effect of piezoelectric–photocatalysis processes, attained a TC degradation rate of 90.8% in 60 min. The kinetic constant (0.0374 min⁻¹) is 1.75 times greater than the combined rates of single photocatalysis and piezoelectric catalysis, indicating a notable synergistic impact. The material has 80% degradation efficiency after five cycles, indicating its remarkable resilience. Mechanistic investigations reveal that nitrogen doping establishes an internal electric field by modulating the S-Mo-S charge distribution. Photogenerated electrons move to generate •O₂⁻, while holes accumulate internally. The ultrasound-induced piezoelectric polarization field interacts with the photogenerated electric field in reverse, thereby synergistically improving carrier separation efficiency and facilitating redox processes. This study emphasizes the viability of non-metal doping as a method for modifying the properties of two-dimensional materials, offering a novel approach to enhance the synergistic attributes of piezoelectric and photocatalytic processes. This technology possesses significant promise for environmental restoration through the utilization of solar and mechanical energy.

1. Introduction

Urbanization and population growth are increasing, resulting in greater discharge of antibiotics into the natural water environment via pharmaceutical wastewater [1,2,3]. Tetracycline antibiotics are widely utilized for the treatment of bacterial infections in both humans and animals, owing to their efficacy and cost-effectiveness [4,5,6,7]. Nonetheless, the residual tetracycline antibiotics present are non-biodegradable, persistent, and ecotoxic, making them threatening to life, health, and ecological equilibrium [8,9]. Consequently, it is essential to conduct quantitative analyses and eliminate antibiotic pollutants from natural aquatic ecosystems. Nonetheless, the total elimination of these antibiotics through conventional wastewater treatment facilities (WWTPs) continues to be a difficulty [10,11]. Consequently, there is an immediate necessity to devise effective and cost-efficient strategies for addressing TC contamination in aquatic ecosystems.

Although semiconductor photocatalysis is an effective method for addressing environmental pollution and energy shortages, its catalytic efficiency is significantly hindered by the rapid recombination of photogenerated electron–hole pairs [12]. Typically, a built-in electric field (BIEF) is formed by the spontaneous polarization of piezoelectric materials in response to external mechanical forces, regulating the transport direction of photogenerated carrier. In addition to solar energy, wind energy, tidal energy, acoustic energy, and other common mechanical force energy sources are classified as green energy [13]. The piezoelectric–photocatalytic synergistic system may efficiently integrate green mechanical energies with solar energy, hence offering a novel technical avenue for wastewater degradation [14,15].

As a two-dimensional material, molybdenum disulfide (MoS₂) has garnered a lot of interest because of its exceptional catalytic activity, adjustable microstructure, and affordability [16,17]. MoS₂ nanosheets consist of a layer of molybdenum atoms, which are connected to two layers of sulfur atoms through covalent bonds [18,19]. Under tensile stress, the molybdenum disulfide nanosheets with asymmetric centers generate a large piezoelectric potential and a large number of polarized charges, which can effectively degrade organic molecules [20,21,22]. Moreover, its narrow bandgap allows it to absorb visible light and generate electron–hole pairs when exposed to light. This unique structure and these excellent physicochemical properties have led to its widespread application in the field of catalysis. Nevertheless, the catalytic activity of MoS₂ is still limited, and the surface migration of excited charges remains suboptimal, requiring further enhancement. Therefore, in order to improve its photocatalytic ability and piezoelectric properties, modification of MoS₂ is necessary.

Non-metallic doping (e.g., sulfur, oxygen, phosphorus, nitrogen, etc.) can alter the band gap of MoS₂ and is an effective way of improving carrier separation efficiency [23,24,25,26,27]. It demonstrates that non-metallic doping can influence the lattice structure and enhance the material’s piezoelectric coefficient (d₃₃) [28,29,30,31]. Up to now, there have been fewer publications on the improvement of piezoelectric–catalytic performance through nitrogen doping. Consequently, it is scientifically important to examine the relation between nitrogen doping and piezoelectric–catalytic efficiency.

This study employs a two-step hydrothermal–calcination method to prepare N-doped MoS₂ piezoelectric–photocatalytic materials, and enhances piezoelectric and photocatalytic functions through structural and electronic modulation. Under the conditions of piezoelectric–photocatalysis coupling, efficient tetracycline degradation was achieved (90.8% degradation within 60 min). Subsequently, material characterization confirmed the enhanced electron–hole separation and charge transfer properties in the N-MoS₂-3 material. Mechanism analysis indicates that the synergistic effect arises from the built-in electric field induced by charge redistribution, which, together with the piezoelectric field, drives the separation of charge carriers. This work provides an effective solution for further developing piezoelectric–photocatalytic technology, promoting light-induced electron–hole separation, and simultaneously utilizing mechanical and light energy.

2. Experimental Section

2.1. Materials

All chemicals used were of analytical grade and employed without further purification (Supplementary Text S1).

2.2. Synthesis of Piezo-Photocatalysts

MoS₂ nanohybrid was prepared using a facile hydrothermal method. Initially, 1.2 g of Na₂MoO₄·2H₂O and 0.76 g of CH₄N₂S were dissolved in 60 mL of 10% (w/w) ethanol solution under constant stirring for 30 min. The mixed solution was placed in a 100 mL Teflon autoclave and heated to 220 °C for 24 h. After cooling to ambient temperature, the black product was separated using centrifugation and then cleaned three times using anhydrous ethanol and deionized water. The obtained samples were dried under vacuum for 12 h at 80 °C.

Certain weights (0.2, 0.4, 1.2, and 2.0 g) of urea and 0.4 g of MoS₂ samples were mixed well and loaded into a porcelain boat. Subsequently, the porcelain boat was placed in a tube furnace and annealed at 900 °C for 120 min in a nitrogen atmosphere. The hierarchical structures of N-doped MoS₂ prepared with 0.2, 0.4, 1.2, and 2.0 g of urea were labeled as N-MoS₂-0.5, N-MoS₂-1, N-MoS₂-3, and N-MoS₂-5, respectively.

2.3. Characterization and Analysis Methods

The X-ray diffractometer (XRD, SmartLab SE, Tokyo, Japan) was used to examine the crystalline phase of MoS₂ with varying N-doped concentrations. Raman spectroscopy was performed using a Horiba HR800 spectrometer(Horiba Scientific, Kyoto, Japan). An integrating sphere and a UV-Vis-NIR spectrophotometer (Shimadzu UV-3600 plus, Kyoto, Japan) were used to record the UV-Vis diffuse reflectance spectra (UV-Vis DRS). The morphology of the samples was examined using a field emission scanning electron microscope (S4800, Hitachi, Tokyo, Japan) and a field emission transmission electron microscope (TEM, FEI-Talos F200S, Hillsboro, OR, USA). X-ray photoelectron spectroscopy (XPS) was used to analyze the materials’ surface properties using a Thermofisher ESCALAB Xi+ device (Thermo Fisher Scientific, Waltham, MA, USA) and a single Al-Kα radiation source (6 mA, 12 kV). The PFM (piezoresponse force microscopy) and KPFM (Kelvin probe force microscopy) modules in an AFM (atomic force microscopy) test system (Bruker Dimension lcon, Billerica, MA, USA) were utilized to monitor the piezoelectric response of the synthesized samples. Electron paramagnetic resonance (EPR) spectroscopy was performed to detect free radical production using an Elexsys E560 system (Bruker, Billerica, MA, USA) with DMPO or TEMP as spin traps.

Additionally, the materials were subjected to photoelectrochemical studies utilizing an electrochemical workstation (PGSTAT204, Metrohm, Herisau, Switzerland). Details of the tests are shown in the Supplementary Text S2.

2.4. Evaluation of the Synergistic Degradation Performance

All experiments were performed in a 200 mL beaker. As the reaction system, 100 mL of reaction solution was used. The beakers were placed in an ultrasonic reactor (CR-070ST, Shenzhen, China) to provide periodic local mechanical strain with an ultrasonic frequency of 40 kHz and a set output power of 480 W. Typically, the degradation phase was preceded by an adsorption–desorption equilibrium phase, where 20 mg of piezoelectric photocatalyst was added to 10 mg/L TC solution for 30 min in the dark. The degradation phase was initiated under different conditions, including visible-light irradiation, ultrasonic action, and visible-light–ultrasonic action. Then, 2 mL portions of solution samples were taken at pre-arranged intervals and filtered through a 0.45 µm filter membrane. The residual concentration of TC was studied and analyzed by measuring the absorbance value of the solution at λ_max = 357 nm using the UV spectrophotometer.

The degradation efficiency was calculated by Equation (1):

$$\mathsf{\eta }(\mathrm{\%})=(1-{\displaystyle \frac{{\mathrm{C}}_{\mathrm{t}}}{{\mathrm{C}}_0}})\times 100$$

(1)

where C₀ and C_t (mg/L) represent the initial concentration of TC and the concentration at a specific time, respectively.

To evaluate the stability and reusability of the catalyst, the catalyst after the catalytic degradation of TC was separated from the solution by centrifugation (10,000 rpm, 5 min), washed multiple times with deionized water and anhydrous ethanol, and then dried in a vacuum oven at 60 °C. The dried catalyst underwent five cyclic experiments under the same reaction conditions.

3. Results and Discussion

3.1. Characterizations of Catalysts

The preparation of MoS₂ and N-MoS₂-x catalysts is shown in Figure 1a. The morphology of bulk MoS₂ and N-MoS₂-x was characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). As shown in Figure 1b–e, the surface of the N-doped samples revealed nanoflake-like aggregation structures identical to pristine MoS₂, suggesting that nitrogen incorporation does not alter the surface morphology (Supplementary Figures S1 and S2, and Table S1) [18]. Moreover, it has been demonstrated that the surfaces of the doped samples harbor minute nanoparticles, indicating the successful incorporation of nitrogen into the MoS₂ lattice. Furthermore, the elemental mapping obtained from scanning electron microscopy reveals that the elements N, Mo, and S are equally distributed across the surface of the nanosheets in the catalyst N-MoS₂-3, as illustrated in Figure 1f–h. The microscopic morphology and lattice stripes of N-MoS₂-x were analyzed using TEM (Figure 1i,j and Supplementary Figures S3–S5) [32]. The TEM images of N-MoS₂-x reveal orderly lattice stripes on the material’s surface.

The lattice spacings align with the reflecting surface (002) of 2H-MoS₂ (JCPDS 37-1492) [33], and a minor increase in the layer spacing of N-MoS₂-3 is also noted. The effective doping of the N element is further evidenced [34].

Raman spectroscopy was used to probe the chemical structure of the prepared catalysts. As shown in Figure 2a, in the MoS₂ spectrum, two distinct peaks were detected at 374.3 and 400.8 cm⁻¹, which were related to the in-plane $E_{2g}^1$ vibrational modes (Mo-S-Mo) and the out-of-plane A_1g vibrational modes (perpendicular to Mo-S-Mo) of 2H-MoS₂, respectively, and no other distinctive characteristic peaks were detected in the spectra, which indicated that N doping did not modify the MoS₂ layered structure [35]. The vibrational peaks of N-MoS₂-3 exhibit a minor blue shift in position, with a frequency difference of around 26.5 cm⁻¹ between the two peaks. This mostly results from the localized stress and alterations in the electrical structure induced by nitrogen doping.

The crystal structure of the synthesized catalysts was analyzed using an X-ray powder diffractometer (XRD). Figure 2b illustrates a succession of distinctive diffraction peaks at 14.4° (002), 29.0° (004), 32.7° (100), 35.9° (102), 39.5° (103), and 58.3° (110), confirming the effective synthesis of hexagonal 2H-MoS₂ (JCPDS 37-1492) [36]. It is important to highlight that, despite the successful doping of the N element into MoS₂, no diffraction peaks corresponding to Mo-N species were detected in the XRD spectra, suggesting that the N doping did not influence the crystal structure of MoS₂. Furthermore, the half-peak width of the characteristic peak (002) of N-doped MoS₂ is obviously reduced, while the peak (102) appears at 35.87°, indicating that the incorporation of N atoms markedly enhances its crystallinity.

X-ray photoelectron spectroscopy (XPS) was used to examine the surface chemical compositions and atomic valence states of MoS₂ and N-MoS₂-x. The full-scan spectra show that the nitrogen doping was successful since the surface of nitrogen-doped MoS₂ has peaks of the N element in addition to the Mo, S, and O elements (Figure 2c). Gas molecules like CO₂ or O₂ adsorbed on the material’s surface could be the source of the O peak among them. The S 2p_1/2 and S 2p_3/2 peaks from S (II) are displayed in Figure 2d. Figure 2e displays the Mo 3d_3/2 and Mo 3d_5/2 peaks from Mo (IV). Furthermore, a specific peak at 235.3 eV associated with Mo (VI) has a very weak signal. Thus, the valence state of Mo remains predominantly Mo (IV). In Figure 2f, a Mo-N peak appears, which further confirms the successful doping of N element in MoS₂. It is evident from Figure 2d–f that charge transfer has taken place in nitrogen-doped MoS₂ since the elemental binding energies Mo 3p, Mo 3d, S 2s, and S 2p in N-MoS₂-x are negatively shifted in comparison to those of MoS₂. These results can be attributed to the doping of element N into the MoS₂ lattice in the form of substituting S to form N-Mo bonds and triggering energy band bending. Moreover, the percentage of N in the optimal catalyst N-MoS₂-3 was about 18.39 at% (Supplementary Table S2).

3.2. Photocatalytic Activity

The effectiveness of a photocatalytic reaction is contingent upon the light-harvesting capacity of the photocatalyst, which exerts a direct influence on the number of charge carriers produced and the efficiency with which natural sunlight is utilized [37,38]. Consequently, the optical characteristics of the catalysts were evaluated using UV-Vis DRS spectroscopy (Figure 3a). The black surface contributes to the increase in the molar adsorption coefficient, thereby exhibiting strong light-harvesting properties over a wide range of 200–800 nm. Compared with pure MoS₂, N-MoS₂-x has enhanced light-harvesting performance in the visible range (400–700 nm), and the enhanced visible-light adsorption ability can improve the light utilization and generate more effective photogenerated electrons to participate in the photocatalytic reaction. In addition, the energy band gap (E_g) of the nitrogen-doped molybdenum disulfide was narrowed from 2.27 eV to 1.95~2.13 eV according to the classical Tauc diagram (Supplementary Equations (S1) and (S2)). In addition, the positions of the conduction (E_CB) and valence bands (E_VB) were determined according to VB-XPS and Mott–Schottky. According to the VB-XPS results, the E_VB values of MoS₂, N-MoS₂-0.5, N-MoS₂-1, N-MoS₂-3, and N-MoS₂-5 are 0.55 eV, 1.09 eV, 1.00 eV, 0.61 eV, and 0.9 eV, respectively. The linear regions of the Mott–Schottky curves are all positively sloped, showing typical n-type semiconductors characteristics. The flat band potentials’ E_fb values are −1.51 V, −1.47 V, −1.4 V, −1.48 V, and −1.27 V (relative to Ag/AgCl). This corresponds to −1.31VNHE, −1.27VNHE, −1.2VNHE, −1.28VNHE, and −1.07VNHE according to Supplementary Equation (S3). In general, for n-type semiconductors, the value of E_fb is about 0.1V higher than the value of E_CB [39]. Therefore, the values of E_fb are about 0.1V higher for MoS₂, N-MoS₂-0.5, N-MoS₂-1, N-MoS₂-3, and N-MoS₂-5, and the E_CB values are estimated to be −1.41 VNHE, −1.37 VNHE, −1.3 VNHE, −1.38 VNHE, and −1.17 VNHE, respectively. The energy band positions are illustrated in Figure 3c.

The charge transfer dynamics at the piezoelectric photocatalyst/solution interface were examined using photoelectrochemical testing [40]. The capability of photoelectric conversion was evaluated utilizing the transient photocurrent density (J). All transient photocurrent density profiles exhibited a quick increase following the illumination for five cycles. The stable J_photo values gradually increased with the increase in N doping, and the J_photo values of MoS₂, N-MoS₂-0.5, N-MoS₂-1, N-MoS₂-3, and N-MoS₂-5 were 0.015 μA·cm⁻², 0.058 μA·cm⁻², 0.066 μA·cm⁻², 0.075 μA·cm⁻², and 0.171 μA·cm⁻² (Figure 4a). Notably, the optimized N-MoS₂-5 sample achieved a remarkable 11.4-fold improvement in photocurrent generation compared to the undoped MoS₂ counterpart. The high and stable J_photo values indicate that the samples may have higher photogenerated carrier separation efficiency and thus exhibit a higher photocurrent response. In this work, the photoelectrochemical stability was evaluated by the photocurrent density retention ratio (PDRE), as shown in Equation (2):

$$PDRE={\displaystyle \frac{i_5}{i_1}}\times 100\%$$

(2)

where PDRE is the photocurrent density retention efficiency; and i₁ and i₅ are the stabilized photocurrent densities at the 1st and 5th cycles, respectively. The calculated efficiencies of all piezoelectric photocatalysts’ PDREs exceeded 80%. Especially for N-MoS₂-1, no obvious photocurrent density attenuation was found. All piezoelectric photocatalysts exhibited satisfactory photoelectrochemical stability during five cycles.

In general, a decrease in the diameter of the capacitive arc represents a decrease in the charge transfer resistance (R_ct), which exhibits a relatively strong electron transport capability [38,41]. The Nyquist curves demonstrate incomplete capacitive arc shapes under both dark and visible-light illumination when fitted using the [R (RQ)([RW] Q)] equivalent circuit model. Under dark conditions, the corresponding capacitive arc radii are in the order of MoS₂ (27.9 Ω) > N-MoS₂-0.5 (25.9 Ω) > N-MoS₂-3 (22.5 Ω) > N-MoS₂-1 (19.9 Ω) = N-MoS₂-5 (19.9 Ω). In comparison with the primitive MoS₂ samples, the arc radius underwent a significant decrease following nitrogen doping, indicating that N doping can effectively enhance the conductivity and reduce the charge transfer resistance of the material (Figure S8). Under visible-light irradiation, the Nyquist curves of all samples displayed a shape akin to those observed in darkness (Figure 4b); however, the R_ct values experienced a further decrease in the dark, indicating that photogenerated electrons can be produced and migrate to the electrode/electrolyte interface under visible-light exposure. A quantitative analysis of the built-in electric field (BIEF) was conducted to elucidate its role as the driving force for the separation of photogenerated electron–hole pairs. As anticipated, the BIEF intensity was found to be enhanced under visible-light irradiation in comparison to dark conditions (Figure 4c and Figure S10).

In summary, nitrogen-doped molybdenum disulfide has enhanced visible-light absorption, hence augmenting its piezoelectric–photocatalytic activity.

3.3. Piezoelectric Performances

Generally, the charge centers of cations and anions become asymmetric under external stress applied to the surface of a piezoelectric material, thereby inducing an internal electric field that promotes the separation of photogenerated charge carriers [42]. The piezoelectric characteristics of MoS₂ and N-MoS₂-3 were confirmed using atomic force microscopy (AFM), employing Kelvin probe force microscopy (KPFM) and piezoresponse force microscopy (PFM) techniques [43]. These catalysts exhibited a distinct disparity in their piezoresponse amplitudes. The topographic pictures of MoS₂ and N-MoS₂-3 samples in PFM measurements distinctly reveal their surface features (Figure 5a,e and Supplementary Figure S12), corroborating the HRTEM results. Simultaneously, the phase diagram in Figure 5c,g, characterized by its congruent topography, demonstrates the homogeneous distribution of polarization within the nanostructures.

As shown in Figure 5b,f, the surface potential of N-MoS₂-3 reaches a maximum value of 157.7 mV, which is 2.94 times higher than that of MoS₂ (53.7 mV). When scanned by applying a bias voltage of −10 to +10 V, the piezoelectric response amplitudes of the samples show a mirror symmetry similarity (Figure 5d,h) [44]. The good piezoelectric performance is further confirmed by the butterfly curve. And the maximum effective piezoelectric coefficient (d₃₃) was calculated from the slope of the butterfly curve (Supplementary Figure S14). The results show that the d₃₃ of N-MoS₂-3 (183 pC V⁻¹) is 1.58 times higher than that of MoS₂. N doping considerably improves the piezoelectric characteristics of MoS₂, according to a comparison of the surface potentials and piezoelectric response amplitudes measured in the PFM. This enhancement of piezoelectric properties can be attributed to the effect of N doping on the crystal structure and electrical conductivity of MoS₂. Regarding the KPFM test, in contrast to the weak BIEF (28.687 mV) produced by MoS₂ under probe tip stress, N-MoS₂-3 produces a positive surface voltage of up to 59.357 mV, showing its excellent piezoelectric properties.

In addition, to further verify the piezoelectric effect. The transient piezoelectric current response maps were investigated at varying ultrasonic power levels (120 to 480 W). Supplementary Figure S15 and Figure 6a illustrate an effective positive connection between J_piezo and ultrasonic power, suggesting that the synthesized piezoelectric catalysts exhibit significant piezoelectric sensitivity. Moreover, the intensity of the piezoelectric-induced BIEF was markedly elevated to 0.2015~0.2418 mV compared to the intensity of the photoelectric-induced BIEF (Figure 6a). The findings indicate that the piezoelectric properties of N-doped MoS₂ enhance electron–hole separation efficiency.

We also analyzed the transient piezoelectric current response maps during the synergistic interaction of piezoelectricity with photovoltaics (Figure 6b and Supplementary Figures S16 and S17). The results indicated that the J_piezo-photo of all materials was elevated in comparison to J_photo and J_piezo across all piezoelectric photocatalysts. Simultaneously, the BIEF intensity within the synergistic system increased to 0.2198–0.2454 mV (Supplementary Figure S18), indicating a significant enhancement in the charge transfer mechanism at the piezoelectric photocatalyst/solution interface.

The preceding studies demonstrate that N-MoS₂-3 can effectively promote the redox process on the surface by generating additional polarization carriers under the effect of applied stress. This effect is accompanied by the material’s excellent piezoelectric properties.

3.4. Performance Evaluation of Piezoelectric–Photocatalytic Degradation of Tetracycline

The significant current response generated under paired ultrasonic and visible-light irradiation indicates excellent piezoelectric–photocatalytic activity, as mentioned above. The photocatalytic and piezoelectric properties of pure MoS₂ and N-MoS₂-x samples were assessed by the degradation efficacy of TC. Before the degradation studies, the catalyst samples were introduced to the aqueous TC solution for 30 min in the absence of light to establish adsorption–desorption equilibrium. The adsorption of tetracycline by N-doped MoS₂ was significantly greater than that of pristine MoS₂, with N-MoS₂-1 exhibiting the highest adsorption, at 2.92 times that of MoS₂. This finding indicates that nitrogen doping may enhance adsorption capacity by increasing surface active sites or optimizing pore structure, thereby indirectly suggesting an improvement in specific surface area.

Figure 7a illustrates that, under visible-light irradiation, 20.0% of pristine MoS₂ was eliminated within 60 min, while the degradation efficiency of N-MoS₂-5 increased to 48.4%, with the kinetic rate constant rising by a factor of 4.44 following nitrogen doping. This outcome confirms that the localized surface plasmon resonance (LSPR) effect, facilitated by an optimal nitrogen modification, enhances the visible-light absorption of MoS₂, thereby augmenting the photocatalytic degradation efficiency of TC. The degradation performance of N-MoS₂-x was significantly improved for piezoelectric–photocatalysis. N-MoS₂-3 exhibited limited photocatalytic (32.2%) and piezoelectric–catalytic (70.5%) activity for TC removal. In contrast, 90.8% of TC was eliminated in 60 min when piezoelectricity and visible light were combined, which was significantly better than other doping ratio materials (Figure 7c).

The kinetic rate constant (Supplementary Figure S20) of N-MoS₂-3 under piezoelectric–photocatalytic coupling is 3.07 times higher than that of MoS₂. Notably, this rate constant exceeds the combined photocatalytic and piezocatalytic rate constants by a factor of 1.75. This indicates that the piezoelectric field generated by ultrasound facilitates the separation and migration of photogenerated electron–hole pairs during the catalytic process. The overall degradation efficiency was markedly improved due to the increased participation of holes and electrons in the piezoelectric–photocatalytic interaction [45].

In order to better understand the TC degradation process, the effects of various operating parameters in the piezoelectric–photocatalytic system were investigated. These parameters included ultrasonic parameters, TC concentration, and initial pH.

Previous research indicates that ultrasonic frequency and power significantly influence piezocatalytic degradation activity. The suitable ultrasonic frequency can optimize the deformation of piezoelectric–catalytic materials by inducing resonance under ultrasonic influence, while augmenting power input can enhance the degree of deformation of the materials. The impact of mechanical force intensity on the efficacy of the catalyst for TC degradation was examined (Supplementary Figure S21a). As the ultrasonic power escalates, the computed k_obs exhibits a corresponding increase. Relative to k_obs at 120 W, k_obs at 480 W exhibited a 77% increase. This suggests that increased ultrasonic power produces greater applied stress, leading to a more robust piezoelectric potential, which facilitates the generation of ROS, hence improving the catalytic efficacy of piezoelectric–photocatalysis.

The degradation rate of TC by N-MoS₂-3 diminished progressively with the increase in starting concentration, yielding rates of 96.0%, 91.7%, 90.8%, 70.8%, and 51.1%, respectively. The migration of carriers to the active site of the photocatalyst was influenced by the abbreviated transport pathway and the transmittance of photogenerated carriers at elevated pollutant concentrations, which concurrently diminished the photocatalytic efficiency of the catalyst [46]. The photocatalytic activity diminished due to intermediate intermediates in the TC photocatalytic degradation process competing with TC molecules for the limited photocatalytic active sites. Although the elevated concentration diminished photocatalytic activity, N-MoS₂-3 demonstrated remarkable photocatalytic efficacy for tetracycline elimination at low pollutant concentrations, achieving a removal rate of 96.0% for 2 mg/L of pollutant. Considering that pollutant concentrations are typically low in actual wastewater, N-MoS₂-3 can be regarded as an efficient photocatalyst for the degradation of the antibiotic TC.

Another crucial element influencing the production of reactive oxygen species (ROS) and photocatalytic degradation in actual wastewater is the pH of the reaction solution. Supplementary Figure S21c illustrates the degradation efficiencies of TC at varying pH levels (3, 5, 7, 9, and 11), which are 84.40%, 92.08%, 70.69%, 60.40%, and 54.06%, respectively. The degradation efficiency is maximized at pH 5, where the reaction rate constant (k_obs = 0.0377 min⁻¹) is likewise the highest. The subsequent pH levels are 3 and 7, exhibiting reaction rate constants of 0.0281 min⁻¹ and 0.0185 min⁻¹, respectively. Under acidic conditions, TC mostly exists in its cationic form [47]. Excess H+ not only depletes surface hydroxyl groups via protonation, so hindering •OH production, but also facilitates the recombination of photogenerated electrons with H+, resulting in diminished TC degradation efficiency. In weakly acidic and neutral conditions, TC is more readily adsorbed onto the catalyst surface, hence improving degrading efficiency. In alkaline conditions, a significant concentration of OH- competes with TC for active sites, diminishing reaction efficiency; concurrently, the surplus OH- facilitates the reaction between H₂O and H+, resulting in a reduction of the primary active species H+ in the photocatalytic process, thereby impairing overall photocatalytic degradation performance.

The stability of N-MoS₂-3 was examined using cycle testing, with the findings presented in Figure 7e. The degradation efficiency of TC exceeded 80% after five cycles, consequently affirming the strong stability and excellent reusability of the N-MoS₂-3 piezoelectric photocatalyst. The crystalline phase and structure of N-MoS₂-3 were preserved during the piezoelectric–photocatalytic degradation of TC under visible light and ultrasound, as evidenced by XRD and SEM analyses conducted before and after recycling.

XRD and SEM measurements performed before and upon recycling demonstrated that the crystalline phase and structure of N-MoS₂-3 remained unchanged during the piezoelectric–photocatalytic breakdown of TC under visible light and ultrasound (Supplementary Figures S22 and S23). This result further confirms the exceptional long-term stability of the catalyst.

To investigate the efficacy of piezoelectric photocatalysts, we performed TC degradation studies in different real water sources (ultrapure water, laboratory tap water, and runoff rainwater) to which TC was introduced (Figure 7f). The piezoelectric–photocatalytic degradation performance of N-MoS₂-3 exhibited variability across diverse water samples, indicating that the composition of these samples influences its catalytic activity. The TC degradation efficacy in the other two water samples was diminished relative to ultrapure water; however, the catalytic activity remained adequate. The runoff precipitation exhibited the lowest degradation efficiency of 72.13%. The reduction in degrading efficiency mostly stemmed from the synergistic influence of intricate constituents in runoff rainfall. Humic acids and suspended particles decreased the efficient use of light energy due to light attenuation, whereas concurrent organic matter competed with TC molecules for reactive oxygen species. Moreover, inorganic salt ions like Cl⁻ further impeded the degradation cycle by quenching free radicals. The experimental results demonstrated that the system not only effectively degraded the target pollutants but also had the capacity to non-selectively degrade additional organic contaminants in the actual water environment.

3.5. Piezocatalytic Mechanism

In Supplementary Figure S24, in order to gain more insight into the degradation mechanism, we performed a series of gradient trapping experiments of free radicals using different trapping agents, thus identifying the species of active species and estimating their respective contributions in the system. The addition of ascorbic acid (AA, detecting •O₂⁻), disodium ethylenediaminetetraacetic acid (EDTA-2Na, detecting h+), tertiary butyl alcohol (TBA, detecting •OH), and furfuryl alcohol (FFA, detecting ¹O₂) resulted in a certain degree of decrease in the degradation efficiency, suggesting that all of the h+, ¹O₂, •O₂⁻, and •OH reactive oxide species were involved in the degradation.

The active oxide species in MoS₂ and N-MoS₂-3 were further identified by EPR testing, as shown in Figure 8a–f [48,49]. Hydroxyl radicals (•OH) were identified utilizing 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) as the trapping agent and deionized water as the solvent. No •OH signals were observed for MoS₂ under any conditions; however, N-MoS₂-3 exhibited characteristic •OH peaks during both sonication and photo-sonication, with no •OH signals found under visible-light irradiation (Figure 8a,d). The presence of •O₂⁻ was identified utilizing DMPO as a trapping agent and methanol as a solvent. In the case of MoS₂, characteristic •O₂⁻ peaks were observed under both visible-light irradiation and photo-ultrasound; however, no •O₂⁻ signal was recorded under ultrasound. The same results were obtained for N-MoS₂-3 (Figure 8d,e). Conversely, ¹O₂ was seen under all conditions.

The findings indicated that •O₂⁻ could be produced during visible-light exposure, but ultrasonic action facilitated the synthesis of ¹O₂, enhancing the degradation capacity. In contrast to MoS₂, N-MoS₂-3 produced OH upon ultrasonic stimulation, and the strength of the distinctive peaks markedly enhanced under light–ultrasonic stimulation. Simultaneously, the concentrations of all active species in N-MoS₂-3 were above those in MoS₂ under all conditions, signifying that nitrogen-doped MoS₂ augmented the generation of active species induced by visible-light and ultrasonic stimulation.

Based on the above analysis, we propose a synergistic degradation mechanism for piezoelectric–photocatalytic degradation of MoS₂ regulated by N doping (Figure 9). In the visible photocatalytic process, nitrogen doping altered the charge distribution of the S_i-Mo_i-S_d structure, enhanced the charge transfer kinetics, expedited the charge transfer from Mo_i to S_i, and established a built-in electric field from the positive surface to the interior under visible-light irradiation. The BIEF facilitated the separation of photogenerated electrons and holes, enabling the photogenerated electrons on the surface of the N-doped MoS₂ nanosheets to react with O₂ in solution, thereby producing the active species •O₂⁻. The breakdown of organic contaminants is then achieved through a reaction with TC. Consequently, h+ could not interact with H₂O or -OH to produce •OH, yet free radical trapping and EPR experiments indicated the presence of •OH in the system, leading to the hypothesis that •OH was created by the reaction between •O₂⁻ and H₂O at the CB location. It is posited that •OH is produced through the reaction of •O₂⁻ and H₂O at the CB location. In piezoelectric catalysis, photogenerated carriers are typically transported in the direction opposite to the electric field, resulting in a gradual decline in potential and an increase in their energy. This phenomenon enhances the carriers’ capacity to evade surface defects and to surpass the overpotential necessary for engaging in redox reactions. In contrast to the sole photocatalytic degradation of organic matter, the piezoelectric–photocatalytic process utilizes ultrasound as an applied mechanical force to activate the piezoelectric–catalytic effect of the material N-MoS₂. The synergistic coupling mechanism between this physical field and the photoexcitation process markedly enhances the separation efficiency of the carriers.

4. Conclusions

In this study, N-doped modified MoS₂ nanosheets were prepared for enhancing the degradation performance of piezoelectric–photocatalytic synergistic system. The experimental results showed that the N-MoS₂-3 sample achieved 90.8% (60 min) degradation efficiency of TC under piezoelectric–photocatalytic synergism, and its reaction kinetic constant was enhanced by 3.66 times compared with that of pristine MoS₂. It also exhibits good morphological and chemical structure stability, and maintains a good catalytic performance after several cycling experiments. In the study of the photocatalytic degradation mechanism of TC, we analyzed four main active species: h+, ¹O₂, •O₂⁻, and ·OH active oxide species are involved in the degradation. This study can provide a case reference for the element-doped modified MoS₂ piezoelectric–photocatalytic removal of new pollutants in water.