Rational Synthesis of a Dual Z-Scheme CdS/Ag₂MoO₄/β-Bi₂O₃ Heterojunction for the Deep Photodegradation of Methylene Blue and Analysis of Its Mechanisms

Abstract

In this work, a novel dual Z-scheme CdS/Ag₂MoO₄/β-Bi₂O₃ (CAB) composite heterojunction was synthesized, with the ultrafine CdS nanoparticles decorating two different-sized particles. In the beginning, the synergistic effect between BO and AMO makes the 10% Ag₂MoO₄/β-Bi₂O₃ (10AB) photocatalyst exhibit an optimal degradation efficiency of 87.1% for methylene blue (MB) of 10 mg·L⁻¹ within 60 min; furthermore, its photocatalytic activity was enhanced by incorporating CdS nanoparticles on the surface of the AB heterojunction. The results showed that the 25% CdS/10% AMO/BO (25C10AB) composite achieved a maximum MB degradation efficiency of 99%. Optical and photoluminescence measurements showed that the dual Z-scheme CAB heterojunction has high crystallinity and efficient charge carrier migration and separation, which makes the samples more efficient for removing pollutants. Theoretical studies (DFT/FEM calculations) were performed to better understand the migration direction of e⁻ and h⁺ in the photocatalytic degradation mechanism. This work provides a feasible approach to obtaining an efficient heterojunction composite photodegradation catalyst.

1. Introduction

The rapid advancement of modern society has led to significant environmental degradation caused by chemical, pharmaceutical, and other industries. Lakes in our country serve as invaluable sources of freshwater and play a crucial role in promoting green, sustainable, and healthy developmental processes [1]. However, many industries discharge significant amounts of contaminated water containing organic and harmful substances. Common organic dyes, such as methyl orange (MO), methylene blue (MB), and rhodamine B (RhB), are often released directly into the environment without adequate treatment, posing risks to both public health and ecological systems. Developing efficient methods for removing these pollutants from water to create a cleaner environment is a major challenge facing humanity [2,3]. Fujisima and Honda’s discovery that TiO₂ electrodes are capable of splitting water in the presence of light has catalyzed significant interest in semiconductor-based, light-driven degradation processes [4]. Consequently, photocatalysis is recognized for its considerable potential in environmental purification and energy development. In recent years, traditional photocatalysts such as CeO₂ [5], BiVO₄ [6], and TiO₂ [7] have increasingly been supplanted by multicomponent composite photocatalysts. This shift is due to the limitations of the former materials, including their exclusive responsiveness to ultraviolet (UV) light, as well as their low charge mobility and separation efficiency [8,9,10,11]. The advantages of semiconductor photocatalytic technology, including its operation at room temperature, the utilization of available sunlight, a broad range of catalyst sources, and the ideal removal of dye pollutants have attracted considerable attention from researchers [12,13,14,15,16]. Currently, extensively studied semiconductor photocatalysts include Bi₂O₂CO₃, CsSnI₃, Ag₂S and TiO₂, etc.; the performance of these materials will be highly achieved by the modification strategy [17,18,19,20].

Bismuth-based compounds are recognized for their superior oxidation performance, making them promising candidates for the degradation of organic pollutants [21,22,23,24]. BO serves as a photocatalyst commonly employed to degrade hazardous substances, including AO7 and RhB [25,26]. However, the utilization of BO is significantly hindered by its wide band gap, limited efficiency under visible light, and poor charge carrier transfer and separation capabilities [27,28,29]. Recent research has focused on constructing composite structures to address these limitations by enhancing the separation of photogenerated electron-hole pairs [30,31,32]. One prevalent approach involves doping BO with other elements to reduce its band gap, improving photocatalytic activity [33]. However, this method is constrained since element substitution is typically only feasible between elements with similar atomic radii, which limits the potential for enhancing the photocatalytic properties. Previous studies have demonstrated that combining BO with other semiconductor heterojunction photocatalysts can significantly enhance the degradation rate of organic pollutants by improving electron transfer rates and reducing carrier recombination [34]. Moreover, heterojunction materials are well-known for their superior photocatalytic performance due to high charge carrier migration and separation efficiencies [35]. Metal molybdates are getting more and more attention at present. Their strong atomic and structural stability and hypotoxicity make them advantageous for applications of the usage of solar energy and storage. AMO is particularly popular in photocatalytic applications due to its unique electronic structure, high conductivity, excellent optical properties [36], and remarkable chemical stability [37,38]. Advancements in synthetic methods have led to the production of morphologically diverse AMO samples, which have been utilized in various photochemical reactions [39,40]. CdS is also receiving increasing attention. However, the instability of charge carriers in pristine CdS limits its widespread application. Thus, building heterojunctions with other semiconductor materials has emerged as a promising strategy to enhance the stability and activity of CdS [41,42,43].

This study presents the development of a novel type of CdS/AMO/BO (CAB) heterojunction composite photocatalyst and investigates the mechanisms between its photoelectrochemical and photocatalytic performance. The dual Z-scheme junction has unique advantages in enhancing the redox properties, whose separation efficiency of carriers is more excellent than other types [35]. Theoretical calculations also indicate that CABs have suitable work functions as well as conduction band and valence band (CB/VB) potentials that can be constructed into the dual Z-scheme CAB heterojunctions, which will be beneficial for the complete degradation of MB. This material, which is zero-dimensional (0D)/three-dimensional (3D)/three-dimensional (3D), has the advantage of having numerous surface-active sites for creating heterojunctions [44]. The phase purity, geometry of sample surfaces, functional group, PL lifetime, charge transfer, elemental species, and separation efficiency of CAB heterojunction photocatalysts were studied. The impact of catalyst levels, dye concentrations, inorganic cations, and inorganic anions on the performance of CAB composites was also evaluated. Finally, a potential photocatalytic mechanism was proposed. This work provides us with an important means of producing superior composite materials with excellent photocatalytic performance.

2. Results and Discussion

2.1. XRD and FTIR Analysis

Figure 1 presents the XRD patterns of xAB (x = 0, 1, 3, 7, 10, 15, and 100) to determine the crystal structures and purity of the samples. The XRD analysis shows the high crystallinity of the samples, as shown in Figure 1a,b. The comparison with the standard XRD patterns of BO (PDF#78–1793) and AMO (PDF#08–0473) indicates that BO crystallizes in a tetragonal structure, while AMO is in a cubic phase. The diffraction peaks of xAB materials align with the tetragonal structure of BO and the cubic structure of AMO, with all AMO peaks visible in the composites, indicating that the composite was successfully formed. The lattice parameters of the AMO and BO composites are almost unchanged. The XRD patterns of CdS, xAB, and yCAB were also observed under similar conditions alongside the standard PDF cards of the XRD of CdS (PDF#80–0019).

The XRD peaks of the pure BO, AMO, and CdS samples were very consistent with the PDF cards. The acuity of peaks in the BO patterns intensely suggests the crystallinity of the nanocomposites, with peaks corresponding to specific crystal planes of the tetragonal BO phase. The most evident peak at 2θ = 27.94° complies with the (201) crystal plane of the tetragonal BO phase. Other peaks obtained at 2θ values of 32.69, 46.2, and 55.49° are also in agreement with the (220), (222), and (421) planes of the BO structure. Furthermore, the extra strong peaks obtained at 31.84, 33.29, 50.91, and 55.8° correspond to the (311), (222), (511), and (440) diffraction planes of the AMO structure. Similar peaks are observed for the AMO crystal structure. The mixture of BO and AMO formed heterojunctions at the interface, evident from the XRD results at 100 °C. In addition, the other AB samples exhibited the characteristic peaks of the AMO sample. Meanwhile, the crystallinity of CdS is poor, therefore the peak is wide. These results explained the successful construction of the CAB composites.

FTIR spectroscopy further revealed the crystal structures and functional groups present on the sample surface [45]. The FTIR spectra of Figure 1c analyzed the chemical bonds of BO, AMO, CdS, 10AB, and 25C10AB samples at room temperature with a range of 460–2500 cm⁻¹. The spectra showed that the bare BO peaks belong to the location of Bi-O vibrations at 483/530/598/631 cm⁻¹ [46]. For the CdS sample, the Cd-S stretching mode can be seen at 618 cm⁻¹ [47], and the peak of S-related groups at 938 and 1197 cm⁻¹ [47,48]. Meanwhile, the FTIR spectrum of AMO indicates the Mo-O feature at 795 and 830 cm⁻¹, which enables the successful preparation of AMO samples [49,50]. At 775 cm⁻¹ [51], the absorption peaks of Mo-O-Mo bond characteristics show signs of bending and vibration. These characteristic FTIR peaks revealed the formation of BO, AMO, and CdS phases. The characteristic peaks of BO for the 10AB material are evident. BO covers the face of AMO, making it unable to observe, resulting in a weak peak strength. The AB and CAB products obviously are clearly visible in the spectra. For all the samples, it can be identified that these samples are bonded with CO₃²⁻ groups (846/880 cm⁻¹ → bending vibration of out-of-plane, 1047,1084, 1120 cm⁻¹ → symmetric stretching vibration type, 1385/1399 cm⁻¹ → anti-symmetric stretching vibration, 1399/1460 cm⁻¹ → anti-symmetric stretching vibration, 1740 cm⁻¹ → metrical stretching is accompanied by in-plane bending) and H₂O molecules (1633 cm⁻¹) [22,46,52]. Compared to bare AMO and 10AB, more CO₃²⁻ groups and H₂O molecules are adsorbed onto the CdS, BO, and 25C10AB samples due to the higher intensity absorption peaks observed for the latter three materials.

2.2. XPS Analysis

The XPS analysis further confirmed the elemental types and chemical states of the synthesized composites, revealing the presence of AMO, BO, and CdS components. The results for XPS (a) Ag 3d, (b) Bi 4f, (c) Cd 3d, (d) Mo 3d, (e) O 1s, and (f) S 2p are shown in Figure 2, indicating that the samples consist of AMO, BO, and CdS.

The Bi 4f XPS testing results of the photocatalysts (Figure 2a) are well-fitted with two peaks with binding energies (BEs) of 164.04–164.48 eV and 158.73–159.17 eV, indicating Bi³⁺ species in BO, 10AB, and 25C10AB [53]. Besides the Bi 4f peaks, the smaller peaks observed at 161.3 and 162.3 eV are attributed to S 2p_3/2/2p_1/2 [22]. The O 1 s XPS atlas in Figure 2b reveal three types of peaks on the O 1 s spectrum originating from lattice oxygen (529.85–529.94 eV), water molecules (532.07–533.37 eV), and chemisorbed oxygen (530.95–531.69 eV) [46,47,52]. Moreover, a mild shift with the binding energies of elements is observed in the O 1 s peak of the 10AB and 25C10AB composites compared to bare BO and AMO, possibly due to carrier transfer between BO and AMO. The Ag 3d spectrum (Figure 2c) displays two peaks at binding energies of 367.91–368.08 eV and 373.94–374.11 eV, which match the Ag 3d_5/2 and Ag 3d_3/2 levels. The binding energies of Mo 3d_5/2/3d_3/2 detected at 235.08–235.32 eV and 231.81–232.15 eV (Figure 2d) confirm the presence of Ag⁺ and Mo⁶⁺ species in AMO [40]. Besides the Mo 3d peaks, a small peak at 229.78 eV is attributed to S 2 s. The Cd 3d_5/2/3d_3/2 binding energies are identified at 405.21–405.35 eV and 411.97–412.11 eV (Figure 2e), while the S 2p_3/2/2p_1/2 binding energies are observed at 161.40–161.49 eV and 162.40–162.49 eV (Figure 2f), explaining the presence of Cd²⁺ and S²⁻ species in the CdS nanoparticles [47]. The peak of Bi 4f_5/2 was observed at 164.03 eV in the S 2p spectra. The binding energy of pure BO, AMO, and CdS are as follows: Bi 4f_7/2/4f_5/2 at 159.17/164.48 eV (Figure 2a), Ag 3d_5/2/3d_3/2 at 368.08/374.11 eV (Figure 2c), Mo 3d_5/2/3d_3/2 at 232/235.2 eV (Figure 2d), Cd 3d_5/2/3d_3/2 at 405.21/411.97 eV (Figure 2e), and S 2p_3/2/2p_1/2 at 161.40/162.40 eV. These characteristic binding energy peaks suggest that Bi³⁺/O²⁻ species exist in BO, Ag⁺/Mo⁶⁺ species in AMO, and Cd²⁺ and S²⁻ species are present in CdS. In the case of 10AB and 25C10AB, a slight shift in the Bi 4f/Ag 3d/Mo 3d/S 2p/Cd 3d peaks was observed due to the interaction between BO, AMO, and CdS.

2.3. Morphology Observation

SEM and TEM testing was performed to illustrate the morphologies of the sample. SEM images reveal that BO exists with the granularity of 75 nm to 460 nm, and the diameter values roughly distributed around 200 nm approximately (Figure 3a). On the other hand, AMO is similar to an ellipsoid shape with sizes ranging from 2.2 μm to 3.6 μm (Figure 3b). As for CdS (Figure 3c), it comprises small particles with a size of a few nanometers.

For the 10AB composite (Figure 3d), AMO particles are observed to be closely assembled with the BO particles. SEM images of 25C10AB (Figure 3e) reveal heterojunction composites composed of ultrafine CdS nanoparticles decorating two different-sized particles. The SEM observational figures reveal that the binary (10AB) and ternary (25C10AB) heterostructures emerge. The EDS spectrum (Figure 3f) shows the presence of Bi/O/Cd/S/Ag/Mo elements in the 25C10AB composite, followed by additional elements such as Cu and C from the sample holder [46]. The atomic content ratio of Mo and Bi is significantly higher compared to other elements. For a more comprehensive understanding, transmission electron microscopy (TEM) analysis was conducted on 25C10AB.

The TEM results presented in Figure 4a and the high-resolution TEM (HRTEM) image in Figure 4b further confirm the formation of the CAB heterojunction, highlighting the lattice fringes associated with the BO (200) facet, AMO (220) facet, and CdS (111) facet within the particles. Additionally, the energy-dispersive spectroscopy (EDS) spectrum illustrates the distribution of constituent elements, including Bi, O, Cd, S, Ag, and Mo, and the signals of these elements are particularly pronounced in the heterojunction composites, as shown in Figure 4c–i.

2.4. Photogenerated Charge Performance

The electron-hole behavior of the well-reflected photoelectrochemical experiments was used to measure the enhancement mechanism of the redox reaction. The photocurrent response test of a transient state (Figure 5a) confirms the efficient separation of e⁻/h⁺ in the composite samples, as evidenced by the noticeable result of the photocurrent in the samples. Additionally, Figure 5b demonstrates that binary and ternary photocatalysts exhibit EIS spectra with notably smaller semicircle diameters compared to pure BO, with 25C10AB exhibiting the smallest semicircle diameter. This suggests a more pronounced carrier separation in the composites. The results of photoelectrochemical testing show that the excellent photocatalytic performance in CAB is primarily driven by the more efficient utilization of photogenerated carriers [44].

The generation and recombination of e⁻/h⁺ pairs under light radiation are well-known phenomena. As the amount of the recombination of e⁻/h⁺ decreases, the emission signal intensity in the photoluminescence (PL) spectra also decreases, leading to improved photocatalyst performance. To assess the impact of CdS nanoparticle modification on AB and validate the proposed mechanism, the PL spectra of CAB were analyzed. Figure 5c displays the tested PL spectra of BO, AB, and CAB composites with excitation wavelengths of 315 nm. The emission band of bare BO at room temperature, centered at 398 nm, is due to the radiative recombination quality of itself. It is clear that the emission peak intensities decrease in binary composites like 10AB compared to pure BO, with the largest decrease observed in the 25C10AB sample. This demonstrates that the recombination of e⁻/h⁺ is limited in composite products, supporting their enhanced photoactivity [45]. These PL results align with the results of an e⁻/h⁺ separation and photocatalytic survey, showing the promising applications of CAB heterojunction photocatalysts in environmental purification.

2.5. Optical Absorption Property

In addition, understanding the optical absorption characteristics of semiconductors is crucial for understanding photocatalytic performance [51]. The UV–vis DRS spectrum was implemented for prepared samples, as shown in Figure 6. Here, we observed the UV–vis diffuse reflectance spectra of AMO, BO, and xAB (Figure 6a), as shown in Figure 6b, displaying the light absorption characteristics of the prepared samples: CdS, xBA, and xCAB.

The pure BO and AMO samples exhibit features to absorb UV light and convert it to visible light, demonstrating visible light-induced redox activity. The 10AB composite samples show a significant increase in absorbing the visible light range compared to the pure samples, with a noticeable red shift after modification with CdS nanoparticles. These changes are attributed to the interactions between AMO, BO, and CdS within the composite sample. All products exhibit strong absorption in the visible light region (400–800 nm), indicating their potential for visible-light-induced photocatalytic characteristics. The UV spectra indicate that the samples were excited by visible light, exhibiting a mild red shift in absorption wavelength relative to pure BO and an absorption boundary in the spectra between 300 and 400 nm. The red shift enhances the activity of samples by reducing the band gap energy of nanocomposites. The band gaps of synthesized materials were determined using Equation (1):

$$E_g\left(\mathrm{e}\mathrm{V}\right)={\displaystyle \frac{1240}{{\lambda }_g\left(\mathrm{n}\mathrm{m}\right)}}$$

(1)

where λ_g are the profiles of the UV–vis DRS spectrum, which reveals interband transition-related absorption edges, and E_g is the optical band gap energy. The values for CdS, BO, and AMO are located at wavelengths of 551.8 nm, 518.6 nm, and 365.4 nm, respectively, which correspond to an E_g of 2.25 eV (CdS), 2.39 eV (BO), and 3.39 eV (AMO) as shown in Figure 6c. These E_g values appear to be different from those reported for these materials [39,42,53,54], indicating the influence of the synthesis methodology and phase composition on the observed E_g values.

2.6. Photocatalytic Activity of Photocatalysts

The properties of the synthesized photocatalysts were evaluated by degrading the target pollutant MB (C_{photocatalyst} = 1 g/L, C_MB = 10 mg/L). In Figure 7, the time-dependent degradation of MB ((a) and (d)), pseudo-first-order kinetic plots ((b) and (e)), decomposition efficiency (η (60 min)), and rate constants (k_app) ((c) and (f)) were analyzed. Photodegradation experiments were performed under visible light using as-prepared photocatalysts 1AB, 3AB, 7AB, 10AB, 15AB, 9C10AB, 18C10AB, 25C10AB, and 30C10AB, along with pure BO, AMO, and CdS samples. The degradation efficiencies were monitored at an absorption wavelength of 664 nm, which corresponds to MB. Notably, the BO and AMO photocatalysts achieved removal efficiencies of 40.6% and 60.4% of MB, within a 60 min period.

The construction of AB heterojunctions significantly enhances the photodegradation of MB. Increase the content of AMO, specifically in the 10AB composite, results in the highest photodegradation efficiency, η = 87.1%. This composite shows improved activity with rate constants 2.145 and 1.442 times higher than those of pure BO and AMO, respectively. However, an increase in AMO content above 10% reduces activity due to the insufficient assembly of excessive AMO particles on the BO surface for forming effective heterojunctions. Photoactivity enhanced is by incorporating CdS nanoparticles onto 10AB, with the 25C10AB composite exhibiting a photoactivity improvement of 1.136 times over that of 10AB. However, excessive AMO in 15AB and CdS in 30C10AB samples results in decreased degradation activity due to enhanced charge carrier recombination, resulting in a reduction of 6.6% and 2% in degradation efficiency compared to 10AB and 25C10AB, respectively. Another reason for the decreased performance could be due to more e⁻/h⁺ being trapped, which increases the likelihood of further recombination, leading to a decrease in carriers transferring to the surface of the samples for redox [55,56]. In order to explore the stability of 25C10AB, we conducted cycling tests [57], and the results showed that the performance of the material slightly decreased after five cycles, but the photocatalytic performance remained strong. This fully demonstrates the structural stability of the 25C10AB sample, and there is still good contact between the three materials. The heterojunction formed by their close contact leads to a high removal rate of MB (seen in Figure 8).

The photocatalytic activity of the 25C10AB photocatalyst with that of other heterojunctions for MB removal is shown in

Table 1. Comparison of the photodegradation performance of 25C10AB with that of other heterojunctions for MB removal.

Photocatalyst	Degradation Efficiency	Reference
25C10AB	60 min, 99%	This work
C3N4/BiVO4	90 min, 94%	[6]
CeO2/MgAl2O4	180 min, 95.5%	[9]
Cu2O/RGO/In2O3	120 min, 95.1	[12]
CdS/Bi4Ti3O12	50 min, 96.3%	[22]
MgAl2O4/CeO2/Mn3O4	180 min, 94.6%	[31]
SrMoO4/SrWO4	180 min, ~90%	[32]
MgS/Ag2MoO4	200 min, 90%	[38]
CdS/Cd-MOF	100 min, 91.9%	[41]
CdS/ZnWO4/ZnS	50 min, 95.6%	[47]
BaTiO3/Bi2WO6	120 min, 88.4%	[58]

. The results further revealed that the 25C10AB photocatalyst prepared in this experiment shows a photocatalytic activity superior to most other Z-scheme and type−II heterojunction photocatalysts.

Under natural conditions, the complexity of the water environment can significantly impact the removal of pollutants. The redox performance of products is highly influenced by the presence of anions and cations in pollutants, as variations in these components can alter the surface features of the semiconductors, affecting their performance. In many actual applications, various factors will impact the photodegradation of MB over the 25C10AB photocatalyst. As depicted in Figure 9a, when the sample content remains unchanged (1 g/L), an increase in MB concentration will cause a removal efficiency increase. Then, when the concentration is over 10 mg/L, the optimal MB concentration, the light usage (especially the UV absorption) of the product reduces. At the same time, this reduces the contact possibility between MB and the photocatalyst, resulting in a reduction in the MB removal rate [56]. Similarly, increasing the photocatalyst concentration initially enhances and subsequently decreases the rate of MB degradation, with the optimal catalyst dosage identified at C_catalyst = 1 g/L (Figure 9b). This behavior can be largely attributed to the higher catalyst concentration providing more catalytic sites for the degradation process. However, as the concentration continues to increase, it can obstruct light penetration at the catalyst’s surface, thereby reducing light absorption. Excessive catalyst content may lead to aggregation, which diminishes the number of available active sites and results in a rapid decline in photocatalytic performance [52]. Additionally, various inorganic anions and cations influence MB degradation, thus the addition of 3 mmol/L of salt substances into the reaction solution. Experiments were conducted on solutions containing anions and cations, with the results presented in Figure 9c,d. The impact of K⁺ and CO₃²⁻ can be disregarded, as they have a minimal impact on MB degradation, while other ions exert a moderate influence.

Cations serve as recombination centers, facilitating the rapid recombination of e⁻/h⁺, while anions consume •OH radicals through specific reactions. The resulting radicals, such as •SO₄⁻, •CO₃⁻, and •Cl, demonstrate lower oxidation potential in comparison to •OH species [57]. Generally, the removal of organic pollutants in photocatalytic reactions is predominantly driven by active free radicals, including •OH and •O₂⁻. The significance of these active species within photocatalytic processes can be illustrated through trapping and annihilation experiments.

2.7. Capture Experiment and EPR Testing

The interaction of photogenerated e⁻/h⁺ in redox reactions with O₂ and H₂O leads to the formation of •O₂⁻ and •OH, which play a crucial role in the removal of MB. To identify the potential presence of these species in the 25C10AB products, three types of scavengers were employed during the reaction process to capture them. Figure 10a illustrates the impact of these scavengers on MB degradation. Notably, benzoquinone (BQ, a scavenger of superoxide •O₂⁻), ammonium oxalate (AO, a scavenger for h⁺), or absolute ethanol (ethanol, a scavenger for hydroxyl •OH) significantly inhibit MB degradation [21,58]. Furthermore, AO or BQ demonstrate a stronger inhibitory effect on MB degradation compared to ethanol, suggesting a strong correlation between MB degradation and the concentrations of •O₂⁻ and h⁺, indicating a high correlation of MB degradation with •O₂⁻ and h⁺. Furthermore, 5,5–Dimethyl–1–pyrroline–N–oxide (DMPO) was used as a capture agent for radicals in this study, enabling us to detect the primary active species produced in the 25C10AB reactions through electron paramagnetic resonance (EPR) technology, as shown in Figure 10b,c. DMPO complexes with •O₂⁻ when dissolved in a methanol environment, while •OH complexes in deionized water [46].

The characteristic absorption peaks of DMPO–•OH and DMPO–•O₂⁻ were not detected in 25C10AB samples without light irradiation, indicating that the free radicals •O₂⁻ and •OH cannot be generated in the dark. However, upon exposure to simulated sunlight, the samples exhibited distinct absorption peaks corresponding to DMPO–•OH and DMPO–•O₂⁻, with stronger signal indicating the production of additional radical species. This confirms the existence of •OH and •O₂⁻ radicals during simulated sunlight irradiation. It is important to note that the scavenging of e⁻ or h⁺ can have a dual impact on the photodegradation process. Firstly, the consumption of e⁻ or h⁺ reduces the formation of •O₂⁻ (or •OH), thus diminishing the photocatalytic effect. Secondly, the lifetime of the charge carriers influences their ability to effectively participate in redox reactions.

2.8. DFT Calculation

To express the exchange correlation interaction between electrons, the generalized gradient approximation (GGA) functional of the Perdew–Burke–Ernzerhof (PBE) form was adopted for initial structural optimization. The cut–off energy was set to 500 eV in the plane–wave expansion. The convergence criteria were set to 10⁻⁵ eV and 10⁻² eV/Å for total energy and the Hellman–Feynman force acting on atoms, respectively. For the calculation of work function, we selected (201), (220), and (111) crystal planes for BO, AMO, and CdS, respectively.

The application of density functional theory (DFT) calculations is crucial for elucidating the work function (Φ) of BO, AMO, and CdS, which is essential for understanding their photocatalytic mechanisms [40,59,60]. Figure 11a–c displays the Φ values of three bare materials obtained through DFT calculations of the first principle. The Φ of our samples, being crucial for governing the charge transfer at heterojunction interfaces, is defined as the difference value between the stationary energy of electrons in a vacuum environment and the Fermi energy (E_F) (Φ = E_VAC − E_F) [52]. The calculated data of the electrostatic potentials for BO (201), AMO (220), and CdS (111) facets are presented in Figure 11a–c, revealing Φ values of 5.229 eV for BO, 5.413 eV for AMO, and 5.83 eV for CdS. The values of the Φ disparity among the materials indicates the carrier comes to transfer at the CAB heterojunction interface.

2.9. Photocatalytic Mechanism

This study investigates the use of three semiconductors, with distinct CB and VB energy levels for redox reactions, to enhance the separation of charge carriers and improve interfacial charge transfer efficiency within the composites. The evaluation of photocatalysis has revealed the exceptional activity of the CAB composite photocatalysts in degrading MB. Based on the analyses, a potential mechanism for the photodegradation is proposed, with the aim of enhancing the redox processes. The experimental findings indicate that the modification of BO particles with AMO significantly enhances photocatalytic activity, with further improvements observed upon the incorporation of CdS nanoparticles into AB. This enhancement is primarily attributed to the successful construction of heterojunctions [61]. Results from the density functional theory (DFT) calculations, referencing the normal hydrogen electrode (NHE) at pH = 7, reveal the E_F levels of the semiconductors, derived from their work functions: 0.91 eV for AMO, 0.73 eV for BO, and 1.33 eV for CdS. The potential levels (versus NHE) of the conduction and valence bands (CB/VB) for the three materials were calculated [35], and the results are as follows: −0.3/+3.09 V for AMO, +0.54/+2.93 V for BO, and −0.43/+1.81 V for CdS. The behavior of the charge carrier migration within the ternary CAB heterojunctions is schematically illustrated in Figure 12a.

When the CAB heterojunction is compactly integrated, thermal equilibrium is established when the E_F levels become consistent. This arrangement results in spontaneous electron diffusion from BO to AMO and CdS, driven by the hierarchy of Fermi levels (E_F(BO) > E_F(AMO) > E_F(CdS)). Consequently, negative charge centers form at the interfaces of AMO and CdS, while positive charged centers form at the interface with BO. This generates an internal electric field directed from BO toward AMO and CdS. Ultimately, the further movement of electrons is constrained by this electric field, leading to a gradual balance in the migration of electrons. In the photocatalytic process under simulated sunlight, BO, AMO, and CdS are photoexcited to create e⁻/h⁺ pairs. Because of the present barriers at the interfaces of CdS/Bi₂O₃ (CB) and AB, the photogenerated carriers could not migrate from the CB of BO to AMO or CdS. Instead, electrons accumulate on the CB of BO, and the holes transfer from the VB of BO to CdS due to the combined electric fields of the materials, as depicted in Figure 12b. The excited electrons gather on the surface of BO and the holes on AS, and as a result, e⁻/h⁺ pairs separate, inhibiting recombination and prolonging the lifetime of holes on the VB of AS and CB electrons on BO. This enhanced charge separation leads to more effective participation in the degradation of MB. Therefore, the ternary CAB heterojunction materials build a dual Z-scheme transfer mechanism for the highly efficient utilization of carriers.

The electrons play a significant role in BO that could reduce the O₂ on the material surface to •O₂⁻. The reduction in oxygen results in the production of highly oxidative •OH [62,63]. Simultaneously, photogenerated holes in AS can react with H₂O, producing •OH groups and ensuring a continuous supply of surface •OH groups. These mechanisms facilitate the photocatalytic degradation of MB. The mechanisms of the proposed reactions for the CAB heterojunction are outlined as follows:

CdS/AMO/BO + hv → CdS(e⁻ + h⁺)/AMO(e⁻ + h⁺)/BO(e⁻ + h⁺)

(2)

CdS(e⁻ + h⁺)/AMO(e⁻ + h⁺)/BO(e⁻ + h⁺) → CdS(h⁺)/AMO(h⁺)/BO(e⁻)

(3)

e⁻(BO) + O₂ → •O₂⁻

(4)

h⁺(CdS)/AMO(h⁺) + OH⁻ → •OH

(5)

h⁺(CdS)/AMO(h⁺) + H₂O → •OH + H⁺

(6)

•O₂⁻, •OH + organic pollutants (MB) → degradation substances

(7)

3. Experimental Section

The main raw materials and reagents used in the present study include bismuth nitrate pentahydrate (Bi(NO₃)₃•5H₂O, analytical grade > 99.0%), silver nitrate (AgNO₃, analytical grade > 99.0%), sodium molybdate dihydrate (Na₂MoO₄•2H₂O, analytical grade > 99.0%), sodium sulfide nonahydrate (Na₂S•9H₂O, analytical grade > 98.0%), cadmium nitrate tetrahydrate (CdN₂O₆•4H₂O, analytical grade > 99.0%), and ethanol (CH₃CH₂OH, analytical grade > 99.7%). All the raw materials and reagents were purchased from Sinopharm Group Chemical Reagent Co., Ltd. (Shanghai, China). for direct use in the experiments. The water used in all the experiments was deionized water (DI water).

3.1. Fabrication of BO

Pure BO nanoparticles were prepared in the experiment as follows: a certain amount of Bi(NO₃)₃•5H₂O was dissolved in a dilute nitric acid aqueous solution of a certain concentration in a beaker. Then, the container was atomized slowly by a spray head at 750 °C. These acquired droplets were subsequently dried using compressed air to remove evaporation, resulting in the formation of BO nanoparticles.

3.2. Fabrication of AMO

Pure AMO particles were prepared as follows: 2 mmol of AgNO₃ and 1 mmol of Na₂MoO₄•2H₂O were weighed and dissolved in 30 mL of ultra-pure water, stirring the mixture for 30 min. Next, the prepared solution was gradually added while continuing to stir for an additional hour to produce the AMO samples. To eliminate impurities from other ions, the solution was rinsed with ultra-pure water and anhydrous ethanol and then dried at 70 °C.

3.3. Fabrication of AMO/BO (AB)

The AB samples were prepared by dissolving 0.1 g of as-prepared BO and a specific mass fraction of AMO in 70 mL of ultra-pure water, followed by processing in ultrasonication equipment for 30 min and magnetic stirring for 30 min. The turbid liquid was then poured into a Teflon liner and heated to 120 °C, reacting for 1 h. The resulting reactions were collected, the supernatant was removed, and the acquired turbid liquid was washed and dried to prepare the final xAMO/BO composites (x = 1%, 3%, 7%, 10%, and 15%), labeled 1AB, 3AB, 7AB, 10AB, and 15AB for different mass fractions of AMO of 1%, 3%, 7%, 10%, and 15%, respectively.

3.4. Fabrication of CdS/AMO/BO (CAB)

The yCdS/10% Ag₂MoO₄/Bi₂O₃ (yC10AB) with varying CdS content (y% = w_CdS/(w_CdS + w_BO@_AMO) × 100%) materials (y = 9%, 18%, 25%, and 30%) were prepared by dissolving 0.2 g of the synthesized 10AB in 40 mL of deionized water, followed by 45 min of magnetic stirring. Stoichiometric amounts of Cd(NO₃)₂•4H₂O and Na₂S•9H₂O were then added and mixed for an additional 60 min using magnetic stirring. The precipitate was rinsed, dried, and labeled with 9C10AB, 18C10AB, 25C10AB, and 30C10AB for different mass fractions of CdS, respectively. Pure CdS nanoparticles were synthesized without the addition of AB. The schematic diagram for the fabrication of samples is shown in Figure 13.

3.5. Characterizations

The phase purity of the samples was characterized by X-ray powder diffraction (XRD) using a Bruker D8 Advance X-ray powder diffractometer (Bruker, Karlsruhe, Germany) with Cu Kα radiation (λ = 0.15406 nm) at a scanning rate of 4°/min. The morphology, microstructure, and chemical composition of the samples were investigated by scanning electron microscopy (SEM) using a JSM-6701F field-emission scanning electron microscope (JEOL Ltd., Tokyo, Japan) coupled with energy-dispersive X-ray spectroscopy (EDS) (JEOL Ltd., Tokyo, Japan) and transmission electron microscopy (TEM) using a JEM-1200EX field-emission transmission electron microscope (JEOL Ltd., Tokyo, Japan). X-ray photoelectron spectroscopy (XPS) measurements were performed using a PHI-5702 multi-functional X-ray photoelectron spectrometer (Physical Electronics, Eden Prairie, MN, USA). Photoluminescence (PL) spectra were recorded using a RF-6000 fluorescence spectrophotometer (excitation wavelength: ~300 nm) (Shimadzu Experimental Equipment Co., Ltd., Shanghai, China). Ultraviolet–visible (UV–vis) diffuse reflectance spectroscopy (DRS) spectra were obtained using a TU-1901 double beam spectrophotometer (Beijing Purkinje General Instrument Co., Ltd., Beijing, China).

3.6. Electrochemical Testing

The transient photocurrents and electrochemical impedance spectroscopy (EIS) spectra of the samples were measured on a CHI660E electrochemical workstation using a three-electrode cell configuration. A platinum foil electrode and a standard calomel electrode (SCE) were used as the counter electrode and reference electrode, respectively. To prepare the working electrode, 15 mg of the photocatalyst, 0.75 mg of poly-vinylidene fluoride (PVDF), and 0.75 mg of carbon black were uniformly mixed using 1-methyl-2-pyrrolidione (NMP) as the solvent. The formed slurry mixture was uniformly coated onto a fluorine-doped tin oxide (FTO) glass substrate with an effective area of 1 cm × 1 cm. A total of 0.1 M Na₂SO₄ aqueous solution was used as the electrolyte. A 300 W PLS-SXE300BF xenon lamp (Beijing Bofulai Technology Co., Ltd., Beijing, China) was employed as the light source to generate simulated sunlight. The photocurrent response was measured at a bias potential of 0.2 V. EIS measurements were conducted by applying a sinusoidal voltage pulse with an amplitude of 5 mV over a frequency range of 10⁻² to 10⁵ Hz.

3.7. DFT Calculations

All the density functional theory (DFT) calculations were performed via the Vienna Ab initio Simulation Package (VASP6.3.), and the projector augmented plane wave (PAW) pseudopotentials were used for the elements involved. The projector augmented wave was used to describe the electron–ion interactions while the Perdew–Burke–Ernzerhof (PBE) functional was used to model the exchange correlation energy. A cut-off energy of 500 eV was used for the structure relaxation and electronic property calculation of monolayer materials. The electronic energy was considered self-consistent when the energy change was smaller than 10⁻⁵ eV, and the geometry optimization was considered convergent until the forces on all atoms were smaller than 0.03 eV/Å.

3.8. Photocatalytic Performance Testing

The photoactivated degradation behavior of the BO, AMO, CdS, xAB, and yCAB composite samples was evaluated by removing MB in an aqueous solution under visible light conditions. The visible light was emitted by a 500 W Xe lamp (Beijing Bofulai Technology Co., Ltd., Beijing, China). During each experiment, a 0.1 g sample was added to a 100 mL MB solution (10 mg/L⁻¹). At the start of the experiment, the suspensions with samples were stirred in a dark environment for 30 min to ensure complete MB absorption onto the catalyst, ensuring a balanced adsorption–desorption (which eliminates the influence of the adsorption effect on the photocatalytic degradation process). Therefore, we speculated that molecules were in a dissociated state in the solution. As for the specific situation of MB molecule adsorption or dissociation in the solution, we currently do not have the software or relevant equipment to explain and test it. At 10 min intervals of irradiation time, the suspensions were collected and centrifuged at 3680 rpm for 6 min to remove the samples. The MB concentrations were measured at 664 nm [45] using a UV–visible spectrophotometer (Shimadzu UV–vis 1700) (Shimadzu Scientific Instruments, Inc., Tokyo, Japan). The degradation rate of the target pollutant (η) was determined via the value of C_t and C₀ using Equation (8):

η = (1 − C_t/C₀) × 100%

(8)

where C_t is the pollutant concentration at time t, and C₀ is its initial concentration.

4. Conclusions

A novel AB heterojunction photocatalyst was synthesized using a one-step hydrothermal method. The 10AB composite demonstrated the highest MB removal efficiency at 87.1%, outperforming BO and AMO by factors of 2.145 and 1.442, respectively. To further enhance its photocatalytic performance, CdS nanoparticles were deposited onto the surface of the AB heterojunction using the co-precipitation method. The results indicated that the 25C10AB composite achieved optimal MB degradation, reaching 99%, thus surpassing the performance of the 10AB composite. The dual Z-scheme CAB heterojunction photocatalyst outperformed the bare AMO, BO, CdS, and AB heterojunction photocatalysts by optimizing the migration paths of charge carriers, facilitating redox reactions at spatially separated active sites, generating increased amounts of •O₂⁻ and •OH radicals, and reducing charge carrier recombination. This study provides a valuable reference for the development of new photocatalysts that are safer and more efficient for the photocatalytic degradation of pollutants.