Efficient Removal of Methylene Blue Using an Organic–Inorganic Hybrid Polyoxometalate as a Dual-Action Catalyst for Oxidation and Reduction

Abstract

An organic–inorganic hybrid polyoxometalate (POM) CoPMoV [PMo^VI₈V^IV₄V^V₂O₄₂][Co(Phen)₂(H₂O)]₂[TEA]₂•H₃O•3H₂O (Phen = 1,10-phenanthroline, TEA = triethylamine) prepared by hydrothermal synthesis was explored as a heterogeneous catalysts to remove methylene blue (MB) through Fenton-like reaction and catalytic reduction. X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive X-ray (EDX), Fourier transform infrared spectroscopy (FT-IR), and X-ray photoelectron spectroscopy (XPS) were employed to characterize CoPMoV. The MB removal rates for the Fenton-like reaction and the catalytic reduction were 91.6% (120 min) and 97.5% (2 min), respectively, under optimum conditions. CoPMoV demonstrated excellent stability and recyclability in the Fenton-like reaction and catalytic reduction, which was confirmed by 5 cycle tests. Plausible mechanisms for MB degradation and reduction have also been proposed. Benefiting from the excellent redox properties of cobalt and [PMo^VI₈V^IV₄V^V₂O₄₂]⁵⁻ anion, CoPMoV could act as a Fenton-like and reductive catalyst for the removal of MB. This study provides a green and facile strategy to design POM-based organic–inorganic material for dye wastewater treatment via oxidation and reduction.

1. Introduction

The development of different industries has led to the discharge of various dye wastewater into the environment, which threatens ecological security [1]. Thus, dealing with dye wastewater before discharge to meet ecological safety requirements is an important issue that must be addressed worldwide. Methods represented by adsorption, coagulation, advanced oxidation processes (AOPs), and membrane separation have been developed to eliminate dyes from industrial effluents [1]. Among them, AOPs have received extensive investigation [2]. The Fenton process is one of the most cost-effective AOPs, and the in situ generated reactive oxygen species (ROS) during the Fenton process can effectively destroy organic pollutants including dyes [3]. However, the traditional Fenton process has the disadvantages of a narrow suitable pH range (3–4), iron sludge formation, and low utilization efficiency of H₂O₂ [3]. Therefore, developing other heterogeneous Fenton-like processes with a wide applicable pH range, high efficiency, higher durability, and simple separation is the best solution to overcome the disadvantages of the traditional Fenton process, which can make the process more practical [4].

Dyes can be catalytically reduced to less toxic and more easily degradable products in the presence of NaBH₄; this discovery has gradually attracted researchers’ attention [5]. So far, the catalysts used in the reduction are mainly focused on noble metal nanoparticles (NPs) [5]. Considering economic benefits, it is better to develop other suitable catalysts to make the method have more potential for practical application.

Polyoxometalates (POMs) are metal–oxygen clusters of transition metals V, Nb, Ta, Mo, and W. POMs are widely studied in the fields of catalysis, materials [6,7], photochemistry [8], and medicine due to their diverse and tunable composition, well-defined structure, nano-size, reversible redox potential, semiconductor-like optical properties, and high thermal stability [9,10]. In addition to their unique physical and chemical properties, POMs are non-toxic and eco-friendly inorganic polymers. When POMs act as catalysts, the catalytic reactions can take place on the surface and in the solid phase of POMs [10]. Although Keggin-type POMs can produce oxygen-containing active substances (e.g., ^•OH) in the presence of H₂O₂, few works about the utilization of POMs-H₂O₂ Fenton-like systems to deal with organic pollutants appeared [10,11]. POMs-H₂O₂ systems can be carried out under neutral and alkaline conditions, which gives them the potential to overcome the disadvantage of low acidity of the traditional Fenton system [10,11]. However, to use POMs as effective and applicable catalysts, their high water solubility must be avoided. Therefore, POMs were loaded on various materials to obtain heterogeneous catalysts based on POMs [12], but this method encountered the aggregation and leakage of catalysts [12]. The hydrothermal synthesis method is the other accepted route to prepare water-insoluble POMs, especially the organic–inorganic hybrids based on POMs (OI-POMs) [13]. The structures and topologies of OI-POMs can be tuned by varying the types of organic ligands and different metal ions [13]. The structural clarity and water resistance properties of OI-POMs have attracted researchers to explore their functions, especially in catalysis [13]. Although OI-POMs were superior in terms of activity and reusability, exploring them as Fenton-like catalysts for dye wastewater treatment has not received sufficient attention yet [13].

Researchers have found that transition metal cobalt is able to undergo Fenton-like reactions for its variable valence state and redox capability, and thus have a facilitating effect on H₂O₂ [14]. Therefore, various cobalt composites have been developed to act as catalysts due to their excellent redox activity [14].

Methylene blue (MB) is widely used in various industries, and the industrial effluents containing MB may have an impact on the health of living organisms.

Based on the above considerations, an OI-POM [PMo^VI₈V^IV₄V^V₂O₄₂][Co(Phen)₂(H₂O)]₂[TEA]₂•H₃O•3H₂O (CoPMoV) (Phen = 1,10-phenanthroline, TEA = triethylamine) was hydrothermally synthesized and explored as a heterogeneous Fenton-like and reductive catalyst for the removal of MB. CoPMoV combined the excellent redox activity of cobalt and [PMo^VI₈V^IV₄V^V₂O₄₂]⁵⁻ and was a good candidate to act as a catalyst. The aim of this study is to provide experimental foundations for dye wastewater treatment using OI-POMs as dual-action catalysts and extend their application to the field of environmental remediation.

2. Results and Discussion

2.1. Characterization of CoPMoV

The coordination mode and 3D framework of CoPMoV are given in Figure 1. The morphology of CoPMoV was characterized by SEM. Figure 2a,b indicate that CoPMoV displayed a bulk morphology with a size range of 0.51 nm–2.14 nm. The EDX result was consistent with the analysis of X-ray single crystal structure characterization (Figure 2c), CoPMoV was composed of C, N, O, P, Mo, V, and Co elements, and each element was uniformly distributed on the surface of CoPMoV (Figure S1a–h).

In the IR spectrum of CoPMoV (Figure 3), the peaks that appeared at 945 cm⁻¹, 855 cm⁻¹, 788 cm⁻¹, 725 cm⁻¹, and 1045 cm⁻¹ were attributed to ν (Mo=O_d), ν (Mo-O_b-Mo), ν (Mo-O_c-Mo), ν (V-O-V), and ν (P-O_a), respectively [15]. The peaks at 1522 cm⁻¹, 1463 cm⁻¹, 1431 cm⁻¹, and 1152 cm⁻¹ originated from the organic ligand [16,17]. The peaks at 3438 cm⁻¹ and 1618 cm⁻¹ were caused by the presence of water [17]. O_a, O_t, O_b, and O_c represent the oxygen atoms coordinated with the central atom, terminal oxygen atoms, corner oxygen atoms shared between M₃O₁₃ units, and edge oxygen atoms shared within M₃O₁₃ units, respectively.

The phase purity and stability of CoPMoV were examined by powder X-ray diffraction (PXRD) technique. Figure 4 indicates that the PXRD pattern of the CoPMoV powder obtained from the experiment was highly consistent with the simulated pattern fitted by the X-ray single crystal structure analysis, proving the high phase purity and stability of the CoPMoV bulk [18,19].

The surface chemical properties of CoPMoV were studied with the XPS technique, with the C1s peak calibrated at 284.8 eV. The XPS spectra of CoPMoV confirm the presence of C, N, O, P, Mo, V, and Co elements (Figure 5a). As shown in Figure 5b, the C1s peak was deconvoluted into two peaks at 284.8 eV and 285.7 eV, corresponding to C-C and C-O bonds, respectively [20,21]. Figure 5c displays the N1s XPS spectrum, with the peak at 397.5 eV attributed to the Co-N bond [22], and the peaks at 399.6 eV and 401.3 eV attributed to pyridine nitrogen and graphitic nitrogen, respectively [23]. The peak belonging to Mo 3p appeared at 395.6 eV. Figure 5d shows the O 1s XPS spectrum, where the peak at 530.2 eV was attributed to M-O (M=Co, Mo) bonds [24], and the peak at 531.6 eV was attributed to O-P/V bonds [25,26]. The peak at 533.5 eV was assigned to O-H bonds, originating from crystal water [27]. As shown in Figure 5e, the P 2p XPS spectrum had a peak located at 134.1 eV [28]. The peaks at 232.2 eV and 235.4 eV in Figure 5f were attributed to Mo 3d_5/2 and Mo 3d_3/2 of Mo⁶⁺, respectively [29]. In the V 2p XPS spectrum, the peaks at 516.4 eV and 523.2 eV were assigned to V 2p_3/2 and V 2p_1/2 of V⁴⁺ [30], and the peaks at 520.2 eV and 524.6 eV were assigned to V 2p_3/2 and V 2p_1/2 of V⁵⁺, respectively (Figure 5g) [26]. The XPS spectrum of Co 2p is given in Figure 5h; the peaks at 779.3 eV and 795.3 eV were assigned to Co 2p_3/2 and Co 2p_1/2 of Co²⁺, while the peaks at 784.5 eV and 801.2 eV were attributed to satellite peaks [31]. In addition, the elemental ratios obtained through the XPS analysis were similar to the results of EDX (Table S1).

2.2. Degradation of MB in the CoPMoV-H₂O₂ Fenton-like System

The heterogeneous Fenton-like catalytic ability of CoPMoV was investigated through MB degradation. The control experimental results indicate that H₂O₂ itself could not degrade MB, the degradation rate MB was 43.0% in the presence of CoPMoV, and the best degradation rate of 91.6% was reached in the CoPMoV-H₂O₂ Fenton-like system (Figure 6a). As shown in Figure S2, the initial pH and concentration of MB solution and the dosage of H₂O₂ and CoPMoV had a significant effect on the degradation rate of MB in the CoPMoV-H₂O₂ Fenton-like system. It is worth mentioning that the best degradation rate could be achieved without adjusting the pH value of the MB solution; this is superior to the traditional Fenton reaction.

The UV-vis spectra of the MB degradation process over reaction time under the optimum conditions are given in Figure 6b. It can be seen that the intensity of the λ_max of MB at 664 nm sharply decreased in 30 min, and then gradually decreased from 30 min to 120 min along with the disappearance of the absorption peaks between 200 nm and 300 nm. The color of the MB solution changed from blue to colorless in 120 min. Such results confirm that the MB molecule was destroyed in the CoPMoV-H₂O₂ Fenton-like system [32].

2.2.1. MB Degradation Intermediates Study

An HPLC-MS analysis was taken to identify the degradation intermediates of the MB produced in the Fenton-like reaction. In the mass spectrum of the Fenton-like system (120 min), 11 intermediates with m/z of 284, 278, 253, 237, 217, 186, 167, 141, 122, 111, and 100 were observed (Figure 7a). The molecular ion at 284 was assigned as C₁₆H₁₈N₃S⁺, which underwent a series of structural changes to become other small molecules. The possible degradation pathway of MB is given in Figure 7b, the structural changes happening in Figure 7b suggest that oxidative radicals should be responsible for the generation of the intermediates. It is evident that demethylation, deamination, and ring-opening reactions occurred in the degradation. The yield of CO₃²⁻, SO₄²⁻, and NO₃ was solidified by the FT-IR spectrum of the MB degradation products which exhibited the characteristic stretching vibrations of CO₃²⁻, NO₃⁻, and SO₄²⁻ at 1448 cm⁻¹, 1380 cm⁻¹, and 1090 cm⁻¹, respectively (Figure S3) [17,33].

2.2.2. Reactive Species Analysis

Generally speaking, reactive oxygen species (ROS) represented by ^•OH radicals play crucial roles in Fenton-like reactions. Therefore, different radical scavengers were added to the CoPMoV-H₂O₂ Fenton-like system to judge the types of ROS. Sodium chloride (NaCl) and isopropanol (IPA) were used as ^•OH scavengers, and 1,4-benzoquinone (BQ) was taken as ^•O₂H scavenger [33,34]. The order of inhibition effect was NaCl ≈ IPA > BQ, indicating that ^•OH plays a decisive role (Figure 8a). A ^•OH-trapping photoluminescence experiment of the Fenton-like system in a terephthalic acid (TA) solution was carried out to further confirm the formation of ^•OH radicals [35]. TA was non-fluorescent, and the PL emission peak of the Fenton-like system in TA solution (0.5 M NaOH +0.5 mM TA) at 450 nm was derived from the highly fluorescent 2-hydroxyterephthalic acid which is the product of the reaction between ^•OH and TA (Figure 8b). The increased intensity of 2-hydroxyterephthalic acid indicates that the longer the reaction time, the higher the yield of ^•OH radicals.

2.2.3. Mechanism of MB Degradation

According to the results of the ROS capture experiments and LC-MS analysis, the possible mechanism of MB degradation in the current Fenton-like system could be summarized as the following Equations (1)–(12). The production of ^•O₂H and ^•OH via Co²⁺→Co³⁺, V⁴⁺→V⁵⁺, and Mo⁶⁺→Mo⁵⁺ redox cycles was similar to the homogeneous Fe²⁺/H₂O₂ or Fe³⁺/H₂O₂ Fenton systems [17,36]. Because ^•O₂⁻ could be converted to ^•OH via Haber–Weiss reaction (Equation (9)) [37], the possible reduction of Mo⁶⁺ by V⁴⁺ might contribute to the production of ^•OH and thus boost the degradation rate. In addition, ¹O₂ might be generated by the disproportionation of ^•OH radicals (Equation (10)) [38], and the production of H₂O₂ could not be excluded (Equation (11)) [39]. The degradation intermediates were generated by the attack of ^•O₂H, ¹O₂, and ^•OH on MB and its degradation products until the generation of various small molecules.

≡Co²⁺ + H₂O₂ → ≡Co³⁺ + ^•OH + OH⁻

(1)

≡Co³⁺ + H₂O₂ → ≡Co²⁺ + ^•O₂H + H⁺

(2)

≡V⁴⁺ + H₂O₂ → ≡V⁵⁺ + ^•OH + OH⁻

(3)

≡V⁵⁺ + H₂O₂ → ≡V⁴⁺ + ^•O₂H + H⁺

(4)

≡Mo⁶⁺ + H₂O₂ → ≡Mo⁵⁺ + ^•O₂H + H⁺

(5)

≡Mo⁵⁺ + H₂O₂ → ≡Mo⁶⁺ + ^•OH + OH⁻

(6)

≡Mo⁶⁺ + ≡V⁴⁺ → ≡Mo⁵⁺ + ≡V⁵⁺

(7)

^•O₂H ↔ ^•O₂⁻ + H⁺

(8)

^•O₂⁻ + H₂O₂ → O₂ + ^•OH + OH⁻

(9)

4^•OH → ¹O₂ + 2H₂O

(10)

2^•O₂H → H₂O₂ + O₂

(11)

MB + ^•O₂H/^•OH/¹O₂ → degradation products

(12)

2.2.4. Reusability of CoPMoV in the Fenton-like Reaction

The reusability of catalysts is one of the important indicators for their practical application. So, repeated cycle experiments were run to evaluate the reuse performance of CoPMoV. In each cycle, CoPMoV was simply filtrated, washed with acetonitrile and deionized water, dried in air, and then used in the reaction. As Figure 9 shows, the MB degradation rate slightly decreased from 91.6% to 89.2% after five consecutive cycles. The comparison between the IR and XRD spectra of CoPMoV before and after the Fenton-like reaction is presented in Figure S4a,b; it is evident that basically, no changes had occurred in the corresponding spectra. Such data indicate that CoPMoV possessed good stability and cycle performance. In addition, the simplicity of the recycling operation made CoPMoV valuable for application.

2.3. Catalytic Reduction of MB

In order to further develop the application performance of CoPMoV, its catalytic activities toward MB reduction were studied. The control experiment results indicate that the UV-vis spectrum of MB did not change in the NaBH₄ or CoPMoV single system (Figure 10a), while the characteristic absorption peaks of MB at 292 nm and 664 nm diminished in the CoPMoV-NaBH₄ binary system in 2 min. That is to say that the reduction of MB could only occur under conditions of the coexistence of CoPMoV and NaBH₄, where CoPMoV and NaBH₄ acted as catalysts and reducing agents, respectively. The catalytic reduction of MB was restricted by the conditions of CoPMoV dosage, NaBH₄ dosage, and initial MB concentration (Figure S5). Under the optimum conditions, the rate of MB reduction could reach 97.5% in 2 min. As shown in Figure 10b, the MB solution changed from blue to colorless in 2 min accompanied by the disappearance of the characteristic absorption peaks of MB, indicating that MB was reduced to less toxic leuco-methylene blue (LMB) [40]. Figure 10c represents a plot of ln (C_t/C₀) against time that was applied at an MB concentration of 0.01 mM (50 mL). The good linear relationship in Figure 10c proves that the MB reduction obeyed pseudo-first-order reaction kinetics due to the excess NaBH₄ [41]. The rate constant (K) calculated by Equation (13) was 1.788 min⁻¹, and the confidence interval for K was −2.1265~−1.4501.

2.3.1. Reduction Product Analysis

An LC-MS analysis was carried out to confirm the generation of the reduction product LMB. The m/z of 284 and 286 were assigned as the MB and LMB molecules, respectively, proving the reduction had truly occurred (Figure 11a). The other m/z of 269, 216, 182, 158, and 110 belonged to the degradation intermediates of LMB under the testing conditions (Figure 11a). The reduction product LMB and its further structural changes are displayed in Figure 11b. These products solidify the conclusion that the reduction product LMB is easier to destroy compared with MB [5].

2.3.2. Reduction Mechanism

The catalytic reduction mechanism of MB in the presence of CoPMoV and NaBH₄ was proposed (Figure 12). In the reduction process, NaBH₄ and MB acted as the electron donor and electron acceptor, respectively. Firstly, NaBH₄ decomposed to BH₄⁻ in solution, and then BH₄⁻ further reacted with H₂O to produce BO₂⁻, H⁺, and e⁻ (Equation (13)) [42]. Secondly, CoPMoV transferred H⁺ and e⁻ to MB and the reduction happened. CoPMoV had intermediate redox potential and thus acted as a transmitter to accept electrons from NaBH₄ and then transfer them to MB [43]. The high catalytic activities of CoPMoV were derived from its POM sub-unit with excellent redox properties. Additionally, the electron hopping between V⁴⁺ and V⁵⁺ enhanced the lifetime of electrons and thus promoted the reduction [44].

BH₄⁻ + 2H₂O → BO₂⁻ + 8H⁺ + 8e⁻

(13)

2.3.3. Reusability of CoPMoV in the Reduction

The reusability of CoPMoV in the reduction was studied to evaluate its application potential, and the recovery process of CoPMoV was the same as that of the Fenton-like reaction. There was no observable loss of the catalytic activity of CoPMoV after five consecutive cycles (Figure 13). Furthermore, the IR and XRD spectra of CoPMoV before and after the reaction remained highly consistent. These results prove that CoPMoV was stable enough to be reused repeatedly in the catalytic reduction of MB.

3. Experimental

3.1. Chemicals and Materials

All the chemicals were analytical grade, commercially available, and used directly. CoPMoV was prepared according to the literature method [16], and the details were as follows:

A mixture of NH₄VO₃ (0.36 g, 3.0 mmol), H₃PMo₁₂O₄₀•χH₂O (0.52 g, 0.28 mmol), CoCl₂•6H₂O (0.42 g, 1.77 mmol), TEA (0.5 mL), Phen•H₂O (0.15 g, 0.75 mmol), and H₂O (10 mL) was stirred for 60min, and then the pH value of the mixture was adjusted to 6 with H₃PO₄. At last, the above mixture was transferred to an 18 mL Teflon-lined reactor and heated for 3 d at 160 °C. After cooling to room temperature, the obtained black block crystals were washed with distilled water and air-dried. Yield: 65% (based on Mo).

3.2. Characterizations

Fourier infrared spectroscopy (Bruker (Billerica, MA, USA) Vertex 80 FT-IR spectrometer), X-ray diffraction (Rigaku (Tokyo, Japan) XRD Ultima IV diffractometer equipped with nickel-filtered Cu Kα radiation (λ = 1.5418 Å)), electron scanning microscope (FEI, Hillsboro, OR, USA, Quanta 20) equipped a device for energy dispersive spectroscopy (EDX) measurements, and X-ray photoelectron spectroscopy (AXIS UltraDLD with AlKa X-ray as the excitation source) were applied to characterize the structure, surface morphology, and chemical composition of CoPMoV, respectively. Using Avantage to analyze the fine spectra of each element in the XPS results, we obtained the relative elemental content in CoPMoV. The UV-vis absorption spectra of MB solutions were measured with a TU-1901 UV-vis spectrophotometer (Beijing Purkinje General Instrument Co., Ltd., Beijing, China). HPLC-MS (Agilent, Santa Clara, CA, USA, 5975c) was used to identify the degradation and reduction products of MB.

3.3. MB Degradation

The pH of the MB solution was adjusted with 0.1 M NaOH/HCl. In the Fenton-like reaction, MB solution was degraded directly, and 3 mL MB solution was sampled, filtered with 0.45 µm membrane, and submitted to UV-vis analysis at a given time. The concentration of MB was determined by monitoring the intensity of the maximum absorbance at 664 nm. Each experiment was repeated at least three times under identical conditions.

3.4. MB Reduction

Freshly prepared NaBH₄ solution was added to the MB solution under stirring and then CoPMoV powder was poured into the solution at room temperature. A total of 3 mL solution was sampled, filtered with 0.45 μm membrane, and submitted to UV-vis analysis at a given time. The following Equation (14) was used to calculate the first-order reaction kinetic rate constant (K, min⁻¹) [15]. All the tests were repeated three times.

ln(C_t/C₀) = −Kt

(14)

where C₀ and C_t are the MB concentrations at time t and 0, respectively, and t is reaction time (min).

4. Conclusions

A hydrothermally synthesized organic–inorganic hybrid CoPMoV based on Keggin-type POM was explored as a heterogeneous catalyst for the Fenton-like degradation and catalytic reduction of MB. As expected, the synergistic effect of cobalt and [PMo^VI₈V^IV₄V^V₂O₄₂]⁵⁻ anion endowed CoPMoV with excellent redox activities. In the Fenton-like reaction, the degradation rate of MB was 91.6% in 120 min, and ^•OH and ^•O₂⁻/^•O₂H were the main ROS. The HPLC-MS and FT-IR analysis of the degradation products confirmed that the MB molecules were degraded into small organic molecules and inorganic ions. In the catalytic reduction of MB, the reduction rate was 97.5% in 2 min, the reduction followed the pseudo-first-order kinetic model, and the first-order reaction kinetic reduction rate constant K was 1.788 min⁻¹. The cycle tests of the Fenton-like reaction and the reduction demonstrated that CoPMoV possessed high stability and reusability. In addition, the mild reaction conditions, easy implementation, and high efficiency of the Fenton-like reaction and the reduction suggest the practical application potential of CoPMoV for the removal of MB. In addition, the other reported OI-POM [Cu^I(phen)₂]₄[PMo^VI₈V^IV₆O₄₂{Cu^I(phen)}₂]•H₅O₂ (CuPMoV) showed better catalytic activities than CoPMoV due to its composition [PMo^VI₈V^IV₆O₄₂]⁷⁻ with a higher degree of reduction. The novelty of the present work lies in our discovery of the organic–inorganic hybrids derived from POM and transition metal with excellent redox activity that can be used either in oxidation or reduction process due to the synergistic redox properties of the building sub-units, and the reduction degree of their POM anions has an effect on the catalytic properties. More work should be carried out to develop the catalytic performances of organic–inorganic hybrids based on different POMs and transition metals in the field of environmental remediation.