Mechanistic Study of Methyl Orange Removal by Fe₃O₄@MIL-53(Fe Cu) Composite Material

Abstract

A novel magnetic composite, Fe₃O₄@MIL-53(Fe Cu), was successfully synthesized and applied for the efficient removal of methyl orange (MO) from aqueous solutions. The ad sorption performance was systematically evaluated under various conditions, including adsorbent dosage, solution pH, coexisting anions, and regeneration cycles. The results demonstrated that an optimal dosage of 20 mg achieved a removal efficiency exceeding 85%, with maximum adsorption observed at pH 3. The presence of common anions (Cl⁻, SO₄²⁻, CO₃²⁻, and PO₄³⁻) showed negligible effects on MO removal. Kinetic studies revealed that the adsorption process followed the pseudo-second-order model. Although minor chemisorption contributions were observed, the Dubinin–Radushkevich (D–R) model confirmed the predominance of physical adsorption. The Freundlich isotherm provided the best fit to the equilibrium data, indicating a maximum adsorption capacity of 193.65 mg/g and suggesting multilayer adsorption on a heterogeneous surface. Thermodynamic analysis confirmed the spontaneous and endothermic nature of the adsorption process. The primary mechanisms governing MO adsorption were identified as electrostatic attraction, π–π interactions, and hydrogen bonding. The composite exhibited excellent reusability over multiple cycles, demonstrating its potential for practical wastewater treatment applications.

1. Introduction

The large-scale discharge of organic pollutants into environmental systems has emerged as a significant global concern. Among these pollutants, synthetic dyes are particularly problematic due to their toxicity and detrimental effects on photosynthetic activity within aquatic ecosystems [1,2]. Wastewater containing dyes—predominantly originating from industries such as paper production, printing, textiles manufacturing, leather processing, food production, cosmetics formulation, and pharmaceuticals [3,4]—contributes significantly to water pollution issues. It is estimated that approximately 700 thousand tons of dyes are disposed annually worldwide [5]. Alarmingly, around 10–20% of these dyes are discharged directly into water bodies without adequate treatment during production or usage processes, posing serious threats to both aquatic ecosystems and human health.

Upon entering aquatic environments, dye pollutants contribute to turbidity, form surface films that obstruct sunlight penetration, and diminish dissolved oxygen (DO) levels. These effects, coupled with increased chemical oxygen demand (COD) and biological oxygen demand (BOD), disrupt the equilibrium of aquatic ecosystems and hinder the growth of photosynthetic organisms, thereby exacerbating pollution in a vicious cycle [6]. In addition to ecological harm, dye wastewater presents significant health risks to humans. Short-term exposure may lead to skin irritation, allergic reactions, eye damage, respiratory issues, and methemoglobinemia. Long-term exposure is even more detrimental since many dyes are highly toxic, mutagenic, teratogenic, or carcinogenic; these properties can adversely affect fetal development and cause damage to internal organs such as the kidneys and intestines [7].

Azo dyes are among the most commonly used synthetic dyes due to their low cost, high stability, and vivid coloration. One notable example is methyl orange (MO), an anionic azo dye extensively utilized in textile dyeing as well as a pH indicator in laboratory experiments. However, MO poses risks not only to human health but also to aquatic organisms because of its potential carcinogenicity and genotoxicity [8].

To address dye wastewater treatment challenges effectively, various physical, chemical, and biological technologies have been developed. These include chemical oxidation methods; electrochemical techniques; membrane separation; coagulation and sedimentation processes; photocatalytic degradation; biodegradation; and adsorption [9,10,11,12]. Among these approaches, adsorption is particularly favored for treating dye wastewater owing to its straightforward process design flexibility operational efficiency low cost controllability minimal sludge production [13,14,15]. Materials employed for dye adsorption encompass activated alumina [16], natural clays [17], resins [18], biomass [19], activated carbon [20], zeolites [21], and metal–organic frameworks (MOFs) [22,23,24]. Among these, metal–organic frameworks (MOFs) have attracted significant attention due to their unique structures. MOFs are a class of crystalline porous materials formed by the coordination of metal ions or clusters with multidentate organic ligands. They exhibit high specific surface areas, tunable pore structures, high porosity, and good thermal stability [25], making them widely applicable in catalysis [26,27], electrochemistry [28,29], energy storage [30], and adsorption [31]. Representative MOF families include ZIF, HKUST, MIL, and UiO [32]. In particular, the MIL family has attracted increasing attention because of its remarkable structural stability, unique breathing effect that allows for tunable pore channels, and abundant metal active sites, which provide distinctive advantages in pollutant adsorption and catalytic degradation. For instance, MIL-53(Fe) has been utilized for Congo red adsorption [33], and its photocatalytic efficiency can be further improved via resin modification [34]. Moreover, mesoporous composites derived from MIL-53(Al) have been developed for photocatalytic applications, and amino-functionalized MIL-101(Al) has been successfully applied for dye removal.

Despite their advantages, the practical application of MOFs is often limited by difficulties in recovery and reuse, particularly when they are in powder form. To address this issue, the integration of MOFs with magnetic materials such as Fe₃O₄ has emerged as a promising strategy. The resulting magnetic MOF composites can be efficiently separated from reaction mixtures using an external magnetic field, significantly improving their recyclability. Various synthesis approaches have been developed to fabricate such materials, including the encapsulation of pre-synthesized Fe₃O₄ nanoparticles within MOF matrices, the deposition of magnetic components via solvothermal or co-precipitation methods, layer-by-layer assembly, and in situ magnetization through electrostatic interactions [35].

In this study, a novel magnetic composite, Fe₃O₄@MIL-53(Fe Cu), was synthesized via a hydrothermal method to address the challenges associated with the recovery of powdered MOFs, aiming to reduce operational costs and broaden their practical applicability. The introduction of Fe₃O₄ nanoparticles imparts strong magnetic responsiveness, enabling facile separation, while the bimetallic MIL-53(Fe,Cu) framework provides abundant active sites for efficient adsorption of methyl orange (MO). The specific objectives of this work are to: (1) optimize key adsorption conditions including solution pH and adsorbent dosage; (2) investigate the adsorption kinetics using pseudo-first-order, pseudo-second-order, and Elovich models; (3) analyze the equilibrium data with Langmuir, Freundlich, and Dubinin–Radushkevich isotherm models; (4) determine the thermodynamic parameters of the adsorption process; and (5) evaluate the reusability and stability of Fe₃O₄@MIL-53(Fe Cu) over multiple cycles.

Compared with previous studies, this work presents a simple, rapid, eco-friendly, and efficient strategy for removing methyl orange (MO) from aqueous solutions using the magnetic Fe₃O₄@MIL-53(Fe Cu) composite. The adsorbent demonstrated excellent adsorption capacity and maintained high removal efficiency over five consecutive adsorption–desorption cycles, highlighting its robustness and reusability. These findings suggest that the proposed composite offers a sustainable and practical solution for the treatment of MO-contaminated wastewater, with potential for reducing environmental pollution.

2. Experimental Component

2.1. Reagents

Ethylene glycol (C₂H₆O₂), ferric chloride hexahydrate (FeCl₃·6H₂O, ≥90.0%), sodium acetate (CH₃COONa), terephthalic acid (C₈H₆O₄, ≥90.0%), N,N-dimethylformamide (HCON(CH₃)₂, ≥99.5%), and cupric chloride dihydrate (CuCl₂·2H₂O, ≥99.0%) were obtained from China National Pharmaceutical Group Chemical Reagent Co., Ltd. (Shanghai, China). Sodium chloride (NaCl, ≥99.5%) was supplied by Tianjin Fengchuan Chemical Reagent Technology Co., Ltd. (Tianjin, China). Sodium sulfate (Na₂SO₄, ≥99.5%) was procured from Tianjin Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Sodium carbonate (Na₂CO₃, ≥99.0%) was sourced from Chengdu Kelong Chemical Reagent Co., Ltd. (Chengdu, China). Sodium phosphate (Na₃PO₄, ≥96.0%) was provided by Tianjin Chemical Reagent Co., Ltd. (Tianjin, China). Anhydrous ethanol (C₂H₆O) was purchased from Tianjin Hengxing Chemical Reagent Manufacturing Co., Ltd. (Tianjin, China). Methyl orange (C₁₄H₁₄N₃NaO₃S) was acquired from Shanghai McLean Biochemical Technology Co., Ltd. (Shanghai, China).

2.2. Material Preparation

2.2.1. Preparation of Fe₃O₄ Nanoparticles

FeCl₃·6H₂O (4.0545 g) was dissolved in 60 mL of ethylene glycol under magnetic stirring assisted by ultrasonication. Sodium acetate (8.23 g) was then added as a stabilizer. The mixture was stirred continuously for 1 h, transferred into a Teflon-lined stainless-steel autoclave, and heated at 200 °C for 10 h. After the reaction, the system was cooled naturally to room temperature. The resulting black product, Fe₃O₄ nanoparticles, was separated by an external magnet, washed repeatedly with deionized water and anhydrous ethanol, and finally dried under vacuum at 50 °C.

2.2.2. Preparation of Fe₃O₄@MIL-53(Fe)

FeCl₃·6H₂O (4.1 g) and terephthalic acid (2.49 g) were added to a beaker containing 40 mL of DMF, followed by ultrasonication and stirring until completely dissolved. Fe₃O₄ nanoparticles (0.4 g) were then introduced and dispersed by ultrasonication for 10 min. The solution was transferred to a reactor and heated at 150 °C for 17 h. Afterward, the mixture was allowed to cool to room temperature. The resulting brown Fe₃O₄@MIL-53(Fe) composite was washed multiple times with hot ethanol and dried at 80 °C.

2.2.3. Preparation of Fe₃O₄@MIL-53(Fe Cu)

FeCl₃·6H₂O (1.3515 g), CuCl₂·2H₂O (1.310 g), and phthalic acid (2.49 g) were dissolved in 40 mL of DMF under ultrasonication and stirring. Fe₃O₄ nanoparticles (0.4 g) were then added and dispersed by ultrasonication for 10 min. The solution was transferred to a reactor and heated at 150 °C for 17 h. After cooling to room temperature, the brown Fe3O4@MIL-53(Fe) composite was washed with hot ethanol and dried.

2.3. Test and Characterization

The morphology of Fe₃O₄@MIL-53(Fe Cu) was characterized using scanning electron microscopy (SEM, JSM-6610LV, JEOL, Tokyo, Japan). The crystal composition of the material was analyzed by X-ray diffraction (XRD, SmartLab 9, Rigaku, Tokyo, Japan). The chemical valence states on the material surface were investigated using X-ray photoelectron spectroscopy (XPS, SmartLab 9, Rigaku, Tokyo, Japan). The specific surface area of the adsorption samples was calculated using the Brunauer–Emmett–Teller (BET) method. Fourier Transform Infrared Spectroscopy (FTIR, Nicolet iS20, Thermo Fisher Scientific, Waltham, MA, USA) was employed to qualitatively identify the functional groups on the material surface. The light absorption properties of the product were measured by UV–Vis diffuse reflectance spectroscopy (UV–Vis DRS, TU-1901, Persee General Instrument Co., Ltd., Beijing, China).

2.4. Adsorption Tests

The adsorption experiments were performed using a water-bath thermostatic shaker maintained at 30 °C and 150 rpm. The initial pH of the MO solution was adjusted with 0.1 M NaOH or H₂SO₄. In each experiment, 20 mg of Fe₃O₄@MIL-53(Fe Cu) was added to 20 mL of MO solution (initial concentration: 200 mg/L) in a series of sample vials. The vials were then agitated in the shaker at 150 rpm for 4 h. At predetermined time intervals, aliquots were withdrawn, filtered through a 0.45 μm membrane, and the residual MO concentration was determined by measuring the absorbance at 464 nm using a UV–Vis spectrophotometer.

Equation (1) is used to calculate the MO removal rate (removal efficiency), as follows:

$$R={\displaystyle \frac{C_o-C_e}{C_o}}$$

(1)

where the MO concentrations before and at reaction times t (min or h) are denoted by the numbers $C_o$ (mg/L) and $C_e$ (mg/L).

The removal capacity of the material for MO is calculated according to Equation (2), as follows:

$$q_e={\displaystyle \frac{(C_0-C_e)\times V}{m}}$$

(2)

where $q_e$ (mg/g) is the amount of MO adsorbed at time t, $m$ (g) is the mass of the adsorbent, and V (L) is the volume of the MO solution.

2.5. Recyclability Analysis Test of Adsorbent

After the adsorption experiment, the solid product containing adsorbed MO was soaked in anhydrous ethanol and separated using magnetic separation. The collected material was then dried in a vacuum drying oven to obtain the regenerated product. The removal performance of Fe₃O₄@MIL-53(Fe Cu) for MO solution (initial concentration: 200 mg/L) was evaluated after five adsorption–desorption cycles.

2.6. Adsorption Kinetics

Kinetic studies provide insights into the characteristics of the adsorption process. The MO adsorption on the adsorbent was investigated using 20 mg of adsorbent and 20 mL of 200 mg/L MO solution. The mixture was shaken at 150 rpm at room temperature under pH 3 conditions. Samples were taken at 5, 10, 20, 30, 60, 90, 120, 180, 240, and 300 min, and the concentration of MO was determined using a UV–Vis spectrophotometer. To evaluate the adsorption process of MO on the adsorbent, the pseudo-first-order kinetic model, pseudo-second-order kinetic model, and Elovich kinetic model were applied, based on Equations (3), (4) and (5), respectively.

$$ln\left(q_e-q_t\right)={lnq}_e-k_{1}t$$

(3)

$$q_t=q_e-{\displaystyle \frac{q_e}{\left(k_2{q}_{e}t+1\right)}}$$

(4)

$$q_t={\displaystyle \frac{ln\left(\alpha \beta \right)}{\beta }}+{\displaystyle \frac{lnt}{\beta }}$$

(5)

where$k_1$ (min⁻¹) is the pseudo-first-order kinetic rate constant, $k_2$ (g·mg⁻¹·min⁻¹) is the rate constant of pseudo-second-order kinetic model, αis the initial adsorption rate constant (mg·g⁻¹·min⁻¹), $\beta$ is the desorption constant (g·mg⁻¹).

2.7. Adsorption Isotherms

The adsorption isotherm represents the non-kinetic relationship between the adsorption capacity at equilibrium $(q_e$) and the MO concentration at equilibrium ($C_e$). The adsorption isotherms were determined for 20 mg of adsorbent by varying the MO concentration across 100, 150, 200, 250, 300, 400, and 500 mg/L. Vials containing 20 mL of solution were stirred at 150 rpm for 4 h at pH = 3. The adsorption data were fitted using isotherm models—Langmuir and Freundlich—by nonlinear regression analysis using Excel 2019. The Dubinin–Radushkevich (D-R) model was analyzed using linear fitting. The corresponding model equations are shown in Equations (6)–(8). Equation (9) is the formula for calculating adsorption energy.

$$q_e={\displaystyle \frac{q_{max}{K}_L{C}_e}{1+K_L{C}_e}}$$

(6)

$$q_e=K_F+{C_e}^{{\displaystyle \frac{1}{n}}}$$

(7)

$${lnq}_e={lnq}_{max}-K_{DR}{\epsilon }^2;\epsilon =RTln\left(1+{\displaystyle \frac{1}{C_e}}\right)$$

(8)

$$E={\left(2K_{DR}\right)}^{-0.5}$$

(9)

where $C_e$ (mg/L) is the equilibrium concentration of the adsorbate; $q_{max}$ (mg/g) is the theoretical maximum adsorption capacity; $K_{DR}$ (mol²/kJ²) is the Dubinin–Radushkevich constant related to adsorption energy; $\epsilon$ (kJ/mol) is the Polanyi potential, where $R$ is the gas constant (8.314 J/mol·K), and $T$ is the absolute temperature in Kelvin. $E$ (kJ/mol) is the mean free energy of adsorption, which is used to determine the nature of the adsorption mechanism. Specifically, $E$ < 8 kJ/mol suggests physical adsorption, 8 ≤ $E$ ≤ 16 kJ/mol indicates ion exchange, and $E$ > 16 kJ/mol corresponds to chemisorption.

2.8. Thermodynamic Study

Thermodynamic parameters, namely, Gibbs free energy ($\Delta G$), enthalpy ($\Delta H$), and entropy ($\Delta S$), are typically used to describe the adsorption process. The thermodynamic parameters were calculated using the following equations:

$$K_p={\displaystyle \frac{C_0-C_e}{C_e}}\times {\displaystyle \frac{V}{m}}$$

(10)

$$ln\left(K_p\right)={\displaystyle \frac{\Delta S}{R}}-{\displaystyle \frac{\Delta H}{RT}}$$

(11)

$$\Delta G=\Delta H-\Delta S·T$$

(12)

where $\Delta G$ (kJ/mol) is the Gibbs free energy change, $\Delta H$ (kJ/mol) is the standard enthalpy change, $T$ (K) is the absolute temperature, $\Delta S$ (J/mol ·K) is the standard entropy change, $K_p$ (L/g) is the solid–liquid distribution coefficient, $R$ = 8.314 J/mol·K is the universal gas constant, $C_0$ (mg/L) is the initial concentration of MO, $C_e$ (mg/L) is the equilibrium concentration of MO in solution, $V$ (mL) is the volume of the solution, and $m$ (g) is the mass of adsorbent used.

Based on the calculated data obtained from the adsorption isotherm experiments, the value of $K_p$ was first calculated using Equation (10), and then ln $K_p$ was derived. A linear fitting was performed by plotting ln $K_p$ as the y-axis and 1/T as the x-axis. From the slope and intercept of the straight line, the values of $\Delta H$ and $\Delta S$ were calculated using Equation (11), and subsequently, ∆G was calculated using Equation (12).

3. Results

3.1. Material Characterization

3.1.1. SEM

Figure 1a–c presents the surface morphologies of Fe₃O₄@MIL-53(Fe), MIL-53(Fe Cu) and Fe₃O₄@MIL-53(Fe Cu), respectively, as observed by scanning electron microscopy (SEM). As shown in Figure 1a,c, Fe₃O₄@MIL-53(Fe) exhibits a typical octahedral morphology [36], with uniform particle size. After the introduction of copper ions, the crystal structure of MIL-53(Fe Cu) maintains the same octahedral framework, indicating that the incorporation of Cu does not disrupt the overall crystal architecture. However, the doping of bimetallic ions results in an elongation of the crystal body, likely due to the competitive coordination between Cu²⁺ and Fe³⁺ during the formation of MOF precursors, which promotes the extension of specific crystal planes [37], thereby increasing the specific surface area of the crystals. A comparison between Figure 1b,c reveals that, in addition to the original structure in Figure 1b, small spherical particles are observed on the surface of Figure 1c. This clearly indicates that Fe₃O₄ nanoparticles have been successfully loaded onto the surface of MIL-53(Fe Cu) [38].

3.1.2. XRD

Figure 2 displays the XRD patterns of Fe₃O₄, Fe₃O₄@MIL-53(Fe), and a series of Fe₃O₄@MIL-53(Fe Cu) composites with systematically varied Fe/Cu molar ratios (denoted collectively as Cu/CuS/BC). Rietveld-refinable diffraction features confirm the preservation of a magnetite phase (Fe₃O₄; PDF# 19–0629) across all composites, with characteristic reflections at 2θ = 30.1°, 35.5°, 43.1°, 55.0°, 56.9°, and 62.6°, consistent with reference data [39]. The coexistence of these reflexes with the MIL-53-type signatures confirms the successful formation of heterostructures without degradation of either component.

The MIL-53(Fe) framework exhibits characteristic Bragg peaks at 2θ = 9.4°, 12.5°, 16.2°, 18.8°, and 25.6°, in agreement with its known crystalline architecture [40]. Incorporation of Cu²⁺ induces a symmetry-lowering distortion, most notably evidenced by the splitting of the (011) peak at 9.4° into two discrete reflections at 9.2° and 10.8°. This phenomenon is indicative of unit cell expansion and reduced local symmetry, attributable to the larger ionic radius of Cu²⁺ (0.73 Å) compared to high-spin Fe³⁺ (0.645 Å) [41], and the consequent competitive metal–ligand coordination during crystallization. Such structural perturbation is consistent with heterometallic cluster formation and may facilitate enhanced surface accessibility [42].

Systematic variation in the Fe/Cu ratio induces progressive intensity modulation between the split peaks. Increasing Cu content suppresses the reflection at 9.2° while enhancing the 10.8° component, suggesting Cu-driven phase evolution and preferential crystallographic orientation. At the highest doping level (Fe:Cu = 1:5), attenuation and broadening of all reflections point to partial loss of long-range order and increased structural disorder, likely due to lattice strain and the formation of defective metal–oxo clusters [43]. This correlates well with the observed increase in specific surface area and suggests amorphization near saturation doping.

3.1.3. BET

Figure 3 presents the N₂ adsorption–desorption isotherms of Fe₃O₄@MIL-53(Fe) and Fe₃O₄@MIL-53(Fe Cu), while

Table 1. Surface area and pore structure of adsorbent.

Physical Property	Fe3O4@MIL-53(Fe)	Fe3O4@MIL-53(Fe Cu)
Specific Surface Area (m2/g)	6.770	46.109
Average Pore Fiameter (nm)	9.174	10.352
Micropore Volume (cm3/g)	0.0008	0.020
Pore Volume (cm3/g)	0.015	0.052

. summarizes the specific surface area parameters of the corresponding adsorbents. As shown in the figure, the pore diameters of the two materials are 9.174 nm and 10.352 nm, respectively—both within the mesoporous range of 2–10 nm—indicating that the materials possess mesoporous structures. The isotherms of both samples are characteristic of type IV curves with H3-type hysteresis loops, further confirming the presence of mesopores within the materials’ frameworks [44]. The relatively low specific surface area of Fe₃O₄@MIL-53(Fe) may be attributed to the “breathing effect” and the presence of inaccessible pores [45]. In contrast, Fe₃O₄@MIL-53(Fe Cu) exhibits a significantly enhanced specific surface area of 46.109 m²/g and a micropore volume of 0.020 cm³/g, compared to the 6.770 m²/g specific surface area and 0.0008 cm³/g micropore volume of Fe₃O₄@MIL-53(Fe). These results indicate an approximate six-fold increase in surface area and a 96% increase in micropore volume following bimetallic incorporation, thereby providing a greater number of exposed active sites for adsorption.

3.1.4. FTIR

The FT-IR spectra of Fe₃O₄@MIL-53(Fe), MIL-53(Fe Cu), and Fe₃O₄@MIL-53(Fe Cu) composites are presented in Figure 4, revealing consistent functional group characteristics and coordination environments across all materials. The spectral similarity suggests that the incorporation of copper does not alter the fundamental chemical structure of the MIL-53 framework.

Key vibrational assignments are as follows: the broad absorption band at 3430 cm⁻¹ corresponds to O–H stretching vibrations from adsorbed water molecules [46]. The peak observed at 1674 cm⁻¹ is associated with C=O stretching vibrations [47]. The asymmetric and symmetric stretching vibrations of carboxylate groups (COO⁻) are identified at 1589 cm⁻¹ and 1381 cm⁻¹, respectively [48]. The peak separation (Δν) of approximately 208 cm⁻¹ between these modes indicates bidentate coordination between carboxylate oxygen atoms and metal centers. The feature at 742 cm⁻¹ is attributed to aromatic C–H bending vibrations, while the band at 557 cm⁻¹ corresponds to Fe–O stretching vibrations, confirming metal-linker bonding within the framework [34].The preservation of these spectral features across all samples demonstrates that the introduction of Cu(II) ions maintains the structural integrity of the MIL-53 architecture, suggesting successful isomorphous substitution without disrupting the primary coordination network.

3.1.5. XPS

The XPS survey spectrum of Fe₃O₄@MIL-53(Fe Cu) confirms the presence of C (285.1 eV), O (532.3 eV), Fe (712.2 eV), and Cu (953.5 eV). The high-resolution C 1s spectrum (Figure 5a) exhibits three peaks corresponding to C–C (284.8 eV), C–O (286.0 eV), and O–C=O (288.7 eV), indicative of aromatic linkers and metal-carboxylate coordination. The O 1s spectrum (Figure 5b) displays contributions from metal–oxygen bonds (530.1 eV), adsorbed water (531.8 eV), and carbonyl groups (533.2 eV). These oxygen species promote hydrophilic interactions and offer active sites for pollutant adsorption. The well-defined chemical states support the structural integrity of the bimetallic system and suggest enhanced adsorption capability through synergistic effects [49].

3.1.6. Zeta Potential Analysis

The pH value significantly influences the adsorption performance of Fe₃O₄@MIL-53(Fe Cu), with the point of zero charge (pHpzc) serving as a key indicator of its surface charge characteristics. As depicted in Figure 6, the zeta potential of Fe₃O₄@MIL-53(Fe Cu) varies across a pH range of 2 to 12. The pHpzc is determined to be 6.0. Under acidic conditions (pH < 6.0), the adsorbent surface carries a positive charge, whereas under line conditions (pH > 6.0), it becomes negatively charged. At pH = pHpzc, the surface is electrically neutral. This charge transition plays a critical role in the electrostatic interactions between the adsorbent and target contaminants, thereby affecting adsorption efficiency.

3.2. Adsorption Properties

3.2.1. Fe₃O₄@MIL-53(Fe Cu) Dosage

The effect of Fe₃O₄@MIL-53(Fe Cu) dosage on the adsorption of methyl orange (MO) is shown in Figure 7. The results indicate that the MO removal rate increases with increasing adsorbent dosage, while the adsorption capacity decreases. This is because the number of available adsorption sites increases with the Fe₃O₄@MIL-53(Fe Cu) dosage, enhancing the potential for MO adsorption. However, at higher dosages, the concentration of MO becomes relatively low, leading to an excess of adsorption sites that exceed the adsorbent’s capacity, resulting in underutilization and an unsaturated state. As the adsorbent dosage increases from 5 mg to 30 mg, the removal efficiency of MO increases from 52% to 96%. Further increases in dosage do not significantly affect the removal efficiency, indicating adsorption saturation, where most of the MO in the solution is adsorbed. Considering both the economic factors and removal efficiency, the optimal dosage of adsorbent is 20 mg, achieving a removal rate of over 85%.

3.2.2. pH

The influence of pH on the adsorption of methyl orange (MO) is shown in Figure 8. The results indicate that the optimal pH for MO adsorption by Fe₃O₄@MIL-53(Fe Cu) is 3, with the maximum adsorption capacity reaching 142 mg/g. As pH increases from 4 to 11, the removal efficiency decreases by 34%, and the adsorption capacity drops from 140 mg/g to 105 mg/g. This phenomenon is attributed to the electrostatic interactions between MO, an anionic dye, and the adsorbent, where negatively charged oxidative functional groups on the adsorbent surface reduce adsorption effectiveness [49]. According to the zeta potential data, the isoelectric point of Fe₃O₄@MIL-53(Fe Cu) is 6. At pH = 3, a positive charge forms on the adsorbent surface, promoting electrostatic attraction of MO. However, when the pH exceeds the isoelectric point, the adsorbent surface becomes negatively charged, repelling MO and promoting the formation of hydroxide ions, which inhibits MO adsorption [50].

3.2.3. Co-Existing Anions

As shown in Figure 9, the presence of four coexisting ions—Cl⁻, SO₄²⁻, CO₃²⁻, and PO₄³⁻—inhibits the adsorption of MO. The inhibitory effect intensifies with increasing ion concentration. In the presence of Cl⁻ ions, the adsorption capacity decreases from 158.37 mg/g to 156.02 mg/g; with SO₄²⁻, the capacity decreases by 3 mg/g; in the presence of PO₄³⁻, the capacity drops from 158.37 mg/g to 155.03 mg/g; and with CO₃²⁻, the capacity decreases by 2 mg/g. Overall, however, the impact of these ions on MO adsorption by Fe₃O₄@MIL-53(Fe Cu) is minimal.

This inhibitory effect may arise because methyl orange (MO), as an anionic dye, primarily binds to the positively charged sites on the composite material surface via electrostatic attraction under acidic conditions. Coexisting inorganic anions (e.g., Cl⁻, SO₄²⁻, CO₃²⁻, PO₄³⁻), which also carry negative charges, compete with MO for the positively charged adsorption sites, thereby reducing the opportunities for electrostatic binding between MO and the adsorbent and resulting in decreased adsorption efficiency. Experimental data indicate that with increasing anion concentration, this competitive effect intensifies, leading to a decline in adsorption capacity (e.g., in the presence of Cl⁻, adsorption capacity decreased from 158.373 mg/g to 156.023 mg/g). Functional groups on the composite surface, such as hydroxyl (-OH) and carboxyl (-COOH), form hydrogen bonds with MO; however, some inorganic anions (e.g., PO₄³⁻, CO₃²⁻) may also interact with these functional groups—through hydrogen bonding or coordination bonding—occupying hydrogen bonding sites and weakening the hydrogen bond interactions between the composite and MO, further inhibiting adsorption. Nonetheless, this inhibitory effect is generally weak, possibly because the affinity of the inorganic anions for the functional groups is lower than that of MO, thus limiting the extent of interference with adsorption.

3.2.4. Recyclability Analysis

The regeneration performance of the adsorbent is crucial for its practical application. In this study, anhydrous ethanol was used to regenerate the adsorbent after methyl orange adsorption. The adsorbent, after adsorption, was soaked in anhydrous ethanol and separated using magnetic attraction. After vacuum drying, the regenerated solid was collected. The cyclic adsorption experiments were conducted under the following conditions: T = 30 °C, initial methyl orange concentration = 200 mg/L, pH = 3, adsorbent dosage = 20 mg, and adsorption time = 4 h.

As shown in Figure 10, after five desorption-adsorption cycles, the removal efficiencies of MO were 94.52%, 90.72%, 87.10%, 81.77%, and 77.15%, respectively. The high removal efficiency observed in the fifth cycle indicates that anhydrous ethanol is effective in desorbing MO from Fe₃O₄@MIL-53(Fe Cu), thereby regenerating the active adsorption sites for subsequent cycles. These results demonstrate that Fe₃O₄@MIL-53(Fe Cu) exhibits excellent reusability, making it a promising and cost-effective material for wastewater treatment.

3.2.5. Adsorption Kinetics Analysis

Based on the optimal conditions identified in the previous experiments (pH = 3, adsorbent dosage = 20 mg), adsorption kinetic experiments were conducted with an initial methyl orange concentration of 200 mg/L at T = 30 °C. Samples were collected at 5, 10, 20, 30, 60, 90, 120, 180, 240, and 300 min, respectively, and the adsorption capacity was subsequently calculated.

Figure 11 shows the results of the three kinetic models. The related parameters are listed in

Table 2. Parameters of adsorption kinetics model for Fe₃O₄@MIL-53(Fe Cu) adsorption of MO.

Pseudo-First-Order	Pseudo-Second-Order	Elovich
$q_e$ = 143.566 mg/g	$q_e$ = 148.425 mg/g	$\alpha$ = 1,264,413.579 mg/mg.min
$k_1$ = 0.256 min−1	$k_2$ = 0.004 g·mg−1·min−1	$\beta$ = 0.115 g/mg
$R^2$ = 0.980	$R^2$ = 0.998	$R^2$ = 0.989

.

Contact time plays a crucial role in the adsorption process. As illustrated in Figure 11, the adsorption capacity for MO increases over time. The adsorption is rapid during the first 30 min, with the capacity rising by nearly 90% and the removal efficiency surpassing 90%. Following this, the rate of adsorption slows down, eventually reaching equilibrium. The rapid uptake in the initial phase can be attributed to the abundant active sites on Fe₃O₄@MIL-53(Fe Cu), which facilitate the transfer of a large amount of MO from the surface into the interior of the adsorbent. In the latter phase, adsorption is primarily influenced by the pore size of Fe₃O₄@MIL-53(Fe Cu).

As shown in Figure 11 and Table 2, the determination coefficient(R²) for the Pseudo-first-order, Pseudo-second-order, and Elovich kinetic models were 0.980, 0.998, and 0.989, respectively. The experimental data were well fitted by all three models. The observed adsorption capacity of Fe₃O₄@MIL-53(Fe Cu) was 150.427 mg/g, which closely matches the theoretical capacity of 148.425 mg/g predicted by the pseudo-second-order model. This indicates that the adsorption of methyl orange onto Fe₃O₄@MIL-53(Fe Cu) involves chemisorption [51].

3.2.6. Adsorption Isotherms Analysis

Based on the optimal conditions obtained from previous experiments (pH = 3, adsorbent dosage = 20 mg), isothermal adsorption experiments were conducted using initial concentrations of methyl orange at 100, 150, 200, 250, 300, 400, and 500 mg/L. The samples were shaken on a constant temperature shaker at three different temperatures (298.15 K, 308.15 K, and 318.15 K) for 4 h. The experimental data were then fitted using three adsorption isotherm models: Langmuir, Freundlich, and Dubinin–Radushkevich (D-R).

The nonlinear fitting results are illustrated in Figure 12, with the corresponding parameters provided in

Table 3. Langmuir, Freundlich, and Dubinin–Radushkevich model fitting parameters for adsorption of methyl orange.

T(K)	Langmuir	Freundlich	Dubinin–Radushkevich
298.15	$q_{max}$ = 154.417 mg/g $K_L$ = 0.237 L/mg $R^2$ = 0.889	$K_F$ = 59.452 mg/g ${\displaystyle \frac{1}{n}}$ = 0.202 $R^2$ = 0.973	$\beta$ = 7.353 × 10−7 mol2/J2
			$q_s$ = 4.908 mg/g
			$E$ = 0.825 kJ/mol
			$R^2$ = 0.675
308.15	$q_{max}$ = 172.500 mg/g $K_L$ = 0.105 L/mg $R^2$ = 0.915	$K_F$ = 68.553 mg/g ${\displaystyle \frac{1}{n}}$ = 0.199 $R^2$ = 0.991	$\beta$ = 4.818 × 10−8 mol2/J2
			$q_s$ = 5.008 mg/g
			$E$ = 1.019 kJ/mol
			$R^2$ = 0.634
318.15	$q_{max}$ = 193.647 mg/g $K_L$ = 0.113 L/mg $R^2$ = 0.899	$K_F$ = 73.429 mg/g ${\displaystyle \frac{1}{n}}$ = 0.215 $R^2$ = 0.998	$\beta$ = 2.432 × 10−8 mol2/J2
			$q_s$ = 5.080 mg/g
			$E$ = 1.434 kJ/mol
			$R^2$ = 0.585

. As shown in Figure 12 and Table 3, both the Langmuir and Freundlich models exhibit a good fit for the adsorption data. At temperatures of 298.15 K, 308.15 K, and 318.15 K, the determination coefficient (R²) for the Langmuir model are 0.889, 0.915, and 0.899, respectively, whereas the R² values for the Freundlich model are 0.973, 0.991, and 0.998. These results suggest that the adsorption process aligns more closely with the Freundlich isotherm, which is indicative of a heterogeneous adsorption mechanism.

In the Langmuir model, as shown in Table 4, the theoretical maximum adsorption capacity (qmax) increases with temperature, reaching values of 154.417 mg/g, 172.500 mg/g, and 193.647 mg/g at 298.15 K, 308.15 K, and 318.15 K, respectively. This indicates that the adsorption process is endothermic. Additionally, the value of 1/n provides insight into the adsorption favorability. When 1/n falls between 0.1 and 0.5, the adsorption is considered favorable, whereas values exceeding 1 suggest unfavorable adsorption. At all three temperatures, the values of 1/n are 0.202, 0.199, and 0.215, which confirms that the adsorption process is favorable under the given conditions.

In the Freundlich model, the KF parameter reflects the adsorption intensity, with higher values indicating a stronger adsorption capacity of the adsorbent [52]. As shown in Table 3, with an increase in temperature, the KF value rises from 59.452 mg/g at 298.15 K to 68.553 mg/g at 308.15 K, and further to 73.429 mg/g at 318.15 K. This trend suggests that elevated temperatures enhance the adsorption of methyl orange (MO) by Fe₃O₄@MIL-53(Fe Cu).

According to the parameters derived from the D-R model, the adsorption energies at temperatures of 298.15 K, 308.15 K, and 318.15 K are 0.825, 1.019, and 1.434 kJ/mol, respectively. The results indicate that MO adsorption by Fe₃O₄@MIL-53(Fe Cu) is predominantly governed by physical interactions.

Table 4 presents a comparison of the adsorption capacities of various adsorbent materials for methyl orange. As shown, Fe₃O₄@MIL-53(Fe Cu) demonstrates a superior adsorption capacity compared to other materials.

Table 4. Comparison of adsorption capacity of methyl orange by different adsorbent materials.

Makings	Adsorption Objects	pH	Adsorption Capacity/mg g−1	Bibliography
Ni-Fe-SO4 LDH	MO	4	82.45	[53]
ZIF-67		/	75.59	[54]
Ni@ZIF-67		6	151.74	[55]
UiO-66		5.90	84.8	[56]
UiO-66/cellulose aerogels		/	71.70	[57]
MIL-53(Al)@CWS		/	99.39	[58]
magnetic lignin-based carbon nanoparticles		5	113	[59]
Fe3O4@MIL-53(Fe Cu)		3	193.68	this study

Table 4. Comparison of adsorption capacity of methyl orange by different adsorbent materials.

Makings	Adsorption Objects	pH	Adsorption Capacity/mg g−1	Bibliography
Ni-Fe-SO4 LDH	MO	4	82.45	[53]
ZIF-67		/	75.59	[54]
Ni@ZIF-67		6	151.74	[55]
UiO-66		5.90	84.8	[56]
UiO-66/cellulose aerogels		/	71.70	[57]
MIL-53(Al)@CWS		/	99.39	[58]
magnetic lignin-based carbon nanoparticles		5	113	[59]
Fe3O4@MIL-53(Fe Cu)		3	193.68	this study

3.2.7. Adsorption Thermodynamics Analysis

To further investigate the adsorption performance of methyl orange on Fe₃O₄@MIL-53(Fe Cu), thermodynamic calculations were performed. The results are summarized in

Table 5. Thermodynamic parameters of adsorption of methyl orange.

T(K)	ΔS (kJ/mol)	ΔH (kJ/mol)	ΔG (kJ/mol)
298.15	0.047	13.886	−0.025
308.15			−0.491
318.15			−0.958

and depicted in Figure 13.

As shown in Figure 13, the linear determination coefficient(R² = 0.942) indicates a strong fit to the thermodynamic equation. From Table 5, the positive enthalpy change (ΔH = 13.886 kJ/mol) suggests that the adsorption process is endothermic. The positive entropy change (ΔS = 0.047 kJ/mol·K) indicates an increase in disorder and enhanced mobility at the solid–liquid interface during the adsorption process. Furthermore, at all three temperatures, the Gibbs free energy (ΔG) values are negative and decrease with increasing temperature, signifying that the adsorption of MO by Fe₃O₄@MIL-53(Fe Cu) is spontaneous and becomes more favorable at higher temperatures.

3.3. Adsorption Mechanism Study

3.3.1. SEM Analysis

Figure 14 shows the SEM images of Fe₃O₄@MIL-53(Fe Cu) before and after the adsorption of MO. Prior to adsorption, the material exhibits a distinct octahedral prismatic crystal structure. After adsorption, the surface of the Fe₃O₄@MIL-53(Fe Cu) becomes significantly rougher and more irregular, with the previously smooth surfaces giving way to a more heterogeneous morphology. This alteration suggests that MO molecules are adsorbed onto the crystal surface, likely due to interactions at the surface folds, leading to the formation of blocky crystal aggregates. Additionally, Fe₃O₄ spheres are dispersed across the interstitial spaces, and the material retains its magnetic properties post-adsorption.

3.3.2. FTIR Analysis

Figure 15 presents the Fourier transform infrared (FTIR) spectra of Fe₃O₄@MIL-53(Fe Cu) before and after the adsorption of methyl orange (MO), measured over the wavenumber range of 4000–500 cm⁻¹. The FTIR analysis reveals a new characteristic peak at 1155 cm⁻¹, which can be attributed to the symmetric stretching vibration of the sulfonate group (–SO₃⁻) in MO, indicating the successful adsorption of MO onto Fe₃O₄@MIL-53(Fe Cu) [60]. Additionally, a redshift (−12 cm⁻¹) is observed at 3398 cm⁻¹ accompanied by an increase in intensity, while a blueshift (+24 cm⁻¹) appears at 1659 cm⁻¹ along with a decrease in intensity, suggesting the involvement of hydroxyl (−OH) and carbonyl (C=O) functional groups in the adsorption process. Furthermore, the intensities of the peaks at 745 cm⁻¹ (aromatic C–H out-of-plane bending) and 548 cm⁻¹ (Fe–O stretching vibration) are weakened, indicating that π–π interactions between the aromatic rings of MO and the adsorbent, as well as coordination to metal sites, play important roles in the adsorption mechanism [61]. The post-adsorption spectra also display other characteristic vibrations of MO, including aromatic C–H in-plane bending (1000–1100 cm⁻¹), aromatic C=C stretching (1480–1600 cm⁻¹), N=N azo group stretching (1400–1500 cm⁻¹), and asymmetric –SO₃⁻ stretching (1030–1050 cm⁻¹), further confirming the adsorption of MO and its associated interaction mechanisms [62].

3.3.3. XPS Analysis

X-ray photoelectron spectroscopy (XPS) was employed to investigate the interaction between methyl orange (MO) and Fe₃O₄@MIL-53(Fe Cu). The results are illustrated in Figure 16.

Figure 16a presents the full XPS spectra before and after the adsorption of MO onto Fe₃O₄@MIL-53(Fe Cu), showing a negligible shift in the overall binding energy. In Figure 16b, a comparison of the C 1s spectra before and after adsorption reveals notable changes in the binding energies of the characteristic bonds. Specifically, the binding energies for C–C, C–O, and O–C=O are 284.7 eV, 285.3 eV, and 288.8 eV, respectively, following adsorption [62]. These binding energies exhibit a shift towards lower values compared to the pre-adsorption state, accompanied by an increase in peak intensities. This suggests that the adsorption process induces the cleavage and rearrangement of C–C and C–O bonds, indicating a direct interaction between MO and these functional groups during the adsorption process [63].

Figure 16c displays the comparison of the O 1s spectra before and after the adsorption of MO. The binding energies of the metal-oxygen bonds, organic C=O groups, and H-O bonds shift to 529.9 eV, 531.8 eV, and 533.6 eV, respectively. This indicates that hydroxyl groups interact with MO through hydrogen bonding during the adsorption process. Figure 16d,e show the comparison of the Fe 2p and Cu 2p spectra before and after adsorption. It can be observed that both the binding energies and characteristic peak positions change, suggesting that Fe and Cu metal ions play an active role in the adsorption process, likely by facilitating the activation of MO [64].

3.3.4. Adsorption Mechanism

The adsorption mechanism of methyl orange (MO) onto the bimetallic Fe₃O₄@MIL-53(Fe Cu) composite was systematically elucidated through integrated experimental analyses and detailed material characterization. The results demonstrate that the adsorption process is governed by a synergistic interplay of multiple mechanisms, including electrostatic attraction, hydrogen bonding, π–π interactions, and mesopore filling.

Zeta potential measurements determined the material’s isoelectric point at pH 6.0. At the optimal adsorption pH of 3, the surface carries a positive charge, promoting electrostatic attraction with the negatively charged sulfonate groups (-SO₃⁻) of MO. Adsorption efficiency decreased significantly at pH values above 6, where the surface charge became negative, causing repulsion and a drop in adsorption capacity from 140 mg/g to 105 mg/g, confirming the crucial role of electrostatics.

Hydrogen bonding was evidenced by FTIR spectral shifts in surface hydroxyl (-OH) and carboxyl (-COOH) groups after MO adsorption, indicating interaction with MO’s oxygen and amino groups. XPS analysis further supported these interactions by showing changes in oxygen-related bonding energies.

π-π stacking interactions also contributed, as the aromatic rings of the MOF’s terephthalic acid linker and the conjugated azo-benzene system of MO facilitated electron cloud overlap. XPS carbon spectra showed binding energy shifts consistent with charge redistribution from these π-π interactions.

Mesoporous structure played an additional role; BET analysis revealed a high surface area (46.109 m²/g) and average pore size (10.35 nm), providing ample space for MO diffusion and physical entrapment. SEM imaging confirmed morphological changes consistent with pore filling after adsorption.

Kinetic analysis indicated that the adsorption followed a pseudo-second-order model, suggesting surface site-controlled physisorption, while the Freundlich isotherm model indicated heterogeneous adsorption consistent with multiple binding sites. Thermodynamics revealed an endothermic and spontaneous process, with increased temperature enhancing adsorption capacity.

In conclusion, the efficient adsorption of methyl orange by Fe₃O₄@MIL-53(Fe Cu) is attributed to the combined effects of electrostatic attraction, hydrogen bonding, π-π stacking, and mesopore filling, highlighting the material’s potential in dye wastewater treatment, see Figure 17.

4. Discussion and Conclusions

Fe₃O₄@MIL-53(Fe Cu) exhibited outstanding adsorption performance for methyl orange (MO), achieving a maximum removal efficiency of 97% at an optimal pH of 3 and a high adsorption capacity of 142 mg/g. The adsorption process was strongly pH-dependent, with significantly enhanced performance observed under acidic conditions. Among common coexisting anions (Cl⁻, SO₄²⁻, CO₃²⁻, and PO₄³⁻), only minor inhibitory effects were observed, with SO₄²⁻ showing the most pronounced influence. Importantly, after five consecutive regeneration cycles using anhydrous ethanol, the composite retained a high removal efficiency of 77.15%, demonstrating its excellent reusability and potential for practical applications in water treatment.

Kinetic analysis indicated that the adsorption process conformed to the pseudo-second-order model, suggesting that physisorption may be the rate-limiting step. The adsorption isotherm was best fitted by the Freundlich model, implying multilayer adsorption on a heterogeneous surface. Thermodynamic studies further revealed that the adsorption process was spontaneous, and the adsorption capacity increased with rising temperature, indicating an endothermic nature.

SEM, FTIR, and XPS analyses confirmed the successful adsorption of MO, which was primarily facilitated through π–π interactions, intermolecular hydrogen bonding, and electrostatic attraction. The composite Fe₃O₄@MIL-53(Fe Cu) demonstrates strong potential for application in wastewater treatment, offering a balance of high adsorption performance and favorable recyclability.