Sustainable Strategy for Microplastic Mitigation: Fe₃O₄ Acid-Functionalized Magnetic Nanoparticles for Microplastics Removal

Abstract

Microplastic (MPs) pollution has emerged as a critical environmental issue due to its persistent accumulation in ecosystems, posing risks to aquatic life, food safety, and human health. In this study, magnetic Fe₃O₄ nanoparticles functionalized with citric acid (Fe₃O₄@AC) were used to remove high-density polyethylene (HDPE), low-density polyethylene (LDPE), and polypropylene (PP) MPs from an aqueous medium. Fe₃O₄@AC was synthesized via the coprecipitation method and characterized by morphology (SEM), crystalline phases (XRD), chemical aspects (FTIR), and surface area (nitrogen sorption isotherms). The MPs removal efficiency of Fe₃O₄@AC was evaluated based on the initial concentration, contact time, and pH. The adsorption isotherm and kinetics data were best described by the Sips and pseudo-second-order models, respectively. Fe₃O₄@AC removed 80% of the MPs at a pH of 6. Based on experimental observations (zeta potential, porosity, and SEM) and theoretical insights, it was concluded that hydrogen bonding, pore filling, and van der Waals forces governed the adsorption mechanism. Reusability tests showed that Fe₃O₄@AC could be reused up to five times, with a removal efficiency above 50%. These findings suggest that Fe₃O₄@AC is a sustainable and promising material for the efficient removal of microplastics from wastewater, offering a reusable and low-impact alternative that contributes to environmentally responsible wastewater treatment strategies.

1. Introduction

Plastic waste pollution has garnered increasing attention over the past decade due to the widespread consumption and inadequate management of plastic materials, resulting in substantial accumulation of plastic particles in the environment. Microplastics (MPs) are emerging pollutants found in various environments on Earth [1,2,3,4,5]. Globally, MPs are defined as synthetic solid particles or polymeric matrices that are insoluble in water, possess regular or irregular shapes and sizes ranging from 1 μm to 5 mm, and originate from primary or secondary manufacturing processes [6]. Primary MPs are intentionally manufactured in microscopic sizes and used as raw materials in the production of textiles and personal care products, such as facial and body scrubs that contain microspheres, fibers, and pellets [7,8]. In contrast, secondary MPs are formed through the degradation of larger plastic items via physical, chemical, and biological processes [7,9]. Given the persistence and pervasiveness of microplastics in the environment, there is an urgent need for sustainable strategies focused on preventing, mitigating, and removing microplastics to reduce ecological risks and promote circular economy principles through innovative materials, improved waste management practices, and environmentally responsible technologies.

The most common MPs are polyethylene (PE), polyvinyl chloride (PVC), polypropylene (PP), polystyrene (PS), and polyethylene terephthalate (PET) [10]. Due to their low density, small size, and resistance to degradation, MPs disperse rapidly throughout terrestrial and aquatic environments. Their physicochemical characteristics, including relatively high surface area-to-volume ratios, hydrophobicity, surface roughness, and chemically diverse functional groups, enhance their ability to adsorb a broad spectrum of environmental contaminants, such as polycyclic aromatic hydrocarbons (PAHs), persistent organic pollutants (POPs), heavy metals, and organochlorine pesticides. This interaction results in a persistent and mobile form of pollution that poses long-term risks to ecological and human health [11,12,13]. The removal of MPs from wastewater has been explored using physical, chemical, and biological methods [14,15]. Among these, adsorption is probably the most favorable approach because of its ease of application and low cost [16]. Various adsorbents, such as biochar [17], sponges [18], aerogels [19], magnetic nanoparticles [17,20,21], carbon nanotubes [22], and metal oxides [23] have been used to remove MPs.

The use of magnetic materials for MP removal is particularly promising because these materials can be rapidly recovered from water by applying a magnetic field. For example, Shi et al. [24] magnetized the surfaces of MPs particles with Fe₃O₄ nanoparticles and achieved an average removal rate of 62–98% for four common types of MPs, including PE, PP, PS, and PET. Similarly, Grbic et al. [25] demonstrated that iron nanoparticles functionalized with hexadecyltrimethoxysilane (a long-chain silane agent) captured PE and PS spheres (<20 µm) with efficiencies of 96% and 88%, respectively. Furthermore, Heo et al. [26] achieved a removal efficiency of 90% for polystyrene particles (<1 µm) using magnetic iron oxide nanoparticles.

Although magnetic Fe₃O₄ nanoparticles have been extensively studied due to their magnetic properties, biocompatibility, resistance, and low toxicity, they are suitable for various environmental applications, such as the removal of pollutants from water [27,28,29]. Despite these advantages, the adsorbents reported to date for MP removal are often limited by rapid saturation and high costs associated with the regeneration process [30]. To address these challenges, research has increasingly focused on modifying the surfaces of magnetic nanoparticles to enhance their adsorption capacity, improve adsorption-desorption kinetics, and facilitate easier separation and regeneration [31]. Modifying the surface of nanoparticles allows for adjustments or improvements in characteristics such as size, shape, and chemical composition [32,33,34]. For instance, citric acid can be adsorbed onto the surface of iron oxide nanoparticles via coordination through one or two of its carboxylate groups, leaving at least one carboxylic acid group uncoordinated. This modification renders the nanoparticle surface more hydrophilic, prevents particle agglomeration, and provides additional functional groups to enhance the adsorption performance [35,36,37].

This study aimed to evaluate the effectiveness of Fe₃O₄@AC in removing different types of MPs (HDPE, LDPE, and PP). These MPs were chosen because their densities are lower than that of water, which complicates their removal using conventional processes. Their low sedimentation rates and increased buoyancy reduce the effectiveness of gravity-based separation methods. Fe₃O₄@AC was synthesized using co-precipitation, followed by the addition of citric acid. Co-precipitation synthesis is the most commonly used method because, in addition to being low-cost and relatively straightforward, it also has a high potential for industrial applications [35,38,39]. This method involves a stoichiometric mixture of ferric and ferrous salts in an aqueous solution, followed by the addition of a base (NaOH, KOH, or NH₄OH) at elevated or ambient temperatures to produce iron oxide [40]. The batch adsorption performance of Fe₃O₄@AC was evaluated by optimizing the operational parameters, such as particle dosage, contact time, and pH. To elucidate the adsorption mechanisms and assess the Fe₃O₄@AC performance, experimental data were analyzed using adsorption isotherm models (Freundlich [41], Langmuir [42], Sips [43], and Dubinin–Radushkevich [44]) and kinetics models (pseudo-first-order [45], pseudo-second-order [46]). Finally, the reusability of Fe₃O₄@AC was evaluated through multiple adsorption and washing cycles. To the best of our knowledge, this is the first study to apply magnetic Fe₃O₄@AC has been applied to remove MPs from an aqueous medium.

2. Materials and Methods

2.1. Chemical Reagents and Analysis

Fe₃O₄@AC from Fe₃O₄ nanoparticles. For this purpose, iron (III) chloride hexahydrate (Sigma Aldrich, Brasilia, Brazil, 98%), citric acid (Sigma Aldrich, Brasilia, Brazil, 99.5%), and iron sulfate pentahydrate (Vetec, Brasilia, Brazil, 99%) were the main chemical reagents used. After synthesis, Fe₃O₄@AC was washed with acetone (Vetec, Brasilia, Brazil, 99%) and methyl alcohol (Dinâmica, Brasilia, Brazil, 99%). The MPs used in this study were derived from three common consumer polymers: high-density polyethylene (HDPE), low-density polyethylene (LDPE), and polypropylene (PP), which were obtained from commercially available plastic products. These polymers were selected because they have a range of lower density values (0.92–0.93 g/cm³, 0.93–0.97 g/cm³, and 0.89–0.92 g/cm³ for LDPE, HDPE, and PP [47], respectively). This is lower than that of water; therefore, their removal from wastewater is not effective. Initially, the plastics were cut into pieces ranging from 3 to 5 cm, mechanically crushed for 5 min at 12,000 RPM, and the resulting product was sieved through a 3 mm mesh.

2.2. Synthesis of Magnetic Fe₃O₄@AC

Fe₃O₄ nanoparticles were synthesized using a co-precipitation method for ferrous (Fe²⁺) and ferric (Fe³⁺) compounds in an aqueous solution [38,39,48,49,50,51,52]. For this, 150 mL of a solution containing 40.5 g (1.0 M) of FeCl₃·6H₂O and 21 g (0.5 M) of FeSO₄·7H₂O was prepared using distilled water that had been previously bubbled with N₂ (g). This procedure was carried out in a round-bottom flask at 70 °C for 30 min under a N₂ atmosphere with constant mechanical agitation (1000 RPM). At a constant temperature of 70 °C, 38 mL of NH₄OH was added to the system to maintain the pH. After 30 min, 3 mL of a 0.5 g/mL citric acid aqueous solution was added. The temperature of the system was then raised to 90 °C under reflux conditions and maintained for 60 min with continuous stirring. The system was then cooled to room temperature, and Fe₃O₄@AC was removed by magnetic separation using a neodymium magnet. The removed Fe₃O₄@AC was washed three times with deionized water, acetone, and methyl alcohol. Figure 1 shows an illustrative scheme of the synthesis procedure for Fe₃O₄@AC using the coprecipitation method.

2.3. Characterization of MPs and Fe₃O₄@AC

The surface functional groups of both MPs and the Fe₃O₄@AC were identified using Fourier Transform Infrared Spectroscopy ATR-FTIR, (Perkin Elmer, Frontier, MA, USA) with an Attenuated Total Reflectance accessory equipped with a ZnSe crystal, operating in the spectral range of 4000 to 650 cm⁻¹. The morphology of the MPs was observed using a digital stereomicroscope Leica EZ4D, (Leica, Wetzlar, Germany). Experimental X-ray diffraction measurements Bruker AXS D8 Focus, (Bruker Co., Karlsruhe, Germany) were conducted at room temperature in the 2θ range of 10° to 70°, with a current of 30 mA, a voltage of 40 kV, a step size of 0.04°, and a scanning speed of 0.1°/min. The average particle diameter (D) was calculated using the Scherrer equation (D = K_pλ/βcosθ, where D is the mean particle diameter (nm), λ is the wavelength of the electromagnetic radiation used for diffraction, θ is the diffraction angle, β is the width at half height of the highest intensity diffraction peak, and K_p is the Scherrer constant, usually considered to be 0.9). The values of θ and β correspond to the highest intensity peaks. The nanoparticles were characterized by X-ray diffraction, by the powder method, using Kα radiation from Copper, with wavelength (λ = 1.5406 Å). The morphologies of MP and Fe₃O₄@CA were observed using scanning electron microscopy (SEM) FEI Quanta 250 (FEI Inc., Eindhoven, The Netherlands). Nitrogen sorption isotherms were recorded using an automated gas adsorption system ASAP 2020 Plus 2.00 (Micromeritics, Norcross, GA, USA). All samples were degassed at 100 °C for 12 h under vacuum prior to N₂ adsorption. Specific areas were estimated using the Brunauer-Emmett-Teller (BET) method, and the pore size distribution was calculated using the Barrett-Joyner-Halenda (BJH) method. The zeta potential (ζ) was measured using a ZetaSizer Nano ZS90, (Malvern Ltd., Malvern, UK) at pH values of 2, 4, 6, 8, and 10. pH measurements were performed using a pH meter and a conductivity meter Metrohm model 914 (Metrohm, Herisau, Switzerland).

2.4. Experimental Procedure for the MPs Removal

To evaluate the performance of Fe₃O₄@AC in removing MPs, a series of experiments were conducted to assess the effects of particle concentration, contact time, and pH. Suspensions of 30 mL were prepared using deionized water and 0.03 g of MPs in a 1:1:1 ratio, containing MPs of HDPE, LDPE, and PP. Different concentrations of Fe₃O₄@AC (0.5, 1.0, 1.5, 2.0, and 3.0 g/L) were added to each solution, and the system was maintained under constant stirring at 500 RPM Orbital Shaker 10 B (Global Trade Technology, São Paulo, Brazil) for 150 min. The Fe₃O₄@AC + MPs set was then removed from the system using a neodymium magnet for magnetic separation. The content was vacuum-filtered through a fiberglass membrane with a pore size of 0.7 µm and thickness of 475 µm (Merck Millipore Ltd., Billerica, MA, USA). After filtration, the membranes containing the removed MPs fraction were placed in an oven at 100 °C for approximately 30 min to dry. Subsequently, the contact times (5, 10, 30, 60, 90, and 150 min) and pH levels (2, 4, 6, 8, and 10) were investigated using the same procedures. The removal percentage, amount of MPs adsorbed at equilibrium, and concentration of MPs removed by Fe₃O₄@AC over time were estimated using Equations (1) and (2), respectively.

$$\mathrm{R}\mathrm{e}\mathrm{m}\mathrm{o}\mathrm{v}\mathrm{a}\mathrm{l}\left(\mathrm{\%}\right)={\displaystyle \frac{({\mathrm{M}}_{\mathrm{M}\mathrm{P}\mathrm{s}}-{\mathrm{M}}_0)}{{\mathrm{M}}_{0,\mathrm{M}\mathrm{P}\mathrm{s}}}}·100$$

(1)

$${\mathrm{q}\mathrm{e}}_{\mathrm{e}\mathrm{x}\mathrm{p}}={\displaystyle \frac{({\mathrm{C}}_0-{\mathrm{C}}_{\mathrm{e}})\mathrm{V}}{\mathrm{m}}}$$

(2)

where, M₀ (g) corresponds to the initial mass of the membrane, M_MPs (g) is the mass of the membrane with MPs after filtration and drying, M_0,MPs (g) is the initial mass of MPs, ${\mathrm{q}\mathrm{e}}_{\mathrm{e}\mathrm{x}\mathrm{p}}$ (mg/L) represents the removal capacity at equilibrium concentration, C_e (mg/L) represents the residual mass concentration, C₀ (mg/L) represents the initial mass concentration of MPs in solution V (L) represents the volume of the solution, and m (g) means the mass of Fe₃O₄@AC.

2.5. Isothermal and Kinetic Studies

The results of the adsorption tests were fitted to the Freundlich [41], Langmuir [42], Sips [42], and Dubinin-Radushkevich [53] (Equations (3)–(6), respectively).

$${\mathrm{q}}_{\mathrm{e}}^{\mathrm{F}\mathrm{r}}={\mathrm{K}}_{\mathrm{f}}{\mathrm{C}}_{\mathrm{e}}^{1/\mathrm{n}}$$

(3)

$${\mathrm{q}}_{\mathrm{e}}^{\mathrm{L}\mathrm{a}\mathrm{n}\mathrm{g}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{m}\mathrm{a}\mathrm{x}}{\mathrm{K}}_{\mathrm{l}}{\mathrm{C}}_{\mathrm{e}}}{(1+{\mathrm{K}}_{\mathrm{l}}{\mathrm{C}}_{\mathrm{e}})}}$$

(4)

$${\mathrm{q}}_{\mathrm{e}}^{\mathrm{S}\mathrm{i}\mathrm{p}\mathrm{s}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{m}\mathrm{a}\mathrm{x}}{{(\mathrm{K}}_{\mathrm{s}}{\mathrm{C}}_{\mathrm{e}})}^{{\mathrm{n}}_{\mathrm{s}}}}{1+{{(\mathrm{K}}_{\mathrm{s}}{\mathrm{C}}_{\mathrm{e}})}^{{\mathrm{n}}_{\mathrm{s}}}}}$$

(5)

$${\mathrm{q}}_{\mathrm{e}}^{\mathrm{D}-\mathrm{R}}={\mathrm{q}}_{\mathrm{m}\mathrm{a}\mathrm{x}}\mathrm{e}\mathrm{x}\mathrm{p}[-({\displaystyle \frac{\mathrm{R}\mathrm{T}\mathrm{l}\mathrm{n}({\mathrm{C}}_{\mathrm{s}}/{\mathrm{C}}_{\mathrm{e}})}{{\mathrm{K}}_{\mathrm{D}\mathrm{R}}}}{)}^2]$$

(6)

where K_f (mg^1−(1/n) g/L^1/n) is the Freundlich constant related to the removal capacity, n is the heterogeneity factor that indicates the intensity of removal, q_max (mg/g) is the maximum adsorption capacity of the adsorbent, and K_l (L/mg) is the Langmuir constant related to the removal energy. C_S and C_e are the equilibrium solubility and concentration, respectively. K_DR (kJ/mol) is the parameter used to estimate the free energy of the system; R (8.314 J/mol·K) is the gas constant; T (K) is the temperature; K_S (L/mmol) is the isothermal constant of the Sips, and ns is the heterogeneity of the system.

The kinetic parameters were evaluated using pseudo-first-order and pseudo-second-order kinetic models (Equations (7) and (8), respectively) [54,55].

$${\mathrm{q}}_{\mathrm{t}}^{\mathrm{P}\mathrm{P}\mathrm{O}}=1-{\mathrm{e}}^{-{\mathrm{k}}_1\mathrm{t}}$$

(7)

$${\mathrm{q}}_{\mathrm{t}}^{\mathrm{P}\mathrm{S}\mathrm{O}}={\displaystyle \frac{{\mathrm{q}}_{\mathrm{e}}^2{\mathrm{K}}_2\mathrm{t}}{1+{\mathrm{q}}_{\mathrm{e}}{\mathrm{K}}_2\mathrm{t}}}$$

(8)

where q_e (mg/g) is the equilibrium removal capacity, q_t (mg/g) is the removal capacity of the adsorbent at a given contact time, t (min) is the contact time, k₁ (min⁻¹) is the rate constant of the pseudo-first-order model, and K₂ (min⁻¹) is the rate constant of the pseudo-second-order model.

2.6. Reuse Testing

To investigate the potential for reuse of Fe₃O₄@AC, after the adsorption experiments, Fe₃O₄@AC nanoparticles charged with MPs were immersed in acetone and subjected to ultrasound (60 Hz) for 30 min. Fe₃O₄@AC was then dried at 100 °C for 2 h and reused in a new MPs removal experiment. The cycle was repeated five times.

3. Results

3.1. Characterization of MPs

The MPs obtained from LDPE, HDPE, and PP polymers are shown in Figure 2. These polymers were selected due to their low-density ranges, 0.92–0.93 g/cm³ for LDPE, 0.93–0.97 g/cm³ for HDPE, and 0.89–0.92 g/cm³ for PP, which are all lower than the density of water (1.00 g/cm³) [47]. In this figure, it can be observed that all MP samples exhibited irregular shapes and a wide range of particle sizes. Specifically, the particle size distribution ranged from 0.1 mm to 3.0 mm. The size range was selected to reflect the variety of MPs in environmental sample.

The ATR-FTIR spectra of LDPE, HDPE, and PP MPs are shown in Figure 3a–c and are consistent with the spectral profiles reported in the literature [56,57]. Bands in the range of 719–721 cm⁻¹ (Figure 3a,b) are attributed to the -CH₂ wagging vibrations, while the band located around 1471 cm⁻¹ is related to CH bending. Moreover, the bands at 2849–2850 cm⁻¹ and 2918 cm⁻¹ are assigned to the symmetric and asymmetric stretching modes of the CH₂ group. An absorption band with a relatively weak intensity at 1377 cm⁻¹ was observed in the LDPE spectrum, associated with the presence of branched methyl groups, which distinguishes LDPE from HDPE [56,58]. Regarding the spectrum of PP, the band at 842 cm⁻¹ is assigned to the C-CH₃ stretching vibration, while the bands at 972, 998, and 1167 cm⁻¹ are attributed to the -CH₃ rocking vibrations. The symmetric bending vibration mode of the CH₃ group is observed at 1376 cm⁻¹. The absorption peak observed at 2951 cm⁻¹ is attributed to the -CH₃ asymmetric stretching vibration. All previously mentioned absorption peaks are associated with the presence of methyl groups in polypropylene.

3.2. Morphological, Crystalline and Chemical Analysis of the Fe₃O₄@AC Nanoparticles

Figure 4a shows a photograph of the Fe₃O₄@AC powder synthesized in this work. As expected, the powder was dark brown, which is characteristic of iron-based materials. Figure 4b shows the SEM images of the Fe₃O₄@AC particles. The material exhibits a broad size distribution, ranging from 20 to 200 µm, and diverse particle shapes, including angular, rounded, and plate-like morphologies. Such heterogeneity is beneficial for MPs removal, as it promotes a variety of interaction mechanisms, including physical trapping, surface adhesion, and interfacial interactions [59]. Figure 4c shows a relatively low-magnification SEM image highlighting the formation of aggregates composed of smaller particles (<20 µm). These aggregates primarily result from magnetic dipole–dipole attractions and van der Waals forces between the Fe₃O₄ nanoparticles during synthesis and subsequent drying. Despite forming agglomerates, these structures do not hinder MPs’ removal in aqueous media [60]. In contrast, they contribute positively by maintaining magnetic responsiveness, which facilitates efficient post-treatment separation using external magnetic fields. Despite this, Fe₃O₄@AC exhibited magnetic characteristics that allowed for its easy separation, as qualitatively demonstrated in Figure 5.

The XRD and FTIR results obtained for the Fe₃O₄@AC nanoparticles are shown in Figure 6a and Figure 6b respectively. The following diffraction peaks were indexed: 2θ = 21.2, 30.2, 35.7, 43.3, 53.5, 57.3, and 62.8°, which are associated with the crystallographic planes (111), (220), (311), (400), (422), (511), and (440), respectively. As expected, the diffractogram confirmed that the Fe₃O₄@AC nanoparticles are fully crystalline. All peaks were indexed using card JCPDS no. 19–629, which corresponds to the magnetite phase (crystalline, face-centered cubic structure of the inverse spinel type) [38]. The Scherrer equation was used to calculate the crystallite size from the highest intensity peak (311) [61], as follows: The average diameter of the nanoparticles was 5 nm. This value is consistent with the crystallite size range (3–9 nm) reported by Omelyanchik et al. [62]. The FTIR spectra (Figure 6b) show that Fe₃O₄@AC nanoparticles presented bands at 568 cm⁻¹, 1400 cm⁻¹, 1622 cm⁻¹, and 3390 cm⁻¹. The band at 568 cm⁻¹ is characteristic of the stretching vibration of the Fe-O bond, while the band located around 1400 cm⁻¹ can be attributed to the stretching vibration of the symmetric and asymmetric carboxyl groups due to the presence of citrate molecules as a shielding agent. The band at 1622 cm⁻¹ is attributed to the C=O vibration (asymmetric stretching) of the COOH group of citric acid, which reveals the binding of a citric acid radical to the surface of the adsorbents by the chemisorption of carboxylate ions. Finally, the band at 3390 cm⁻¹ indicates simple binding and vibration of OH stretching due to the presence of water molecules absorbed on the surface of Fe₃O₄ [38,49].

The specific surface area, total pore volume, pore size distribution, and average pore diameter of the Fe₃O₄@AC nanoparticles were determined using the nitrogen adsorption–desorption isotherms (Figure 7). A typical IV isothermal curve was observed with well-defined steps in the adsorption-desorption curves occurring between the relative pressure, P/P0, of 0.4–1.0, suggesting the existence of a narrow distribution of mesopores [63,64]. This was confirmed by the presence of a desorption hysteresis loop (type H3/H4) associated with slit-shaped pores, revealing the simultaneous existence of micropores (<2 nm) and mesopores (2–50 nm) according to the IUPAC notation [65]. The specific surface area determined by the BET method was 110 m²·g⁻¹. The total pore volume and pore size distribution were determined using the BJH method. The results indicate that the Fe₃O₄@AC nanoparticles synthesized in this work presented a total pore volume of 0.21 cm³·g⁻¹. The pore size distribution confirms that the material is mesoporous, with pore sizes ranging from 42 to 160 nm. The average pore size of BJH was 7.5 nm.

3.3. Removal of the MPs

3.3.1. Effect of Fe₃O₄@AC Nanoparticles Concentration, Contact Time and pH

The efficiency of MPs removal was obtained as a function of the initial concentration of Fe₃O₄@AC nanoparticles (C₀), contact time (t), and pH (Figure 8a–c). The initial concentration of MPs into suspension was 0.3 g/L, with HDPE, LDPE and PP mixed in equal proportions (1:1:1). The efficiency of MPs removal increased when C₀ increased from 0.5 to 1 g/L. However, the efficiency stabilized at 74 ± 7.2% when the concentration ranged from 1.0 g/L to 3 g/L (Figure 8a). These results are consistent with those of Shi et al. (2022) [24], who found that increasing the concentration of Fe₃O₄@AC nanoparticles did not further enhance the removal of polyethylene MPs. This behavior may be attributed to the saturation of the available adsorption sites on the nanoparticle surfaces at higher dosages. Moreover, the removal efficiency increased significantly between 5 and 10 min and remained relatively constant (80 ± 7.9%) between 10 and 150 min (Figure 8b). Therefore, a Fe₃O₄@AC nanoparticle concentration of 1.0 g/L and a contact time of 30 min were selected for further experiments.

The pH plays a fundamental role in the adsorption process because it influences the surface charge of both the adsorbent and adsorbate. The effect of pH on MP removal is illustrated in Figure 8c. The removal of MPs increased as the pH increased from 2.0 to 6.0, stabilized in the range of 6.0 and 8.0, and decreased for pH values greater than 8.0. Figure 9 compares the zeta potential (ζ) values of the Fe₃O₄@AC nanoparticles and HDPE, LDPE, and PP MPs as a function of pH. These measurements help elucidate the influence of surface charge interactions on the adsorption behavior at different pH values. Both Fe₃O₄@AC nanoparticles and MPs are negatively charged. Therefore, the removal of MPs by Fe₃O₄@AC nanoparticles cannot be explained by electrostatic interactions. Instead, these results suggest that the formation of hydrogen bonds could be the interaction mechanism due to the presence of polar groups such as COOH⁻ and OH⁻ of citric acid, which become weaker at acidic pH. MPs can form hydrogen bonds with polar functional groups, enhancing their adhesion and transport in various media [66]. The reduction in MPs removal at pH values above 8.0 is attributed to the dilution of Fe₃O₄@AC nanoparticles, which results in the release of iron ions into the solution, as evidenced experimentally by its brown color. This phenomenon is attributed to the decrease in the stability of the iron oxide structure at high pH due to the deprotonation of surface groups, which leads to the solubilization of the adsorbents, culminating in a significant reduction in the removal efficiency.

Table 1. Comparison of MPs removal with different magnetic materials.

Magnetic Materials	Size (μm)	MPs	Size MPs (μm)	Removal Capacity (mg/g)	Removal	References
Fe3O4@AC	-	HDPE, LDPE, PP	1–3000	23.72	80%	This work
NPs-Fe	0.025	PE PS	<20	-	92%	[25]
Fe3O4	-	PS	1	2799.2	90%	[26]
PEG/Fe3O4	-	PE	13–149	2.202	97%	[16]
Corn cob magnetic biochar	63.5	PA	27–307	1145	97%	[17]
Nano-Fe3O4	<0.03	PE, PP, PS, PET	200–900	367.62	98%	[24]
Magnetite and cobalt ferrite nano ferrofluid	-	PE, PVC, PES	0.74–1.88	-	55%	[67]
Fly ash modified with Fe	-	PS	0.08	83.1	94%	[68]
Nano-Fe3O4 with hydrophobic coating	~0.1	PS	0.1–1	-	93%	[23]
Microrobotsγ-Fe2O3/Pt/TiO2	-	PS	0.05	-	97%	[69]
Fe3O4 NPs	0.01	PE, PP, PVC, PS, PET	20–800	-	100%	[70]
Fe3O4@SiO2-PAC 18	1.72	PE PS	10 μm	2.41	92%	[71]

summarizes the performances of the most prominent magnetic removal agents reported in the literature. In general, the Fe₃O₄@AC nanoparticles used to remove HDPE, LDPE, and PP demonstrated good removal performance compared to those in previous studies. However, the material evaluated in this work had a lower removal capacity than those reported in other studies. This difference can be attributed to factors such as the size of the adsorbent and MPs. Nevertheless, the material was synthesized through a more cost-effective and greener process, highlighting the need for further optimization, such as preventing particle aggregation to enhance its performance.

3.3.2. Adsorption Isothermal and Kinetic Study

The adsorption behavior of MPs on Fe₃O₄@AC nanoparticles with different particle concentrations (0.5, 1.0, 1.5, 2, and 3 g/L) was investigated (Figure 10a). The Freundlich, Langmuir, Sips, and Dubinin-Radushkevich equations were fitted to the experimental data, and the results are summarized in

Table 2. Parameters obtained by fitting the data with the Freundlich, Langmuir, Sips, and Dubinin-Radushkevich isothermal models and pseudo-first- and pseudo-second-order kinetic models.

Isotherm	Parameters	Fe3O4@AC
Freundlich ${\mathbf{q}}_{\mathbf{e}}^{\mathbf{F}\mathbf{r}} = {\mathbf{K}}_{\mathbf{f}}{\mathbf{C}}_{\mathbf{e}}^{1/\mathbf{n}}$	Kf (mg1−1/n g/L1/n)	40.84
	n	5.02
	R2	0.969
Langmuir ${\mathbf{q}}_{\mathbf{e}}^{\mathbf{L}\mathbf{a}\mathbf{n}\mathbf{g}} = {\displaystyle \frac{{\mathbf{q}}_{\mathbf{m}\mathbf{a}\mathbf{x}}{\mathbf{K}}_{\mathbf{l}}{\mathbf{C}}_{\mathbf{e}}}{1+{\mathbf{K}}_{\mathbf{l}}{\mathbf{C}}_{\mathbf{e}}}}$	qmax (mg/g)	27.27
	Kl (L/mg)	107.88
	R2	0.987
Sips ${\mathbf{q}}_{\mathbf{e}}^{\mathbf{S}\mathbf{i}\mathbf{p}\mathbf{s}} = {\displaystyle \frac{{\mathbf{q}}_{\mathbf{m}\mathbf{a}\mathbf{x}}{\left({\mathbf{K}}_{\mathbf{s}}{\mathbf{C}}_{\mathbf{e}}\right)}^{{\mathit{n}}_{\mathit{s}}}}{(1+{\mathbf{K}}_{\mathbf{s}}{\mathbf{C}}_{\mathbf{e}})}}$	qmax (mg/g)	23.72
	Ks (L/mg)	84.93
	R2	0.991
	ns	2.922
Dubinin-Radushkevich ${\mathbf{q}}_{\mathbf{e}}^{\mathbf{D}-\mathbf{R}} = {\mathbf{q}}_{\mathbf{m}\mathbf{a}\mathbf{x}}\mathbf{e}\mathbf{x}\mathbf{p}[-({\displaystyle \frac{\mathbf{R}\mathbf{T}\mathbf{l}\mathbf{n}({\mathbf{C}}_{\mathbf{s}}/{\mathbf{C}}_{\mathbf{e}})}{{\mathbf{E}}_{\mathbf{D}\mathbf{R}}}}{)}^2]$	qmax (mg/g)	36.24
	EDR (KJ/mol)	0.05
	R2	0.85
Pseudo-first-order ${\mathbf{q}}_{\mathbf{t}}^{\mathbf{P}\mathbf{P}\mathbf{O}} = 1-{\mathbf{e}}^{-{\mathbf{k}}_1\mathbf{t}}$	k1 (min−1)	0.33
	qe (mg/g)	22.89
	R2	0.998
Pseudo-second-order ${\mathbf{q}}_{\mathbf{t}}^{\mathbf{P}\mathbf{S}\mathbf{O}} = {\displaystyle \frac{{\mathbf{q}}_{\mathbf{e}}^2{\mathbf{K}}_2\mathbf{t}}{1+{\mathbf{q}}_{\mathbf{e}}{\mathbf{K}}_2\mathbf{t}}}$	K2 (g/(mg·min))	0.035
	qe (mg/g)	23.445
	R2	0.998

. The correlation coefficient (R²) was used to describe the best mathematical fit of the data. The mathematical adjustment performed with the Sips equation was considered the most satisfactory, with a value of R² (0.991), compared with the calculated values for the Freundlich, Langmuir, and Dubinin-Radushkevich equations (0.969, 0.987, and 0.846, respectively). Furthermore, the value of $q_{max}^{cal}$ determined by the Sips model was 23.72 ± 0.75 mg/g; this value is close to the $q_{max}^{exp}$ (22.55 ± 0.66 mg/g). Therefore, the Sips isotherm provided a good correlation to characterize the removal of MPs by Fe₃O₄@AC nanoparticles and indicated that the process occurred in stages, since the Sips equation is based on the theory that at low concentrations of adsorbate, adsorption is described by the Freundlich equation. At high adsorbate concentrations, adsorption is characterized by the Langmuir equation; that is, it occurs through monolayers. The isothermal exponential ns of the Sips model (ns) in this study indicates that the adsorption of MPs involves several inhomogeneous adsorption behaviors on the surface of the adsorbent. Furthermore, the heterogeneity factor obtained from the Freundlich model indicates that physical processes govern the nature of adsorption [72,73].

The pseudo-first- and pseudo-second-order kinetics models were adjusted to the experimental data (Figure 10b and Table 2) using 30 mg of Fe₃O₄@AC nanoparticles and a time interval of 0 to 150 min. The experiments were conducted at room temperature and at a pH of 6. Pseudo-first-order and pseudo-second-order models were fitted to experimental adsorption kinetics data (Figure 10b). The R² values for the pseudo-first-order and pseudo-second-order models were 0.998 and 0.999, respectively. Furthermore, the $q_e^{cal}$ values with the pseudo-first order and pseudo-second-order models were 22,886 ± 0.409 mg/g and 23,445 ± 0.502 mg/g. It was considered that the pseudo-second order model best fitted the experimental data because it presented the highest R² value and because it presented $q_e^{cal}$ close to $q_e^{exp}$. This result indicates that the removal of MPs by Fe₃O₄@AC nanoparticles is theoretically associated with chemisorption, and that the removal rate depends on the time and availability of binding sites [73,74].

3.4. Insight into the Mechanisms of MPs Removal by Fe₃O₄@AC Nanoparticles

Electrostatic and hydrophobic interactions, pore-filling mechanisms, Van Der Waals forces, and hydrogen bonds are the predominant mechanisms in MPs removal [68,75]. Given the nature of the average pore size, specific surface area, and size range of MPs, no adsorption of MPs in the interstices of Fe₃O₄@AC nanoparticles was observed. The Fe₃O₄@AC nanoparticles were inserted on the surface of the MPs, making it an MP remnant. Figure 11 shows a set of plausible mechanisms for MPs removal by Fe₃O₄@AC nanoparticles based on the results presented (kinetic-isothermal analysis, zeta potential, SEM analysis, and BET investigations). As discussed, considering the zeta potential values depicted in Figure 9, both the Fe₃O₄@AC nanoparticles and MPs were negatively charged. Therefore, the MPs removal process by Fe₃O₄@AC nanoparticles cannot be explained by electrostatic interactions [76].

The hydrophobic interaction mechanism also does not describe the investigated adsorption process because the adsorption band identified at 3390 cm⁻¹ in the FTIR spectra of Fe₃O₄@AC nanoparticles (Figure 6b) indicates the hydrophilic character of this material, as this absorption band is related to the presence of water molecules absorbed on their surface [38,49]. Indeed, it is known that MPs are predominantly hydrophobic [77]. Hydrophobic interaction occurs when both materials involved in the adsorption process exhibit hydrophobic characteristics [57,78].

Figure 12a,b shows SEM images of the surface of HDPE-MPs after the removal process using 1 g/L of Fe₃O₄@AC nanoparticles for 30 min. As shown in Figure 12a, Fe₃O₄@AC nanoparticles were observed on the surface of the HDPE-MPs. The interactions between the Fe₃O₄@AC nanoparticles and MPs investigated in this study can be explained in terms of pore filling, van der Waals interactions, and hydrogen bonding. Regarding the pore-filling mechanism, Figure 12a,b shows that the HDPE-MP surface presented a relatively high concentration of defects (pores, roughness, and cracks). The same behavior was observed for LDPE and PP-MPs. This supports the pore-filling mechanism, which can explain the removal process by Fe₃O₄@AC. In addition, the presence of pores in the Fe₃O₄@AC nanoparticles (Figure 12) increased the surface area available for interactions, which favored the adsorption of these particles on the MPs. The van der Waals interaction mechanism also contributes to the removal of MPs by Fe₃O₄@AC nanoparticles. It is known that aliphatic polymers (such as HDPE, LDPE, and PP) interact with different compounds through van der Waals forces due to the presence of non-specific functional groups [78]. When a polar compound (Fe₃O₄@AC) approaches a non-polar compound (HDPE, LDPE, or PP), a spontaneous dipole moment can form, favoring the connection between Fe₃O₄@AC and HDPE, LDPE, or PP [79,80].

The hydrogen bonding mechanism originates from the processing of MPs, as the applied mechanical stress favors the creation of defects on the surface, thereby increasing the concentration of oxygenated functional groups, such as carboxyl and hydroxyl [81,82]. These functional groups increase the hydrophobic character of MPs, as well as their polarity, charge, and specific surface area, favoring the interaction between MPs and the environment [78,83].

3.5. Reuse Testing of Fe₃O₄@AC Nanoparticles

Figure 13a shows the results of the Fe₃O₄@AC nanoparticle reuse tests. After the adsorption test, the Fe₃O₄@AC nanoparticles were immersed in acetone and subjected to ultrasound (60 Hz) for 30 min. The Fe₃O₄@AC nanoparticles were then dried (100 °C for 2 h) and reused in a new MPs removal experiment. The results showed that the adsorption efficiency of MPs gradually decreased from 83% to 70% after three cycles. The removal capacity of MPs in the fifth cycle was significantly reduced, with a decrease of approximately 39% compared to the initial removal percentage. This degradation may have resulted from the partial oxidation or structural deterioration of the Fe₃O₄@AC nanoparticles during the regeneration process. Regarding the chemical stability of the material, the FTIR spectra (Figure 13b) show that the main functional groups remain largely unchanged after five reuse cycles, indicating that the core chemical structure is preserved despite some mass loss. To enhance the durability of the material, more robust surface modifications, such as polymeric coatings or silica encapsulation, should be explored to protect the magnetic core from oxidation. Furthermore, alternative regeneration protocols (solvent-free thermal treatments or mild chemical washing) may improve the recovery efficiency and extend the material lifespan. Addressing these limitations is essential, as regeneration directly impacts the economic and environmental viability of reuse strategies by minimizing material replacement and disposal issues [84].

4. Conclusions

Magnetic Fe₃O₄@AC nanoparticles were successfully synthesized using the co-precipitation method, followed by citric acid functionalization. XRD experiments confirmed the predominant presence of the magnetite phase. Fe₃O₄@AC nanoparticles synthesized proved to be efficient in removing MPs particles (HDPE, LDPE, and PP) from aqueous solutions, as 1 g L⁻¹ of Fe₃O₄@AC nanoparticles removed more than 80% of MPs at room temperature and pH 6. The conclusions obtained from the nitrogen adsorption isotherms, SEM, and zeta potential experiments, supported by theoretical insights, indicate that the process involves the removal of MPs by Fe₃O₄@AC nanoparticles, and the mechanisms of surface hydrogen bonding, pore filling, and van der Waals forces were predominant in this process. Reuse tests indicated that the synthesized magnetic materials removed >50% of the MPs after five adsorption cycles. These results underscore the potential of Fe₃O₄@AC nanoparticles as a sustainable, efficient, fast-acting, and cost-effective material for MPs removal, contributing to environmentally responsible wastewater treatment solutions. Despite these promising results, the application of Fe₃O₄@AC nanoparticles in real wastewater treatment systems remains a challenge. Wastewater contains a complex mixture of dissolved organic matter, surfactants, and colloidal particles, which may compete for adsorption sites or alter the surface properties of both MPs and Fe₃O₄@AC nanoparticles. In particular, the presence of organic contaminants can lead to fouling or partial deactivation of the magnetic composite, thereby reducing the removal efficiency. Therefore, additional studies are needed to assess the performance of the material under realistic wastewater treatment conditions, including variations in pH, ionic strength, and organic load, in order to validate its practical scalability and long-term stability. Future research should focus on optimizing the synthesis process (particularly minimizing particle agglomeration) to further enhance performance. Moreover, testing the material in real wastewater systems is essential to assess its robustness in the presence of coexisting contaminants. Finally, scaling up from batch to continuous systems is a crucial step toward the practical and sustainable application of this technology. Future work should focus on optimizing the synthesis process, particularly by preventing particle aggregation, to enhance performance. Additionally, it is recommended to assess the material’s effectiveness in removing MPs from real wastewater to better understand the potential interference from other coexisting contaminants in complex real systems. Finally, transitioning from batch experiments to continuous mode would further advance the practical application of the material.