Removal of Amoxicillin Antibiotic from Polluted Water by a Magnetic Bionanocomposite Based on Carboxymethyl Tragacanth Gum-Grafted-Polyaniline

Abstract

Removal of antibiotics from contaminated water is very important because of their harmful effects on the environment and living organisms. This study describes the preparation of a bionanocomposite of carboxymethyl tragacanth gum-grafted-polyaniline and γFe₂O₃ using an in situ copolymerization method as an effective adsorbent for amoxicillin antibiotic remediation from polluted water. The prepared materials were characterized by several analyses. The vibrating sample magnetometer and thermal gravimetric analysis showed that the carboxymethyl tragacanth gum-grafted-polyaniline@ γFe₂O₃ bionanocomposite has a magnetization saturation of 25 emu g⁻¹ and thermal stability with a char yield of 34 wt%, respectively. The specific surface area of bionanocomposite of about 8.0794 m²/g was obtained by a Brunauer–Emmett–Teller analysis. The maximum adsorption capacity (909.09 mg/g) of carboxymethyl tragacanth gum-grafted-polyaniline@ γFe₂O₃ was obtained at pH 7, an agitation time of 20 min, a bioadsorbent dose of 0.005 g, and amoxicillin initial concentration of 400 mg/L. The Freundlich isotherm and pseudo-second-order kinetic models were a better fit with the experimental data. The kinetic model showed that chemical adsorption is the main mechanism for the adsorption of amoxicillin on the bioadsorbent. In addition, the maximum adsorption capacity for amoxicillin compared to other reported adsorbents showed that the prepared bionanocomposite has a higher maximum adsorption capacity than other adsorbents. These results show that carboxymethyl tragacanth gum-grafted-polyaniline@ γFe₂O₃ would be a favorable bioadsorbent for the remediation of amoxicillin from contaminated water.

1. Introduction

Pharmaceuticals, particularly antibiotics, represent a concerning new class of pollutants in the environment, not to mention their effects on human health with regard to several infectious illnesses [1,2]. Most of these chemicals are found in surface water at elevated amounts [3]. Their bioresistant character and ability to elude traditional sewage treatment procedures are strongly tied together. It harms the ecology by making bacteria resistant and interfering naturally with the growth, development, and movement of a variety of microorganisms [4]. One of the biggest obstacles to a sustainable water future is the presence of this material in the aquatic environment, particularly in arid nations where water recycling is crucial. As a result of their extensive use in both human and animal medications, antibiotics play a significant role in water contamination. They are employed to treat bacterial infections and are extremely resistant to being inactivated until they have completed their intended function, which causes their incomplete metabolism in the organism [5]. More than 90% of medications taken orally don’t decompose, thus they become active compounds. Since antibiotics are highly soluble in water, conventional treatment procedures cannot eliminate them, which poses a significant obstacle to their removal [6].

Amoxicillin is a commonly used antibiotic whose structure is based on a β-lactam antibiotic categorized as penicillin, which is the cause of its great bacterial resistance to many microorganisms [7]. Systemic, bacterial, and gastrointestinal illnesses are treated with amoxicillin in both human and veterinary medicine. It is widely recognized that amoxicillin is utilized in modern medicine, and its ecotoxicity contributes to the danger of medical wastewater. Due to the difficulty of breaking down this antibiotic, the residue is eliminated in the urine and feces. Consuming too much amoxicillin creates resistant bacteria because it accumulates in the body and feeds the organisms [8]. Consequently, an effective technique to remove it is essential before it is released into the aquatic environment.

The most common techniques for removing antibiotics include electrochemical degradation [9], Fenton oxidation process [10], UV radiation [11,12], ozonation [13], membrane filtration [14,15], photolysis [16], biological degradation [17], and adsorption [18]. The most appealing technology, nevertheless, is the adsorption process, which has a flexible and straightforward design, is simple to use, is inexpensive, and is highly effective [19,20]. The adsorbent used in industrial applications should be able to quickly absorb the target material and be ecologically benign [21]. The choice of an appropriate adsorbent for antibiotic elimination was the subject of several investigations. In order to effectively remove amoxicillin from water, a range of micro/nanostructures were deemed acceptable adsorbents, either as single phases or composites. Several works have reported the use of various adsorbents for the removal of amoxicillin from aqueous solutions. For example, a metal–organic framework (MIL-53(Al)) was prepared with a hydrothermal technique and used as an adsorbent for the removal of amoxicillin antibiotics from water [22]. It was reported that the adsorption capacity of MIL-53 is 758.5 mg g⁻¹ in experimental conditions due to its high surface area. In another work, an adsorbent based on NH₄Cl-induced activated carbon was employed for the removal of amoxicillin from water [23]. It was expressed that the removal percentage of amoxicillin is above 99%, owing to the high specific surface area (1029 m²/g) of NH₄Cl-induced activated carbon. Amoxicillin removal from water was reported by an adsorbent based on activated carbon produced from pomegranate peel/iron nanoparticles [24]. The high removal percentage (97.9%) was obtained at a pH of 5 during a contact time of 30 min. A green magnetic adsorbent based on functional CoFe₂O₄-modified biochar was used for amoxicillin removal from water. It was reported that the maximum adsorption capacity (99.99 mg/g) is obtained at a pH of 7 and at ambient temperature [25].

Of those, given their biodegradability, availability, reasonable cost, and minimal toxicity to biological systems, biopolymer-based composites have recently been used often as an ecologically friendly adsorbent to remove water contaminants. Natural polymer hydrogels were created using guar gum, karaya gum, xanthan gum, ghatti gum, and tragacanth gum. These natural polymers were capable of efficiently capturing pollutants in their three-dimensional (3D) network. As a result, the amount of interaction between contaminants and the hydrogel’s surface functional groups is increased, improving adsorption effectiveness. They also have several industrial uses because of their capacity to hydrate in cold or hot water, either by stabilizing emulsion systems or by gel formation.

Tragacanth gum (TG), as a colorless and odorless natural polymer, is a highly complex heterogeneous anionic polysaccharide that forms from the stems and branches of Astragalus gummier and other Asian Astragalus species [26]. Within a few weeks, the exudate can be recovered after it has solidified into flakes or coils of ribbon [27]. Tragacanth gum is a substance that is found in Iran, India, Afghanistan, and Turkey. Neutral and anionic sugars, such as D-galacturonic acid, D-galactose, D-xylose, L-arabinose, L-fucose, and d-glucose are found in TG [26,28]. This natural polymer is widely used in the food industry, in medicine, and in cosmetics [29]. In addition, it is inexpensive, readily accessible, and has great solubility, strong thermal stability, and a long shelf life [30]. The existence of the hydroxyl- and carboxylic-acid-reactive functional groups in the TG can be used as chelating sites for the removal of pollutants from water. One method for improving the adsorption capacity of natural polymers is copolymerization with functional monomers and adding inorganic fillers.

The aforementioned advantages can be strengthened by developing bionanocomposite materials that contain polyaniline and magnetic nanoparticles, such as γFe₂O₃. Consequently, in this study, a three-step oxidative polymerization procedure using ammonium persulfate oxidizer was used to create an antibacterial magnetic bionanocomposite based on carboxymethyl tragacanth gum-grafted-polyaniline, and then various aspects of it were examined. The bionanocomposite was subsequently applied for amoxicillin antibiotic removal from polluted water.

2. Materials and Methods

2.1. Materials and Instruments

Distilled aniline, ammonium persulfate, monochloroacetic acid, hydrochloric acid, iron(III) chloride hexahydrate, iron(II) chloride tetrahydrate, and other solvents were supplied by Merck Company (Darmstadt, Germany). Tragacanth gum (TG) with high-quality in translucent flakes was purchased from Rad Kimia-Garan Company (Tehran, Iran).

The chemical structure of prepared materials was studied by Fourier transform infrared (FTIR, Equinox 55, Bruker, Leipzig, Germany), elemental analysis (CHNSO, ECS 4010, NC technologies) and energy-dispersive X-ray (EDX, MIRA 3-XMU, Tescan, Kohoutovice, Brno-Kohoutovice Czech Republic). The crystallinity and morphology of the samples were assessed by X-ray diffraction (XRD, D8 Advance X-ray diffractometer, Bruker, Leipzig, Germany) and a field emission scanning electron microscope (FESEM, MIRA 3-XMU, Tescan, Kohoutovice, Brno-Kohoutovice Czech Republic). The specific surface area of the samples was studied by the Brunauer–Emmett–Teller (BET, Belsorp mini II, MicrotracBEL, Osaka, Japan) technique. The thermal behavior of the samples was investigated by thermogravimetric analysis (TGA, L81A1750, Linseis, Selb, Germany).

2.2. Preparation of Carboxymethyl Tragacanth Gum (CMT)

The carboxymethyl tragacanth gum (CMT) was synthesized according to our previous literature with slight modifications [31]. We fully dissolved 1 g of powdered tragacanth gum (TG) in a mixture of water/ethanol (85 mL/15 mL). Then, 1.2 g of NaOH in 10 mL of distilled water was added to the reaction mixture. The solution was kept at 50 °C for 30 min under a magnetic stirrer (Heidolph, Schwabach, Germany). Afterward, 1.3 g of monochloroacetic acid in 10 mL of distilled water was added to the reaction solution, and the final solution was stirred for 4 h at 50 °C. After cooling, the solution was poured into a double volume of ethanol or methanol. The resultant CMC sediment was separated using filter paper and dried at 40 °C (Figure 1).

2.3. Synthesis of γFe₂O₃

The co-precipitation technique was employed for the γ-Fe₂O₃ nanoparticles synthesis [32]. A solution of NaOH (1.0 M) was added to 100 mL of deionized water, and the reaction mixture was stirred magnetically for 15 min under an inert atmosphere. Afterward, a solution of iron(III) and iron(II) salts was dropped into the previous solution, and the final solution was kept under vigorous stirring at ambient temperature for 70 min. The resulting brown precipitate was separated and washed several times with distilled water and ethanol. Lastly, the precipitate was dried and the obtained powder was calcined at 300 °C for 2 h to gain the γ-Fe₂O₃ nanoparticles.

2.4. Preparation of Carboxymethyl Tragacanth Gum-Grafted-Polyaniline@γFe₂O₃

In a 250 mL round-bottom flask, 0.62 g of carboxymethyl tragacanth gum (CMT) was dissolved in 50 mL of distilled water. Then, 1.5 mL of HCl was added to the solution, and the flask was kept under N₂ gas at 0–5 °C. Subsequently, 10 wt% (an optimal amount) of γFe₂O₃ nanoparticles were dispersed in 20 mL of distilled water for 30 min using an ultrasonic bath and then added to the above solution. After that, within 30 min, 3 g of ammonium persulfate in 10 mL of distilled water was added to the solution. The flask was then placed on a magnetic stirrer for 12 h under the aforementioned conditions after 1.25 mL of aniline monomer was added. The nanocomposite was separated by a magnet and rinsed with 10 mL of N-methyl pyrrolidone, water, and acetone before they were dried at 50 °C (Figure 1).

2.5. Adsorption Experiment

The carboxymethyl tragacanth gum-grafted-polyaniline@γFe₂O₃ (CMT-g-PANI@Fe₂O₃) bionanocomposite’s ability to remove amoxicillin antibiotic from aqueous solutions was tested in a few different ways. The calibration curve of amoxicillin was prepared by defining the absorbance at 228 nm with a series of standard solutions (1−8 mg/L) achieved from reducing the stock solution at pH 7. Then, the amoxicillin initial or equilibrium concentrations were measured with the calibration curve [22].

Amoxicillin’s initial concentration in an aqueous solution, solution pH, the amount of adsorbent, the agitation time, and other important parameters were all examined for their effects on adsorption capacity. The pH was changed from 4 to 9 using HCl (0.1 N) and NaOH (0.1 N). The optimal adsorption conditions were then investigated using a range of CMT-g-PANI@Fe₂O₃ biosorbent doses (0.005–0025 g), contact periods (5–30 min), and initial amoxicillin concentrations (50–400 ppm). By comparing the outcomes of the experimental data with those predicted by the Freundlich and Langmuir models, the adsorption isotherms were also explored. To assess the adsorption kinetics, the pseudo-first-order and pseudo-second-order models were also applied. The experimental tests were carried out three times, and an average of the results was provided. A UV-visible spectrometer was used to assess the amoxicillin concentration. Amoxicillin’s capacity and adsorption efficiency onto CMT-g-PANI@Fe₂O₃ bionanocomposite were calculated using Equations (1) and (2), respectively [20].

$$\mathrm{R}\%=\left(\frac{{\mathrm{C}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{e}}}{{\mathrm{C}}_{\mathrm{i}}}\right)\times 100$$

(1)

$${\mathrm{Q}}_{\mathrm{e}}=\left(\frac{{\mathrm{C}}_{\mathrm{i}}-{\mathrm{C}}_{\mathrm{e}}}{\mathrm{m}}\right)\times \mathrm{V}$$

(2)

where C_i and C_e are the amoxicillin initial and the equilibrium concentrations in the solutions (mg/L), correspondingly. m is the weight (g) of CMT-g-PANI@Fe₂O₃ bionanocomposite and V is the solution volume (L).

2.6. Isotherm Study

Langmuir and Freundlich’s isotherm models are used to evaluate the maximum adsorption capacity and equilibrium adsorption isotherms. The Langmuir isotherm model measured the single-layer adsorption of contaminants onto the adsorbent surface, whereas the Freundlich isotherm model measured the multilayer adsorption of pollutants on the adsorbent surface. The mathematical expression of Langmuir’s (Equation (3)) and Freundlich’s (Equation (4)) isotherm models are as follows [22,33]:

$$\frac{{\mathrm{C}}_{\mathrm{e}}}{{\mathrm{Q}}_{\mathrm{e}}}=\frac{1}{{\mathrm{K}}_{\mathrm{L}}{\mathrm{Q}}_{\max }}+\frac{1}{{\mathrm{Q}}_{\max }}{\mathrm{C}}_{\mathrm{e}}$$

(3)

$${\mathrm{LnQ}}_{\mathrm{e}}={\mathrm{LnK}}_{\mathrm{F}}+\frac{1}{\mathrm{n}}{\mathrm{LnC}}_{\mathrm{e}}$$

(4)

where, C_e, Q_e, and Q_max are the equilibrium concentration (mg/L), the equilibrium, and maximum adsorption capacity (mg/g) correspondingly; K_L (L/mg) is the Langmuir constant calculated from the plot between C_e/Q_e and C_e; K_F (L/mg) is the Freundlich constant calculated from the plot between Ln Q_e and Ln C_e. n is a parameter to define the adsorption process favorability; once n > 1, the amoxicillin adsorption onto bioadsorbent is anticipated at high concentrations.

2.7. Kinetic Study

The famous kinetics models were used for studying the effect of time on the adsorption process. The mathematical expression of pseudo-first-order (PFO, Equation (5)) and pseudo-second-order (PSO, Equation (6)) models are shown as follows [22,33]:

$$\mathrm{Ln}\left({\mathrm{Q}}_{\mathrm{e}}-{\mathrm{Q}}_{\mathrm{t}}\right)={\mathrm{LnQ}}_{\mathrm{e}}-{\mathrm{k}}_1\mathrm{t}$$

(5)

$$\frac{\mathrm{t}}{{\mathrm{Q}}_{\mathrm{t}}}=\frac{1}{{\mathrm{k}}_2{\mathrm{Q}}_{\mathrm{e}}^2}+\frac{1}{{\mathrm{Q}}_{\mathrm{e}}}\mathrm{t}$$

(6)

where Q_t (mg/g), and Q_e (mg/g) are the adsorption capacity at time t and equilibrium, correspondingly. k₁ (1/min) and k₂ (g/mg⋅min) are the rate constants of the PFO and PSO, correspondingly.

2.8. Desorption and Reusability

To examine the desorption and reusability of the CMT-g-PANI@Fe₂O₃ bionanocomposite, the amoxicillin adsorbed onto CMT-g-PANI@Fe₂O₃ bionanocomposite was floated in ethanol and stirred at ambient temperature for 1 h. After that, the bionanocomposite was separated by a magnet. The quantity of released amoxicillin in the elution medium was measured afterward employing a UV-visible spectrophotometer. The following equation was employed to obtain the desorption percentage.

$$\%\mathrm{D}=\frac{\mathrm{A}}{\mathrm{B}}\times 100$$

(7)

where A is the amoxicillin desorbed (mg) in the medium and B is the amoxicillin adsorbed (mg) on the CMT-g-PANI@Fe₂O₃ bionanocomposite.

3. Results and Discussion

Currently, the removal of antibiotic drugs from polluted water due to their destructive effect on humans and other living beings is very important. In this regard, we employed an antibacterial magnetic nanobiosorbent based on a carboxymethyl tragacanth gum-grafted-polyaniline and γFe₂O₃ nanoparticles (CMT-g-PANI@Fe₂O₃) for the elimination of amoxicillin from contaminated water.

3.1. Nanobiosorbent Characterization

FTIR spectra are employed for studying the chemical structure of prepared materials and are shown in Figure 2A. In the FTIR spectrum of the TG, the absorption bands at 3470 cm⁻¹, 2934 cm⁻¹, and 1740 cm⁻¹ are related to the stretching vibrations of the OH, –CH, and ester carbonyl groups, respectively [34,35]. In addition, the absorption band at 1604 cm⁻¹ is due to the existence of the carboxylate anion of d-galacturonic acid. In the FTIR spectrum of the CMT, the absorption band at 1622 cm⁻¹ is associated with the –COO⁻ asymmetric vibrations which overlapped with the carbonyl groups of acidic and ester existing in the TG parts of the CMT [31,34]. Moreover, the absorption band that appeared at 1433 cm⁻¹ is related to the bending vibrations of –CH₂. The FTIR spectrum of γFe₂O₃ nanoparticles displayed an absorption band at 550 cm⁻¹ that was related to the stretching vibrations of Fe–O. Moreover, the bands at 1620 cm⁻¹ and 3400 cm⁻¹ were associated with the –OH stretching vibrations on the surface of the γFe₂O₃ nanoparticles [36,37]. The FTIR spectrum of CMT-g-PANI displayed a broad absorption band at ~3205 cm⁻¹ and is related to the stretching vibrations of the N–H and OH of PANI and CMT, respectively. The absorption bands at 3050 cm⁻¹ and 2949 cm⁻¹ are attributed to aromatic and aliphatic C–H in the PANI and CMT, respectively. The two bands appearing at 622 cm⁻¹ and 1470 cm⁻¹ were related to the asymmetric vibrations of the –COO⁻ and stretching vibration of the benzenoid ring, respectively [38]. The presence of characteristic absorption bands of CMT, PANI, and Fe₂O₃ in the FTIR spectrum of CMT-g-PANI@Fe₂O₃ revealed that the bioadsorbent was synthesized.

The XRD analysis was employed for the investigation of the crystallinity nature of the samples as shown in Figure 2B. The XRD pattern of CMT showed an amorphous nature as compared to the XRD pattern of TG [31,39]. This crystallinity reduction may be related to the the –OH group’s replacement by the –COO⁻ groups in the CMT [34]. A crystalline nature with diffraction peaks located at 2theta = 23°, 27°, 34°, 35°, 41°, 49°, 53°, 57°, 62°, and 63° was observed in the XRD pattern of Fe₂O₃ [39]. Synthesized Fe₂O₃ has an ordered cubic structure, which is consistent with JCPDS file no. 39-1346 [40]. XRD patterns comparison of CMT-g-PANI and CMT-g-PANI@Fe₂O₃ showed that the crystallinity of CMT-g-PANI improved in the presence of Fe₂O₃.

The elemental analysis of TG, CMT, and CMT-g-PANI was performed, and the percentage mass contents of the elements are shown in

Table 1. Elemental analysis of TG, CMT, and CMT-g-PANI.

Sample	C	H	N	O
TG	42.19	5.71	-	52.1
CMT	29.66	4.07	-	66.27
CMT-g-PANI	39.88	4.38	8.03	47.71

. In CMT, the oxygen content reaches 66.27% as compared to TG which means that the substitution of –COO⁻ was carried out in the TG, successfully. Correspondingly, in CMT-g-PANI, the oxygen content decreased to 47.71%. This could be related to the grafting reaction between CMT and PANI. Furthermore, the presence of nitrogen content of 8.03% in CMT-g-PANI is related to the PANI.

EDX examination was also applied for the chemical structure study of the prepared samples as shown in Figure 3A. The existence of C, O, Cl, and Na elements in the EDX spectrum of CMT confirmed the preparation of CMT and the substitution of –COONa in the TG backbone. The existence of a tiny amount of Cl is related to chlorine removal from monochloroacetic acid due to the nucleophilic substitution reaction between the hydroxyl of TG and monochloroacetic acid. The presence of Fe and O elements in Fe₂O₃ and C, N, O, Cl, and Na elements in the CMT-g-PANI expressed that the Fe₂O₃ and CMT-g-PANI@Fe₂O₃ were prepared. In addition, the presence of C, O, N, Fe, Cl, and Na in CMT-g-PANI@Fe₂O₃ showed that the bionancomposite was prepared successfully.

Figure 3B shows the FESEM micrographs of CMT, Fe₂O₃, CMT-g-PANI, and CMT-g-PANI@Fe₂O₃. The FESEM image of CMT shows polyhedral particles with microsize. It appears that the chemical modification of TG leads to the formation of polyhedral microparticles. The nano-spherical and amorphous structures were observed in the FESEM images of Fe₂O₃ and CMT-g-PANI. The presence of Fe₂O₃ nanoparticles lead to a granular structure in the FESEM image of CMT-g-PANI@Fe₂O₃.

VSM analysis was employed to evaluate of magnetic properties of Fe₂O₃ and CMT-g-PANI@Fe₂O₃ as shown in Figure 4A. The magnetization saturation (Ms) values of Fe₂O₃ and CMT-g-PANI@Fe₂O₃ were 70 emu g⁻¹ and 25 emu g⁻¹, respectively, and showed superparamagnetic properties for both. In addition, the Ms value of Fe₂O₃ decreased with the coating of Fe₂O₃ by CMT-g-PANI.

TG analysis was applied to study the thermal stability of prepared materials as seen in Figure 4B. The TGA thermogram of TG displays two steps of mass loss from 50 to 178 °C and from 200 to 580 °C. The first mass loss (10 wt%) was related to the removal of moisture absorbed in the TG, and the second mass loss (65 wt%) corresponded to the branched heterogeneous structure degradation of the TG [34,39]. The char yield of TG at 800 °C was 25 wt%. Four mass losses at 30–150 °C, 150–300 °C, 300–550 °C, and 550–780 °C were observed in the TGA thermogram of CMT. The first (10 wt%) and second (32 wt%) mass losses were related to the moisture evaporation and combination of the saccharide ring degradation, the C–O–C breaking, and the CO₂ elimination from the CMC, respectively [34,39]. The two latter mass losses were attributed to the CMT backbone breaking [39]. The char yield of CMC at 800 °C was 33 wt%. In the TGA thermogram of CMT-g-PANI, four mass losses were observed. The first mass loss (10 wt%) at 5–200 °C was caused by the evaporation of water and solvent entrapment in the copolymer chains [41]. The second mass loss (25 wt%) at 200–300 °C probably corresponded to the loss of hydrochloric acid, saccharide ring, and the CO₂ elimination from the CMT-g-PANI. The third mass loss (10 wt%) at 300–500 °C and the fourth mass loss (20 wt%) at 500–750 °C was ascribed to the breakdown of sugar units in the CMT structure and PANI, respectively. The char yield of CMT-g-PANI at 800 °C was 34 wt%. Compared with CMT-g-PANI, the thermogram of the CMT-g-PANI@Fe₂O₃ showed good thermal stability owing to the existence of Fe₂O₃ nanoparticles in the CMT-g-PANI matrix.

Brunauer–Emmett–Teller (BET) analysis was applied to define the specific surface area of materials and evaluate the effect of the presence of Fe₂O₃ nanoparticles within the CMT-g-PANI matrix, as seen in Figure 5. The specific surface area (a_s,BET) values of CMT-g-PANI@Fe₂O₃ and Fe₂O₃ were 8.0794 m²/g and 2.8278 m²/g, respectively. This confirms that the CMT-g-PANI@Fe₂O₃ had better surface properties in terms of surface area and micropore area compared to Fe₂O₃. Thus, the presence of Fe₂O₃ together with CMT-g-PANI led to an improvement in the a_s,BET value of CMT-g-PANI@Fe₂O₃ which could be effective in antibiotic adsorption by bioadsorbent.

3.2. Optimization of Effective Parameters for Amoxicillin Adsorption

Solution pH: It is well known that the changes in the medium pH can directly affect the removal percentage of pollutants from an aqueous solution. In this regard, a solution pH range of 4–9 for finding the best pH for amoxicillin adsorption by the CMT-g-PANI@Fe₂O₃ was examined (Figure 6A). Results revealed that by growing the pH values from 4 to 7, the Q_e increased from 54.18 mg/g to 65.39 mg/g and then decreased at high pH (8 and 9). At pH 7, amoxicillin antibiotic occurs as zwitterions, while at pH 4 and 5, it performs in cationic species, and at pH 8 and 9, it appears in anionic species. Thus, amoxicillin removal was efficient under pH 7 (neutral). The hydroxyl, carboxyl, and amine groups in the copolymer together with a high surface-to-volume ratio of Fe₂O₃ in the CMT-g-PANI@Fe₂O₃ adsorbent lead to the rapid diffusion of amoxicillin molecules into the adsorbent and interaction with its functional groups for efficient amoxicillin.

Amount of biosorbent: Adsorption examinates were carried out with changing quantities (0.005–0.025 g) of the CMT-g-PANI@Fe₂O₃ at pH 7 (optimal) to investigate the association between the adsorbent amount and its adsorption capability for amoxicillin. According to Figure 6B, the Q_e decreased from 189.73 to 34.50 mg/g as the adsorbent amount increased from 0.005 g to 0.025 g. At a lower adsorbent amount, the higher the amount of amoxicillin adsorbed by the CMT-g-PANI@Fe₂O₃. The optimal quantity of CMT-g-PANI@Fe₂O₃. for subsequent examinations was found to be 0.005 g.

Agitation time: The agitation time effect on the Q_e of the CMT-g-PANI@Fe₂O₃ adsorbent for the removal of amoxicillin was examined (Figure 6C). Results showed that the Q_e increased up to 195.49 mg/g by increasing the agitation time from 5 to 20 min at pH 7 with an adsorbent dose of 0.005 g (optimal). Furthermore, the Q_e decreased slightly to 183.93 mg/g once the agitation time reached 30 min. Consequently, an agitation time of 20 min was chosen to be the perfect time for subsequent examinations. Generally, at the beginning of the adsorption process, numerous vacant sites of adsorbent are vacant to interact with amoxicillin molecules in solution. The functional group’s interaction of the adsorbent and the amoxicillin grow stronger once the agitation time is increased to 20 min. At longer time frames (>20 min), there was no more improvement in Q_e, which was probably related to the existence of adsorbent active sites and the equilibrium state approach.

Amoxicillin initial concentration: The association between the amoxicillin initial concentration and the Q_e of the CMT-g-PANI@Fe₂O₃ adsorbent was studied by altering the amoxicillin concentration from 50 to 400 mg/L at the optimal pH (7.0), dose (0.005 g), and agitation time (20 min). Figure 6D shows that the amoxicillin initial concentration influences the Q_e of CMT-g-PANI@Fe₂O₃ adsorbent. With increasing the amoxicillin initial concentration from 50 to 400 mg/L, Q_e of adsorbent with an intensity rose to 788.83 mg/g with a steep slope. Thus, the Q_e increases once the amoxicillin initial concentration is increased, whereas the adsorbent amount remains constant, and this pattern holds until the amoxicillin initial concentration reaches an equilibrium level.

3.3. Adsorption Isotherm

To find the interaction between amoxicillin and CMT-g-PANI@Fe₂O₃ bioadsorbent, the isotherm study was employed. It is well known that Langmuir and Freundlich’s isotherm models are used to evaluate the maximum adsorption capacity and equilibrium adsorption isotherms. Figure 7A,B demonstrates the Langmuir and Freundlich isotherm. The obtained parameters of isotherm models are tabulated in

Table 2. Isotherm parameters, for amoxicillin adsorption onto the CMT-g-PANI@Fe₂O₃ bioadsorbent.

Isotherm	Parameters
Freundlich	KF	88.8633
	n	1.99
	R2	0.9918
Langmuir	Qmax	909.09
	KL	0.0502
	R2	0.9097

. Consistent with obtained correlation coefficient (R²) of isotherm models, the Freundlich isotherm was found to be more reliable with the experimental data than the Langmuir model. This result shows that the amoxicillin adsorbed over the CMT-g-PANI@Fe₂O₃ bioadsorbent surface as a multilayer. In addition, the obtained “n” in the Freundlich model is an important parameter to define the adsorption process favorability. The value of n > 1 in Table 2 shows that the amoxicillin adsorbed over the CMT-g-PANI@Fe₂O₃ bioadsorbent surface is desirable at high concentrations.

The comparison of Q_max for amoxicillin to other adsorbents reported in recent years showed that the CMT-g-PANI@Fe₂O₃ biosorbent has a higher maximum adsorption capacity (909.09 mg∙g) than other adsorbents (

Table 3. Assessment of the maximum adsorption capacity of CMT-g-PANI@Fe₂O₃ bionanocomposite with other reported studies.

Adsorbent	Experimental Conditions	Qmax (mg/g)	Ref.
MIL-53(Al) metal-organic framework	pH: 7.5, adsorbent dosage: 0.1 g/L, contact time: 12 h, T: 303 K, initial concentration: 50 mg/L	758.5	[22]
NH4Cl-induced activated carbon	pH: 6, adsorbent dosage: 0.8 g/L, contact time: 30 min, T: 303 K, initial concentration: 100 mg/L	437	[23]
Activated carbon/iron nanoparticles	pH: 5, adsorbent dosage: 1.5 g/L, contact time: 30 min, T: 298 K, initial concentration: 10 mg/L	40.282	[24]
Functional CoFe2O4-modified biochar	pH: 7, adsorbent dosage: 0.05 g/L, contact time: 60 min, T: 298 K, initial concentration: 50 mg/L	99.99	[25]
Activated carbon	pH: 5.5, adsorbent dosage: 0.06 g/L, contact time: 240, T: 298 K, initial concentration: 100 mg/L	4.4	[42]
Mg-Al layered double hydroxide/cellulose nanocomposite	pH: 7, adsorbent dosage: 0.1 g /L, contact time: 20 min, T: 298 K, initial concentration: 120 mg/L	138.3	[43]
Concrete-based hydrotalcite	pH: 5, adsorbent dosage: 0.1 g/L, contact time: 12 h, T: 298 K, initial concentration: 400 mg/L	49.7	[44]
Chitosan-coated Fe3O4@Cd-MOF microspheres	pH: 8, adsorbent dosage: 0.05 g/L, contact time: 240 min, T: 298 K, initial concentration: 50 mg/L	103.09	[45]
Fe3O4/SiO2/CTAB–SiO2	pH: 5, adsorbent dosage: 0.009 g/L, contact time: 60 min, T: 298 K, initial concentration: 25 mg/L	362.66	[46]
Fe3O4@activated carbon	pH: 6, adsorbent dosage: 0. 5 g/L, contact time: 90 min, T: 298 K, initial concentration: 200 mg/L	238.1	[47]
CMT-g-PANI@Fe2O3	pH: 7, adsorbent dosage: 0.005 g/L, contact time: 20 min, T: 298 K, initial concentration: 400 mg/L	909.09	Current work

). This could be related to the presence of Fe₂O₃ nanoparticles and numerous active sites, for example, hydroxyl, carboxylate, and amine groups in the nanocomposite, which can effectively interact with amoxicillin (via electrostatic interaction and hydrogen bonding), led to eliminating the amoxicillin.

3.4. Adsorption Kinetic

For discovering and evaluating possible adsorption routes, equilibrium time, and adsorption rate-limiting phase of amoxicillin adsorbed over the CMT-g-PANI@Fe₂O₃ adsorbent used the PFO and PSO kinetics models. In addition, kinetic models are employed to define the adsorption process mechanism, for example, surface and chemical adsorption. The PFO is based on the adsorbent capacity and employed once adsorption happens via diffusion in a border layer, whereas the PSO expresses that the chemical adsorption procedure is the main controlling procedure. Figure 8A,B and

Table 4. Kinetic parameters, for amoxicillin adsorption on the CMT-g-PANI@Fe₂O₃.

Models	Parameters
	k1	0.0743
Pseudo-first-order	Qe calculated	38.122
	Qe experimental	788.73
	R2	0.9851
	k2	0.000001
Pseudo-second-order	Qe	833.33
	Qe experimental	788.83
	R2	0.9900

exhibit kinetic linear plots, and their calculated parameters. According to R², the difference between the estimated Q_e and observed Q_e values, the PSO is more appropriate for considering the amoxicillin adsorption kinetics on the CMT-g-PANI@Fe₂O₃. Consequently, chemical adsorption is the main mechanism in the adsorption of amoxicillin on the adsorbent and it displays that, in addition to amoxicillin molecules, the CMT-g-PANI@Fe₂O₃ adsorbent is involved in the adsorption procedure.

3.5. Desorption and Reusability

The reusability of adsorbent is a significant factor to create the adsorption procedure economically. The reusability of adsorbent can be effective for its use in numerous consecutive cycles without considerable performance debility. In the present work, we evaluated the amoxicillin desorption and reusability of CMT-g-PANI@Fe₂O₃ bioadsorbent in three consecutive cycles, and an ethanolic solution was investigated. In this regard, amoxicillin adsorbed onto the CMT-g-PANI@Fe₂O₃ was performed in optimal conditions and was immersed into ethanol (10 mL) under stirring at room temperature for 1 h to desorb amoxicillin. Afterward, CMT-g-PANI@Fe₂O₃ was collected by a magnet, washed with distilled water, and then dried for successive adsorption/desorption processes. Then, the released amoxicillin amount in the elution medium was measured using a UV-visible spectrophotometer. Figure 9 demonstrates that the adsorption percentage decreased from 98.59 % to 94.03%, and the desorption percentage decreased from 94.41% to 92.001% after three consecutive cycles, these results show that the CMT-g-PANI@Fe₂O₃ could continue to remove amoxicillin after three consecutive cycles of adsorption/desorption process without considerably losing adsorption capacity.

3.6. Suggested Mechanism of Amoxicillin Adsorption

The pollutant adsorption mechanism by the adsorbent depends on the functional groups and morphology of the adsorbent, which can be created a synergistic effect in increasing the pollutant adsorption by the adsorbent. Considering that the CMT-g-PANI@Fe₂O₃ bioadsorbent has amine, carboxylate, and hydroxyl functional groups, it can create intermolecular interactions such as hydrogen bonding, electrostatic and π–π interactions with the amoxicillin antibiotic, and cause its efficient adsorption on the CMT-g-PANI@Fe₂O₃ bioadsorbent as shown in Figure 10.

4. Conclusions

A magnetic bionanocmposite was synthesized by an in situ copolymerization method from carboxy methyl tragacanth gum in the presence of aniline monomer and γ Fe₂O₃ nanoparticles (CMT-g-PANI@Fe₂O₃). The CMT-g-PANI@Fe₂O₃ bionanocmposite was characterized by several analyses and employed as a bioadsorbent for the removal of amoxicillin antibiotic from contaminated water. A granular structure with good thermal stability (char yield 34 wt%), magnetization saturation (25 emu g⁻¹), and a specific surface area (8.0794 m²/g) observed for CMT-g-PANI@Fe₂O₃ bionanocmposite. The maximum adsorption capacity (909.09 mg/g) was found at pH 7, an agitation time of 20 min, an adsorbent dose of 0.005 g, and an amoxicillin initial concentration of 400 mg/L. The Freundlich isotherm and pseudo-second-order kinetic models fit closely with the experimental data. The adsorption/desorption process showed that the CMT-g-PANI@Fe₂O₃ could continue to remove amoxicillin after three consecutive cycles without considerably losing adsorption capacity. The intermolecular interactions such as hydrogen bonding, electrostatic, and π–π interactions between amoxicillin antibiotic and CMT-g-PANI@Fe₂O₃ were suggested for the adsorption mechanism of amoxicillin antibiotic by bionanocomposite.