Heterogeneous degradation of amoxicillin in the presence of synthesized alginate-Fe beads catalyst by the electro-Fenton process using a graphite cathode recovered from used batteries

The presence of human antibiotics in the environment has become a serious environmental concern. The concentrations of these compounds have been reported to be in the range of (Elizalde-Velázquez et al. 2016). Even at low concentrations, the presence of antibiotics in surface and groundwater constitutes a great threat for the ecosystem (Kadji et al. 2021). Thus, exposure to antibiotics can lead not only to growth stunting and bodyweight reduction in aquatic organisms, but also to the development of antibiotic resistance genes in pathogenic bacteria, which constitutes a dangerous threat to human health, since the effectiveness of the treatment of bacterial infections is reduced (Kumar et al. 2021). The beta lactams represent the major pharmaceutical products; they constitute 65% of the world market for antibiotics (Mojiri et al. 2019). Among them, amoxicillin (AMX) is a broad-spectrum beta-lactam antibiotic, and it is one of the most prescribed antibiotics. AMX is widely used for the treatment of a variety of bacterial infections. However, about 60% of the prescribed and taken AMX is excreted unchanged (Dogan & Kidak 2016) and hence is released in wastewater. This presents a potential risk for the environment and human beings (Li et al. 2015). Chemical and physicochemical techniques have been widely studied to remove antibiotics from water such as adsorption (Boudrahem et al. 2019; Saidi et al. 2019; Akkouche et al. 2021). Adsorption is one of the most effective and low-cost methods, but this technique has some disadvantages such as being non-destructive and merely transferring pollutants from one phase to another, which always results secondary pollution (Madi-Azegagh et al. 2018a, 2018b). The biological methods such as activated sludge process and anaerobic treatment, the most cost-effective for wastewater treatment, which are destructive and have been extensively studied, do not always appear relevant for the removal of pollutants, owing to their low biodegradability (Ikhlef-Taguelmimt et al. 2020; Kadji et al. 2021). However, these techniques are not adapted for the treatment of antibiotics residues owing to their recalcitrance (Annabi et al. 2016; Dogan & Kidak 2016).

Recently, scientists give more importance to heterogeneous electro-Fenton process, which is considered as an environmentally friendly approach, since the catalyst is supported on a solid material. This permits easy separation and recovery of the catalyst and minimizes the production of waste sludge (Zárate-Guzmán et al. 2019).

This study, in great detail, focused in a first part on the recovery of graphite rod from used batteries to use it as a cathode in the electro-Fenton process, since the graphite cathode is the best material for H₂O₂ electro-generation (Nayebi & Ayati 2021) and to reduce the electrode cost. In a second part, the immobilization of iron in the biomaterial by using alginate beads, which rectifies some of the disadvantages of the conventional electro-Fenton process. Furthermore, the efficiency of iron alginate beads (Fe-Alg) and the graphite rod for the degradation of AMX by the heterogeneous electro-Fenton treatment was tested. For this purpose, the effects of operational parameters such as the applied current intensity, the temperature and the initial AMX concentration on the degradation of AMX were studied.

One of the most toxic and chemical-rich wastes is used batteries. The implementation of measures for the control of these waste batteries is still insufficient in developing countries; furthermore, techniques for the recycling of this waste are not yet available. The electrode is made of graphite, and has many applications if we mention only a few: it is used as a lubricant in motor oils, it is used as a mould in the manufacture of mineral compounds (ferro-alloys), and it is also used in pencil mines. The aim of this study is therefore to recover the electrode to use it as a cathode in the electro-Fenton process. The samples to be studied are cylindrical R20 batteries with an electromotive force of 1.5 V (32 mm × 61 mm). The steps for dismantling the used battery are listed below:

• Use the screwdriver to remove the plastic cover.

• Saw off the top part along its width in order to remove the plastic cover and the cap.

• Saw the casing lengthwise to remove the top part, remove the washer and then the plastic layer to access the black powdery mixture between the zinc cylinder and the graphite electrode.

• Recover the graphite electrode (see figure S1, supplementary material), wash it and sand it with abrasive paper.’

The target compound was amoxicillin (C₁₆H₁₉N₃O₅S) with purity of 99%; it was purchased from Sigma-Aldrich and its characteristics are presented in table S1 (supplementary material). Na₂SO₄ (99% purity), H₂SO₄ (96% purity), Fe₂O₁₂S₃ (96% purity), CaCl₂ (96% purity), CH₄O (99% purity), KH₂PO₄ (99.5% purity), and sodium acetate anhydrous (99% purity) were purchased from Biochem Chemopharma. Sodium alginate and 1,10- phenanthroline (99% purity) were obtained from Sigma-Aldrich. hydroxylamine hydrochloride (NH₂OH·HCl, 98% purity), was purchased from Labosi.

Two electrodes (see description in our previous work (Kadji et al. 2021) were connected to a DC supply (Model GW insTEK GPS-2303) and an electric field was applied. All electrolyses were carried out in a 600 mL undivided cylindrical Pyrex glass cell (see figure S2, supplementary material). The electrodes were centred in the bulk, the inter-electrode distance was 1 cm. The synthetic aqueous solutions treated were prepared by dissolving a certain amount of AMX in distilled water (2.5, 5 and 15 mg of AMX was used to prepare a solution with a concentration of 0.0136, 0.027 and 0.082 mM, respectively). And then the pH solution was adjusted to 3 using sulphuric acid. Prior to the electro-Fenton experiment, the solution was saturated by oxygen contained in the air by means of a pump, and then 3.1 g of Alg-Fe beads were added to the electrolysis cell.

Numerous research papers report the synthesis of Alg-Fe beads as catalyst, employing different conditions and preparation methods (Rosales et al. 2012; Iglesias et al. 2014; Hammouda et al. 2016).

The Alg-Fe beads were prepared by extrusion (see figure S3, supplementary material). A mixture containing the gelation agent, calcium chloride (2% w/v) and the solution of sulfate iron was prepared and maintained under vigorous stirring. The concentration of ferric ions in the solution was 0.02 M. The synthesis of the alginate beads was ensured by dropwise addition of alginate sodium solution (3% w/v) to the solution containing the crosslinking agent using a syringe. A rapid reaction occurred between the alginate and the crosslinking agent on the surface, leading to instantaneous formation of beads with spherical shape. The drip was done through a needle of 3 mm internal diameter.

The maturation time of alginate beads was fixed to 24 hours. This period is largely sufficient to ensure complete gelation of the alginate. After that, the beads were filtered and washed with distilled water to remove excess chloride and calcium ions. Wet beads obtained were dried at 40 °C during 48 h.

Surface morphology of beads was analysed with scanning electron microscopy (SEM) coupled to energy dispersive spectrometric (EDS) to determine the chemical composition of the beads (model JOEL JSM7_7610F PLUS).

Based on related studies (Hammouda et al. 2016), the total iron contained in the beads was determined by acid digestion using concentrated sulfuric acid. The colorimetric o-phenanthroline method was used to determine the concentration of iron; 1,10-phenanthroline reacts with the ferrous ion Fe²⁺ to form the tris (1,10-phenanthroline) Fe II, also called ferroin: [Fe(o-phen)3]²⁺, a red-orange complex. The absorbance was measured at by UV visible spectrophotometer (Thermo-scientific evolution 2001). The full method was described by Saywell & Cunningham (1937). This method was used to determine the iron concentration during the treatment and the iron content in the beads.

The quantification of the concentration of AMX was determined by high performance liquid chromatography (HPLC ACC 3000 HPLC). Samples were taken and filtered through 0.22 μm membrane syringe filter (Satorius Stedim biotech Gmbh,Germany). The HPLC was equipped with a standard degasser (LPG 3400 SD), an autosampler, pump (Model LPG 3400 SD) and a detector with visible ultraviolet ray (UV/Vis detector VWD 3400 RS). The separation was achieved on a Thermo Fisher scientific (Germany) C18 (5 mm; 4.6 ×150 mm) reversed-phase column. The mobile phase consisted of CH₄O/KH₂PO₄ (5/95 v/v) with a flow rate of 0.5 mL/min and the detection of AMX was carried out at 232 nm.

All BOD₅ measurements were duplicated and average results were used. The determination of BOD₅ was carried out in Oxitop in the presence of added nutrients according to EN 1899-1-H51. Additionally, the probable influence of nitrification processes was inhibited by N-allythiourea. The incubation of the samples was carried out directly in Oxitop in the presence of activated sludge from a waste treatment plant (Sidi Ali Labhar, Béjaia, Algeria) and the determination of oxygen dissolved in water was carried out after 5 days at 20 °C. The chemical oxygen demand (COD) was measured by means of Kits Nanocolor^® 15–160 mg/L COD according to DIN ISO 15705 at 148 °C. The amount of oxygen required for the oxidation of the organic and mineral matter at 148 C for 2 h was quantified after oxidation with K₂Cr₂O₇ at acidic pH and heating. COD was measured by Nanocolor 500D photometer type (Macherey-Nagel, Hoerd, France) (Yahiaoui et al. 2016, 2018).

Calculation of the iron retained in the alginate revealed that the composition in iron beads was 4.63% w/w.

The images in Figure 1(a) and 1(c) show that all beads have a spherical shape, in agreement with the related literature (Fundueanu et al. 1999; Rosales et al. 2012). The iron-free alginate (Alg) beads (Figure 1(a) and 1(b)) have a homogeneous and smooth surface with undulations. On the opposite, a change in the texture of the alginate-iron beads (Alg-Fe) was noticed as it is illustrated in Figure 1(c)–1(e), due most likely to the presence of iron in the Alg-Fe beads. These beads had a rough and heterogeneous surface with less apparent fissures or fractures. This difference in texture shows that the iron interacted and was efficiently dispersed in the alginate beads by modifying their initial texture through physico-chemical interactions.

The composition of iron, calcium and sodium contained in the beads was determined by EDS. The results of this analysis for the different synthesized beads (Alg and Alg-Fe) are depicted in Figure 2. The EDS spectrum of iron-free alginate beads Figure S4(a) showed a significant peak of calcium, while the peak sodium ions was very weak. Indeed, the sodium ions initially present in the alginate were replaced by Ca²⁺ ions by ion exchange. The analyses performed on the Alg-Fe beads are shown in Figure 2(b). The spectrum shows the presence of iron in the sample and the absence of calcium, while some of the sodium ions were still present Figure S4(b), but the remaining amount was almost insignificant. This difference in composition shows that the iron encapsulation was effective.

The adsorption test was performed in the dark. A synthetic solution of 0.082 mM AMX was prepared. The pH was adjusted to 3 since the optimum pH value for the Fenton reaction is about 3. According to previous research studies on the electro-Fenton process, it is found that acidic pH values close to 3 are the most favourable for the electro-Fenton oxidation (Huang et al. 2017; Rezgui et al. 2018), in fact, iron acts as a catalyst with maximum catalytic activity at acidic pH (Miklos et al. 2018). Thus, the oxidation potential of hydroxyl radicals decreases with increasing pH. This is due, on the one hand, to the fact that the decomposition of H₂O₂ into water and oxygen is accelerated at pH values above 5 and, on the other hand, to the formation of ferric hydroxides at a pH above 4, which reduces the degradation rate (Ghanbari & Moradi 2015; Aramyan & Moussavi 2017; Mirzaei et al. 2017). The catalyst was added to the reaction medium after pH adjustment. After 150 min the removal yield was only 1%; this clearly indicates that adsorption can be neglected, the same findings were reported by Rosales et al. (2012) and Hammouda et al. (2016) using Alg-Fe beads as catalyst in the the electro-Fenton process for the decolourisation of dyes and the degradation of indole, respectively.

The first parameter examined was the influence of the current intensity, since it exerts a significant effect on the degradation ability of the EF-Alg-Fe process (Electro-Fenton Alg-Fe process) (Mansour et al. 2012; Annabi et al. 2016). Trials were performed by varying the applied current intensity from 50 to 600 mA. The Figure 3(a) exemplified the results of AMX removal. The degradation yield was improved as the current intensity increased, from 86.22% to an almost total degradation at 400 mA, which should be related to a higher production of •OH in the system. In addition, increasing the current intensity should lead to an improvement of the ferrous ion regeneration by cathodic reduction. Accordingly, the COD abatement also increased (Figure 3(b)) with the current intensity, until 82.35% for 600 mA. Meanwhile, the optimal operating current intensity was 600 mA.

The variation in iron concentration as a function of the applied electric current was followed during treatment (see figure S5, supplementary material). The results indicated that the increase in current intensity in the 50–600 mA range strongly affects the total iron concentration. Indeed, the amount of iron increased with current intensity, from a negligible amount for 50 mA to79.71 mg·L⁻¹ for 600 mA. This significant increase of iron concentration observed is probably due to the fact that the release of iron from the beads is favored at high current intensities. The amount of ferrous ions released into the solution increased at high currents leading to an increase in the number of •OH radicals, which is associated with an increase in the COD abatement rate. This suggests that a homogeneous fenton reaction has been occurred (Mirzaei et al. 2017).

To check the influence of this parameter on the AMX's degradation, assays were carried out at 25, 45 and 55 °C, leading to an almost total removal after 120 min of reaction time in the range of temperatures tested, as evidenced by Figure 4(a). However, a slight increase of the degradation rate for increasing temperatures should be noted, as it is confirmed at the examination of the rate constant, the values were 0.023, 0.0269, and 0.0292 for 25, 45, and 55 °C, respectively.

It can be seen from Figure 4(a), the increase of the temperature did not significantly improve the AMX's removal and irrespective of the temperature total removal was observed at 120 min reaction time. This might be attributed to the decomposition of hydrogen peroxide to oxygen and water at high temperatures according to (Equation (9)) (Mirzaei et al. 2017). In addition, oxygen dissolution diminishes as the temperature rises, so less hydrogen peroxide is produced via (Equation (3)) (Mansour et al. 2012); this slows down the Fenton's reaction, and subsequently there are fewer hydroxyl radicals in the solution. Moreover, high temperatures may deteriorate the catalyst to small fragments (Mirzaei et al. 2017).

The variation of iron concentration at different temperatures was examined. The total iron concentration after 120 min of reaction time was 79.71 and 76.19 mg·L⁻¹ for 25 and 45 °C, respectively, which does not show a significant impact between 25 and 45 °C; while further increase to 55 °C considerably increased the iron concentration to 108.12 mg·L⁻¹. The increase in iron concentration at a high temperature could be attributed to the increased leaching of iron from the beads into the solution, as ferric ions are easily leached from the beads at high temperature. In addition, we found that the apparent degradation rate constants increased with temperature. This can be explained that degradation takes place by the homogeneous Fenton reaction resulting from the presence of high concentrations of iron in the solution (Mirzaei et al. 2017).

The line giving the variation of ln(k_app) versus (1/T) is illustrated in Figure 4(b). The energy of activation was found to be 6.42 KJ/mol. It is in agreement with the activation energy value (7.5 kJ/mol) reported by Mansour et al. 2012, for the degradation of sulfamethazine by electro-Fenton process. Some researchers calculated the activation energy for the degradation of some refractory organic compounds by using a dark Fenton-like process, a photo Fenton-like process (Ifelebuegu et al. 2016) and thermally activated persulfate (Zhao et al. 2019); the results were 104 kJ/mol, 42 kJ/mol and 126.9 kJ/mol, respectively. Comparison of these values to the one found in this study allowed to conclude that a low energy is enough to realize the degradation's reaction of AMX by a heterogeneous electro-Fenton process.

Concerning the effect of the initial AMX concentration, the results obtained by conducting experiments at 0.082 and 0.027 and 0.0136 mM are represented in Figure 5. As observed, the degradation rate of the AMX was inversely proportional to its initial concentration. Indeed, for 0.0136 mM the degradation rate was 100% within 90 min of reaction time. The time became longer with higher AMX concentration, as revealed for 0.027 and 0.082 mM, the AMX disappeared only after 120 min of treatment, which is expected because more organic matter is oxidized by the same amounts of generated hydroxyl radicals, causing then competitive consumption of hydroxyl radicals between intermediate products and the target molecule.

Based on the kinetic results illustrated in Figure 6 and Table 1, the curves (Figure 6(a) and 6(b)) suggest that the model of pseudo-first order kinetic was the most appropriate for the concentrations of 0.082 and 0.027 mM. These results are in an agreement with other research studies carried out on the degradation of refractory organic compounds by the heterogeneous electro-Fenton process, which employed different catalysts and showed that the degradation kinetic follows a pseudo-first-order model (Kalantary et al. 2018; Liu et al. 2018). In addition, the calculated first-order rate constants in this study are in the same range compared to other studies. For example, Kalantary et al. (2018), in their study of AMX degradation by a heterogeneous electro-Fenton process using Nano-Fe₃O₄ as a catalyst, indicated that 89.1% of AMX was degraded in 1 h of reaction time, with a rate constant of 0.0457 min⁻¹. Liu et al. (2018) treated a pharmaceutical contaminant, ibuprofen using iron supported on activated carbon fiber cathode, the rate constant was determined to be 0.0288 min⁻¹. Otherwise, hydroxyl radicals do not accumulate in the reaction medium (Antonin et al. 2015), since they are very reactive species, with very short half-life (10⁻³ μs) (Antonin et al. 2015; Mirzaei et al. 2017).

As presented in Table 1, it is worth noting that the kinetic degradation of AMX was impacted by the initial concentration of AMX; the rate constant of degradation diminished as the initial concentration of AMX increased. The same observations were reported in the literature studies (Mansour et al. 2012; Annabi et al. 2016). This behaviour results, on the one hand, from the non-selectivity of the hydroxyl radicals that react with the target pollutant and the generated intermediate compounds. On the other hand, the amount of produced hydroxyl radicals is constant, while the quantity of organic molecules present in the solution bulk increases (Annabi et al. 2016).

The reusability of the Alg-Fe catalyst was studied. To reveal the impact of this crucial factor, assays were conducted in the optimal operating conditions. As illustrated in Figure 7, the degradation efficiency after each cycle was nearly similar to that of the first cycle; after 5 cycles of use, the Alg-Fe catalyst depletion was only 8%. Based on these results, the performance of Alg-Fe catalyst was demonstrated.

To assess the biodegradability after the electro-Fenton process, the BOD₅/COD ratio was examined for two initial concentrations of AMX, 0.0136 and 0.082 mM. The corresponding results are presented in Figure 8(a) and 8(b), respectively. When the initial concentration of AMX was 0.0136 mM, the corresponding initial BOD₅/COD ratio was 0.07, which diminished to reach 0 after 60 min of reaction time, as demonstrated in Figure 8(a). A total removal of AMX was obtained within 90 min of pretreatment (Figure 5), the corresponding COD abatement and BOD₅/COD ratio were 50% and 0.36, respectively.

From the results displayed in Figure 8(b), it can be seen that the BOD₅/COD ratio of the AMX solution (before treatment) was 0.01, indicating very low biodegradability. During electro-Fenton treatment, the removal of refractory contaminants can be achieved by means of hydroxyl radicals, which break down the parent molecules of the pollutants into more biodegradable intermediates which are more easily degraded by microorganisms. As a result, the biodegradability was improved to 0.22 after 2 hours of reaction time and the corresponding COD abatement yield reached 82.3%.

To understand the low biodegradability of the degradation by-products generated during the oxidation process, the related literature was explored. Trovó et al. (2011) studied the degradation of AMX by the photo-Fenton process, they noticed that the intermediates generated during treatment are more toxic than the AMX. The results indicated that diketopiperazine-2,5 is the principle degradation product of AMX. The same finding was reported by Lamm et al. (2009). The photocatalytic degradation of amoxicillin was investigated by Çağlar Yılmaz et al. (2020), the identification of the degradation by-products revealed the formation of two main products: phenol followed by benzene. These two compounds are known to be toxic. This clearly explains the low biodegradability of the electrolyzed solutions of amoxicillin. The increase of the BOD₅/COD ratio after the electro-Fenton treatment indicates that a biological treatment could be envisaged, even though the limit of biodegradability, 0.4, was not attained (Carboneras Contreras et al. 2020). However, the COD abatement corresponding was 82.3%, which suggests that a large amount of organic matter had been oxidized. From this, subsequent biological treatment may not be relevant for the degradation of AMX by-products owing to their low biodegradability and the high oxidation of the solution, namely the low organic compounds remaining in the solution.

Developing stable, active and efficient catalyst material for environmental application is necessary. In this work, the synthesis, characterization and the application of iron supported by alginate material were reported in this study. The iron content of the Fe-Alg beads was determined to be 4.63% w/w. The use of this heterogeneous catalyst demonstrated its efficiency in overcoming excess iron ions. Indeed, after 5 cycles of use, the depletion of the Alg-Fe catalyst was only 8%, and thus the amount of sludge generated is reduced.

Furthermore, a complete degradation of AMX by the electro-Fenton treatment was obtained after 5 degradation cycles under the following conditions: initial AMX concentration 0.0136 mM, 600 mA current intensity, and 90 minutes of reaction time. In addition, a removal efficiency of 100% and a COD reduction of 85.7% were obtained after only 90 and 150 minutes of reaction time, respectively.

The results of the kinetic modelling using the non-linear method showed that at low concentration, the degradation reaction follows a pseudo-second-order kinetic. Biodegradability tests showed that after treating the AMX solution with the electro-Fenton process, the biodegradability ratio (BOD₅/COD) was increased from 0.07 to 0.36 namely close to the limit threshold (0.4) revealed the improvement of the biodegradability of the solution.

The results obtained in the present study indicate that the heterogeneous electro-Fenton process is an efficient method for the removal of refractory organic compounds.

All relevant data are included in the paper or its Supplementary Information.