Pencil graphite supported nano zero-valent iron for removal of levofloxacin from aqueous solution: Effects of pH, kinetic and biological activity

2.1 Chemicals

Ferric chloride hexahydrate (FeCl₃·6H₂O, ≥ 99 %), Absolute ethanol (C₂H₅OH, ≥ 99 %), Sodium borohydride (NaBH₄, ≥ 97 %), Pencil lead graphite (HB, 0.5 mm), and Levofloxacin (C₁₈H₂₀FN₃O₄, ≥ 99 %), Milli-Q water (Resistivity = 18.2 MΩ.cm), Nitrogen gas N₂ ≥ 99 %, Hydrochloric acid (0.1 M HCl), Sodium hydroxide (0.1 M NaOH), and sodium chloride (0.1 M NaCl). All bacterial strains (Escherichia coli, Staphylococcus aureus, Proteus mirabilis, Klebsiella pneumonia, Staphylococcus epidermidis, and Enterococcus faecalis) were donated by the Department of Biology and Biochemistry at Birzeit University.

2.2 Synthesis of pencil graphite supported zero-valent iron nanoparticles

Pencil graphite supported nano zero-valent iron (PG-Fe⁰) was synthesized using the liquid-phase reduction method (Chen et al., 2011).

• 4Fe³⁺ + 3BH₄^- + 9H₂O → 4Fe⁰ + 3H₂BO^3- + 12H⁺ + H₂↑(g).

The previously established method was used to synthesize the PG-Fe⁰ composite (Al-Rimawi et al., 2022). The preparation consisted of PG-Fe⁰ with an iron/pencil graphite mass ratio of (1:1). 4.84 g of Ferric chloride hexahydrate (FeCl₃·6H₂O) was dissolved in 100 mL of miscible liquids (Milli-Q water and absolute ethanol at a volume ratio of 1:4 v/v), followed by 4.84 g of milled pencil graphite were added to the solution. The mixture was stirred vigorously under a nitrogen atmosphere for 15 min, then 100 mL 0.47 M NaBH₄ solution was added dropwise to the mixture at a rate of 1–2 drops per second. After all of the NaBH₄ solution had been added, the mixture was continuously stirred for a further 20 min under a nitrogen atmosphere to completely deplete NaBH₄ and FeCl₃·6H₂O The composite was thoroughly washed with water and ethanol before being collected and dried under vacuum at room temperature. The composite was then sealed in a bottle for future use, and the amount of Fe⁰ NPs on the surface of PG was estimated to be 160 mg/g.

2.3 Characterization techniques

The diffraction patterns of PG-Fe⁰ composite were obtained using mini-XRD (Rigaku Ultima, Tokyo, Japan) with a CuK α source and 2θ angles in the range from 0−65° at a slow scan rate of 2θ/ min. A UV–vis Spectrophotometer (Hp 8453, Agilent, USA) was used to measure solution absorption over a wavelength range of 190–1100 nm. Fourier transform infrared (FTIR) spectra were recorded in the range of 4000–230 cm⁻¹ using (a Bruker TENSOR II Spectrometer), and sample pellets were prepared by combining the analyte with KBr in a corresponding ratio of 1:200 (w/w). The composite images were captured using the JOEL-2100F field emission transmission electron microscope (FE-TEM) and scanning electron microscopy (SEM) coupled with energy dispersive X-rays (SEM-EDX) (Jeol 6700LV).

3.1 Effect of pH

At different initial pH values (3.0, 5.0, 6.5, 8.0 and 10.0) 100 mL of 20 mg/L LEV solutions were prepared. The solutions were adjusted to the appropriate pH using 0.1 M NaOH and 0.1 M HCl solutions. Each solution received 0.5 g of PG-Fe⁰ composite before being shaken in a water bath shaker for 3 h at a constant temperature of 298 K and 125 rpm. When equilibrium was established, the concentration of LEV was determined by taking a 5 mL aliquot from each solution and measuring the absorption of LEV with a UV–visible spectrophotometer. Then, the concentration of LEV in each solution was calculated.

3.2 Effect of adsorbent dose

To evaluate the optimal amount of adsorbent dose required for LEV adsorption, a different dose of the PG-Fe⁰ composite (0.1, 0.5, 1.0, and 1.5 g) was applied to 100 mL of 25 mg/L LEV solution. The mixtures were shaken at 125 rpm for 3 h at 298 K and pH 6.5. After equilibrium was established, each solution was filtered through a 0.45 µm filter, and the removal efficiency in each solution was calculated.

3.3 Effect of contact time and adsorbate concentration

The kinetic study was carried out at two different pH values (pH = 6.5 and pH = 8). For that, concentrations of LEV solutions were prepared (10, 20, 25, 30, and 35 mg/L) at appropriate pH, followed by the addition of 0.5 g of adsorbent material. After that, all solutions were shaken for 3 h at 125 rpm and 298 K. All samples were analyzed at various time intervals (5, 10, 15, 20, 30, 40, 80, and 320 min). At the end of the contact time, the adsorbent composite was collected by an external magnetic field and the aliquot was filtered through 0.45 µm membrane filters. The LEV absorption in the supernatant solution was analyzed before and after the adsorption process, and the concentration of LEV was calculated using a calibration curve (Fig.S1). The removal efficiency of the adsorbent was determined according to Eq. (1),

3.4 Effect of temperature

The removal efficiency of LEV was investigated as a function of temperature for 100 mL of 20 mg/L LEV solution at 278, 288, 298, and 308 K. At each temperature and pH 6.5, 0.5 g of PG-Fe⁰ composite was added to the corresponding LEV solution. The solutions were shaken for 3 h in a thermostatic shaker at 125 rpm. Samples from each solution were collected at various time intervals (5, 10, 15, 20, 30, 40, 80, and 320 min), then the collected samples were filtered through a 0.45 µm filter, and the absorbance of the aliquot from each sample was measured to determine the LEV concentrations.

3.5 Fenton catalyst

LEV removal was also investigated using a Fenton-like reaction, which involved exposing 0.5 g of PG-Fe⁰ composite (equivalent to 80 mg Fe⁰) to 90.0 mL of 10 mg/L LEV solution and 10 mL of freshly prepared H₂O₂. The solution was then kept in a thermostatic shaking water bath at 298 K for two hours before the absorbance of the separated LEV solution was measured.

3.6 Antimicrobial activity

The antimicrobial activity of PG-Fe⁰-LEV was investigated using the well-agar diffusion method. Gram-positive (E. faecalis, S. epidermidis, and S. aureus) and gram-negative (E. coli, K. pneumonia, and P. mirabilis) bacteria were used. The PG-Fe⁰-LEV composite was made by combining 0.5 g of PG-Fe⁰ composite with 100 mL of 25 mg/L LEV solution and shaking it continuously at 125 rpm. After 300 min of adsorption, the composite was collected and dried using an external magnetic field, and a UV–vis spectrophotometric analysis method was used to determine the concentration of adsorbed LEV. The optimal concentrations of PG-Fe⁰-LEV, PG-Fe⁰ composite, and LEV were calculated, and Milli-Q water was used as a negative control. The bacterial strains were evenly distributed across the Mueller-Hinton agar plate. After piercing a 6 mm hole with a sterile head, 50 μL of each antimicrobial agent was injected into the wells of agar, and for 24 h the plates were incubated at 310 K. Finally, the samples were duplicated, and the average diameter of the final inhibition zone was measured.

4.1 Characterization of PG-Fe⁰ composite

To characterize the surface morphology of synthesized materials, SEM, TEM, EDS, XRD, and FTIR techniques were used. TEM and SEM images for aggregated chains of Fe⁰ NPs with an average diameter of about 60 nm are shown in Fig. 2 (a, b). Aggregation of Fe⁰ NPs reduces the surface area and limits the dispersion of Fe⁰ NPs in the solution causing a drop in pollutant removal efficiency. SEM images of milled pencil graphite that appeared as bonded sheets and flakes are shown in Fig. 2 (c), and SEM images of the PG-Fe⁰ composite are shown in Fig. 2 (d). The image depicts Fe⁰ NPs adhering and dispersing on the PG surface. Furthermore, the Fe⁰ NPs adhere tightly to the PG surface, and the composite is strongly attracted to an external magnetic field. The aggregation of the Fe⁰ NPs was reduced as soon as the Fe⁰ NPs spread across the PG surface. The elemental analysis of PG, Fe⁰, and PG-Fe⁰ composite was evaluated using energy-dispersive X-ray spectroscopy (EDS). The EDS spectrum of the PG-Fe⁰ composite is depicted in (Fig.S2), and the analysis confirmed the presence of the main elements in the composite.

Close

Fig. 2

(a) and (b) are TEM and SEM images of Fe⁰ NPs respectively. (c) and (d) are SEM images of PG and PG-Fe⁰ composite, respectively.

Export to PPT

The XRD characterization technique was used to determine the crystalline morphology of the Fe⁰ and PG-Fe⁰ composite; the diffraction pattern of the Fe⁰, as shown in Fig. 3 (a), demonstrated the presence of crystalline zero-valent iron in the sample with a basic reflection peak at 2θ = 44.9° (110 reflectance), which illustrates the distribution of small size particles for a metallic iron crystal (Fe⁰), the reflectance peak of iron oxide was very weak, implying that Fe⁰ is the main component of iron nanoparticle materials. While Graphite showed characteristic peaks at 2θ = 26.7° and 2θ = 55°, as shown in Fig. 3 (b) (Sun et al., 2006). While Fig. 3 (c) depicts the XRD spectrum diffraction angles of the PG-Fe⁰ at (2θ) 26.7°, 44.9°, and 55°, respectively. These reflections from (0 0 2), (1 1 0), and (0 0 4), are assigned to the Fe⁰ and PG and demonstrate the presence of Fe⁰ on the PG surface (Yuan et al., 2008; Skrzypczynska et al., 2018; Xu et al., 2018). After LEV adsorption via Fe⁰ loaded on the PG surface, no change in the position of the distinctive peak for PG (0 0 2) was observed, implying that no phase transformation occurred. Whereas the distinctive peak of Fe⁰ (1 1 0) was shifted to the right, indicating LEV attachment to the surface of Fe⁰ NPs (Fig. 3 d) (Ben Arfi et al., 2017).

Close

Fig. 3

The XRD patterns of (a) Fe⁰ NPs, (b) PG, (c) PG-Fe⁰ composite, and (d) Fe⁰-PG-LEV.

Export to PPT

The coordination mode between ligand (LEV) and metal (Fe⁰) was identified using FTIR spectroscopy after LEV adsorption (Fig. 4). For this purpose, the sodium levofloxacin (Na_levo), PG-Fe⁰, and PG-Fe⁰-LEV FTIR spectra were obtained in the range of 4000–230 cm⁻¹. The probable modes are monodentate (Δν(COO^–)_complex ≫ Δν(COO^–)_Na), bidentate chelating (Δν(COO^–)_complex ≪ Δν(COO^–)_Na), or bidentate bridging (Δν(COO^–)_complex ≈ Δν(COO^–)_Na). The Na_levo spectrum in Fig. 4 (a) showed the absorption peak for the pyridone ν(C⚌O)_p at 1616 cm⁻¹ and the peak of the carboxylate stretch ν(C⚌O)_carb at 1721 cm⁻¹. In PG-Fe-LEV composite spectrum, the peak of the carboxylic group ν(C⚌O)_carb has disappeared upon binding and the two characteristic peaks asymmetric ν(COO^–, 1586 cm⁻¹)_asym and symmetric ν(COO^–,1392 cm⁻¹)_sym stretching vibrations was shifted to 1620 cm⁻¹ and 1395 cm⁻¹, respectively. The difference between the two peaks Δν(COO^–) = 225 cm⁻¹ which is greater than in Na_levo Δν(COO^–) = 194 cm⁻¹ indicating the monodentate coordination mode of the carboxylate group. Similarly, the carbonyl pyridone ν(C⚌O)_p is shifted from 1616 cm⁻¹ to 1580 cm⁻¹ upon bonding. The overall changes indicate that levofloxacin was bonded to the PG-Fe⁰ composite through the pyridone oxygen atom and one of the oxygen atoms from the carboxylate group as shown in Fig. 5 (Turel, 2002; Uivarosi, 2013; Mimouni et al., 2018). Generally, the ν(M−N) and ν(M−O) for metal complexes bands appear in the range of (230–700) cm⁻¹ (Alabdali and Ibrahim, 2014). The broad split peak between 3500 and 3100 cm⁻¹ is assigned to the O—H and N—H stretching of H₂O molecules and piperazinyl moiety, respectively (Qassim, 2015).

Close

Fig. 4

FTIR spectra of (a) NaLEV, (b) PG-Fe⁰ composite (c) PG-Fe⁰-LEV composite.

Export to PPT

Close

Fig. 5

Schematic represents the suggested adsorption mechanism of LEV onto the surface of the PG-Fe⁰ composite (M), pH = 6.5.

Export to PPT

4.2 The effect of pH and adsorbent dose on LEV removal

As shown in Fig. 6 (a), different composite doses of PG-Fe⁰ ranging from 0.1 to 1.5 g were investigated for LEV removal efficiency. When an adsorbent dose is increased, the removal efficiency increases from 30 % to 78 % due to the availability of the free locations and the unsaturation adsorption sites (i.e., more surface area). The adsorption efficiency also decreased from 78 % to 50 % at 1.5 g, which could be explained by the ability of particles to aggregate by increasing their amount, resulting in a decrease of free adsorbent sites. However, an increase in performance occurs only after a certain amount of adsorbent dose is added, after that, performance becomes constant.

Close

Fig. 6

Effect of (a) amount of adsorbent, (b) pH, and (c) initial adsorbate concentration on removal efficiency at 298 K.

Export to PPT

The chemical structure of LEV is pH-dependent, and the pH value is an important parameter that influences LEV adsorption using the PG-Fe⁰ composite. Fig. 6 (b) depicts the effect of pH on the adsorption of LEV in an aqueous solution with a composite adsorbent. As shown, the removal of LEV increases as the pH rises from 3 to 5, and LEV molecules appear in the cationic form LEV⁺. The cationic form of LEV⁺ gradually decreases as the pH rises, and LEV molecules are converted to the Zwitterionic form. The maximum adsorption of LEV onto PG-Fe⁰ composite occurs when the chemical structure is neutral/zwitterion (pH 5–7). As the pH of the LEV solution increased from 7 to alkaline, a gradual decrease in LEV removal was observed, indicating that LEV molecules have the anionic form (LEV^-). The aspect of maximum exclusion of LEV molecules at pH > 7 or at pH > 7 can be attributed to ion exchange. Moreover, as the pH increases or decreases, the removal efficiency decreases due to electrostatic repulsion between similar charges. Furthermore, as shown in Fig. 6 (c), the removal efficiency of LEV by composite in different initial concentrations is more efficient at pH 6.5 than at pH 8. At both pH values, the maximum removal efficiency was at 10 mg/L of LEV and decreased as the initial concentration increased; further kinetic and thermodynamic analysis will be performed in the following sections. On the other hand, the performance of the milled PG is insufficient to remove LEV from aqueous solutions for all concentrations at both pH values.

4.3 LEV removal kinetic

Kinetic experiments were carried out at two different pH values (pH = 6 and pH = 8) to investigate the adsorption rate as well as the kinetic mechanism of the adsorption process. The amount of LEV adsorbed onto the PG-Fe⁰ composite surface was calculated using the mass balance equation Eq.(2)(Thilagan et al., 2013).

Where Q is the adsorbent's concentration of LEV removal (mg/g). C₀ represents the initial concentration of LEV (mg/L). C_e indicates the concentration of LEV in solution at a given time. V indicates the volume of the solution (L), and m indicates the mass of the adsorbent (mg). Lagergren pseudo-first-order and pseudo-second-order models were used to test the experimental data of LEV adsorption by PG-Fe⁰ composite (Vijayakumar et al., 2012; Ho, 2014).

The linear Lagergren pseudo-first-order model is represented as Eq.(3) (Simonin, 2016; Fu et al., 2021):

Where $q_e$ and $q_t$ are the amounts of LEV that are adsorbed by the adsorbent in equilibrium and at any time (mg/g), respectively. $k_1$ is the first-order rate constant (min⁻¹) and $t$ is the contact time (min). The relationship was obtained by plotting $\ln \left(q_e-q_t\right)$ versus contact time. The values ​​of $k_1$ and $q_e$ were obtained from the slope and the axis intercept, respectively.

Another kinetic model, the linearized pseudo-second-order based on the Ho relation was investigated to fit the experimental data Eq.(4) (Ho, 2014; Benjelloun et al., 2021).

Where Q denotes the amount of LEV adsorbed on the adsorbent (mg/g). $Q_e$ is the amount of LEV adsorbed on the adsorbent at equilibrium (mg/g). $k_2$ is the rate constant of pseudo-second-order adsorption g/(mg·min), and the contact time is denoted by $t$ (min). Fig. 7 depicts the linear fits obtained using the pseudo-second-order kinetic model.

Close

Fig. 7

Pseudo-second-order linear fits for the removal of LEV by PG-Fe⁰ composite at (a) pH = 6.5 and (b) pH = 8, and 298 K.

Export to PPT

The Kinetic parameters extracted from Fig. 7 are listed in Table 1. The parameters show that the correlation coefficient of data (R² values) fits better at both pH levels for the pseudo-second-order model than for the pseudo-first-order model. Furthermore, at all concentrations, the amount of LEV removed at equilibrium is greater at pH 6.5 than at pH 8, and the rate of adsorption (k) decreased as LEV concentration increased for both pH values.

4.4 Adsorption isotherm

The sorption of LEV on the surface of PG-Fe⁰ composite was quantified using Langmuir and Freundlich isotherm models. Langmuir describes the adsorption of a monolayer surface on a solid with specific sites that are strongly identical and is based on the kinetic principle (Hall et al., 1966). To fit the experimental data, the linear form of Langmuir adsorption isotherm was used as shown in Eq.(5).

$C_e$ denotes the LEV solution's equilibrium concentration (mg/L). The maximum amount of LEV (mg) adsorbed per gram of adsorbents for complete monolayer coverage is defined as $Q_m$ (mg/g). The Langmuir constant is denoted by the symbol $K_L$ (L/mg).

The slope and intercept for the linear plot of $\frac{C_e}{Q_e}$ versus $C_e$ , as shown in Fig. 8 (a), were used to calculate the adsorption parameters $Q_m$ and $K_L$ , respectively. The Freundlich isotherm is another commonly used adsorption model that implies multilayer adsorption and can be used for heterogeneous adsorbent surfaces with different sites of variable adsorption energy (Saadi et al., 2015). The linearized form of the Freundlich isotherm model is represented as:

Close

Fig. 8

Linear plots of (a) Langmuir isotherm model and (b) Freundlich isotherm model for LEV adsorption via PG- Fe⁰ composite.

Export to PPT

The Freundlich constant ( $k_f$ ) reflects the adsorption affinity and n is the Freundlich constant related to adsorption linearity. Both constants can be calculated from the slope and intercept of the linear plot of $\mathit{ln}{Q}_e$ versus $\mathit{ln}{C}_e$ , as shown in Fig. 8 (b). The parameters extracted from the Langmuir and Freundlich models at both pH values are shown in Table 2.

The R² values indicated that the experimental data at pH 6.5 was well fitted by both the Langmuir and Freundlich isotherm models. This data is consistent with the adapted LEV charge structure (zwitterionic), which allows for the formation of monolayer or multilayer adsorption between opposite ionic charges. At pH 8, the Langmuir isotherm model described the data better than the Freundlich isotherm, indicating monolayer adsorption between anionic LEV and composite. Furthermore, these results suggest a relation between the Freundlich model isotherm and the zwitterionic form of LEV which allows for multilayer adsorption.

The Q_m values confirmed these findings; the calculated maximum capacity of a composite at pH 6.5 is approximately 66.3 mg/g, while it is 11.4 mg/g at pH 8. These values are low when compared to the surface area of nano zero-valent iron adsorbents (131.3 m²/g) (Al-Rimawi et al., 2022), but this is to be expected with bulky LEV molecules adsorbed in an aqueous solution versus physical adsorption of N₂ gas molecules on a solid surface. The LEV uptake capacity using PG-Fe⁰ was compared to the capacities of other adsorbents reported in the literature (Table S1). The presented capacity demonstrated a moderate removal efficiency for LEV as toxic and biological active water pollutants; however, in terms of recovery, cost, and eco-friendly, the PG-Fe⁰ composite still outperforms other adsorbents.

In the Freundlich model, the linearity of the adsorption or the degree of curvature for the isotherms over the concentration range is described by 1/n, which typically ranges from 1 to 0, and the value of 1/n less than 1 indicated a favorable adsorption process (Saadi et al., 2015).

4.5 Adsorption thermodynamic

The activation energy for LEV adsorption on the surface of PG-F⁰ composite was calculated using the Arrhenius equation based on the kinetic data obtained at different temperatures, Eq.(7) (Logan, 1982).

Where $k_2$ is the pseudo-second-order rate constant g/(mg·min), A is the pre-exponential factor, E_a is the activation energy (kJ/mol), R is the gas constant 8.314 J/(K·mol), and T is the absolute kelvin temperature (K). The slope for the linear plot of $\mathit{ln}{k}_2$ versus 1/T was used to calculate the value of $E_a$ (Fig.S3). Other thermodynamic parameters were estimated including the heat of adsorption (standard enthalpy change, Δ $H^{\mathrm{o}}$ ), standard entropy change ( $\mathrm{\Delta }{S}^{\mathrm{o}}$ ), and standard change in Gibbs free energy ( $\mathrm{\Delta }{G}^{\mathrm{o}}$ ). To evaluate the presented parameters, the thermodynamic Eqs. (8–9) were used.

The $\mathrm{\Delta }{H}^{\mathrm{o}}$ and $\mathrm{\Delta }{S}^{\mathrm{o}}$ values were determined using van't Hoff equation (Lima et al., 2020):

Where the K_eq is the equilibrium constant expressed as $\left(\frac{Q_e}{C_e}\right)$ , and the value of $\mathrm{\Delta }{G}^{\mathrm{o}}$ was obtained from the following equations:

The values of $\mathrm{\Delta }{H}^{\mathrm{o}}$ and $\mathrm{\Delta }{S}^{\mathrm{o}}$ were determined using the slope and y-axis intercept obtained from the linear plot of $\mathit{ln}{K}_{\mathit{eq}}$ versus 1/T as shown in (Fig.S4), and the thermodynamic parameters were determined and listed in Table 3.

The nature of sorption (chemisorption or physisorption) could be determined by the value of $E_a$ or from the value of $\mathrm{\Delta }{H}^{\mathrm{o}}$ . The magnitude of activation energy indicates whether the adsorption mechanism is physisorption (5 to 40 kJ/mol) or chemisorption (40 to 800 kJ/mol) (Abramian and El-Rassy, 2009). However, the negative activation energy makes a difficult decisive conclusion about the nature of adsorption (Atkins et al., 2014). Depending on the value of $\mathrm{\Delta }{H}^{\mathrm{o}}$ , the calculated value (44.4 kJ/mol) is in the range of chemisorption (>40 kJ/mol) which was also confirmed by the FTIR analysis. Furthermore, the positive value of ΔH^o implies that the adsorption process is endothermic.

Previous research has led to the conclusion that a negative value for ΔG^o indicates that the adsorption process is spontaneous. In this study, the negative value of ΔG^o indicates that LEV is naturally adsorbed by the adsorbent at 288 K, and the value of ΔG^o decreases as the temperature increases, indicating that LEV adsorption is more favorable at high temperatures.

4.6 Removal of LEV by a Fenton-like process.

One of the most effective methods for degrading organic contaminants is the Fenton catalyst. The Fenton reagent can be generated by a cyclic process in an aqueous solution from the interaction of Fe²⁺ and H₂O₂. When ferrous ions are replaced by Fe⁰ or iron oxides, a Fenton-like reaction occurs. The radicals are produced by reactions that occur during a Fenton-like cycle in which Fe⁰ is used as a source of Fe²⁺ ions, and the radical can chemically oxidize and mineralize the organic pollutants (Neyens and Baeyens, 2003):

• (i) Fe⁰ + 2H₂O₂ → Fe²⁺ + 2•OH + 2OH^–

• (ii) Fe²⁺ + 2H₂O₂ → Fe⁰ + 2•OOH + 2H⁺.

Fig. 9 depicts the results of related experiments, the results show that using PG-Fe⁰ composite to eliminate the LEV has nearly the same removal efficiency as a Fenton-like reaction of (Fe⁰ + H₂O₂) and is better than using bare Fe⁰. On the other hand, Using PG-Fe⁰ in presence of H₂O₂ reduced the removal efficiency to 18 % under the same conditions. This behavior could be attributed to the LEV degradation cycle being hindered by the trapping of the radical produced during the Fenton-like process, or the adsorption of LEV molecules on the surface of Fe⁰ prevents the formation of radical, additionally, the Fenton-like process is very sensitive to pH value (Zhang, 2003; Poursaberi et al., 2012; Nazari et al., 2013). However, when the removal efficiency of PG-Fe⁰ is compared to that of the Fenton-like reaction, the proceeding method using PG-Fe⁰ as an adsorbent significantly outperforms the Fenton-like performance because the Fenton-like process produces a lot of more toxic degraded substances that harm the environment than the parent compound.

Close

Fig. 9

Removal % of LEV using Fe⁰ and PG-Fe⁰ as adsorbents and as a Fenton-like catalyst.

Export to PPT

4.7 Antibacterial activity

The future of PG-Fe⁰ composite can be used to reduce environmental pollution and control the release of antibiotics in drug delivery systems. Therefore, there is a great opportunity for the recovery, reusability, and recycling of a large number of antibiotics using the synthesized PG-Fe⁰ composite.

The reusability of the adsorbed LEV and the antibacterial activity of PG-Fe⁰-LEV composite against various bacterial strains was investigated. The lack of growth of bacteria or microorganisms is observed in the area around the hole known as the inhibition zone, and the effectiveness of the LEV antibiotic increases at a specific concentration on various types of bacterial strains. Fig. 10 depicts the antibacterial activity of PG-Fe⁰ composite, LEV, PG-Fe⁰-LEV, and water (negative control) in terms of zone inhibition as measured by the agar well diffusion method at pH 6.5. The antibacterial activity was investigated by comparing the effectiveness of pure LEV and PG-Fe⁰-LEV for inhibiting the effect of the bacterial strain. The calculated suppression zone diameter is shown in Table 4, in all tested samples, no antibacterial activity was detected for the PG-Fe⁰ composite. The antibacterial activity of PG-Fe⁰-LEV on the bacterial strains was irregular and did not show a significant effect on the Enterococcus faecalis strain. In addition, the results show that the LEV activity after adsorption on the surface of the PG-Fe⁰ composite varies from that of the pure LEV drug. Despite the PG-Fe⁰-LEV composite having a lower effect than pure LEV, the new combination between the active LEV molecules and the PG-Fe⁰ composite does not destroy LEV's bacterial inhibition mechanism as an antibiotic. Furthermore, the biological activity test shows the possible influence of LEV on bacterial life in the aquatic environment.

Close

Fig. 10

Photographs of Petri plates used in the agar well diffusion method; (A) negative control (water), (B) PG-Fe⁰, (C) PG-Fe⁰-LEV, and (D) LEV, at pH = 6.5.

Export to PPT