Materials

Raw kaolin clay was obtained from Hadiya, Ethiopia. Iron (III) chloride hexahydrate (FeCl₃·6H₂O), iron (II) sulfate heptahydrate (FeSO₄·7H₂O), concentrated sulfuric acid (H₂SO₄, 18.4 mol/L), hydrochloric acid (HCl, 12 mol/L), ammonia (NH₃, 13.4 mol/L), and hydrogen peroxide (H₂O₂, 9.79 mol/L) were purchased from Merck, India, and used as received. Double distilled water was used throughout the whole experimental process.

Acid activation of kaolin

Kaolin was activated by H₂SO₄ based on a protocol reported in previous work.^34,35 Particularly, the raw kaolin sample was purified by washing with distilled water 3 times followed by drying in an oven at 105°C for 3 hours. The dried sample was crushed to obtain a fine powder that was able to pass through a 63-µm mesh size sieve. The sieved sample was further washed with distilled water and the dispersed clay separated by centrifugation. The washing process was continued 4 times to obtain pure kaolin. The wetted samples were dried in an oven at 105°C for 3 hours. Then, the powdered kaolin (6 g) was mixed with 180 mL of sulfuric acid (0.5 mol/L) and refluxed for 12 hours at 70°C for activation. The slurry was washed with distilled water 3 times till the supernatant became neutral with no trace amount of sulfate ion present. The slurry was separated and dried in an oven at 105°C for 3 hours. The final powder product (<63 µm size) is designated as activated kaolin (AK).

Synthesis of magnetite nanoparticles

The facile coprecipitation technique was used to synthesize Fe₃O₄ NPs. In particular, aqueous solutions of ferric and ferrous salts were mixed together in a molar ratio of 2:1, respectively. The mixture was heated with continuous stirring under reflux setup, and then NH₃ was added drop by drop to adjust the pH of the solution to 10. The solution was heated at 70°C for 45 minutes. Finally, the product was washed 3 times with distilled water to remove excess ammonium and dried under vacuum.

Synthesis of AK-Fe₃O₄ composite

The magnetically separable AK-Fe₃O₄ composite was synthesized by in situ coprecipitation of Fe₃O₄ NPs onto AK. The solution of FeCl₃·6H₂O and FeSO₄·7H₂O in a 2:1 molar ratio, respectively, was added drop by drop to 100 mL of AK solution. The mass ratio of iron salts to AK was fixed at 2:1, 4:1, and 6:1, respectively. After 30 minutes of continuous stirring, the ammonia solution was added to adjust the pH to 10. Then, the solution was refluxed for 8 hours at 70°C. The dark brown precipitate was washed 3 times with distilled water and dried under vacuum.

Characterization techniques

The functional groups present on the synthesized materials were investigated by FTIR spectroscopy. The FTIR was recorded using the Perkin Elmer FT-IR (spectrum RX-I) spectrometer in the range 400 to 4000 cm⁻¹ by mixing the samples with spectroscopy-grade KBr in 1:2 mass ratio. X-ray diffraction patterns of samples were recorded using SHIMADZU XRD, excellent in the science of XRD (model number XRD-7000; Shimadzu, Kyoto, Japan), which was operated at 40 kV and 45 mA, with a Cu-Kα (λ = 1.54056 Å) radiation. Scanning electron microscopy studies of the morphology of AK-Fe₃O₄ were analyzed by microstructure analysis with a Nova NanoSEM analyzer operated at an accelerating voltage of 10 keV. The samples were degassed at 150°C under vacuum. The magnetic property AK-Fe₃O₄ was investigated using a VSM (Lakeshore 665; Lake Shore Cryotronics, Inc., Westerville, OH, USA) at room temperature. The pH of point of zero charge (pH_PZC) of AK-Fe₃O₄ was determined to know the surface charge of AK-magnetite nanoparticles. For this purpose, 50 mL of sodium chloride (NaCl) solution was placed in a 100-mL Erlenmeyer flask. The pH was then adjusted to successive initial values between 3 and 9 using either sodium hydroxide or hydrogen chloride (0.1 mol/L), and 0.15 g of adsorbent (AK-Fe₃O₄) was added to the solution. After a contact time of 24 hours, the final pH was measured and plotted against the final pH. The pH at which the curve crosses the line pH (final-initial) is taken as pH_PZC.

RSM-BBD optimization of the Fenton-like degradation variables

The Fenton-like degradation of 4-NP was carried out in a batch reactor at varying concentrations of 4-NP. Specifically, 50 mg of the Fenton catalyst was added to 50 mL of 4-NP solution (10-30 mg/L) and stirred continuously. After adding the oxidation catalyst, 0.5 mL of H₂O₂ (9.79 mol/L) was added to the solution, which is considered as the initial (time t = 0) for the reaction. The pH of the solution was adjusted using NaOH (0.1 mol/L) and HCl (0.1 mol/L). The degradation (%) of 4-NP was analyzed by measuring the absorbance at λ_max = 400 nm using an ultraviolet-visible spectroscopy. The RSM-BBD optimization of 4-NP degradation mainly focused on 3 independent variables: pH, contact time (minutes), and concentration of 4-NP (mg/L). These 3 variables are selected because catalysis processes are highly dependent on them. The other remaining variables including 0.5 mL of 9.79 mol/L H₂O₂ (97.9 mmol/L), 50 mg of catalyst dose in 50 mL solution (1 mg/mL), and room temperature (298 K) were fixed throughout the whole reaction process. Design Expert 11 software program was used to compute the statistical analysis. The experimental levels of factors used for RSM-BBD are given in Table 1.

Table 1.

The experimental levels of factors used for RSM-BBD.

Characterizations

FITR analysis

Figure 1 shows the FTIR spectra of AK, Fe₃O₄ NPs, and AK-Fe₃O₄ composite. Activated kaolin (Figure 1A) shows 2 clearly identifiable bands at 3690 and 3620 cm⁻¹. The peak at 3690 cm⁻¹ is due to stretching vibrations of surface hydroxyl groups, whereas the other peak at 3620 cm⁻¹ is ascribed to the vibrations of inner hydroxyl groups.³⁶ The absorption bands observed at 3420 cm⁻¹ could be assigned to the OH vibrational mode of water. The band associated with Si-O deformation is detailed to be 1040 cm^-1.³⁷ The band at 796 cm⁻¹ is assigned to the Si-O-Si symmetric stretch.³⁸ Besides, a strong peak at 544 cm^-1 indicates the presence of Si-O-Al stretching vibrations.³⁹ Figure 1B shows the FTIR spectrum obtained from pure magnetite nanoparticles. The 2 intense strong peaks observed at 557 and 637 cm⁻¹ are attributed to the symmetric stretching vibration mode associated with Fe-O bond in the crystalline lattice of Fe₃O₄.⁴⁰ The FTIR spectrum of AK-Fe₃O₄ composite (Figure 1C) shows the peak due to both AK and Fe₃O₄ NPs, which implies the formation of the composite. But the intensity of inner and outer hydroxyl (OH) of kaolin becomes weak in the composite, indicating interruption of the kaolin structure.⁴¹ This weak intensity is due to the exchange of Al³⁺ of kaolin structure with Fe³⁺ from Fe₃O₄.⁴¹

Figure 1.

Fourier transform infrared spectra of (A) activated kaolin (AK), (B) Fe₃O₄, and (C) AK-Fe₃O₄ composites.

XRD analysis

The powder XRD analysis identified the unknown clay to be kaolin, and crystallite size and lattice cell parameters of Fe₃O₄ NPs, AK, and AK-Fe₃O₄ composites. The raw clay obtained from Hadiya, Ethiopia, was found to be kaolin. As shown in Figure 2A, raw kaolin is exactly matched with kaolinite Al₂H₄O₉Si₂ (COD: 00-900-9230).⁴² The raw kaolin after chemical activation did not show a significant change in the XRD pattern compared with the raw kaolin ( Supplemental Figure 1). Figure 2B shows the formation of spinel-type Fe₃O₄ NPs with 14 nm average crystallite size (D) calculated using Scherrer formula (equation (1)).⁴³ The characteristic plans are observed at (220), (311), (400), (422), (511), (440), and (553), which are in agreement with the inverse cubic spinel structure of Fe₃O₄ (JCPDS card no. 85-1436, space group Fd3m). Figure 2C presents the XRD patterns of the AK-Fe₃O₄ composite. The characteristic peaks due to Fe₃O₄ and AK were observed concurrently in the AK-Fe₃O₄ composite. The analysis illustrates that inverse spinel Fe₃O₄ NPs were successfully precipitated with activated clay to form the AK-Fe₃O₄ composite.⁴⁴ Besides, the average crystallite size of Fe₃O₄ NPs was found to be 11 nm which is less than naked Fe₃O₄ NPs. Moreover, the intercalation of Fe₃O₄ into AK layer is plausibly investigated by calculating the interplanar distance (d) before and after Fe₃O₄ NP treatment. As shown in  Supplemental Figure 2, 2θ and d values at (100) of AK are different before and after Fe₃O₄ NP treatment. The shift of diffraction angle from 2θ = 12.3776° to 2θ = 12.0862° and expanding of the d value from 7.14531 to 7.31692 Å are attributed to the intercalation of Fe₃O₄ to the AK layer:⁴⁵

where λ is the wavelength of the x-ray source (λ = 0.15418 nm), β is the peak width of the diffraction peak profile at half maximum height resulting from small crystallite size in radians, and θ is the Bragg angle.

Figure 2.

The powder x-ray diffraction patterns of (A) activated kaolin (AK), (B) Fe₃O₄, and (C) AK-Fe₃O₄ composite.

SEM, VSM, and pH_PZC analyses

The rough texture of the AK-Fe₃O₄ composite was investigated by the SEM image (Figure 3A). AK-Fe₃O₄ due to the formation of an infinite number of pores on the surface increases the surface area for adsorption of 4-NP. The magnetization value of naked Fe₃O₄ and AK-Fe₃O₄ was recorded at 300 K using VSM to investigate their magnetic nature. As shown in Figure 3B, both naked Fe₃O₄ and AK-Fe₃O₄ composites show superparamagnetic nature with zero remanence and coercivity. The saturation magnetization (MS) was found to be 57 and 25 emu/g for the naked Fe₃O₄ NPs and AK-Fe₃O₄ composite, respectively. In this sense, in the AK-Fe₃O₄ composite, the non-magnetic kaolin is responsible for weakening the magnetization value. Besides, those magnetization values are significantly lower than bulk Fe₃O₄ (MS ~82 emu/g).⁴⁶ The plot of ΔpH versus initial pH of the solution of AK-Fe₃O₄ is presented in  Supplemental Figure 3. It is revealed that the AK-Fe₃O₄ composite shows a positive charge below pH_PZC = 6 and shows a negative charge as the pH increases.

Figure 3.

(A) Representative scanning electron microscopy image and (B) vibrating sample magnetometer curve of AK-Fe₃O₄ composite. AK indicates activated kaolin; NPs, nanoparticles.

Optimization of 4-NP degradation using RSM-BBD

During catalysis reaction, optimizations of the process variables using statistical approaches are crucial for the economic utilization of time and resources. Moreover, the statistical analysis gives a clear understanding of the reaction process and minimizes the experimental errors. Herein, the RSM-BBD approach was applied to optimize the Fenton-like degradation of 4-NP using AK-Fe₃O₄ composite. Box-Behnken design is a second-order design (equation (2)) based on 3-level factorial designs.⁴⁷ The complete BBD matrix for the 3 independent variables (contact time (A), pH (B), and 4-NP concentration (C)) and the response of both coded and real values are presented in Table 2:

where Y is the predicted response (degradation, %) of 4-NP; b is a constant; and b_i, b_ii, and b_ij refer to the linear, quadratic, and interaction coefficients, respectively. X_i and X_j are lists of the coded values of factors.

Table 2.

The RSM-BBD design matrix of the 3 variables with coded and real values, and response with actual and predicted values.

The final equation (equation (3)) generated from 4-NP degradation using H₂O₂-AK-Fe₃O₄ based on BBD was found to be well fitted to the quadratic model:

Equation (3) shows how the 3 factors in a quadratic or interactive model affect the 4-NP degradation using the AK-Fe₃O₄ composite. As shown in equation (3), those with negative coefficients affect 4-NP degradation inversely, whereas those with positive coefficients enhance the degradation of 4-NP. Moreover, the adequacy of the quadratic model to sufficiently explain the 4-NP degradation process is validated via statistical parameters such as R² and analysis of variance (ANOVA). The detailed ANOVA results of 4-NP degradation using AK-Fe₃O₄ are presented in Table 3. As shown in Table 3, R² (0.9964) is much closer to unity, which implies that the data are well fitted to the quadratic model. In particular, 99.64% of the Fenton-like degradation of 4-NP can be described by the proposed quadratic model (equation (3)). Besides, Figure 4 shows that the predicted R² (0.9823) has a reasonable agreement with adjusted R² (0.9917), that is, the difference is less than 0.2. It is affirmed that the model can sufficiently explain the relationship between factors and 4-NP degradation (%).

Table 3.

The analysis of variance of 4-nitrophenol degradation based on quadratic model.

Figure 4.

The comparison of actual (experimental) and model prediction values of 4-nitrophenol degradation (%).

The ANOVA (Table 3) of the model F value is 214.69, which implies that the quadratic model is significant and there is only a 0.01% chance that an F value this large could occur due to noise. The P value (<.0001) of the model also shows the significance of the model. Moreover, the significance of the model parameters is also validated by P values. The P value less than .05 (5%) indicates the model terms are significant. Herein, all the linear (A, B, and C), quadratic (A², B², and C²), and interaction (AC) terms are significant, whereas model terms greater than 0.1 are not significant; therefore, the AB and BC interaction model terms are not insignificant.

The lack of fit (F = 0.40) from ANOVA (Table 3) implies that the lack of fit is not significant compared with the pure error. There is a 76.17% chance that a lack of fit F value this large could occur due to noise. This implies that the Fenton-like degradation of 4-NP is well explained by our quadratic model with a 95% confidence level.

The 2-dimensional (2D) and 3-dimensional (3D) surface plots of the response as a function of process variables are useful for a clear understanding of the reaction process. The 2D-3D plots (Figure 5) show the 4-NP degradation (%) with varying contact time, pH, and 4-NP concentrations. The degradation (%) of 4-NP differs with varying interaction variables and is found to be maximum with AB interaction terms than with AC and BC at a fixed 1 mg/mL concentration of AK-Fe₃O₄. As shown in Figure 5A, the AB interaction optimized at a fixed 4-NP concentration (20 mg/L) is the most important interaction parameter in predicting the maximum degradation of 4-NP. Hence, the optimum 4-NP degradation in 50 mL solution was found to be 96.01% ± 1.89% using 1 mg/mL of AK-Fe₃O₄, 20 mg/L of 4-NP, 97.9 mmol/L of H₂O₂, and pH of 3 at the end of 75 minutes of reaction time. Moreover, the RSM-BBD optimization with other scenarios gives maximum degradation of 4-NP, and based on considering the available resource it is possible to adapt different combinations of contact time, pH, and 4-NP concentration.

Figure 5.

The 2-dimensional and 3-dimensional surface plots of the effect of (A) contact time and pH, (B) initial 4-NP concentration and contact time, and (C) 4-NP initial concentration and pH of solution on the degradation (%) efficiency of 4-NP by AK-Fe3O4 composite. AK indicates activated kaolin; 4-NP, 4-nitrophenol.

In general, the numerical optimization of the desired goal is expected to maximize, minimize, target, be within the range, and be none of the responses and adjust to a precise value for the factors.⁴⁸ Herein, the numerical optimization found a point that maximizes the desirability function in 4-NP degradation. For example, the criteria set for the 3 factors in the range including contact time (10-90 minutes), pH of the solution (3-8), and initial 4-NP concentration (10-30 mg/L) and 96.08% maximum degradation of 4-NP were found to be the maximum desirability. Figure 6 shows ramp desirability that was generated from 100 optimum points via numerical optimization.

Figure 6.

The desirability ramp for optimization of 3 factors (contact time, pH solution, and 4-NP concentration) for the maximum degradation (%) of 4-NP using AK-Fe₃O₄ composite. AK indicates activated kaolin; 4-NP, 4-nitrophenol.

Mechanism and kinetics of 4-NP degradation

 Supplemental Figure 4(a) shows the removal of 4-NP with and without H₂O₂. AK, Fe₃O₄, AK-Fe₃O₄, and H₂O₂ show very less removal of 4-NP, whereas AK-Fe₃O₄ composite in combination with H₂O₂ shows a significant degradation of 4-NP. This is attributed to the removal of 4-NP, which was predominantly through Fenton-like degradation of 4-NP. Moreover,  Supplemental Figure 4(b) shows the comparative degradation efficiency of AK/H₂O₂, Fe₃O₄/H₂O₂, and AK-Fe₃O₄/H₂O₂. Fe₃O₄/H₂O₂ and AK-Fe₃O₄/H₂O₂ due to the presence of Fe₃O₄, which activates H₂O₂, show greater degradation efficiency than AK/H₂O₂.²⁰ For Fenton-like reaction on the iron-based heterogeneous catalyst, the activation of H₂O₂ using Fe²⁺ and Fe³⁺ proceeds via adsorption of H₂O₂ on the catalyst surface, which in turn depends on the surface area of the catalyst.⁴⁹ Fe₃O₄ nanoparticles show a strong affinity toward H₂O₂, which facilitates the formation of active oxidizing agents (HO^• and $\mathrm{H}{\mathrm{O}}_2^{•}$ ).^25,49 The plausible mechanism for the degradation of 4-NP using the AK-Fe₃O₄/H₂O₂ system can be suggested as shown in Figure 7. Hence, H₂O₂ is initially adsorbed on the AK-Fe₃O₄ composite where it is activated by Fe₃O₄ for the formation of HO^• and HOO^•. HO^• is a powerful oxidant that can mineralize 4-NP into H₂O and CO₂.

Figure 7.

The plausible Fenton-like degradation of 4-NP using AK-Fe₃O₄ composite as a catalyst for the activation of H₂O₂. AK indicates activated kaolin; 4-NP, 4-nitrophenol.

The kinetics of the Fenton-like degradation of the 4-NP can be investigated using the pseudo-first-order equation (equation (4)):²⁹

where C_o and C_t are the initial and final concentrations at time “t” of 4-NP. Therefore, a plot of ln(C_t/C_o) with respect to reaction time gives a straight line whose slope is k ( Supplemental Figure 4(c)). The kinetic analysis was performed by varying the initial concentrations of 4-NP while maintaining others constant. As shown in  Supplemental Figure 4(d), the rate constant (k) decreases with an increase in 4-NP concentrations. This implies that 4-NP degradation is relatively fast at low concentration. Moreover,  Supplemental Table 1 shows the comparison of AK-Fe₃O₄ with other catalysts used for 4-NP Fenton-like degradations. AK-Fe₃O₄ shows faster 4-NP degradation efficiency than others that are working at room temperature (298 K).

Reusability and stability of AK-Fe₃O₄

Reusability and stability are very important criteria for evaluating the practical application of catalysts. Successive cycles of the Fenton degradation were carried out to evaluate the stability and reusability of our catalyst under the same reaction conditions. The optimum condition of degradation parameters in 50 mL solution such as 1 mg/mL of AK-Fe₃O₄, 20 mg/L of 4-NP, 97.9 mmol/L of H₂O₂, and pH of 3 at the end of 75 minutes of reaction time was used to investigate the stability and reusability of the catalyst. After each successive cycle, the AK-Fe₃O₄ was collected by centrifugation, washed at least 3 times, and dried in the oven at 100°C for 12 hours.⁵⁰ As shown in Figure 8, the catalyst could be successfully reused for 5 successive degradation cycles with more than 90% degradation efficiency, indicating that the AK-Fe₃O₄ Fenton-like catalyst is stable. Moreover, the stability of the AK-Fe₃O₄ before and after 5 times use was examined by XRD. As can be seen from Figure 9, the XRD patterns of AK-Fe₃O₄ do not change before and after degradation. This implies the stability of the catalyst crystal structure after use.⁵¹ But the peak intensity decreased after use, which is possibly due to catalyst loss during recovery.

Figure 8.

The reusability of AK-Fe₃O₄/H₂O₂ for the degradation (%) of 4-NP in 50 mL solution (1 mg/mL of AK-Fe₃O₄, 20 mg/L of 4-NP, 97.9 mmol/L of H₂O₂, pH of 3; contact time, 75 minutes). Error bar represents 5% error. AK indicates activated kaolin; 4-NP, 4-nitrophenol.

Figure 9.

The x-ray diffraction patterns of AK-Fe₃O₄ (A) before and (B) after Fenton-like degradation of 4-NP. AK indicates activated kaolin; 4-NP, 4-nitrophenol.