Process design and optimization of EDTA-biojarosite – a treatment approach in the Box–Behnken framework

The use of pesticides has increased dramatically over the past few decades due to their efficacy in pest control and insect management (Kawahigashi 2009). Pentachlorophenol (PCP), a member of the pentachlorobenzene family of organochlorine pesticides, is composed of an aromatic ring with five chloro groups in place of phenol. As a pesticide and wood preservative, PCP has been widely used in agriculture for many years. However, when PCP leaches out of wood or is carried away by agricultural runoff, it interacts with organic matter in the soil, leading to a decline in fertility (Martınez et al. 2000). Depending on the pH-value and presence of solvents, PCP can cause surface and groundwater contamination (Tanjore & Viraraghavan 2016). The inability to trace the non-point sources of pesticide pollution has become a significant environmental concern (Nitschke & Schossler 1998). Due to its toxicity, PCP has been included in the Environmental Protection Agency's list of pollutants (Rodrigues & Kyuma 2013).

Various treatment methods, such as activated carbon adsorption, biological degradation, and advanced oxidation processes, have been extensively explored over the years for the degradation of PCP. Among these techniques, Fenton oxidation has emerged as the most efficient method (Liou et al. 2004; Mecozzi et al. 2006; Zimbron & Reardon 2009; Rodrigues & Kyuma 2013).

Microporous and mesoporous substances are known for their versatile adsorption. Microporous materials such as zeolites and metal-organic frameworks have pores with sizes on the angstrom order (less than 2 nm). These small pores make them highly selective for the adsorption of molecules of specific size and shape while mesoporous materials with pore sizes between 2 and 50 nm provide a large internal surface area. This large surface area is beneficial for adsorption as it gives the molecules more opportunity to interact with the material (Atiyah et al. 2022b, 2022a; Khadim et al. 2022).

Fenton oxidation involves the generation of highly reactive hydroxyl radicals, which attack the aliphatic and aromatic components of organic compounds, leading to their decomposition. Numerous researchers have investigated the degradation of PCP using Fenton oxidation (Brown et al. 1986; Weavers & Hoffmann 2000; Wolski et al. 2006).

In Fenton oxidation, iron, a divalent cation, acts as a catalyst to facilitate the reaction. However, due to cost considerations, alternative catalysts have replaced commercial iron in the treatment process (Bhaskar et al. 2019, 2021, 2022b; Bhaskar et al. 2020; Rashmishree et al. 2022). Mecozzi et al. (2006) explored the use of PCP EAF dust as a catalyst for Fenton degradation (Mecozzi et al. 2006). In the Fenton oxidation process, Karale et al. (2013) introduced an alternative catalyst by substituting chemically leached laterite iron (Karale et al. 2013). Similarly, Bhaskar et al. (2021) aimed to replace commercial iron with bleached laterite iron in the Fenton oxidation process (Bhaskar et al. 2021). Successful attempts have also been made to use biojarosite, an iron hydroxy sulfate, as a catalyst, both in the form of laterite and as a synthesized bioleached laterite nano iron catalyst (Bhaskar et al. 2021, 2022a).

The utilization of statistical response surface methodology for process optimization is a recognized approach for developing prediction models. In the context of PCP removal, factorial design has been employed in optimization efforts (Mathialagan & Viraraghavan 2010; Hassen et al. 2021). Specifically, A 2² factorial design known as Box–Behnken design has been utilized, featuring three interlocking points on the surface sphere surrounding the center, to explore and optimize the impact of key process parameters as independent variables on the response, which serves as the dependent variable (Tripathi et al. 2009; Tak et al. 2015; Shokri et al. 2019). The Box–Behnken design ensures that all design points fall within the selected safe operating zone, providing reliable coefficient estimates close to the center of the design space and eliminating the need for axial points. This design method also enables the identification of significant factors and interactions between variables.

The current study investigates the role of EDTA, acting as a chelating agent, on the dissolution of iron biologically synthesized jarosite in the degradation of pentachloride via Fenton's oxidation process, also evaluating the role of biojarosite as a catalyst in the Fenton's oxidation of PCP at near neutral and acidic pH-value. A comparison is made between this approach and the conventional Fenton oxidation of pentachloride. The regression model is constructed using catalyst loading, H₂O₂ dosage, and pH-value as independent variables, with the removal efficiency of PCP is considered as the response variable.

The synthesis and characterization of biojarosite were conducted by employing a novel acidophilic strain known as Acidithiobacillus ferrooxidans (Accession No. MG271840), following the procedure outlined in a previous study (Bhaskar et al. 2019). To investigate the Fenton degradation of PCP catalyzed by biojarosite, two separate experiments were performed with initial target compound concentrations of 10 mg/L, one with the addition of EDTA and the other without EDTA. In each case, biojarosite was gradually added to the Erlenmeyer flask containing the PCP solution, and the pH-value was adjusted to the desired value with 0.5 M H₂SO₄ or 1 M NaOH. The study encompassed a range of experimental conditions, including different dosages of biojarosite (ranging from 0.1 to 1 g/L) and H₂O₂ (ranging from 100 to 1,000 mg/L), separately for the sets with and without EDTA (Bhaskar et al. 2022b). Samples were collected at regular intervals for analysis. The concentration of PCP was determined using an HPLC Agilent 1200 system equipped with a C18 reverse phase column (pore size 3.5 μm, 100 × 0.46 cm) and a diode array detector (DAD) (Bhaskar et al. 2019). The chemical oxygen demand (COD) was measured using the APHA Method 4500-F: 1992. pH measurements were taken using a digital pH meter (Model – edge, HANNA make). The ferric iron content was determined using the potassium thiocyanate method (Woods & Mellon 1941), while the H₂O₂ concentration was measured using a UV-spectrophotometer (Eisenberg 1943).

This study utilized the Box–Behnken design of response surface methodology. The independent variables were chosen at three levels (−1, 0, +1), while the removal of PCP (pentachlorophenol) served as the dependent variable (Table 1). A total of 15 experimental runs were conducted. The variable parameters were optimized within specific ranges: 0.1–1.0 g/L for biojarosite, 100–1,000 mg/L for H₂O₂ dosage, and pH-value ranging from 2.5 to 3.5 without EDTA, and 6.5 to 7.5 with 100 mM EDTA. The focus was on PCP removal, which was considered a countermeasure. Experimental data were used to develop a regression model based on the response, and 3D contour graphs were employed to assess the influential parameters (Bhaskar et al. 2022b).

Throughout the study, the pH-value remained slightly acidic, ranging from 3.0 to 4.7, in the absence of EDTA. This acidic pH range favored the oxidation process by promoting the formation of hydroxyl radicals (Kang & Hwang 2000; Burbano et al. 2005). The maximum reduction in chemical oxygen demand (COD) observed was 61.23% without EDTA addition and 87.61% with EDTA addition, indicating optimal removal. The adsorption process likely contributed to the removal of organic contaminants, leading to a decrease in COD (Masomboon et al. 2009). The leaching of ferrous iron from jarosite directly influenced the oxidation rate. Maximum degradation occurred within the initial 60 min of treatment, after which the process slowed down. The amount of catalyst added also played a role in the heterogeneous Fenton oxidation process, indicating the involvement of adsorption, which contributed to the enhanced removal of target compounds (Boussahel et al. 2007; Kang et al. 2016; Chen et al. 2017). It should be noted that the rate of oxidation in the Fenton process depends on the rate of dissolution of the metallic iron released from the catalyst. The amount of catalyst increases the degradation efficiency due to the adsorption on the surface of the jarosite catalyst particles, making the reaction process heterogeneous (Boussahel et al. 2007). Bhaskar et al. (2022b) studied the EDTA-based catalytic degradation of dyes using biojarosite, achieving complete dye removal.

According to Fukushima & Tatsumi (2001), tetrachlorocatechol and tetrachloro-hydroquinone were identified as the two major byproducts of Fenton oxidation in the case of PCP. It was observed that at a pH value higher than 5, octachlorodibenzodioxin was formed, while this compound was not detected as an intermediate at pH value 3. The study suggests that the formation of coupling compounds resulting from Fenton oxidation binds with the humic acid used in the experiment, thereby suppressing the formation of octachlorobenzodioxin (Fukushima & Tatsumi 2001).

In another study conducted by Ye and their team (2019), eight intermediates were identified: tetrachlorophenol, trichlorophenol, trichlorohydroquinone, tetrachloro-hydroquinone, hexachlorodibenzobenzofuran, pentachlorodibenzo-p-dioxin, hexachlorodibenzo-p-dioxin, and octachlorodibenzo-p-dioxin. It was proposed that the attack of hydroxyl radicals (OH•) on aromatic rings led to the formation of hydroxylated products, such as tetrachlorophenol, trichlorophenol, and trichlorohydroquinone. Direct attack of OH• radicals on the Cl–Cl bond resulted in tetrachloroquinone formation. Furthermore, the coupling reaction between PCP radicals formed during Fenton oxidation could lead to the generation of hexachlorodibenzofuran and octachlorodibenzo-p-dioxin through further dichlorination processes, resulting in the production of hexachlorodibenzo-p-dioxin and pentachlorodibenzo-p-dioxin (Ye et al. 2019). Thus, mineralization and the formation of dimers from intermediates are significant factors in the degradation of PCP.

During the first cycle of reuse, degradation efficiencies of 81.5 and 89.7% were achieved, with corresponding COD removal rates of 60 and 85.5% for Fenton oxidation and EDTA-based Fenton oxidation, respectively. In the second cycle of reuse, the efficiency decreased to 77 and 79.7%, with COD removal rates of 57.7 and 71.2%, respectively. For the third cycle of reuse, degradation efficiencies were measured at 71.5 and 72.3%, with COD removal rates of 55.5 and 56.5%. In the fourth cycle, the degradation efficiencies dropped to 68.6 and 69.8%, with COD removal rates of 51 and 51.7%, respectively. The reuse of the catalyst resulted in a reduction in efficiency by 14.8 and 23.7% for Fenton oxidation and EDTA-based Fenton oxidation of PCP, respectively.

The significance of PCP removal through Fenton oxidation is indicated by the F-value of the model, which is 42.57, with a noise level of 0.03. The model is highly significant, as evidenced by a probability value of <0.0001. The R-squared value obtained, 0.9871, indicates a strong correlation between the predicted and observed values, suggesting that 98.71% of the variance can be attributed to the variables, highlighting their high significance in the Fenton oxidation of PCP. The model accounts for 98.71% of the total variance, leaving only 1.29% unexplained. The low probability (p < 0.0001) of the F-test, along with the ANOVA quadratic regression model, demonstrates that the model is highly significant for the Fenton oxidation of PCP and shows no lack of fit. The lack of fit F-value of 11.41 implies that the lack of fit is not significantly related to pure error. There is an 8.16% chance that the lack of fit F-value could be attributed to noise. The predicted R-squared values of 0.8037 are reasonably close to the adjusted R-squared value of 0.9639, with a difference of less than 0.2, indicating good agreement between them (Tables 4 and 5).

The significance of PCP removal through EDTA-based Fenton oxidation is indicated by the F-value of the model, which is 47.77, with a noise level of 0.03 (Table 6). The model is highly significant, as evidenced by a probability value of <0.0001. The R-squared value obtained, 0.9885, indicates a high degree of correlation between the predicted and observed values, suggesting that 98.85% of the variance can be attributed to the variables, highlighting their high significance in the EDTA-based Fenton oxidation of PCP. The model accounts for 98.85% of the total variance, leaving only 1.15% unexplained. The ANOVA quadratic regression model demonstrates that the model is highly significant for the EDTA-based Fenton oxidation of PCP, with a low probability (p < 0.0001) for the F-test, and it shows no lack of fit. The lack of fit F-value of 0.53 implies that the lack of fit is not significantly related to pure error. There is a 70.45% chance that the lack of fit F-value could be attributed to noise. The predicted R-squared values of 0.9040 show good alignment with the adjusted R-squared values of 0.9678, indicating that the difference between them is less than 0.2.

The coefficient of variance obtained in the Fenton oxidation model is 3.65%, indicating the model's reproducibility. In the case of EDTA-based Fenton oxidation, the coefficient of variance obtained in the model is 3.0%, suggesting possible reproducibility. These low coefficient of variance values demonstrate the high precision and reliability of the experiments conducted in both Fenton oxidation and EDTA-based Fenton oxidation. The p-values of the regression coefficients indicate that the linear, quadratic, and interaction effects of the independent variables significantly influence both Fenton oxidation of PCP and EDTA-based Fenton oxidation of PCP. These effects are highly significant, further supporting the robustness of the models.

In the present study, the model terms used for Fenton oxidation of PCP were represented by A, B, C, A², B², and C². The statistical analysis conducted on the experimental data revealed that catalyst loading, H₂O₂ dosage, and pH-value had a significant impact on Fenton oxidation. Specifically, jarosite loading and H₂O₂ dosage exhibited a more linear influence on Fenton oxidation, indicating their contribution to PCP removal. Previous research has also reported a higher positive linear effect of iron dosage compared to H₂O₂ in Fenton oxidation. Moreover, H₂O₂ dosage demonstrated the highest positive linear influence (with a high coefficient) among the variables, surpassing the influences of biojarosite and pH-value. The analysis revealed that the independent variables acted in a quadratic manner, influencing the Fenton oxidation of PCP.

The model terms used for the EDTA-based Fenton oxidation of PCP include A (Biojarosite dosage), B (H₂O₂ dosage), C (pH), A², B², and C². Analysis of the experimental data consistently revealed that both jarosite loading and EDTA addition significantly influenced the UV-Fenton oxidation process. The EDTA-based Fenton oxidation showed a positive correlation with jarosite loading and H₂O₂ dosage, indicating that the removal of PCP was enhanced by increasing jarosite loading and H₂O₂ dosage. Among the variables, H₂O₂ dosage exhibited the highest positive linear influence (coefficient of 14.22), followed by jarosite loading (coefficient of 3.46) and pH-value (coefficient of 1.74). However, the quadratic effect of the independent variables had a negative impact on the oxidation of pentachlorophenol in the EDTA-based Fenton reaction.

The catalytic efficacy of biojarosite, synthesized from Acidithiobacillus ferrooxidans, was assessed in the presence and absence of the chelating agent EDTA, confirming its role. The results indicated that PCP degradation through Fenton oxidation with biojarosite was more pronounced at acidic pH-levels, while at neutral pH-value, the addition of EDTA facilitated the dissolution of iron from biojarosite, enhancing its catalytic activity in PCP degradation. Significant removal efficiencies of 93.5 and 83.4% were observed with and without EDTA, respectively. The degradation of PCP followed a pseudo-first-order reaction with R-square values of 0.9607 and 0.9688 for the reactions with and without EDTA, respectively. The efficiency of PCP degradation was further enhanced by increasing the catalyst dosage and H₂O₂ concentration. Both models exhibited excellent fit between the actual and predicted responses, demonstrating the high precision of the Box–Behnken method. The influential parameters were discussed, and quadratic equations were derived for both models. These findings suggest that the utilization of biojarosite as a replacement for commercial iron can enhance the stability of Fenton degradation for organic compounds.

All the authors have consented to publish this article

The authors have consented to participate.

The authors approve the article for publication.

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

The authors declare there is no conflict.