Removal of 17α-ethinylestradiol (EE2) from aqueous solutions by peanut shells (Arachis hypogaea): adsorption kinetic, isothermal, and thermodynamic studies

Contamination of water bodies caused by chemical substances is a global challenging issue for humankind. Since the 20th century, rapid progress in industrialization and urbanization, uncontrolled population growth and climate change have led to the unsustainable use of water. These factors, combined with anthropogenic actions threaten and affect significantly the quality of ecosystems, biodiversity, food, energy and the mineral and water resources of the planet (Hejazi et al. 2011; Silva et al. 2014; Barbosa et al. 2015). The availability of clean water is one of the most threatening issues that our society has been facing. Unfortunately, it has affected more than 3.3 billion people around the world who do not have access to drinking water and/or live with water scarcity and this number may increase 52% by 2025 (Hejazi et al. 2011; Okoli et al. 2017; Olu-Owolabi et al. 2017).

Part of environmental contaminations are related to the emerging contaminants (ECs), which include a wide range of substances capable of causing harm to the surroundings and human health, even in small amounts (ng L⁻¹ to μg L⁻¹) (Han et al. 2010; Choina et al. 2013; Candido et al. 2016). ECs are chemical compounds present in several commercial products (pharmaceuticals, pesticides, personal care products, hormones, illicit drugs, nutritional supplements, diagnostic agents, among others) or microorganisms that can be found in environmental matrices at trace levels. These compounds are hardly monitored and can provide a potential risk to human health and to the environment (Collins & Silva 2011; Choina et al. 2013; Birch et al. 2015; Patiño et al. 2015; Candido et al. 2016).

Endocrine disruptors (EDs), organic micropollutants, are compounds capable of causing undesirable biological effects in live organisms and serious damage to the environment (Choina et al. 2010; Barreiros et al. 2016; Monneret 2017; Wan et al. 2022). In this context, removing EDs from water has been attracting the attention of activist groups, the global scientific community, government agencies, and the private sector. They are able to deregulate, alter or affect the functions of the endocrine system in the following ways: blocking, imitating or minimizing the normal effect of a specific hormone (Dezotti & Bila 2007;,Jardim & Ghiselli 2007; Braga et al. 2011; Fernandes et al. 2011; Ifelebuegu 2012; Silva et al. 2012; Dzieweczynski & Hebert 2013).

Within this class are steroid estrogens (natural or synthetic), a group of biologically active compounds and highly persistent in aquatic environments. They are continuously released into water bodies by human urinary and/or fecal excretions, partially or totally metabolized (Ifelebuegu 2012; Choina et al. 2013; Sadowski & Kopciuch 2013; Birch et al. 2015; Candido et al. 2016; Leonard et al. 2017). Synthetic estrogen, 17α-ethinylestradiol (EE2), the target substance of this study, proceeding from natural estrogen 17β-estradiol (E2), found in pharmaceuticals, mainly in contraceptive pill formulations and hormone replacement therapies for menopause, osteoporosis treatment, menstrual disorders and cancer of the prostate cancer is considered a potent ED (Sodré et al. 2007; Mazellier et al. 2008; Choina et al. 2010; Fernandes et al. 2011; Silva et al. 2012; Han et al. 2013; Aris et al. 2014; Laurenson et al. 2014; Barreiros et al. 2016; Leonard et al. 2017). Furthermore, EE2 has been found in various environments (wastewater, surface water or groundwater), and is a problematic pollutant for aquatic systems because it is highly environment persistent and bioaccumulates in living organs (De Wit et al. 2010; Liu et al. 2021; Wan et al. 2022).

Recent studies have shown that environmental issues due to ED contamination are associated with water and effluent treatment plants (Sodré et al. 2007; Birch et al. 2015; Wan et al. 2022). Therefore, efficiency of treatment plants in removing EE2 is greatly imporatant in combating pollution and reducing EE2 contamination by humans and the environment. According to Wan et al. (2022), several EE2 removal studies in effluent treatment plants are performed all over the world. They use inefficient or inappropriate treatment methods for the complete removal of these pollutants, therefore, the search for alternative and sustainable technologies, processes and techniques becomes a continuous and arduous task (Jardim & Ghiselli 2007; Ifelebuegu 2012; Richardson & Ternes 2014; Patiño et al. 2015; Diagboya & Dikio 2018).

Several technologies have been employed for removing organic pollutants from water, such as filtration, coagulation, chemical precipitation, ion exchange, reverse osmosis, membrane technology, electrochemical, enzymatic treatment, photocatalysis, photolysis, solvent extraction, advanced oxidative processes, activated sludge, ozonization, adsorbents, among others (Aquino et al. 2013; Patiño et al. 2015; Diagboya & Dikio 2018). Tang et al. in a review study report occurrence and removal of EE2 in municipal wastewater treatment plants (WWTPs), totaling 282 WWTPs in 29 countries. 64% of WWTPs use the activated sludge process, anaerobic-anoxic-oxic/anoxic-oxic 6%, oxidation ditch 5%, member bioreactor 3%, biofilm 6%, stabilization pond 7%, primary treatment 3% and other processes 6%, with EE2 removal averages in the range of 47.5–83.6%, and observed that the main mechanisms of EE2 removal in municipal WWTPs include mainly adsorption and biodegradation. Another review study reports that several physical, chemical, biological and hybrid processes applied as treatment methods for endocrine disrupting compounds have been extensively studied, in addition to promising adsorption technologies, especially those that use agricultural residues as adsorbents. Other works on EE2 removal, degradation and adsorption can be found in the literature (Heo et al. 2011; Ifelebuegu & Ezenwa 2011; Valladares Linares et al. 2011; Ifelebuegu et al. 2015; Li et al. 2016; Goswami et al. 2018; Liu et al. 2019; Procópio et al. 2020).

However, unconventional techniques like the adsorption methods have emerged with great potential and economic viability for the treatment of effluents and contaminated water, mainly involving natural adsorbents from wastes from agro-industrial activities and biodegradable materials from by-products from agricultural and industrial operation (Fernandes et al. 2011; Han et al. 2013). Removal of EE2 and EDs has always been difficult to perform and the use of these materials as a cost-effective alternative, robustness, ease of use and application, good efficiency are of great interest to the scientific community and a great option for traditional adsorbents (Braga et al. 2011; Fernandes et al. 2011; El-Sayed et al. 2014; Barros et al. 2017). However, they are discarded in large quantities, making them viable, renewable and ecologically correct, reducing the direct release of pollutants into the environment and assisting in the wastewater treatment. China and USA are the world's largest producers of peanut shells and generate huge amounts of tons of by-products per year and at low cost (Brown et al. 2000; Zhu et al. 2009; Al-Othman et al. 2012).

In this context, the objective of this work was to evaluate the use of peanut shells, a lignocellulosic agro-industrial residue, as an alternative means of effective biomass adsorption and low cost, for the removal of the synthetic hormone EE2 from the aqueous medium. Furthermore, experimental data were analyzed using pseudo-first-order, pseudo-second-order and Elovich adsorption kinetic models, and kinetic constants were calculated as a function of temperature. Adsorption equilibrium was expressed by Langmuir, Freundlich and Sips isotherms. Activation energy of the adsorption process, which is a very important indicator, was also evaluated.

With the results obtained in this work, it is possible to show the technical potential that the material has for removing contaminants emerging from the aqueous medium. In addition, it is possible that it is an economically viable alternative, since it is an industrial waste that is normally discarded.

Ultrapure deionized water (18.2 MΩ*cm 298 K resistivity) was obtained from a Direct-Q (Millipore^®) purification system. 0.1 mol L⁻¹ of HCl solution and 0.1 mol L⁻¹ of NaOH solution were used to adjust pH of the solutions. Acetonitrile (ACN) [≥99.9%, HPLC grade] and methanol [≥99.9%, HPLC grade] were purchased from Sigma–Aldrich^®, and the EE2 standard [≥99.4% purity] was purchased from Fluka Analytical^®. Standard EE2 solution of 10 mg L⁻¹ in methanol was prepared for the evaluation of the experimental parameters (selectivity, linearity, precision, accuracy, detection limit, quantification limit, and robustness) of the chromatographic methodology. 10.00 mg of the EE2 standard was weighed, dissolved in methanol, and quantitatively transferred to a 100.00-mL volumetric flask, completing the volume with methanol and homogenizing. Subsequently, a 10.00 mL aliquot of the standard solution, previously prepared, was pipetted into a 100.00-mL volumetric flask, completing the volume with methanol and homogenizing. For the adsorption experiments, a solution of EE2 was obtained from Ciclo 21 contraceptive (União Química^®). One pill was macerated and dissolved with deionized water, and then transferred to a 100.00-mL volumetric flask. According to the manufacturer, each tablet contains 0.03 mg of EE2, however, for greater reliability of the results, the concentrations of EE2 in the prepared solutions and after the adsorption process were determined by the chromatographic method. Solutions were stored in a polyethylene bottle at 277 K.

EE2 identification and quantification analyses were performed in a high-performance liquid chromatograph (Infinity 1260 – Agilent Technologies^®, Germany) composed of a quaternary pump, an automatic injector, a column heating module, and a molecular fluorescence detector. An Eclipse Plus C8 chromatographic column (4.6 mm × 150 mm – 5 μm) from Agilent Technologies was used. Chromatographic condition used for the analysis was mobile phase (MP) in isocratic mode – ACN:water (3:2), MP flow – 0.5 mL min⁻¹, injection volume – 10 μL, chromatographic column temperature – 318 K, wavelength of the fluorescence detector (FLD) – 230 nm (excitation) and 310 nm (emission). The MP was previously filtered through cellulose ester (Millipore^®) and nylon membranes (Supelco Analytical) with 0.45 μm porosity and deafened in an ultrasonic bath (SoniClean 2 Model – Sanders Medical). The quality control and quality assurance of the chromatographic method can be consulted in our other study (Procópio et al. 2020), however, in summary, the evaluated parameters were selectivity, linearity, precision, accuracy, limit of detection, limit of quantification, and robustness.

To save disposal costs and possible environmental problems, in this work an agro-industrial residue or by-product agricultural abundant, cheap and renewable was used as a natural adsorbent for the adsorption tests in the EE2 removal. Raw peanut shells, free from washing, drying, grinding, sieving and chemical treatments, were used. The materials were collected at the local market in the city of Itajubá, Minas Gerais, Brazil. A lignocellulosic compound, peanut shells are potential adsorbents in the pollutant removal due to their physical–chemical characteristics. According to Brown et al. (2000), raw peanut shells have 8–10% humidity, 6–7% protein, 60–70% fiber, 34–45% cellulose, 27–33% lignin, 2–3 ash and 61% porosity. Also due to their chemical composition and complex structures, lignocellulosic materials can be used to adsorb pollutants with no modification or physical and chemical treatment (Ossman et al. 2014).

Adsorption experiments were performed at room temperature (approximately 293 K). Beakers of 150 mL were used containing the mass of peanut shells and 100 mL of the solution of EE2 contraceptive Cycle 21. Experiments were carried out at a constant agitation for 6 h. The parameters used and evaluated in the adsorption process were adsorbent mass (0.5, 1 and 2 g), pH (4, 6 and 8) and stirring speed (300, 400 and 500 rpm). For calculation purposes, the possible errors arising from the changes in the volume of the solution due to the water absorption by the different mass of peanut shell were disregarded.

1.0 mL of aliquots of the solution were collected at time intervals of 1, 2, 3, 4, 5 and 6 h. All aliquots were filtered using qualitative paper to remove larger particles and impurities. Furthermore, for the removal of smaller particles, a syringe (10 mL – Art Glass) coupled to a filter (Allcrom) with 0.45-μm porosity was used.

Experimental conditions of the EE2 adsorption process for the kinetic study were as follows: 2 g (adsorbent mass), 6 (pH) and 500 rpm (stirring speed). The experiments were performed in triplicate and its preparation follows the adsorption experiments step mentioned above. However, aliquots were collected at times 0, 1, 2, 3, 4, 5, 6 and 24 h. The temperatures used for the experiments were 309, 329 and 339 K.

Isothermal study was performed under the same experimental conditions as the kinetic study, but varying the EE2 concentration in the solution. For this, 500 mL of solution containing 15 EE2 contraceptive Cycle 21 pills was prepared. Aliquots of 5, 10, 20, 30, 40 and 50 mL were transferred to 100-mL volumetric flasks, completing the volume with deionized water. Subsequently, the solutions were transferred to beakers for the adsorption process. 1 mL of aliquots were collected at time 0 and 24 h and the EE2 remaining concentrations in the solutions were determined by Equation (2). The temperatures used for the experiments were 306, 326 and 333 K.

A multivariate optimization technique using the Box–Behnken model was used for the adsorption experiments as in our other study (Procópio et al. 2020) with the aid of Statistica^® software, and an Excel spreadsheet or Origin^® software for the treatment of the data. The parameters (stirring speed – 300, 400 and 500 rpm; adsorbent mass – 0.5, 1 and 2 g and pH – 4, 6 and 8) were varied in three levels, totaling 15 experiments (Table 1).

From Table 1 it was observed that experiment 8 (stirring speed = 500 rpm, mass = 1 g and pH = 8), experiment 12 (stirring speed = 400 rpm, mass = 2 g and pH = 8), and experiment 4 (stirring speed = 500 rpm, mass = 2 g and pH = 6) showed the best results, with 60, 63 and 64% of EE2 removal, respectively. It is noted that the adsorbent mass and the stirring speed were the main factors that affected the results in a more pronounced way.

The stirring speed had great influence on the removal process because it is directly linked to the distribution of the analyte in the solution and the contact of the adsorbent mass (Esmaeli et al. 2013; Fontana et al. 2016a). However, the adsorbent mass was the most important parameter in EE2 removal, as it was the determinant factor in the adsorbent/adsorbate equilibrium. This phenomenon can be related to the increase in available sites for EE2 adsorption as the adsorbent mass in the aqueous medium increases, which in this case was 0.5–2 g of peanut shells (Fontana et al. 2016b). The pH of the solution had little significance in the EE2 removal process, although it showed better results varying the values from 6 to 8 (Wang et al. 2015).

Therefore, according to the preliminary data in our other study (Procópio et al. 2020) and the results of Table 1, the best experimental condition found for EE2 adsorption and removal by peanut shells was 500 rpm of stirring speed, 2 g of adsorbent mass and pH of 6. In this optimum condition, a new experiment was carried out in triplicate (100 mL EE2 263 μg L⁻¹), whose stirring period was 24 h, obtaining a total EE2 removal by the peanut shells close to 90%.

The EE2 removal rate found in this study was considered significant, when compared to other studies involving EE2 adsorption and removal. Researches involving peanut shells in EE2 adsorption have not been found, but it is possible to find some interesting studies about EE2 removal by other materials, for example Fernandes et al. (2011). In this work, the authors evaluated the EE2 removal rate from aqueous solutions using decomposed peat as adsorbent and, as a result, obtained approximately 55% of removal using an EE2 initial concentration of 0.1 mg L⁻¹ and varying the adsorbent mass from 50 to 200 mg and with the adsorption process operating through magnetic stirring for 36 h. Wang et al. (2018) reported EE2 removal using magnetic ion exchange resins and, as a result, obtained a maximum removal of 75% for EE2 at the initial concentration of 20 μg L⁻¹. On the other hand, Clara et al. (2004) evaluated EE2 adsorption in activated and inactivated sludge, EE2 at 1 mg L⁻¹ and adsorbents at concentrations of 1–7 g L⁻¹, stirred over a period of 24 h. Even at very high initial concentrations, a great adsorption affinity to the adsorbent could be observed by the results obtained from the parameters of partition or distribution coefficient, organic matter, and the organic carbon content of the sorbent. Rudder et al. (2004) studied the removal of EE2 from water by an advanced treatment using three upstream bioreactors with a capacity of 2 L that were filled through sand, granulated activated carbon and MnO₂ granules, obtaining respectively, 17.3, 99.8 and 81.7% of EE2 removal by varying the initial concentration of EE2 from 5 to 20 μg L⁻¹. Cheng et al. (2018) investigated microalgae mutant Chlorella PY-ZU1 under 15% CO₂ in the EE2 removal in wastewater. The authors used EE2 concentrations in the range of 0.01–5 mg L⁻¹ to stimulate microalgae growth, obtaining a 94% removal rate by Chlorella PY-ZU1 at an EE2 concentration of 5 mg L⁻¹. Finally, Almeida et al. (2022) synthesized a nanocomposite based on graphene oxide, magnetic chitosan and organoclay (GO/mCS/OC) for EE2 adsorption. They evaluated the effects of dosage, sonication time and pH on the process, resulting in a material with a maximum adsorption capacity of 50.5 mg g⁻¹ at 30 °C.

In this sense, comparing the rate of adsorption and/or removal of this contaminant with other studies, there is a great alternative for the peanut shells used in this work for applications in processes of EE2 adsorption, a natural adsorbent of residues from agro-industrial activities, biodegradable, available in large quantities and ecologically correct.

Therefore, the sharp start of the adsorption kinetics may be related to the phenomena of diffusion and convection, external mass transfer and internal diffusion. These phenomena are directly linked to the transport of EE2 molecules within the solution to the adsorbent external surface, known as the hydrodynamic limit layer, and the transport of the solute by diffusion through the limit layer to the pores of the adsorbent referring to external diffusion. After 6 h, the adsorption kinetics may be related mainly to the internal diffusion of EE2 molecules in the adsorbent, in which the adsorption process occurs on the surface and internal pores of the adsorbent, a region with low availability of active sites and mass transfer (Balsamo & Montagnaro 2019; Baup et al. 2000; Dabrowski 2001; Fernandes et al. 2011). According to Balsamo & Montagnaro (2019) the kinetics of the process does not depend on the stage in which the adsorbate (contaminant) is adsorbed on the external surface of the adsorbent, however, the internal and external diffusion step of the adsorbate on the surface and internal pores of the adsorbent is determinant in the kinetics. Therefore, the kinetics is dependent on the intrinsic characteristics of the adsorbate (molecule size) and adsorbent (porosity) that affect the total rate of adsorption, beyond to the surface area, pore size distribution and the chemical nature of the adsorbent, for example, greater than the pore size of the adsorbent cause its inactivation, reducing the useful surface of adsorption (Carolin et al. 2017).

It was observed that 227.5 μg L⁻¹ of EE2 was adsorbed at an equilibrium of 309 K. When the temperature is increased from 309 to 329 K, only 147 μg L⁻¹ of EE2 was adsorbed by the peanut shells, a decrease in the percentage of EE2 removal from 95.6 to 62.0%. The increase in temperature favored the inverse reaction, that is, in the direction of the reactants formation and the solute (EE2) residual concentration increased in solution, as seen in Figure 1. At 309 K, the EE2 concentration in solution was 10.5 μg L⁻¹, while at temperatures of 329 and 339 K, the concentrations were 90.2 and 65.4 μg L⁻¹, respectively, in 24 h of process. However, an experimental anomaly was observed in the efficiency of the process, which may be related to the temperature variation and the agitation of the solution by the heating plate used for the process. Also a more appropriate way to minimize this experimental error would be to use a thermostatic bath with agitator in a controlled manner.

Table 2 presents the kinetic parameters and correlation coefficients (R²) for the kinetic models applied at the different temperatures studied. The experimental data obtained from the EE2 adsorption process on peanut shells were better adjusted to the Elovich kinetic model at 309 K (R² = 0.9802) (Figure 2(b)), pseudo-first-order at 329 K (R² = 0.9665) (Figure 2(c)), and pseudo-second-order at 339 K (R² = 0.9920) (Figure 2(d)), which presented the highest R² at the respective temperatures. However, the values of q_e calculated from the nonlinear adjustment of the pseudo-first-order kinetic model for the temperatures 309, 329 and 339 K were concordant to the experimental q_e values. The pseudo-second-order model showed good R² values, however, the q_e values of calculated were relatively different from the experimental q_e values. Based on these results, it is suggested that the adsorption system between EE2/peanut shells is more a first-order reaction than a second-order one. Therefore, from the experimental data (Figure 1) and what was said earlier, it is assumed that the adsorption process does not occur by a single mechanism, but rather by a mixture of mechanisms. The adsorption kinetics of EE2 molecules occurs mainly by the mechanism of external diffusion, observed by the abrupt onset of kinetics up to 6 h of reaction, a region with high availability of active sites and mass transfer. Such a mechanism is dependent on the amount of adsorbate adsorbed and on the active sites on the adsorbent surface (Krishnan & Anirudhan 2002; Yang & Al-Duri 2005; Fontana et al. 2016a, 2016b; Haro et al. 2017; Silva et al. 2018).

In the literature, the study of isotherms is focused on gas/solid and liquid/solid systems, in this case, for a gas/solid system, the experimental data showed a V-type isotherm behavior, in which the adsorbent–adsorbate interactions are weak, but they are obtained with certain porous adsorbents of mesoporous structure with pores from 2 to 50 nm (Brunauer et al. 1938; Sing et al. 1985); however, for a liquid/solid system it has characteristics of a sigmoidal isotherm, in which there is a moderate attraction between adsorbate and adsorbent, and competition of solvent and solute molecules for the active sites of the adsorbent (Moreno-Castilla 2004).

From Table 3 and Figure 4, the experimental data of the adsorption process were better adjusted to the Sips isothermal model for the respective temperatures of 306 K (R² = 0.9936) (Figure 4(b)), 326 K (R² = 0.9984) (Figure 4(c)) and 333 K (R² = 0.9656) (Figure 4(d)) however, it presented n values greater than 1, a parameter that predicts the heterogeneity of the adsorbent surface and indicates whether the adsorption process can occur in monolayers or multilayers. Langmuir model presented calculated qmax values totally inconsistent with the experimental values and unfavorable R² values. However, the EE2 adsorption on peanut shells adjusted the Freundlich model at 306 K (R² = 0.9840, n = 0.6181) and the increase in temperature disfavors the model, as the R² values decrease and the n values increase, but the adsorption is still favorable in the interval 1 < n < 10, and for 1/n > 1 the isotherms present upward or concave curvature where the marginal sorption energy is proportional with the surface concentration (Delle Site 2001; Febrianto et al. 2009; Eren et al. 2010). Therefore, the adsorption phenomenon can be described by the Freundlich isotherm, an indicating that the EE2 adsorption on peanut shells can occur in multilayers and present a heterogeneous surface of the active sites of the adsorbent, which can be caused by the presence of different functional groups and the various interactions between adsorbent–adsorbate. However, as the Sips isotherm is the combination of the Langmuir and Freundlich model, and based on the results of R², it can't be disregarded, even with n > 1, the adsorption can also occur in a set of monolayers and the adsorbent presents characteristics of a homogeneous surface (Baláz et al. 2005; Deng et al. 2009; Eren et al. 2010).

Fei et al. (2017), in a study of the characterization of EE2 adsorption behavior in marine sediments, obtained better EE2 adsorption isotherms adjusted by Freundlich model. In the tested concentration range, a K_F from 15.8 to 39.8 L kg⁻¹ and a 1/n value from 0.5 to 0.8 were found, suggesting that EE2 adsorption would become saturated when EE2 residual concentration in aqueous medium increased. However, Han et al. (2013), used polyamide particles 612 (PA612) of industrial grade as adsorbents for EE2 removal obtaining values for qmax, K_F and 1/n equals to 25.4 mg g⁻¹, 16.1 (mg g⁻¹) (L mg⁻¹)^1/n and 0.87, respectively, indicating that the PA 612 particles have affinity for EE2 solutes in water. Joseph et al. (2013), studying the EE2 removal by coagulation and combined adsorption using single/multiple-walled carbon nanomaterials and powdered active charcoal in landfill leachate matrices, obtained excellent results. EE2 adsorption capacity was in the log K_F range from 3.41 to 1.38 and 1/n value from 0.32 to 1.36, representing over a 90% removal, which the mechanism is mainly related to the hydrophobic interactions and donor /receptor electron interactions π-π. Besides, studies on the adsorptive capacity and use of peanut shells in solution to remove contaminants such as metals were found in the literature, but not for EE2 (Brown et al. 2000; Johnson et al. 2002; Romero et al. 2004; Zhu et al. 2009; Al-Othman et al. 2012; Ossman et al. 2014).

According to the results in Table 4, the ΔG° negative values indicate that the EE2 adsorption on peanut shells is a spontaneous process, that is, the process was thermodynamically favorable and directly proportional to the temperature. ΔH° positive value indicates the endothermic nature of the adsorption process, where energy is absorbed by the EE2/adsorbent system. The value of 168.9 kJ mol⁻¹ (ΔH°) is characteristic of the existence of strong intermolecular interactions between EE2/adsorbent and a high probability of the process occurring by chemisorption. The positive value of ΔS° (0.6022 kJ K⁻¹ mol⁻¹) suggests a small increase in randomness at the adsorbate-adsorbent interface (system) and solution, a greater affinity of the adsorbent for the adsorbate and the adsorbate is less hydrated on the surface of the adsorbent than in solution (Liu & Xu 2007; Ghosal & Gupta 2017; Fang et al. 2021; Tang et al. 2021).

In this study, raw peanut shells as adsorbents showed performance in removing EE2 from the aqueous medium. The high removal and adsorption capacity even with a small amount of adsorbent and low concentration of adsorbate, suggests peanut shells as a promising adsorbent for application in wastewater treatment. From the best adsorption condition (agitation speed = 500 rpm, adsorbent mass = 2 g and pH = 6), 90% removal was obtained and an adsorptive capacity of 17.3 μg g⁻¹ at 306 K. Kinetic data adjusted the pseudo-first-order model, however, it indicated that adsorption can occurs by a mixture of mechanisms, in which the mechanism of external mass transfer and external diffusion were the main steps of the adsorption process, directly related to the abrupt drop in kinetics after 6 h of the process. The system presented an E_a of 16.69 kJ mol⁻¹, indicative of a slow adsorption kinetics because of the high reaction potential barrier. The Sips isothermal model was able to describe the sorption behavior for the adsorbent and the EE2 as it presented a better fit to the experimental data, suggesting that peanut shells have heterogeneous characteristics and that there is formation of monolayers and multilayers by chemisorption. ΔG° negative values indicated that the EE2 adsorption process was a favorable and spontaneous process. ΔH° positive value (168.9 kJ mol⁻¹) indicated the endothermic and chemical nature of the adsorption process. ΔS° positive value (0.6022 kJ K⁻¹ mol⁻¹) indicated a small increase in the degree of disorganization and affinity of the EE2/adsorbent system. Therefore, the compilation of results points to peanut shells as a promising adsorbent alternative for the removal of endocrine interferents, with impactful benefits for nature and suggests an evaluation of this adsorbent in real aqueous matrices for possible wastewater treatment.

The authors would like to thank the Coordenação de Aperfeiçoamento de Pessoal do Nível Superior (CAPES) for a fellowship and the Rede Mineira de Química (RQ-MG) and Multicentric Chemistry Postgraduate Program for financial support.

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

The authors declare there is no conflict.