Enhanced removal of estrogens from simulated wastewater by biochar supported nanoscale zero-valent iron: performance and mechanism

The intensification of estrogen non-point source pollution has drawn global attention due to their contribution to ecological environment problems worldwide, and it is critical to develop effective, economic and eco-friendly methods for reducing estrogens pollution. To address the agglomeration and oxidation of nano zero-valent iron (nZVI), biochar-nanoscale zero-valent iron composite (nZVI-biochar) could be a feasible choice for estrogens removal. This study summarized biochar and nZVI-biochar preparation, characterization, and unusual applications for estrone (E1), 17β-estradiol (E2), and estriol (E3) removal. The properties of biochar and nZVI-biochar in characterization, effects of influencing factors on the removal efficiency, adsorption kinetics, isotherm and thermodynamics were investigated. The experiment results showed that nZVI-biochar exhibited the superior removal performance for estrogens pollutants compared to biochar. Based on the quasi-second-order model, estrogens adsorption kinetics were observed, which supported the mechanism that chemical and physical adsorption existed simultaneously on estrogens removal. The adsorption isotherm of estrogens could be well presented by the Freundlich model and thermodynamics studies explained that nZVI-biochar could spontaneously remove estrogens pollutants and the main mechanisms involved π-π interaction, hydrophobic interaction, hydrogen bonding and degradation through ring rupture. The products analyzed by GC–MS showed that estrogens degradation was primarily attributed to the benzene ring broken, and Fe3+ promoted the production of free radicals, which further proved that nZVI-biochar had the excellent adsorption performances. Generally, nZVI-biochar could be employed as a potential material for removing estrogens from wastewater. Porous and functional nZVI-biochar was prepared and applied for estrogens removal. nZVI-biochar exhibited high adsorption capacities to estrogens pollutants due to π–π interaction, hydrophobic interaction and hydrogen bonding. The estrogens adsorption process of nZVI-biochar conformed to quasi-second-order kinetics and the Freundlich model. The production of free radicals contributed to the degradation of estrogens. Porous and functional nZVI-biochar was prepared and applied for estrogens removal. nZVI-biochar exhibited high adsorption capacities to estrogens pollutants due to π–π interaction, hydrophobic interaction and hydrogen bonding. The estrogens adsorption process of nZVI-biochar conformed to quasi-second-order kinetics and the Freundlich model. The production of free radicals contributed to the degradation of estrogens.


Introduction
The increasing of endocrine-disrupting chemicals (EDCs) detected in agricultural, industrial, and domestic sewage is becoming a serious problem threatening human health and ecosystems (Sutherland and Ralph 2019).Compared to other EDCs, E1, E2 and E3 were considered to the most overbearing due to their strong estrogenic activity (Silva et al. 2012), which led to endocrine disorders, abnormalities and dysplasia (Li and Zhang 2014;Hayes et al. 2002;Campbell et al. 2006) at only ng•L −1 level.According to previous studies, municipal wastewater was is a hotspot for estrogens, and the presence of E1 and E2 was discovered in sewage treatment plants (STPs) effluent due to the incomplete removal from STPs (Ben et al. 2017).Livestock production and animal waste caused high amounts of estrogens transferred to environment (Adeel et al. 2017).As reported, up to roughly 17,000 μg of estrone (E1) and 384 μg of estradiol (E2) from farm animals per day were released to environment (Matthiessen et al. 2006).Alarmingly, estrogens were even detected in drinking water treatment plants, which meant estrogens posed a serious threat to human health (Deich et al. 2021).Recently, the maximum limit of E2 in water has been set by the European Union to be 0.4 ng•L −1 as the Environment Quality Standard.
To reduce estrogens pollution in the environment, various methods were explored to remove estrogens, such as adsorption (Rovani et al. 2016), membrane filtration (Fang et al. 2017), biodegradation (Lucas and Jones 2006), and advanced oxidation processes (Ren et al. 2017).A novel iron-doped graphitic biochar based AOPs was applied to remove estrogens in water, where activated persulfate contributing to degradation of estrogens (Zhang et al. 2019).However, strict operational requirements were needed and secondary pollution occurred simultaneously in the environment by AOPs.Membrane filtration was another method to remove estrogens and 100% of estrogens removal were obtained by microporous polyethersulfone hollow fiber membranes (Hu et al. 2003).Nevertheless, filtration devices were easily blocked, which required higher operating cost.Compared to these approaches, adsorption was regarded as an promising method for estrogens removal from aqueous solutions due to its accessibility and high efficiency (Jawad et al. 2022a(Jawad et al. , 2022b;;Abdulhameed et al. 2022;Zain et al. 2023).Many sorbent materials have been reported for estrogens removal from water, including molecularly imprinted polymers (Noir et al. 2007), powdered activated carbon (Paune et al. 1998), and carbon nanotubes (Joseph et al. 2011).However, it has several limitations to be applied on a large scale due to low adsorption capacity, high cost, and difficult regeneration.In this regard, it is necessary to develop an economical and renewable adsorbent with the excellent adsorption performances for removal of estrogens pollutants based on reuse of waste resources (Zhang et al. 2023;Wu et al. 2022;Huang et al. 2023).
Nano-zero-valent iron (nZVI) has gained significant attention as a potential material for environmental remediation, which could give a large number of adhesion sites for pollutants due to their large specific surface area (Zhao et al. 2023).nZVI with a low reduction potential (E 0 ) and high reactivity could remove various environmental pollutants by reduction and sorption, such as organic contaminants (Kakavandi et al. 2019), and heavy mentals (Liu et al. 2015).However, nZVI particles were known to face challenges in practical applications such as aggregation and oxidation.Aggregation could lead to reduce contact between nZVI particles and pollutants because of its small size and inherent magnetism, limiting the remediation potential of nZVI.And nZVI was easily aged in the environment and easily oxidized to Fe 2+ /Fe 3+ in aerobic environments, which greatly reduced its reactivity.For further applications, many researchers developed effective methods of immobilizing nZVI on carriers (biochar, metal oxide, zeolite, etc.) (Li et al. 2018;Pei et al. 2020;Zhang et al. 2020), which enhanced stability and reactivity of nZVI.For example, nZVI was modified with mesoporous activated carbon to absorb bisphenol A rapidly with lower diffusion resistance (Xu et al. 2022a), and zeolite for ammonia-nitrogen (NH4 + -N) removal (Eljamal et al. 2022).
Biochar is a low cost and effective carbonaceous material for dispersing and stabilizing nZVI because of its surface charge, porous structures and abundant functional groups (Karunaratne et al. 2022), which is obtained through carbonization of waste biomass in an oxygenlimited environment.Agricultural wastes, crustacean shells and animal feces are common biochar precursor materials (Li et al. 2022;Sun et al. 2022).Up to 1.5 million ton of walnut shell per year by 2020 was produced in China (Zhang et al. 2015), with superior hydrophilicity, high specific surface area (Xu et al. 2021) and wear resistance (Yu et al. 2016), which was considered as an promising raw material of biochar.Combination of biochar and nZVI could effectively protect nZVI from agglomeration and rapid oxidation and increase dispersion on biochar, thus enhance the removal of pollutants (Wang et al. 2019).As reported, nZVI-biochar has been explored for pollutants removal, such as antibiotics (Wei et al. 2019), Cr (VI) (Liu et al. 2022a), and nitro and chlorinated contaminants (Ahmad et al. 2021).However, there were few studies on estrogens removal by nZVI-biochar.
Hence, the purpose of this study was an advanced overview of efficient estrogens removal from simulated wastewater by a magnetic nZVI-biochar composite.The structural properties of nZVI-biochar were investigated through various characterizations (SEM-EDS, TEM, XRD, FTIR, BET).Removal effects of nZVI-biochar on estrogens were systematically studied, including initial pH, reaction temperature, Fe/C ratio, and contact time.In addition, adsorption kinetics, isotherm and thermodynamic were further investigated for better understanding of the adsorption data.The GC-MS was used to monitor degradation products, which better explained estrogens removal pathway and mechanisms.The outcomes of this research would help to select an effective component of nZVI-biochar, which had practical application value for the degradation of estrogen in simulated wastewater.

Biochar and nZVI-biochar preparation used in this experiment
Biochar was prepared by slow pyrolysis of the walnut shell under nitrogen protection (Liu et al. 2022b).Walnut shell was chosen as raw material for biochar production, and was ground into smaller than 2 mm powder by a grinder.
After washing with deionized water to remove residual sugar, walnut shell powder was dried in an electric blast drying furnace for 12 h at 65 °C.Then, the feedstock was pyrolyzed at 700 ℃ using a glass tube furnace in an oxygen-free environment (BR-12NT-120/300-2, China) for obtaining biochar.The heating rate was 15 ℃•min −1 , and the pyrolysis continued for 1 h after reaching 700 ℃. nZVI-biochar particles were prepared using walnut shell biochar as a carrier.As described in the previous study (Xu et al. 2021), nZVI-biochar could be synthesized by liquid-phase reduction method.24.88 g FeSO 4 •7H 2 O and 5 g biochar were dissolved in 200 ml ethanol-water (ethanol: water = 1:3, v/v) to prepare nZVI-biochar and stirred continuously for 30 min.Afterward, 50 ml NaBH 4 reducing agent was added to the solution at a flow rate of 1.5 mL•min −1 under vigorous magnetic stirring (150 rpm) for 1 h to ensure the completion of nZVI growth (Wang et al. 2006;Ghauch et al. 2009).Filtered nZVI-biochar was rinsed three times alternately with ethanol and degassed reverse osmosis water (Wang et al. 2006) to avoid nZVI oxidation.This synthesis procedure was conducted by nitrogen continuously bubbled into the reaction system.The obtained samples were dried overnight in a vacuum at 70 ℃ (Sun et al. 2006;Lv et al. 2011).

Reagents
nZVI-biochar was prepared using FeSO 4 •7H 2 O, NaBH 4 , and ethanol (99%, v/v), all from Shanghai MACKLIN Reagent Co. Ltd.Estrone, 17β-estradiol, and estriol used in the adsorption experiment were purchased from Sigma Aldrich (USA), and the purity was greater than 98%.All organic solvents adopted for high-performance liquid chromatography came from American Tedia company.The compound was dissolved in methanol to prepare a target estrogen and internal standard with a concentration of 1000 mg•L −1 , and stored in refrigerator at 4℃.The Wahaha (China) company provided distilled water.

Characterization of materials
The configuration of different materials was observed by scanning electron microscopy (SEM-EDS, Inspect F50, USA) and transmission electron microscopy (TEM, technai G2 F20, USA).The BET specific surface area and pore size distribution of different materials were determined using a static N 2 adsorption instrument (Autosorb-iQ2-MP, USA).XRD (Smartlab9, Japan) was used to analyze the crystal structure and crystallinity of different materials.NICOLET-380 FTIR spectrometer (USA) was used to record Fourier transform infrared (FTIR) spectra in the range of 4000 to 400 cm −1 using KBr disk.

Sorption experiments
All estrogens removal experiments were accomplished in conical flasks, and 100 mg•L −1 NaN 3 was added to restrain microbial degradation.The reaction mediums were stirred with a temperature-controlled water bath oscillator (SHA-B) at 180 rpm to maintain the adsorbent uniformly dispersed in the water environment water environment.To remove oxygen from all the solutions, nitrogen was injected for 15 min.Unless otherwise required, the reaction temperature was room temperature.The initial pH of the solution was adjusted through the addition of 0.1 mol•L −1 HCl or 0.1 mol•L −1 NaOH.nZVI-biochar was prepared using various mixing ratios of nZVI to biochar (1:1; 2:1; 3:1) in the tests for finding the optimum addition ratios of nZVI to biochar.The adsorption study was conducted with different materials, pH variations (4 ~ 9), different estrogen concentrations (500, 1000, 1500, 2000, 2500 μg•L −1 ), different iron-carbon ratios (1:1; 2:1; 3:1), and different temperatures (279, 284, 298, 324, 349 K).A 0.22 μm membrane was used to filter 1 mL of samples periodically during batch experiments for analysis of estrogens.Each experiment was conducted in triplicate .
The adsorption kinetics of estrogens was studied at 25 ℃.The different absorbents were added alone to investigate the adsorption of estrogens.Three estrogens were initially added to the water solution at an initial concentration of 1000 μg•L −1 .All the flasks were stirred at 180 rpm in the dark.Water samples were extracted after 10, 30, 60, 120, 360, 480, and 720 min of shaking, and immediately filtered through a 0.22 μm membrane for estrogens analysis.Each experiment was conducted in triplicate.
Adsorption isotherms and adsorption kinetics were performed under the same conditions.After adding different concentrations of estrogen solution (500 ~ 2500 μg•L −1 ), all the flasks were stirred at 180 rpm in the dark and sampled after 1 h based on the kinetics results.The samples were immediately filtered through a 0.22 μm membrane for further analysis, and all experiments were conducted in triplicate.The estrogens adsorption by glassware and soluble organic matter, and photodegradation were regarded to be minimal.

Determination of estrogens concentration and products
The concentration of estrogens was obtained by High Performance Liquid Chromatography (Ultimate 3000 Series), Thermos C18 column (150 mm*4.6 mm, 5 mm).0.22 μm hydrophilic polyvinylidene fluoride filter membrane was used first for estrogen filtration before testing samples.The wavelength was 200 nm and the column temperature was 25 °C.The injection volume was 50.0 μL and the flow rate was 1.0 mL•min −1 .The mobile phase was 50% acetonitrile and 50% deionized water.The content of estrogens was determined adopting a standard curve featuring the correlation coefficient greater than 0.99.
The degradation products of estrogens were analyzed using Gas Chromatography Mass Spectrometry (GC Varian CP3800/MS Saturn2200, USA), and water samples were extracted by Sep-Pak C18.Firstly, 10 mL deionized water was added to the Sep-Pak C18 small column to remove impurities.Then, the filler was soaked in 10 mL methanol for 5 min.Finally, the column was rinsed 3 times with 15 mL deionized water at a flushing rate of 1-2 mL•min −1 .Then 1 L water sample passed through the C18 with a flow rate of 5 mL•min −1 .The tested item remained on the C18 small column and was vacuum dried for 5 min to remove any residual moisture.Using a nitrogen blowing instrument, the concentrated pollutant was completed by washing with 10 mL acidified methanol and 20 mL dichloromethane at 1 mL•min −1 after 30-min drying at 40 °C, and then transferred to a 1.5 mL injection bottle.Afterward, 50μL of BSTFA and pyridine was added, and the concentrate was derivatized at 70 °C for 45 min and 1 μL was taken for analysis.

Data analysis
The adsorption kinetics behavior was evaluated by Quasi-first-order (QFO), Quasi-second-order (QSO) kinetics models and Elovich model (Qiu et al. 2009).The results from the adsorption isotherm experiment of different materials were studied adopting Langmuir and Freundlich models.The experiment data were also analyzed with thermodynamic model.The specific formulas and parameters of six models are presented in Table 1.
Where K 1 / (min −1 ) and K 2 /[g (mg•min) −1 ] represent the rate constant of the QFO kinetic model and the QSO kinetic model separately.q t / (mg•g −1 ) is the amount of estrogen adsorbed by different materials at time t, q e / (mg•g −1 ) indicating the equilibrium adsorption capacity.α / ( mg•g −1 •h −1 ) and β / (g•mg −1 ) are initial adsorption rate, constants related to surface coverage, respectively.K F /[mg (g•L) −1 ] and n indicate the Freundlich constant and heterogeneity factor, separately; C e / (mg•L −1 ) represents the equilibrium content of estrogen pollutants, and q m / (mg•g −1 ) and K L / (L•mg −1 ) represent the maximal adsorption capacity and Langmuir binding constant, separately.G (kJ/mol), S/J/ (mol•K), H / (kJ/mol), K / (L/g) and T / (K) represent the Gibbs free energy change, the entropy change, the enthalpy change, the thermodynamic equilibrium constant, and the Kelvin temperature, respectively.R is the ideal gas constant, and R = 8.314 J/ (mol•L).

Materials characteristics
Figure 1A, B, D, E exhibit the SEM images of biochar, and nZVI-biochar particles before and after estrogens adsorption.The experimental results presented that the porosity of biochar was obvious and rich, which provided a favorable possibility for the capture of nZVI in the biochar, and the surface morphology and surface area of biochar did not change significantly after usage for estrogens removal, indicating that biochar might have the potential for recycling (He et al. 2014;Zhang et al. 2010aZhang et al. , 2011)).Compared with the unmodified nZVI, nZVI particles got immobilization on the surface or pore of biochar.Obviously, the immobilization of nZVI employing biochar avoided the aggregation of nZVI beneficially to keep their high surface and reactivity.In addition, the aggregation of nZVI disappeared after usage for estrogens removal, indicating that nZVI was involved in the degradation of estrogen pollutants.
The surface element compositions of the biochar and nZVI-biochar were analyzed by EDS spectroscopy.The distribution of elemental C, O, Mg, Si, S, K,Ca was observed from the corresponding SEM-EDS mapping images of biochar (Fig. 1 C), and elemental C and O were the main component elements, mainly from the biochar substrate.Moreover, the peaks of Fe were found from the EDS spectra of nZVI-biochar (Fig. 1F), further confirming the successful attaching of nZVI on biochar.In summary, according to their elemental mapping images, nZVI was evenly distributed on the surface of the biochar, which was in accordance with the results by Karunaratne et al. (2022).
The inside morphology and particle size distribution of the original biochar and nZVI-biochar were studied by transmission electron microscopy (TEM).As shown in Fig. 1G, H, the unique spherical structure of nZVI confirmed its agglomeration tendency.The clear edges and the uneven units of smooth surface of biochar contributed to the loading and aggregation of nZVI.TEM results were consistent with the description of SEM, confirming the successful synthesis of the nZVI-biochar.
The XRD pattern of nZVI-biochar is illustrated in Fig. 1I.The XRD pattern of nZVI-biochar exhibited the apparent peak at 2θ = 44.65°,indicating that nZVI Table 1 Formulas and parameters used in kinetics, isotherm and thermodynamic models

Kinetics
Quasi-first-order q t = q e (1 − e −K1t ) particles were successfully loaded on the surface of biochar (Dai et al. 2016).Furthermore, a weak oxidation peak was detected at 2θ = 34.65°,which was assigned to oxidation of nZVI during preparation, drying and storage (Jiang et al. 2011).
The BET specific surface area and pore volume of the nZVI-biochar composites were calculated to be 756.59m 2 •g −1 and 0.47 cm 3 •g −1 , respectively, much larger than those of biochar (116.87 m 2 •g −1 and 0.19 cm 3 •g −1 ) due to the deposition of magnetic nanoparticles on the surface and the macropore's inner wall of biochar, which exposed that nZVI-biochar had more adsorption capacity than biochar.By increasing BET specific surface area and pore volume, nZVI-biochar may have a greater contact Figure 1L shows the FTIR spectra of nZVI-biochar adsorbents before and after estrogens removal.Obvious changes of FTIR band characteristics occurred in nZVI-biochar after interaction with estrogens.Firstly, the band for O-H groups vanished, emphasizing the importance of hydrogen bonding in estrogen adsorption.According to Jia et al. (2020), the hydroxyl groups could act as hydrogen donors, bonding with oxygen atoms from estrogens.Secondly, the intensity of aromatic C-H out-of-plane vibrations was found to be reduced, which indicated that the estrogens-π interaction occurred during estrogens adsorption (Wei et al. 2020).A stable combination of a π from aromatic C-H with another π from estrogens anions could improve the adsorption onto nZVI-biochar.Thirdly, the stretching vibrations of oxygen-containing groups such as C = O and C-O were also weakened with slight shift in peak position, implying that the oxygen containing functional groups on nZVIbiochar also participated in estrogens adsorption via surface complexation reaction (Zhu et al. 2021).Finally, a band wavenumber at 613 cm −1 of nZVI-biochar attributed to Fe-O transverse vibration, which confirmed the successful introduction of the Fe 0 (Ji et al. 2022).After the reaction, the stretching vibration of Fe-O band was weakened, which suggested that nZVI participated in the reaction of the estrogens on nZVI-biochar.

Effects of different influencing factors
Figure 2 shows that nZVI exhibited a certain estrogens removal rate (only 11.9%, 13.5%, 11.1% of E1, E2, E3 were removed after 60 min), which was attributed to oxidation and aggregation of nZVI (Song et al. 2023).Whereas, 46.7%, 49.1%, 39.6% of E1, E2, E3 were removed by biochar alone, mainly due to the adsorption of estrogens by biochar.In contrast, 79.7%, 83.9%, 63.3% of E1, E2, E3 were quickly removed by nZVI-biochar within 10 min, which were significantly higher than the removal rates of biochar alone.Furthermore, the adsorption of E1, E2, and E3 on different absorbent materials occurred rapidly in the first 10 min, and then slowly before reaching adsorption equilibrium within 60 min.
Biochar and nZVI-biochar composites with larger specific surface area could provide more estrogen absorption sites than nZVI alone.In addition, distribution of nZVI particles on the surface or porous structure of biochar made nZVI aggregation highly unlikely, which could offer more active sites than pure nZVI (Luo et al. 2014).When nZVI-biochar was added into the reaction system, the primary cell could be formed spontaneously, which improved the removal efficiency of estrogen pollutants, and meanwhile, inhibited the oxidation of nZVI (Dou et al. 2010;Lv et al. 2014) and improved the stability of nZVI-biochar.
The mass ratio of iron to carbon (Fe/C ratios) was crucial in the synthesis of nZVI-biochar, because the combination of nZVI and biochar could offset their limitations, maintain mutual stability and nZVI dispersion in the removal of pollutants.Proper Fe/C ratios promoted the rational arrangement pattern of nZVI inside or outside the carbon skeleton, which may play a role in the reactivity of nZVI-biochar.The influences of iron-carbon mass ratio (1:1, 2:1, 3:1) on estrogens removal were studied and the results are exhibited in Fig. 3A.In 60 min, the removal rates of E1, E2, E3 were 11.9%, 13.5%, 11.1% by nZVI and 46.7%, 49.1%, 39.6% by biochar alone.The removal rates of E1, E2, E3 increased with increasing iron-carbon mass ratio (60.5%, 71.2%, 48.8% for 1:1 ironcarbon mass ratio, 79.7%, 83.9%, 63.3% for 2:1 iron-carbon mass ratio respectively), which could be attributed to two main aspects.Firstly, the increased adsorption sites made high availability of the nZVI and estrogens for the reaction, which exhibited better removal performance (Jia et al. 2019).Secondly, the synthesized nZVI-biochar may also acquire substantial galvanic cells with high iron-carbon mass ratio due to the formation of cementite (Wang et al. 2016).However, when the iron-carbon mass ratio was further increased to 3:1, the removal rates of E1, E2, E3 decreased to 74.3%, 78.5%, 58.2%, because nZVI particles distributed on the surface of biochar led to less active sites presented to adsorb estrogen pollutants (Devi and Saroha 2014).Furthermore, a higher dose of nZVI might promote particles agglomeration (Ai et al. 2022), and block reactive sites for contaminants (Zhu et al. 2023) and impede electron transfer, which made a decrease in the removal rate of estrogens.Zhang investigated almost complete removal of CB with nZVI-biochar system, and the mass ratio of 2:1 was conducive to the highest removal rate, indicating that high or low mass ratio of iron to carbon was not suitable for removing CB due to iron's influence on the electron transfer efficiency (Zhang and Wu 2017).Speciation of the composite and surface charge properties of the adsorbent were affected by the pH of initial solution, and estrogen's pKa value (acid dissociation constant) would affect its ionization state under different pH conditions, which influenced the interaction with adsorbent materials, so pH was the most significant parameter in the adsorption process.Estrogens removal by nZVI-biochar (2:1) at different pH values from 4 to 9 is investigated in Fig. 3B.The outcomes exhibited estrogens removal rate by nZVI-biochar presented little difference at pH 4 and 5.However, the removal rate of estrogens decreased as the solution pH increased from 6 to 9, indicating that acidic conditions were conducive to the highest estrogens removal.This phenomenon may be due to the fact that H + could enhance iron oxide corrosion on nZVI surface and suppress nZVI passivation under acidic conditions (Diao et al. 2016), and Fe (OH) 3 passivation layers were easily formed on nZVI surface under alkaline conditions, which prevented the contact between nZVIbiochar and estrogens and the electron transfer channel.Additionally, the formation of primary cells between nZVI and biochar was suffocated (Xie et al. 2017;Lai et al. 2012), resulting in a decreased removal rate of estrogens when the solution pH was greater than 7.
Figure 3C reveals the effects of different initial concentrations of estrogens on the removal rate by 0.1 g•L −1 nZVI-biochar composites at the equilibrium time of 60 min.The outcomes exhibited the capacities of E1, E2 and E3 removal were increased by increasing estrogens concentration; and removal rate dropped from 83.2%, 87.4%, 69% to 67.2%, 72.9%, 54.2% as the initial concentration increased from 500 to 2500 μg•L −1 , respectively.The similar trend was clearly observed in previously reported studies (Lai et al. 2013;Liu et al. 2013;Xu et al. 2020).Collision times increased with the increase of estrogens concentration, which increased removal rate of estrogens (Wan et al. 2017;Jawad et al. 2022cJawad et al. , 2020)).Maintaining the dosage of the nZVI-biochar led to the restricted active adsorption sites on nZVI-biochar, and thereby excessive estrogen was difficult to be completely removed (Jawad et al. 2022c), which meant that the active nZVI-biochar sites to the estrogens pollutants at low concentrations were available.
Temperature held the key to the removal of estrogen pollutants by nZVI-biochar.The removal effects of nZVI-biochar on estrogen pollutants at various temperatures are exhibited in Fig. 3D.Results indicated that 298 K was the optimum temperature for E1, E2, and E3 removal by nZVI-biochar with final removal efficiency of 79.7%, 83.9%, and 63.3%.These results were consistent with previously reported studies, in which the optimum temperature for bisphenol A by NZVI@MAC composites was 298 K (Xu et al. 2022a).Therefore, a cost-effective and eco-friendly material of nZVI-biochar performed excellent potential for practical applications in the wastewater treatment.

Adsorption kinetics
The kinetic studies gave the equilibrium time and described the removal process, which was an important approach for determining the mechanism.On the basis of previous reports (Maamoun et al. 2018(Maamoun et al. , 2021)), the quasi-first-order model (QFO), quasi-second-order (QSO) model, and Elovich model were conducted to test the kinetic data.parameters of different models.Figures 4 and 5 exhibits the relationships of q e versus t, q t versus lnt of estrogens adsorption on biochar, nZVI, and nZVI-biochar, respectively.Compared with the correlation coefficient (R 2 ) of the QFO model, QSO model and Elovich model, the QSO model was more suitable to characterize the experiment data of estrogens adsorption on nZVI-biochar with the higher R 2 and the closer values of q e to experimental data.This meant that the adsorption of estrogens was a chemical adsorption (Blázquez et al. 2011;Hameed 2009).The Elovich model was used to describe chemical sorption occurring on a heterogeneous surface (Diagboya et al. 2015).Meanwhile, the Elovich model with a lower R 2 indicated the physical adsorption might involve in the main mechanism.In addition, nZVI-biochar could improve the removal ability in the reduction system of biochar combined with estrogens considering the positive correlation.Table 2 shows that the removal efficiency was in the order of nZVI-biochar > biochar > nZVI.

Adsorption isotherms
To further investigate the adsorption capacity of biochar, nZVI, and nZVI-biochar, the adsorption isotherms were analyzed using the Langmuir and Freundlich models  3 summarizes the fitting equations and parameters.The Langmuir model was adopted for describing the homogeneous adsorption process with a limited number of identical sites, and each adsorbate molecule features a permanent enthalpy and adsorption activation energy (Foo and Hameed 2010).Freundlich model was broadly adopted for multilayer adsorption on heterogeneous surfaces with nonuniform adsorption enthalpy distribution (Ng et al. 2002).Based on the relatively high correlation coefficient R 2 ranging from 0.993 to 0.999 of Freundlich models for different materials, it appeared that the energy distribution in real solid surfaces was not quite uniform and monolayer (Vimonses et al. 2009).The values of K F meant the maximum adsorption capacities of different materials, with the K F of biochar and nZVI-biochar much higher than that of nZVI, which indicated that nZVI-biochar had the highest removal ability of estrogen, followed by biochar and nZVI.The high removal ability of nZVI-biochar for estrogens was attributed to its high specific surface (Morawski and Inagaki 1997).Furthermore, compared with the unsupported nZVI particles, the stability and activity of nZVI were improved by employing biochar as a carrier.

Adsorption thermodynamics
In spite of the lower correlation coefficient of the thermodynamic analysis, the results still could provide key information.Thermodynamic fitting parameters of three materials for estrogens adsorption are shown in Table 4.The negative values of ΔG at all reaction temperatures confirmed that the reactions between estrogen pollutants and adsorptive material were spontaneous and favorable excluding nZVI (Peng et al. 2018).Furthermore, the ΔH with negative values disclosed that the estrogens adsorption was exothermic, which indicated that high temperatures were not required for efficient removal performance consistent with the previously mentioned temperature effect.In general, The theoretical ranges of enthalpy changes (ΔH) corresponding to hydrogen bonding, electrostatic interactions and van der Waals forces are 2 ~ 40, 20 ~ 80 and 4 ~ 10 kJ/mol, respectively (Machado et al. ).Thus, hydrogen bonding and van der Waals forces were the main mechanisms in the adsorption process.In the same temperature, the value of |ΔG| of nZVI-biochar was always higher than that of biochar and nZVI, indicating that nZVI-biochar was more effective for absorbing estrogens.Combined with kinetic studies the results showed that chemical adsorption and physical adsorption existed simultaneously in the process of estrogens adsorption by three materials.

Adsorption mechanism of estrogens
The mechanism of estrogens removal is primarily based on π-π interaction, hydrophobic interaction, electrostatic interaction and hydrogen bonding among nZVIbiochar and estrogens, and responsible for removal performance (Fig. 8).The higher adsorption capacity of estrogens on nZVI-biochar and biochar confirmed that π-π interaction was of great importance for estrogens removal, which was closely related to the degree of graphitization (Sun et al. 2020).Hydrophobic interaction played an leading role on the adsorption of hydrophobic pollutants (Xie et al. 2020).Hence, the estrogens (2.6 ≤ logK ow ≤ 4) adsorption on nZVI-biochar was strongly influenced by hydrophobic interactions, and it was an effective adsorbent for removing organic pollutants that were hydrophilic.As reported, the hydrogen bonding was weaken with the increase of solution pH (Zhang et al. 2010b).As pH increased, the removal rate of estrogens decreased slightly, indicating that the strong interaction, hydrophobic interaction and weak hydrogen bonding between estrogens and nZVI-biochar were responsible for the high removal performance.Besides, Xu et al. reported that the oxygen-centered PFRs formed during preparation led to the degradation of estrogens (Xu et al. 2022b).Thereby, the degradation products of estrogens by nZVI-biochar were analyzed by GC-MS.
According to the outcomes the degradation of E1, E2 and E3 by nZVI-biochar was mainly through ring rupture and the degradation products were Methyl palmitate, Diisobutyl phthalate, 1-Butoxybutane.

Conclusion
In this work, nZVI-biochar was prepared using walnut shell waste as precursor by liquid-phase reduction approach.The results indicated that the dispersion and pore structural parameters were significantly improved compared with nZVI.The adsorption experiments confirmed that nZVI-biochar exhibited good adsorption performance toward E1, E2, and E3 with the maximum adsorption capacity of 7.97 mg•g −1 , 8.39 mg•g −1 , 6.33 mg•g −1 , respectively.The kinetic fittings suggested that chemical and physical adsorption existed simultaneously on estrogens removal.The isothermal fitting results showed that the adsorption of E1, E2 and E3 occurred as multilayer adsorption on a local non-uniform surface.
Combining the characterization results, the adsorption process for E1, E2 and E3 was controlled by various strong-weak interactions including π-π interaction, hydrophobic interaction, hydrogen bonding.In addition, degradation of E1, E2 and E3 by nZVI-biochar was mainly through ring rupture and the degradation products were Methyl palmitate, Diisobutyl phthalate, and 1-Butoxybutane.This study provided a material with excellent adsorption performance for removal of organic pollutants, which realized resource reuse and was in accord with the goal of carbon neutrality.

Fig. 1
Fig. 1 Materials characteristics.A, B SEM images of biochar before and after adsorption, D, E SEM images of biochar before and after adsorption.C, F SEM-EDS spectra of biochar and nZVI-biochar.G, H TEM images of biochar and nZVI-biochar.I XRD pattern of nZVI-biochar.J BET specific surface area and pore volume.K N 2 adsorption isotherms.L FTIR spectra of nZVI-biochar before and after adsorption

Fig. 3
Fig. 3 Effects of different influencing factors.A removal rate in different iron-carbon ratio, B estrogens removal rates at different pH, C initial content effects on estrogen contaminants removal, D removal rates of estrogen contaminants at different temperatures.The values shown in the figure are means ± SD (n = 3)

Table 2
Table 2 exhibits the calculation Kinetic analysis results of estrogens removal by different materials

Table 3
Parameters of Langmuir and Freundlich isothermal adsorption models

Table 4
Thermodynamics analysis results of estrogens removal by different materials