Assessment of schwertmannite, jarosite and goethite as adsorbents for efficient adsorption of phenanthrene in water and the regeneration of spent adsorbents by heterogeneous fenton-like reaction

Chemosphere 244 (2020) 125523

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Assessment of schwertmannite, jarosite and goethite as adsorbents for efficient adsorption of phenanthrene in water and the regeneration of spent adsorbents by heterogeneous fenton-like reaction Xiaoqing Meng a, Chunmei Zhang a, Jing Zhuang a, Guanyu Zheng a, b, *, Lixiang Zhou a, b a b

Department of Environmental Engineering, College of Resources and Environmental Sciences, Nanjing Agricultural University, Nanjing, 210095, China Jiangsu Collaborative Innovation Center for Solid Organic Waste Resource Utilization, Nanjing, 210095, China

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


Schwertmannite, jarosite and goethite can rapidly adsorb phenanthrene.
The adsorption process was a spontaneous and exothermic process.
Schwertmannite and jarosite had better separation ability than goethite.
The regeneration efficiency of schwertmannite was as high as 97.9 e99.7%.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 October 2019 Received in revised form 26 November 2019 Accepted 29 November 2019 Available online 30 November 2019

Schwertmannite, jarosite or goethite are commonly used to remove metals and/or metalloids from contaminated water via adsorption processes, but it is still unclear whether they can be used as adsorbents to remove hydrophobic organic pollutants (HOCs), such as polycyclic aromatic hydrocarbons (PAHs), from groundwater or wastewater. Here, the feasibility of using these iron (oxyhydr) oxide minerals as adsorbents for phenanthrene (a model PAH) adsorption and regenerating the spent adsorbents via heterogeneous Fenton-like reaction was investigated. Results showed that they exhibited rapid adsorption rates and considerable adsorption capacities for phenanthrene. The maximum Langmuir capacities (Qmax) for phenanthrene adsorption at 28  C were in an ascending order of goethite (567 mg$g-1) < schwertmannite (727 mg$g-1) < jarosite (2088 mg$g-1). The adsorption process was a spontaneous and exothermic process along with the decrease of randomness at the solid/liquid interfaces, which was influenced by temperature, adsorbent dosage, and the coexistence of inorganic anions. Both schwertmannite and jarosite were superior to goethite in light of their easy separation from the bulk solution after the adsorption processes. A multi-cycle experiment demonstrated that the regeneration efficiency of schwertmannite (97.9e99.7%) was much higher than that of jarosite (80.1 e87.2%), and the mineral structure, morphology and functional groups of schwertmannite were not changed during the successive adsorption-regeneration processes. Therefore, among the investigated three iron (oxyhydr) oxide minerals, schwertmannite was an attractive and regenerable adsorbent for the removal of phenanthrene from water owing to its high adsorption capacity, good separation ability, and excellent reusability. © 2019 Elsevier Ltd. All rights reserved.

Handling Editor: Y Yeomin Yoon Keywords: Schwertmannite Jarosite Goethite Phenanthrene Adsorption Regeneration

* Corresponding author. Department of Environmental Engineering, Nanjing Agricultural University, Nanjing, 210095, China. E-mail address: [email protected] (G. Zheng). https://doi.org/10.1016/j.chemosphere.2019.125523 0045-6535/© 2019 Elsevier Ltd. All rights reserved.

2

X. Meng et al. / Chemosphere 244 (2020) 125523

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) are a group of hazardous organic compounds that consist of at least two fused benzene rings. They are highly toxic and carcinogenic to living organisms (Mojiri et al., 2019) and can cause a serious threat to human being health through the food chain (Johnsen and Karlson, 2005; Manoli and Samara, 1999). Due to the increased anthropogenic activities, the PAHs contamination is frequently detected in groundwater or wastewater, where they can persist and induce severe environmental problems (Nizzetto et al., 2008; Zambianchi et al., 2017). Thus, efficient techniques should be developed to tackle the contamination of PAHs in groundwater or wastewater (Hu et al., 2017; Ligaray et al., 2016; Ravindra et al., 2008). A diversity of approaches has been developed to remove or degrade PAHs in groundwater or wastewater, which include physical, chemical, and biological remediation methods (Khairy et al., 2018). Among the various remediation methods, adsorption is one of the most effective and widely used techniques for the removal of PAHs from contaminated groundwater or wastewater. However, the adsorption process only transfers pollutants from aqueous phase onto the surface of solid adsorbents rather than eliminating them. Besides, the spent adsorbents need to be regenerated before reuse to avoid secondary environmental pollution (Braschi et al., 2016). The commonly used methods for regenerating the spent (PAHs-loaded) adsorbents include thermal treatment, chemical solvent extraction, advanced oxidation processes (AOPs), and so on (Nahm et al., 2012; Wang et al., 2017). The thermal treatment usually requires high energy cost and may cause mass loss and/or structural damage of adsorbents (Gao et al., 2016). The chemical solvent extraction would consume large amounts of extraction agents, and the generated solutions that contain high contents of PAHs are hazardous to the environment and need ~ o et al., 2016). It is noteworthy that AOPs further treatment (Patin can degrade or even mineralize the adsorbed PAHs to restore the adsorption ability of the spent adsorbent (Cabrera-Codony et al., 2015). For instance, Gao et al. (2016) showed that heterogeneous photo-Fenton process was efficient to regenerate the spent iron modified bentonite (FeMB) and the regeneration efficiency was as high as 79% after five consecutive adsorption-regeneration cycles. Therefore, it is of great interests to search for adsorbent materials that can effectively adsorb PAHs from water and can be easily regenerated without losing their adsorption properties. Schwertmannite [Fe8O8(OH)8-2x(SO4)x, 1  x  1.75] and jarosite [KFe3(SO4)2(OH)6] are ubiquitous iron oxyhydroxysulphate minerals, which are abundant in acid mine drainage, acid-sulfate soils, or sludge bioleaching environments (Bigham et al., 1996; Regenspurg et al., 2004; Zhu et al., 2013). Goethite (a-FeOOH), a typical iron oxide mineral, occurs widely in natural environments and is the dominant reactive mineral in lake and marine sediments (Chen et al., 2017). Previous studies reported that iron (oxyhydr) oxides, such as schwertmannite, jarosite, and goethite, can be used as adsorbents for remediating groundwater or wastewater contaminated with metals and/or metalloids (e.g., As, Cr, Cu and Pb), due to their relatively wide availability, specific crystal structures, and large surface areas (Mamindy-Pajany et al., 2009; Richmond et al., 2006; Rout et al., 2012). For instance, Liao et al. (2011) used schwertmannite to adsorb As(III) from groundwater and found that the maximum adsorption capacity of schwertmannite for As(III) was as high as 113.9 mg g1. Gan et al. (2015) found that schwertmannite modified by AlPO4 can effectively adsorb Cr(VI) and Cu(II) from contaminated water. To date, it is still unclear whether schwertmannite, jarosite, and goethite can be used as adsorbents to remove hydrophobic organic

pollutants (HOCs), such as PAHs, from groundwater or wastewater. Although Müller et al. (2007) proposed that pyrene may be adsorbed on goethite-coated quartz via cation-p bonding and Meng et al. (2017) observed the adsorption of phenanthrene on schwertmannite, the adsorption behavior of PAHs on schwertmannite, jarosite, and goethite still need to be further clarified. Recent studies also revealed that schwertmanite, jaosite and goethite can be used as catalysts to catalyze the heterogeneous Fenton-like reaction to degrade or mineralize organic pollutants including phenanthrene, phenol and sulfamethazine (Meng et al., 2017; Wang et al., 2013; Yan et al., 2017; Zhou et al., 2013). Although this provides the possibility of regenerating spent adsorbents through heterogeneous Fenton-like reactions catalyzed by them, it is still necessary to find out how the regeneration processes based on heterogeneous Fenton-like reactions impact their adsorption abilities for PAHs. Therefore, the objectives of the present study are to (1) study the adsorption behavior of phenanthrene as a model compound of PAHs on schwertmannite, jarosite and goethite; (2) investigate the influences of adsorbent dosage and the coexistence of inorganic anions on the adsorption abilities of schwertmannite, jarosite and goethite for phenanthrene; and (3) explore the ease of separating spent adsorbents from adsorption solutions and the adsorbent stabilities during the successive adsorption-regeneration processes. This study will be useful for extending the application of iron (oxyhydr) oxide minerals as low-cost and regenerable adsorbents to remediate PAHs-contaminated water. 2. Materials and methods 2.1. Reagents Phenanthrene (purity  99%) was purchased from Aladdin Co., Ltd. Acetonitrile was purchased from Tedia Co., Ltd at HPLC grade. Dichloromethane and other reagents were supplied by Sinopharm Chemical Regent Beijing Co., Ltd at analytical grade and used without further purification. All solutions were prepared using deionized water. 2.2. Preparation and characterization of schwertmannite, jarosite and goethite In the present study, schwertmannite and jarosite were biosynthesized through oxidizing ferrous sulfate by Acidithiobacillus ferrooxidans LX5 cell suspensions (Liao et al., 2011; Liu et al., 2015), because of the high synthesis yields and the good catalytic activities of biosynthetic schwertmannite and jarosite in Fenton-like reactions (Meng et al., 2017; Yan et al., 2017). Goethite was synthesized through mixing 1 M Fe(NO3)3 with 5 M KOH at 70  C for 60 h (Manning et al., 1998), since the biosynthesis of goethite is very time-consuming and inefficient (Jiang et al., 2013). The detailed procedures can be found in Text S1. The properties of schwertmannite, jarosite and goethite were characterized by powder X-ray diffraction (XRD, ThermoFisher X’TRA), scanning electron microscope (SEM, Hitachi S-3400 N) and flourier transform infrared spectroscopy (FTIR, NEXUS870). The detailed characterization methods are shown in Text S2. 2.3. Phenanthrene adsorption experiments To avoid the interference of solvent (i.e. methanol or acetone) on the adsorption of phenanthrene by schwertmannite, jarosite, or goethite, the stock solution of phenanthrene was prepared with deionized water. First, 20 mg phenanthrene powder was added to 1 L of deionized water. The mixture was stirred using magnetic

X. Meng et al. / Chemosphere 244 (2020) 125523

stirrers for 1 h at 80  C and then filtered through a glass column packed with glass wool to remove any undissolved phenanthrene particles. The obtained filtrate was used as the stock phenanthrene solution for adsorption experiments. The phenanthrene concentration in this stock solution was quantified using high performance liquid chromatography (HPLC), and it was 614.7 mg L1. Batch adsorption experiments were performed in 50-mL glass centrifuge tubes with Teflon-coated screw caps under dark conditions. A certain amount of schwertmannite, jarosite or goethite was introduced to glass centrifuge tubes, each containing 15 mL of phenanthrene stock solution. The mixtures were then shaken at 250 rpm and 28  C in a rotary shaker. At certain time intervals, samples were sacrificially collected, centrifuged for 5 min at 3000 g, and then filtered through 0.22 mm glass fiber filter paper. The residual phenanthrene in the filtrate was extracted using dichloromethane (Manoli and Samara, 1999; Meng et al., 2017), and then its amount was determined using HPLC. In the adsorption kinetics experiments, the adsorption of phenanthrene on the tested iron (oxyhydr) oxides was investigated by varying the contact time from 0 to 120 min at 28  C. The adsorption thermodynamic experiments were carried out at 18, 28 and 38  C, and the initial phenanthrene concentration for the adsorption isotherm experiments ranged from 0 to 614.7 mg L1. The influences of adsorbent dosage and coexistence of inorganic anions on the adsorption of phenanthrene were investigated by changing the dosages of adsorbents in a range  of 0.1e2 g L1 and the respective concentration of Cl, SO2 4 , NO3  and H2PO4 in a range of 100e300 mg$L-1, respectively. The ease of separating phenanthrene-loaded adsorbents from adsorption solutions was assessed through a settlement experiment. First, 0.1 g of schwertmannite, jarosite, or goethite was introduced to 100 mL of stock solution with 614.7 mg L1 phenanthrene. After the complement of adsorption process, solutions were maintained stationary for 24 h to let the adsorbents settle. The separation of adsorbents from the solution was visualized at given time intervals, and meanwhile the transmittance of supernatant was measured by UVevis spectroscopy at 600 nm. The efficiency of regenerating phenanthrene-loaded schwertmannite or jarosite, the reusability of adsorbents, and the adsorbent stability were investigated through a multi-cycle experiment, with each cycle consisting of an adsorption process and a regeneration process. During the regeneration process, the spent adsorbents were used as catalysts to catalyze heterogeneous Fenton-like reaction to degrade the adsorbed phenanthrene (Meng et al., 2017). In each cycle, 0.1 g of schwertmannite or jarosite was used to adsorb phenanthrene from 100 mL of stock solution with 614.7 mg L1 phenanthrene. After the completion of adsorption process, residual phenanthrene in the adsorption solution was determined to calculate the adsorption efficiency, while the spent adsorbents were collected and placed in another 50 mL glass tube containing 10 mL of deionized water. The pH value of the mixture containing spent adsorbents was adjusted to 3.0 by using 0.1 M H2SO4 (Meng et al., 2017; Yan et al., 2017), and then 30 mL H2O2 solution (30%, v/v) was added to induce heterogeneous Fenton-like reaction (Meng et al., 2017). After 5 h of reaction, the adsorbents were collected. A part of collected adsorbents was used to determine the amount of residual adsorbed PAHs to calculate the degradation (regeneration) efficiency, and the other adsorbents were washed with deionized water for three times, dried enough, and subsequently used in the next cycle. In the present study, all experiments were conducted in triplicate and data presented are the mean values of the triplicate samples with standard deviations. 2.4. Analytical methods The concentration of phenanthrene was analyzed using a HPLC

3

(HPLC, Agilent-1260) equipped with UV254 detection. The detailed analytical methods are provided in Text S3. 3. Results and discussion 3.1. Characterizations of prepared schwertmannite, jarosite and goethite The XRD patterns of prepared iron (oxyhydr) oxide minerals are shown in Fig. S1. The XRD pattern of prepared schwertmannite displayed a weak crystalline structure with seven typical broad characteristic peaks (2q: 18.24, 26.27, 35.16, 39.49, 46.53, 55.29, 61.34 ), and all diffraction peaks well matched with those of the standard diffraction data for schwertmannite (PDF#47e1775) (Fig. S1a). The XRD pattern of prepared jarosite exhibited a good crystalline structure with strong and sharp diffraction peaks at angle 2q of 14.9, 15.42, 17.4, 24.36, 28.68, 28.96, 30.06, 31.14, 35.2, 39.28, 45.84, 46.76, 47.48, 49.98, 60.06, and 62.54 , and all these peaks matched with those of the standard diffraction data for jarosite (PDF#10e0443) (Fig. S1b). The XRD pattern of prepared goethite also matched well with those of the standard diffraction data for goethite (PDF#29e0713). The results suggested that the prepared iron (oxyhydr) oxides were pure schwertmannite, jarosite and goethite, respectively. It can be seen from the SEM images that the morphologies of schwertmannite (Fig. 1a) and jarosite (Fig. 1b) were agglomerate spherical particles with “hedge-hog” and smooth surfaces, respectively, while goethite crystals (Fig. 1c) were acicular and elongated along the crystallographic c-axis direction (Schwertmann and Cornell, 2008). The BET analysis revealed that the specific surface areas of prepared schwertmannite, jarosite and goethite were 54.10, 1.47 and 44.52 m2 g1, respectively. 3.2. Adsorption kinetics Adsorption kinetics of phenanthrene on schwertmannite, jarosite or goethite is illustrated in Fig. 2. It can be seen from the kinetics curves that schwertmannite, jarosite and goethite had similar adsorption behaviors for phenanthrene and all of these adsorption processes were controlled by two stages. In general, phenanthrene was rapidly adsorbed within the initial 15 min, and then the adsorption processes continued at much lower rates. Apparent adsorption equilibrium of phenanthrene on schwertmannite, jarosite or goethite was achieved within 30 min of contact time, and their equilibrium adsorption capacities were 600, 528 and 502 mg g1, respectively. Obviously, their equilibrium adsorption capacities did not correlate with their respective specific surface areas, indicating that the adsorption of phenanthrene on these iron (oxyhydr) oxide minerals was not solely controlled by their n, 2016; Mansouri et al., 2017). specific surface areas (Marba Adsorption kinetics models including the pseudo-first-order model, pseudo-second-order model and the intraparticle diffusion model (Text S4) were usually employed to analyze the kinetics behavior and mechanisms during adsorption processes (Ho et al., 2000; Cao et al., 2014). The adsorption data of phenanthrene on these three iron (oxyhydr) oxide minerals were found to fit well with the pseudo-second-order model, and the respective correlation coefficients (R2) were 0.9995 for schwertmannite, 0.9996 for jarosite, and 0.9997 for goethite. However, the pseudo-first-order model and the intraparticle diffusion model performed worst for all of the adsorbents (Table S1). These results suggested that the chemical adsorption occurred during the adsorption of phenanthrene by schwertmannite, jarosite and goethite, which may control the uptake rate of phenanthrene (Ho and McKay, 1999). Furthermore, the pseudo-second-order rate constants (K2) of

4

X. Meng et al. / Chemosphere 244 (2020) 125523

Fig. 1. SEM images of (a) schwertmannite, (b) jarosite, (c) goethite, (d) schwertmannite repeatedly used for five successive adsorption-regeneration cycles, and (e) jarosite repeatedly used for five successive adsorption-regeneration cycles.

molecular sieve MCM-41 (0.06 g mg1$min1) (Hu et al., 2014). The high K2 values demonstrated that these three iron (oxyhydr) oxide minerals possess abundant binding sites to adsorb phenanthrene. Thus, schwertmannite, jarosite and goethite exhibited rapid adsorption rates and considerable adsorption capacities for phenanthrene. 3.3. Adsorption isotherms and thermodynamics

Fig. 2. Adsorption kinetics of phenanthrene on schwertmannite, jarosite and goethite. (Experimental conditions: initial phenanthrene ¼ 614.7 mg L1, adsorbent dosage ¼ 1 g L1, 28  C).

phenanthrene adsorption on schwertmannite, jarosite and goethite were 0.67, 0.37, and 0.35 g mg1$min1, respectively. The K2 value of phenanthrene adsorption on schwertmannite was nearly twice of that of goethite, suggesting that the adsorption of phenanthrene on schwertmannite was much faster than that of goethite. Besides, the K2 values of phenanthrene adsorption on these three iron (oxyhydr) oxide minerals were very close to or much higher than that of other reported adsorbents, such as fibric peat (0.36e1.40 g mg1$min1) (Tang et al., 2010) and mesoporous

Equilibrium adsorption isotherms are usually used to determine the capacities of adsorbents, adsorbent surface properties, and the relationship between adsorbates and adsorbents (Lamichhane et al., 2016). The adsorption isotherms of phenanthrene by schwertmannite, jarosite and goethite are shown in Fig. S2, and jarosite offered the much higher adsorption capacity for phenanthrene than schwertmannite and goethite. The parameters of the Linear, Langmuir and Freundlich models (Text S5) are presented in Table 1. The adsorption isotherm data for schwertmannite well fitted with the Langmuir model and the correlation coefficients (R2) were 0.991e0.998 at 18e38  C, implying the monolayer adsorption of phenanthrene on the surface of schwertmannite. However, the adsorption isotherm data for jarosite and goethite well fitted with all three models and the respective correlation coefficients (R2) showed no apparent differences among these isotherm models. This result suggests that the adsorption of phenanthrene by jarosite or goethite may be a partitioning process into the region near to the mineral surfaces as well as a sorption process to mineral surface domains with different free energies. This result is in agreement with the results of Müller et al. (2007), who investigated the adsorption of phenanthrene, pyrene and benzo(a)pyrene by goethite-coated quartz and proposed that pyrene may be adsorbed on goethite-coated quartz via cation-p bonding. The maximum Langmuir capacities (Qmax) for phenanthrene adsorption on these three iron (oxyhydr) oxide minerals at each

X. Meng et al. / Chemosphere 244 (2020) 125523

5

Table 1 Adsorption equilibrium parameters for the adsorption of phenanthrene by schwertmannite, jarosite and goethite. Adsorbents

Schwertmannite

Jarosite

Goethite

T ( C)

18 28 38 18 28 38 18 28 38

Linear isotherm

Langmuir isotherm

Freundlich isotherm

Kd (L$g1)

R2

Qmax (mg$g1)

KL (L$mg1)

R2

KF (g$mg1)

n

R2

7.775 4.801 3.722 2.376 2.315 1.873 6.926 5.388 3.350

0.970 0.939 0.917 0.961 0.931 0.970 0.986 0.975 0.984

1128 727 586 2931 2088 1392 636 567 511

0.021 0.018 0.016 0.008 0.011 0.014 0.026 0.023 0.019

0.997 0.998 0.991 0.993 0.979 0.980 0.975 0.982 0.950

38.55 30.78 26.07 2.64 1.75 1.61 23.54 21.01 17.32

1.43 1.67 1.76 0.94 0.97 1.01 1.41 1.45 1.50

0.957 0.982 0.941 0.975 0.949 0.965 0.981 0.977 0.975

temperature were in the following ascending order: goethite < schwertmannite < jarosite (Table 1). At 28  C, the Qmax values for phenanthrene adsorption on schwertmannite, jarosite and goethite obtained in the present study were 567e2088 mg g1, relatively higher than those of other geosorbents including organoclays (around 620 mg g1) (Lee et al., 2004), bentonite (around 210 mg g1) (Changchaivong and Khaodhiar, 2009) and palygorskite (around 2.50 mg g1) (Zhao et al., 2016). In addition, Liu et al. (2016a, 2016b) reported that the maximum Langmuir capacity for phenanthrene adsorption on activated carbon was 1.1 mg g1, which was slightly higher than that of schwertmannite but was much lower than that of jarosite. The adsorption of phenanthrene by each of these three minerals decreased when increasing the temperature from 18 to 38  C, indicating that the adsorption of phenanthrene by these three iron (oxyhydr) oxides was favored at lower temperatures. In order to evaluate the temperature effect on the adsorption of phenanthrene on schwertmannite, jarosite and goethite, thermodynamic parameters including Gibbs free energy (DG, kJ$mol1), enthalpy (DH, kJ$mol1), and entropy (DS, J$mol1$K1) were calculated (Text S6). The calculated DG, DH and DS for the adsorption of phenanthrene on these three iron (oxyhydr) oxide minerals are presented in Table S2. The negative values of DG at 18, 28, and 38  C suggested that the adsorption process was feasible and spontaneous. The decrease in the magnitude of DG at higher temperatures implied the diminishing of the spontaneous of the process, and thus the adsorption of phenanthrene by these three iron (oxyhydr) oxide minerals was not favorable at higher temperatures. The negative DH values indicated the adsorption of phenanthrene on schwertmannite, jarosite or goethite was an € € exothermic process at the tested temperatures (Ozcan and Ozcan, 2005; Senturk et al., 2009). The DS values were 78.77 J mol1$K1 for schwertmannite, 23.06 J mol1$K1 for jarosite, and 77.15 J mol1$K1 for goethite, demonstrating the decreased randomness at the solid/liquid interfaces during the adsorption of phenanthrene by these minerals (Hu et al., 2014; € € Ozcan and Ozcan, 2005). 3.4. Effect of adsorbent dosage and coexistence of inorganic anions on the adsorption of phenanthrene by schwertmannite, jarosite and goethite The adsorbent dosage is an important factor influencing the adsorption efficiency, and thus the effect of the dosage of schwertmannite, jarosite or goethite on the adsorption of phenanthrene was investigated. As shown in Fig. 3, the equilibrium adsorption capacity for phenanthrene was remarkably enhanced when increasing jarosite dosage from 0 to 0.2 g L1 or schwertmannite/goethite dosage from 0 to 1.0 g L1, most probably due to the increased amounts of available binding sites for phenanthrene

Fig. 3. Effect of adsorbent dosage on equilibrium adsorption of phenanthrene by schwertmannite, jarosite and goethite. (Experimental conditions: initial phenanthrene ¼ 614.7 mg L1, adsorbent dosage ¼ 1 g L1, 28  C).

adsorption (Ibrahim et al., 2010). However, it is noteworthy that excessive adsorbent dosages can hardly further increase their equilibrium adsorption capacities for phenanthrene owing to the unsaturated adsorption sites of excessive adsorbents (Hu et al., 2014; Zhu et al., 2016). Groundwater and wastewater commonly contain many inor  ganic anions, such as Cl, SO2 4 , NO3 and H2PO4 , which may affect the adsorption of HOCs through competing for adsorption sites (Liu et al., 2016a, 2016b; Zhang et al., 2017a, 2017b). It can be seen from   Fig. 4 that Cl, SO2 4 , NO3 and H2PO4 showed distinct potential in reducing the adsorption of phenanthrene by schwertmannite, jarosite and goethite, and their inhibitory effects were enhanced along with the increase of their respective concentrations from 100 to 300 mg L1. For instance, the equilibrium adsorption capacity of schwertmannite for phenanthrene dropped by 7.5%, 20.1%, 29.1% and 38.5%, respectively, when the respective concentrations of Cl, 2  1 NO 3 , SO4 and H2PO4 increased from 0 to 300 mg L . The inhibitory effect of inorganic anions on the adsorption of phenantherene by these three iron (oxyhydr) oxide minerals was in a sequence of    SO2 4 z H2PO4 > NO3 > Cl . This observation may be due to the competition of these inorganic anions with phenantherene at the adsorption site on the surface of iron (oxyhydr) oxides through forming inorganic anions-ferric complexes (Deliyanni et al., 2007; Gan et al., 2015; Kuang et al., 2017; Lee et al., 2004), which may result in less available binding sites for phenanthrene.

6

X. Meng et al. / Chemosphere 244 (2020) 125523

3.5. Regeneration of spent adsorbents by heterogeneous fenton-like reactions

  Fig. 4. Effect of the coexistence of Cl, SO2 4 , NO3 or H2PO4 on equilibrium adsorption of phenanthrene by (a) schwertmannite, (b) jarosite and (c) goethite. (Experimental conditions: initial phenanthrene ¼ 614.7 mg L1, adsorbent dosage ¼ 1 g L1, 28  C).

The ease of separating spent adsorbents from bulk solutions is very crucial for the application of the adsorbents (Kuang et al., 2017), and thus a settlement experiment was carried out to evaluate the possibility of separating schwertmannite, jarosite and goethite from the adsorption systems. As shown in Fig. S3, schwertmannite and jarosite in the suspensions settled within only 2 h, while no obvious settlement of goethite was observed even after 24 h. This result was consistent with the supernatant transmittances measured by UVevis spectroscopy at 600 nm (Fig. S4). Within 60 min of settlement, the supernatant transmittance decreased from 0.840 to 0.008 a.u. for schwertmannite system and from 0.672 to 0.005 a.u. for jarosite system, but the supernatant transmittance of goethite system was still as high as 2.696 a.u. even after 24 h of settlement. Indeed, previous study reported that FeOOH nanoparticles in suspension formed fine colloid, thus being difficult to separate and collect (Kuang et al., 2017). Therefore, either schwertmannite or jarosite is superior to goethite in light of the easy separation from the bulk solution after the adsorption processes. A multi-cycle experiment using schwertmannite or jarosite as an adsorbent for phenanthrene adsorption and regenerating the spent adsorbents via heterogeneous Fenton-like reaction was carried out, and the degradation efficiency of the adsorbed phenanthrene and the stabilities of schwertmannite and jarosite in the successive adsorption-regeneration processes were assessed. As shown in Fig. 5a, the adsorption efficiency of phenanthrene on schwertmannite and jarosite was 86.1e88.6% and 81.3e85.2%, respectively, during five consecutive adsorption-regeneration processes. The degradation efficiency of phenanthrene adsorbed on the spent schwertmannite was as high as 97.9e99.7% (Fig. 5b), demonstrating that the spent schwertmannite can be effectively regenerated by the heterogeneous Fenton-like reaction catalyzed by it. However, only 80.1e87.2% of phenanthrene adsorbed on the spent jarosite can be effectively degraded by the heterogeneous Fenton-like reaction catalyzed by jarosite, probably due to the much lower catalytic activity of jarosite in the heterogeneous Fenton-like reaction. In fact, our previous study has revealed that schwertmannite exhibited a much higher catalytic capacity than jarosite in heterogeneous Fenton-like reactions, which mainly resulted from the easier activation of schwertmannite by H2O2 and the much more rapid transfer of iron from the surface of schwertmannite into solution during the reactions (Yan et al., 2017). In addition, it is noteworthy that the system volume of the regeneration process can be further reduced once the pH value of the reaction system was adjusted to 3.0e4.5 and the added amount of H2O2 is maintained as that of the present study. Therefore, the adsorption of phenanthrene by schwertmannite and the regeneration of spent schwertmannite via heterogeneous Fenton-like reaction would be a promising approach for the removal of phenanthrene from groundwater or wastewater. XRD patterns of schwertmannite (Fig. S1d) and jarosite (Fig. S1e) that were repeatedly used for five cycles still matched the standard diffraction data for schwertmannite (PDF#No.47e1775) or jarosite (PDF#10e0443). Additionally, the SEM images of schwertmannite (Fig. 1d) or jarosite (Fig. 1e) being repeatedly used for five successive adsorption-regeneration cycles showed that the morphologies of schwertmannite was not changed by the repeated uses, and the surface of jarosite became much smoother. Obviously, the mineral structures of both schwertmannite and jarosite were not drastically changed during the successive adsorption-regeneration processes. As shown in Fig. 6, FTIR spectrums of schwertmmnite and jarosite that were repeatedly used for five cycles were still in agreement

X. Meng et al. / Chemosphere 244 (2020) 125523

Fig. 5. (a) Adsorption efficiency of phenanthrene and (b) the regeneration efficiency of spent adsorbents during five successive adsorption-regeneration processes. (Experimental conditions: initial phenanthrene ¼ 614.7 mg L1, adsorbent dosage ¼ 1 g L1, 28  C).

with those of pristine minerals. However, it is noteworthy that a new absorption peak at 1440 cm-1 was observed after the adsorption of phenanthrene by either schwertmannite or jarosite, which represents the stretching vibration of C-C in benzene rings (Zhang et al., 2017a, 2017b). Following the adsorption of phenanthrene, the intensity of eOH stretching vibration in these two minerals slightly decreased, suggesting that p-p interaction may be the major mechanism responsible for the adsorption of phenan~ o et al., threne on the surfaces of schwertmannite and jarosite (Patin 2016; Tang et al., 2015). After five successive adsorptionregeneration processes, the absorption peak at 1440 cm-1 was not observed for schwertmannite, most probably because of the complete mineralization of phenanthrene by heterogeneous Fenton-like reactions catalyzed by schwertmannite (Meng et al., 2017). However, the absorption peak at 1440 cm-1 can still be observed for jarosite due to the incomplete degradation of phenanthrene adsorbed on jarosite. These results clearly elucidate that

7

Fig. 6. FTIR spectra of (a) schwertmannite and (b) jarosite: red line, pristine schwertmannite or jarosite; green line, schwertmannite or jarosite being used for phenanthrene adsorption; blue line, schwertmannite or jarosite being repeatedly used for five successive adsorption-regeneration processes. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

schwertmannite is an excellent adsorbent for the adsorption of phenanthrene, and its mineral structure, morphology and functional groups were not changed during the successive adsorptionregeneration processes consisting of the adsorption of phenanthrene and the heterogeneous Fenton-like regeneration. 4. Conclusions The feasibility of using iron (oxyhydr) oxide minerals including schwertmannite, jarosite and goethite as adsorbents for phenanthrene adsorption and regenerating the spent adsorbents via heterogeneous Fenton-like reaction was investigated. Schwertmannite, jarosite, and goethite exhibited rapid adsorption rates and considerable adsorption capacities for phenanthrene, and

8

X. Meng et al. / Chemosphere 244 (2020) 125523

the adsorption process was a spontaneous and exothermic process along with the decrease of randomness at the solid/liquid interfaces. The efficiency of phenanthrene adsorption on these three iron (oxyhydr) oxide minerals was influenced by temperature, adsorbent dosage, and the coexistence of inorganic anions. A settlement experiment revealed both schwertmannite and jarosite were superior to goethite in light of their easy separation from the bulk solution after the adsorption processes. The stability and reusability/regenerability of adsorbent during successive adsorption-regeneration processes demonstrated that the regeneration efficiency of schwertmannite was much higher than that of jarosite, and the mineral structure, morphology and functional groups of schwertmannite were not changed during the repeated uses. Therefore, among the investigated iron (oxyhydr) oxide minerals, schwertmannite was an attractive and regenerable adsorbent for the removal of HOCs from water due to its high adsorption capacity, good separation ability, and excellent reusability. Author contributions Xiaoqing Meng: Conceptualization, Methodology, Formal analysis, Investigation, Writing - original draft, Visualization. Chunmei Zhang: Methodology, Formal analysis, Investigation. Jing Zhuang: Methodology. Guanyu Zheng: Conceptualization, Validation, Formal analysis, Writing - review & editing, Visualization, Supervision. Lixiang Zhou: Conceptualization, Supervision. Acknowledgements This research was financially supported by the National Key Research and Development Program of China (2017YFD0801000) and the National Natural Science Foundation of China (21477055, 21637003). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.chemosphere.2019.125523. References Bigham, J.M., Schwertmann, U., Traina, S.J., et al., 1996. Schwertmannite and the chemical modeling of iron in acid sulfate waters. Geochem. Cosmochim. Acta 60 (12), 2111e2121. Braschi, I., Blasioli, S., Buscaroli, E., et al., 2016. Physicochemical regeneration of high silica zeolite Y used to clean-up water polluted with sulfonamide antibiotics. J. Environ. Sci. 43, 302e312. Cabrera-Codony, A., Gonzalez-Olmos, R., Martín, M.J., 2015. Regeneration of siloxane-exhausted activated carbon by advanced oxidation processes. J. Hazard Mater. 285, 501e508. Cao, J.S., Lin, J.X., Fang, F., et al., 2014. A new absorbent by modifying walnut shell for the removal of anionic dye: kinetic and thermodynamic studies. Bioresour. Technol. 163, 199e205. Changchaivong, S., Khaodhiar, S., 2009. Adsorption of naphthalene and phenanthrene on dodecylpyridinium-modified bentonite. Appl. Clay Sci. 43 (3e4), 317e321. Chen, Y., Bylaska, E.J., Weare, J.H., 2017. Weakly bound water structure, bond valence saturation and water dynamics at the goethite (100) surface/aqueous interface: ab initio dynamical simulations. Geochem. Trans. 18 (1), 3. Deliyanni, E.A., Peleka, E.N., Lazaridis, N.K., 2007. Comparative study of phosphates ite and hybrid removal from aqueous solutions by nanocrystalline akagane ite. Separ. Purif. Technol. 52 (3), 478e486. surfactant-akagane Gan, M., Sun, S., Zheng, Z., et al., 2015. Adsorption of Cr (VI) and Cu (II) by AlPO4 modified biosynthetic schwertmannite. Appl. Surf. Sci. 356, 986e997. Gao, Y., Guo, Y., Zhang, H., 2016. Iron modified bentonite: enhanced adsorption performance for organic pollutant and its regeneration by heterogeneous visible light photo-Fenton process at circumneutral pH. J. Hazard Mater. 302, 105e113. Ho, Y.S., McKay, G., 1999. Pseudo-second order model for sorption processes. Process Biochem. 34 (5), 451e465. Ho, Y.S., Ng, J.C.Y., McKay, G., 2000. Kinetics of pollutant sorption by biosorbents.

Separ. Purif. Methods 29 (2), 189e232. Hu, T., Zhang, J., Ye, C., et al., 2017. Status, source and health risk assessment of polycyclic aromatic hydrocarbons (PAHs) in soil from the water-levelfluctuation zone of the Three Gorges Reservoir, China. J. Geochem. Explor. 172, 20e28. Hu, Y., He, Y., Wang, X., et al., 2014. Efficient adsorption of phenanthrene by simply synthesized hydrophobic MCM-41 molecular sieves. Appl. Surf. Sci. 311, 825e830. Ibrahim, S., Wang, S., Ang, H.M., 2010. Removal of emulsified oil from oily wastewater using agricultural waste barley straw. Biochem. Eng. J. 49 (1), 78e83. Jiang, S., Kim, M.G., Kim, I.Y., et al., 2013. Biological synthesis of free-standing uniformed goethite nanowires by shewanella sp. HN-41. J. Mater. Chem. 1 (5), 1646e1650. Johnsen, A.R., Karlson, U., 2005. PAH degradation capacity of soil microbial communitiesddoes it depend on PAH exposure. Microb. Ecol. 50 (4), 488e495. Khairy, M., Ayoub, H.A., Rashwan, F.A., et al., 2018. Chemical modification of commercial kaolin for mitigation of organic pollutants in environment via adsorption and generation of inorganic pesticides. Appl. Clay Sci. 153, 124e133. Kuang, L., Liu, Y., Fu, D., et al., 2017. FeOOH-graphene oxide nanocomposites for fluoride removal from water: acetate mediated nano FeOOH growth and adsorption mechanism. J. Colloid Interface Sci. 490, 259e269. Lamichhane, S., Krishna, K.C.B., Sarukkalige, R., 2016. Polycyclic aromatic hydrocarbons (PAHs) removal by sorption: a review. Chemosphere 148, 336e353. Lee, S.Y., Kim, S.J., Chung, S.Y., et al., 2004. Sorption of hydrophobic organic compounds onto organoclays. Chemosphere 55 (5), 781e785. Liao, Y., Liang, J., Zhou, L., 2011. Adsorptive removal of As(III) by biogenic schwertmannite from simulated As-contaminated groundwater. Chemosphere 83 (3), 295e301. Ligaray, M., Baek, S.S., Kwon, H.O., et al., 2016. Watershed-scale modeling on the fate and transport of polycyclic aromatic hydrocarbons (PAHs). J. Hazard Mater. 320, 442e457. Liu, F., Zhou, J., Zhang, S., et al., 2015. Schwertmannite synthesis through ferrous ion chemical oxidation under different H2O2 supply rates and its removal efficiency for arsenic from contaminated groundwater. PLoS One 10 (9), e0138891. Liu, W., Cai, Z., Zhao, X., et al., 2016a. A high-capacity and photo-regenerable composite material for efficient adsorption and degradation of phenanthrene in water. Environ. Sci. Technol. 50 (20). Liu, Y., Zhou, A., Gan, Y., et al., 2016b. Effects of inorganic anions on carbon isotope fractionation during Fenton-like degradation of trichloroethene. J. Hazard Mater. 308, 187e191. Mamindy-Pajany, Y., Hurel, C., Marmier, N., et al., 2009. Arsenic adsorption onto hematite and goethite. Compt. Rendus Chem. 12 (8), 876e881. Manning, B.A., Fendorf, S.E., Goldberg, S., 1998. Surface structures and stability of arsenic (III) on goethite: spectroscopic evidence for inner-sphere complexes. Environ. Sci. Technol. 32 (16), 2383e2388. Manoli, E., Samara, C., 1999. Polycyclic aromatic hydrocarbons in natural waters: sources, occurrence and analysis. Trac. Trends Anal. Chem. 18 (6), 417e428. Mansouri, M., Nademi, M., Olya, M.E., 2017. Study of methyl tert-butyl ether (MTBE) photocatalytic degradation with UV/TiO2-ZnO-CuO nanoparticles. Int. J. Pharm. 7 (1), 19e32. n, Gregorio, 2016. BET adsorption reaction model based on the pseudo Marba steady-state hypothesis for describing the kinetics of adsorption in liquid phase. J. Colloid Interface Sci. 467, 170e179. Meng, X., Yan, S., Wu, W., et al., 2017. Heterogeneous Fenton-like degradation of phenanthrene catalyzed by schwertmannite biosynthesized using Acidithiobacillus ferrooxidans. RSC Adv. 7 (35), 21638e21648. Mojiri, A., Zhou, J.L., Ohashi, A., et al., 2019. Comprehensive review of polycyclic aromatic hydrocarbons in water sources, their effects and treatments. Sci. Total Environ. 133971. €gel-Knabner, I., 2007. Sorption of polycyclic aromatic Müller, S., Totsche, K.U., Ko hydrocarbons to mineral surfaces. Eur. J. Soil Sci. 58 (4), 918e931. Nahm, S.W., Shim, W.G., Park, Y.K., et al., 2012. Thermal and chemical regeneration of spent activated carbon and its adsorption property for toluene. Chem. Eng. J. 210, 500e509. Nizzetto, L., Lohmann, R., Gioia, R., et al., 2008. PAHs in air and seawater along a NortheSouth Atlantic transect: trends, processes and possible sources. Environ. Sci. Technol. 42 (5), 1580e1585. € € Ozcan, A., Ozcan, A.S., 2005. Adsorption of Acid Red 57 from aqueous solutions onto surfactant-modified sepiolite. J. Hazard Mater. 125 (1e3), 252e259. ~ o, Y., Díaz, E., Ordo n ~ ez, S., 2016. Pre-concentration of nalidixic acid through Patin adsorptionedesorption cycles: adsorbent selection and modeling. Chem. Eng. J. 283, 486e494. Ravindra, K., Sokhi, R., Van Grieken, R., 2008. Atmospheric polycyclic aromatic hydrocarbons: source attribution, emission factors and regulation. Atmos. Environ. 42 (13), 2895e2921. Regenspurg, S., Brand, A., Peiffer, S., 2004. Formation and stability of schwertmannite in acidic mining lakes. Geochem. Cosmochim. Acta 68 (6), 1185e1197. Richmond, W.R., Cowley, J.M., Parkinson, G.M., et al., 2006. An electron microscopy ite) nanorods and nanotubes. CrystEngComm 8 (1), study of b-FeOOH (akagane 36e40. Rout, K., Mohapatra, M., Anand, S., 2012. 2-Line ferrihydrite: synthesis, characterization and its adsorption behaviour for removal of Pb (II), Cd (II), Cu (II) and Zn (II) from aqueous solutions. Dalton Trans. 41 (11), 3302e3312. Schwertmann, U., Cornell, R.M., 2008. Iron Oxides in the Laboratory: Preparation and Characterization. John Wiley & Sons.

X. Meng et al. / Chemosphere 244 (2020) 125523 Senturk, H.B., Ozdes, D., Gundogdu, A., et al., 2009. Removal of phenol from aqueous solutions by adsorption onto organomodified Tirebolu bentonite: equilibrium, kinetic and thermodynamic study. J. Hazard Mater. 172 (1), 353e362. Tang, J., Lv, H., Gong, Y., Huang, Y., 2015. Preparation and characterization of a novel graphene/biochar composite for aqueous phenanthrene and mercury removal. Bioresour. Technol. 196, 355e363. Tang, X., Zhou, Y., Xu, Y., et al., 2010. Sorption of polycyclic aromatic hydrocarbons from aqueous solution by hexadecyltrimethylammonium bromide modified fibric peat. J. Chem. Technol. Biotechnol. 85 (8), 1084e1091. Wang, M., Li, G., Huang, L., et al., 2017. Study of ciprofloxacin adsorption and regeneration of activated carbon prepared from Enteromorpha prolifera impregnated with H3PO4 and sodium benzenesulfonate. Ecotoxicol. Environ. Saf. 139, 36e42. Wang, W.M., Song, J., Han, X., 2013. Schwertmannite as a new Fenton-like catalyst in the oxidation of phenol by H2O2. J. Hazard Mater. 262, 412e419. Yan, S., Zheng, G., Meng, X., et al., 2017. Assessment of catalytic activities of selected iron hydroxysulphates biosynthesized using Acidithiobacillus ferrooxidans for the degradation of phenol in heterogeneous Fenton-like reactions. Separ. Purif. Technol. 185, 83e93. Zambianchi, M., Durso, M., Liscio, A., et al., 2017. Graphene oxide doped polysulfone

9

membrane adsorbers for the removal of organic contaminants from water. Chem. Eng. J. 326, 130e140. Zhang, C., Li, Y., Wang, F., et al., 2017a. Performance of magnetic zirconium-iron oxide nanoparticle in the removal of phosphate from aqueous solution. Appl. Surf. Sci. 396, 1783e1792. Zhang, Z., Hou, X., Zhang, X., et al., 2017b. The synergistic adsorption of pyrene and copper onto Fe (III) functionalized mesoporous silica from aqueous solution. Colloid. Surf. Physicochem. Eng. Asp. 520, 39e45. Zhao, S., Huang, G., Wei, J., et al., 2016. Phenanthrene sorption on palygorskite modified with gemini surfactants: insights from modeling studies and effects of aqueous solution chemistry. Water, Air, Soil Pollut. 227 (1), 17. Zhou, T., Wu, X., Zhang, Y., et al., 2013. Synergistic catalytic degradation of antibiotic sulfamethazine in a heterogeneous sonophotolytic goethite/oxalate Fenton-like system. Appl. Catal. B Environ. 136, 294e301. Zhu, J., Gan, M., Zhang, D., et al., 2013. The nature of Schwertmannite and Jarosite mediated by two strains of Acidithiobacillus ferrooxidans with different ferrous oxidation ability. Mater. Sci. Eng. C 33 (5), 2679e2685. Zhu, M., Yao, J., Dong, L., et al., 2016. Adsorption of naphthalene from aqueous solution onto fatty acid modified walnut shells. Chemosphere 144, 1639e1645.