Wenxiang Ni,Luyang Yang,Xiaolin Zhang,Hui Qiu,*

1 Collaborative Innovation Center of Atmospheric Environment and Equipment Technology,Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control,School of Environmental Science and Engineering,Nanjing University of Information Science &Technology,Nanjing 210044,China

2 State Key Laboratory of Pollution Control and Resource Reuse,School of the Environment,Nanjing University,Nanjing 210023,China

Keywords:Iron oxides Polymer Sulfate Adsorption Ternary complex

ABSTRACT Incorporating of hydrous ferric oxide(HFO)inside porous supports with large sizes has become an effective way to decontaminate the water from heavy metals.Ubiquitous anions like sulfate are usually present in high concentrations in water,and might greatly affect adsorption behavior of hybrid HFO.Here,a polymer-based HFO-CPS was fabricated by encapsulating nano-HFO inside a chloromethylated polystyrene polymer(CPS)and the reactivity of HFO-CPS with Cu(II)was evaluated in the presence of sulfate ions.Surface complexation theory was firstly employed to describe the effect of sulfate on Cu(II) adsorption edges of hybrid HFO-CPS,where constant capacitance model (CCM) was adopted.The available weak adsorption site Fe(2)OH of hybrid HFO-CPS was found to decrease from 20% Fe to 5% Fe,which might be caused by the pore plugging effect after HFO encapsulation.With the assumption that a ternary complex was formed,the effect of sulfate on Cu(II) adsorption by HFO-CPS were successfully described by CCM using the optimized Fe(2)OH site under different sulfate concentrations (1 or 10 mmol∙L-1) and Cu/Fe ratios (0.0042 or 0.0252).It is confirmed that the formation of FeOHCuSO4 ternary surface complexes played an important role in enhancing Cu(II) adsorption on HFO-CPS in the presence of sulfate.

1.Introduction

Water pollution is an significant environmental issue and becoming more serious all over the world,such as nitrogen,phosphorus,algal blooms,and heavy metals pollution [1–5].Excessive presence of heavy metals in waters has caused widespread concern because of the severe adverse effects on ecosystem and human health[6–9].Nanosized metal(hydr)oxides have attracted increasing attention as promising sorbents due to their large specific surface area and high activities towards heavy metals [10–13].In particular,hydrated ferric oxides(HFO)have been extensively used in the remediation of natural aquatic systems or wastewater containing toxic metals,because of high adsorption capacity,low cost,excellent environmental compatibility,and chemical stability over a wide pH range [14].Unfortunately,as ultrafine particles,metal(hydr)oxides are associated with undesirable technical issues such as activity loss caused by agglomeration,difficult separation as well as excessive pressure drops when directly used in fixed-bed columns or any other flow-through system [15–18].Loading of nanosized HFO on porous supports with large sizes,such as activated carbon [19],zeolite [20],mesoporous silicates [21] and porous polymeric resins[22,23],is an effective approach to overcome above-mentioned shortcomings.In particular,porous polymeric resins have attracted more and more attention in the field application due to their adjustable pore size and surface chemistry [24–26].Cumbal and SenGupta [24] demonstrated that porous polymeric resin with nondiffusible charged functional groups would enhance or reject permeation of target ionic pollutions due to Donnan membrane effect and could serve as excellent supports for HFO encapsulation.In general,the resultant hybrid materials retain the inherent properties of ferric oxides nanoparticles as well as the outstanding hydrodynamic performance and mechanical strength of the polymeric support.

To the best of our knowledge,almost all works focused on the single-sorbate or multimetallic cation system without any reactive co-existed anions in background,thus the obtained conclusions might not be directly applicable in natural aquatic systems.Few studies have focused on evaluating the adsorption behavior in the presence of sulfate or phosphate,which could act as inorganic ligands possessing strong complexing ability with metals oxides in adsorption process [27,28].Particularly,sulfate is one of the most common ligand anions in natural waters and normally occurs in high concentrations in sulfate-rich systems such as groundwater,marine waters as well as acid mine drainage [29,30] It was reported that sulfate ions could dramatically increase the heavy metals removal of ferric oxides by forming metal-sulfate ternary surface complexes or by modifying electrostatic environment at the oxide-solution interface,which was already verified by spectroscopic characterizations and modeling studies [31–34].Ali and Dzombak [29] confirmed that sulfate could potentially play an important role in regulating sorption of goethite toward Cu(II)ions through the formation of a ternary surface complexFeOHCuSO4at low pH values (pH <5.5) based the surface complexation model.Swedlund and Webster [35] indicated that the adsorption of Cu(II) and Zn(II) onto ferrihydrite were dramatically enhanced in the presence of sulfate,and could be accurately modeled through surface complexation theory only if ternary surface complexes,FeOHCuSO4orFeOHZnSO4,are taken into consideration.Until now,the underlying adsorptive mechanism of bulky HFO in the presence of sulfate can be well explained by the surface complexation models,which has been developed to calculate parameters of thermodynamic properties mathematically and has become an effective tool to elucidate and predict the reactions of oxidesolution interface [27].

Nevertheless,no attempt has been made to quantitatively describe and predict the adsorption behavior of the new fabricated hybrid HFO with surface complexation models in sulfate-rich systems [36].Considering the varied nano-pore structures of polymeric resins,the physicochemical properties of HFO might change significantly in specific surface area and adsorption sites after encapsulated within the inner pore of supports [37].It is reported by Pan et al.[38] that the pore clogging effect after ferric oxides encapsulation dramatically decreased the pore volume and the average pore diameter of the polymeric support,which might be difficult for heavy metals to get access to the surface of oxides loaded and thereby decreasing available adsorption site density[39].Therefore,it is of great importance to establish a specific modeling system of new hybrid nano-HFO for heavy metals removal in sulfate-rich systems.

This work was undertaken to investigate the effect of the major ligand sulfate on Cu(II)adsorption by the HFO encapsulated inside polymer hosts and to elucidate the underlying mechanism of adsorption process with modeling method.Here a commercial chloromethylated polystyrene polymer (CPS) was employed as the host material which is an inert macroreticular material with physicochemical properties well described and has been widely used in catalysis and adsorption [40,41] The effect of sulfate on Cu(II) adsorption of the resultant HFO hybrid for a wide range of sulfate concentrations and Cu/Fe ratios were experimentally determined,and the possible formation of ternary surface complexes was explored using surface complexation model to predict the adsorption data for the range of conditions examined.

2.Materials and Methods

2.1.Materials

All chemicals involved in this research are of analytical grade,and solutions were all prepared with deionized water(18.25 MΩ∙cm).Copper nitrate,sodium sulfate,and Sodium nitrate were purchased from Sigma-Aldrich,Inc.Copper nitrate and sodium sulfate were employed as the sources of Cu(II) and sulfate solution and dissolved in deionized water as stock solutions.The CPS polymers were obtained from Zhengguang Resin Co.(Hangzhou,China)and presented as spherical beads with the sizes ranging from 0.55 to 0.63 mm in diameter.Prior to use,the CPS was firstly subjected to extraction with ethanol for 6 h in a Soxhlet apparatus,followed by rinsing with 1 mol∙L-1HCl,1 mol∙L-1NaOH to remove residue impurities,then were dried under vacuum desiccated at 333 K for 24 h for further use.(Their basic structural information is listed in Table 1)

Table 1The physicochemical properties of the samples

Table 2Intrinsic constants for surface acid base reactions of HFO and the Cu(II) adsorption

2.2.Preparation of polymer-supported nano-hydrated ferric oxides

The preparation of HFO-CPS followed the synthesis procedure described in our previous work and with minor modification[36].In detail,10.0 g of dry CPS beads were added into 300 ml of 1:2 (v/v) ethanol–water solution containing 2 mol∙L-1Fe(NO3)2-∙9H2O.The mixture was stirred for 12 h to ensure that Fe(III) ions were fully impregnated insided the nanopores of CPS beads.After filtration,the polymer beads were immersed into 200 ml of 0.1 mol∙L-1NaOH–NaCl solution at 298 K and then stirred for another 12 h.The Fe(III)preloaded on CPS beads have been precipitated as Fe(III) hydroxides onto the inner surface of polymer beads.Finally,the resultant particles were washed with the deionized water until neutral pH,followed by thermally treated at 323 K for 24 h until reaching a constant weight to obtain the hybrid HFOCPS.

2.3.Adsorption experiments

The Cu(II) adsorption on HFO-CMPS in the absence or presence of sulfate was performed in batch experiments containing 0.01 mol∙L-1NaNO3as the background electrolyte.To be specific,0.0125 g of HFO-CMPS sample was added in 50 ml polypropylene centrifuge tubes.Then,the Cu(II)and Na2SO4stock solutions were added to obtain the preset concentration with a total volume of 25 ml solution.A negligible volume of HNO3or NaOH was used to achieve the desired solution pH.The sealed tubes were then transferred to a thermostatic shaker and vibrated under 200 r∙min-1for 48 h at 298 K.It is sufficient to reach adsorption equilibrium under such experimental conditions demonstrated by pre-experiment.After equilibrium,the supernatant solutions were further filtered through 0.45 μm syringe filters for Cu(II)and sulfate analysis,and the pH was measured immediately.The Cu(II)or sulfate sorption after equilibrium can be determined from the difference between initial and final solution concentrations.

2.4.Characterization

The flame atomic adsorption spectrophotometry (TAS-990,PGENERAL) was employed to measure Cu(II) concentrations in solution.The crystalline structure of HFO-CPS was analyzed by an X-ray diffraction analysis instrument (XRD,XTRA,Switzerland)with CuKaradiation (40 kV,25 mA).Transmission electron microscopy(TEM,JEOL,JEM-2100 HR,Japan)was used to investigate the morphology of HFO encapsulated.The specific surface area and pore size distribution of HFO-CPS were determined by N2adsorption and desorption test at 77 K (Micromeritics ASAP-2020,USA).The concentration of SO4in solution was analyzed by ionchromatographic separation (ICS 1100,Dionex,USA).To measure Fe(III)loadings inside D201,the hybrid HFO-CPS was first digested by nitric-perchloric acid solution and then analyzed by inductively coupled plasma atomic emission spectrometry (ICP-AES,PerkinElmer,USA).

2.5.Modeling scheme

The surface complexation model(SCM)has been widely used to explore the adsorption mechanism of heavy metal ions on the surface of oxide minerals.Unlike the empirical methods of Langmuir and Freundlich isotherm models,SCM was proposed to describe the adsorption process at the solid–liquid interface based on the acid-base reactions and surface complexation reaction which governed by mass law equations [42].According to the various description methods of electrostatic interaction on the oxide surface during adsorption,SCM could be divided into diffusion layer model (DLM),basic stern model (BSM),three-layer model (TLM)and constant capacitance model (CCM).Among these models,CCM assumes that all surface species are inner-sphere complexes,which consistent with the previous spectroscopic studies that the formation of inner-sphere complexes would be the primary adsorption mechanism of heavy metals adsorption on the HFO surfaces.It is also supposed that no surface complexes are formed with ions from the background electrolyte,only one charged layer exists on the surface as well as that surface charge density is linearly related to the surface potential[43,44].Therefore,the simple format of CCM makes it convenient to describe the interfacial reactions of loaded HFO in the absence or presence of sulfate ligand.Dzombak and Morel [42] have demonstrated that the surfaceactive sites of HFO are amphoteric and the surface ionization reactions can be stated as

The adsorption data of Cu(II) in single-sorbate system were modeled using the reactions and equilibrium constants determined by Ali and Dzombak [29] with CCM and the least-squares fitting program FITEQL 4.0.Binary-component sorption of Cu(II)in the presence of sulfate was modeled by using of Cu-SO4ternary surface complexes.Unlike other well-formed goethite or hematite,HFO has the characteristics of surface chemical heterogeneity.Generally,adsorption sites on the surface of HFO could be divided into two types,Type 1 and Type 2.Type 1(Fe(1)OH)represents the high affinity sites which is a small fraction of the total sites;Type 2(Fe(2)OH)represents the most abundant weak affinity sites of the total sites[35].The density of Type 1 and Type 2 sites derived from Dzombak and Morel [42] is 0.005 and 0.2 mol∙mol-1Fe respectively,and it is assumed that the adsorption of metal cation ions occurs at both sites.The surface acid-base reactions and complex reactions occurring in the system were listed in Table 2,and the corresponding reaction equilibrium constants used for fitting adsorption data of Cu(II)were derived from the classical literature of the surface complexing models.

3.Results and Discussion

3.1.Characterization

Fig.1.TEM images of the samples:(a) CPS and (b) HFO–CPS.

TEM images of HFO-CPS depicted in Fig.1 illustrated that HFO was successfully impregnated inside the inner pore of CPS and dispersed uniformly as nanoparticles,which expected to display larger accessible surface areas and stronger activity than the bulky ones.It can be also confirmed by STEM-HAADF images illustrated in Fig.2.The bright spot showed in the left picture represents HFO nanoparticles,and the region of the orange rectangle was measured by EDS mapping.From the distribution of Fe and O elements,it was demonstrated that HFO are dispersed uniformly as nanoparticles.The XRD spectra clearly show the loaded HFO nanoparticles were amorphous in nature.The Fe(III) loading content of hybrid HFO-CPS was 12.17% in Fe mass according to ICP-AES analysis.The impregnation of HFO decreased pore volume of CPS beads from 0.126 to 0.089 cm3∙g-1,and decreased pore diameter from 28.9 to 11.2 nm.However,BET result indicated that the specific surface area of hybrid HFO-CPS increased from 27.7 to 45.8 m2∙g-1,probably resulting from the inclusion of nano-sized HFO.Similar observation was also available in recent study concerning polymerbased nanoparticles for fluoride ions removal [38].The physicochemical properties of hybrid HFO-CPS prepared in this study were summarized in Table 1.

3.2.Adsorption of Cu(II) by HFO-CMPS in sulfate-free system

Fig.2.HAADF image of HFO-CPS (left)and the corresponding EDS elemental maps(right).

Fig.3.Experimental data for Cu(II)adsorption onto HFO-CPS in sulfate-free system of low Cu/Fe ratio ([Cu(II)]:0.05 mmol∙L-1;Cu/Fe ratio:0.00420).

The adsorption of Cu(II)by HFO-CPS in sulfate-free system was determined as a function of pH with a low Cu/Fe ratio of 0.0042.The solution background electrolyte NaNO3concentration was 0.01 mol∙L-1.It can be seen from Fig.3 that the percentage of Cu(II) adsorbed by HFO-CPS increases as solution pH rises gradually,consistent with bulky HFO [36].The experimental data of Cu(II)adsorption was fitted by the CCM model.The surface acid-base reaction constants and intrinsic adsorption constants were listed in Table 2 derived from Dzombak and Morel [42],where the site density of strong adsorption sitesFe(1)OH and weak adsorption siteFe(2)OH was 0.5% Fe and 20% Fe,respectively [35],and other parameters are shown in Table 3.

The fitting results were shown in Fig.4(a),it can be seen that the simulated data was apparently higher than experimental data.According to previous research,the impregnation of HFO could dramatically decrease the pore volume and the average pore diameter of CPS host due to the pore clogging effect,resulting in decrease of available active sites of HFO for Cu(II) ions adsorption[38].Jia et al.[37]demonstrated that the ion exchange sites of the polymer-based titanium phosphate composite became lower than the theoretical values calculated by acid-base titration since titanium phosphate was immobilized inside the porous polystyrene polymer.Therefore,a higher prediction in adsorbed amount of Cu(II)could be obtained since the site density of HFO was directly employed to fit the adsorption behavior of hybrid HFO-CPS.Based on that,it could be necessary to optimize the adsorption sites density of the hybrid HFO-CPS.It was found that when the density of the weak adsorption sitesFe(2)OH is reduced from 0.2 to 0.05 mol∙mol-1Fe,the fitting result of Cu(II) adsorption using CCM are in good agreement with experimental data,as shown in Fig.4(b).It can be concluded that the density of effective adsorption site for HFO might decline because Cu(II)ions could be getting more difficult to get access to the inner surface of composite due to pore clogging effect after HFO immobilized inside the inner pore of porous support.

According to results obtained by low Cu/Fe system above,the high Cu/Fe ratio system with Cu/Fe=0.0252 continued to be fitted by CCM with the optimized density of adsorption sites.The results illustrated in Fig.5 showed that the model fitting data agrees well with the experimental data.It can be seen that the optimized parameters of active sites density were verified to be suitable for the new prepared polymer-based HFO-CPS.It is of great importance to optimize adsorption parameters for hybrid HFO composite in order to obtain a better fit result with experimental data in case changes happen due to the pore clogging effect caused by HFO inclusion.Next,the sulfate effect on Cu(II) adsorption by HFOCPS could be examined and fitted by CCM with the density of surface active site optimized by sulfate-free system.

3.3.Adsorption of Cu(II) by HFO-CMPS in sulfate systems

The effect of sulfate on the Cu(II) adsorption on HFO-CPS was shown in Figs.6 and 7.In general,adsorption of Cu(II) was increased apparently in the presence of sulfate both in low and high Cu/Fe ratio systems.This phenomenon for low Cu/Fe ratio systems was much more evident than high Cu/Fe ratio systems.For example,the presence of sulfate increased Cu(II) adsorption upto 21% at pH 4.5 when the Cu(T)/Fe ratio was lower at 0.00420[Fig.6(a)].In contrast,the increased Cu(II)adsorption was less than 5%when the Cu(T)/Fe ratio was higher at 0.0252[Fig.7(a)].Moreover,the Cu(II) adsorption was dramatically increased as the sulfate concentration rises,as shown in Figs.6(b) and 7(b).For example,The increased Cu(II) adsorption for low Cu/Fe ratio systems at pH 4 was increased from 21% to 30% when the concentration of sulfate rises from 1 mmol∙L-1to 10 mmol∙L-1.The adsorption behaviors of Cu(II)on HFO-PS in the presence of sulfate were fitted by CCM using the intrinsic adsorption constants reported by Dzombak and Morel[42].Since the XPS results in previous studies have confirmed the formation of Cu-SO4ternary complexon the surface of the HFO-CPS composite in sulfate systems,the adsorption reaction of sulfate onto the surface of HFO and the coordination reaction of ternary complexshould also be taken into account.The relevant new reactions and coordination constants in sulfate systems are shown in Table 4.Swedlund and Webster[35]have confirmed that the ternary surface complex formed on strong adsorption sites has no significant effect on Cu(II)adsorption fitted by the surface complex model.Therefore,the formation of ternary complexon the weak adsorption site was only considered here and the coordination reaction constant lgKINT(7.59) was quoted directly from the results reported by Swedlund and Webster [35].According to the optimization results obtained in the sulfatefree system,the density of strong adsorption siteOH and weak adsorption site—Fe(2)OH was 0.5%Fe and 5%Fe,respectively.The fitting results by CCM were shown in Figs.6 and 7.In the low Cu/Fe ratio system(Fig.6),it can be seen that the fitting results of Cu(II) adsorption on HFO-CPS agrees well with the experimental data for a broad range of pH value,either in 1 mmol∙L-1or in 10 mmol∙L-1of SO4background.In the high Cu/Fe ratio system(Fig.7),the fitting results are highly coincident with the experimental data for both low and high sulfate concentration background as well.This effect can only be accurately modeled using CCM and surface complexation theory if ternary surface complexes,are taken into consideration.Therefore,it can be confirmed that the CCM can successfully fit the adsorption of Cu(II) by HFO-PS0under different concentrations of sulfate and Cu/Fe ratio,with the assumption that the ternary surface complexwas formed on the surface of the HFO-PS composite.

Table 3Parameter values used in CCM modeling

Table 4Constants for surface complex reactions of HFO in sulfate systems

Fig.4.Experimental data and fitting curves for Cu(II) adsorption onto HFO-CPS in sulfate-free systems of low Cu/Fe ratios ([Cu(II)]:0.05 mmol∙L-1;Cu/Fe:0.00420).(a):OH,20%;(b):OH,5%.

Fig.5.Experimental data and fitting curves for Cu(II) adsorption onto HFO-CPS in sulfate-free systems of high Cu/Fe ratios ([Cu(II)]:0.3 mmol∙L-1;Cu/Fe:0.0252).

Fig.6.Effect of sulfate on the Cu(II) adsorption onto HFO-CPS in low Cu/Fe ratio system ([Cu(II)] 0.05 mmol∙L-1;Cu/Fe:0.00420):Experimental data and fitting curves.(a)[sulfate]=1 mmol∙L-1;(b) [sulfate]=10 mmol∙L-1.

Fig.7.Effect of sulfate on the Cu(II) adsorption onto HFO-CPS in high Cu/Fe ratio system ([Cu(II)] 0.3 mmol∙L-1;Cu/Fe:0.0252):Experimental data and fitting curves.(a) 1 mmol∙L-1;(b) S 0 mmol∙L-1.

3.4.Distributions of adsorbed Cu(II) species in sulfate systems

Fig.8 shows the morphological distribution of the adsorbed Cu(II)species calculated by CCM,taking the system with a high Cu/Fe ratio(0.0252)for example.As the value of pH decreasing,the proportion of ternary complexin the total adsorption amount gradually increased,while the proportion of Cu(II)adsorbed in the form ofFe(1)OCu+andFe(2)OCu+get much lower.The ternary complexdominates below pH=3.5 for the 1 mmol∙L-1of sulfate background.When the SO4concentration was up to 10 mmol∙L-1,the ternary complexdominates below pH=5.4.It is also confirmed that the increase in background SO4concentration could significantly promote the formation of ternary surface complexes,which resulting in tremendous enhancement in adsorbed amount of Cu(II) by hybrid HFO-CPS.

4.Conclusions

Fig.8.Distributions ofadsorbedCu(II) species on HFO-CPS in binary-component systems.(a) Total [Cu(II)]=0.3 mmol∙L-1,Total ==1 mmol∙L-1;(b) Total [Cu(II)]=0.3mmol∙L-1,Total==10mmol∙L-1.

A new composite HFO-CPS is fabricated by anchoring nanosized iron oxides within a commercial chloromethylated polystyrene polymer (CPS) in order to evaluate the effect of sulfate on the Cu(II) adsorbed by hybrid HFO through surface complex theory and CCM model.It is confirmed that the available weak adsorption sitedecreases from 20% Fe to 5% Fe caused by the pore plugging effect after HFO immobilization.Based on that,the CCM model can successfully describe the adsorption of Cu(II) by HFO-PS under different sulfate concentrations (1 or 10 mmol∙L-1)and Cu/Fe ratio(0.0042 or 0.0252)with the assumption that a ternary surface complex was formed.The microscopic adsorption mechanism is verified that the addition of sulfate can promote Cu(II)adsorption obviously due to the formation of the ternary surface complexon the surface of hybrid HFO-CPS.The approach by using surface complex theory to clarify the underlying adsorption mechanism on hybrid HFO-CPS in sulfate systems is firstly established and very inspiring.The possible utilization of surface complex theory on predicting adsorption behaviors of hybrid metal oxides deserves systematically investigation in future work.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ackn(2o)wledgements

This work was financially supported by National Natural Science Foundation of China (21607080),Natural Science Foundation of Jiangsu Province(BK20160946).Meanwhile,this research was supported by Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD) and Jiangsu Engineering Technology Research Center of Environmental Cleaning Materials.