Arsenate Adsorption by Hydrous Ferric Oxide ... - ACS Publications

Jan 14, 2016 - Hongchao Li, Chao Shan,* Yanyang Zhang, Jianguo Cai, Weiming Zhang, and Bingcai Pan*. State Key Laboratory of Pollution Control and ...
1 downloads 0 Views 3MB Size
Download PDF
Research Article www.acsami.org

Arsenate Adsorption by Hydrous Ferric Oxide Nanoparticles Embedded in Cross-linked Anion Exchanger: Effect of the Host Pore Structure Hongchao Li, Chao Shan,* Yanyang Zhang, Jianguo Cai, Weiming Zhang, and Bingcai Pan* State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210023, China S Supporting Information *

ABSTRACT: Three composite adsorbents were fabricated via confined growth of hydrous ferric oxide (HFO) nanoparticles within cross-linked anion exchangers (NS) of different pore size distributions to investigate the effect of host pore structure on the adsorption of As(V). With the decrease in the average pore size of the NS hosts from 38.7 to 9.2 nm, the mean diameter of the confined HFO nanoparticles was lessened from 31.4 to 11.6 nm as observed by transmission electron microscopy (TEM), while the density of active surface sites was increased due to size-dependent effect proved by potentiometric titration. The adsorption capacity of As(V) yielded by Sips model was elevated from 24.2 to 31.6 mg/g via tailoring the pore size of the NS hosts, and the adsorption kinetics was slightly accelerated with the decrease of pore size in background solution containing 500 mg/L of Cl−. Furthermore, the enhanced adsorption of As(V) was achieved over a wide pH range from 3 to 10, as well as in the presence of competing anions including Cl−, SO42−, HCO3−, NO3− (up to 800 mg/L), and PO43− (up to 10 mg P/L). In addition, the fixed-bed working capacity increased from 2200 to 2950 bed volumes (BV) owing to the size confinement effect, which did not have adverse effect on the desorption of As(V) as the cumulative desorption efficiency reached 94% with 10 BV of binary solution (5% NaOH + 5% NaCl) for all the three adsorbents. Therefore, this study provided a promising strategy to regulate the reactivity of the nanoparticles via the size confinement effect of the host pore structure. KEYWORDS: sorption, composite, ion exchanger, nanoparticle, size confinement effect, pore structure

■

INTRODUCTION Metal oxide nanoparticles have shown great potential as promising adsorbents in water treatment process, and their reactivity has been proved to be highly size-dependent.1−7 It has been reported that the size of ferric (hydr)oxides could influence their surface charge and specific surface area.1−4,6,7 As such, these size-dependent properties would pose great impact on their reactivity such as adsorption performance. For instance, Yavuz et al.5 demonstrated that, with particle size of Fe3O4 decreased from 300 to 12 nm, the removal efficiency of As(III) increased from 24.9 to 99.2% while that of As(V) increased from 29.2 to 98.4%. Yean et al.6 found that the adsorption capacity of As(III) on the 20 nm magnetite nanoparticles was approximately 18 times larger than that on the 300 nm ones. Auffan et al.2 also found that the adsorption of As(III) was elevated almost four times with the particle size of maghemite decreased to less than 20 nm. Nevertheless, as a © 2016 American Chemical Society

result of their ultrafine nature, nanoparticles are not suitable for direct use in flow-through systems due to the excessive pressure drop, which considerably limited their application in practical water treatment processes. Porous materials, such as silica, activated carbon, and polymeric ion exchangers, have attracted enormous attention because of their appropriate size and competence as the supporting host for nanoparticles.8−13 Recently, increasing interests have arisen concerning the effect of the pore size of porous host on the size of the confined nanoparticles. Richardson et al.14 showed that smaller pore size in alumina resulted in a smaller final Ni particle size. Bore et al.15 also demonstrated the average size of the Au particles increased Received: October 15, 2015 Accepted: January 14, 2016 Published: January 14, 2016 3012

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces

Arsenate [As(V)] was employed as a model anionic contaminant to compare the adsorption capacity and adsorption kinetics of the HFO−NS composites. Moreover, the adsorption of As(V) by the HFO−NS composites was also investigated at different solution pH as well as in the presence of competing anions. Fixed-bed adsorption and regeneration were conducted to evaluate the size confinement effect on the practical application performance.

from 3.9 to 7.2 nm with the pore size of mesoporous silica increasing from 2.9 to 6.5 nm. Tripathi et al.16 found that the size of boron−carbon−oxy nitride particles were tunable by employing mesoporous silica with different pore sizes as the hosts. Therefore, the hosts with smaller pores confined the growth of nanoparticles and thus decreased their size. Naturally, considering the size-dependent effect mentioned above, the reactivity of the nanoparticles supported by the porous hosts has great potential to be enhanced via the size confinement effect of the host pore structure. Prior studies17−20 have shown that interface confinement of the host materials has significant influence on the catalytic reactivity of the nanoclusters. However, few studies have systematically elucidated the effect of the host pore structure on the surface chemical properties of the confined nanoparticles and the adsorption performance of the composite. In our previous studies, a series of nanocomposites (i.e., metal oxide nanoparticles immobilized by macroporous ion exchangers) were fabricated as promising adsorbents for water purification.21−27 The synthesized hybrid materials exhibit great potential for practical application due to their appropriate size for fixed-bed packing, mechanical strength, large adsorption capacity, and preference for the target contaminants. For example, by employing the macroporous anion exchanger as the polymer host, the anionic contaminants such as arsenate,26 phosphate,21−23 and fluoride24 could be efficiently removed from water through preconcentration via Donnan membrane effect9 and the subsequent uptake by the nanoparticles such as hydrous ferric oxide (HFO). For instance, the adsorption of As(V) by the anion-exchanger supported HFO nanoparticles have been investigated,9,28,29 and the major mechanism of As(V) adsorption by the host anion-exchanger and the embedded HFO nanoparticles could be briefly expressed as eqs 1−3.30,31

■

[−N+(CH3)3 ]Cl− + H 2AsO4 − → [−N+(CH3)3 ]H 2AsO4 − + Cl−

(1)

2[−N+(CH3)3 Cl−] + HAsO4 2 − → [2 − N+(CH3)3 ]HAsO4 2 − + 2Cl−

EXPERIMENTAL SECTION

Materials. All chemicals used in this study were of analytical grade unless specified. The stock solutions As(V) (1000 mg/L) were prepared by dissolving Na2HAsO4·7H2O (Sigma−Aldrich) with ultrapure water (18 MΩ·cm). The chloromethylated polystyrene (CS) beads, provided by Zhengguang Industrial Co., Ltd. (Hangzhou, China), were used as the precursor to fabricate the NS hosts. Prior to use, the CS beads were sieved to the diameter range of 0.45−0.55 mm, and were subjected to extraction in a Soxhlet apparatus with ethanol for 8 h to remove possible impurities. Preparation of Adsorbents. The strongly basic anion exchangers (NS) were synthesized according to the typical protocol.32,33 In brief, 30 g of CS beads were soaked in 180 mL nitrobenzene at 298 K for 12 h. Anhydrous ZnCl2 (6 g) was then introduced into the mixture as the catalyst for post-cross-linking reaction. After 0.5−5 h of thermal treatment (353 K) in oil bath, the beads were filtered out and Soxhlet extracted with ethanol for 8 h to remove the residual nitrobenzene. Then, the obtained beads were immersed into 250 mL of ultrapure water for 12 h to ensure entire swelling. A certain volume of 33% trimethylamine was added under rapid stirring and the temperature was increased gradually up to 60 °C with the heating rate of 2 °C/h and remained for 6 h. Finally, after cooled to room temperature, the beads were filtered and repeatedly washed to obtain the NS hosts. NS hosts with identical exchange capacities but different pore size distributions were obtained by adjusting the dosage of trimethylamine and the reaction time of post-cross-linking process, respectively. For a convenient description, the three synthesized host materials were denoted as NS-L, NS-M, and NS-H, respectively, where L, M, and H represent the different reaction time, following the order of H > M > L. The HFO−NS composites were synthesized according to our previous study.23 Briefly, 10 g of NS host were added into mixed solution of HCl, NaCl and FeCl3, followed by continuous agitation for 12 h. The Fe(III)-loaded NS hosts were filtered out and introduced into 100 mL of 5% NaOH solution with continuous stirring for another 12 h. Then, the resultant beads were subjected to thermal treatment at 335 K to obtain the HFO−NS composites. The content of Fe loading was controlled by adjusting the dosage of FeCl3. Correspondingly, the three synthesized hybrid adsorbents were denoted as HFO−NS-L, HFO−NS-M, and HFO−NS-H, respectively. Characterization of Adsorbents. The pore size distribution along with specific surface area of the NS hosts and HFO−NS composites were determined using surface and pore analyzer (Nova 3000, Quantachrome, Boynton Beach, FL). The samples were outgassed at 60 °C for 24 h prior to the measurements, and the surface area were fitted with Brunauer−Emmett−Teller (BET) model. The morphology and size of the HFO nanoparticles confined in the HFO−NS composites was characterized with TEM (JEM-200CX, JEOL, Japan), and the size distribution was statistically counted from 150 particles in the TEM images for each sample. X-ray diffraction (XRD, X’TRA, ARL, Switzerland) was used to investigate the mineralogy of the HFO−NS composites with Cu Kα radiation. The contents of Fe loaded within the HFO−NS composites were determined by an atomic absorption spectrophotometer (AA-7000, Shimadzu, Japan) after acid digestion. Scanning electron microscope (SEM, S-3400 II, Hitachi, Japan) equipped with energy dispersive spectroscopy (EDS) was employed to visualize the distribution of Fe along the diameter on the cross section of the HFO−NS hybrid beads. A particle crushing device (YHKC-2A, Yinhe Instruments, Taizhou, China) was employed to determine the mechanical strengths of the

(2)

>FeOH + H 2AsO4 − + H+ → >FeOAsO(OH)2 + H 2O (3)

However, the pore size distribution of the macroporous ion exchangers was mainly in the macro- and meso-pore regions (average diameter d > 10 nm, in dry state). Inspired by the aforementioned size confinement effect, pore structure of the polymeric ion exchanger can be tuned via post-cross-linking to further improve the adsorption capabilities of the composite. Meanwhile, it is also important to investigate the confinement effect on the adsorption kinetics. Hence, the objective of this study is to elucidate the effect of host pore structure on the adsorption performance of the nanoparticles confined in the cross-linked ion exchanger. Three composite adsorbents were fabricated via confined growth of HFO nanoparticles within strongly basic anion exchangers (NS) of different pore size distributions. The size confinement effect on the diameter of the embedded HFO nanoparticles was visualized by transmission electron microscopy (TEM) while that on the surface chemical properties of the nanoparticles was characterized by potentiometric titration and zeta potential. 3013

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces Table 1. Characteristics of the NS Hosts and Corresponding HFO−NS Composites material NS-L NS-M NS-H HFO−NS-L HFO−NS-M HFO−NS-H

matrix poly(styrenedivinylbenzene)

functional group

exchange capacity (mmol/g)

BET surface area (m2/g)

average pore diameter (nm)

−N+(CH3)3

1.18 1.18 1.17

29 78 113 55 167 214

38.7 14.4 9.2 15.7 7.8 6.0

prepared materials following the procedure described in our previous study.34 The surface chemical properties of the NS hosts and HFO− NS composites were characterized by potentiometric titration31 and zeta potential (Nano ZS90, Malvern Instruments, U.K.). An automatic titration system (T50, Mettler Toledo, Switzerland) with a combined glass pH electrode (DGi115-SC, Mettler Toledo, Switzerland) was employed for the titration experiments, and the detailed operation procedure along with the calculation method was available in the Supporting Information (Text S1). Batch Adsorption. In batch adsorption experiments, 40 mg of adsorbent were introduced to 250 mL conical flasks containing 100 mL of As(V) solution. The sealed conical flasks were shaken in a thermostatic incubator under 160 rpm at 298 K for 24 h to ensure equilibrium. Adsorption isotherms were obtained by varying the initial As(V) concentration while the solution pH was adjusted with 0.1 M HCl or NaOH solution. To examine the effects of the competing anions on As(V) uptake, NaCl, Na2SO4, NaHCO3, NaNO3, or NaH2PO4/Na2HPO4 were introduced into the solution. Adsorption kinetic experiment was performed in 500 mL As(V) solution with 0.2 g of adsorbent, and a 0.5 mL aliquot was sampled at a series of time intervals for As analysis. Atomic fluorescence spectrometer (AF-640A, Rayleigh, Beijing, China) was employed to analyze As concentration and the detection limit for As was 0.20 μg/L. The amount of As loaded onto the adsorbents was calculated by conducting mass balance between the initial and final concentration. Fixed-Bed Adsorption and Regeneration. Fixed-bed column experiments were conducted in a glass column (12 mm in diameter and 130 mm in length) packed with 5 mL (∼2.1 g) of adsorbent. Synthetic solution containing As(V) as well as competing anions flowed down through the column, and the flow rate was controlled by a peristaltic pump (BT100-2J/YZ1515x, LongerPump, China). The empty bed contact time (EBCT) of adsorption experiment was 4 min. After breakthrough, the in situ regeneration of adsorbent was conducted with binary solution of NaOH−NaCl (both 5 wt %) and the EBCT for regeneration was 60 min.

Fe content (wt %)

7.7 7.5 7.4

mechanical strength (N) 7.23 7.06 6.97 7.07 6.84 6.75

± ± ± ± ± ±

2.48 1.88 1.59 1.71 1.58 2.04

Figure 1. Pore size distribution of the polymer hosts NS-L, NS-M, and NS-H.

The three HFO−NS hybrid materials had similar Fe loading contents (7.4−7.7 wt %). The BET specific surface areas were 55, 167, and 214 m2/g for HFO−NS-L, HFO−NS-M, and HFO−NS-H, respectively. The increased BET specific surface areas of the HFO−NS hybrid materials compared with their corresponding NS hosts could be attributed to the embedded HFO nanoparticles. A similar observation was also reported in our recent study concerning anion-exchanger supported hydrous zirconium oxide nanoparticles.24 The Fe element scanning with SEM-EDS along the diameter on the cross section of the HFO−NS beads (Figure S1) suggested that HFO nanoparticles were uniformly embedded in the NS hosts, owing to the well-distributed FeCl4− precursor anchored onto the quaternary ammonium functional groups via ion exchange in the preparation process.23 The XRD patterns of the HFO− NS composites (Figure S2) implied that the HFO within the NS hosts were all amorphous in nature, which would be favorable for As(V) adsorption relative to crystalline phases.36 The mechanical strength of the NS host slightly decreased with the increase of post-cross-linking time, while the impregnation of the particles with HFO nanoparticles inside the host also resulted in slight decrease of the mechanical strength (Table 1). Nevertheless, none of the decreases was statistically significant (P = 0.524−0.803), suggesting neither the post-cross-linking nor the HFO loadings had obvious effect on the mechanical strength of the particle. TEM images of the HFO−NS composites as well as the statistical size distribution of the confined HFO nanoparticles (N = 150) were displayed in Figure 2. Clearly, with decrease of the host pore size, the major fraction peak shifted toward the smaller size direction. The mean diameter of the HFO nanoparticles was 31.4, 24.7, and 11.6 nm for HFO−NS-L, HFO−NS-M, and HFO−NS-H, respectively. This observation provided direct evidence that decreasing the pore size remarkably improved the dispersion of

■

RESULTS AND DISCUSSION Characterization of the Adsorbents. The major properties of the NS hosts were characterized, and the results are summarized in Table 1. The average pore size gradually decreased from 38.7 to 9.2 nm with the increase of the postcross-linking reaction time due to the conversion of large pores into small ones. Accordingly, the BET specific surface area increased from 29 to 113 m2/g with the decrease in pore diameter. The pore size distributions of the three NS hosts were shown in Figure 1. The pore diameter of NS-L mainly ranged from 10 to 75 nm, whereas, NS-H and NS-M had wider pore size distribution from 1.6 to 75 nm. In particular, NS-H and NS-M were evidently rich in micropores (d < 5 nm) relative to NS-L. Prior study35 also showed that the post-crosslinking reaction significantly affected the pore structure, and longer reaction time was favorable for the formation of NS hosts with smaller pore size and larger surface area. Additionally, the three NS hosts had almost the same ion-exchange capacity (1.17−1.18 mmol/g). Therefore, the NS hosts had no substantive differences except pore size distribution. 3014

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces

Figure 2. TEM images of (a) HFO−NS-L, (b) HFO−NS-M, (c) HFO−NS-H, and (d) statistical size distribution of the HFO nanoparticles embedded inside the polymeric hosts (N = 150).

whereas HFO−NS-L had the flattest one. Considering the major surface functional group of the HFO nanoparticles were the hydroxyl groups, this titration result implied that the density of hydroxyl groups of the HFO nanoparticles increased as the average pore size diameters of the NS hosts decreased from 38.7 to 9.2 nm.7 Herein, the potentiometric titration analysis proved that the surface chemistry of the embedded nanoparticles could be tuned through size confinement effect of the host pore structure. Specifically, smaller HFO nanoparticles tend to have larger surface area with more active sites, which would benefit the adsorption capability of the composite. Figure S4 showed that the zeta potentials of the three HFO− NS composites were all positive and above 10 mV in the pH range of 3−11 due to the fixed positive charge of ammonium groups.22 The zeta potentials basically followed the same order of HFO−NS-H > HFO−NS-M > HFO−NS-L, as a result of the increased active sites of embedded HFO nanoparticles. Adsorption Isotherm. The adsorption isotherms of As(V) on the three HFO−NS composites were shown in Figure 4.

the HFO nanoparticles inside the NS hosts as a result of the size confinement effect. The influence of host pore structure on the surface chemistry of the confined HFO nanoparticles was investigated by potentiometric titration. Considering the immobilized quaternary ammonium groups might consume the added OH− through ion exchange, the NS hosts were also titrated as references. The net surface charge QH (mmol/g) was plotted versus pH in Figure 3, where the detailed method of QH

Figure 3. Net surface charge (QH) of HFO−NS-L, HFO−NS-M, and HFO−NS-H as a function of pH. Data collected via potentiometric titration of 2 g/L adsorbent in 0.1 M NaNO3.

calculation can be found in Text S1 (Supporting Information) and literature.22 As can be seen in Figure S3, the titration curves of the NS hosts overlapped with that of blank water, suggesting that the added HNO3 or NaOH did not react with the surface groups of NS hosts but was only used to change the pH of the solution. This observation was reasonable because ion-exchange process was completely masked in the presence of 0.1 M NaNO3. Thereby, the titration curves of HFO−NS composites (Figure 3) exclusively reflected the surface chemical properties of the encapsulated HFO nanoparticles. Among all the three composites, HFO−NS-H exhibited the steepest titration curve,

Figure 4. Adsorption isotherm of As(V) by HFO−NS-L, HFO−NSM, and HFO−NS-H at 298 K. Initial pH 7.0 ± 0.2, adsorbent dosage 0.4 g/L. 3015

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces

attributed to the enhanced adsorption capacity as discussed above. To quantitatively describe the adsorption kinetics, pseudo-first-order model and pseudo-second-order model39−41 (described in Text S3, Supporting Information) were employed to fit the kinetic data of all the three composites, and the fitted parameters were summarized in Table S3. Although the coefficients of determination (R2) were higher than 0.989 for both kinetic models, yet the calculated qe values of pseudo-first order kinetics model were closer to the experimental data. Therefore, the pseudo-first-order kinetics model could better describe the kinetics of As(V) uptake onto the HFO−NS composites. A previous study21 reported that the adsorption of phosphate by Zr(IV) oxide nanoparticles confined in anion-exchanger also followed the pseudo-first order kinetics model. Thus, under the same experimental condition, the adsorption kinetics of the three composites could be compared by the pseudo-first order rate constant (k1). It could be seen from Table S3 that the k1 values of the three HFO−NS composites were very close to each other (1.82− 1.91 h−1). This observation demonstrated that the As(V) adsorption kinetics was not obviously affected by the size confinement effect of the host pore structure. The unchanged adsorption kinetics should be owing to the immobilized positively charged quaternary ammonium groups,24 which would enhance the permeation and preconcentration of As(V) via Donnan membrane effect.9 In addition, the presence of macropores within the NS hosts could also facilitate the intraparticle diffusion of As(V).42,43 The adsorption kinetics of As(V) by the HFO−NS hosts were also investigated in the presence of 500 mg/L of chloride, and the results were shown in Figure 5b. Due to the strong competing effect of high concentration of chloride, the ionexchange effect of the NS hosts was masked (Figure S6). Under such circumstance, the qe values could be exclusively attributed to the adsorption by the embedded HFO nanoparticles. The experimental qe values were 9.56, 11.9, and 14.4 mg/g for HFO−NS-L, HFO−NS-M, and HFO−NS-H, respectively. The increasing qe values with decrease of host pore size also reflected the enhanced adsorption capacity with improved dispersion of the confined nanoparticles. The kinetics curves were also fitted with both pseudo-first order and pseudo-second order kinetics models, and the results were listed in Table S4. Similarly, all the kinetics curves were better fitted with the pseudo-first order kinetics model judging from the experimental and calculated qe values. Moreover, the pseudo-first order rate constants (k1) slightly increased from 0.568 to 0.613 h−1 with the decrease of the host pore size. The observation showed that in the presence of competing anion (Cl−), the adsorption kinetics was not inhibited by size confinement of the host pore structure. On the contrary, the adsorption kinetics was slightly accelerated, which should also be attributed to the better dispersion and smaller size of the confined HFO nanoparticles. Effect of Solution pH. The influence of solution pH on As(V) adsorption by the three HFO−NS composites was depicted in Figure 6. It could be seen that the adsorption performance of As(V) was highly pH-dependent. However, the optimal pH range for As(V) removal was 6−8 for all the three HFO−NS composites. Further decrease of pH from 6 to 3 or increase from 8 to 11 resulted in an obvious drop in adsorption capacity. The effect of pH could be explained in terms of both As(V) speciation and the surface chemistry of the HFO nanoparticles. With pH decreased from 6 to 3, As(V) was

Clearly, the adsorption capacities of As(V) followed increasing order from HFO−NS-L to HFO−NS-H. The data were further analyzed by Langmuir, Freundlich, and Sips models37,38 (described in Text S2 in the Supporting Information), and the obtained parameters were summarized in Table S1. Results showed that all the three isotherms were best described by the Sips model (R2 from 0.993 to 0.997), followed by Langmuir model (R2 from 0.964 to 0.984). From NS-HFO-L to NSHFO-H, the adsorption capacity (qmax) yielded by Langmuir model increased from 23.1 to 27.8 mg/g, and those by Sips model increased from 24.2 to 31.6 mg/g. This observation verified the hypothesis that the adsorption capacity can be promoted via the size confinement effect of the host pore structure. Furthermore, this result could be supported by the result of surface chemistry characterization mentioned above. The enhanced qmax can be ascribed to the higher density of surface adsorption sites due to decrease of the size of the confined HFO nanoparticles. Apart from As(V), the adsorption isotherms of As(III) on the three HFO−NS composites were also determined, and the results were shown in Figure S5 and Table S2. Similarly, the adsorption capacity of As(III) also increased with the decrease of the pore size of the NS hosts. These above results both proved that the size confinement effect of the host pore structure played a major role in the adsorption capacity of the nanoparticles confined in the crosslinked anion exchanger. As(V) Adsorption Kinetics. Adsorption kinetics of As(V) by the three HFO−NS composites were investigated, and the results were displayed in Figure 5a. Apparently, the amount of As(V) adsorbed at equilibrium (qe) increased from 21.5 to 26.3 mg/g with the decrease of the host pore size, which was

Figure 5. Adsorption kinetics of As(V) by HFO−NS-L, HFO−NS-M, and HFO−NS-H at 298 K (a) in Cl− free solution and (b) in background solution containing 500 mg/L of Cl−. Initial As(V) 20 mg/L, initial pH 7.0 ± 0.2, adsorbent dosage 0.4 g/L. 3016

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces

Figure 6. Effect of solution pH on the adsorption of As(V) by HFO− NS-L, HFO−NS-M, and HFO−NS-H at 298 K. Initial As(V) 20 mg/ L, adsorbent dosage 0.4 g/L.

Figure 7. Effect of competing anions on the adsorption of As(V) by HFO−NS-L, HFO−NS-M, and HFO−NS-H at 298 K. Initial As(V) 10 mg/L, initial pH 7.0 ± 0.2, adsorbent dosage 0.4 g/L.

protonated to the less negatively charged species (i.e., H3AsO4 or H2AsO4−; Figure S7), which would reduce the electrostatic attractions between As(V) and the fixed positively charged quaternary ammonium groups of the NS hosts.23 At higher pH from 8 to 11, the surface hydroxyl groups of the HFO nanoparticles underwent a deprotonation process and turned negatively charged, which is unfavorable for As(V) sequestration.44−46 The competition induced by OH− anions would also inhibit the uptake of As(V). However, under the same pH condition, HFO−NS-H always exhibited the highest adsorption capacity, followed by HFO−NS-M and that of HFO−NS-L was always the lowest, only except that the performance of HFO− NS-H was equivalent to that of HFO−NS-M at pH 10 and 11. The above results demonstrated that the size confinement effect of the host pore structure on the adsorption of As(V) by the HFO−NS composite was significant not only at pH 7, but also over a broad pH range from 3 to 10. Prior studies29,47 also reported that ferrihydrite and polymer-supported HFO could have wide applicable pH range for the uptake of As(V). In addition, the release of Fe in the treated water was not detected (detection limit 0.03 mg/L) throughout the pH range from 3 to 11, suggesting the stability of the HFO−NS composites. Effect of Competing Anions. Considering practical applications for As(V) removal in water treatment, it is important to investigate the possible competitive effects of the competing anions on the adsorption of As(V). The effects of four common anions including Cl−, SO42−, HCO3−, and NO3− (up to 800 mg/L) were investigated for the three HFO−NS composites. As shown in Figure 7, an obvious decline in the As(V) adsorption capacity of all HFO−NS composites was observed with the increase of the four competing anions from 0 to 400 mg/L. In comparison, further increase of the concentration of competing anions to 800 mg/L did not result in obvious decrease in the adsorption capacities of the three HFO−NS composites. The decline was due to the fact that competing anions greatly occupied the ion-exchange sites of the NS hosts and even completely masked the ion-exchange effect at high concentrations, as proved in Figure S6, while the unchanged adsorption performance in the presence of high concentration of competing anion (400−800 mg/L) was ascribed to the tight inner-sphere complexation of As(V) with the HFO nanoparticles. In contrast, Cl− and NO3− could only form outer-sphere complex with Fe(III) oxides,9,48,49 and SO42− would also be adsorbed predominantly via outer-sphere

complexation at pH > 6.50,51 Although HCO3− could form inner-sphere complex with Fe(III) oxides,52 yet the affinity was much lower than As(V).53 Thus, the four competing anions could not evidently inhibit the sequestration of As(V) by the confined HFO nanoparticles. In comparison with the above four common competing anions, phosphate could result in significant decline of As(V) adsorption in a relatively low concentration range (0−10 mg P/L), as shown in Figure S8, because phosphate could also form tight inner-sphere complex with Fe(III) oxides.23,48 These effects of the competing anions observed in the present study were in agreement with the previous literature.22,24,54,55 Moreover, under the same concentration of one specific competing anion, HFO−NS-H always exhibited the highest adsorption capacity, whereas HFO−NS-L had the lowest one. To further illustrate the adsorption preference of the HFO− NS composites, the distribution coefficient,25 Kd (L/g), was derived according to eq 4: c − ce V Kd = 0 ce m (4) where c0 and ce (mg/L) are the initial and equilibrium As(V) concentrations, respectively, V (L) is the volume of the solution, and m (g) represents the mass of the adsorbent. The calculation results (Table S5) showed that HFO−NS-H always presented the largest Kd values, followed by HFO−NS-M and then HFO−NS-L. Furthermore, the disparities among the three composites were especially evident at high concentrations of competing anions. Under such conditions, the ion-exchange effect of the NS hosts was masked and the adsorption of As(V) mainly relied on the inner-sphere complexation with the HFO nanoparticles. Therefore, the disparity of Kd values particularly reflected the size-dependent reactivity of the HFO nanoparticles. The above results proved that the size confinement effect of the host pore structure on As(V) adsorption by HFO− NS composites was also significant in the presence of common competing anions, which laid foundation for practical application with respect to the complicated components of actual waters. Fixed-Bed Adsorption and Regeneration. Fixed-bed adsorption was conducted to evaluate the practical application potential of the three HFO−NS composites for treatment of As-contaminated water. Synthetic feeding solution containing 3017

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces 200 μg/L of As(V) and various competing anions (Cl−, NO3−, SO42−, HCO3−, and SiO32−, concentration ranged from 5 to 150 mg/L) was employed to simulate contaminated water, and the detailed composition is shown in Figure 8. According to the

BV, which did not have adverse effect on the desorption of As(V). Therefore, this study provides a promising strategy to regulate the reactivity of the nanoparticles via the size confinement effect of the host pore structure.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b09832. Detailed operation procedure and calculation method of the potentiometric titration; adsorption isotherm and kinetics models; adsorption isotherm fitting parameters of As(V) and As(III) by HFO−NS composites; kinetics curve fitting parameters for adsorption of As(V) by HFO−NS composites in Cl− free solution and in background solution containing 500 mg/L of Cl−; distribution coefficient (Kd) of As(V) uptake onto HFO−NS composites at 298 K in the presence of different competing anions; distribution of Fe along the diameter on the cross section of the HFO−NS beads; XRD spectra of the HFO−NS composites; Net surface charge (QH) of the NS hosts as a function of pH; zeta potentials of the HFO−NS composites; adsorption isotherm of As(III) by the HFO−NS composites; effect of Cl− on the uptake of As(V) by the NS hosts; aqueous species distribution of As(V) as a function of pH; effect of phosphate on the adsorption of As(V) by the HFO− NS composites. (PDF)

Figure 8. Breakthrough profiles of As(V) removal from synthetic feeding solution by HFO−NS-L, HFO−NS-M, and HFO−NS-H at 298 K, EBCT = 4 min. (Inset) Regeneration of the laden HFO−NS composites at 298 K. Regenerant: 5% NaOH + 5% NaCl, EBCT = 60 min.

maximum contaminant level of As recommended by the World Health Organization drinking water guideline, the breakthrough point was set at 10 μg/L. The effective treatment volumes were ∼2200, 2450, and 2950 bed volumes (BV) for HFO−NS-L, HFO−NS-M, and HFO−NS-H, respectively. Therefore, the size confinement effect of the host pore structure resulted in significant enhancement of the fixed-bed working capacity of the HFO−NS composite. Furthermore, the in situ regeneration of the laden HFO−NS composites was conducted with the binary solution (5 wt % NaOH + 5 wt % NaCl). The results showed that the cumulative desorption efficiency of As(V) from all the three HFO−NS composites reached 94% with 10 BV of the above regenerant. The desorption performances of the three composites were quite similar, suggesting that the size confinement of NS host pore structure did not have adverse effect on the fixed-bed regeneration of the HFO−NS composites. The Fe leaching into the regenerant was below 0.023% for all the three HFO−NS composites, and no obvious mass loss was detected during regeneration, indicating their durability for long-term cyclic utilization.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Tel: +86-25-8968-0390. *E-mail: [email protected]. Tel: +86-25-8968-0390. Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China under Grant No. 51578280. REFERENCES

(1) Arcibar-Orozco, J. A.; Avalos-Borja, M.; Rangel-Mendez, J. R. Effect of Phosphate on the Particle Size of Ferric Oxyhydroxides Anchored onto Activated Carbon: As(V) Removal from Water. Environ. Sci. Technol. 2012, 46, 9577−9583. (2) Auffan, M.; Rose, J.; Proux, O.; Borschneck, D.; Masion, A.; Chaurand, P.; Hazemann, J.; Chaneac, C.; Jolivet, J.; Wiesner, M. R. Enhanced Adsorption of Arsenic onto Maghemites Nanoparticles: As(III) as a Probe of the Surface Structure and Heterogeneity. Langmuir 2008, 24, 3215−3222. (3) Madden, A. S.; Hochella, M. F.; Luxton, T. P. Insights for SizeDependent Reactivity of Hematite Nanomineral Surfaces through Cu2+ Sorption. Geochim. Cosmochim. Acta 2006, 70, 4095−4104. (4) Wang, X.; Li, W.; Harrington, R.; Liu, F.; Parise, J. B.; Feng, X.; Sparks, D. L. Effect of Ferrihydrite Crystallite Size on Phosphate Adsorption Reactivity. Environ. Sci. Technol. 2013, 47, 10322−10331. (5) Yavuz, C. T.; Mayo, J.; Y, W. W.; Prakash, A.; Falkner, J. C.; Yean, S.; Cong, L.; Shipley, H. J.; Kan, A.; Tomson, M. Low-Field Magnetic Separation of Monodisperse Fe3O4 Nanocrystals. Science 2006, 314, 964−967. (6) Yean, S.; Cong, L.; Yavuz, C.; Mayo, J.; Yu, W.; Kan, A.; Colvin, V.; Tomson, M. Effect of Magnetite Particle Size on Adsorption and Desorption of Arsenite and Arsenate. J. Mater. Res. 2005, 20, 3255− 3264.

■

CONCLUSIONS In this study, the size confinement effect of host pore structure on the adsorption of As(V) by the HFO nanoparticles embedded in cross-linked NS anion exchanger was systematically elucidated. The mean diameter of the confined HFO nanoparticles decreased with the decrease of the average pore size of the NS hosts, and the density of surface sites was increased owing to the size-dependent effect. The adsorption capacity of As(V) was elevated via tailoring the pore size of the NS hosts while the adsorption kinetics slightly accelerated with the decrease of pore size in background solution containing 500 mg/L of Cl−. Furthermore, the enhanced adsorption of As(V) by the HFO−NS composites was achieved over a wide pH range from 3 to 10 and in the presence of competing anions including Cl−, SO42−, HCO3−, NO3−, and PO43−. In addition, the fixed-bed working capacity increased from 2200 to 2950 3018

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces (7) Zeng, H.; Singh, A.; Basak, S.; Ulrich, K.; Sahu, M.; Biswas, P.; Catalano, J. G.; Giammar, D. E. Nanoscale Size Effects on Uranium(VI) Adsorption to Hematite. Environ. Sci. Technol. 2009, 43, 1373−1378. (8) Jang, M.; Shin, E. W.; Park, J. K.; Choi, S. I. Mechanisms of Arsenate Adsorption by Highly-Ordered Nano-Structured Silicate Media Impregnated with Metal Oxides. Environ. Sci. Technol. 2003, 37, 5062−5070. (9) Cumbal, L.; SenGupta, A. K. Arsenic Removal Using PolymerSupported Hydrated Iron(III) Oxide Nanoparticles: Role of Donnan Membrane Effect. Environ. Sci. Technol. 2005, 39, 6508−6515. (10) Guo, X.; Chen, F. Removal of Arsenic by Bead Cellulose Loaded with Iron Oxyhydroxide from Groundwater. Environ. Sci. Technol. 2005, 39, 6808−6818. (11) Jang, M.; Min, S.; Kim, T.; Park, J. K. Removal of Arsenite and Arsenate Using Hydrous Ferric Oxide Incorporated into Naturally Occurring Porous Diatomite. Environ. Sci. Technol. 2006, 40, 1636− 1643. (12) Jang, M.; Min, S.; Park, J. K.; Tlachac, E. J. Hydrous Ferric Oxide Incorporated Diatomite for Remediation of Arsenic Contaminated Groundwater. Environ. Sci. Technol. 2007, 41, 3322−3328. (13) Jang, M.; Chen, W.; Cannon, F. S. Preloading Hydrous Ferric Oxide into Granular Activated Carbon for Arsenic Removal. Environ. Sci. Technol. 2008, 42, 3369−3374. (14) Richardson, J. T.; Propp, J. L. Pore Size Effects on Sintering of NiAl2O3 Catalysts. J. Catal. 1986, 98, 457−467. (15) Bore, M. T.; Pham, H. N.; Switzer, E. E.; Ward, T. L.; Fukuoka, A.; Datye, A. K. The Role of Pore Size and Structure on the Thermal Stability of Gold Nanoparticles within Mesoporous Silica. J. Phys. Chem. B 2005, 109, 2873−2880. (16) Tripathi, N.; Yamashita, M.; Uchida, T.; Akai, T. Observations on Size Confinement Effect in B-C-N Nanoparticles Embedded in Mesoporous Silica Channels. Appl. Phys. Lett. 2014, 105, 014106. (17) Martinez de la Hoz, J. M.; Balbuena, P. B. Small-Molecule Activation Driven by Confinement Effects. ACS Catal. 2015, 5, 215− 224. (18) Fu, Q.; Li, W. X.; Yao, Y. X.; Liu, H. Y.; Su, H. Y.; Ma, D.; Gu, X. K.; Chen, L. M.; Wang, Z.; Zhang, H.; Wang, B.; Bao, X. H. Interface-Confined Ferrous Centers for Catalytic Oxidation. Science 2010, 328, 1141−1144. (19) Fu, Q.; Yang, F.; Bao, X. H. Interface-Confined Oxide Nanostructures for Catalytic Oxidation Reactions. Acc. Chem. Res. 2013, 46, 1692−1701. (20) Zhang, F.; Jiao, F.; Pan, X.; Gao, K.; Xiao, J.; Zhang, S.; Bao, X. Tailoring the Oxidation Activity of Pt Nanoclusters via Encapsulation. ACS Catal. 2015, 5, 1381−1385. (21) Chen, L.; Zhao, X.; Pan, B.; Zhang, W.; Hua, M.; Lv, L.; Zhang, W. Preferable Removal of Phosphate from Water Using Hydrous Zirconium Oxide-Based Nanocomposite of High Stability. J. Hazard. Mater. 2015, 284, 35−42. (22) Pan, B.; Han, F.; Nie, G.; Wu, B.; He, K.; Lu, L. New Strategy to Enhance Phosphate Removal from Water by Hydrous Manganese Oxide. Environ. Sci. Technol. 2014, 48, 5101−5107. (23) Pan, B.; Wu, J.; Pan, B.; Lv, L.; Zhang, W.; Xiao, L.; Wang, X.; Tao, X.; Zheng, S. Development of Polymer-Based Nanosized Hydrated Ferric Oxides (HFOs) for Enhanced Phosphate Removal from Waste Effluents. Water Res. 2009, 43, 4421−4429. (24) Pan, B.; Xu, J.; Wu, B.; Li, Z.; Liu, X. Enhanced Removal of Fluoride by Polystyrene Anion Exchanger Supported Hydrous Zirconium Oxide Nanoparticles. Environ. Sci. Technol. 2013, 47, 9347−9354. (25) Zhang, Q.; Du, Q.; Hua, M.; Jiao, T.; Gao, F.; Pan, B. Sorption Enhancement of Lead Ions from Water by Surface Charged Polystyrene-Supported Nano-Zirconium Oxide Composites. Environ. Sci. Technol. 2013, 47, 6536−6544. (26) Zhang, Q.; Pan, B.; Zhang, W.; Pan, B.; Zhang, Q.; Ren, H. Arsenate Removal from Aqueous Media by Nanosized Hydrated Ferric Oxide (HFO)-Loaded Polymeric Sorbents: Effect of HFO Loadings. Ind. Eng. Chem. Res. 2008, 47, 3957−3962.

(27) Hua, M.; Jiang, Y.; Wu, B.; Pan, B.; Zhao, X.; Zhang, Q. Fabrication of a New Hydrous Zr(IV) Oxide-Based Nanocomposite for Enhanced Pb(II) and Cd(II) Removal from Waters. ACS Appl. Mater. Interfaces 2013, 5, 12135−12142. (28) Sherman, D. M.; Randall, S. R. Surface Complexation of Arsenic (V) to Iron(III) (hydr)oxides: Structural Mechanism from Ab Initio Molecular Geometries and EXAFS Spectroscopy. Geochim. Cosmochim. Acta 2003, 67, 4223−4230. (29) Zhang, Q.; Pan, B.; Chen, X.; Zhang, W.; Pan, B.; Zhang, Q.; Lv, L.; Zhao, X. S. Preparation of Polymer-Supported Hydrated Ferric Oxide Based on Donnan Membrane Effect and Its Application for Arsenic Removal. Sci. China, Ser. B: Chem. 2008, 51, 379−385. (30) Korngold, E.; Belayev, N.; Aronov, L. Removal of Arsenic from Drinking Water by Anion Exchangers. Desalination 2001, 141, 81−84. (31) Nie, G.; Pan, B.; Zhang, S.; Pan, B. Surface Chemistry of Nanosized Hydrated Ferric Oxide Encapsulated inside Porous Polymer: Modeling and Experimental Studies. J. Phys. Chem. C 2013, 117, 6201−6209. (32) Pan, B.; Du, W.; Zhang, W.; Zhang, X.; Zhang, Q.; Pan, B.; Lv, L.; Zhang, Q.; Chen, J. Improved Adsorption of 4-Nitrophenol onto a Novel Hyper-Cross-Linked Polymer. Environ. Sci. Technol. 2007, 41, 5057−5062. (33) Abbasian, M.; Ghaderi, A.; Namazi, H.; Entezami, A. A. Preparation of Anion-Exchange Resin Based on Styrene-Divinylbenzene Copolymer Obtained by Suspension Polymerization Method. Polym.-Plast. Technol. Eng. 2011, 50, 1606−1612. (34) Zhang, Q. R.; Pan, B. C.; Zhang, S. J.; Wang, J.; Zhang, W. M.; Lv, L. New Insights into Nanocomposite Adsorbents for Water Treatment: A Case Study of Polystyrene-Supported Zirconium Phosphate Nanoparticles for Lead Removal. J. Nanopart. Res. 2011, 13, 5355−5364. (35) Sun, Y.; Chen, J. L.; Li, A. M.; Liu, F. Q.; Zhang, Q. X. Adsorption of Resorcinol and Catechol from Aqueous Solution by Aminated Hypercrosslinked Polymers. React. Funct. Polym. 2005, 64, 63−73. (36) Waychunas, G. A.; Kim, C. S.; Banfield, J. F. Nanoparticulate Iron Oxide Minerals in Soils and Sediments: Unique Properties and Contaminant Scavenging Mechanisms. J. Nanopart. Res. 2005, 7, 409− 433. (37) Foo, K.; Hameed, B. Insights into the Modeling of Adsorption Isotherm Systems. Chem. Eng. J. 2010, 156, 2−10. (38) Shafqat, M. N.; Pierzynski, G. M. The Freundlich Adsorption Isotherm Constants and Prediction of Phosphorus Bioavailability as Affected by Different Phosphorus Sources in Two Kansas Soils. Chemosphere 2014, 99, 72−80. (39) Ho, Y.; McKay, G. A Comparison of Chemisorption Kinetic Models Applied to Pollutant Removal on Various Sorbents. Process Saf. Environ. Prot. 1998, 76, 332−340. (40) Vadivelan, V.; Kumar, K. V. Equilibrium, Kinetics, Mechanism, and Process Design for the Sorption of Methylene Blue onto Rice Husk. J. Colloid Interface Sci. 2005, 286, 90−100. (41) Ho, Y. Review of Second-Order Models for Adsorption Systems. J. Hazard. Mater. 2006, 136, 681−689. (42) Yu, M.; Li, Y.; Zhang, S.; Li, X.; Yang, Y.; Chen, Y.; Ma, G.; Su, Z. Improving Stability of Virus-like Particles by Ion-Exchange Chromatographic Supports with Large Pore Size: Advantages of Gigaporous Media beyond Enhanced Binding Capacity. J. Chromatogr. A 2014, 1331, 69−79. (43) Ruckenstein, E.; Vaidyanathan, A.; Youngquist, G. Sorption by Solids with Bidisperse Pore Structures. Chem. Eng. Sci. 1971, 26, 1305−1318. (44) Möller, T.; Sylvester, P. Effect of Silica and pH on Arsenic Uptake by Resin/Iron Oxide Hybrid Media. Water Res. 2008, 42, 1760−1766. (45) Pierce, M. L.; Moore, C. B. Adsorption of Arsenite and Arsenate on Amorphous Iron Hydroxide. Water Res. 1982, 16, 1247−1253. (46) Giménez, J.; Martinez, M.; de Pablo, J.; Rovira, M.; Duro, L. Arsenic Sorption onto Natural Hematite, Magnetite and Goethite. J. Hazard. Mater. 2007, 141, 575−580. 3019

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020

Research Article

ACS Applied Materials & Interfaces (47) Raven, K. P.; Jain, A.; Loeppert, R. H. Arsenite and Arsenate Adsorption on Ferrihydrite: Kinetics, Equilibrium, and Adsorption Envelopes. Environ. Sci. Technol. 1998, 32, 344−349. (48) Sparks, D. L. Environmental Soil Chemistry. 2nd ed.; Academic Press: San Diego, CA, 2003, p 133. (49) Meng, X.; Bang, S.; Korfiatis, G. P. Effects of Silicate, Sulfate, and Carbonate on Arsenic Removal by Ferric Chloride. Water Res. 2000, 34, 1255−1261. (50) Wijnja, H.; Schulthess, C. P. Vibrational Spectroscopy Study of Selenate and Sulfate Adsorption Mechanisms on Fe and Al (Hydr)oxide Surfaces. J. Colloid Interface Sci. 2000, 229, 286−297. (51) Lefèvre, G. In Situ Fourier-Transform Infrared Spectroscopy Studies of Inorganic Ions Adsorption on Metal Oxides and Hydroxides. Adv. Colloid Interface Sci. 2004, 107, 109−123. (52) Appelo, C. A. J.; Van Der Weiden, M. J. J.; Tournassat, C.; Charlet, L. Surface Complexation of Ferrous Iron and Carbonate on Ferrihydrite and the Mobilization of Arsenic. Environ. Sci. Technol. 2002, 36, 3096−3103. (53) Brechbühl, Y.; Christl, I.; Elzinga, E. J.; Kretzschmar, R. Competitive Sorption of Carbonate and Arsenic to Hematite: Combined ATR-FTIR and Batch Experiments. J. Colloid Interface Sci. 2012, 377, 313−321. (54) Arcibar-Orozco, J. A.; Josue, D.-B.; Rios-Hurtado, J. C.; RangelMendez, J. R. Influence of Iron Content, Surface Area and Charge Distribution in the Arsenic Removal by Activated Carbons. Chem. Eng. J. 2014, 249, 201−209. (55) Zhang, W.; Singh, P.; Paling, E.; Delides, S. Arsenic Removal from Contaminated Water by Natural Iron Ores. Miner. Eng. 2004, 17, 517−524.

3020

DOI: 10.1021/acsami.5b09832 ACS Appl. Mater. Interfaces 2016, 8, 3012−3020