Arsenic sorption onto an aluminum oxyhydroxide-poly[(4-vinylbenzyl)trimethylammonium chloride] hybrid sorbent

Abstract

In this work, interpenetrated hybrid polymers consisting of aluminum oxyhydroxide (AlOOH) and a quaternary ammonium reagent were synthesized, and their capabilities of arsenite and arsenate retention from aqueous solution were studied. The hybrids were characterized by Fourier transform infrared spectroscopy (FTIR); ¹³C, ²⁷Al and ²⁹Si solid-state nuclear magnetic resonance; scanning electron microscopy (SEM-EDS); FTIR microspectroscopy; and X-ray diffraction (XRD), among other techniques. Arsenic(III) and (V) sorption experiments were conducted under different experimental conditions (i.e., isotherms, kinetics, and selectivity). The hybrid sorbent shows a poor affinity toward arsenite; however, the hybrids with higher AlOOH exhibited improved As(III) sorption without preoxidation. The opposite trend was observed for arsenate. Kinetic experiments showed that the sorption of arsenate occurs more rapidly than that of arsenite.

---

1. Introduction

Groundwater constitutes 97% of the freshwater in the world and becomes an important freshwater source during periods without rain. Currently, millions of people depend on aquifers, and 40% of the world's food is produced via irrigated agriculture, which depends on groundwater. Arsenic (organic and inorganic) exhibits among the highest toxicities, and the World Health Organization (WHO) thus recommends a maximum contaminant level (MCL) of 10 μg L⁻¹. There are a large number of water treatment technologies for arsenic removal discussed in the literature.¹ Among them, adsorption and ion exchange are the most commonly employed technology because of their relatively low cost and simple implementation and operation to achieve high removal efficiencies. Despite the advantages of adsorption processes, high removal efficiency is achieved when the predominant species is pentavalent arsenic. Owing to the natural circumneutral pH of water, arsenate occurs in its anionic form, which allows for removal by ion exchange or electrostatic interactions with an adsorbent, such as strong base anion exchange resins. However, in natural effluents, arsenic is also found in oxidation state 3+ as arsenite, a non-charged species at natural pH, which decreases the efficiency of sorption by ion exchange.² Moreover, arsenite is the dominant form in typical anaerobic groundwater and is more toxic than the oxidized species.³

Aluminum-based sorbents include activated alumina, gibbsite, aluminum hydroxide precipitated from aluminum salts, and layered double hydroxides, with activated alumina (AA) being the most frequently studied sorbent.^4–7 Sign et al. studied the adsorption of As(III) and found that AA was a suitable adsorbent for the removal of As(III) from drinking water, with pH and contact time being the two most relevant parameters in the sorption.⁴ Recently, Lescano et al. investigated the adsorption of arsenate and arsenite using different metal oxides, including AA. The results showed that As(V) is more easily adsorbed than As(III) for AA.⁸ Aluminum oxyhydroxide (boehmite oxide types) has also been used for arsenate removal.^9,10 Regarding the mechanism involved in the sorption of As(III and V) onto aluminum-based sorbent, Goldberg and Johnston indicated that arsenate forms inner-sphere surface complexes on amorphous aluminum oxides, whereas arsenite forms outer-sphere surface complexes.¹¹

Considering the sorbent properties of aluminum-based sorbents for arsenate and arsenite, this work aims to combine the sorption properties of strong base anion exchange resins and aluminum oxyhydroxide. Here, interpenetrating hybrid polymer/inorganic oxide networks (IPNs) made of a covalently crosslinked network and an inorganic oxide material were synthesized to investigate the effectiveness of sorption properties for As(III and V). In contrast to other composite sorbents in which (nano)particles are embedded in a porous polymer matrix,¹² in this work, both classes of material networks are covalently connected.¹³ For this purpose, hybrid sorbents were synthesized using a (4-vinylbenzyl)trimethylammonium chloride monomer and aluminum tri-sec-butoxide as starting materials. The monomer bears a quaternary ammonium group attached to an aromatic group in the same way as anion exchange resins (i.e., IRA-400), whereas the inorganic precursor is a well-known compound that suffers a sol–gel process, forming a hydrated aluminum oxide.¹⁴ We expect that the polymeric part remove arsenate by ion exchange mechanisms, while the inorganic achieve the arsenic uptake through a different mechanism, forming inner- and outer-sphere complexes. This type of interaction could provide selectivity to the sorbent and allowing the one-step sorption of arsenite as well (without pre-oxidation). The hybrid sorbents were prepared by varying the mole ratio of the aluminum oxide precursor and the monomer to provide different surface compositions, and their effect on arsenic sorption was investigated.

2. Materials and methods

2.1 Materials

The interpenetrated hybrids were obtained using aluminum tri-sec-butoxide (Al(O-s-Bu)₃, 97%, Merck, Darmstadt, DE) as starting material. The [3-(methacryloyloxy)propyl]trimethoxysilane (MPS, 95%, Aldrich) was used as a coupling reagent, and the 4-(vinylbenzyl)trimethylammonium chloride was used as a monomer (ClVBTA, 99%), both of which were provided by Sigma-Aldrich, St. Louis, MO, USA. Acetylacetone (acac, 99%, Merck, Darmstadt, DE) and benzoyl peroxide (BPO, 99%, Sigma-Aldrich, St. Louis, MO, USA) were used as a retardant compound initiator, respectively. The other reagent used in the synthesis and sorption experiments included 2-butanol (anhydrous, >99.5%) as a synthesis solvent, NaAsO₂ (0.05 mol L⁻¹), Na₃AsO₄ (1000 mg L⁻¹), potassium chloride, hydrochloric acid (37%) and sodium hydroxide (99%), all of which were provided by Merck, Darmstadt, DE.

2.2 Synthesis of hybrid materials

The synthesis of the hybrid was previously reported.^13,15 The inorganic precursor Al(O-s-Bu)₃, ClVBTA, MPS, acac, and BPO (2 mol% based to ClVBTA) were dissolved in 10 mL of 2-butanol. The amount of ClVBTA was maintain fixed to 10 mmol, MPS/Al(O-s-Bu)₃ and MPS/acac mole ratios were 1. To obtain sorbents with different organic–inorganic content Al(O-s-Bu)₃/ClVBTA + Al(O-s-Bu)₃ mole ratios of 0.8, 0.6, 0.4, 0.2 were used. For example, to prepare the hydrid of mole ratio = 0.8 the following amounts were used: 40 mmol Al(O-s-Bu)₃, 10 mmol ClVBTA, 40 mmol MPS, 40 mmol acac, 0.2 mmol BPO and dissolved in 10 mL of 2-butanol. Subsequently, the dissolution was degassed with N₂ and polymerized at 80 °C for 24 h. Afterward, the polymeric dissolution was transferred to a Teflon beaker and cooled to room temperature, and deionized water (5 mL) was added and mechanically mixed. The mixture was left for 8 h at room temperature and was later dried at 80 °C in the oven until a fine white powder was formed. All composites were washed with deionized water under magnetic stirring for 12 h and dried at 70 °C to remove the unreacted reagents. For comparison purposes, the control compounds were prepared; namely, AlOOH, was obtained following the same procedure but without the addition of the monomers, whereas the copolymer (P(ClVBTA-co-MPS)) was obtained by radical polymerization of the ClVBTA and MPS in the presence of a protic solvent to favor the hydrolysis and condensation of methoxysilane groups leading to the formation of crosslinked polymer. The sorbents were grinded and sieved to obtain a particle size of 100 μm × 180 μm for sorption experiments. The hybrids hereinafter will be referred as IPN(φ), where φ represents the mole ratio (Al(O-s-Bu)₃/ClVBTA + Al(O-s-Bu)₃).

2.3 Sorption experiments

Arsenic sorption studies were performed under different experimental conditions. The effect of the pH, isotherms, sorption kinetics, and SO₄²⁻ concentration were studied as a function of φ. The sorption experiments were carried out following this general procedure unless stated otherwise: 30 mg of sorbent was combined with 5 mL of arsenate solution in a thermo-regulated water bath at 140 rpm for 24 h. The samples were then filtered, and the arsenic concentrations were determined. The sorption capacities were calculated by the difference between the initial and final arsenic concentrations. For the study of the pH effect, arsenic solutions of 50 mg L⁻¹ were adjusted to pH values of 2, 4, 6, 8, 10 and 12 and combined with the hybrid. The isotherms were conducted at 25 °C while keeping the amount of resin constant and varying the concentration of the arsenic in the range of 10–1000 mg L⁻¹.

The effect of time on arsenic retention was also studied. In 500 mL of arsenic (100 mg L⁻¹) on a plate heater at 25 °C under constant agitation of 140 rpm, 2.0 g of hybrid sorbent was added, and 5.0 mL of the samples were then withdrawn at different time intervals (1 to 240 min). The samples were filtered using a Millipore filter with a pore size of 0.2 microns. To evaluate the selectivity of the hybrids, a set of arsenate solutions containing sulfate anions was prepared. The arsenic concentration was maintained at 100 mg L⁻¹, and the SO₄²⁻ concentration varied from 50–500 mgL⁻¹.

Elution studies were conducted by contacting the selected sorbents with concentrated arsenic solution (500 mg L⁻¹) following the same procedure described above, later the exhausted sorbents were dispersed in HNO₃ 1 mol L⁻¹ or NaOH 1 mol L⁻¹ solution to elute the arsenic oxyanions for 24 h at room temperature.

2.4 Physicochemical characterization

The morphology and structure of the hybrids were elucidated by infrared spectroscopy (FTIR, Perkin Elmer 1760-X spectrometer using a range of 4000–400 cm⁻¹ and KBr pellets) and scanning electron microscopy (SEM, SEM-PROBE CAMECA model SU-30 equipped with an energy dispersive X-ray device, EDX). Nuclear magnetic resonance of ¹³C, ²⁷Al, and ²⁹Si nucleus (Bruker Ascend™ 400 MHz CP-MAS solid state NMR) and FTIR microspectroscopy (Perkin Elmer, Spotlight 400 FTIR Imaging System model Spectrum Frontier mid-IR spectral range, reflectance mode) were used to evaluate the band at 1385 cm⁻¹ to obtain a map of its distribution in the hybrid surface, and the obtained images were analyzed using the Spectrum (version 10.03.06.0100) and SpectrumIMAGE (version R1.7.1.0401) software packages. X-ray diffraction (XRD, RIGAKU, Geigerflex-model diffractometer (CuKα tube, 40 kV, 20 mA) with a range of 2° to 90° 2θ at 1° min⁻¹) was also performed. According to the concentrations of the arsenic in solutions, this element was determined using flame or hydride generation atomic absorption spectrometry (Perkin Elmer, Pinaacle 900F) with detection limits of 150 μg L⁻¹ (flame) and 0.03 μg L⁻¹ (hydride generation).

3. Results and discussion

3.1 Hybrid sorbent characterization

The synthesis procedure used to obtain the hybrid consisted of a two-stage process. The first stage corresponds to a radical copolymerization between the monomers, ClVBTA and MPS. The presence of anhydrous solvent (2-butanol) avoided the reaction of methoxysilane groups both between themselves and with the oxide network, whereas the addition of acetylacetone retards hydrolysis and condensation reaction of aluminum alkoxides.¹⁶ The addition of deionized water after the polymerization drives the formation of the inorganic network and also allows the formation of Si–O–Al bonds by hydrolysis reaction with the inorganic precursor, setting up the covalent bond between organic and inorganic phases. The second stage consisted of a sol–gel reaction that comprises the hydrolysis of the aluminum alkoxide to form hydroxyl groups followed by condensation to polymerize with the oxide network. The hydrolysis of aluminum tri-sec-butoxide in excess of water gives aluminum oxyhydrate (AlOOH), which contains octahedrally coordinated aluminum, and, when annealed, loses water to form Al₂O₃.¹⁷ After the hybrids were obtained, the materials were washed with distilled water to remove the unreacted reagents.

With the aim of confirming the structure of the IPN hybrids, the spectroscopic techniques of FTIR, solid-state ²⁹Si, ²⁷Al and ¹³C-NMR analyses were performed. Fig. 1 shows the FTIR spectra of synthesized hybrids and control sorbents. The P(ClVBTA-co-MPS) compound exhibited the main vibrational bands ascribed to the functional groups used in the copolymerization—namely, 1711 cm⁻¹ of the carbonyl group, 1634 cm⁻¹ corresponding to the stretching of the carbon double bond of the aromatic ring, and 1471 cm⁻¹ of the ammonium group. The signal at 1110 cm⁻¹ is ascribed to the Si–O vibration mode. However, the aluminum-based control sorbent exhibited the signals at 1016 cm⁻¹ and 983 cm⁻¹, which are attributed to the symmetric bending vibration mode of Al–O–H, whereas the signals at 776 and 529 cm⁻¹ are associated with the stretching of tetrahedral and octahedral coordinated aluminum.¹⁸ The spectrum of IPN(0.2) exhibited vibration bands at 1705, 1637, and 1475 cm⁻¹, ascribe to the stretching vibrations of carbonyl, C=C of the aromatic ring, and quaternary ammonium group, respectively. For the hybrids with higher mole ratios, the signals associated with the polymer phase (1637 and 1475 cm⁻¹) tend to disappear or to be overlapped by new signals. For instance, the bands at 1589 and 1529 cm⁻¹ are attributed to O–H vibration of water molecules physically adsorbed on the oxide, and the bands at 1385 and 1012 cm⁻¹ are attributed to the bending vibration hydroxyl groups on metal oxide. The observed vibrational bands confirm the existence of the functional groups in the hybrid material, and the change of intensities are in agreement with the different mole ratio used in the feed.

Fig. 1 Infrared characterization of hybrid materials (1) spectrum of control sorbent, (2) spectrum of interpenetrated hybrids (a) IPN(0.2), (b) IPN(0.4), (c) IPN(0.6), and (d) IPN(0.8).

The ¹³C-NMR spectra of IPNs and control polymer are shown in Fig. 2a. The spectrum of copolymer exhibits the signals corresponding to both monomers used in the synthesis. The signals in the region of δ = 10.4–22.7 ppm are attributed to alkane carbons of the main chain polymer and carbons of the propyl chain of MPS, whereas the signal at 68.7 ppm is ascribed to the carbon of methoxy groups bound to silicon. The carbonyl appears at 178.1 ppm. For the ClVBTA pending group, the signals at δ = 52.9, 127, and 133.1 ppm are attributed to methyl groups of ammonium and the alkene carbons of the aromatic ring in ClVBTA. All mentioned signals were observed for the spectra of IPNs, with a slight increase in the intensity of the signals ascribed to the MPS compound. This result could be a consequence of the greater content added in the synthesis. The spectroscopic characterization confirms the structure of the hybrid. ²⁹Si-NMR analysis provides information about how the methoxysilane groups react with the inorganic networks and allows confirmation of whether the organic and inorganic networks are covalently bound displaying anchoring signals called T¹, T², and T³ ascribed to single silanol, OSi(OH)₂R (−50 to −45 ppm); geminal silanol, O₂Si(OH)R (−55 ppm); and T³ siloxane, O₃SiR; structures (−70 to −60 ppm), respectively.¹⁹ Fig. 2b displays the ²⁹Si-NMR of IPN(φ) exhibiting the presence of the three anchoring signals in all hybrids, with the T² and T³ bridges being the most abundant. The signal observed at −37.1 ppm in the spectrum of IPN(0.8) is ascribed to unreacted methoxysilane groups of the coupling reagent. These results confirm that both inorganic and the organic polymer components are covalently bound by means of the organosilane reagent.²⁰ The ²⁷Al-NMR is very sensitive to the coordination number of the aluminum atoms; hexa-coordinated aluminum atoms (Al^VI) present peaks in the range of 3 to 7.5 ppm, whereas the chemical shifts of tetra-coordinated aluminum atoms (Al^IV) have been found in a wider region from 35.4 to 66.5 ppm. For penta-coordinated aluminum atoms (Al^V) in alkoxides, the chemical shifts seem to range from 29 to 44 ppm.²¹ For the present hybrids, it is observed in Fig. 2c that IPN(0.8) and IPN(0.6) exhibit signals in the range of 0 to 10 ppm, which are ascribed to octahedrally coordinated aluminum. For IPN(0.4) and IPN(0.2), signals between 60 and 70 ppm are observed, indicating tetrahedrally coordinated aluminum. A signal of approximately 14 ppm is also observed and can be associated with the Al^VI structures.

Fig. 2 Solid-state NMR analyses for IPN(φ) hybrids. (a)¹³C NMR, (b) ²⁹Si NMR and (c) ²⁷Al NMR.

Powder XRD patterns of the synthesized materials are shown in Fig. S1.† The XRD pattern of the AlOOH control material showed several diffraction peaks attributed to mixed aluminum oxyhydroxide phases, as has been reported by Lee et al.²² The intensity and sharpness of the diffraction peaks show that the formed Bayerita and Gibbsita phases are well crystallized. However, the increase of polymer loading in the hybrid composition resulted in the formation of amorphous materials. For IPN(0.6) and IPN(0.8), the aluminum oxyhydroxide phases are poorly crystalline, and as a result, it has not been possible to undertake any detailed structural analysis of them (Fig. S1b†).

A detailed study of the surface characteristics of the adsorbent can provide important information that allows us to understand the sorption results. Considering that the hybrid sorbent is formed in two phases, the quantification and distribution on the surface is an important issue. For this purpose, the sorbent was characterized by SEM-EDX and FTIR microspectroscopy. Fig. S2† displays the scanning electron microscopy images of the hybrids. With the aim of confirming the change in the surface composition, the concentration of elements was determined by energy dispersive X-ray elemental analyses.

The sorbent specimens at the xerogel state exhibit a uniform surface without evidence of roughness or porosity. The figures also display the EDS spectrum corresponding to a section surface area of the sorbent particles and a table with the wt% for representative elements. It is observed that the elements ascribed to the polymer phase C and Cl increase their concentration on the surface as φ decreases, which is clearly attributed to the high polymer content. Conversely, Al and Si, corresponding to inorganic precursor and coupling reagent, respectively, increase their concentration with φ. These results reveal the presence of polymer and oxide phases on the surface of the IPN hybrids and that their surface concentration changes with the variation of the mole in the synthesis. With the aim to provide more insight into the distribution of the phases on the surface of the hybrid FTIR microspectroscopy analyses were performed. This characterization technique allows determination of the infrared spectrum of the surface image; by plotting the absorbance intensity at a determined wavenumber as a function of position, it is then often possible to create distribution maps for the selected frequency and provide insights into the distribution of a particular functional group on the surface of the adsorbent.²³ Fig. 3 shows the results of infrared microspectroscopy of the hybrid samples evaluating the absorbance of the band 1385 cm⁻¹, which is attributed to the inorganic phase—specifically, to the deformation vibration of Al–O–H.¹⁸ IPN(0.8) and IPN(0.6) exhibit high concentrations of inorganic phase without a homogenous dispersion, whereas IPN(0.4) reveals a better dispersion but a lower concentration of the aluminum oxide phase. These results are consistent with those obtained from EDS analysis. The presence of oxide on the surface can be explained by the procedure used to obtain the hybrid. During the second step (sol–gel), the product formed in the first step is swollen with the added water, leading to hydrolysis and condensation reactions of the inorganic precursor.

Fig. 3 Surface FTIR microspectroscopy images of the hybrids evaluated at 1385 cm⁻¹. (a) φ = 0.8, (b) 0.6, (c) 0.4 and (d) 0.2.

3.2 Sorption studies

3.2.1 Equilibrium sorption experiments. Inorganic oxides are capable of varying the pH in solutions and then changing the speciation of arsenic anions. Additionally, the evaluation of the pH mixture at equilibrium corresponds to a simple way to estimate the point of zero charge (PZC) of a sorbent. Fig. S3† shows the pH drift after equilibrium, showing that all of the sorbent exhibits the same behavior. This plots are useful for a quickly determination of the pHpzc, the pH at which the curve crosses the pH_initial = pH_final can be taken as pHpzc.²⁴ Then, for the present system the pHpzc should be located between pH 4–6. When pH > pHpzc, the terminating hydroxyl groups at the surface deprotonate and reduce the solution pH by diminishing the OH⁻ concentration. According to these results, the pH value of 6.0 was chosen to perform the arsenic sorption experiments in equilibrium.

Fig. 4 displays the adsorption isotherms obtained at 25 °C carried out for all synthesized hybrid and control sorbents. Arsenite sorption clearly shows a low affinity of arsenite toward the sorbent; however, the hybrids with a major inorganic component exhibit higher sorption. As expected, the copolymer control sorbents containing only ammonium groups display the lowest arsenite sorption. It is well known that ammonium-based sorbents do not retain arsenite owing to the absence of charge at the pH of natural effluents and can retain arsenate ions only by ion exchange.²⁵

Fig. 4 Arsenic sorption isotherms using the hybrid sorbent, (a) arsenite and (b) arsenate.

Arsenate removal exhibited the opposite trend; the sorbents with higher polymer content showed the highest sorption and a better affinity in general. In this case, the sorbent with lower φ tended to be more similar to ammonium-based sorbent. Previous investigations reported that arsenate has long been known to be strongly adsorbed by aluminum(oxy)hydroxides, whereas arsenite is considerably less readily adsorbed.²⁶ Moreover, arsenite and arsenate can interact with Al-based sorbent by forming monodentate mononuclear, bidentate mononuclear and bidentate binuclear surface complexes.²⁷

For quantification purposes, the experimental data were fitted to the Langmuir and Freundlich isotherm models. The Langmuir isotherms are valid for the monolayer sorption onto a surface containing a finite number of identical sites and are commonly applied to homogeneous surfaces with non-interaction between the adsorbed molecules. The Langmuir isotherm is given as

where C_e is the equilibrium concentration (mg L⁻¹), q_e is the amount of arsenic removed at equilibrium (mg g⁻¹), and q_m and b are the Langmuir constants related to the sorption capacity and energy of sorption, respectively. The Freundlich isotherm is derived to model the multilayer adsorption and the adsorption on heterogeneous surfaces. The form of the Freundlich isotherm model is given as

q_e = k_f(C_f)^1/n (2)

where q_e is the amount of adsorbed analyte per unit weight of the solid phase at equilibrium concentration C_e, k_f is the Freundlich constant related to the sorption capacity, and 1/n represents the sorption intensity. The Freundlich constant n should have a value in the range of 1 to 10 for the classification of adsorption to be favorable.

Table 1 displays the isotherm parameters obtained using a nonlinear fit. For arsenite sorption, it was observed that Freundlich isotherms exhibit the best fit. The k_f values revealed that only Al-based control sorbent showed the highest sorption capacity, whereas k_f for the hybrids significantly decreased. Despite the value for φ = 4.0, k_f decreased with φ, which could be attributed to the increase of the polymer phase, which does not interact favorably with arsenite anions. The arsenate sorption showed a better agreement with the Freundlich isotherm as well, but in this case, the Langmuir isotherm also shows an acceptable concordance with the experimental data. Langmuir sorption parameters clearly reveal that as the polymer content increases, the arsenate sorption also increases. Similarly, k_f increases as φ decreases. This result could be attributed to the quaternary ammonium groups that retain arsenate by the ion exchange mechanism. A comparison of the sorption capacities for arsenite and arsenate achieved for the present sorbent were compared with those values reported in literature for diverse nanocomposite and hybrid sorbents (Table S1†).

Table 1 Langmuir and Freundlich isotherm parameters (non-linear fitting)

x	Langmuir			Freundlich
	qm (mg g^−1)	b (L g^−1)	r^2	1/n	kf (mg g^−1)	R^2
a For this sorbents was not possible to obtain the Langmuir fitting (too low correlation coefficient).
Arsenite
Al-control	123.6	0.0144	0.9716	0.62	4.384	0.9835
0.8	197.6	2.58 × 10^−3	0.9587	1.04	0.442	0.9589
0.6^a	—	—	—	1.54	0.037	0.9620
0.4	81.8	6.23 × 10^−3	0.9423	0.78	0.90	0.9806
0.2^a	—	—	—	1.78	0.025	0.9827
Polymer control	—	—	—	1.64	0.012	0.9098
Arsenate
Al-control	1.13	0.024	0.9053	0.52	3.05	0.9378
0.8	38.1	0.28	0.8956	0.29	10.56	0.9177
0.6	55.2	2.84	0.8727	0.17	29.40	0.9059
0.4	80.6	1.61	0.8954	0.21	40.06	0.9559
0.2	102.8	0.31	0.9556	0.33	27.28	0.9060
Polymer control	93.8	0.15	0.9762	0.41	16.91	0.9259

---

3.2.2 Kinetic experiments. Kinetic studies were conducted to evaluate the effect of time on arsenic sorption. The studies were carried out for the hybrids that exhibited the highest sorption arsenite/arsenate capacities—namely, φ = 0.2 and φ = 0.8 for arsenate and arsenite, respectively (see Fig. 5). The kinetic curves were also obtained for control sorbents for comparison purposes. The curves for arsenite sorption revealed slower kinetics than arsenate and a lower maximum sorption. As expected from isotherm experiments, IPN(0.8) and Al-control showed similar arsenite sorption capacities that were higher than the polymer control. In contrast, the arsenate was quickly adsorbed by IPN(0.2) and the polymer control, with a sorption capacity much higher than the AlOOH blank.

Fig. 5 Sorption kinetic experiments (a) arsenite, (b) arsenate.

The pseudo-second-order (PSO) equation has been widely used owing to the excellent fit of experimental data for the entire sorption period of many systems.²⁸ The experimental data were studied using a nonlinear fitting of the PSO model given by the following equation:²⁹

where q_e and q are the arsenic amounts adsorbed (mg g⁻¹) at equilibrium and at time t, respectively, and k₁ (min⁻¹) and k₂ (g mg⁻¹ min⁻¹) are the rate constants of sorption.

Table 2 shows the PSO parameters and confirms the good agreement of the experimental data with the kinetic model. The pseudo-second-order rate expression is useful to describe chemisorption processes. These processes involve valency forces through the sharing or exchange of electrons between the adsorbent and adsorbate as covalent forces, and ion exchange.³⁰ The rate constant (k₂) of arsenite sorption using the Al-control exhibited the highest value, whereas k₂ for arsenate has the same order of magnitude. Regarding the sorption capacity (q_e), the values confirm what was explained above.

Table 2 Pseudo-second-order model parameters

	As(III)			As(V)
	Polymer control	Al-control	HC(0.8)	Polymer control	Al-control	HC(0.2)
qe (mg g^−1)	3.5	15.5	16.1	34.5	6.97	37.6
k2 (g mg^−1 min^−1)	0.014	0.032	0.008	0.032	0.022	0.083
R^2	0.7872	0.9315	0.9732	0.9984	0.9441	0.9983

---

3.2.3 Selectivity. Sulfate anion is a recognized interferer for arsenic sorption, especially in ion exchange processes, where the concentration of SO₄²⁻ should be <20 mg L⁻¹ to carry out an efficient process.³¹ The selectivity sequence for strong base anion exchange resins is SO₄²⁻ > HAsO₄²⁻ > CO₃²⁻ ≈ Cl > H₂AsO₄²⁻, with sulfate causing the greatest interference. The interference is a consequence of both structural and charge similarities with the arsenate anion.³² With the aim to evaluate the selectivity of the sorbent, experiments were conducted in the presence of sulfate ions ranging from 50 to 500 mg L⁻¹. Fig. 6 compared the sorption performance for arsenite and arsenate using the sorbent with the best results. Arsenite exhibited a dramatic decrease of sorption in the presence of sulfate. Wijnja et al. studied the adsorption of sulfate onto aluminum (hydr)oxide surfaces by Raman and ATR-FTIR, indicating that both inner- and outer-sphere surface complexes of SO₄²⁻ occur.³³ In the case of sorbent, arsenite sorption is mainly carried out on the aluminum moiety, and the presence of SO₄²⁻ prevents the arsenite sorption. Typically, arsenate sorption onto polymer control exhibits a constant diminishing as the sulfate concentration increases; however, for IPN(0.2), the sorption remains constant up to 150 mg L⁻¹ of SO₄²⁻, revealing that this sorbent possesses selectivity for arsenate uptake. In a previous work, arsenate sorption on aluminum oxide in the presence of SO₄²⁻ demonstrated that sorption is slightly affected when sulfate concentration ranged between 0 and 250 mg L⁻¹.³⁴ Consequently, these results demonstrated the selectivity properties provided by the metal oxide phase, especially for arsenate sorption, as a consequence of the mechanism involved.

Fig. 6 Arsenic sorption in the presence of SO₄²⁻ anions.

3.2.4 Sorption mechanism and elution experiments. According to the surface complex model theory, metal hydroxyl groups on the surface of many metal oxides are the most abundant and provide reactive adsorption sites for anions. These vibrational bands can be detected by FTIR spectroscopy, and their change after sorption can be determined. Fig. 7 displays the FTIR of hybrid φ = 0.4 and φ = 0.8 before and after arsenic sorption. The FTIR of the hybrid φ = 0.4 loaded with arsenate and arsenite ions shows the characteristic bands associated with the hybrid structure and the appearance of strong new bands at 1381 cm⁻¹ and in the region of 890–870 cm⁻¹, which correspond (both bands) to the stretching of As–O bonds.³⁵ Because this hybrid has a higher content of PClVBTA, the sorption of arsenite is carried out mainly by ion exchange, and barely any evidence of sorption could be detected. Conversely, for the hybrid major content of AlOOH (φ = 0.8), the FTIR shows significant changes in the spectrum. After arsenic(III and V) sorption, the vibration bands of 1589 and 1529 cm⁻¹ attributed to water molecules physically adsorbed on the oxide significantly decreased in intensity, as did the band at 1385 cm⁻¹, which was attributed to the bending vibration hydroxyl groups on metal oxide. These results indicate the formation of inner sphere complexes of arsenic species onto the surface of AlOOH phase.

Fig. 7 FTIR of hybrid sorbent before and after arsenic sorption.

Arsenic elution is a critical consideration in order to achieve a successful desorption process and restore the sorbent to be reused in a sorption process. This process must be performed when the sorbent is exhausted by the use of a suitable eluent, usually an acidic or alkaline media. Fig. S4† shows the elution percentages achieved for a selected exhausted hybrid sorbent using nitric acid and sodium hydroxide dissolutions, the results reveal a better elution of arsenate in acidic and alkaline media than arsenite. Probably, due the mainly electrostatic interaction between arsenate ions and IPN(0.2) sorbent (mostly composed of quaternary ammonium polymer), in alkaline media the hydroxyl groups can effectively displace the arsenate ions, while in acidic media the arsenic species can turn into uncharged species decreasing the sorption.

4. Conclusions

By suitable synthesis controlling the mole ratio of inorganic precursor and monomer, interpenetrating hybrid sorbents with different surface compositions were obtained. The sorbents with higher aluminum oxyhydroxide content exhibited moderate arsenite sorption, whereas those with higher polymer content showed better sorption capacities for arsenate. The results suggest that the mechanism involved in the sorption process is the formation of inner or outer sphere complexes for arsenite onto aluminum oxyhydroxide, whereas for arsenate, the mechanism is due to the ion exchange process on the polymer phase and the formation of a complex on the inorganic phase. The results suggest that interpenetrated hybrids could be an alternative for one-step sorption of arsenite/arsenate-contaminated waters.

Acknowledgements

The authors thank FONDECYT Initiation No 11121291, FONDECYT Postdoc No 3140130, FONDECYT No 1150510, and CIPA-CONICYT Regional R08C1002.

References

1. D. Mohan and J. C. U. Pittman, J. Hazard. Mater., 2007, 142, 1–53.
2. L. Dambies, Sep. Sci. Technol., 2004, 39, 603–627.
3. B. K. Mandal and K. T. Suzuki, Talanta, 2002, 58, 201–235.
4. T. S. Singh and K. K. Pant, Sep. Purif. Technol., 2004, 36, 139–147.
5. S. S. Tripathy and A. M. Raichur, Chem. Eng. J., 2008, 138, 179–186.
6. S. Bang, M. Pena, M. Patel, L. Lippincott, X. Meng and K.-W. Kim, Environ. Geochem. Health, 2011, 33, 133–141.
7. J. Dou, W. Qin, A. Ding, E. Xie, L. Zheng and W. Ding, Environ. Technol., 2015, 1–8,  DOI:10.1080/09593330.2015.1046950.
8. M. R. Lescano, C. Passalía, C. S. Zalazar and R. J. Brandi, J. Environ. Sci. Health, Part A: Toxic/Hazard. Subst. Environ. Eng., 2015, 50, 424–431.
9. F. Ogata, N. Kawasaki, T. Nakamura and S. Tanada, J. Colloid Interface Sci., 2006, 300, 88–93.
10. B. Doušová, V. Machovič, D. Koloušek, F. Kovanda and V. Dorničák, Water, Air, Soil Pollut., 2003, 149, 251–267.
11. S. Goldberg and C. T. Johnston, J. Colloid Interface Sci., 2001, 234, 204–216.
12. S. Sarkar, P. K. Chatterjee, L. H. Cumbal and A. K. SenGupta, Chem. Eng. J., 2011, 166, 923–931.
13. C. H. Campos, B. F. Urbano and B. L. Rivas, Mater. Lett., 2014, 131, 198–202.
14. J. Liu and J. C. Berg, J. Mater. Chem., 2007, 17, 4430–4435.
15. C. H. Campos, B. F. Urbano and B. L. Rivas, Composites, Part B, 2014, 57, 1–7.
16. Ł. John and P. Sobota, in Ceramic Materials, ed. W. Wunderlich, InTech, Rijeka, 2010, ch. 4, pp. 69–86.
17. S. Kureti and W. Weisweiler, J. Non-Cryst. Solids, 2002, 303, 253–261.
18. A. Boumaza, L. Favaro, J. Lédion, G. Sattonnay, J. B. Brubach, P. Berthet, A. M. Huntz, P. Roy and R. Tétot, J. Solid State Chem., 2009, 182, 1171–1176.
19. F. Bauer, H.-J. Glasel, U. Decker, H. Ernst, A. Freyer, E. Hartmann, V. Sauerland and R. Mehnert, Prog. Org. Coat., 2003, 47, 147–153.
20. G. Kickelbick, Prog. Polym. Sci., 2003, 28, 83–114.
21. F. Babonneau, L. Coury and J. Livage, J. Non-Cryst. Solids, 1990, 121, 153–157.
22. J. S. Lee, H. S. Kim, N.-K. Park, T. J. Lee and M. Kang, Chem. Eng. J., 2013, 230, 351–360.
23. R. Bhargava, S. Q. Wang and J. L. Koenig, Adv. Polym. Sci., 2003, 163, 137–191.
24. Y. Yang, Y. Chun, G. Sheng and M. Huang, Langmuir, 2004, 20, 6736–6741.
25. L. Dambies, Sep. Sci. Technol., 2005, 39, 603–627.
26. J. G. Hering and S. Dixi, in Advance in Arsenic Research, ed. P. A. O'Day, D. Vlassopoulos, X. Meng and L. G. Benning, American Chemical Society, Washington, 2005, ch. 2.
27. Y. Arai, E. J. Elzinga and D. L. Sparks, J. Colloid Interface Sci., 2001, 235, 80–88.
28. Y. S. Ho and G. McKay, Process Biochem., 1999, 34, 451–465.
29. K. V. Kumar, J. Hazard. Mater., 2006, 137, 1538–1544.
30. Y.-S. Ho, J. Hazard. Mater., 2006, 136, 681–689.
31. U. S. EPA, Arsenic Treatment Technologies for Soil, Waste, and Water, U.S. EPA/National Service Center for Environmental Publications, 2002.
32. D. A. Clifford and G. L. Ghurye, in Environmental Chemistry of Arsenic, ed. W. T. Frankenberger, Marcel Dekker, Inc., New York, 1st edn, 2002, ch. 9.
33. H. Wijnja and C. P. Schulthess, J. Colloid Interface Sci., 2000, 229, 286–297.
34. J. Youngran, M. Fan, J. Van Leeuwen and J. F. Belczyk, J. Environ. Sci., 2007, 19, 910–919.
35. G.-S. Zhang, J.-H. Qu, H.-J. Liu, R.-P. Liu and G.-T. Li, Environ. Sci. Technol., 2007, 41, 4613–4619.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra01230b

---