Abstract
Abstract
Use of layered double hydroxides (LDHs) in the environmental field is gaining popularity due to their potential to sorb toxic anions, attributed to their large surface area, high anion exchange capacity, and good thermal stability. In this study, four different LDHs (i.e., Cu-Al-, Mg-Al-, Mg-Fe-, and Zn-Al-LDH) were synthesized to select one or more efficient sorbents, capable of removing arsenite [As(III)] from contaminated waters. In particular, we studied the following: (1) X-ray diffraction patterns and specific surface area of the synthesized LDHs; (2) sorption isotherms of As(III) at pH 7.0; and (3) sorption of As(III) on LDHs, in the presence of inorganic anions [carbonate (CO3), chloride (Cl), fluoride (F), phosphate (PO4), sulfate (SO4)] commonly present in aquatic environments. The poorly crystalline LDHs (i.e., Cu-Al-LDH and Mg-Fe-LDH) sorbed greater amounts of As(III) than the well-crystalline LDHs (i.e, Mg-Al-LDH and Zn-Al-LDH). The efficiency of the competing anions at inhibiting As(III) sorption by the LDHs was Cl ≤ F < SO4 << CO3 << PO4, regardless of initial ligand/As(III) molar ratios (R) or LDH. Although Cu-Al-LDH sorbed lower amounts of As(III) than the Mg-Fe-LDH, it showed, surprisingly, a higher affinity for As(III). This surprising behavior puts this LDH in the forefront as a potential sorbent for the treatment of arsenic-contaminated waters.
Introduction
A
Arsenic pollution, as a worldwide problem, has been reported in a many countries, including Bangladesh, India, China, Australia, Japan, United States, Argentina, and lately Italy (Mandal and Suzuky, 2002; Pigna et al., 2014). Consequently, the development and use of techniques to remove arsenic from water have received extensive attention (Smedley and Kinniburgh, 2002; Choong et al., 2007); solar oxidation and removal of As, for example, seem to be economic and feasible (Hug et al., 2001). Nevertheless, among the various treatment technologies, the sorption technique is considered one of the most popular due to its high efficiency, easy operation, low cost, and little risk (Choong et al., 2007). Many sorbents have been developed, including Al and Fe (hydr)oxides, anionic clays, and zeolites (Violante and Pigna, 2002; Choong et al., 2007; Liu et al., 2001; Pigna et al., 2006; Zhu et al., 2011).
Recently, layered double hydroxides (LDHs), a class of multifunctional nanostructured anionic clays, have been reported to serve as sorbents for anionic contaminants due to their large surface area, high anion exchange capacity, and good thermal stability (Goh et al., 2008; Violante et al., 2009; Caporale et al., 2011, 2013; Chetia et al., 2012). The general formula of the LDHs is [M2+1-x M3+ x (OH)6]x+ [(An−)x/n m H2O]x−, where M2+ are the divalent cations (Mg, Ca, Cu, Zn, Ni, Fe), M3+ are the trivalent cations (Al, Fe, Mn), x is the molar ratio M3+/(M2+ + M3+), taking values between 0.20 and 0.33, and A is an interlayer anion (e.g., Cl, NO3, ClO4, CO3, SO4) of valence n (Cavani et al., 1991).
They are composed of positively charge brucite-like sheets and the positive charges are balanced by intercalation of anions in the hydrated interlayer regions, analogous to the cationic clays whose negative charge of the aluminosilicate layers are counterbalanced by cations (Cavani et al., 1991; Costantino and Pinnavaia, 1995; Violante et al., 2009; Caporale et al., 2011). The LDHs have relatively weak interlayer bonding and, as a consequence, the original anions sorbed in the interlayer are easily exchanged with many inorganic and organic anions by simple ion exchange (Goh et al., 2008); their point of zero charge was reported to be in the range 7.0–9.0 (Violante et al., 2009).
Over the last 20 years, studies have been carried out on the removal of arsenic from contaminated waters using different forms of LDHs as sorbents (You et al., 2001; Goh et al., 2008; Violante et al., 2009; Caporale et al., 2011, 2013; Chetia et al., 2012) as well as on the use of LDHs in natural, synthetic, and anthropogenic mining systems (Douglas et al., 2010; Gomez et al., 2013; Paikaray et al., 2014). However, the removal of As(III) by LDHs in the presence of competing anions has received scant attention.
In this study, four different LDHs (i.e., Cu-Al-, Mg-Al-, Mg-Fe-, and Zn-Al-LDH), were synthesized to determine their efficiency in removing arsenite [As(III)] in the presence of competing inorganic anions [i.e., carbonate (CO3), chloride (Cl), fluoride (F), phosphate (PO4), sulfate (SO4)], commonly present in water bodies.
The LDHs were characterized by X-ray diffraction (XRD) and determining their specific surface area. Arsenite sorption isotherms, in the presence and absence of competing anions, were collected to measure the efficiency of an anion in inhibiting As(III) sorption by the LDHs and give insight into the sorption process.
Materials and Methods
Synthesis of Cu-Al-, Mg-Al-, Mg-Fe-, and Zn-Al-LDHs
LDHs of Cu-Al (Cu-Al-LDH), Mg-Al (Mg-Al-LDH), Mg-Fe (Mg-Fe-LDH), and Zn-Al (Zn-Al-LDH) were synthesized by the coprecipitation method of Costantino and Pinnavaia (1995). Suitable amounts of solutions containing either MgCl2·6H2O, FeCl2·6H2O, CuCl2·2H2O, ZnCl2·2H2O, and/or AlCl3·6H2O (initial Mg/Al, Mg/Fe, Cu/Al, or Zn/Al molar ratio equal to 2) were slowly added with stirring to NaOH solutions at pH 10.0 and 20°C.
The pH of each suspension was maintained constant for 24 h, at 20°C, by adding 0.5 mol/L NaOH using an automatic titrator (Potentiograph E536 Metrom Herisau) in conjunction with an automatic syringe (burette 655 Dosimat). After 24 h, the suspensions were washed five times with deionized water, followed by centrifugation at 10,000 g/min for 0.5 h. The suspensions were then dialyzed (molecular weight cut off of 15,000) for 21 days, freeze dried, and lightly ground to pass through a 0.315 mm sieve.
Characterization of the LDHs
Cu-Al-, Mg-Al-, Mg-Fe-, and Zn-Al-LDH precipitates were characterized by XRD. XRD patterns of randomly oriented samples were obtained using a Rigaku diffractometer (Rigaku Co.) equipped with Cu Kα radiation generated at 40 kv and 30 mA and a scan speed of 2° 2θ min−1. The XRD traces were the results of eight summed signals. In addition, the specific surface area of each LDH sample (in triplicate) was determined by H2O sorption at 20% relative humidity, according to the method described by Quirk (1955).
Arsenite sorption isotherms
One hundred milligrams of each sample were equilibrated at 20°C with 0.01 mol/L KCl at pH 7.0. Suitable amounts of a 0.1 mol/L NaAsO2 stock solutions were added to obtain an initial As(III) concentration in the range 5 × 10−4 to 10−2 mol/L. The pH of each suspension was maintained constant at pH 7.0 by the addition of 0.1 or 0.01 mol/L HCl or KOH using a stirred pH-stat apparatus. The final volume of the suspensions was adjusted to 20 mL with 0.01 mol/L KCl. The suspensions were shaken for 24 h at 20°C on a magnetic stirrer. Finally, the suspensions (20 mL) were centrifuged at 10,000 g/min for 20 m and filtered through a 0.22-μm membrane filter. The filtrates were stored at 2°C until analysis.
Arsenite sorption in the presence of inorganic ligands
In flasks containing 100 mg of each sample were added simultaneously suitable amounts of As(III) from a 0.1 mol/L NaAsO2 and the inorganic anions Cl, F, SO4, CO3, and PO4 to have initial ligand/As(III) molar ratios (R) of 1, 2, and 3. The amount of As(III) added to each LDHs was about 50% of surface coverage based on the sorption isotherms. The final volume of the suspensions was adjusted to 20 mL with 0.01 mol/L KCl and then they were shaken for 24 h at 20°C on a magnetic stirrer. The pH of each suspension was maintained constant at pH 7.0 by the addition of 0.1 or 0.01 mol/L HCl or KOH using a stirred pH-stat apparatus. Finally, the suspensions were centrifuged at 10,000 g/min for 20 min and filtered through a 0.22-μm membrane filter. The filtrates were stored at 2°C until analysis.
Arsenite determination
As(III) concentration in the filtrates was determined by flow-injection hydride generation atomic absorption spectrometer (HG-AAS), using a Perkin-Elmer AAnalist 700 interfaced with the FIAS 100 hydride generator. The reagents used for HG-AAS were 10% (v/v) HCl and 0.2% NaBH4 in 0.05% NaOH. The samples were reduced before analysis with 5% (w/v) KI and 5% (w/v) ascorbic acid to improve sensitivity. The As detection limit provided by this method was 2.7 × 10−8 mol/L.
The amount of As(III) sorbed was determined by the difference between the amount initially added and that determined in the filtrates. The intraday repeatability study was carried out by the analysis of the same standard solution five consecutive times (n = 5) in the same day under the same conditions. The interday precision was carried out for three successive days using the same conditions. The relative standard deviation of these measurements ranged from 1.5% to 3.6%.
Results and Discussion
Nature of the LDHs
All the LDHs were initially intercalated with chloride ions and water molecules. The XRD patterns of the LDHs are shown in Fig. 1. All LDHs showed peaks near 0.760, 0.380, 0.260, 0.153, and 0.150 nm and some asymmetric peaks at high angles (>30° 2θ), characteristic of hydrotalcite (Cavani et al., 1991; Costantino and Pinnavaia, 1995; Caporale et al., 2011). Costantino and Pinnavaia (1995) calculated that the gallery height of a Mg-Al-LDH was 2.93 Å, smaller than the diameter of the chloride ion (3.62 Å), suggesting that there was a strong electrostatic interaction between the layers of high charge density and the chloride ion, even in the presence of interlayered water. The fact that the basal spacing for all the LDHs was similar (d003 = 7.6 Å), implied that the electrostatic interaction between the layers and chloride ion were similar in all the LDHs.

Powder X-ray diffraction patterns of Cu-Al, Mg-Al, Mg-Fe, and Zn-Al layered double hydroxides (LDHs).
The peaks in the XRD patterns of the Mg-Al-LDH and Zn-Al-LDH were narrow and sharp, indicating that they were well crystalline, whereas the peaks of the Mg-Fe-LDH and Cu-Al-LDH were broad, indicating that the latter LDHs were less crystalline, there was a disorder in the layers stacking, and there was a separate phase of metal oxides and/or the particles were generally smaller (Costantino and Pinnavaia, 1995; Khan and O'Hare, 2002). In accordance, the surface area of the two poorly crystalline LDHs was about 30% higher compared with the two well-crystalline LDHs (Table 1).
LDH, layered double hydroxides.
Arsenite sorption isotherms
The 24 h As(III) sorption isotherms on the LDHs at pH 7.0 are reported in Fig. 2. The As(III) sorbed onto the LDHs conformed to the Langmuir equation in the following form:

Sorption isotherms of As(III) onto Mg-Al-LDH, Mg-Fe-LDH, Cu-Al-LDH, and Zn-Al-LDH, at pH 7.0 and 20°C, after 24 h of reaction.
where S is the amount of the As(III) sorbed per unit mass of sorbent (i.e., LDHs) (mmol/kg), Sm is the maximum amount of As(III) that may be bound to the sorbent (sorption capacity), c is the equilibrium solution concentration (mmol/L), and K is a constant related to the binding energy (Giles et al., 1974).
The shape of the As(III) sorption isotherms at pH 7 for the Mg-Al-LDH and Zn-Al-LDH was typical of an L-type curve (Giles et al., 1974), while for the Cu-Al-LDH and Mg-Fe-LDH it was that of an H-type curve (Fig. 2). The H-type curve is a special case of the L-type curve in that at low As(III) concentrations [<∼400 mmol As(III) kg−1], a greater percentage of the As(III) initially added was sorbed by the LDH. This indicated that As(III) had a higher affinity for the sorption sites on Cu-Al-LDH and Mg-Fe-LDH than on Mg-Al-LDH and Zn-Al-LDH (Limousin et al., 2007). This higher affinity could be due to the higher surface area and/or poor crystallinity of the Cu-Al-LDH and Mg-Fe-LDH, suggesting that there were more sorption sites available for As(III) sorption on these LDHs than on Mg-Al-LDH and Zn-Al-LDH.
The K value, an indication of binding energy, was highest for Cu-Al-LDH (K = 9.79) >> Mg-Fe-LDH (K = 4.02) >> Mg-Al-LDH (K = 2.73) ≥ Zn-Al-LDH (K = 2.28) (Table 1). This indicates that the higher affinity of As(III) for the sorption sites on Cu-Al-LDH and to a lesser extent Mg-Fe-LDH may also be due to the interaction between the sorption sites and the As(III). Nevertheless, Mg-Fe-LDH (Sm = 1,312 mmol/kg) had the highest sorption capacity, nearly double that of Cu-Al-LDH (Sm = 760 mmol/kg), and almost thrice that of Mg-Al-LDH and Zn-Al-LDH (Sm = 562 and 569 mmol/kg, respectively), suggesting that Mg-Fe-LDH had more sites available for As(III) sorption compared with the Cu-Al-LDH and/or had the higher affinity for the sorption sites at high As(III) concentrations.
Note that the surface area of Cu-Al-LDH was similar to that of Mg-Fe-LDH and about 30% larger compared with Mg-Al-LDH and Zn-Al-LDH (Table 1). Hence, stereochemical and/or electrostatic effects also appear to play a role in the sorption of As(III) by these LDHs. Sanderson et al. (2013) observed that with increased isomorphous substitutions of Al for Mg, there was an increase in the charge density, binding the layers together more tightly. This could hinder/block the sorption of the anion into the interior interlayer sites, particularly if the sorption happened on the outer edge of the interlayer. This may explain why the sorption capacity of Cu-Al-LDH was smaller compared with Mg-Fe-LDH at high As(III) concentration.
Arsenite sorption in the presence of inorganic ligands
All the LDHs were initially intercalated with chloride ions and water molecules. The sorption of As(III) on the Cu-Al-, Mg-Al-, Mg-Fe-, and Zn-Al-LDH at pH 7.0, in the presence of increasing initial ligand/As(III) molar ratios (R = 1, 2, 3), after 24 h of reaction, are shown in Tables 2–4. The efficiency (%) of the ligands to compete with As(III) was calculated by subtracting the amount sorbed in the presence of each anion from that initially added, divided by the amount initially added:
Initial molar ratio anions/As(III) = 1.
Numbers in parenthesis indicate the efficiency (%) of the anion in inhibiting As(III) sorption on the LDH.
Initial molar ratio anions/As(III) = 2.
Numbers in parenthesis indicate the efficiency (%) of the anion in inhibiting As(III) sorption on the LDH.
Initial molar ratio anions/As(III) = 3.
Numbers in parenthesis indicate the efficiency (%) of the anion in inhibiting As(III) sorption on the LDH.
The efficiency (%) of the anions to compete with As(III) for the sorption sites with the four LDHs was in the order Cl ≤ F ≤ SO4 << CO3 << PO4, regardless of the initial R value or LDH (Tables 2–4). It was apparent that there was a relationship between the charge of the competing anion and the efficiency, the higher the charge of the anion the more competitive it was with the As(III) for the sorption sites. Other researchers have observed a similar relationship between charge and competitiveness (You et al., 2001; Goh and Lim, 2010; Caporale et al., 2013). There was no significant difference in the competitiveness between the Cl and F with As(III) for the sorption sites, attributed to the fact that they are both monovalent and nonspecific anions in solid–liquid interfacial reactions.
The large difference between the divalent anions to compete with As(III) for the sorption sites was due their ability to form inner-sphere complexes. Sulfate only forms inner-sphere complexes at more acidic pH values (3.5–4.5), whereas CO3 and PO4 form inner-sphere complexes at pH 7 (You et al., 2001; Violante and Pigna, 2002). Interestingly, You et al. (2001) found that if CO3 was the interlayer anion, instead of Cl, As(III) was not sorbed by an Mg-Al-LDH. Phosphate competes more effectively than CO3 because it has a higher affinity at pH 7, and/or can form trivalent bonds with the surface.
The efficiency of a particular anion to compete with As(III) was similar for the Zn-Al-LDH and Mg-Al-LDH, and for the Mg-Fe-LDH and Cu-Al-LDH for all R values (Tables 2–4), with the efficiency being considerably higher for the Zn-Al-LDH/Mg-Al-LDH versus the Mg-Fe-LDH/Cu-Al-LDH. Interactions between the negatively charged anion and the positively charged LDH determine the separation between the layers (Khan and O'Hare, 2002). Also, anions must intercalate between adjacent layers to maintain charge neutrality. Neither of the latter two statements in themselves explain why the competition between an anion and As(III) varied between LDHs.
Another important consideration is the geometry of the coordinating site in the interlayers and the anion. Dynes and Huang (1997) studied the competition of organic acids with selenite for sorption sites on Al hydroxides. Generally, the larger the stability constant of the Al-organic solution complexes (KAl-L), the more effective the organic acid was in competing with selenite for the sorption sites of Al hydroxides. However, some of the organic acids competed less successfully than expected based on their KAl-L values. This was attributed to the sterochemical and electrostatic effects originating from both the surface of the Al hydroxides and the organic acid. It is proposed that the coordination groups in the interlayer of Mg-Fe-LDH/Cu-Al-LDH are of similar geometry as that of the As(III) compared with the Zn-Al-LDH/Mg-Al-LDH.
Competition between As(III) and the inorganic anions for the sorption sites increased with increasing R for all LDHs (Tables 2–4). For the less competitive anions such as Cl, the efficiency increased slightly with increasing R (e.g., Mg-Al-LDH, 5.2% [R = 1], 9.4% [R = 2], 16% [R = 3]), whereas for the more competitive anions such as PO4, the efficiency increased significantly with increasing R (e.g., Mg-Al-LDH, 48.6% [R = 1], 63.2% [R = 2], 78% [R = 3]). The increase in efficiency with increasing R was attributed to the direct occupation of the sorption sites by the anions and/or surface charge modification through the sorption of the anions. In essence, the higher anion concentration was able to maintain the charge neutrality in the interlayer, such that it was more difficult for the As(III) to approach the sorption sites.
Conclusions
This study confirms that LDHs are reliable and efficient As(III) sorbents, able to remove harmful anions from contaminated waters in a cost-effective way. The nature and reactivity of the LDH sorbents toward the removal of the As(III) anion depended on the divalent (Mg vs. Cu or Zn) and trivalent (Al vs. Fe) cations used to synthesize the LDH. In the absence of competing anions, the Mg-Fe-LDH had the highest sorption capacity, whereas in the presence of competing anions, the Cu-Al-LDH showed a higher affinity for As(III) (i.e., lowest efficiency %). Hence, in contaminated waters, where competing anions occur in high molar ratios to As(III), Cu-Al-LDH may be better choice for the sorbent than Mg-Fe-LDH. Upcoming studies with X-ray adsorption spectroscopy have been recently scheduled to obtain crucial information on the nature of the bonds occurring between As(III) and the various LDHs, as well as As(III) desorption surveys by competing anions at different pH's and temperatures.
Footnotes
Acknowledgment
This work was supported by the Italian Research Program of National Interest (PRIN), year 2010–2011 (Grant number 2010JBNL17_005).
Author Disclosure Statement
No competing financial interests exist.
