Arsenic(V) Removal from Water by Resin Impregnated with Cyclodextrin Ligand

: More rigorous environmental rules on hazardous metals release require the implementation of diverse methods to remove them from contaminated streams. The adsorption process is the most attractive method for removal of toxic metal ions from different aqueous solutions. In the present study, batch and ﬁxed bed column experiments were carried out to investigate the performance of permethylated β -cyclodextrin impregnated resin for the removal of As(V) ions from aqueous solutions. Sorption of As(V) ions was performed in groups under a variety of pH and metal concentrations. It was noted that removal efﬁciency was the highest (98%) at pH 6.0, and at an initial As(V) concentration equal to 0.1 mg/L. Sorption model use for experimental data indicates that the removal mechanism of As(V) by the permethylated β -cyclodextrin impregnated resin was the physical sorption. The adsorption capacity of the cyclodextrin ligand immobilized onto Amberlite XAD-4 resin was 19 mg/g. In addition, the impregnated resin was regenerable, thus was able to be used over and over.


Introduction
Heavy metal and metalloid presence in the environment, due to their toxicity, is harmful, and mitigating this problem is now of great importance. Contamination of water by toxic metal ions has gradually increased over recent years due to overpopulation, insufficient hazardous waste storage, and industrial activity. Arsenic is one of the toxic metals of environmental concern due to occupational or domestic exposure, and is listed in the US Agency for Toxic Substances and Disease Registry's (ATSDR) "Top 20 Hazardous Substances" in cooperation with United States Environmental Protection Agency (US EPA) [1].
Arsenic can filter into aquatic environments naturally, for example dissolution of minerals by weathering, microbial activity, and complexation with natural organic materials. However, anthropogenic activities, including industrial mining and metallurgical activity, fossil fuel combustion, arsenic pesticide usage, herbicides, and crop desiccants, can lead to arsenic contamination in soils and surface water [2,3].

Preparation of Permethylated β-Cyclodextrin (β-CD)
β-CD (1.135 g; 0.001 mol) was dissolved in dry DMF (120 mL) and cooled to 0 • C. Sodium hydride (1 g) (60% in mineral oil) was carefully integrated, and the mixture was agitated at this temperature for 1 h, then at room temperature for 1 h. The reaction mixture was then cooled to 0 • C, and methyl iodide (3.17 mL; 50.94 mmol) was added drop by drop. The reaction mixture was stirred at 0 • C for 1 h, and then left to warm gradually to room temperature. After 24 h of stirring, the mixture was cooled to 0 • C and methanol (25 mL) was integrated. The mixture was transferred into icy water and extracted with chloroform or dichloromethane (5 × 80 mL). The organic layer was rinsed with 3% sodium thiosulfate and deionized water, dried over Na 2 SO 4 , and the solvent was evaporated. The residue can be purified by column chromatography on silica eluting with ethyl acetate containing 0.1% methanol [28] to produce permethylated β-CD (ligand 1) as a white solid.
The procedure of preparation of permethylated β-CD was shown in Figure 1. 1

Preparation of Permethylated β-Cyclodextrin (β-CD)
β-CD (1.135 g; 0.001 mol) was dissolved in dry DMF (120 mL) and cooled to 0 ° Sodium hydride (1 g) (60% in mineral oil) was carefully integrated, and the mixture w agitated at this temperature for 1 h, then at room temperature for 1 h. The reaction mixtu was then cooled to 0 °C, and methyl iodide (3.17 mL; 50.94 mmol) was added drop b drop. The reaction mixture was stirred at 0 °C for 1 h, and then left to warm gradually room temperature. After 24 h of stirring, the mixture was cooled to 0 °C and methanol (2 mL) was integrated. The mixture was transferred into icy water and extracted with chl roform or dichloromethane (5 × 80 mL). The organic layer was rinsed with 3% sodiu thiosulfate and deionized water, dried over Na2SO4, and the solvent was evaporated. Th residue can be purified by column chromatography on silica eluting with ethyl aceta containing 0.1% methanol [28] to produce permethylated β-CD (ligand 1) as a white soli The procedure of preparation of permethylated β-CD was shown in Figure 1

Preparation of Impregnated Resin
A 2 g portion of dry Amberlite XAD-4 was placed in 20 mL of dichloromethane co taining ligand 1 (1.0 × 10 −4 mol) and stirred for 24 h. A sintered glass filter was used separate the resin, which was rinsed with water to remove the solvent. The gravimetr assay was applied to determine the amount of the ligand immobilized onto Amberli XAD-4 resin [14]. The maximum value of ligand 1 saturation reached 1.0 × 10 −5 mol/g.

Preparation of Impregnated Resin
A 2 g portion of dry Amberlite XAD-4 was placed in 20 mL of dichloromethane containing ligand 1 (1.0 × 10 −4 mol) and stirred for 24 h. A sintered glass filter was used to separate the resin, which was rinsed with water to remove the solvent. The gravimetric assay was applied to determine the amount of the ligand immobilized onto Amberlite XAD-4 resin [14]. The maximum value of ligand 1 saturation reached 1.0 × 10 −5 mol/g. The scanning electron microscope (VEGA 3, TESCAN, Brno, Czech Republic) images before and after immobilization of permethylated β-CD onto Amberlite XAD-4 resin ( Figure 2) indicates the change in surface area of the "raw" and the impregnated resin, indicating the presence of the ligand 1 onto resin.

Batch Sorption Procedures
Batch experiments were carried out by putting of 0.1 g of impregnated resin into glass bottles and adding of 100 mL of arsenic solution at a 0.1 mg/L (up to 1 mg/L in the case of concentration change) for pH range = 3-9. Arsenic(V) solutions were prepared by dissolving Na2HAsO4·7H2O in deionized water to the desired concentration and used without further purification. The pH of As(V) solution was adjusted within the range 3-7 (before contact with the adsorbent) with 0.1 M HCl or NaOH. The mixture was shaken at room temperature for 60 minutes, the adsorbent was filtered off, and arsenic concentration in aqueous phase was measured by Inductively Coupled Plasma Mass Spectrometry (ICP MS) (Elan 6000, PerkinElmer, Waltham, Massachusetts, United States).
The effectiveness of arsenic removal (%) and metal uptake (mg of metal ion/g of the permethylated β-CD) were determined as the following: where qe is adsorption capacity at equilibrium (mg/g), C0 and Ce are the initial and the final As(V) concentration (mg/L), respectively, V is the volume of solution (L), and m is the sorbent weight (g) in dry form No notable adsorption capacity of Amberlite XAD-4 resin was observed in the blank experiments when ligand 1 was non attending.
The reported values correlated with the average of three replicates, and the observed standard deviation was less than 2%.
The Langmuir, Freundlich, and Temkin adsorption models, as well as the van't Hoff equation, were applied to interpret data obtained for the sorption of As(V) ions onto the impregnated resin.

Langmuir Model
The Langmuir isotherm is a generally used model for adsorption on a completely homogenous surface where adsorbed molecules show little interplay [14]. The model assumes uniform adsorption energies onto the surface and maximum adsorption controlling by the monolayer's saturation level.
Langmuir model can be represented by the following linear equation [29]:

Batch Sorption Procedures
Batch experiments were carried out by putting of 0.1 g of impregnated resin into glass bottles and adding of 100 mL of arsenic solution at a 0.1 mg/L (up to 1 mg/L in the case of concentration change) for pH range = 3-9. Arsenic(V) solutions were prepared by dissolving Na 2 HAsO 4 ·7H 2 O in deionized water to the desired concentration and used without further purification. The pH of As(V) solution was adjusted within the range 3-7 (before contact with the adsorbent) with 0.1 M HCl or NaOH. The mixture was shaken at room temperature for 60 minutes, the adsorbent was filtered off, and arsenic concentration in aqueous phase was measured by Inductively Coupled Plasma Mass Spectrometry (ICP MS) (Elan 6000, PerkinElmer, Waltham, Massachusetts, United States).
The effectiveness of arsenic removal (%) and metal uptake (mg of metal ion/g of the permethylated β-CD) were determined as the following: where q e is adsorption capacity at equilibrium (mg/g), C 0 and C e are the initial and the final As(V) concentration (mg/L), respectively, V is the volume of solution (L), and m is the sorbent weight (g) in dry form No notable adsorption capacity of Amberlite XAD-4 resin was observed in the blank experiments when ligand 1 was non attending.
The reported values correlated with the average of three replicates, and the observed standard deviation was less than 2%.
The Langmuir, Freundlich, and Temkin adsorption models, as well as the van't Hoff equation, were applied to interpret data obtained for the sorption of As(V) ions onto the impregnated resin.

Langmuir Model
The Langmuir isotherm is a generally used model for adsorption on a completely homogenous surface where adsorbed molecules show little interplay [14]. The model assumes uniform adsorption energies onto the surface and maximum adsorption controlling by the monolayer's saturation level.
Langmuir model can be represented by the following linear equation [29]: Processes 2022, 10, 253 where q e represents the mass of adsorbed metal per unit of resin (mg/g), V m is the monolayer capacity, k is the equilibrium constant, and C e is the equilibrium concentration of the solution (mg/L). The Langmuir plot was used to determine k and V m .

Freundlich Model
The Freundlich model is considered the earliest empirical equation and is consistent with exponential spreading of active centers, which is characteristic of heterogeneous surfaces [14]. The Freundlich equation (linear form) is [30]: where K f and n represent adsorption capacity and favor, respectively. The magnitude of the n indicates the favorability of adsorption [31].

Temkin Model
The Temkin model reveals that a decrease in the heat of adsorption is linear, and adsorption is defined by an uniform distribution of binding energies. The Temkin isotherm is represented by the following equation [27]: where K T is the equilibrium binding constant (L/mg), b refers to heat of adsorption (kJ/mol), R is the gas constant (8.314 × 10 −3 kJ/(K·mol) and T is the absolute temperature (K).

Van't Hoff Equation
The van't Hoff equation relates the change in the equilibrium constant of the adsorption, to the change in temperature, given the changes in enthalpy (∆H) and entropy (∆S) of the process [32]. The van't Hoff equation (linear form) is [32]: The values of ∆H and ∆S were calculated from the slope and intercept of the linear plot of ln k versus 1/T.

Column Experiment Design
A flow-through column test (using Millipore column, Tygon tubing, and peristaltic pump Manostat Carter) was used to investigate the effectiveness of ligand 1 immobilized onto Amberlite XAD-4 resin for the removal of arsenic ions from water. The 1.0 g of impregnated resin was placed in a Millipore column (length 250 mm, internal diameter 10 mm). The column was subsequently saturated by the upward flow of distilled water. The water was displaced with the upward flow of the synthetic arsenic solution (at concentration of 1.0 mg/L) once saturation was achieved at a constant flow rate of 4.0 mL/min. A spur at the top of the column was used to collect effluents into covered glass collection vessels and analyze them to determine the arsenic concentration by Inductively Coupled Plasma Mass Spectrometry (ICP MS) (Elan 6000, PerkinElmer Waltham, Massachusetts, United States). Effluents were taken until the value of proportion of effluent concentration (c) to initial concentration (c 0 ) was 1.
The moment of breakthrough emergence and the profile of the breakthrough curve are crucial features in determining the dynamic sorption capacity. The loading behavior of a sorbent in a fixed bed is generally expressed in terms of adsorbed metal concentration (c ad = inlet metal concentration (c 0 ) − effluent metal concentration (c)) or normalized concentration defined as the ratio of effluent metal concentration to inlet metal concentration (c/c 0 ) as a function of time or volume of effluent for a given bed height. The area under the breakthrough curve (A) is obtained by integrating the adsorbed concentration (c ad ; mg/L) versus t (min). A graph of the breakthrough curve can be used to determine the total adsorbed metal quantity (maximum column capacity) [33].
Total adsorbed metal quantity (q t ; mg) in the column for a given feed concentration and flow rate is calculated from the following equation: where Q is the flow rate (mL/min). Equilibrium metal uptake (q e ) (or maximum adsorption capacity) in the column is defined by the total amount of metal sorbed (q t ) per gram of sorbent (m) after total flow time: The breakthrough is generally defined as the point when the effluent concentration from the column is approximately 3-5% of the influent concentration [34].
The impregnated resin was washed with deionized water so that the As(V) ions which were not sorbed could be displaced. Arsenic(V) ions sorbed on the impregnated resin were subsequently eluted using 0.1 M NaOH.

Effect of pH
A crucial operational parameter in the sorption process is pH, as it determines the solubility of the metal ions, concentration of the counter ions on the sorbent's functional groups, and the extent of ionization of the sorbate during reaction [35].
As shown in Figure 3, the removal of As(V) using permethylated β-cyclodextrin impregnated resin depends notably on pH of water. The highest removal As(V) efficiency (more than 90%) was at pH ranges from 5 to 7. As(V) mainly exists then in the form of H 2 AsO 4 − or HAsO 4 2− [36,37]. The As(V) removal efficiency decreased significantly at pH < 5, because of the high ionic mobility of [H 3 O]+ in water as compared with most other ions [38]. Metal ions are able to be eliminated from water as they interact with the protonated or deprotonated hydroxyl groups of CDs, placed at both rims of the CD framework. For derivated CDs, the complexation mechanism can be more complicated. For instance, alkylated CDs can mostly bind via coordination bond from lone pair electron of oxygen, other forces such as van der Waals, dipole-dipole may together catch metal ions [18]. Under neutral conditions, the surface of the studied sorbent becomes more positively charged, causing adsorbent-adsorbate electrostatic attraction [39]. In turn, when the adsorbent was placed in an alkaline environment, its surface is net negatively charged. Considering arsenic's dominant form (H 2 AsO 4 − ), there will be electrostatic repulsion due to the adsorbent's negative surface charge at alkaline pH [40]. Therefore, the performance of the adsorbent for arsenic(V) ions was diminished in an alkaline medium.

Effect of Initial Arsenic(V) Concentration
The effect of initial concentration on uptake of As(V) ions by permethylated β-CD immobilized onto Amberlite XAD-4 resin is shown in Figure 4. The quantity of As(V) sorbed uniformly increased as metal concentration increased up to the value of 1.0 mg/L. The uptake of As(V) was 6.8 mg/g at a metal concentration of 0.1 mg/L, while at a concentration of 1 mg/L-19.0 mg/g. It was found that immobilized permethylated β-CD performs well in sorption of As(V) from water, a result of the larger surface area provided by the framework of Amberlite resin to the ligand for interaction with As(V) ions, along with inclusion complexation via macrocyclic ring [1,18]. Moreover, at initial concentration of 0.1 mg/L the highest As(V) removal efficiency (98%) was achieved; the arsenic concentration was then below the limit in drinking water (10 μg/L).

Adsorption Isotherms
The Langmuir, Freundlich, and Temkin isotherms for a temperature of 293 K, and a van 't Hoff plot (in the range of temperature 293-313 K) are shown in Figure 5. All value constants determined from adsorption isotherms and van't Hoff plot are given in Table 1.

Effect of Initial Arsenic(V) Concentration
The effect of initial concentration on uptake of As(V) ions by permethylated β-CD immobilized onto Amberlite XAD-4 resin is shown in Figure 4. The quantity of As(V) sorbed uniformly increased as metal concentration increased up to the value of 1.0 mg/L. The uptake of As(V) was 6.8 mg/g at a metal concentration of 0.1 mg/L, while at a concentration of 1 mg/L-19.0 mg/g. It was found that immobilized permethylated β-CD performs well in sorption of As(V) from water, a result of the larger surface area provided by the framework of Amberlite resin to the ligand for interaction with As(V) ions, along with inclusion complexation via macrocyclic ring [1,18]. Moreover, at initial concentration of 0.1 mg/L the highest As(V) removal efficiency (98%) was achieved; the arsenic concentration was then below the limit in drinking water (10 µg/L).

Effect of Initial Arsenic(V) Concentration
The effect of initial concentration on uptake of As(V) ions by permethylated β-CD immobilized onto Amberlite XAD-4 resin is shown in Figure 4. The quantity of As(V) sorbed uniformly increased as metal concentration increased up to the value of 1.0 mg/L. The uptake of As(V) was 6.8 mg/g at a metal concentration of 0.1 mg/L, while at a concentration of 1 mg/L-19.0 mg/g. It was found that immobilized permethylated β-CD performs well in sorption of As(V) from water, a result of the larger surface area provided by the framework of Amberlite resin to the ligand for interaction with As(V) ions, along with inclusion complexation via macrocyclic ring [1,18]. Moreover, at initial concentration of 0.1 mg/L the highest As(V) removal efficiency (98%) was achieved; the arsenic concentration was then below the limit in drinking water (10 μg/L).

Adsorption Isotherms
The Langmuir, Freundlich, and Temkin isotherms for a temperature of 293 K, and a van 't Hoff plot (in the range of temperature 293-313 K) are shown in Figure 5. All value constants determined from adsorption isotherms and van't Hoff plot are given in Table 1.

Adsorption Isotherms
The Langmuir, Freundlich, and Temkin isotherms for a temperature of 293 K, and a van 't Hoff plot (in the range of temperature 293-313 K) are shown in Figure 5. All value constants determined from adsorption isotherms and van't Hoff plot are given in Table 1. As(V) ions' adsorption parameters onto impregnated resin are consistent with the Langmuir and van't Hoff models. R 2 values are considered a measure of the correspondence of experimental data to the isotherm model. These values for As(V) ions are very similar to 1. Therefore, the following order to fit isotherms is indicated by determination coefficients: Langmuir > van't Hoff > Temkin > Freundlich, for adsorption of As(V) ions onto impregnated resin. A Freundlich constant n value larger than 1 shows the beneficial character of adsorption [14].
The Temkin isotherm shows that (i) there is a linear decrease in adsorption heat of all the molecules in the layer with coverage due to adsorbent-adsorbate interactions, and that (ii) a uniform distribution of binding energies characterizes adsorption. Ion-exchange mechanisms typically have binding energies in the order of 8-16 kJ/mol, while physisorption processes are mentioned to have low adsorption energies [41]. Low b and ∆H values (1.2 and 4.1 kJ/mol, respectively) noted in the present study show minimal sorbate-adsorbent ionic interactions, indicating the physisorption of arsenic(V) ions. Moreover, the positive magnitude of ∆H indicates the endothermic character of As(V) adsorption process and the positive magnitude of ∆S suggests the affinity of the immobilized permethylated β-CD for arsenic, as well as increase of randomness at solid-solution boundary by way of metal ion uptake [42].  Table 1. Isotherms and van't Hoff equation parameters calculated for As(V) ions adsorption onto resin impregnated with permethylated β-CD. As(V) ions' adsorption parameters onto impregnated resin are consistent with the Langmuir and van't Hoff models. R 2 values are considered a measure of the correspondence of experimental data to the isotherm model. These values for As(V) ions are very similar to 1. Therefore, the following order to fit isotherms is indicated by determination coefficients: Langmuir > van't Hoff > Temkin > Freundlich, for adsorption of As(V) ions onto impregnated resin. A Freundlich constant n value larger than 1 shows the beneficial character of adsorption [14].

Langmuir Isotherm Parameters Freundlich Isotherm Parameters Temkin Isotherm Parameters Van't Hoff Equation Parameters
The Temkin isotherm shows that (i) there is a linear decrease in adsorption heat of all the molecules in the layer with coverage due to adsorbent-adsorbate interactions, and that (ii) a uniform distribution of binding energies characterizes adsorption. Ionexchange mechanisms typically have binding energies in the order of 8-16 kJ/mol, while physisorption processes are mentioned to have low adsorption energies [41]. Low b and ∆H values (1.2 and 4.1 kJ/mol, respectively) noted in the present study show minimal sorbateadsorbent ionic interactions, indicating the physisorption of arsenic(V) ions. Moreover, the positive magnitude of ∆H indicates the endothermic character of As(V) adsorption process and the positive magnitude of ∆S suggests the affinity of the immobilized permethylated β-CD for arsenic, as well as increase of randomness at solid-solution boundary by way of metal ion uptake [42].
Many different adsorbents used to eliminate arsenic(V) ions from aqueous solutions have been demonstrated in the literature. A summary of the published adsorption capacity of different adsorbents for trace and low initial concentrations of arsenic(V) (up to 2 mg/L) is presented in Table 2, including the results from this work. As is shown in Table 2, the adsorption capacity of As(V) ions for permethylated β-CD impregnated Amberlite XAD-4 resin was, in most cases, higher than for different adsorbents reported in the literature.

Column System
The breakthrough curve of As(V) adsorption is presented in Figure 6. It can be seen that breakthrough occurred after 30 minutes (feeding solution flow of 120 mL). The loading of column was achieved after 150 minutes (feeding solution flow of 600 mL). The calculated adsorption column capacity value for arsenic(V) was 19.1 mg As(V) per g of permethylated β-CD immobilized onto resin. This is a similar value to that obtained in batch experiment at the identical (1.0 mg/L) concentration of arsenic(V) solution. A significant reduction of the overall cost of water treatment process would be achieved by an adsorbent with good adsorption capacity, as well as showing favorable regeneration properties. The successful regeneration of an adsorbent can therefore be considered an essential action for determining the applicability of an adsorbent in water remediation practice [50]. The results of regeneration studies indicated that 0.1 M NaOH solution is worthy for this operation in the case of immobilized CD ligand and can be applied as a regenerating factor. The desorption rate was high; when a volume of 60 mL of regeneration solution was passed through column, more than 95% of As(V) ions were desorbed.
The stability and potential reusability of the impregnated resin were also evaluated by monitoring the change in recoveries of As(V) ions through several sorption and desorption cycles, for initial As(V) concentration of 1.0 mg/L. The results of adsorption/desorption cycles using the column mode with permethylated β-CD impregnated resin are shown in Figure 7. It was found that, during five cycles, no noticeable change in sorption capacity of the resin occurred and comparable percentage values for the desorption of As(V) ions were achieved. These results indicate that the permethylated β-CD impregnated Amberlite XAD-4 resin was stable and easily regenerable without any physical or chemical damage, thus could be re-used. A significant reduction of the overall cost of water treatment process would be achieved by an adsorbent with good adsorption capacity, as well as showing favorable regeneration properties. The successful regeneration of an adsorbent can therefore be considered an essential action for determining the applicability of an adsorbent in water remediation practice [50]. The results of regeneration studies indicated that 0.1 M NaOH solution is worthy for this operation in the case of immobilized CD ligand and can be applied as a regenerating factor. The desorption rate was high; when a volume of 60 mL of regeneration solution was passed through column, more than 95% of As(V) ions were desorbed.
The stability and potential reusability of the impregnated resin were also evaluated by monitoring the change in recoveries of As(V) ions through several sorption and desorption cycles, for initial As(V) concentration of 1.0 mg/L. The results of adsorption/desorption cycles using the column mode with permethylated β-CD impregnated resin are shown in Figure 7. It was found that, during five cycles, no noticeable change in sorption capacity of the resin occurred and comparable percentage values for the desorption of As(V) ions were achieved. These results indicate that the permethylated β-CD impregnated Amberlite XAD-4 resin was stable and easily regenerable without any physical or chemical damage, thus could be re-used.
The stability and potential reusability of the impregnated resin were also evaluated by monitoring the change in recoveries of As(V) ions through several sorption and desorption cycles, for initial As(V) concentration of 1.0 mg/L. The results of adsorption/desorption cycles using the column mode with permethylated β-CD impregnated resin are shown in Figure 7. It was found that, during five cycles, no noticeable change in sorption capacity of the resin occurred and comparable percentage values for the desorption of As(V) ions were achieved. These results indicate that the permethylated β-CD impregnated Amberlite XAD-4 resin was stable and easily regenerable without any physical or chemical damage, thus could be re-used.

Conclusions
The permethylated β-CD impregnated Amberlite XAD-4 resin has been found efficient for adsorption of As(V) ions from water.
It was reported that the removal of As(V) using the impregnated resin depends particularly on pH and initial metal ion concentration; the highest As(V) removal efficiency (98%) was achieved at pH 6.0 and an initial concentration of 0.1 mg/L. The arsenic concentration was below the limit in drinking water (10 µg/L) then.
The adsorption parameters of As(V) ions onto impregnated resin were consistent with the Langmuir model. Moreover, using sorption models on the experimental results shows that the arsenate ion extraction mechanism onto permethylated β-CD impregnated resin was physisorption. The adsorption capacity, both in batch and column tests, was 19 mg As(V) per g of the CD ligand immobilized onto resin.
Thus, the novel adsorbent could help in extracting arsenic(V) at trace and low concentrations from the polluted water. Moreover, the permethylated β-CD impregnated Amberlite XAD-4 resin is regenerative and can be used several times.