The polyanion was prepared according to the reported literature procedure [18]. Na₂WO₄⋅4H₂O (182 g, 0.55 mol) is dissolved in distilled water (300 mL) and the solution is heated to reflux. Aqueous 4M HCl (165 mL) is added dropwise under vigorous stirring to the heated solution over a period of 30 min. Crucially, any precipitate (tungstic acid) formed is allowed to re-dissolve before further acid addition. An aqueous solution of sodium metasilicate (11 g, 50 mmol in 100 mL distilled water) is added followed by quick addition of aqueous 4M HCl (50 mL), giving a solution pH of 5.0. The solution is refluxed for 1 h.

After cooling to room temperature, KCl (150 g) is added, and a colorless precipitate is formed. The solid product is collected by filtration, washed with two 50 mL portions of aqueous KCl solution (1 M) followed by two 30 mL portions of cold water. The crude product is recrystallized from water, filtered and the colorless crystalline product is dried in the desiccator under vacuum.

The undecatungstophosphate was prepared using the same method according to the literature procedure. [18]

In 300 mL beaker, 54 g of Na₂WO₄·2H₂O (0.164 mole) and 2.1 g of Na₂HPO₄ (0.015 mole) were dissolved in 150 mL distilled water. The solution was heated and stirred at 90˚C and 60 mL of (4 M) HCl was added dropwise by burette over one hour. The final pH of the solution should be adjusted to 5.5 - 6.5, if necessary. The solution was kept at 90˚C for another hour and the pH was checked again. If the pH remained unchanged, 21 g of solid KCl was added to the solution and dissolved. Upon cooling to 0˚C the white precipitate was isolated from the solution, and it was recrystallized twice from hot water. The salt was checked with IR and EDS and the results were in agreement with those reported in literature.

In a round-bottom flask, a solution of K_y [α-XW₁₁O₃₉]∙xH₂O (1.00 eq.) was dissolved in 50 mL of water, heated to 50˚C and a solution of adequate amount of TOA-Br (8.00 eq for {SiW₁₁O₃₉} and 7.00 for {PW₁₁O₃₉}) in toluene (80 mL) was added. The mixture was vigorously stirred for 5 minutes and the organic layer was separated. After removal of the solvent under reduced pressure, the light yellow highly viscous liquid was solvent-stripped once with 50 mL toluene and three times with 50 mL chloroform.

0.5 g of [N(C₈H₁₇)₄]_yXW₁₁O₃₉ was dissolved in 50 mL of acetone and then 4 g of Bentonite were added. The suspension was gently shaken for 30 minutes and the solvent was removed by rotary evaporator. Further 50 mL of acetone were added, and the shaking/solvent removal procedure was repeated three times. The final product was obtained as a solid, free-flowing powder. The POM-IL loading corresponds to 12.5 wt%.

Adsorption test for MB removal from aqueous solutions was examined using adsorption batch experiments to explore the adsorption properties and the factors influencing the adsorption. Batch experiments were performed by varying contact time, mass of adsorbents, concentration of dyes, pH, ionic strength and temperature.

The following conditions were preserved for the different sets of experiments:

1) The effect of pH was conducted in the range from 2.0 to 10.0, 30 mg of SWB and 40 mg of PWB, 100 mL solution volume, 50 ppm MB concentration and T = 298 K. HCl(0.01 M) or NaOH (0.01 M) was added to set the desired pH value.

2) Effect of contact time was carried out through the time intervals (10, 20, 30, 40, 50, and 60 min), 30 mg of adsorbent, 100 mL solution, 50 ppm MB concentration, 298 K, initial non modified pH.

3) The effect of adsorbent mass was examined using different doses (20, 30, 40 and 50 mg) of adsorbent, initial non modified pH, time 60 min, 50 ppm MB concentration, 100 mL solution volume, and temperature 298 K.

4) The effect of initial concentration was conducted using 50, 100, 150 and 200 mg/L of MB, initial non modified pH, time 60 min, adsorbent mass 30 mg of SWB and 40 mg of PWB, and temperature 298 K.

5) The Effect of ionic strength was conducted by adding NaCl, KCl and NaNO₃ salts at different concentrations ranging from 0.1 to 0.4 M, initial non modified pH, time 60 min, 50 ppm MB concentration, 100 mL solution volume, and temperature 298 K.

6) The influence of temperature on the adsorption process was studied under different temperatures of 298, 313, 328 and 343 K, initial non modified pH, time 60 min, 50 mg/L MB concentration, adsorbent mass 20 mg, solution volume 100 mL.

In all experiments, a water bath shaker at 300 rpm was used to set the desired temperature and duration of the studied mixture. After adsorption, the dye was kept apart from the adsorbent by centrifugation with 3000 rpm speed and use UV-VIS spectrophotometric technique to determine the residual concentration C based on Beer-lambert equation.

The percentage removal and adsorption capacity at time (Q_t) were calculated by using Equations (1) and (2), respectively.

R % = C 0 − C f C 0 × 100 (1)

Q t = C 0 − C f m × V (2)

C₀ and C_f: initial and final concentrations (mg/L) of dye solutions. V: volume (L). m: mass (g) of the adsorbent.

The FTIR spectra of raw bentonite, SWB and PWB in the range of 4000 - 400 cm⁻¹ were shown in Figure 2. The FTIR transmittance bands in the low frequency region (1200 - 400 cm⁻¹) of the modified Bentonite and raw Bentonite are largely comparable indicating that the clay mineral has not changed upon modification by polyoxometalate.

Absorption due to Al-O-Si bending at 527 cm⁻¹, Si-O-Si bending at 468 cm⁻¹, and Al-Al-OH bending at 914 cm⁻¹ confirm the presence of montmorillonite [21] [22]. The very strong characteristic peak observed at 1032 cm⁻¹ was ascribed to asymmetric stretching vibrations of Si-O-Si bonds. The broad band in the 3430 - 3623 cm⁻¹ region can be ascribed to the symmetric stretching of (Si-OH, Al-OH) structural hydroxyl groups OH and of the physically adsorbed water OH. The shoulder at 3694 cm⁻¹ and the weak band at 695 cm⁻¹ suggest the presence of kaolinite. The absorption band at 1639 cm⁻¹ is attributed to the

angular vibration of the OH group and related to the adsorbed water and the hydration water present in the clay.

New absorption bands emerged at 2950, 2854, 1468 and 1382 cm⁻¹ which confirm the presence of TOA cations. The fingerprint of POM is not clearly visible in FT-IR spectra of SWB and PWB due to their overlapping with FT-IR bands of Bentonite.

After adsorption of MB dye some bands appeared, such as bands at 1488 cm⁻¹ and peaks at 1394 - 1335 cm⁻¹ region related to the −CH₃ and the feature conforming to the C=C skeleton stretching at 1600 cm⁻¹ characteristic of the aromatic ring vibrations of MB dye.

The SEM images of the raw Bentonite showed an irregular porosity and disperse surface with large number of cavities Figure 3(A). In contrast, the morphology of SWB sample showed clear changing in the surficial properties of the clay Figure 3(B). The surface becomes smoother, flake like morphology with smaller cavities than raw bentonite, this result is confirmed by BET analysis. The Scanning Electron Microscopy showed that the bentonite layer-stacking structure was retained even after incorporating polyoxometalate ionic liquid into the bentonite matrix.

EDS analysis Table 1 showed that natural Saudi bentonite is mainly composed of silicon dioxide, Aluminum oxide, iron oxide, sodium oxide, in addition to calcium, magnesium, titanium and potassium oxides. Comparison of raw and modified bentonite revealed no significant change of the chemical composition of the bentonite matrix.

The X-ray patterns of raw Bentonite, PWB and SWB are illustrated in Figure 4.

The XRD analysis of raw bentonite sample showed that bentonite is mainly composed of, quartz (SiO₂), calcite (CaCO₃) and Feldspar as non-clay minerals.

XRD patterns diffractograms revealed three main clay minerals, montmorillonite, kaolinite and illite. Moreover, the characteristic diffraction signals were found to be identical for both raw bentonite and (TOA)_yXW₁₁@Bentonite which indicate that the POM-IL remains on the surface of the clay.

The X-ray diffractogram recorded in a range of small angle (2θ) ranging from 0˚ to 10˚ Figure 5 shows the presence of a characteristic diffraction peak located at (2θ) = 5.68˚ from which the interlayer distance was found to be 14.08 Å. This peak is associated with the (001) diffraction (d₀₀₁) of the montmorillonite phase. [23] [24] The introduction of POM to the Bentonite has not affected the d₀₀₁ diffraction peak which appears at 14.86 Å for SWB and 15.58 Å for PWB since there is no significant expansion of interlayer space Δd (4.38Å and 5.05 Å).

This finding confirmed that the structure of Bentonite is retained after POM-IL incorporation.

Moreover, the presence of POM in the prepared adsorbent is confirmed by the presence of structural characteristic peaks at 2θ: 37.20˚, 65.80˚, 66.83˚ and 78.77˚ in agreement with the Keggin structure [25] [26]. The data also show that quartz major component is not affected during POM incorporation to the clay.

TEM characterization carried out on the raw bentonite and POM modified bentonite are presented in Figure 6(A). TEM analysis confirmed the XRD results and reveal the existence of layers at basal space d₀₀₁ = 14.08Å for bentonite and d₀₀₁ = 14.75 Å for POM modified bentonite. Moreover, TEM shows that POM molecules are located on the surface of Bentonite (dark spots in Figure 6(B)) and confirm the nanosized Keggin structure (~1 nm).

Nitrogen sorption was used to investigate the specific surface area (BET method), pore diameter and specific pore volume (BJH method) of the raw bentonite and

POM-IL modified bentonite to establish the consequences of POM-IL adsorption on the bentonite pore structure. The data in Table 2 show that upon POM-IL adsorption in POM-IL@bentonite, an overall reduction in pore diameter, specific pore volume and specific surface area is observed. This verifies that the POM-IL does not only coat the exterior surface but also coats the internal pore structure while leaving the pores still accessible.

The effect of contact time on Methylene Blue removal by SWB, PWB and bentonite is shown in Figure 7. It could be seen that the MB adsorption increases with increasing contact time for modified and raw bentonite and occurred in two stage: first fast stage and equilibrium stage. SWB adsorbent has higher removal than PWB during the rapid first stage and reached 93% removal after 10 min compared to 66% for PWB for the same time. The difference can be explained by the nature of the heteroatom in the polyoxometalate which plays a role in the overall charge of the polyanion making [SiW₁₁O₃₉]⁸⁻ more negative than [PW₁₁O₃₉]⁷⁻ which in turn attracts more cationic MB dye since the adsorption is due to electrostatic attraction between negative adsorbent surface and positive dye molecules. The rapid MB removal during the first stage (10 min) is due to presence of high number of available vacant sites on the adsorbent surfaces, the adsorption reached equilibrium within almost 30 min with total removal. Further increase in contact time will no longer improve the adsorption percent removal. There is also a significant difference in percent removal between raw bentonite, SWB and PWB which clearly confirms the enhancement in bentonite adsorptive performance with POM-IL modification.

The effect of adsorbent mass was studied by variation of SWB and PWB dose from 20 to 60 mg. Figure 8(A) clearly shows that the % removal of MB increased with increasing dose of adsorbents. This is due to the increasing available vacant sites which, therefore increases the percent removal. However, the adsorption capacity (Q_e) Figure 8(B) decreased as the adsorbent dose increased. This is because, the large adsorbent mass effectively decreases the number of covered sites per unit mass comes down, which results in a reduced capacity.

By comparison with raw bentonite, we noticed that the capacity uptake of POM-IL bentonite for both adsorbents SWB and PWB is higher than that of raw bentonite in the range of 20 - 60 mg. such a result shows clearly that the incorporation of POM-IL in the bentonite matrix enhances the adsorptive capacity of the bentonite at low adsorbent dose.

The effect of initial concentration on the removal of methylene blue on SWB,

PWB and bentonite is studied in the range from 50 - 200 mg/L. The results are reported in Figure 9. Several studies have documented the effect of initial dye concentration on dye uptake by clays. In most cases the increase of initial concentration resulted in an increase of the adsorption capacity and a decrease of the overall removal efficiency. In this study, the % removal of dye decreased with increasing initial dye concentrations. At low concentration (50 ppm), the removal reached almost 100% for both POM-IL modified bentonite and raw bentonite. Over loading of methylene blue above the saturation level did not find available vacant site resulting on the formation of accumulation of methylene blue molecule on the surface of adsorbent and the % removal decrease. At concentrations higher than 100 ppm, SWB adsorbent has higher % removal when compared to PWB and Bentonite which have similar results. The adsorption capacity (Q_e) increased gradually with the increase in concentration Figure 9(A). However, at lower dye concentration, the adsorption is independent on initial concentration since the ratio of the number of MB cations to the number of available adsorption sites is small.

The pH effect on the adsorption of MB dye on SWB and PWB was investigated in the range of 2 - 10. As shown in Figure 10, there is no significant effect on the adsorption of methylene blue while changing the pH solution values, SWB and PWB are stable adsorbents over a wide pH range. Independent of the origin of this insignificant effect of the pH, this finding is quite meaningful in the adsorption process application since it makes complicated pH adjustment of the effluent before treatment unnecessary.

As a result, pH was not adjusted during adsorption experiments to simulate real treatment condition of industrial effluents.

The effect of temperature on MB dye adsorption onto SWB and PWB was conducted at 5 different temperatures ranging from 303 to 343 K.

Figure 11 shows that the removal percentage of MB increased with the increase in temperature, indicating that MB adsorption on both adsorbent surfaces was favored at higher temperatures. The increase in adsorption with increasing

temperature indicated the endothermic nature of the adsorption process.

The study the effect of salts on the adsorption of dyes is important since wastewater contains different concentrations of salts Figure 12 shows the influence of salt concentration on MB adsorption onto SWB and PWB. As a whole, the presence of salts reduced the capacities of SWB and PWB on adsorbing MB dye. Both the % removal and the adsorption capacity of the POM-IL modified bentonite began to decrease when high concentrations of salts were added. This could be partly ascribed to the competitive adsorption between Na⁺, K⁺ and MB cations for available active sites. As shown in Figure 12, the reduced affinity between negatively charged POM-IL modified bentonite and MB cationic dye observed in presence of KCl is due K⁺ smaller hydrated radius (2.32 Å) than Na⁺ (2.76 Å); in this case K⁺ could easily occupy more adsorption sites and could not as easily be leached. Similar results were reported by De Castro et al. when studying the adsorption of MB onto EDTA modified bentonite [27].

It is noticed that KCl inhibiting effect on the adsorption of MB is more pronounced in the case of SWB compared to PWB (Figure 12 and Figure 13).

The effect of chloride and nitrate anions on the MB adsorption is comparable Figure 13, because these negatively charged anions can compete with the bentonite as sites for electrostatic attraction and disrupt MB adsorption [27].

This model is one of the most extensively used models to describe the adsorption kinetics. The equation for pseudo-first order is:

d t d Q = k 1 ( Q e − Q t ) (3)

Q_t and Q_e are dye amount adsorbed (mg/g) at time, t (min) and at equilibrium, respectively, k₁ is the rate constant of adsorption (min⁻¹).

The linear form for pseudo first order can be expressed as:

ln ( Q e − Q t ) = ln Q e − k 1 t (4)

When plot of ln ( Q e − Q t ) vs t drawn, Q_e and k₁ values can be obtained from intercept and slope respectively.

Pseudo-second order suggests that the uptake process is due to chemisorption and has the general form as:

Q t = k 2 Q e 2 1 + k 2 Q e t (5)

Mathematically it can also be expressed as follows:

t Q t = t Q e + 1 k 2 Q e 2 (6)

where k₂ ((g/mg)/min) is the constant for second order, Q_e (mg/g) and Q_t (mg/g) the amount of dye adsorbed at equilibrium and any time respectively.

The kinetics of MB dye adsorption onto SWB and PWB were analyzed using pseudo-first-order PFO model (Equation (4)) and pseudo-second order PSO model (Equation (6)).

By applying the two kinetic model, it was found that the plot of t/q_t vs. t for PSO model fit for the whole range of contact time for both adsorbent, with excellent linearity Figure 14. On the other hand plots of ln ( Q e − Q t ) vs. t displayed less-linearity.

The kinetic parameters for PFO and PSO were calculated and tabulated in Table 3.

The real test for the validity of the kinetic model arises from the comparison between the experimentally determined q_e and the calculated q_e [28] [29]. According to the reported results it is clear that in the PSO kinetic model, the Q_ecal is very close to the q_e exp values. In addition, the R² values > 0.99 for the PSO model were higher than those obtained in PFO model. Moreover, in pseudo-second-order a trend was observed with the rate constants, k₂ generally decreased as the initial concentration increases, which is due to the occupation of active sites. As a result of the above-mentioned reasons, it can be concluded that

the pseudo-second-order model best fit the experimental equilibrium data.

The Langmuir isotherm assumes monolayer adsorption, absence of interaction between adsorbate molecules and homogeneous distribution of active sites on the adsorbent surface [22]. The general equation for Langmuir is:

Q e = Q max k L C e 1 + k L C e (7)

where, Q_e is the dye adsorbed amount at equilibrium (mg/g), Q_max is the maximum adsorption capacity (mg/g). k_L is the constant of Langmuir and C_e is dye concentration at equilibrium.

It can also be expressed as:

C e Q e = 1 Q max k L + C e Q max (8)

The value of (1/Q_max) and K_L can be obtained from linear curve of (C_e/Q_e) versus C_e.

The R_L separation factor values provide information about the affinity between adsorbate and adsorbent which can be calculated using the following Equation (9):

R L = 1 1 + k L C 0 (9)

where K_L is the Langmuir constant, C₀ (mg /L) is the initial concentration of dye.

Table 4 shows the types of Langmuir isotherm according to R_L Values.

The Freundlich equation is empirical and assumes heterogeneous multilayer adsorption, its mathematical equation is express as:

Q e = k f C e 1 / n (10)

where C_e is the equilibrium dye concentration (mg/L), Q_e is the dye adsorbed amount at equilibrium (mg/g), k_f are the Freundlich constants and n indicative of the extent of the adsorption and the degree of nonlinearity between solution concentration and adsorption intensity.

It can also be linearized in logarithmic form as shown below:

log Q e = log k f + 1 n log C e (11)

The value of (1/n) and logk_f can be calculated from slope and the intercept respectively from linear plot of logQ_e verses logC_e.

Table 5 summarizes the calculated coefficients of determination (R²) and model parameters of the Langmuir and Freundlich isotherms. The results show clearly that the Langmuir isotherm (R² > 0.98 for SWB and R² > 0.97 for PWB) best fits the equilibrium data for both adsorbents. This indicates that MB adsorption onto SWB and PWB is homogeneous. The maximum adsorption capacities for MB onto SWB and PWB were 113 and 277 mg∙g⁻¹, respectively.

To predict the adsorption efficiency, the essential characteristics of Langmuir isotherm was explained in term of a dimensionless constant separation factor R_L.

As shown in Figure 15(B), the R_L values calculated for both SWB and PWB

were all in the range of 0 - 1. Moreover, the R_L values were also observed to decrease when the MB dye initial concentration were increasing which means that the adsorption becomes more favorable. This could be due to the increase in driving force when more MB solute was present.