Characterization of expanded vermiculite
The crystalline structure of expanded vermiculite was investigated using X-ray diffraction analysis. The result is shown in Fig. 2a. The XRD pattern was similar to previous publications on expanded vermiculite, the peaks at 2θ = 27.2° and 34.32° correspond to d-spacing of 3.3 Å, and 2.6 Å, respectively [34]. The diffraction peak at around 60o is ascribed to the (060) plane indicating the trioctahedral type kind of the vermiculite structure [34–36]. Typically, vermiculite can exist in either dioctahedral or trioctahedral forms, although the trioctahedral variety is more prevalent in soils that have a similar structure to mica [37]. The other diffraction peaks are attributed to the other impurities such as talc, cordierite, and other clays in the vermiculite material.
Figure 2b displays the results of the thermal-gravimetric (TG) derivative thermogravimetric (DTG) study of wasted vermiculite, showing a 19.63% overall weight loss. The evaporation of physical water absorbed on the surface causes the initial 6% mass loss to occur at temperatures between 20 and 150°C [28]. Between 150 and 550 oC, the enlarged vermiculite's structure shows an 11% weight loss, which is explained by the elimination of water molecules that come into contact with the cations in the interlayer area [35]. There is a 2.6% weight loss in the range of 550 to 850 ºC, which is explained by dehydroxylation. As previously reported, the endothermic peak at 904 ºC may indicate the creation of a new enstatite crystalline phase [36].
The SEM images show the morphologies of expanded vermiculite, the expanded vermiculite appeared as a flake-like structure with the thickness in nanoscale, as shown in Fig. 3 (a,b). Expansion is related to the separation of layers as agents penetrate and then release from the solid mass of the mineral rock. The expansion process leads to significant changes in the surface morphology of vermiculite [28]. The expansion procedure results in a vermiculite surface area reaching 73.953 m2/g instead of 17.718 m2/g before expanding (Fig. 3c). The enlarged vermiculite EDX spectrum (Fig. 3d) matches the elemental chemical composition of vermiculite, which was mainly made up of SiO2, MgO, Al2O3, Fe2O3, CaO, K2O, and TiO2 [38]. Furthermore, the percentage of this element as oxide is consistent with previously published percentages [39].
The efficiency in removing Alizarin Red S
Effect of pH
Because H+ ions strongly impact both the adsorbent and the adsorbed substance, the pH directly affects the adsorption equilibrium. The ability to remove alizarin red S from expanded vermiculite decreased when increasing pH from 3 to 7 (Fig. 4a). The results show that pH affects the nature of the investigated adsorption process. The adsorption mechanism may be related to the electrostatic attraction of the dye's functional groups to the adsorbent's functional groups. Because of its negatively charged sulfonate groups (-SO3−) in an aqueous solution, ARS is an anionic dye. The surface of the enlarged vermiculite particles will be positively charged at pH < pHpzc (7.086), primarily due to (-COOH2+), (-OH2+), and extra active sites (Met-OH2+) (Fig. 4b).
Effect of adsorbent dose
Fixed: alizarin red S concentration of 10 mg/L, ambient temperature, and pH = 3.0. By varying the adsorbent range from 0.25-2.0 g/L, the adsorption capacity of ARS on expanded vermiculite was investigated. Figure 5 shows that the performance does not increase linearly with the increase in adsorbent dosage. The graph illustrates the rapid increase in the adsorption rate of a 10 g/L dye solution as the adsorbent dosage rises from 0.25 to 0.50 g/L. Then, when increasing the adsorbent dosage, the adsorption rate increased insignificantly. Sometimes, increasing the dosage of expanded vermiculite reduces its adsorption capacity. Therefore, the optimal amount of adsorbent used for further experiments is 0.5 g/L.
Adsorption kinetics
Researching the adsorption process's kinetics is crucial to considering the possibilities of employing solid adsorbents. The pseudo-second-order equation and the Lagergren pseudo-first-order equation, two of the most popular kinetic models, have been utilized to analyze the adsorption kinetic behavior of ARS on expanded vermiculite.
One way to express Lagergren's pseudo-first-order kinetic equation is as follows [14]:
$$ln\left({q}_{e}-{q}_{t}\right)=ln{q}_{e}- {k}_{1}t \left(3\right)$$
Furthermore, the pseudo-second-order kinetic rate equation is as follows [14]:
$$\frac{t}{{q}_{t}}= \frac{1}{{k}_{2}{q}_{e}^{2}}+ \frac{t}{{q}_{e}} \left(4\right)$$
where qe and qt (mg/g) represent the adsorption capacities at equilibrium and time t, respectively; k1 denotes the pseudo-first-order adsorption rate constant (min− 1); k2 denotes the pseudo-second-order adsorption equilibrium rate constant (g/mg.min).
The time dependent adsorption data of ARS on expanded vermiculite was applied with the two kinetic models mentioned. Figure 6 presents the graph representing the pseudo-first-order kinetic model and the experimentally calculated parameters. The pseudo-first-order kinetic model's correlation coefficient is low (R2 = 0.7693), with a significant difference between the experimental qe (7.64 mg/g) and the calculated qe (57.74 mg/g).
Table 2
Parameters of pseudo-first-order and pseudo-second-order kinetics.
Pseudo 1st order | Pseudo 2nd order |
k1 (min− 1) | qe (mg/g) | R2 | k2 (g/mg.min) | qe (mg/g) | R2 |
0.0454 | 57.74 | 0.7693 | 0.0045 | 8.31 | 0.8964 |
In this model calculation, the adsorption capacity reached equilibrium at 8.31 mg/g, and the pseudo-second-order kinetic rate constant (k2) reached 0.0045 g/mg.min. The correlation coefficient in the pseudo-second-order kinetic model was close to the standard value (R2 = 0.8964). The adsorption process of alizarin red S on expanded vermiculite may be a physical adsorption mechanism due to electrostatic attraction.
Adsorption isotherm
Three mathematical models (Temkin, Freundlich, and Langmuir) have been used to describe the equilibrium data. The experimental data were used to solve five linear equations in Table 1 using the Langmuir isotherm adsorption models, displayed in Table 3 and Fig. 7. Because of their optimal error distribution, they are widely used. While Langmuir equations 1 and 5 give quite similar results and high linear coefficient (R2 = 0,9905). However, the Langmuir equations 2, 3, and 4 have low value of linear coefficients (R2 = 0,717; 0,5613 and 0,5613; respectively), along with the qo values also having large deviations, the difference between the lowest and highest values is 3 times apart. In addition, the Langmuir constant (KL) indicates the degree of contact between the adsorbate and the surface. A higher KL value suggests a more substantial contact between the adsorbent and the adsorbate, whereas a lower value suggests a weaker relationship. However, the KL values in Table 3 demonstrate that the interaction between adsorbent and adsorbate is a weak interaction even though the qo results do not reflect the same result. Because systematic mistakes occur when non-linear functions are transformed into linear models, there is strong evidence that choosing the best-fit model for linearized equations can be done using criteria other than the linear coefficient.
Table 3
The parameters of the linear Langmuir isotherm model for ARS adsorption.
Isotherm model | qo (mg/g) | KL (L/g) | R2 |
Langmuir-1 | 163.93 | 0.0056 | 0.9905 |
Langmuir-2 | 85.47 | 0.0119 | 0.7170 |
Langmuir-3 | 60.10 | 0.0189 | 0.5163 |
Langmuir-4 | 100.45 | 0.0098 | 0.5163 |
Langmuir-5 | 182.50 | 0.005 | 0.9905 |
The Freundlich and Temkin equations were also linearized to represent the data (Table 1), and Table 4 and Fig. 8 displays the determined model parameters. According to the Freundlich and Temkin isotherm models, the linear coefficient (R2) provides good results compared to the linear Langmuir equations. However, it is marginally less than the values of the Langmuir-1 and Langmuir-5 equations, 0.9761 and 0.9802 for the Freundlich and Temkin equations, respectively.
Table 4
Parameters of the Temkin isotherm and linear Freundlich model for ARS adsorption.
Freundlich isotherm | Temkin isotherm |
n | KF (L/g) | R2 | AT (L/g) | B | R2 |
1.168 | 1.240 | 0.9761 | 0.2339 | 11.489 | 0.9802 |
The Freundlich isotherm model states that the degree of adsorption was correlated with the constant KF, and the approximate intensity of the adsorption was given by n [38]. Adsorption capacity was considered good if its value fell between 2 and 10, moderate between 1 and 2, and poor if it fell below 1. The current study's magnitude of n (1,17) indicates that the adsorption process is relatively challenging [31]. Applying the data to the Temkin isotherm model reveals that the adsorption of ARS on expanded vermiculite is exothermic, as indicated by the constant B (11.489) positive value and the adsorption energy parameter bT (216 kJ/mol) positive value variation. Low bT values also suggest a lack of interaction between the adsorbent and ARS molecules. The evaluation of the Freundlich isotherm adsorption model is similar to this.
The most appropriate model is the Langmuir isotherm adsorption model, which is evaluated just by the linear coefficient. However, the difference in data calculated from the 5 linear equations of the Langmuir model gives different results. While the KL value reflects the weak interaction of adsorbent and surface, the q0 value gives the opposite result (60–180 mg/g), and this value does not match the experimental data. The coefficient of the data computed using the Freundlich and Temkin models is smaller than that of the Langmuir-1 and Langmuir-5 equations, indicating a feeble interaction between the adsorbent and adsorbate. It illustrates how weakly the adsorbent and adsorbent interact, and how moderately difficult it is for the adsorption process to match the experimental results.
The maximal adsorption capacity of ARS by expanded vermiculite was calculate from the Langmuir model to be 182.5 mg/g, which is comparable to other absorbents reported previously (Table 5). However, vermiculite is more cost-effective and its production process is easily scalable. This study suggests that expanded vermiculite might be more readily used in practical applications compared to other adsorbents.
Table 5
Adsorption capacity of various adsorbents for the ARS dye.
Adsorbent | ARS maximal adsorption capacity, mg/g | References |
Activated carbon | 8.29 | [39] |
APTES grafted sonicated vermiculite | 18.2 | [28] |
Mesoporous hybrid gels | 30.7 | [40] |
Activated clay modified by iron oxide | 32.7 | [11] |
polyethyleneimine (PEI)-functionalized magnetic carbon nanotubes | 174.9 | [21] |
NiFe2O4/polyaniline magnetic composite | 186 | [41] |
Fe3O4 nanoparticles | 140.8 | [42] |
CoFe2O4 nanoparticles | 192.3 | [42] |
Ionic liquid-Fe3O4 nanoparticles | 256.4 | [42] |
Expanded vermiculite | 182.5 | This study |
Recyclability of expanded vermiculite for ARS
The reusability of a saturated adsorbent is a critical aspect for the practical application, as regenerating the adsorbent is essential for the sustainability of the process. In addition, the process of restoring the adsorbent is more cost-effective than replacing it, which eliminates the need for directly disposing of the spent adsorbent. For this purpose, the suitable solution for the ARS desorption from adsorbed vermiculite was investigated. Several solutions including 0.1M NaOH, 0.1M HCl, acetone:water (3:2, v/v), anhydrous ethanol (EtOH), and 0.1 M NaOH/EtOH (1:1, v/v) were employed to remove ARS from adsorbed vermiculite. The experimental results indicated that the removal efficiency following the initial regeneration with the aforementioned desorbing agents was 85%, 71%, 90%, 72%, and 95%, respectively. This suggests that the NaOH/EtOH solution can be considered the most effective desorbing agent for the removal of ARS from vermiculite. To carrying out the recycling test, 0.1 g of expanded vermiculite was introduced to the 50 mL of 10 mg/L ARS solution at pH 3 for 60 minutes at room temperature. The NaOH/EtOH solution was employed to regenerate the adsorbent after each testing cycle. The removal percentage of ARS was measure to evaluate the recyclability of the expanded vermiculite after each cycle. The results are exhibited in Fig. 9. The result shows that the reduction in ARS removal efficiency by the adsorbent after 5 cycles is less than 10%, demonstrating that the expanded graphite is durable and efficient for the practical application.