3.1. Characterization of the adsorbent
The X-ray diffraction (XRD) analysis was conducted to characterize the precursor material, NC. The XRD patterns revealed the presence of several mineral phases, including smectite at 2θ values of 6.53 and 12.6, illite at 2θ value of 17.9, quartz at 2θ values of 21 and 27, feldspar at 2θ value of 27.95, and calcite at 2θ values of 29.3 and 23.2, as illustrated in Fig. 1. The chemical composition analysis indicates that approximately 20% by mass of the NC sample comprises calcite. These findings provide crucial information regarding the composition and structure of the precursor material, which may have implications for its potential applications [20]. Clay impregnated with Fe3O4 (MON) is indicated by the appearance of the characteristic peak for the magnetite at (2Ɵ = 35.93) which confirms the successful impregnation [21].
The Fourier-transform infrared spectroscopy (FTIR) analysis was conducted to characterize the untreated raw material, NC, and the modified adsorbent, MON, impregnated with Fe3O4. The FTIR spectrum of NC displayed several characteristic peaks. The bands at 3620 cm− 1 and 3570 cm-1 are associated with the stretching vibrations of structural hydroxyls. The stretching vibrations of adsorbed water are represented by the peaks at 3434, 1648, and 1428 cm− 1. The stretching vibrations of Si-O bonds in the tetrahedral layer are responsible for the prominent band at 1029 cm− 1, whereas the deformation of Si-O-Al and Si-O-Si groups is evident from the strong bands at 526 and 466 cm− 1, respectively. The shoulder at 629 cm− 1 corresponds to the perpendicular vibrations of octahedral cations (RO-Si), where R may be Al, Fe, or Mg. The closely spaced peaks at 912 cm− 1 (Al-Al-OH) and 874 cm− 1 (Al-Fe-OH) are attributed to the isomorphic replacement of aluminum by iron in the octahedral sheets. The vibration of silica Si-O bonds is evidenced by the peaks at 1096, 796, and 697 cm− 1, which indicate that quartz is the main contaminant in the samples. The calcite is indicated by the bands at 2520 cm− 1. The FTIR spectrum of the modified adsorbent, MON, impregnated with Fe3O4, exhibited some changes compared to the FTIR spectrum of NC. Several peaks were lost or shifted, and new peaks appeared at the surface of the impregnated adsorbent. The appearance of a new characteristic peak at 433 and 579 cm− 1 is attributed to the stretch vibration band of Fe-O in the MON, indicating the presence of clay impregnated with Fe3O4. These findings provide crucial information regarding the structural and compositional changes in the adsorbent after modification with Fe3O4, which may have implications for its potential applications. [22]. The narrow and broadband formed at 3150 cm− 1 (Fe3O4) in the FTIR spectra can be attributed to the N-H stretching vibration mode of the free NH2 group, which arises from the use of NH4OH during the synthesis of Magnetic Oxide Nanoparticles. A comparison of the spectra indicates that the interaction of Fe3O4 results in the shifting of the typical Si-O-Al and Si-O-Si group deformation bands from 529 to 579 cm-1 and 466 to 474 cm− 1, respectively [21].
Furthermore, the N2 adsorption-desorption isotherms of both the untreated natural clay (NC) and Magnetic Oxide Nanoparticles impregnated natural clay (MON) are presented in Fig. 3 and Table 1, and display a similar shape, indicating that both materials exhibit type H3 hysteresis, which is like Type II adsorption-desorption isotherms. The presence of hysteresis strongly suggests the presence of porosity, with the Type II nature at high relative pressure indicating the existence of large-size meso- and macro-pores that are not filled as saturation is approached [23]. H3 hysteresis is associated with plate-like or layered structures, which is consistent with both NC and MON, as they contain silicate layers. The key characteristic of H3 hysteresis is the sharp and abrupt closure of the desorption isotherm to the adsorption, which, for nitrogen adsorption at 77K, occurs at about 0.45 relative pressure. However, it should be noted that this closure is due to the onset of cavitation or bubble formation during the adsorption process, which makes the desorption isotherm less useful for pore size characterization. In addition, the amounts of specific surface area and pore volume of the NC and MON samples were determined to be 13.02 and 18.034 m2 g− 1 and 2.9913 and 4.1434 cm3g− 1 STP, respectively, confirming the occurrence of modification of Magnetic Oxide Nanoparticles (Fe3O4) impregnated in natural clay. The specific surface area (SBET, Ss.Lang., Sp), mean pore diameter, and pore volume (Vm, Vp) of the NC and MON samples were all calculated using Langmuir, BJH, and BET curves, from which different quantities of specific surface area were derived. Furthermore, the mean pore sizes for NC and MON were determined to be 9.5946 and 16.375 nm, respectively.
Table 1
N2 Adsorption-desorption isotherm parameters at 77 obtained for NC and MON
Sample
|
BET
|
Langmuir
|
BJH
|
SBET (cm3g-1)
|
Vm
(cm3g-1)
STP
|
Total pore volume
(cm3 g-1)
|
Mean pore diameter (nm)
|
Ss Lang. (cm3g-1)
|
Vm
(cm3g-1)
STP
|
Sp
(cm3g-1)
|
rp
(nm)
|
Vp
(cm3g-1) STP
|
NC
|
13.02
|
2.9913
|
0.03123
|
9.5946
|
18.388
|
4.2247
|
10.598
|
1.21
|
0.029753
|
MON
|
18.034
|
4.1434
|
0.073825
|
16.375
|
23.221
|
5.3351
|
18.381
|
1.21
|
0.073026
|
Furthermore, the surface morphology of NC was examined using SEM analysis, revealing a honeycomb-like microstructure, numerous curled flakes with open-air spaces that have tiny interfacial zones and mutual connections, microspheres of fibrous smectite, and flaky particle morphology [24]. The impregnation of natural clay with magnetic oxide nanoparticles was confirmed by the SEM scans of MON, as shown in (Fig. 4). The surface of NC was observed to be covered with Fe3O4 nanoparticles, increasing porosity and surface area. The embedded clay exhibited a fluffy texture and a rough, porous surface. The produced nanocomposite appeared as thicker cleaved platelets and flatter blocks due to the natural macro-porousness of the adsorbent surface, which are nanoscale iisize and cannot be easily observed. Finally, high magnification values revealed that the small magnetic nanoparticles were grouped on the talc's exposed surfaces on the outside of the nanocomposite, forming larger, shiny-looking particles [21][25].
Moreover, the energy-dispersive X-ray spectroscopy (EDS) spectrum obtained for NC and MON, as depicted in Fig. 5, shows the presence of several elements, including oxygen, aluminum, silicon, calcium, iron, magnesium, potassium, and sodium. The results of the elemental analysis of the composite are presented in the Table and the figure inset. As reported in the analysis, the weight percentages of iron for NC and MON are 5.28% and 16.58%, respectively[26]. This observation confirms the successful impregnation of Fe3O4 into the clay matrix (MON).
3.2. Adsorption studies
3.2.1. Effect of agitation time and pH
The result of the study on adsorption time which is demonstrated in (Fig. 6), is that extracellular binding led the AFD molecules to be promptly removed in the first 15 minutes. After 30 minutes, the maximum adsorption efficiency of AFD molecules increased gradually from (330.67 and 347.52) mg. g-1 for NC and MON, respectively, to approach the constant value (of 336.29 and 350.52) mg. g-1 at 120 minutes.
Additionally; The adsorption process is significantly influenced by the pH factor, which also affects the chemistry of the aqueous medium and the surface of the adsorbent. The dye solution's pH was initially determined to be about pH = 9. After being mixed with NC and MON, the pH was once again determined to be around pH = 9, which indicates that pH remained constant. Therefore, using 0.1 g of adsorbent (NC and MON) was added to 50ml of 750 mg L− 1 (AFD) for 120 min. separately and adjust pH values around (2.0–10.0) then adsorption capacity was determined and the results are illustrated in (Fig. 6). The initial pH shows no significant effect on the adsorption process.
3.2.2. Kinetic study
Various kinetic models have been proposed in the literature to identify the rate-determining step and the controlling mechanism, including mass transfer and chemical reaction mechanisms. In the present investigation, experimental data were analyzed using the pseudo-first-order equation (proposed by Lagergren, Eq. 2) and the pseudo-second-order equation (proposed by Hoo, Eq. 3) to determine the appropriate kinetic model. The fitting quality of these models was evaluated by analyzing the correlation coefficient (R2), which was used as an error analysis parameter. The results of the comparative analysis demonstrate the suitability of the chosen kinetic models for the present study.
Pseudo-first-order kinetics:\({{q}}_{{t}}={{q}}_{{e}} (1-{e}^{-{K}_{1 }t}) \left(2\right)\)
Pseudo-second-order kinetics:\({{q}}_{{t}}=\frac{{{{K}_{2 }q}_{e }}^{2}t}{1+{{{K}_{2 }q}_{e }}^{2}t} \left(3\right)\)
Where: qe (mg g− 1) is the adsorption capacity at equilibrium; qt (mg g− 1) is the adsorption capacity at time t; k1 (min− 1) is the first-order kinetic rate constant, and k2 (g mg− 1 min− 1) is the second-order kinetic model rate constant.
The information collected estimates the kinetic behaviors of the AFD adsorption processes by NC and MON utilizing the pseudo-first-order and pseudo-second-order as shown in (Fig. 7). The pseudo-first or pseudo-second-order model parameters that were used to determine the kind of adsorption (chemical or physical adsorption) were listed in (Table 2).
Allof the information and the results from (Fig. 7) and (Tables 2 & 3), like the correlation coefficient values (R2), the rate constants of the kinetic models (K1 and K2), and the calculated maximum adsorption capacity (qe) from the experiment compared with that obtained from the kinetics model, show that the data were better fitted with the Pseudo-Second-Order model as compared to the Pseudo-First-Order model.
The present study investigated the kinetic behaviors of AFD adsorption processes by NC and MON using pseudo-first-order and pseudo-second-order models. The adsorption capacity at equilibrium (qe) and at time t (qt), as well as the first-order rate constant (k1) and the second-order kinetic model rate constant (k2), were calculated. The model parameters were used to determine the type of adsorption, whether chemical or physical. The obtained results, including correlation coefficient values (R2), rate constants of the kinetic models (K1 and K2), and maximum adsorption capacity (qe), were compared to show the best-fit kinetic model.
The Pseudo-Second-Order model was found to be more appropriate for describing the kinetic behavior of AFD adsorption by NC and MON, as compared to the Pseudo-First-Order model. The correlation coefficient values of the Pseudo-Second-Order model were higher than those of the Pseudo-First-Order model, indicating a better fitting of the former. The experimental values of qe were found to be closer to the values obtained from the Pseudo-Second-Order kinetics model than to those obtained from the Pseudo-First-Order model. The rate constants from both models indicated fast-kinetic adsorption in the first 15 minutes, followed by gradual adsorption. The Pseudo-Second-Order model rate constants were higher than those of the Pseudo-First-Order model, suggesting that the faster rate of adsorption was more consistent with the former. The results also demonstrated that the adsorption of AFD by NC and MON was predominantly physical in nature.
Table 2
Kinetic Adsorption Parameters of adsorption of AFD on NC.
Kinetic
Models
|
Kinetic
Parameters
|
Temperature (K)
|
298
|
308
|
318
|
|
qexp (mg g− 1)
|
336.28696
|
336.28696
|
336.84348
|
1st Order
|
qcalc (mg g− 1)
|
330.88286 ± 2.22866
|
331.87535 ± 2.23054
|
332.41422 ± 2.17143
|
k1 (min− 1)
|
0.4444 ± 0.03403
|
0.44606 ± 0.03426
|
0.44998 ± 0.034
|
R2
|
0.83938
|
0.83758
|
0.84003
|
2nd Order
|
qcalc (mg g− 1)
|
337.89909 ± 0.31037
|
338.86834 ± 0.27311
|
339.26714 ± 0.33672
|
k2 (g mg− 1 min− 1)
|
0.00441 ± 8.82783E-5
|
0.00443 ± 7.83667E-5
|
0.00453 ± 1.00516E-4
|
R2
|
0.99814
|
0.99855
|
0.9977
|
Table 3
Kinetic Adsorption Parameters of adsorption of AFD on MON.
Kinetic
Models
|
Kinetic
Parameters
|
Temperature (K)
|
298
|
308
|
318
|
|
qexp (mg g− 1)
|
350.51884
|
350.51884
|
350.74493
|
1st Order
|
qcalc (mg g− 1)
|
348.17309 ± 0.76874
|
348.73582 ± 0.62866
|
349.29832 ± 0.56788
|
k1 (min− 1)
|
0.65977 ± 0.03385
|
0.67091 ± 0.02924
|
0.66543 ± 0.02565
|
R2
|
0.8522
|
0.88591
|
0.90995
|
2nd Order
|
qcalc (mg g− 1)
|
350.7455 ± 0.38234
|
351.09174 ± 0.39275
|
351.65255 ± 0.45773
|
k2 (g mg− 1 min− 1)
|
0.01316 ± 0.00084881
|
0.01422 ± 0.00101
|
0.01405 ± 0.00115
|
R2
|
0.97784
|
0.97298
|
0.96449
|
3.2.3. Adsorption Isotherms Study
The study of adsorption isotherms offers insights into the extent of adsorbate and adsorbent surface interaction at equilibrium, which governs the partitioning of the adsorbate between the solution and the adsorbent. In particular, the Langmuir adsorption isotherm is commonly employed to describe monolayer adsorption, which is mathematically represented by equation (Eq. 4) [27].
Where Ce is the concentration of the adsorbate (mg L− 1) at equilibrium, qm is the monolayer maximum adsorption capacity of the adsorbent (mg g− 1) and KL is the Langmuir adsorption constant (L mg− 1).
Additionally; Freundlich adsorption isotherm (Eq. 5) was formulated for multilayer adsorption on heterogeneous surfaces [28].
Where KF [(mg g− 1 (mg L− 1)1/n] is the adsorption capacity and n is the nonlinearity coefficient.
Langmuir Model: \({q}_{e}=\frac{{q}_{m}{K}_{L}{C}_{e}}{1+{K}_{L}{C}_{e}}\)(4)
Freundlich Model: \({{q}_{e}= {K}_{F}{C}_{e}}^{1/n}\) (5)
The adsorption experiments were conducted with various initial concentrations of Acid Fuchsin Dye (AFD), using a fixed amount of adsorbent (NC and/or MON), optimal pH, and a stirring period of 120 minutes at different temperatures. Langmuir and Freundlich adsorption isotherms were plotted in Fig. 8, and the parameters for the Langmuir and Freundlich models for NC and MON onto AFD were tabulated in Tables 4 and 5, respectively. The analysis of the correlation coefficients for the adsorption isotherms revealed that the Freundlich model provided a superior fit for the data compared to the Langmuir model. The magnitude of the adsorbent's adsorption is associated with the Freundlich constant (1/n). When (0.1 < 1/n < 0.5), adsorption occurred easily; (0.5 < 1/n < 1) represented a significant challenge, and (1/n > 1) signified a considerable difficulty in adsorption [29]. The low values of 1/n of the adsorbents (NC and MON) indicate that the adsorption process is favorable and easily occurred.
In this study, it was found that MON exhibits a higher Langmuir adsorption capacity, compared to NC. The enhanced removal efficiency of MON can be attributed to the variations in the lamellar dimensions of the clay substrates, and the concentration of surface charges. Furthermore, the type of layered support used was found to have a considerable effect on the nanostructure characteristics such as fractal dimensions, surface area, and porosity of the formed hybrid solids. It also affected the phase development of iron oxide crystals.
Table 4
Adsorption Isotherm Parameters of adsorption of AFD on NC.
Isotherm
Models
|
Isotherm
Parameters
|
Temperature (K)
|
298
|
308
|
318
|
Langmuir
|
qm (mg. g− 1)
|
636.51047 ± 58.69342
|
648.70967 ± 61.28684
|
659.32853 ± 62.37366
|
k1 (L. mg− 1)
|
0.07954 ± 0.03944
|
0.08029 ± 0.04079
|
0.08616 ± 0.04419
|
R2
|
0.82546
|
0.82144
|
0.82322
|
Freundlich
|
n
|
4.47221 ± 0.59528
|
4.39816 ± 0.57486
|
4.41416 ± 0.55195
|
Kf
|
151.88876 ± 27.92988
|
152.15333 ± 27.86075
|
156.83392 ± 27.26164
|
R2
|
0.9472
|
0.94933
|
0.95369
|
Table 5
Adsorption Isotherm Parameters of adsorption of AFD on MON.
Isotherm
Models
|
Isotherm
Parameters
|
Temperature (K)
|
298
|
308
|
318
|
Langmuir
|
qm (mg. g− 1)
|
666.56108 ± 63.40409
|
676.52491 ± 67.45065
|
687.57997 ± 67.95498
|
k1 (L. mg− 1)
|
0.09426 ± 0.04851
|
0.11615 ± 0.06539
|
0.13756 ± 0.07781
|
R2
|
0.82152
|
0.80648
|
0.80962
|
Freundlich
|
n
|
4.46795 ± 0.55723
|
4.60146 ± 0.52235
|
4.79285 ± 0.5048
|
Kf
|
162.64299 ± 27.74383
|
175.36688 ± 26.26707
|
190.52425 ± 25.21574
|
R2
|
0.95393
|
0.96304
|
0.96824
|
3.2.4. Thermodynamic study
The adsorption of AFD onto NC and MON was investigated at various temperatures (25, 35, and 45°C), and the thermodynamic properties were determined using the following equations to compute the Gibbs free energy change (∆G°), enthalpy change (∆H°), and entropy change (∆S°):
$$\varDelta {G}^{^\circ }= -RT\text{ln}{K}_{L} \left(6\right)$$
$$\varDelta {G}^{^\circ }= \varDelta {H}^{^\circ }-T\varDelta {S}^{^\circ } \left(7\right)$$
$$ln{ K}_{c}= \frac{{\varDelta S}^{o}}{R}- \frac{{\varDelta H}^{o}}{RT} \left(8\right)$$
Where KL: Langmuir constant, the values of KL were estimated from the intercept of the plots of ln(qe/Ce) vs. qe. Additionally; (∆H° & ∆S°) were evaluated from the slope and intercept of the linear plot of (ln KL) vs (1/T) which is illustrated in (Fig. 9) and the thermodynamic parameters are represented in (Table 6).
Table 6
Thermodynamic Parameters for the adsorption of AFD on (NC and MON)
|
NC
|
MON
|
Temp. (K)
|
∆H°
(J mol− 1)
|
∆S°
(J K− 1 mol− 1)
|
∆G°
(kJ mol− 1)
|
∆H°
(J mol− 1)
|
∆S°
(J K− 1 mol− 1)
|
∆G°
(kJ mol− 1)
|
298
|
-3.1222
|
-10.6552
|
6.2720
|
-14.9025
|
30.4100
|
-5.8513
|
308
|
6.4584
|
-5.5129
|
318
|
6.4815
|
-5.24460
|
The thermodynamics study examined the adsorption of AFD onto NC and MON at different temperatures (25, 35, and 45°C). The thermodynamic parameters, including Gibbs free energy change (∆G°), enthalpy change (∆H°), and entropy change (∆S°), were calculated using appropriate equations. The results indicated that AFD adsorption onto the NC adsorbent was nonspontaneous at all temperatures due to the positive values of ∆G°. Furthermore, the negative values of ∆H° and ∆S° indicated that the adsorption reaction was exothermic and led to a decrease in entropy in the system, respectively. On the other hand, the negative ∆G° values suggested that AFD adsorption onto the MON adsorbent was spontaneous at all temperatures. The exothermic nature of the adsorption process was indicated by the negative value of ∆H°. Additionally, the positive value of ∆S° demonstrated the affinity of MON to AFD and suggested an increase in molecular randomization at the solid/liquid interface after adsorption.