Many techniques were used to characterize the different adsorbents. XRD of the raw materials is retrieved on an XG actinism diffractometer applying CuKα emission (λ = 1.5418 Å) through 30,000 V/15mA with a scanning speed of 2˚·min⁻¹ and the 2θ angle between 5˚ - 80˚. The FT-IR spectra are obtained from a spectrometer brand spectrum 100 (USA) in the central zone of (4000 - 400 cm⁻¹). The TG-DSC spectra are obtained from an apparatus in which the increase in the heating rate is from 1˚C·min⁻¹ at ambient temperature upper limit of 800˚C in atmospheric air. SEM coupled with EDX was performed on the JOEL JBN-5600 instrument. Brunauer-Emmett-Teller (BET) surface area of raw and modified clay materials was determined by micrometrics ASAP 2020 V3.00 H. Pore size distribution was calculated by the Barrett-Joyner-Halenda (BJH) method.

The adsorption studies in batch mode were carried out using the 250 mL conical flasks (Kimax, USA) as batch reactors. To perform the kinetic study, eight 250 mL conical flasks each containing 0.1 g of adsorbent and 20 mL AO52 dye solution with a concentration of 10 mg·L⁻¹, were stirred on a multistation shaker set at a speed of 250 rpm at ambient temperature (25˚C ± 2˚C) at different times 3, 5, 10, 15, 20, 25, 30 and 40 min. Adsorption isotherm studies were performed at AO52 concentrations varying from 4 - 12 mg/L. After the adsorption process, to determine the residual concentration, the filtrate was assayed using UV-visible spectrophotometer (Model-S23A) at the wavelength of the maximum adsorption of 465 nm. The Metrohm brand 744 pH meter was used to measure the pH of

AO52 dye solution. To study the influence of the pH of the AO52 solution, the solutions of HCl or NaOH (0.1 N) were used to adjust the pH of the initial solution of dye by varying from 2 - 12. The reactors contained 0.1 g sample and 20 mL of 10 mg·L⁻¹ AO52 dye that was reacted for a specific time at ambient temperature. The effect of temperature on the adsorption kinetics of AO52 dye was taken into account at variable temperatures 25˚C, 30˚C, 40˚C and 60˚C, 0.5 g of adsorbent was added in 100 mL of 10 mg/L of AO52 dye and brought to stirring for the equilibrium time of each adsorbent. The amount of AO52 dye adsorbed at equilibrium (q_e) and adsorption percentage (%R) of all the clay materials are calculated according to the equations below.

q e = C o − C e m × V (1)

% R = C o − C e C o × 100 (2)

where q_e (mg·g⁻¹) is the equilibrium amount of dye adsorbed, C_o and C_e are the initial and equilibrium AO52 dye concentrations respectively, m (g) is the sample weight, V is the dye solution volume (mL).

The kinetic models used to describe the adsorption kinetics of AO52 dye onto clay materials in aqueous solutions are given in Table 2.

Langmuir and Freundlich isotherm models were used to describe the adsorption mechanism of AO52 dye as shown in Table 3.

The essential characteristic (R_L) of the Langmuir isotherm is defined by the following equation [16] .

R L = 1 1 + K L C o (3)

The nature of the adsorption process depends on the R_L values: If R_L > 1: unfavorable;

R_L = 1: linear; R L ∈ ] 0 ; 1 [ : favourable and R_L = 0: irreversible.

The thermodynamic constants such as Gibbs energy (ΔG^°), enthalpy (ΔH^°) and entropy (ΔS^°) were calculated by using both the thermodynamics and Van’t Hoff equations [17] .

K d = C a d s C s o l u t i o n (4)

Δ G ° = − R T ln K d (5)

ln K d = Δ S ° R − Δ H ° R T (6)

In which R is the universal gas constant, T is the temperature in (K), and K_d is the partition measure, C_ads is the concentration of the solute adsorbed on the samples in (mg·g⁻¹) and C_solution is the first measured initial concentration of the solute (mg·g⁻¹).

Figure 2 shows the FT-IR spectra of Kao (a), Na-MMT (b), F13 (c) and F23 (d) respectively. Figure 2(a) and Figure 2(b) were utilized to validate the formation of formulated clay lime mixtures F13 and F23. Figure 2(a) indicates that the absorption bands at 3694 and 3619.88 cm⁻¹ in the Ourlago-kaolin sample, were attributed to the intermediate layer hydroxyl (O-H) assemblage and the group at 3433.14 cm⁻¹ was assigned to the O-H bonds pointing in the layers of the tetrahedral sheets [18] . Absorption bands at 993.33 cm⁻¹ were related to the stretching Si-O bonds, whereas the absorption band at 523.28 cm⁻¹ characterized the Al-O-Si bond in kaolinite. The very low absorption band at 795.22 cm⁻¹ shows the presence of quartz in the clay materials.

As shown in Figure 2(b), the characteristic bands of bending vibration bond Si-O of quartz appear at 524.75 cm⁻¹ and 644.75 cm⁻¹; 690.50 and 419.39 cm⁻¹ [19] . In this figure, the absorption band at 3696 cm^-1 can be attributed to the stretching vibration of the inner OH located on the edge of the layers. Absorption bands at 3619.87 and 1631.75 cm⁻¹ correspond to the O-H stretching vibration and bending vibration of H₂O of sodium montmorillonite. The Si-O and Si-O-Si bonds can be attributed to the absorption bands at 996.16 cm⁻¹ and

774.28 cm⁻¹ respectively [20] .

Figure 2(c) and Figure 2(d) show the characteristic peaks around 1300 cm⁻¹ centered at 1214 and 1219 cm⁻¹; the one at around 730 cm⁻¹ centered at 750.87 cm⁻¹ and 743.94 cm⁻¹ and the wide and strong bond at around 500 cm⁻¹ centered at 529.51 cm⁻¹ and 527.07 cm⁻¹ respectively for F13 and F23 related to bending vibration Ca-O bonds. Figure 2(d) shows that the absorption bands at 3691 cm⁻¹ and 3620.02 cm^−1, can be attributed to the interlayer hydroxyl (O-H) and the O-H bonds present in the layer of tetrahedral sheets respectively. The most intensive bands 1026.94 and 1001.15 cm⁻¹ for F13 and 1003.20 cm⁻¹ for F23 spectra were assigned to the stretching Si-O vibrations of the layer of tetrahedral sheets and those at 461.94 cm⁻¹ for F13 and 461 cm⁻¹ for F23 are assigned to the bending Si-O-Si vibrations. The characteristic bands at 909.40; 910.41; 909.38 and 909.34 cm⁻¹ for Kao, Na-MMT, F13 and F23 respectively are related to Al-O-H vibration. The formation of the formulated clay lime mixtures was confirmed by the disappearance of the absorption bands at 684.71 and 690.50 cm⁻¹ for Kao and Na-MMT respectively and the appearance of those at 1214 and 1219 cm⁻¹ on the F13 and F23 FT-IR spectra respectively.

It can be observed, from the above results, an excellent correlation between FT-IR analysis and XRD analysis (see Figure 4 and Figure 5 Section 3.1.3) of natural clay samples which are the raw materials of those modified clay materials. The quartz was detected in all the samples.

The SEM micrographs of Kao (a), Na-MMT (b), F13 (c) and F23 (d) are shown in Figure 3. In the SEM micrographs 3a and b, the bright spots show the rough and porous surface of the raw materials, which is one of the factors increasing the preparation yield of F13 and F23 adsorbents. The SEM images 3c and d show the irregular structure of the clay materials and disorder in the pores distributions which can facilitate the intraparticle diffusion of the acid dye. The SEM images 3c and d also show that during the heat treatment, the calcium hydroxide contained in the clay mixtures was dehydrated and converted to calcium oxide, thus leading to the formation of porous calcium-based materials which are responsible for the dye removal. Similar result was reported by [21] .

XRD analysis of kaolin shows that kaolinite is the principal constituent. Figure 4 shows that the main reflections of Ourlago-kaolin (Kao) have appeared at 2θ = 12.36˚, 19.94˚, 24.90˚, 35.98˚, 38.46˚, 45.66˚, 55.12˚ and 62.34˚ [22] . However, this material possesses some impurities. In this clay material, the characteristic peak of maghemite (Fe₃O₄ (F)) appears at 2θ = 35.71˚ (d = 2.51Å) [23] . Some crystalline phases can also be observed at the characteristic peaks [24] at 2θ = 20.89˚ (d = 4.23 Å) and 73.67˚ (1.28) for quartz (Q); 2θ = 34.65˚ (d = 2.58 Å) and 62.24˚ (1.49 Å) for montmorillonite and 2θ = 34.65˚ (d = 2.58 Å) for illite (I). Figure 4 gives XRD pattern of Kao.

Figure 5 represents the XRD pattern of sodium montmorillonite used in this study as raw material to obtain modified materials. The XRD analysis of this clay shows that montmorillonite is the principal constituent. The main reflections of sodium montmorillonite were observed at 2θ = 6, 24, 23.62, 25.91, 27.88, 34.65, 41.37, 62.07˚ respectively at the characteristic corresponding interlayered spacing d(Å): 15.037, 4.50, 3.76, 3.44, 3.19, 2.58, 2.16, 1.49. Other crystalline phases such as quartz; dolomite; illite; calcite; and muscovite in this clay are considered as impurities. They can be observed at the characteristic peaks: 2θ = 20.89, 36.55, 40.34, 45.85, 68.34, 73.67˚ at d(Å) = 4.23, 2.45, 2.23, 1.97, 1.45, 1.37, 1.28 respectively for the quartz (SiO₂); 2θ = 22.10, 30.92, 59.97˚ at d(Å) = 4.02, 2.89, 1.54 for

dolomite; 2θ = 29.99, 39.58˚ at d(Å) = 2.99, 2.27 for calcite (C) and 2θ = 34.65˚, 54.08˚ at d(Å) = 2.58, 1.69 for illite (I) [24] . Kaolinite (K) and muscovite (Mu) also appeared at 2θ = 12.21˚ (d = 7.25 Å) and 38.53˚ (2.33 Å) and 2θ = 45.61˚ (d = 1.99 Å) respectively according to our database.

Figure 6 represents the XRD patterns of combinations F13 and F23. From these, compared to those of the two raw materials, new crystalline phases emerge. The main reflections of glauconite were observed at 2θ = 50.07 and 55.29˚

respectively at the characteristic corresponding inter-reticular distances d(Å): 1.82 and 1.66; potassium, magnesium, aluminium silicate hydroxide appeared at: 8.88, 19.85 and 37.69˚ respectively at d(Å): 9.94, 4.46 and 2.38., Sanidine also appeared at 26.75˚ (3.33 Å) [24] ; Orthoclase appeared at: 13.32˚ (d = 6.63 Å) [25] ; sodium aluminium silicon oxide appeared at: 19.71˚ and 20.04˚ respectively at d(Å): 4.50 and 4.42 [26] . From the above, with regard to diffractograms of F13 and F23, an additional peak of glauconite was detected at 55.29˚ (1.66 Å) in F13 and the high intensities of the various characteristic peaks of different crystalline phases observed testify to the quality of the treatment carried out.

Figure 7 represents the TG-DSC graphs of Kao (a), Na-MMT (b), F13 (c), F23 (d). Figure 7(a) shows that between 20˚C - 200˚C, there is an endothermic peak at 59˚C in DSC graph, here, the mass loss of Kao can be assigned to the departure of water present on the clay material surface. Figure 7(b) shows that in the temperature range of 20˚C - 200˚C, there is an endothermic peak at 53˚C in DSC graph, the mass loss of Na-MMTin this step can be due to the departure of the physically adsorbed water at the clay surfaces and water molecules around sodium ion on the sites where the exchange is possible in MMT [20] . Figure 7(b) also shows that a thermal bump between 300˚C and 330˚C is concomitant with a significant weight loss.

Figure 7(c) and Figure 7(d) show that between 20˚C - 200˚C, the endothermic peaks precisely at 65˚C for F13 and F23 are associated to the liberation of the O-H of the physisorbed H₂O followed by the mass losses of 4.90% for F13 and 4.54% for F23. Figure 7(a), Figure 7(c) and Figure 7(d) show that above 450˚C, very marked endothermic phenomena centered at 483˚C, 503˚C and 502˚C respectively, are associated with significant weight loss. This leads to the dehydroxylation of kaolinite [23] according to the equation:

Si 2 Al 2 O 5 ( OH ) 4 (Kaolinite) → Si 2 Al 2 O 7 (Metakaolin)   +   2H 2 O

Table 4 indicates that the total weight loss of F13 and F23 is higher than those of the Ourlago kaolin (Kao) and the sodium montmorillonite (Na-MMT). This can be due to the transformation of calcium hydroxide into calcium oxide during the calcination step. This increase of the total weight loss confirms that the modification was accomplished.

Figure 8(a), Figure 8(c) and Figure 8(d) show that the main elements of Kao, F13 and F23 are silicium (Si), oxygen (O), iron (Fe), potassium (K) and aluminium (Al). Figure 8(b) shows that the main elements of Na-MMT are silicium (Si), oxygen (O), sodium (Na), iron (Fe), and aluminium (Al). The silicium (Si) presence in these samples also induced the presence of the quartz which is confirmed by XRD and FT-IR analyses in the previous section.

Table 5 shows that the ratio of Al/Si is around 1.5 times greater of Na-MMT than those of Kao, F13 and F23 due to the purification step of the natural montmorillonite clay. This elemental composition also shows that these materials are aluminosilicates.

In many applications, porosity of clay materials is one of the desired parameters. The BET surface area and pore volume of adsorbents were measured and results are reported in Table 6. This table compares the porosity of both raw and modified clay materials. Modified clay materials F13 and F23 give ~ 4.4 to 9.6 times

increases in surface area. A substantial decrease in BJH adsorption average pore radius can also be observed and a rise in pore volume suggests the formation of new pore through the synthesis procedure. Similar results were reported by Rehman et al. [27] .

Figure 9 shows the agitation time intervals ranging from 0 to 40 min. The uptake of AO52 dye (mg/g) onto Kao, Na-MMT, F13 and F23 increased rapidly within the first 3 min for each adsorbent and then increased slowly until reached the equilibrium and remained nearly constant. The equilibrium times are 5 min for Kao and F23; 10 min for Na-MMT and F13. The very high adsorption capacities observed during this first step can lead to the disponibility and accessibility of many active sites which also increased after the modification step of the natural clays. A rise in agitation time and the adsorption rate could induce a reduction of the boundary layer resistance, the transport of the AO52 dye from the solution to the sorbent surfaces, the migration of the AO52 dye at the interface of adsorbent - solution as well as its intraparticle diffusion in the adsorbents internal pores and into the cavities present on the surface during an adsorption process. Similar results were reported by Abbas et al. [28] .

Figure 10 shows that with the natural clays, when the pH increases from 2 to 12, the quantity of AO52 dye removed at equilibrium (q_e) varied from 0.6789 to 0.1396 mg/g for Na-MMT and 0.4995 to 0.0282 mg/g for Kao. As the pH of the solution increases, the electrostatic force repulsions are higher between the adsorbent and the acid dye, thereby decreasing the extent of adsorption. The higher adsorption of AO52 dye onto natural clays could also be attributed to hydrophilic interaction, the positive charge of the edge sites of the clay materials surface enhances the electrostatic interaction and the lone pair electrons on the nitrogen may be attracted to the positively charged samples surfaces at the low pH [29] . Conversely, the amount of AO52 adsorbed onto calcined mixed clays and lime F13 and F23 increase when the pH increases, according to this equation:

SiO 2 ( s ) + x   Ca 2 + ( aq ) + 2 x   OH − ( aq ) + ( y − x ) H 2 O → x   CaO ⋅ SiO 2 ⋅ y   H 2 O

where x is 0.8 - 1.5 and y is 0.5 - 2.5, when the hydroxide ion (OH⁻) concentration is higher, the SiO₂ contained in the structure of clay dissolves and combines with calcium ions to give porous structure and large specific surface materialsnamely calcium silicates [21] . The resulting porous materials in this

medium enhance the removal of the AO52 dye molecules. So at high pH, the solution in contact with the basal oxygen surface of the tetrahedral sheet will contain excess hydroxyls and thus the surface exhibited a CEC [30] .

The influence of the Kao, Na-MMT, F13 and F23 dosage is shown in Figure 11. This figure shows that when the mass of adsorbent varied from 0.1 - 0.8 g/20mL, the variation of 0.578 to 0.2288 mg/g for Kao; 0.8899 to 0.5157 mg/g for Na-MMT; 5.1579 to 1.4122 mg/g (64.47% to 88.26%) for F13 and 2.0589 to 0.6635 mg/g (51.47% to 82.94%) for F23 was observed after each sample equilibrium time. These results show that the AO52 dye removal capacities increase when the adsorbent dosage decreases. At this adsorbent dosage, there is a variation of the AO52 dye solution pH. Here; it can be also observed that the exchange of anion with acid dye and the adsorbent active surface functional groups is the possible reaction. Similar results were reported by Saad et al. [31] . A higher percentage of removal of pollutants may be observed by increasing the dosage of the clay materials. This is probably because of the availability of a greater number of active sites for AO52. However, no significant increase in the percentage removal of AO52 was observed as the dosage of adsorbent increased, probably because of the attainment of equilibrium. Similar results were reported by Zahir et al. [32] .

Figure 12 shows that the initial AO52 dye concentration strongly affects adsorption capacity. Increasing of AO52 concentration will also increase the AO52 dye diffusion around the clay material surface due to overcoming the various mass transfer resistances at the solution-adsorbent interface, thereby resulting in strong collisions between the dye and the catalytic area as a consequence of an extension in the strength of the propulsion of the facilitated diffusion [33] [34] .

Temperature is one of the kinetic parameters that increases through the outer layer and internal broods of the adsorbent materials, the diffusion of the adsorbate molecules. Figure 13 shows that the uptake ability of AO52 dye on Na-MMT increases when the temperature increases with values of 0.792 to 1.9468 mg·g⁻¹ at 298 to 333 K, this can be due to the great mobility of the AO52 dye molecules, thus facilitating their penetration and their accessibility at the level of the catalytic area of the porous surface of Na-MMT. This will result in suppression of the energy of activation and furthermore increase the diffusion rate of the intra-particle. Similar results were reported by [35] .

Figure 13 also indicates that the capacities of adsorption at equilibrium of Kao, F13 and F23 decrease from 0.3815 to 0.0715 mg·g⁻¹; 3.1158 to 0.5052 mg·g⁻¹ and 1.4105 to 0.941 mg·g⁻¹ respectively with the temperature increasing from 298 to 333 K and the adsorption phenomenon controlled by exothermic process. The decrease in the adsorption of AO52 dye by clay materials Kao, F13 and F23 with the increase in the temperature of the reaction medium could be due to the improvement in desorption during the adsorption process as well as to the undermining of the physical attraction contraint at the interface of catalytic area and CI Acid Orange 52. Similar result was obtained by [36] . According to the above results, physical adsorption could play an important role in this system. Similar result was reported by Chen et al. [20] .

Langmuir isothermal plot for the removal of AO52 by Kao, Na-MMT, F13 and F23 is shown in Figure 14.

Figure 15 shows the plot of the Freundlich isotherm for the removal of AO52 by Kao, Na-MMT, F13 and F23.

Table 7 represents the Langmuir and Freundlich adsorption isotherms for the removal of AO52 dye using Kao, Na-MMT, F13 and F23.This table shows that the uptake of AO52 dye by the clay mixtures F13 and F23 exceeded the AO52 dye adsorption capacity for single-layer adsorption, is overwhelmed, which could explain why even multilayer adsorption is produced or any other possible mechanism. For example, precipitation would also occur during this process. Values of n greater than one made it possible to conclude that the Freundlich model for F13 and F23 allows a better adaptation of this study [21] . Values of R² are higher for the Langmuir model, this suggests that the adsorption sites possess the same energy and there is monolayer of AO52 dye molecules adsorbed at the clays surface.

Table 8 shows that, at different concentrations of AO52 dye at 25˚C, the R_L values are amidst 0 - 1, thus confirming therefore that favorable adsorption process. These low R_L values prove that the interactions between the AO52 dye molecules and the clay materials used are strong. A similar result was reported by Kumar and Tamilarasan [36] .

Figure 16 shows pseudo first order kinetic model for AO52 dye removal onto Kao, Na-MMT, F13 and F23.

Pseudo second order kinetic models for AO52 dye adsorption onto Kao, Na-MMT, F13 and F23 are shown in Figure 17.

Table 9 represents the various constants resulting from these two kinetic models used in this study. In view of the results presented in Table 6, the pseudo second order kinetic model makes it possible to describe with more precision the mechanism of adsorption of the AO52 dye molecules. This is due to the fact that the real values (q_e,exp) are close to the theoretical (q_e,cal) values. In addition, from this model, it emerges that the values of the correlation coefficients (R²) are very high and almost equal to one for all the adsorbents.