The XRD data of bentonite and organoclays prepared with different surfactant loadings were shown in Fig. 1. The patterns of the modified bentonites indicate the modification of clay by surfactants.

As shown in Fig. 1, d₀₀₁ increased significantly after treatment with surfactants. Pristine bentonite had an interlayer space of 1.23 nm, which increased to 1.88 nm and 1.78 nm for H-Bt and B-Bt, respectively. As the surfactant loading increased, the basal spacing of the organoclays gradually increased, which can be attributed to changes in the structural configuration of the molecules within bentonite [27–28]. Based on the observed interlayer spacing of bentonite and organoclays, the basal spacing of 1.45 nm with a shoulder of 1.72 nm indicated the presence of more than one molecular arrangement (a monolayer and a lateral bilayer at the edge) in the bentonite interlayer at 0.5 CEC for H-Bt. The basal spacing of 1.84 nm at 1.0 CEC implied a lateral bilayer arrangement, and the orientation of the molecular layer remained the same when the surfactant loading was increased up to 1.5 CEC.

Compared to exchange of the HDTMAB cationic surfactant, the use of BDHAC cationic surfactant having a benzene ring resulted in a significantly increased basal spacing of 1.78 nm at 0.5 CEC. This represents the formation of a lateral bilayer in the interlayer of the clay. Furthermore, one more expansion was observed at 1.81 nm for 1.0 CEC-BDHAC, reflecting a lateral bilayer arrangement. However, no further expansion occurred, and the spacing decreased marginally to 1.78 nm at 1.5 CEC. This phenomenon can be explained based on the fact that at 1.5 CEC, organoclays with excess surfactant can be adsorbed on the surface of bentonite by the arrangement of surfactant, which was simulated using density functional theory (DFT) calculation in the previous study. [29, 30]. The molecular dynamic (MD) simulation demonstrated that monolayer surface arrangements were evidently observed at lower surfactant loadings and more complicated surface arrangements such as bilayer or paraffin-type initiated by increased surfactant molecules (> 100% of the CEC), affecting the thickness of the organic cation surface layers [30]. Thus, the clay interlayer was not significantly improved. According to this result, the structural properties of bentonite clay layers differ with different surfactant loadings (low surfactant loading at 0.5 CEC vs. high surfactant loading up to 1.5 CEC) and surfactant types (HDTMAB vs. BDHAC). In particular, organoclays prepared with higher surfactant loadings (i.e., 1.5 CEC) have a larger basal spacing, and they may have a greater number of sorption sites within the interlayer space, as well as on the clay surface; these aspects will be further examined by performing BET analysis.

Other structural parameters such as specific surface area, pore volume, and average adsorption-desorption pore diameter values of the organoclays are summarized in Table 1.

As summarized in Table 1, the specific surface area, total pore volume, and average pore diameter of the organoclays are lower than those of bentonite. The surface areas of bentonite, H-Bt at 1.5 CEC, and B-Bt at 1.5 CEC were 42.96, 1.79, and 9.22 m²/g, respectively. The total pore volumes of bentonite, H-Bt, and B-Bt were 0.95, 0.004, and 0.015 cm³/g, respectively, and the average pore diameters were 8.84, 8.06, and 6.59 nm, respectively. Bentonite has the capability to expand the interlamellar space when ion exchange occurs with surfactant molecules, and it is widely known that the larger organic cations in the surfactants increase the spacing between the tetrahedral sheets [31], as confirmed by the XRD result obtained herein. Based on the increased d₀₀₁ spacing of the modified clays with two cationic surfactants, the relatively large surfactant cations of HDTMAB and BDHAC were intercalated into the aluminosilicate sheets, resulting in expansion of the interlamellar space of bentonite [14]. This behavior led to increase the packing density of organoclays by increasing the surfactant loading in the clays and altering the arrangement of surfactant molecules from parallel to surfactant chains oriented at certain angles to the silicate surface [30]. This reduced the BET surface area and pore volume of the organoclays. For instance, the relatively large HDTMAB and BDHAC cations may have been compactly loaded and packed in the interlamellar spaces in the clay, and severe pore blocking may have occurred, which may have led to a decrease in the pore volume, average pore diameter, and BET surface area.

Interestingly, the pore packing of the organoclays prepared with HDTMAB was larger than those prepared with BDHAC, possibly because of a difference in the hydrophilicity levels of the two surfactants. The hydrophilic bentonite and the hydrophobic functional groups of BDHAC possessing a benzene ring caused the two surfactants to repel each other. This repulsion may have prevented deep insertion of BDHAC into the bentonite pores during synthesis. Particularly, the presence of the larger benzene ring (8.29Å) on BDHAC compared to the alkyl chain length (3.65 Å) led to the loading of a more significant amount of surfactant on the clay surface than in the interlayer spacing of the clay. For this reason, the average pore diameter of B-Bt decreased more significantly than that of H-Bt. In other words, a greater amount of HDTMAB can be loaded in the interlayer space than BDHAC, even at the same CEC level, and the pore volumes of the organoclays prepared with HDTMAB were smaller, although the pore diameter did not decrease significantly. This trend of average pore size is consistent with the tendency observed in the XRD patterns. Additionally, an increase in the cation coverage on the clay surface tended to decrease the hydrophobicity of organoclays [30], which may alter the rheological and swelling properties of the modified clays, particularly for those with the higher organic cation loading (> 1.0 CEC). Eventually, it leads to the weaker sorption capacity of organoclays to uptake organic pollutants. It appears that the loaded surfactant and the type of surfactant molecule influence the structure distribution of organoclays [29], as will be further verified with FT-IR.

FT-IR can be used to determine the molecular environment of intercalated surfactant molecules within organoclays [11, 19, 32] and several distinguishing regions: (i) OH stretching region (3700–3000 cm⁻¹), (ii) CH stretching region (2900–2800 cm⁻¹), (iii) HCH bending vibration (1520–1400 cm⁻¹), and (iv) Si-O region [5].

Here, the functional groups of the adsorbents prepared using HDTMAB and BDHAC were examined with FT-IR, as presented in Fig. 2. A sharp, intense peak at 3627 cm⁻¹ for bentonite was assigned to OH stretching vibrations of the structural hydroxyl group, and the hydroxyl band of the organoclays was relatively steady. The CH stretching region of the organoclays (i.e., H-Bt and B-Bt) prepared with 1.5 CEC loading is shown in Fig. 2. Compared to bentonite, the bands at 2917, 2919, and 2850 cm⁻¹ were attributed to the methylene group in the surfactant molecules, and the organoclays modified with HDTMAB and BDHAC had antisymmetric V_as(CH₂) and symmetric V_s(CH₂) stretching modes [33]. The bands observed at 1468 and 1471 cm⁻¹ for the organoclays were ascribed to the scissoring modes of the methylene group (−CH₂) and HCH deformation of the surfactant molecules intercalated into bentonite, respectively [19]. Broad bands at 1113 and 998 cm⁻¹ were distinctively observed for bentonite owing to the Si-O groups in the tetrahedral sheets, and the peak at 916 cm⁻¹ represented the Al-Al-OH groups in bentonite [32]. These peaks were observed in cases of the organoclays as well, which confirmed the presence of bentonite clay even after loading the surfactant molecules. The observed FT-IR results clearly indicated the presence of surfactant molecules loaded on the bentonite, and the surfactant modification of bentonite was successfully achieved.

The zeta potential was measured for surface charge analysis of the clays before and after modification. As shown in Fig. 3, the measured zeta potential of unmodified bentonite was −32.7 mV due to its negatively charged nature [34]. By contrast, the measured zeta potentials of the organoclays modified with cationic surfactants, namely H-Bt and B-Bt, were +47.8 mV and +54.2 mV, respectively. This shift in surface charge indicates not only the presence of intercalated surfactant molecules within the interlayer spacing of bentonite but also the great potential of the electrostatic adsorption mechanism for anionic materials.

Based on the characteristics of bentonite and the organoclays, we conducted adsorption experiments using the organoclays to remove anionic OG dye from aqueous solutions. The adsorption capacities of the organoclays prepared with different surfactant loadings of 0.5–2.0 CEC were tested for the removal of OG after 24 h, and the maximum percentages of OG adsorbed by the organoclays are presented in Fig. 4.

Fig. 4 shows the maximum OG removals achieved using H-Bt and B-Bt at 1.5 times the theoretical CEC (110 meq/100 g). Because similar or lower OG removal rates were achieved with ratios higher than 1.5, the optimal removal efficiency of OG was considered to be achieved at 1.5 CEC. At 1.5 CEC, H-BT removed 99% of OG while approximately 67% OG removal was achieved by B-Bt and this level remained unchanged, even when the surfactant loading was increased. These phenomena agree well with the XRD data presented in Fig. 2. Although the d-spacing of H-Bt increased up to 1.5 CEC, B-Bt exhibited the maximum value at 1.0 CEC. This can be ascribed to the hydrophilic affinity resulting from the structural differences between the two surfactants, as discussed earlier in the BET analysis section. Thus, HDTMAB can be loaded continuously in the interlayer space, whereas BDHAC, which has less affinity to hydrophilic bentonite, is partially modified between the interlayer as well as on the surface of the clay, which was not further improved by our current synthesis method. However, because both organoclays yielded the maximum removal rate at 1.5 CEC surfactant loading, 1.5 CEC can be considered the optimal surfactant loading ratio for the organoclays. Therefore, the organoclays prepared using both surfactants at 1.5 CEC were used for all subsequent adsorption experiments to test the uptake of OG dye from aqueous solutions.

The adsorption isotherms of OG for the two organoclays were obtained in batch experiments conducted using different initial concentration ranges, and they are shown in Fig. 5. To determine the adsorption capacities and adsorption surface properties of the adsorbents for OG, the equilibrium experimental data were fitted to the Langmuir (Eq. 5) and Freundlich (Eq. 6) isotherm equations. The Langmuir equation applies to a homogeneous adsorption system, and it refers to a chemical monolayer adsorption site, while the Freundlich equation is suitable for a heterogeneous system with physical multilayer adsorption [35–36]. It is also well known that the Langmuir equation is suitable for electrostatic adsorption [37]. The values of K_L, q_m, K_F, n, and the correlation coefficient (R²) for the Langmuir and Freundlich isotherms are given in Table S1.

As listed in Table S1, both H-Bt and B-Bt showed higher R² values in the Langmuir model (0.976 and 0.972, respectively) than the Freundlich model (0.896 and 0.900). Hence, OG adsorption by H-Bt and B-Bt fitted better into the Langmuir model than the Freundlich model, indicating that the adsorbents are suitable for chemical monolayer adsorption. In relation to the positive surface charge of the organoclays examined in section 3.1.2, the primary mechanism of removing OG can be associated with electrostatic and chemical sorption. The maximum sorption capacity (q_m) of H-Bt (114.3 mg/g) was higher than that of B-Bt (65.56 mg/g), which is consistent with the result presented in 3.2.1 (Fig. 4). Moreover, it provides substantial information on the mechanism that a greater amount of surfactants exists between the H-Bt interlayers, which is in good agreement with the results of XRD and BET analyses. Consequently, the q_m of H-Bt was higher than that of B-Bt due to the larger amount of modified surfactant, and for this reason, it yielded a higher OG removal rate. According to Seniha Elmen et al. (2012) [38], the q_m of HDTMA-Bentonite for OG adsorption was 104.48 mg/g, and the q_m used in this study yielded similar results.

Adsorption kinetics were determined to compare the adsorption rates of H-Bt and B-Bt. Kinetics studies are performed to obtain important indicators for understanding the adsorption behavior of OG. Both pseudo first-order and pseudo second-order kinetic models were used to fit the experimental data by using an initial concentration of 200 mg/L. The results are shown in Fig. 6 and summarized in Table S2. According to the results obtained using the two different kinetic models, both organoclays followed a pseudo second-order model with high correlation coefficient (R²) values of 0.993 for H-Bt and 0.982 for B-Bt, whereas the pseudo first-order model yielded R² values of 0.956 and 0.938, respectively. The high correlation coefficients obtained for the pseudo-second order model suggest that the adsorption of anionic OG dye onto H-Bt and B-Bt followed the pseudo second-order model more closely than the pseudo first-order model in this study. It is known that physical adsorption fits the pseudo first-order kinetic model, and chemical adsorption fits the pseudo second-order kinetic model [39, 40]. These kinetic fitting results indicated that OG adsorption on organoclays was mediated by chemical interaction, which correlates to the isotherm results.

The kinetic constants, as obtained using the pseudo second-order model, were 0.002509 g/mg·min and 0.00083 g/mg·min for H-Bt and B-Bt, respectively. The kinetic constant of H-Bt was approximately three times higher than that of B-Bt, indicating that H-Bt tended to adsorb more anionic dye than B-Bt.

An adsorption test was conducted using a mixed sample of cationic and anionic dyes to investigate the ion selectivity of the adsorbents (Fig. 7). Methylene blue (MB), a cationic dye with a strong blue color, has a strong absorbance band at a wavelength of 662 nm, and the relatively pale orange anionic dye Orange G (OG) has a maximum absorbance peak at 478 nm. A comparison of the absorbance of OG and MB dyes before and after the adsorption test was conducted using the prepared organoclays, and changes in the absorbance of OG and MB were observed at 478 and 662 nm, respectively.

Both H-Bt and B-Bt organoclays removed considerable amounts of anionic dye molecules at 478 nm, suggesting that the organoclays prepared using two different types of surfactants (HDTMAB and BDHAC) can adsorb the anionic OG dye effectively (almost up to 100%). However, in terms of the uptake of cationic MB dye molecules, the removal efficiency of B-Bt was not significant, while H-Bt was able to adsorb more than 50% of the MB molecules in water. Based on these results, it can be concluded that the adsorption capacities of the organoclays intercalated with HDTMAB and BDHAC are close to 100% for anionic OG, and both organoclays are more efficient at adsorbing anionic contaminants from aqueous solutions than cationic pollutants. In other words, both H-Bt and B-Bt removed anionic pollutants more selectively than cations due to the high surface charges of the adsorbents (+47.8 mV for H-Bt, +54.21 mV for B-Bt), which have a tendency to adsorb anions electrostatically. Apparently, the pHs of dyes prepared separately and in the mixed form were measured at pH 6.7(OG)/7.2 (MB) and 7.0, respectively, and the effect of pH on the adsorption of cationic and anionic dye molecules by the pristine bentonite was expected to be minimal in the range of pH 3 – 12, in agreement with previous observations on other clays [41, 42].

Although both adsorbents commonly adsorbed almost all of the OG, as shown in Fig. 7, the adsorption behaviors of the two organoclays were different in the removal of cationic MB dye molecules. Additionally, the optimized organoclays can be an effective sorbent for the removal of anionic heavy metals (e.g., Cr(VI), Mo(VI), As(III), As(V)) due to the presence of positively charged exchange sites [43–45] via ion exchange, electrostatic interaction, and surface complexation. To a certain extent, the enhanced adsorption of Cu(II) and Zn(II) was reported as well as co-adsorption of organic pollutants and heavy metal ions were achieved by surfactant-modified clays [46, 47]. Therefore, from a practical perspective, it can be suggested that the organoclays synthesized with HDTMAB provide a higher potential for the uptake of a greater amount of environmental pollutants from real wastewater due to the obtained structural properties of the H-Bt organoclays, particularly the larger basal spacing of which increased during the synthesis. In particular, the utilization of modified clays can be appropriately considered for the adsorption of textile effluents, including dyes and heavy metals (e.g., Cu, Cd, Zn, Pb, and Ni originating mainly from color-pigments) during the fabric manufacturing processes [31, 48, 49].