Porosity refinement and intensification of triarylmethane dyes adsorption on bauxite residue via chemical activation: parametric optimization and theoretical insights

2.1. Materials and reagents

Anhydrous zinc chloride (ZnCl₂) (analytical grade) and sodium hydroxide pellets (NaOH) (analytical grade) were purchased from Himedia Laboratories Pvt. Ltd, India. All chemicals were used without any further purification. MG and BG were obtained from SRL Pvt. Ltd., BR was collected from the National Aluminum Company (NALCO), Damanjodi, Odisha, India.

2.2. Preparation of sorbent (BR-Zn)

The zinc-treated BR sorbent was prepared by a simple soft chemical method [32]. 10 g of BR was dispersed in a 500 mL beaker containing 100 mL of distilled water. The beaker was then placed on a bath sonicator for 15 min. In a separate beaker, 1 g of ZnCl₂ was added to 10 mL of distilled water. 30 mL of 2 M NaOH was added dropwise to the ZnCl₂ solution until a milky white-colored suspension was formed. The beaker containing the BR suspension was then placed on a magnetic stirrer at a temperature of 80°C with a stirring speed of 300 rpm. After 15 min, a ZnCl₂ solution was added to the BR suspension, and it was wrapped with aluminum foil. The suspension was stirred for 90 min and then cooled to room temperature. The suspension was filtered by Whatman filter paper. The residue was kept in a hot-air oven at 50°C for 12 hrs. Then, the sample was collected and placed in a furnace for 120 min at 500°C (Figure 1). The powder material was then stored in an air-tight, sealed container for further application.

Close

Figure 1.

Schematic of adsorbent preparation, process, and regeneration of the adsorbent.

Export to PPT

2.3. Sorbent characterization

As the sorption of any organic molecule depends on the surface chemistry of the sorbent material, several analytical techniques were employed to examine the characteristics of the sorbent. The crystal structure and phases of the sorbent were examined using an X-ray diffractometer (XRD) (Panalytical) set up with Cu Kα radiation (λ = 0.1541 nm). Data were collected in a 2θ range of 20 to 70° at a scan rate of 2°/min. Fourier transform infrared (FTIR) was employed to determine the surface functional groups before and after adsorption, using an FTIR Spectrophotometer (IR Affinity-1S, Shimadzu). Transmission electron microscopy (TEM, JEOL JEM 2100 PLUS) was used to examine the morphology and elemental composition of the sorbent. Thermal properties of the sorbent were investigated using thermogravimetric analysis (TGA 8000, Perkin Elmer). Surface area, pore volume, and pore diameter were investigated with the help of Brunauer-Emmett-Teller (BET) (Quantcrome Autosorb iQ-x). The surface charge of the sorbent was measured with the help of dynamic light scattering (Litesizer DLS, Anton Paar).

2.4. Batch adsorption study

A batch sorption study of Triarylmethane dyes (MG and BG) on the surface of zinc-treated BR was carried out using the following procedure. A 50 mL solution of MG-BG dye was prepared in a 100 mL Erlenmeyer conical flask to conduct all adsorption experiments. The desired amount of sorbent zinc-treated BR was added to the solution and placed in a shaker incubator. The pH of the solution was adjusted using 1 M HCl and 1 M NaOH with the help of a pH meter (Systronics-361). After every 15 minutes, a 2 mL sample was collected and centrifuged. The supernatant solution from all experiments was analyzed using a UV/Vis spectrophotometer (Agilent Cary 60). The UV spectra for the MG dye were collected at 617 nm, and for the BG dye, at 625 nm. The maximum adsorption efficiency of the sorbent was determined by conducting a series of experiments. Variables such as sorbent dose, initial concentration, stirring speed, pH, and temperature were adjusted to establish the optimal conditions for MG and BG dye adsorption on the sorbent. Ultimately, the adsorption efficiency (Ae %) was calculated using Eq. (1). Loading of both dyes on the sorbent surface at equilibrium (q_e) was calculated using Eq. (2) [33].

Where A_e % = adsorption efficiency, q_e = Loading at equilibrium, C₀ = Initial concentration, C_e = Concentration at equilibrium, V = Volume of the reaction, m = mass of the sorbent.

2.5. Design of experiment using response surface methodology

RSM represents a suite of mathematical tools pivotal in delineating relationships between diverse independent factors and one or more responses in experimental settings [34]. Key techniques within RSM for experimental designs encompass Box-Behnken, Central Composite Design (CCD), and Doehlert Designs. In this study, the CCD was used in all adsorption experiments conducted, and the statistical model generated by Design-Expert Software (Version 6.0, Minneapolis, Stat-Ease, USA) was employed [35]. Five crucial parameters, namely Adsorbent Dose, MG and BG concentration, String speed, pH, and temperature, acted as input variables, while the efficiency of MG and BG separation (%) using zinc-treated BR as the adsorbent served as the response variable. The minimum (-1) and maximum (+1) values for the input variables were as follows: 10 to 90 g/L of adsorbent dose, 20 to 60 mg/L of MG and BG concentration, 50 to 250 rpm of stirring speed, 3 to 11 of pH of the solution and 10 to 50°C temperature. When it comes to experimental design, the CCD approach is the most appropriate RSM method available [36]. The factorial runs, axial runs, and center runs make up the majority of its contents. To calculate the number of experiments for each process variable, one can use the following formula (Eq. 3):

Where N is the number of experiments at each run required, and n is the number of independent variables, whereas c is the central point.

The CCD procedure of the RSM technique involves a step-wise process of designing tests, calculating model coefficients, and anticipating model behavior and acceptance, followed by model development to understand the functional behavior and relationship between input parameters. To realize the process actions, a prototype of a quadratic regression equation has been designed as (Eq. 4):

There are constants bi, linear coefficients bii and bij, quadratic coefficients i and j, and a linear coefficient b_i.

Table 1 presents the coded table corresponding to these values, whereas Table 2 illustrates the statistical experimental design formulated by Design Expert Software, predicated on multiple input factors. This design underwent evaluation employing the CCD technique, deemed the most appropriate RSM technique for experimental design scenarios. The efficacy of the model was measured based on R², Adjusted R², and Predicted R² values. A higher correlation coefficient (R²) implies a more robust interpretation of the experimental dataset, validating the model’s effectiveness.

2.6. Point of zero charge

The adsorption technique was utilized to study the surface charge of zinc-treated BR [37]. Initially, 10 Erlenmeyer conical flasks (100 mL) each containing 20 mL of water were adjusted to a pH range of 2 to 11 with 1M HCl and NaOH. 20 mg of zinc-treated BR sorbent was added to each flask. All flasks were placed inside the incubator shaker after 24 h, and the pH of the solution was measured. A graph was plotted between the initial pH and the change in pH.

2.7. Adsorption isotherm

A graph that represents the variation in the interaction of adsorbate on the surface of the sorbent at different pressures at a constant temperature is known as an adsorption isotherm. Conventionally, many adsorption isotherms are used to understand the nature of the sorption [38]. The Langmuir isotherm model is one of the most used models in the adsorption process. It assumes the adsorbate and sorbent interaction occurs physically, and all the adsorbent sites are homogeneous [39]. A monolayer of adsorbate forms over the surface of the adsorbent due to electrostatic and dipole interactions. At the same time, the Freundlich isotherm assumes the formation of multiple layers on the surface and the heterogeneous nature of the adsorbent sites [40]. The linear mathematical equations for both isotherms are given in Eqs. (5) and (6), respectively [41].

Where C_e = Concentration of adsorbate at equilibrium (mg/L), q_e = Loading at equilibrium (mg/g), q_m = maximum loading capacity (mg/g), K_L = Langmuir adsorption constant.

Where K_f and n represent the adsorption capacity and adsorption intensity, respectively.

Except for the Langmuir and Freundlich isotherm models, the Temkin and Dubinin-Radushkevich (D-R) isotherm model was also utilized to evaluate adsorbate-sorbent interaction. The Temkin isotherm is generally used to determine whether the process is endothermic or exothermic [42]. According to the literature, the D-R isotherm model is used to predict the adsorption energy [43]. The mathematical equations of the Temkin and D-R isotherms are given in Eqs. (7) and (8), respectively.

Where R = Universal gas constant, T = Temperature in kelvin, b_T = Temkin isotherm constant.

Where K = D-R isotherm constant, ε = Adsorption potential.

2.8. Adsorption kinetics

Adsorption kinetics refers to the detailed reaction pathway and speed of a reaction, which helps determine the possible reaction mechanism. The adsorption process was evaluated using several models, including the pseudo-first-order, pseudo-second-order, intraparticle diffusion model, and Elovich kinetic model. The mathematical form of these kinetics is illustrated below in Eqs. (9-12), respectively [44-47].

Where qe = Loading at equilibrium (mg/g), q_t = amount adsorbed at time T, k₁ = rate constant.

Where q_e = Loading at equilibrium (mg/g), q_t = amount adsorbed at time T, k₂ = rate constant.

α =initial adsorption rate, β = desorption constant during each experiment, qt = amount adsorbed at time t.

Where $k_{diff}$ is denoted as the diffusion rate constant (mg/g.min^0.5), and C denotes the intercept of $q_t$ vs $t^{0.5}$ plot.

2.9. Thermodynamics study

Temperature is known to have a significant impact on the rate and degree of sorption; therefore, a thermodynamic investigation of the adsorption of MG and BG on zinc-treated BR was essential. The study may also offer some additional important insights, such as the spontaneity of the reaction. The overall process is either exothermic or endothermic, and its nature can be predicted through a thermodynamic study. Thermodynamic investigation can be used to assess the randomness of the process when adsorption takes place on a specific adsorbent. The Van’t Hoff equation is employed to determine each of these influencing variables. The relation between enthalpy, entropy, and free energy can be deduced by plotting a graph between lnK vs. 1/T. The mathematical form of the Van’t Hoff equation is given below (Eq. 13) [48].

Where $\frac{{\text{q}}_{\text{e}}}{{\text{C}}_{\text{e}}}$ = ln K_d = Adsorption equilibrium constant, T = Temperature, R = Universal Gas constant, ΔH = Change in Enthalpy, ΔS = Change in Entropy.

The following formula is used to calculate the change in free energy based on thermodynamic data (Eq. 14).

Where ΔG = Change in Gibb’s free energy, R = Universal gas Constant, ln K = Adsorption equilibrium constant.

2.10. Reusability of the sorbent

Investigation of the reusability of the sorbent may provide insight into the overall economics and potential for large-scale application [49]. After the sorption study, the zinc-treated BR particles were transferred into a 5 mL centrifuge tube using a 5 mL syringe for collection. After cleaning three times with distilled water, the zinc-treated BR particles were centrifuged (at 7000 rpm for 10 min) and then dried at 80°C in a hot air oven for 6 h (Figure 1). Using the dried zinc-treated BR material, subsequent sorption studies were conducted.

2.11. Residual antibacterial activity of the treated water

Aquaculture has made considerable use of MG dye as an antibacterial and fungicidal agent. MG-containing treated water was tested for toxicity using the agar well diffusion technique [50]. The effect of treated water on Escherichia coli (E. coli) was evaluated by taking contaminated water and normal distilled water as the control sample. E. coli was cultured aerobically for 20 h, at 37°C in a shaker incubator using an autoclaved nutrient mixture. To keep the concentration of bacteria below one million per mL in suspension, 0.9% saline solution was utilized for dilution. Subsequently, the inoculum was evenly spread over the nutrient agar plate. A specific amount of three types of water samples (contaminated water, treated water, distilled water) was then introduced into the well. After overnight incubation, the inhibitory zones were meticulously observed. The inhibitory zones indicate the extent of bacterial growth. To verify the accuracy and completeness of the results, all the experiments were performed three times.

3.1. Sorbent characterization

The microstructure of the sorbent may provide insight into the adsorption mechanism; therefore, the XRD technique was employed to investigate the structural features of BR and BR-Zn. The X-ray spectrum was collected in the 2θ range of 20° to 70°, and the results are shown in Figure 2(a). XRD pattern of BR indicated the presence of Fe₂O₃, SiO₂, Al₂TiO₅, and TiO₂. The sharp peaks at 2ϴ = 26.66, 29.38, and the minor peaks at 21.38, 24.27, 27.55, 40.94, and 54.08 have been corroborated by the SiO₂ (JCPDS card number 01-081-0069). Similarly, peaks at 2ϴ =33.17, 35.70, 40.80, 49.59, 62.60, 64.06, 68.69 have been complemented with the iron oxide (JCPDS card number 01-089-0597). Additionally, minor peaks of Al₂TiO₅ and TiO₂ are also identified. The quantification data for BR and zinc-treated BR were provided in Figure S1. No Zn or ZnO peak was observed in the zinc-treated BR sample after chemical activation with ZnCl₂. The average crystallite size (D) was calculated using Scherrer’s equation (Eq. 15) and given in Table 3. The table also included other information such as d-spacing values and HKL values [27].

Close

Figure 2.

(a) XRD and (b) FTIR of zinc-treated BR and BR.

Export to PPT

Where D = Average crystallite size, K= Scherrer Constant, λ = X-rays wave length, β = line broadening at FWHM, ϴ = Bragg’s angle in degrees.

The surface chemistry of the sorbent may have a consequential impact on its ability to retain specific molecular species and, therefore, to identify the functional group on the surface of BR-Zn, FTIR was taken in the wave number range of 400 to 4000 cm^-1 (Figure 2b). A very small broad peak between 3000 and 3300 cm-1 signifies the presence of the OH group in both BR and zinc-treated BR samples [51]. A more intense peak near 979 cm^-1 represents the presence of SiO₂ [52]. A reasonably weak peak can be observed at 2325 cm^-1 in BR attributed to the presence of H-O-H weakly bound water. However, after activation, that peak vanishes, which may be due to calcination at 360°C. Again, two peaks near 1634 and 1671 cm^-1 support the presence of water, which also decreases with calcination [53]. A sharp, broad peak near 979 cm^-1 represents the presence of Si-O-Si bond stretching [54]. The 870 cm^-1 peak suggests the C-O stretching [55]. The peaks near 503 and 750 cm^-1 correspond to the stretching of Fe-O. The FTIR of zinc-treated BR before and after adsorption is provided in Figure S2.

The morphology of the sorbent may provide useful information about the material surface on which adsorption is expected to happen. Morphological change induced in the BR due to ZnCl₂ activation was investigated by Transmission Electron Microscopy (TEM) micrographs using the Image J application [56]. In Figure 3, sub figure 3(a) and (e) represent the Selected Area Electron Diffraction (SAED) images taken from a particular area of the specimen for zinc-treated BR and BR, respectively. Figure 3(b) and (f) represent the Energy-Dispersive X-ray Spectroscopy (EDX) of BR-Zn and BR. The obtained d-spacing value from SAED, shown in Figure 3(c) and (g), matching with the d-spacing value of the JCPDS card 01-089-0597, confirms the presence of Fe₂O₃. Figure 3 (d) and (h) represent the TEM images of zinc-treated BR and BR, respectively. From the figure, it can be observed that the zinc-treated BR and BR consist of various cylindrical, spherical, and rod-shaped particles (Liao and Chen, 2002). The cylindrical and spherical particles may be attributed to the presence of SiO₂, whereas the rod-shaped particles are the Fe₂O₃. The particle size distribution graph for zinc-treated BR and BR was calculated and given in Figure 3(j) and (k), showing the average particle size for zinc-treated BR and BR is 14.67 nm and 11.99 nm, respectively. The calcination temperature i.e., 400°C might be one of the probable reasons for this increment of particle size. The EDX elemental mapping for zinc-treated BR and BR was given in Figure. 3 (i) and (l), showing the presence of oxygen, silicon, and iron in the sorbent. The TEM data is following X-ray diffraction (XRD) data.

Close

Figure 3.

(a) SAED pattern of BR-Zn; (b) EDX of BR-Zn; (c) d-spacing calculation of BR-Zn; (d) TEM image of BR-Zn; (e) SAED pattern of BR; (f) EDX of BR; (g) d-spacing calculation of BR; (h) TEM image of BR; (i) elemental mapping of BR-Zn; (j) particle size distribution of BR-Zn; (k) particle size distribution of BR; (l) elemental mapping of BR.

Export to PPT

The porous nature of the sorbent may influence the adsorption efficiency, and hence, the BET technique was utilized to investigate the porosity of the zinc-treated BR (Figure 4). At the initial stage, the N₂ gas adsorption over zinc-treated BR and BR shows a small convex shape near relative pressure 0.2. A steep increase in nitrogen adsorption can be seen between a relative pressure of 0.6 and 0.9, which corresponds to the presence of narrow pores [57]. It is evident from the figure that both the graphs for zinc-treated BR, Figure 4(a), and BR, Figure 4(c), under H₄ Type of Hysteresis loop. BJH was employed to determine the pore size of the material, showing that most of the pores fall within the range of 40 nm, confirming the microporosity of the material [58]. After the activation with ZnCl₂, a change can be observed in surface area from 17.91 to 19.68 m²/g. A slight change in pore volume and pore diameter is evident from the graph. The initial and final pore volume for BR was estimated to be 0.045 and 0.050 cc/g, respectively.

Close

Figure 4.

(a) N₂ adsorption-desorption graph for BR-Zn; (b) BJH pore size distribution of BR-Zn; (c) N₂ adsorption-desorption graph for BR; (d) BJH pore size distribution of BR.

Export to PPT

The surface charge of a material plays a primary role in sorption studies; therefore, the surface charge of BR and zinc-treated BR was evaluated. The Zeta potential of the sorbent was measured by choosing water as the solvent. In Figure 5, subparts Figures 5(a) and (c) represent the Zeta potential for zinc-treated BR and BR. From the zeta potential curve, it is confirmed that the average Zeta potential for BR and BR Zn was -18.9 eV and -20.9 eV, respectively. These findings indicate that both BR and zinc-treated BR are of negative charge and can adsorb triarylmethane dyes, such as MG and BG. The slight improvement in surface charge may be attributed to the presence of a negative functional group on the sorbent’s surface. As NaOH is used in the treatment process, we may conclude that the OH^- ion from the NaOH may contribute to the increase in the surface charge of BR-Zn. The point of zonal charge (PZC) of the sorbet was found to be 7.91 and 7.75 for zinc-treated BR and BR, respectively, in Figures 5(b) and (d). This shift in PZC may be attributed to the use of NaOH during the surface modification process. During the activation process, some OH^- ions may have been attached to the surface of the BR, making the surface more negative, which is reflected in the PZC data [59,60]

Close

Figure 5.

(a) Zeta potential BR-Zn; (b) PZC of BR-Zn; (c) Zeta potential BR; (d) PZC of BR.

Export to PPT

The thermal stability of the material was examined, and a graph was plotted showing the relationship between percent mass loss and temperature. A hump can be observed between 150 to 250°C corresponds to water loss (approx. 4%) of the material [61]. A step can be observed between 350-500°C, which may be attributed to the evaporation of organic matter (approximately 3%) present in the substance [62]. The final mass loss of ∼3% can be observed at 600 to 770°C; after this step, the material is thermally stable [63]. For zinc-treated BR a, a straight line is observed on the graph, indicating no mass loss step.

3.2. Effect of process parameter on the sorption of MG on BR-Zn

The impact of several process parameters on the adsorption operation was carefully investigated in this study for both the dyes MG and BG. The various process parameters are stirring speed, sorbent dosages, dye concentration, pH, and temperature. The left-hand side of Figure 6 represents the various process parameters for MG adsorption, and the right-hand side is for BG adsorption. To evaluate the impact of the stirring speed on the sorption process, 50 mL of MG dye solution was taken in a 100 mL conical flask. By maintaining other parameters at constant values and varying the stirring rate in a range of 50 to 250 rpm, the sorption experiment was conducted. The graph between adsorption efficiency and time was plotted to identify the maximum adsorption efficiency at a particular string speed. A very minute difference can be observed in Figures 6(a) and (e), which leads to the conclusion of the negligible effect of stirring speed on the sorption process. The highest adsorption efficiency was observed at a stirring speed of 150 rpm for both MG and BG dyes. By simply varying the stirring speed, a decrease in adsorption efficiency can be noticeable. With an elevation in stirring speed from 50 to 150 rpm, the turbulence may disrupt the liquid boundary layer surrounding the sorbent particle, thereby affecting the overall adsorption rate [64]. However, a further increase in stirring speed from 150 to 200 rpm results in a decline in adsorption efficiency, which may be attributed to the reduction in contact time between the dye molecule and the zinc-treated BR sorbent. This result aligns well with previously reported literature [65]. Increased agitation generally enhances the dispersion of sorbent particles and reduces mass transfer resistance, facilitating better contact between dye molecules and the adsorbent surface. However, in the present study, the adsorption efficiency plateaued beyond 150 rpm, indicating that an optimal hydrodynamic condition was already achieved at this speed. Beyond this point, excessive turbulence may have led to desorption or hindered the establishment of equilibrium by decreasing the effective interaction time. This observation aligns with the principles of external film diffusion control, where further increments in stirring do not contribute significantly to mass transfer but may instead destabilize adsorptive interactions. Moreover, the negligible difference observed between 150 and 200 rpm indicates that the adsorption system transitioned from being externally diffusion-controlled to intraparticle diffusion-dominated. This hypothesis is further supported by the relatively stable adsorption efficiency trend across the higher agitation range, indicating that pore diffusion has become the limiting step. Thus, selecting an appropriate stirring speed is crucial—not only to enhance the mass transfer kinetics but also to prevent negative impacts arising from high shear rates, especially when using chemically modified porous adsorbents like BR-Zn.

Close

Figure 6.

Effect of process parameters (a) effect of stirring on adsorption of MG; (b) effect of adsorbent dose on adsorption of MG; (c) effect of concentration on adsorption of MG; (d) effect of pH on adsorption of MG; (e) effect of stirring on adsorption of BG; (f) effect of adsorbent dose on adsorption of BG; (g) effect of concentration on adsorption of BG; (h) effect of pH on adsorption of BG.

Export to PPT

Many previously reported scientific findings indicate that sorbent dosage is one of the crucial parameters for the sorption process. To comprehensively examine the sorption dynamics across varied sorbent doses, spanning from 0.2 to 1.8 g/L, a series of batch experiments was conducted for MG and BG dyes. A constant concentration of MG (40 mg/L), maintained at pH 7, under 150 rpm at 30°C, facilitated the analysis of collected samples using a UV/Vis spectrophotometer. The resulting dataset, shown in Figures 6(b) and (f) and supported by additional information in Figure S3, demonstrates a clear trend that with increasing sorbent dose, the rate of adsorption increases, which may be due to the presence of an active site on the surface of zinc-treated BR [66]. This trend suggests that higher adsorbent dosages contribute to an increased number of accessible binding sites, resulting in more effective dye uptake per unit volume of solution. However, beyond a threshold dosage (1.6-1.8 g/L), the adsorption efficiency tends to plateau, indicating that equilibrium saturation has been approached and further addition of sorbent does not proportionally enhance removal efficiency. This could be attributed to the agglomeration of adsorbent particles at higher dosages, which potentially reduces the total surface area available for adsorption due to the overlapping of active sites. Moreover, the decreased driving force for mass transfer at lower residual dye concentrations may also limit further sorption. Such behavior aligns with Langmuir-type monolayer adsorption, corroborating the finite number of uniform active sites on the zinc-treated BR surface.

By varying the initial dye concentration between 20 and 60 mg/L under constant experimental circumstances such as sorbent dosage (1 g/L), stirring speed (150 rpm), pH=7, and temperature (30°C), the effect of dye concentration on adsorption was examined in Figures 6(c) and (g). The maximum adsorption efficiency of the zinc-treated BR was achieved at 40 mg/L. Furthermore, an increase in the initial concentration from 40 to 60 mg/L results in a decrease in adsorption efficiency from 95 to 81%, which can be attributed to the filling of active sites on the surface of BR-Zn. However, at lower concentrations, the adsorption efficiency is also low, which may be due to the fewer adsorbate molecules available to adsorb at the active site of the adsorbent[67,68]. From Figure 6(c), it can also be inferred that the contact time for the sorption process is 3 h. At about 180 min, the sorbent zinc-treated BR shows the maximum adsorption efficiency, which is approximately 95%. After 180 min, a decrease in adsorption efficiency was noticed, which was attributed to the desorption of the dye molecules. In the case of BG, the adsorption efficiency decreases with increasing dye concentration. The optimal dye concentration for the sorption process of BG was fixed at 40 Mg/L. At 180 min, the efficiency of the zinc-treated BR sorbent to adsorb BG is nearly 70%. However, a 96% removal of BG dye is achieved with an extent of contact time of 330 min. These findings are consistent with previously reported literature. These results suggest that dye adsorption is highly dependent on initial concentration, which influences the mass transfer driving force. The optimal performance at 40 mg/L reflects a balance between the available active sites and dye molecules. A further rise in concentration leads to site saturation, reducing efficiency. The slight drop in removal after 180 min indicates possible dye desorption due to weak surface interactions. BG showed lower and slower adsorption compared to MG, likely due to its larger molecular size and limited pore accessibility. The extended time for BG removal points to diffusion constraints, consistent with previous studies on dye adsorption kinetics and intra-particle transport mechanisms.

The effect of pH on zinc-treated BR adsorption was studied by varying the pH of the adsorbate solution in the ranges of pH 5 to 9 for MG dye and pH 3 to 11 for BG dye, and the results are presented in Figures 6(d) and (h). By increasing pH from 7 to 9, the adsorption efficiency increases from 93 to 96%, whereas the adsorption efficiency reduces as pH decreases from 7 to 5 for MG dye. As the pH increases from 7 to 11, the adsorption efficiency of BG increases from 93% to 97%. At pH 3, the adsorption efficiency is 34%. The PZC for the zinc-treated BR was found to be pH 7.91; by increasing the pH higher than PZC, the surface becomes more negative due to the accumulation of -OH groups, thus favoring high adsorption efficiency [69]. However, at pH less than PZC, the H⁺ ion concentration increases on the surface of zinc-treated BR and repels the positively charged MG dye molecules, which will reduce the adsorption efficiency [70]. As the difference in adsorption efficiency at pH 7 and pH 9 is only 3% for both dyes, we choose pH 7 as the optimal condition for the adsorption of MG and BG dyes on BR-Zn. Furthermore, from an upscaling perspective, maintaining the system pH at pH 9 may lead to corrosion in the container; therefore, pH 7 was chosen as the optimal pH for the sorption process. Additionally, the variation in removal efficiency with pH suggests the key role of surface charge modulation and dye speciation. At lower pH values, competitive adsorption between H⁺ ions and dye cations for the same active sites likely limits uptake. On the other hand, at higher pH, the surface deprotonation not only enhances negative charge density but also facilitates multilayer dye adsorption due to weakened repulsion among adjacent dye molecules. The sharper increase in BG adsorption efficiency at alkaline pH compared to MG may also be attributed to its larger π-conjugated system, enabling stronger π–π interactions with the activated BR surface. These results collectively confirm that electrostatic forces, surface chemistry, and molecular structure intricately govern the pH-dependent adsorption behavior.

Temperature is known to have a significant impact on most physicochemical processes. Therefore, the effect of temperature on the adsorption efficiency of MG and BG was examined, and the results are provided in Figure 7, subparts Figures 7(a) and (c), respectively. All three experiments were carried out by varying temperature in a range of 30 to 50°C and keeping other parameters constant, such as adsorbent dosages (1 g/L), adsorbate concentration (50 mg/L), stirring speed (150 rpm), and pH = 7. It can be observed that the adsorption efficiency increases by increasing temperature from 30 to 50°C, which suggests the sorption process is endothermic for MG and BG dyes. [71]. The probable reason for this effect may be attributed to the increase in kinetic energy at high temperatures, which can lead to preferential diffusion into the adsorbent pores. Moreover, the rise in temperature causes the internal pores of the sorbent matrix to enlarge, allowing the dye molecules to enter the internal pores of the sorbent [72]. Moreover, the thermodynamic parameters such as ΔH°, ΔG°, and ΔS°, calculated from Van’t Hoff plots, supported the endothermic and spontaneous nature of the adsorption. The positive ΔH° values further confirmed the endothermic behavior, while the negative ΔG° values across the studied temperature range implied the feasibility and spontaneity of the dye uptake process. The positive entropy change (ΔS°) suggested increased randomness at the solid–solution interface during adsorption, possibly due to the displacement of water molecules by dye molecules. These findings underline the thermally activated mechanism of adsorption and the potential for better performance at elevated temperatures.

Close

Figure 7.

(a) Effect of temperature on adsorption for MG; (b) effect of temperature on adsorption for BG; (c) thermodynamic study of MG; (d) thermodynamic study of BG.

Export to PPT

3.3. Statistical analysis through the RSM model

In this investigation, the five most important operational parameters on MG and BG uptake, such as adsorbent dose, MG concentration, string speed, pH, and temperature, were taken as input variables, and the MG and BG removal (%) was considered as the outcome of this experiment design. The statistical coefficients for the MG and BG Removal model were generated using Design Expert Software (Version 8.0.6), as shown in Eq. (16) and Eq. (17) respectively.

RSM utilizes the supplied data to evaluate the impact of various parameters on adsorption efficiency using a regression model. To illustrate the connection between the response variable (here, adsorption efficiency) and the predictor variables (adsorbent dose, concentration, string speed, pH, and temperature), RSM usually employs a second-order polynomial equation. Each component and its interactions are represented by their coefficients in the data, which together determine the total response. The initial level of adsorption efficiency when all parameters are zero is represented by the constant term, which is 98.564 and 97.0095 for MG and BG, respectively.

To evaluate the model of the regression equation, the Analysis of Variance (ANOVA) method was used to determine the impact of key parameters. The equilibrium and initial MG concentrations were calculated to generate the actual data, and mathematical model equations were used to predict the values. Based on the actual and predicted data, the value of the correlation coefficient (R²) was calculated to validate the model’s success. The study’s findings are presented in four figures: Figures 8(a-d). In Figure 8(a), the impact of MG concentration and adsorbent dose on MG removal efficiency is illustrated. The study showed that increasing the adsorbent dose (30-70) g/L and the MG concentration (30-50 mg/L) led to an improvement in MG removal rate initially from (90% - 98%) followed by decreasing from (90% to 80%) while keeping the stirring speed, pH and temperature constant. Figure 8(b) presents the effect of pH and adsorbent dose on MG removal efficiency. The study showed that even with a low dose of adsorbent (30 g/L) and at 9 pH, the removal efficiency of MG was found to be maximum, which is almost 100%, while the other parameters, such as, stirring speed (150 rpm), MG concentration (40 mg/L) and temperature (30^°C) keeping constant. Similar type results have been found in Figures 8(c-d).

Close

Figure 8.

Mutual interaction and desirability analysis obtained from RSM-CCD of BG dye adsorption efficiency over BR-Zn. The mutual Interaction of (a) Concentration and Adsorbent dosage; (b) pH and adsorbent dosage; (c) Temperature and adsorbent dosage; (d) Temperature and pH.

Export to PPT

Similarly, Figures 9(a-d) illustrate the interactions of different parameters, providing insight into their combined effects on adsorption efficiency. Figure 9(a) depicts the interaction between initial dye concentration and adsorbent dosage. Increasing the adsorbent dosage generally enhances dye removal efficiency, as it provides more active sites for adsorption. However, at higher dye concentrations, the removal efficiency may plateau or decrease if the adsorbent’s capacity becomes saturated. Figure 9(b) illustrates the combined effect of pH and adsorbent dosage on BG dye adsorption. The solution’s pH significantly influences the surface charge of the adsorbent and the ionization state of the dye, affecting adsorption efficiency. At lower pH levels, increasing the adsorbent dosage enhances removal efficiency; however, this effect may diminish at higher pH values due to changes in electrostatic interactions. Figure 9(c) examines how temperature and adsorbent dosage jointly affect BG dye adsorption. Elevated temperatures can increase adsorption efficiency by enhancing the mobility of dye molecules and the active sites’ accessibility. However, the extent of this effect depends on the adsorbent dosage; higher dosages may lead to a more pronounced improvement in efficiency with temperature increases. Figure 9(d) explores the interaction between temperature and pH. The adsorption efficiency varies with changes in temperature and pH, indicating that optimal conditions exist for maximum dye removal.

Close

Figure 9.

Mutual interaction and desirability analysis obtained from RSM-CCD of BG dye adsorption efficiency over BR-Zn. Mutual Interaction of (a) Concentration & Adsorbent Dosage; (b) pH & Adsorbent Dosage; (c) Temperature & Adsorbent Dosage; (d) Temperature & pH.

Export to PPT

To validate the model, the researchers conducted an ANOVA with various process parameters using activated clay as the adsorbent. The Sum of Squares (1907 for MG removal and 3063 for BG removal) is a statistic used in RSM to quantify the total variability in the data and the extent to which the observed data points differ from the mean value. The extent to which the model accounts for the observed variance in the response variable is demonstrated by an increasing sum of squares, which also signifies greater variability in the data. This example uses 20 data points to determine the variance of the response variable, as indicated by the Degrees of Freedom (df = 20), which denotes the number of independent values that can vary in the study. To provide an average measure of variation per degree of freedom, the Mean Square (95.63 and 153.16, respectively, for MG and BG) is calculated by dividing the Sum of Squares by the Degrees of Freedom. This number represents the dispersion of the data. Tests for general model significance are indicated by the F-value (53.99 and 64.28, respectively, for MG and BG), which is a ratio of the model’s mean square to the error’s mean square. A larger F-value suggests that the model can explain a considerable amount of the response variable’s variability. The possibility that the reported results are attributable to chance is indicated by the p-value, which is less than 0.0001. It is quite improbable that the findings in the model happened inadvertently, given the statistical significance of the data and the low p-value. In terms of statistical significance, these metrics point to a very reliable and highly significant model that the RSM analysis employed to explain the response variable’s variability. In addition to that, the ANOVA findings revealed an average removal effectiveness of 98% for MG and 97% for BG, a standard deviation of 1.45 (average), and a coefficient of variation (C.V. %) of 1.55 (average). The adjusted and predicted R2 values for MG removal were 0.955 and 0.906, respectively, while the adjusted and predicted R2 values for MG and BG removal were 0.962 and 0.916. The model’s correlation coefficient value (R2) was found to be 0.989 and 0.977 for MG and BG, respectively.

The close agreement between the adjusted R² and the predicted R², with a difference of less than 0.2, suggests the model’s reliability. Table 2 shows the actual and expected removal efficiency values estimated using the Response Surface Methodology (RSM) model. By applying response surface methodology (RSM), it was found that under certain conditions, the MG and BG dyes could be effectively removed, with removal efficiencies of 98% and 96.5%, respectively. The ideal settings involved a starting MG concentration of 40 ppm and a contact duration of 180 min. An adsorbent dose of 40 g/L, a pH of 7, a temperature of 30°C, and a stirring speed of 150 rpm were used in the procedure. The treatment procedure was found to be effective in the given experimental parameters, as these conditions maximized the removal efficiency of MG and BG (Figure S4).

3.4. Adsorption isotherm

Adsorption isotherms can provide useful information about the physicochemical interaction between the adsorbent and adsorbate. Therefore, a range of experiments were executed by varying the concentration of MG and BG dyes from 10 to 70 mg/L. The required values for linear model isotherm fitting were calculated, and model fitting (Langmuir, Freundlich, Temkin, D-R) was carried out using the experimental data. The respective data are shown in Figure 10, and the value obtained from the fitting was summarised in Table 4. The R² values for MG and BG dyes are 0.99 and 0.98, respectively, which suggests that the experimental data are in good agreement with the Langmuir isotherm model. From the isotherm point of view, it may be predicted that the adsorbent surface is homogeneous, and a monolayer sorption may have occurred on its surface [73]. The obtained K_L values are 0.27 and 5.42 for MG and BG dyes, respectively, indicating that the interaction between BG and zinc-treated BR sorbent is stronger compared to MG and BR-Zn. The maximum loading at equilibrium (qm) for the process was evaluated at 64.93 mg/g and 41.84 mg/g for MG and BG, respectively. The above results are attributed to the fact that although the adsorbent exhibits a stronger affinity for BG dye, it has a higher adsorption capacity for MG dye. This could be due to differences in the molecular size, shape, or interaction mechanism of the two dyes, which affect the packing density and coverage on the adsorbent surface. The R_L value, as obtained from experimental data, is 0.067 for MG and 0.0036 for BG, indicating a favorable sorption process, as the value of RL lies between 0 and [74].

Close

Figure 10.

Adsorption isotherm models fitting for MG adsorption (a) Langmuir; (b) Freundlich; (c) Temkin; (d) D-R, for BG adsorption (e) Langmuir; (f) Freundlich; (g) Temkin; (h) D-R.

Export to PPT

3.5. Adsorption kinetics

Sorption is a physicochemical process where mass transfer of solute takes place from a solution to the surface of the sorbent. Although it is an arduous task to elucidate the exact sorption mechanism, a kinetic study may elucidate the probable reaction pathway for the process [75]. Pseudo first order, Pseudo second order, Intra particle diffusion model, and Elovich kinetic model for the sorption of MG and BG over zinc-treated BR were provided in Figures 11(a-h). The highest R² value (≈ 0.99) was obtained from the Eovich kinetic model for MG. Parameter such as α (initial adsorption rate) and β (desorption rate during each experiment) of the Elovich kinetic model were calculated from the graph. However, BG follows pseudo-second order reaction pathways Figure 11(f). All the calculated data from kinetic models were summarised in Table 5 for both MG and BG dyes.

Close

Figure 11.

Linear kinetics model fitting MG adsorption (a) Pseudo 1^st order; (b) Pseudo 2^nd order; (c) intraparticle diffusion model; (d) Elovich kinetic model. BG adsorption (e) Pseudo 1^st order; (f) Pseudo 2^nd order; (g) intraparticle diffusion model; (h) Elovich kinetic model.

Export to PPT

3.6. Thermodynamic study

A thermodynamic study of the sorption process may offer insight into the mechanism of sorption. ΔH, ΔS, and ΔG were calculated using Vant Hoff’s equation. From Figure 7(a), it is evident that with an increase in temperature, the adsorption efficiency value increases for both the dyes, indicating the sorption process is endothermic in nature. If the change in enthalpy value lies between 4 and 40 kJ/mol, then the process can be termed physisorption [76]. The ΔH value for the sorption process for MG dye and BG dye was found to be 14.04 KJ/mol and 47.42 KJ/mol, respectively, which suggests the sorption process is physisorption in nature. The calculated ΔG values were -10.53, -10.53, and -10.304 KJ/mol for MG and -6.72, -10.22, and -14.02 KJ/mol for BG dye, which demonstrates that the process is spontaneous in nature. Finally, the ΔS value of both dyes was calculated; the value comes from thermodynamic modeling for MG dye was 78.317 KJ/mol, and for BG dye was 178.48 kJ/mol, which suggests a decrease in randomness after adsorption of MG on the zinc-treated BR surface [77]. Possibly, the positive value suggests the orderly arrangement of the solute dye molecules on the surface of the sorbent. The nearly similar structure of both dyes may be one of the probable reasons that both dyes exhibit a spontaneous and endothermic nature after adsorbing on the surface of the RM-Zn.

3.7. Proposed mechanism

A control UV/Vis spectrum of MG and BG uptake over BR (before activation), and zinc-treated BR (after activation) was represented by Figure S3. From the spectrum, it can be observed that the adsorption of MG dye over BR gets saturated beyond 180 min The A_e value for BR is 50% whereas after activation with zinc, the treated BR at 180 min has increased to 95%. However, for BG, the maximum adsorption efficiency for BR is 46%, and for zinc-treated BR, it is 93% at 330 min. From the zeta potential and PZC data, it can be concluded that after activation, a significant amount of negative charge deposition has occurred on the surface of BR. A high adsorption efficiency value was obtained although the solution pH was maintained at 7, which is lower than the PZC of 7.91. This suggests not only charges but also some ions could be interfering with the sorption process, probably the OH^- ion. The experimental data best fit with the Langmuir isotherm, which points to the physisorption of the MG and BG over BR-Zn. The thermodynamic data ΔH=14.04 KJ/mol and 47.42 KJ/mol for MG and BG dye also coincide with isotherm fitting, suggesting physisorption of MG and BG over BR-Zn.

3.8. Reusability

When assessing the commercial potential and economic viability of sorbent systems, the reusability of the adsorbent is a crucial factor to consider. As seen in Figure S5, the sorption efficiency of zinc-treated BR remains almost the same up to the 4^th cycle. Beyond that, the process requires an additional 15 to 20 minutes for the MG dye and the same for the BG dye. To unveil the commercial potential of the sorbent zinc-treated, BR, further processing and scale-up can be performed.

3.9. Ecotoxic analysis

Ensuring the safe reuse or discharge of treated water into natural water bodies is one of the primary reasons for removing MG and BG dyes from water. The study evaluated the treated water’s continued ability to inhibit E. coli growth. The well containing MG-contaminated water showed a distinct 1.07 cm antibacterial zone Figure12(a). Whereas in the case of BG, it gave a 1.6 cm zone Figure 12(b), indicating the antibacterial agent’s efficacy against bacteria. On the other hand, complete bacterial growth was seen close to the treated and distilled water well, suggesting that the treated water had no negative effects [78]. This implies that MG and BG dyes may be successfully removed from water by zinc-treated BR without causing further contamination. Moreover, this suggests that zinc-treated BR adsorption performs efficiently without leaving any negative residues, which makes it an effective water-purifying technique. Further optimization, continuous monitoring, and pilot-scale studies could enhance the efficiency of the technique.

Close

Figure 12.

(a) Toxicity test of MG dye after adsorption; (b) Toxicity test of BG dye after adsorption. (DW: Distilled water; CW: Contaminated water (containing MG dye); TW: Treated water).

Export to PPT

3.10. Comparison with other adsorbents

The comparative analysis of various adsorbents for the removal of triarylmethane dyes, such as MG and BG, highlights the superior performance of the zinc-treated BR developed in this study, exhibiting maximum adsorption capacities of 38.04 mg/g for MG and 37.05 mg/g for BG under optimized operational conditions. In contrast, traditional adsorbents, such as bagasse fly ash (9.98 mg/g) and fly ash from a thermal power plant (1 mg/g), demonstrated significantly lower adsorption efficiencies. Among the mineral and nanocomposite-based materials, zeolite-magnetite nanocomposites and pumice stone showed moderate adsorption capacities of 21.05 mg/g and 22.57 mg/g, respectively. Functionalized MWCNTs and industrial solid wastes offered relatively lower adsorption capacities, while synthetic polymers like poly(divinylbenzene) achieved 14.61 mg/g. Interestingly, agricultural waste-based adsorbents such as pistachio shells and tea powder exhibited remarkably high capacities of 46.42 mg/g and 101.01 mg/g, respectively, indicating the potential of low-cost biomass for dye remediation. However, when considering a balance between adsorption capacity, availability, cost, and environmental impact, zinc-treated BR stands out as a promising and sustainable alternative for effective dye removal from aqueous media. Table 6 compares the adsorption capacity of different zinc-treated BR adsorbents from the literature with current study.