Due to the fact that layered double hydroxides consist in positive layers of mixed metal hydroxides, the pH of working solutions plays an important role in the adsorption behavior of these materials. Depending on the value of the initial pH of the working solution, the surface of the materials can be either negatively charged, neutral at a particular pH, or positively charged. The value of pH at which the surface of the materials is neutral (the so-called pH of point of zero charged pH_pzc) was determined by contacting 0.0050 g of solid materials with 50 cm³ of solutions of HNO₃ or NaOH with defined pH_i values. After 24 h of mixing, the values of final pH (pH_f) of the solutions were determined. By plotting pH_f = f(pH_i), the pH_pzc can be determined as the intersection point of the obtained curve with the theoretical curve pH_i = pH_f. The pH of point of zero charged for the studied materials is presented in Figure 3.

From Figure 3 it can be observed that the behavior of the materials (LDH – R, prepared from analytical grade reagents and LDH – W, prepared from waste) is similar and the pH = 9.5 represents the point of zero charge of the surfaces for both materials. At pH values lower than pH_pzc = 9.5 the surface of LDH became positive due to the protonation effect, while at pH values greater than pH_pzc = 9.5 the surface of LDH became negative due to the deprotonation effect (Xu et al., 2008). From a practical point of view, if the initial pH of the working solution has values lower than pH = 4, or greater than pH = 12, the mixed hydroxides can be dissolved.

On the other hand, the Orange II dye exists in aqueous solutions in various forms (Bourikas et al., 2005), but in the pH range of 2 – 9 only the partial ionized species (HL⁻) is present; the fully ionized species being predominant at pH > 12 (Zaheer et al., 2019).

In order to determine the maximum adsorption capacities of the studied materials, equilibrium studies were performed with solutions containing initial concentration of Orange II between 10 and 100 mg/L, a S:L ratio of 1 g/L, at three different temperatures (293, 308, and 323 K). Due to the fact that, at 323 K, the equilibrium was not reached with the same starting dye solutions as in previous experiments, we performed the equilibrium studies with an initial concentration of Orange II between 40 and 200 mg/L.

The experimental data were fitted with the non-linear form of the Langmuir and Freundlich isotherms in order to describe the equilibrium between the adsorbent materials and Orange II. The non-linear mathematical expression of Langmuir model is represented in Equation (5) (Tran et al., 2016):

where q_max (mg/g) is the maximum saturated monolayer adsorption capacity under the given conditions and K_L (L/mg) is the Langmuir constant related to the affinity between an adsorbent and adsorbate.

The non-linear form of Freundlich isotherm is described by Equation (6):

where K_F (mg/g)(L/mg)^1/n is the Freundlich constant and 1/n (0 < n < 10) is a Freundlich intensity parameter.

The equilibrium isotherms of Orange II adsorption onto LDH-R and LDH-W at the studied temperatures are shown in Figures 6–8. The equilibrium isotherms parameters, together with the obtained correlation coefficients, are presented in Table 2.

The adsorption performance of the studied materials increases with the increase of the working temperatures. At 323 K the maximum adsorption capacities developed by LDH-R and LDH-W is 117 and 126 mg/g, respectively. These values are almost 3 times higher than the maximum adsorption capacities resulted at 293 K. Due to the fact that the isotherms present an S shape at 323 K the initial concentration of Orange II was increased until the equilibrium was reached. According to Giles and et al., this shape is obtained when the compounds containing a sulfonate group and an aromatic ring are absorbed vertically to the adsorbent surface, implying strong intermolecular forces (Giles et al., 1960). At a higher temperature, the diffusion and mobility of the Orange II particles are increased leading to a higher mass transfer rate from solutions to the surface of the studied LDHs. Furthermore, this behavior suggests that the adsorption of Orange II onto LDHs materials is a chemisorption. By increasing the temperature, the rate of the chemical reaction increases, the favorable intermolecular forces between the LDHs surface and dye are much stronger than those between adsorbate and solvent (Kaur et al., 2013; Gupta et al., 2017). The increase in adsorption capacities of adsorbents with an increase in temperature may also be attributed to the pore size enlargement (Demirbas et al., 2008; Banerjee and Chattopadhyaya, 2017).

It is observed that for both adsorbents the curves obtained from the two models reproduce, with sufficient accuracy, the experimental data. Also, for various temperatures and both models, the values of the correlation coefficient R² are higher for LDH-R and, for the same type of adsorbent, these values are higher for the Langmuir model (Table 2). Furthermore, the maximum adsorption capacities developed by the studied adsorbent and experimentally determined are close to those calculated from the Langmuir isotherm, regardless of the working temperatures. These data indicate that the Langmuir model described the adsorption of Orange II onto the studied LDHs on the entire concentration range, suggesting that the adsorption takes place from a monolayer onto the adsorbent surface (Tran et al., 2016). The LDH obtained from waste developed a higher maximum adsorption capacity than the LDH synthesized from reagents (Table 2).

The strength of adsorption is indicated by the magnitude of the Freundlich constant (K_F), which is the highest at the highest temperature (for both adsorbents) and for LDH-W (at the same temperature). For all the cases the Freundlich parameters 1/n present sub unitary values which indicates the affinity of Orange II for the studied adsorbents (Gupta et al., 2017).

In order to determine if the Orange II adsorption takes place due to a physisorption or a chemisorption process, the adsorption energy E was estimated by applying the Dubinin-Radushkevich (D-R) isotherm. The linearized form of this isotherm is given in Equation (7) (Hadi et al., 2010):

where B_D is the D-R model constant (mol²/kJ²) and ε is the Polanyi potential (Equation 8):

The adsorption energy E whose magnitude can provide information on adsorption mechanisms is determined from Equation (9), B_D being determined from the linear representation of lnq_e=f(ε²):

The D-R isotherm parameters are presented in Table 2. The low values of the adsorption energy confirm that the dye adsorption on both adsorbents is physically achieved. The maximum adsorption capacities calculated from the D-R isotherms present values lower than those experimentally obtained.

Due to the fact that the Langmuir isotherm best described the adsorption of Orange II onto the studied LDHs, the equilibrium experimental data were also fitted with its four linear forms which are presented in Equations (10)–(13) (Tran et al., 2016):

Langmuir linearization (type 1):

Langmuir linearization (type 2):

Langmuir linearization (type 3):

Langmuir linearization (type 4):

The symbols have the same meaning as described before. The applicability of these isotherms was compared based on the error function values: correlation coefficient R², chi-square test χ² (Equation 14) (Hadi et al., 2010), average relative error ARE (Equation 15) (Li et al., 2017), and normalized standard deviation Δq (Equation 16) (Lv et al., 2007):

The Langmuir parameters determined from its linear representations (figures omitted for the sake of brevity) are presented in Supplementary Tables 3, 4.

Each of the four linear forms of the Langmuir isotherm will result in different parameter estimates as shown in Supplementary Tables 3, 4. The adsorption of Orange II onto LDH-R and LDH-W adsorbents is well-represented by the Langmuir type I model with highest coefficient of determinations; R² values as shown in Supplementary Tables 3, 4, indicate that there is strong evidence that the sorption of the chosen dye onto the prepared samples follows the Langmuir isotherm.

The thermodynamic parameters were calculated according to the following equations (Kaur et al., 2013; Banerjee and Chattopadhyaya, 2017; Gupta et al., 2017):

where: ΔG° is Gibbs free energy (KJ/mol), ΔS° is the entropy and heat of adsorption (J/mol K), ΔH° is enthalpy (kJ/mol), T is the absolute temperature (K), R is universal gas constant (8.314 J/(mol·K)), and K_d is the distribution coefficient.

ΔH° and ΔS° were obtained from the slope and intercept of van't Hoff plots of lnK_d vs. 1/T (Figure 9) and the resulted values, together with the Gibbs free energy calculated with Equation (17) and the correlation coefficients, are presented in Table 3.

The positive value of ΔH° thermodynamically substantiates the assumption that the adsorption of Orange II on the studied LDH is endothermic in nature. The feasibility and spontaneity of the adsorption process was confirmed by the obtained negative values of ΔG°. The positive value of ΔS° suggested an increase in randomness at the solid/liquid interface during adsorption of Orange II dye onto the studied materials. Adsorption is then likely to occur spontaneously at a high temperature because ΔH° > 0 and ΔS° > 0 (see Table 3) (Kaur et al., 2013; Banerjee and Chattopadhyaya, 2017; Gupta et al., 2017).

The XRD patterns of exhausted adsorbents are presented in Figure 10. The diffraction peaks of materials after Orange II dye adsorption are characteristic for layered double hydroxides with carbonate as an interlayer anion, since no modification in the position of the main peaks and basal spacing occurred. The basal spacing d₀₀₃ and the d₁₁₀ distances (cation-cation distance in the hydroxides layer) of the as-synthetized materials were d₀₀₃ = 7.605 Å (d₁₁₀ = 3.082 Å) for Zn₂Al – LDH prepared from zinc ash and d₀₀₃ = 7.598 Å (d₁₁₀ = 3.080 Å) for Zn₂Al – LDH prepared from analytical grade reagents (Cocheci et al., 2018). For the materials separated after Orange II adsorption, the basal spacing and d₁₁₀ distances are: d₀₀₃ = 7.593 Å (d₁₁₀ = 3.079 Å) for LDH – W (prepared from waste) and d₀₀₃ = 7.587 Å (d₁₁₀ = 3.077 Å) for LDH – R (prepared from analytical grade reagents). Studies regarding Orange II intercalation in the interlayer space of layered double hydroxides find that the basal spacing became larger due to this intercalation (Geraud et al., 2007; Liu et al., 2007). The unchanged basal spacing in our case suggests that, due to the presence of a carbonate anion in the interlayer space of LDH (due to the strong electrostatic attraction and the hydrogen bonds that the carbonate creates with the hydroxides layer), the organic dye was adsorbed on the surface of the layered double hydroxides.

The FTIR spectrum of Orange II (Figure 11) presents several absorption bands between 1,700 and 1,000 cm⁻¹; the bands with maxima at 1,662 and 1,622 cm⁻¹ correspond to a combination of phenyl ring vibrations with stretching of the C = N groups (Stylidi et al., 2004; Zhang et al., 2007). The band with a maximum at 1,614 cm⁻¹ correspond to carbonyl (C = O) stretching vibrations and the maximum at 1,473 cm⁻¹ are attributable to N = N stretching vibrations. The bands with maxima at 1,396 cm⁻¹ and 1,320 cm⁻¹ could be attributed to O – H bending vibrations (Stylidi et al., 2003). The absorption band with a maximum at 1,090 cm⁻¹ could be attributed to the ν(N – N) stretching vibrations, and the maxima located at 1,153 cm⁻¹ and 1,045 cm⁻¹ correspond to the coupling between the phenyl ring and the ν_s(SO₃) mode of the sulfonate group (Zhang et al., 2007).

The FTIR spectra of exhausted adsorbents present the main absorption bands that are characteristic of layered double hydroxides but also show some small shoulders that correspond to the vibrations of sulfonate groups of Orange II dye (Figure 11). The main FTIR absorption bands of LDH – R (Figure 11B) and LDH – W (Figure 11C) could be attributed as follows: the strong band at 3,380 cm⁻¹ is attributable to O – H stretching vibrations of Zn and Al hydroxides from the layer, the band with a maximum at 1,360 cm⁻¹ could be assigned to the stretching ν₃ mode of carbonate. The small shoulders at 1,504 cm⁻¹ can be attributed to a lowered carbonate symmetry that activated the ν₁ vibration mode of carbonate (Carja et al., 2001). The lowering of the carbonate symmetry activated the appearances of some weak shoulders corresponding to the ν₂ and ν₄ vibration mode of carbonate, visible at around 920 cm⁻¹ and 600 cm⁻¹, respectively (Kloprogge et al., 2002). The band located around 770 cm⁻¹ and the band at around 550 cm⁻¹ may be due to the Al – O or Zn – Al – O bond vibrations in the brucite-like layer (Cocheci et al., 2018). The small shoulders that are located at 1,120 cm⁻¹ and 1,005 cm⁻¹ could be assigned to the bands of sulfonate group of Orange II dye that are shifted to lower frequencies due to the interactions with the hydroxides layer (Geraud et al., 2007).

The adsorption mechanism of Orange II onto the LDH-R or LDH-R adsorbents might involve two paths: external surface adsorption of dye via electrostatic interactions or replacement of carbonate anions from the interlayer region by the sulfonate group via anionic exchange (Lei et al., 2017; Srilakshmi and Thirunavukkarasu, 2019). From the structural characterization of exhausted materials after adsorption, it could be concluded that, in this case, the mechanism of Orange II adsorption onto the studied materials involves electrostatic interactions. This conclusion is in agreement with reports of other researchers who studied the adsorption of different anionic dyes on similar adsorbent materials (Geraud et al., 2007; Khosla et al., 2013; García et al., 2014; Jin et al., 2014). The schematic representation on Orange II adsorption onto the surface of adsorbents is presented in Figure 12.