Selective and Highly Ecient Adsorption of a Mixture of Anionic and Cationic Dyes as Synthetic Wastewater Absorbents on Layered Double Hydroxide: Experimental and Computational Studies

Background: In the current work, the removal of cationic and anionic dyes from water was studied both experimentally and computationally. We examine the selectivity of the adsorbent, Zn – Fe layered double hydroxide (LDH) toward three cationic and anionic dyes. Methods: The chemical and physical properties of the prepared adsorbent before and after adsorption were investigated using FT-IR, X-ray diffraction, zeta potential, energy dispersive X-ray, X-ray photoelectron spectroscopy, particle size, HRTEM, and FESEM analysis; optical and electric properties were estimated. The influence of pH on the adsorption process was estimated. Monte Carlo simulations were performed to understand the adsorption mechanism and compute the adsorption energies.

Absorption of visible light from the top of the valence band (which is mainly composed of oxygen (O) 2p orbitals hybridized with Fe or Zn 3d orbitals) to the bottom of the conduction band (which is mainly composed of Fe or Zn 3d orbitals hybridized with O 2p orbitals) is the reason for the electronic band gap transition [18].
The process of degradation has been recommended as an effective environmental strategy for remediating organic pollutants such as dyes, especially when using low-cost semiconducting metal oxides as photocatalysts [19,20]. Therefore, after an experimental study of the adsorption process, we examined the LDH applied as a photocatalyst for the MO dye. In this study, we aim to analyze a multiadsorbate system by studying the selectivity of the main dye in the ternary system and then the interaction and behavior of two model cationic dyes (MB and MG) and one model anionic dye, MO [21], with Zn-Fe LDH applied as an effective adsorbent material. The Zn-Fe LDH prepared was well characterized by FT-IR, XRD, FESEM, HRTEM, UV-Vis spectroscopy, N2 adsorption/desorption, zeta potential, partial size analysis, and XPS. The adsorption mechanism of the electrical behavior was analyzed. This study highlights the potential application of Zn-Fe LDH as an efficient adsorbent of anionic and cationic dyes and its electrical properties, which extend its scope for application in environmental remediation processes.

Materials
Zn(NO3)2.6H2O was purchased from Chem-Lab NV, Belgium, Fe(NO3)3.9H2O. Hydrochloric acid was supplied by Carlo Erba reagent while NaOH was supplied from Piochem for laboratory chemicals, EGYPT. MB, MO and MG powder were purchased from Oxford laboratory reagents (India) Table 1. All mentioned chemicals had been used without any purification. The experiments and preparation of the material were performed using deionized water which free from CO2. Table 1 Chemicals and physical characteristics of methylene blue, methyl orange and malachite green.

Synthesis of Zn-Fe LDH
In our work Co-precipitation method was used to prepare Zn-Fe LDH. The added solution of Zinc and iron as nitrate precursors by 4:1 molar ratio. Slow flow rate 0.10 ml/min of NaOH (2M) solution was added till pH 10 for completes the precipitation. The resulting material was aged and kept at 60±0.5 °C for 12 hr and then was filtered and washed several times using distilled water to get rid of excess OHand then washing using ethanol. Finally, the resulted adsorbent sample had been dried at 80±0.5 °C for 24 hr [22].

Characterizations of the prepared material
The formed LDH/nitrate type had been characterized by XRD (PANalytical Empyean, Sweden). Accelerating voltage used was 40  The BET specific pore volume, specific surface area and pore size distribution of the Nanoadsorbents were determined by N2 adsorption applying an automatic surface analyzer (TriStar II 3020, Micrometrics, USA). For analyze the elemental composition of the prepared material (Kratos-England), the X-ray photo electron spectroscopy (XPS) with Al-KαX-ray mono chromatic source (hυ=1486.6eV) had been used. Zeta potential and hydrodynamic particle size were investigated by Nano-Zeta sizer (Malvern Instruments Ltd, United Kingdom). Using highresolution transmission electron microscopy (HRTEM, JEOL-JEM 2100) to determine the microstructures of the used LDH. The procedure of sample preparation for zeta potential measurements was as explained in our previous work [14] powder powder powder

Investigation of optical properties
The optical band gap of the sample material was performed with the Kubelka-Munk (K-M) function using the following equation [23]: where F(R), R, K(λ), and S(λ) are the K-M or re-emission functions, the diffuse reflectance of the sample, the absorption coefficient, and the scattering coefficient, respectively.
The absorption coefficient α was calculated using the Lambert Law from the measured absorbance results [24,25]: The following expression, suggested by Tauc, Davis, and Mott, is used: Equilibrium conditions were investigated by isotherm models and discussed in terms of nonlinear equations. We demonstrated the significance of our results using the statistical parameters R 2 and χ 2 : solutions of dye were run to confirm the reliability of data from the spectrophotometer. All experiments were performed in triplicate to ascertain reproducibility, and the average concentration was estimated by applying the mean and standard deviation (±SD) obtained from SPSS version 16. A p-value of less than 0.05 was taken to be statistically significant. (OH)50(NO3)5, and the cations distribution was adopted as reported by Fan et al. [27] for the 4 (M 2+ /M 3+ ) molar ratio. The cell and the studied dyes were optimized using the Universal forcefield [43], and the QEq charge method [28] was applied. The optimization process was done by the Forcite module. The convergence tolerance quality was set to be ultra-fine.

Monte Carlo (MC) simulation
The adsorption of MO, MB and MG molecules on the Zn-Fe LDH surface was carried out using MC simulation, by using the adsorption locator module that uses the Metropolis MC method to obtain the lowest-energy conformers between the adsorbate and adsorbent surface.
This module calculates (ΔEads). Two surfaces were cleavage from the optimized constructed cell, that is, LDH (001) and (010) surfaces. A 35 Å-thick vacuum slab above the LDH surfaces was created, and the two models are shown Fig. 1. As shown in this figure, the (001) and (010) planes represent the hydroxyl, and Zn-Fe LDH facets, respectively. The van der Waals force and electrostatic interaction were handled by the atom-based and Ewald methods, respectively.

Results and discussion
3.1 Morphology study FESEM images were applied to perform the morphology of the synthesis LDH as shown in Fig.2 (a-d). It displayed characteristic sheets, layers and hexagonal like morphology of LDH.
Using precipitation technique for preparation of the adsorbent may be the reason of the layers accumulation in this statue and also the reaction rate and time have been an important role on the thickness and shape of prepared sample. It could be due to the slow nucleation for precipitate formation [29]. EDX analysis was performed to confirm the presence of Zn, Fe and O as seen in

Surface property
As shown in Fig. 3f, N2 adsorption/desorption isotherms of synthesized LDH we find the existence of small closed loop at relative pressure >0.8, and the absence of a plateau at the P/Po near to 1 are followed to type II isotherm that is matches with H3-type hysteresis loop and that related to the mesopores/macroporous construction and capillary condensation process. The prepared sample has surface properties that estimated by (BET) method. BET surface area, the total pore volume and average pore size and of the sample are 71.61 m 2 /g, 0.078 cm 3 /g and 2.61 nm, respectively. The average pore size is <50 nm and there is extensive spreading of pore size up to16 nm (Fig.3h). (XPS) was applied for analysis the composition of LDH and assure the bonding of containing atoms. As observed from (Fig. 4) the XPS spectrum proved the existence of Fe, O and Zn. The Zn 2p spectra of XPS displayed two peaks asymmetric assigned to Zn 2p 3/2 and Zn 2p 1/2 core levels at 1021.8 and 1044.7 eV respectively, which related to that Zn charge were in 2+ oxidation state in LDH [14]. The O 1s peak as showed in Fig 4 of pure material (LDH) was at 531.80 eV which due to the presence of the -OH group of Zn-Fe LDH [31]. Moreover, the signal of Fe 2p of Fe 3+ peaks intricate by Fe 2p3/2 (711.7 eV) and Fe 2p1/2 (725.5 eV) refer to small positive change [32].

Fig. 4.
X-ray photoelectron spectroscopy spectra of the prepared Zn-Fe LDH.

X-ray diffraction investigation
The spectra of FTIR for Zn-Fe nitrate LDH was presented in Fig. 5a. The infrared peaks were around 3386 cm -1 which could be ascribed to characteristic free OH-stretching of the LDH structure vibration [33] with the interlayer water molecules and hydrogen bonding [34,35] and at 1621.72 cm -1 , the band exhibited the bending vibration of the interlayer H2O water molecules.
The peak situated at 1396.  [36,37]. The bands showed around 903 cm −1 could be related to the anti-symmetric deformation mode (ν4) of nitrate, and the weak out of plane symmetric deformation type (ν2) [38].
The XRD analysis of Zn-Fe LDH showed a crystalline layered phase (Fig. 5b structure of the Zn-Fe LDH. It could be observed that the bands were thin, proving that the exchange of Fe and Zn into the Zn-Fe LDH crystalline structure were achieved [39]. Since the peaks of (015) and (009) are higher intense than that for (003) which related to the creation of partly delaminated Zn-Fe LDH is probable this coincides with the innovative morphology through SEM images [35] (Fig. 5c)

Optical properties
The optical or photon properties of Zn-Fe LDH samples, such as the band gap energy, were identified using UV-Vis (NIR spectroscopy, DRS), and the resulting spectrum is displayed in   Its variety with photon energy is depicted in Fig. 9. From Fig. 9, we can see that the dielectric loss and dielectric constant diminish with photon energy analogs according to the behavior reported in the literature [40]. The magnitudes of the real dielectric constant are higher than the imaginary dielectric constant since they are reliant on n and k values. The real part of the dielectric constant contains a term that describes the amount by which it will impede the speed of light in the material, and the imaginary part shows how a dielectric absorbs energy from an electric field because of dipole movements [41]. The optical conductivity, which is related to the refractive index and absorption coefficient as given below, is then determined: where c is the speed of light in a vacuum. The reliance of the optical conductivity on the incident photon energy for various Zn-Fe LDH nanoparticles is displayed in Fig. 10. It can be observed that absorption is moderately low at high wavelengths, indicating a lower probability of electron transition to higher energy bands. On the contrary, at low wavelengths (i.e., at high energies), absorption is high, demonstrating greater opportunities for electron transitions.  Dipoles at this point do not comply with the field, and ε′= ε∞ (high frequency values of ε′) at extremely high frequencies (ω >>>1/τ). At low frequency, the dielectric constant is very high, and it is initially found to diminish with frequency and then to become somewhat stabilized. The high value of ε′ at frequencies less than 1 kHz, which increases as the frequency diminishes and the temperature increases, corresponds to the system's bulk effect. The issue of interfacial charge carriers is an important factor for the improvement of dielectric values in the frequency region. The requirement for the high value of the dielectric constant in the low-frequency area can be obstructing the charge carriers at the electrode. At low frequency, the dielectric loss is extremely high, but with increasing frequency, it falls rapidly. The dielectric loss increases with increasing temperature, analogous to the temperature reliance of the dielectric constant as shown in Fig.   11b. At chosen frequencies, the dielectric loss value is found to increase as a function of temperature. This mechanism can be joined with the lagging behind of charged ion species with the applied energy that prompts polarization.  Figure 12 shows the variety of AC electrical conductivity σac of Zn-Fe LDH as a function of frequency at various temperatures. The conductivity plot has the accompanying characteristics:

Conductivity
(i) scattering at lower and converging at higher frequencies of conductivity spectra with increased temperature. With increasing temperature, the plot shows that conductivity increments.
In the low-frequency region, frequency independent conductivity behavior is noticed, but that becomes sensitive in the high-frequency region, generally known as hopping frequency, moved to the higher-frequency side with increment of temperature. The conductivity increments in the higher-frequency region, because of the hopping of charge carriers in finite clusters.

Adsorption of dyes onto Zn-Fe LDH
The pH has a great effect on the adsorption process, where above a pH of 4, MO develops a negative charge owing to its dissociation constant of 3.47 (Fig. 13a). At pH 7, it reaches the maximum adsorption value, and beyond this value, the adsorption decreases. In acidic media, MO was protonated at its nitrogen-nitrogen double bond, and so the adsorption percent decreased owing to the forces of electrostatic repulsion. Reaching equilibrium in alkaline media was difficult owing to a high quantity of OH -, which competes with anionic MO and thereby prevents adsorption equilibrium [42]. Zeta potential is a technique to study the stability of the prepared material and dispersion in solution (Fig. 13b). The high stability of the Zn-Fe LDH nanoparticle dispersions is related to the high positive zeta potential over a large pH range (2-10), leading to strong particle-particle repulsions. Moreover, in considering the surface charge and surface properties for the adsorption behavior of prepared LDH with different dyes, the positive zeta potential of LDH is consistent with the electrostatic attraction between MO and LDH and supports the process of adsorption. By increasing the pH, the positive surface charge of the adsorbent decreases, thus leading to a decrease in the attraction of the negatively charged MO species, which increases the adsorption of anionic dyes (Fig. 13a(. In the acidic medium, the surface of the Zn-Fe LDH has a positive charge, and the MG and MB exist as cationic species (with a positive charge) [22]. Therefore, at low pH, adsorption is decreased owing to electrostatic repulsion (Figs. 13(c-d)). The MG and MB removal sharply increased to 81% and 84%, respectively, at a pH of 6 for MG and 9 for MB. At a pH higher than 6-8, the adsorption of MG likely increases owing to OH − groups that are attracted to the positive molecules of MG. It was reported that the PZC value of Zn-Fe LDH is 6.72 (Fig. 13e) [22], which is consistent with electrostatic attraction either between MG molecules or between MB and Zn-Fe LDH. To demonstrate the stability of the partial in several media, we estimate the practical size distribution (DLS) measurements. As observed in Fig. 13f, using Zn-Fe LDH with a smaller hydrodynamic size led to better aquatic stability and higher zeta potential. Moreover, this could allow its long-term application as a potential adsorbent for different pollutants in aquatic systems. Furthermore, to prove the stability of the adsorbent in low pH, we investigate the XRD spectrum of LDH in acidic media (Fig. 13g). The results showed that the material was maintained at characteristic peaks of LDH [43], as presented in Fig. 13g.

Adsorption isotherm studies
Adsorption isotherms explain how molecules of the adsorbate are distributed between the solid and liquid phases as the adsorption process reaches an equilibrium state. Modeling is crucial to compare and predict the LDH for which two-or three-parameter isotherm models apply well. Two-parameter models are commonly applied owing to simplicity and ease of fitting, and because the two-parameter     for this LDH is carefully compared with those for other adsorbents. Considering the high adsorption capacity, it seems that the Zn-Fe LDH prepared in this study could potentially be used as a cost-effective adsorbent for dye-polluted aquatic systems.

Ternary adsorption isotherm studies
As shown in Fig. 16, the adsorption capacity (qe) of dyes is negatively affected when the concentration of each dye in the mixture is increased in the range of 10-1000 mg/L. The decrease in the adsorption capacity of MB and MG is lower than MO, which is probably due to the affinity of MO toward the positively charge adsorbent surface (Table 5). On the other hand, MG and MB show the opposite behavior from that shown by MO since Zn-Fe LDH presents a high affinity toward anionic dyes (Fig. S1). We can conclude that interactions are being favored basically for the removal of anionic dyes like MO rather than cationic, as is the case for MB or MG. Langmuir, Freundlich, and Langmuir-Freundlich adsorption isotherms are applied to study the adsorption capacity exhibited by LDH used and the concentration of dyes at equilibrium. The Langmuir adsorption isotherm assumes homogeneous monolayer adsorption, whereas the Freundlich isotherm assumes heterogeneous multilayer adsorption. The adsorption isotherms assist in investigating the maximum adsorption capacity and the adsorption mechanism ( Table   5). As shown from the calculated parameters (Table 5) and the isotherm plots for dye adsorption ( Fig. 16) [30], which agrees with the adsorption results (Table 5) [40][41][42][43]46]. [40][41][42][43][44].

Kinetic studies
The equilibrium time is considered one of the important factor that effect on the cost and applicability of the adsorption process. Several experiments happened at different time to investigated the equilibrium time of the adsorption process. After that fitting the obtained data to three kinetic models. Figure S2 shows the effect of the time on the adsorption process of MO solution (pH 7) 30 mg/L onto Zn-Fe LDH (0.01 mg/50 ml) at 30 ± 0.5 °C. We observed from first order, pseudo-second order and Avirma were the best fit kinetic models with R 2 =0.997 (Table 6). These radical groups of •OH and O2 − will result in the decomposition of MO. The effect of initial MO concentration on photodegradation efficiency had been achieved by varying the initial MO concentration from 10 and 20 mg L −1 with other parameters such as catalyst concentration, reaction temperature and pH value remaining constant, and the result is shown in Fig. 17. It could be shown that the photodegradation efficiency decreases with an increase in the initial concentration of MO. The assumed reason is that equilibrium adsorption of reactants on the surface sites of catalyst increases with the MO concentration and with the increase of the initial concentration of MO, the dye molecules absorbed the light much more than the catalyst did, which is thought to have an inhibitive effect on the photodegradation process, and so the rate of photocatalytic reaction decreases [55].

Monte Carlo (MC) simulation
To understand the interactions between the dyes and LDH surface, MC study was performed using Zn-Fa LDH (001) and (010)  molecule on the LDH (010) surface displayed that the benzene-1-sulfonate moiety was located between two LDH layer as given in Fig 19 (a) and sulfonic oxygens were formed HBs with the hydroxyl hydrogens of both layers. A diazinyl-moiety nitrogen atom was found to form HB with hydroxyl hydrogen of one LDH layer. A cation-π interaction was observed between Zn cation and the phenyl group as shown in that figure. Additionally, the nitrogen atom of the dimethylamino group was found to be located at a distance of 2.51 Å from a Fe atom, indicating a coordination bond might be formed between them. In the case of MB, the aromatic nitrogen formed HBs with hydroxyl hydrogen of two LDH layers, as shown in Fig. 19

Conclusions
In this research, a coprecipitation method was applied to synthesize Zn-Fe-LDH, then the Zn-Fe-LDH was used for dye adsorption in single and ternary systems after investigating the structure of the prepared material using physical and chemical methods. For the single system, the maximum adsorption capacities were 230.68, 133.29, and 57.34 mg/g for MO, MB, and MG, respectively; for the ternary solution, the respective values were 217.97, 93.122, and 49.57 mg/g.
Experimental isotherm data fitted well with nonlinear isotherm models. Furthermore, pseudofirst-order, pseudo-second-order, and Avrami models described the adsorption kinetic data for MO, demonstrating chemisorption and physisorption properties. The optimum pH was 7, 9, and 6 for MO, MB, and MG, respectively. The adsorption mechanisms were investigated for dyes through XRD and FT-IR analyses and Monte Carlo simulation. Moreover, LDH proved that it could be applied as a photocatalyst for dye-polluted water.

*Competing interests
The authors confirm no competing interests