Hydrogel Nanocomposite of Acrylic Acid-Grafted Sodium Alginate-Based Zinc Oxide: Synthesis, Characterization, and Applicability for Crystal Violet Dye Removal

INTRODUCTION 
Textile dyes industries operations utilize plenty of water and thus cast large amounts of wastewater. The harmful effects of dyes on the body in the water are not confirmed to be toxic but may cause a reduction of dissolved oxygen and obstruction of the photosynthetic processes [1]. In this regard, several treatment methods such as ozonation [2], coagulation/flocculation [3], filtration membrane [4], liquid-liquid extraction [5], photocatalysis [6], and adsorption [7] have been applied in removing dyes from aqueous media. Thus, the adsorption process is of utmost preferred in water treatment due to its high efficiency, design simplicity, low cost, and non-toxicity. The adsorptive removal of the water contaminants depends on several significant factors such as the molecular structure of the contaminants, the nature and functionality of the adsorbent, and key operational conditions. The network structure is the main characteristic of super-adsorbent polymers that have a suitable degree of cross-linking. These polymers can form a very stable hydrogel by absorbing and retaining very large amounts of aqueous solutions or water. Thus retained water is hard to remove under pressure. Currently, they are utilized in several applications having hygienic products agriculture, sealing, food additives, pharmaceuticals, drug release systems, regenerative medicines, coal de-watering, artificial snow, biomedical applications tissue engineering, wound dressing separation of biomolecules or cells, diagnostics, and barrier materials to regulate biological adhesions biosensor [8-11].
Different prepared composites were used in the removal of dyes or other pollutants. Nejhad and co-workers utilized Poly(vinyl alcohol)/Montmorillonite nanocomposite hydrogels in the removal of the Methylene Blue from aqueous solution solutions [12]. In their study, they found that the dependency of gel content and swelling of the nanocomposite hydrogel adsorbent is in direct and inverse proportion to the amount of the incorporated montmorillonite to the adsorbent, respectively. Jawad and co-workers synthesized a cross-linked chitosan-epichlorohydrin/TiO2 bioadsorbent nanocomposite and used it to remove the reactive red 120 dye from an aqueous solution with an adsorption capacity of 210 mg/g at 30 °C [13]. They also synthesized a new composite by the modification of a cross-linked chitosan-glyoxal with kaolin clay as an inorganic filler [14]. The prepared composite was applied in the color removal and COD reduction of the dye remazol brilliant blue R with the highest adsorption capacity of 447.1 mg/g at 30 °C. Our previous work reported the synthesis and DFT studies of new hydrogel nanocomposite SA-g-p(AAc-co-MA)/TiO2. The synthesized nanocomposite was applied as an adsorbent to remove Brilliant green dye from aqueous media, which showed promising results under the optimized conditions [15].
Due to chemical stability, low cost, very low toxicity, photostability, Highly oxidizing, and exclusive semiconducting properties, ZnO nanoparticles are vastly utilized in several applications like photo-catalysis, dye-sensitized solar cells, pigments, drug delivery, etc. Malik and co-workers studied the effect of the high stability mechanical prepared CuS-ZnO nanocomposite on the photodegradation of Congo red and phenol pollutants [16]. Under visible light, the experimental degradation of Congo red and phenol showed that CuS could aid ZnO in this process. However, there are some problems with using ZnO nanoparticles as an adsorbent in an aqueous solution such as; recovery from the solution after the reaction completion, easy aggregation, and time consumption [17]. These problems have reduced the applications of ZnO nanoparticles in water treatment. Here, we incorporated ZnO nanoparticles in the (SA-g-(PAAC-co-VBS) hydrogel matrix and utilized the product (SA-g-(PAAC-co-VBS)/ZnO for the removal of CV dye in an aqueous solution. Further, the newly synthesized adsorbent showed a significant recycling performance with an efficiency of >80%, other than its other studied properties. 

MATERIALS AND METHODS
Sodium alginate (SA), Acrylic acid (AAC), N,N-methylene-bis acrylamide(MBA) , Sodium-4-vinylbenzenesulfonate(VBS), Potassium persulfate (KPS), Hydrochloric acid (HCl), Soduim hydroxide (NaOH), Ethanol(C2H5OH) and Zinc Oxide (ZnO) were supplied by Sigma–Aldrich. Crystal Violet(CV) is a textile dye obtained from Al- Hila factory (Iraq) for textiles (Molecular weight 407.98 g/mol , M.F C25H3OCl and ʎmax=590 nm). A standard solution (1000 mg L−1) was prepared by dissolving of 1 gm of CV in 1000 mL of distilled water.
FT-IR technique was utilized to identify functional groups on the surface which responsible for CV dye adsorption. Adsorbent surfaces (SA-g-(PAAC-co-VBS), (SA-g-(PAAC-co-VBS)/ZnO), and the after-adsorption-dye-loaded adsorbent sample was placed in an oven at 65 ◦C for 4 h. The samples were made as pellets, and the spectra of the adsorbent were recorded before and after the adsorption process in the range 4000–400 cm−1 using an Infrared spectrophotometer (FTIR, 8000, Shimadzu-Japan).

Preparation of SA-g-(PAAC-co-VBS)/ZnO) hydrogel 
The hydrogel was prepared by the free radical graft copolymerization method. Sodium alginate (SA) 1 gm was dissolved in distilled water (20 mL). In the reaction flask, a quantity of ZnO NPs (0.1g) was suspended in 20 ml of distilled water and shaken for 5 h followed by ultra-sonic for 2 h. Then, it was added to the solution sodium alginate, followed by stirring at 25 oC for 2 h until a well-dispersed homogeneous gel formed. At that point, Sodium 4-vinylbenzenesulfonate VBS (0.5 g) followed by Acrylic acid AAC (3 ml) was added with consistent blending at 25 °C. Then, potassium persulfate (0.03 g) in 1ml of distilled water was added with continuous mixing for 5 min. After that, N,N-methylene-bisacrylamide (MBA), (0.05 g) was added to the mixture with the a stream of N2. Finally, the reaction flask was set in a 70 °C water bath to complete the co-polymerization and obtain the hydrogel nanocomposite (SA-g-(PAAC-co-VBS)/ZnO). The obtained hydrogel was cut into small pieces and washed with plenty of distilled water, and then with ethanol to remove unreacted chemicals, dried in an oven at 60 °C, and then kept in a desiccator.
The synthesis steps are summarized in Fig. 1a. The decomposition of the initiator generates the free radicals that are transferred to SA to initiate the polymerization of AAC and VBS. Then, the monomers are grafted onto the SA backbone. After that, the addition of the cross-linker resulted in the formation of a three-dimensional (3D) network. The proposed adsorption process of CV is shown in Fig. 1b.

Adsorption study 
Batch adsorption experiments were carried out in 100 mL conical flasks, in a 100 ml solution with a 200 mg/L concentration of CV dye. To the CV solution, an amount of hydrogel 0.04 g was added. The conical flasks were shaken with a speed of 160 rpm for 2 h in a shaker water bath at room temperature. The effects of different parameters such as pH solution (3-10), contact time (2-240 min), adsorbent dosage (0.01-0.1 g), and solution temperature (25-40 oC) were studied. To analyze the residual concentration of CV dye in the solution, the samples were collected from the flask at pre-determined time intervals. Then, the residual of CV was determined using a UV-Visible spectrophotometer at λmax= 590 nm. The following equation was used to obtain the amounts of adsorbed and removed CV:

                                                 
Where qe = The adsorbed dye per unit of adsorbent (mg/g).  
C0= Initial concentration (mg.L-1), Ce= Equilibrium concentration (mg.L-1).
m= Dose of adsorbent (g), VL= Volume of solution (L).
The removal percentage (E%) of CV was calculated based on the reduction in absorbance at λmax value of the dye as follows [18]:

                                            
 
RESULT AND DISCUSSION 
FTIR spectral analysis 
Infrared analysis results are shown in fig. 2. In the alginate spectrum, a broad band 3000 to 3600 cm-1 is attributed to the stretching of COOH groups, which showed a little change in the intensity after modification with ZnO. The bands appeared at 2923-2925 cm-1 can be attributed to the stretching of C–H from glucose units in alginate [19]. In the spectrum of SA-g-(PAA-co-VBS)/ZnO, the characteristic band that appeared at 1718 cm-1 is assigned to the stretching of C=O in poly(acrylic acid), which significantly supports of grafting of AAC on SA. Moreover, the broadband at 450–800 cm-1 is assigned to the stretching of Zn-O-Zn groups [20, 21]. FT-IR spectra of ZnO NPs generally show two main bands, the first one at 627 cm-1 (corresponding to stretching of Zn–O) and the second at 1404 cm-1 (corresponding to stretching of Zn–O–Zn), which confirm the presence of ZnNP in SA-g-(PAA-co-VBS)/ZnO, thus the successful graft co-polymerization of SA with the ZnO NPs [22, 23].     

TGA analysis
Thermogravimetric analysis of the nanocomposite was investigated at a heating up to 600 oC, with a rate of 5 ◦C/min under dry nitrogen. The result of TGA of the as-obtained Architecture-like (a) SA-g-(PAAC-co-VBS), (b) SA-g-(PAAC-co-VBS)/ZnO NPs. Architectures further provide evidence that ZnO NPs incorporated into SA-g-(PAAC-co-VBS)/ZnO NPs architectures, as shown in Fig. 3. In the TGA curve {W(mg) vs temperature T(oC) plots} [24-26], a weight loss observed in the range of 25–600 °C, in which the weight loss of 42.4%, and 52.2% in the range of 25–300 °C for SA-g-(PAAC-co-VBS) and SA-g-(PAAC-co-VBS)/ZnO NPs, respectively. This is attributed to the desorption and then evaporation of bound water. The weight loss of 25.8% in SA-g-(PAAC-co-VBS, and 49.1% in SA-g-(PAAC-co-VBS)/ZnO NPs at the range 460-580 oC and 300-580 oC indicates the loss of cross-linked network structure, and the formation of anhydride with the removal of water molecules from carboxylic groups in polymer chains. The decomposition rate of SA-g-(PAAC-co-VBS)/ZnO NPs is less than that of SA-g-(PAAC-co-VBS). TGA results predicted that ZnO could behave as a thermal barrier, therefore it increases the thermal stability of synthesized SA-g-(PAAC-co-VBS)/ZnO NPs hydrogel [27-30].

FE-SEM and TEM analysis
FE-SEM technique was used as a tool in characterizing the physical properties and surface morphology of the new adsorbent. The synthesized ZnO NPs arranged in a flower-like shape pattern as shown in Fig. 4a. The pores give a substantial chance for ZnO NPs to entangle inside them [31]. The rough porous surface of (SA-g-(PAAC-co-VBS) and (SA-g-(PAAC-co-VBS)/ZnO) revealed changes in the morphology of phase for new irregular bulky particles presence on the surface (Fig. 4(b,c)). This lead to increase in surface texture protuberance and coarseness, and thus increasing the surfaces area of the absorbents that facilitate the water diffusion into the absorbent. FE-SEM of adsorbent was taken before and after dye adsorption on (SA-g-(PAAC-co-VBS)/ZnO) surface.
The physical structure of zinc oxide clearly noticed as the ZnO nanoparticles were linked to the outer and inner surfaces and filled the cracks and holes in the sample of SA-g-(PAAC-co-VBS). Therefore, the spherical-like shape of ZnO NPs could be considered to enhance the uptake of CV dye because of the enhancement of the surface area of SA-g-(PAAC-co-VBS) [17]. In the morphology of SA-g-(PAAC-co-VBS)/ZnO, the internal pores can be seen which favors the intra-particle diffusion of CV as appears in fig. 4d. ZnO NPs provided well-developed pores with a uniform distribution that led to a porous structure and large surface area. The pores and surface of SA-g-(PAAC-co-VBS)/ZnO were entirely occupied by CV, and this confirms the successful adsorption process. Fig. 4e shows the TEM images of SA-g-(PAAC-co-VBS)/ZnO, where ZnO NPs observed embedded inside the SA-g-(PAAC-co-VBS). Also, the incorporation of ZnO NPs into SA-g-(PAAC-co-VBS) is confirmed by the presence of Zn and O peaks in the EDS of SA-g-(PAAC-co-VBS)/ZnO as shown in Fig.4f [32]. The synthesized nanocomposite contains C, O, Zn, S, Ca, and, Al elements, which confirms the presence of ZnO onto SA-g-(PAAC-co-VBS)/ZnO. The values of the highest and lowest elements that existed in the modified SA-g-(PAAC-co-VBS)/ZnO were 63.4 wt.% and 0.3 wt.%, respectively [33].

XRD analysis
The XRD spectra of ZnO NPs, SA-g-(PAAC-co-VBS)/ZnONPs, and SA-g-(PAAC-co-VBS) are shown in Fig. 5. The ZnO NPs crystalline peaks that appear at 32o, 34o, and 36o confirm the presence of ZnO NPs. The broadening in the peak is due to graft co-polymerization in the chains of poly acrylic acid onto SA [34]. The vanishing of crystalline peaks was observed after the addition of ZnO NPs to the hydrogel. The conversion from crystalline to amorphous phase may be due to the diffusion of ZnO NPs into the micro-pores and macro-pores of SA-g-(PAAC-co-VBS)/ZnONPs. The XRD patterns of SA-g-(PAAC-co-VBS)/ZnO NPs showed a very little peak observed in the case of ZnO which confirm the incorporation of ZnO NPs within the SA-g-(PAAC-co-VBS) matrix [35]. 

Adsorption process
Effect of contacting time
The contacting time is one of the significant parameters in the assessment of the practical application of the adsorption process. The experimental results of adsorption of chosen CV dye concentration 200 mg.L-1 on the adsorbent surfaces of the (SA-g-(PAAC-co-VBS)/ZnO) with contact time (Fig. 6) revealed that the adsorption increased quickly during the first 15 min, and then gradually till the equilibrium [36]. Also, the adsorption capacity increases with the increase of contacting time till reaches the equilibrium, which is come from the saturation of the active sites of absorbent surfaces. An acceptable efficiency was found at a contact time of two hours and a concentration of 200 mg.L-1, as the increase in time will have a slight effect on the removal rate due to the saturation of the active sites [37, 38].

Effect of the adsorbent dose
Investigation of the removal percentage E% of CV dye on the adsorbent dose of hydrogel was performed with unadjusted pH using several doses (0.01- 0.1 g) of SA-g-(PAAC-co-VBS)/ZnO at a constant concentration of CV dye of 200 mg L-1. The amount of adsorbed CV dye decreased from 200 to 159 mg/g, with increasing of adsorbents dosage from 0.01 to 0.1 g, and the removal efficiency E% of CV dye increased from 73.4 to 99.6% (Fig. 7). The incremental dye removal becomes constant of the adsorbent SA-g-(PAAC-co-VBS)/ZnO at amounts more than 0.04 g as the amount of dye in the solution diminished because of semi-stable removal, thus the chosen optimum dose of the adsorbents was 0.04 g. Increasing the adsorption percentage with increasing adsorbent dose can be attributed to the increased adsorbent surface area which leads to increasing of adsorption sites [21, 39, 40].

Effect of pH 
Initial pH is an important factor in adsorption as it plays the main role in the protonation of functional groups, and in changing the surface charge of the adsorbent [17]. As shown in Fig 8, when the pH increases from 3 to 6.6, the adsorption capacity increases significantly, then followed by a slight decrease with the pH continuously increasing from 6.6 to 10. Complete removal of CV dye at the concentration of 200 mg.L-1 (100 mL) is achieved at pH= 6.6, and can reach the adsorption capacity of about 452.2 mg/g [41]. This is attributed to the de-protonation of functional groups in SA-g-(PAAC-co-VBS)/ZnO at this pH, which can simplify the electrostatic adsorption towards CV dye. It is worth to be mentioned that despite the adsorption behavior of CV has been weakened in the extremely acidic solution (pH= 3), the adsorption efficiency is still as high as 55.5 mg/g, which is better than the most common use adsorbents [40].

Kinetics model 
The adsorption kinetics model is a very important factor to estimate the rate-controlling step and adsorption mechanism. Thus, the three utmost widely utilized adsorption kinetics the first model, the second model, and the Elkovich model are applied to further analyze the adsorption result. Their nonlinear equations calculate as follows [42]:

where qt is the total amount of adsorbate that is adsorbed at time t (mg-1), qe is the equilibrium adsorption potential (mg. g-1), kf is the rate constant of the first model (min-1), and t is the contact time (min). k2 is the second-order rate constant. The fitted parameters of the adsorption kinetics model of CV (200 mg/L) at 30 oC were calculated from the non-linear regressions of the integrated Eqs. (8-10)[43]. The study of the fitted curve of the kinetics model of dye is shown in Fig. 9 and Table 1.
The related factors and correlation coefficients (R2=0.9977), (R2=0.9315), and (R2=0.843) can be obtained via the regression of the experimental results. The values of qe are pretty close to qexp of the first model, comparative to the sound and Elkovich model indicating that the adsorption process of CV dye onto the adsorbents fitted the first model. Also, the correlation coefficient obtained from the equation First model was equal to 0.9977, which further proposed that the total adsorption process of the SA-g-(PAAC-co-VBS)/ZnO obeyed the kinetic First model [44, 45].

Adsorption Isotherm 
Adsorption results are utmost often represented by the value of the equilibrium isotherm, in which the adsorption capacity (qe mg/g) is the concentration of the sorbent (the solid-phase) versus the concentration of the sorbate in the liquid phase (Ce mg/L). At the point of dynamic isotherm of the adsorption method, there is a requirement to determine the adsorbate spreading amidst the solid phase and liquid phase; this information can be gutted through adsorption equilibrium studies. In this study, two of the utmost generally utilized isotherms, namely the isotherm Langmuir and isotherm Freundlich were utilized to describe the adsorption isotherm. The non-linear forms of the Langmuir and Freundlich models [46, 47] are given as:

 
qe: The amount of adsorbed per unit weight (mg/g) of adsorbent at equilibrium.   Ce: Concentration of adsorbate in solution after adsorption at equilibrium (mg.L-1), (mol/L). qm is the empirical Langmuir constant that represents the maximum adsorption capacity (mg/g) of the total number of surface sites per mass of adsorbent. It may vary in different compounds because of differences in adsorbate sizes.     KL: Empirical Langmuir constant (L/mg) or the equilibrium constant of the adsorption reaction. Kf: Empirical Freundlich constant or capacity factor (L/mg) or the dye quantity adsorbed for unit equilibrium concentration. 1/n: Freundlich exponent, if n=1, the adsorption is linear; if n˂1, then the adsorption is chemical, and if the value is more than 1, then the adsorption will be physical [48-50]. 
A plot of qe vs  Ce (Fig. 10), where the values of KF and 1/n are obtained from the intercept and slope in the same order. The correlation coefficient (R2) value for the Freundlich model at 30 oC is R2= 0.9969 of CV onto (SA-g-(PAAC-co-VBS)/ZnO) [51, 52].      

Thermodynamic Study 
The results obtained from the study of temperature were utilized for the thermodynamic parameters. Gibbs free energy (∆G⁰) is presented in Eq. 5, which can be well expressed by means of ∆H and ∆S, at a constant temperature (Eq. 6). The linearized form of equations 5 and 6 results in Eq. 7, which is the van’t Hoff equation:

         
             
where ΔG° (kJ/mol) changes in Gibbs free Energy, ΔH° (kJ/mol) changes in enthalpy, ΔS° (J/mol. K) changes in the entropy of CV dye adsorption, R gas constant (8.314J/mol K), T absolute temperature (K). 
The data reveals that the equilibrium adsorption efficiency of CV dye was increased with the increasing temperature of the solution for all initial concentrations [53]. The values of ΔH and ΔS respectively can be calculated from the slope and intercept of vant Hoff equation by plotting of ln Xm vs 1000 K/T (Fig. 11). The thermo-dynamic parameters are given in Table 2. 
As shown in the table, the ΔG ⁰ values are negative for SA-g-(PAAC-co-VBS)/ZnO, which indicates the spontaneous nature of the sorption method. For all the sorbents, the positive value of ΔH suggests the endothermic nature of the adsorption method. Moreover, the positive values of ΔS indicate the increase in randomness at the interface of solid/liquid during the sorption of CV onto SA-g-(PAAC-co-VBS)/ZnO [54].

Regeneration of (SA-g-(PAAC-co-VBS)/ZnO).
The regeneration of hydrogels after sorption is one of the economically significant parameters. It aids in predicting the mechanism of CV dye removal from dye-loaded adsorbents, and recycling of used adsorbents, which in turn may reduce the costs and protect the environment from secondary pollution. For the desorption studies of CV, different agents were used at 0.01 N such as H2SO4, NaOH, H3PO4, HCl, HNO3, and water [55-58]. The synthesized adsorbent (SA-g-(PAAC-co-VBS)/ZnO) was regenerated with 100% in a diluted HCl solution. This is because of the conversion of –COO− to –COOH in acidic solutions which leads to the weakening of the electrostatic interactions between (SA-g-(PAAC-co-VBS)/ZnO) and CV. 
The reusability and performance of (SA-g-(PAAC-co-VBS)/ZnO) after adsorption of CV dye were investigated using HCl solution for up to 3 steps under optimum conditions. After the 3 cycles of using the adsorbent SA-g-(PAAC-co-VBS)/ZnO, the efficiency is still significant (>80%) and this reveals that SA-g-(PAAC-co-VBS)/ZnO is a probable renewable adsorber [32, 59] as shown in Fig. 12.

CONCLUSION
In the current research, we report the synthesis of the new superabsorbent (SA-g-(PAAC-co-VBS)/ZnO) using the free radical graft copolymerization method. Characterization techniques revealed the presence of ZnO in the hydrogel matrix which was confirmed via FT-IR and XRD. Introducing ZnO NPs into the hydrogel matrix of SA and AAC grafted onto VBS resulted in a unique structure, high pore volume, and high surface area which led to a high adsorption efficiency of superabsorbent. The superabsorbent exhibited good capacities for CV removal from an aqueous solution, as it successfully removed about 98.9%. The kinetics model result fits well with the model First-order, and the adsorption mechanism was controlled by intraparticle diffusion as well as boundary layer diffusion. The isotherm model which followed by the adsorption process in this work is Freundlich one with a maximum adsorption efficiency about of 1150 mg/g. The adsorption method was found to be spontaneous and endothermic physical adsorption. Hence, the synthesized adsorbents can be used effectively in the water treatment for removal of selective cationic dyes. Furthermore, desorption studies predicted that the superabsorbent can be utilized repeatedly for the regeneration of CV dye. In summary, the superabsorbent has the potential to be utilized as an environmentally friendly and efficient adsorbent for the removal of CV from an aqueous solution.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.