An efficient reusable perylene hydrogel for removing some toxic dyes from contaminated water

The synthesis of adsorbents that meet the need for large-scale production at relatively low cost and are capable of removing anionic and cationic toxic dyes from aqueous solutions, with high sorption capacity and reusability, is urgently needed from an environmental and industrial viewpoint. In this context the identification of hydrogels that remove dyes efficiently under ambient conditions and at near-neutral pH without the necessity of pre-treatment is an imperative. In this study we report the preparation of two hydrogels using the redox polymerisation of acrylamide, hydroxyethylmethacrylate (HEMA) and N-isopropylacrylamide (H1) and acrylamide, HEMA, N-isopropylacrylamide and perylene-5-ylpent-3-yne-2-methylprop-2-enoate-co-2-methyl-2-(prop-2-enoylamino) propane-1-sulfonic acid (PePnUMA-co-AMPS) (H2). These hydrogels proved to be effective for the removal of methylene blue (MB), fuchsin acid (FA) and Congo Red (CR) from aqueous solution at near-neutral pH where their adsorption behaviour was in keeping with the Langmuir model having qmax values of 769.2 mg g −1 (MB), 1666.7 mg g (FA) and 2358.2 mg g (CR). The adsorption of MB and FA by these hydrogels follows pseudo-first-order kinetics, whilst the adsorption of CR follows pseudo-second-order kinetics. Detailed thermodynamic analysis indicated that the dye–adsorbent interaction is primarily one of physisorption in nature. Finally, desorption studies carried out in 1.0 mol L NaClO4 indicated that these adsorbents could be recycled at least four times using a variety of dyes while maintaining their mechanical properties. © 2021 The Authors. Polymer International published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.


INTRODUCTION
Due to the significant development and scale up of various industrial activities, such as cosmetics, leather, textiles, paper and paint manufacture, enormous quantities of wastewater containing toxic and carcinogenic dyes have been produced as a by-product. In many cases such discharges render groundwater unfit for human consumption 1,2 or remain present in both raw and treated river water, despite prior effluent treatment by manufacturing facilities. 3 These substances pose a threat to human health and have been linked with significant carcinogenic and mutagenic consequences for humans and aquatic species. 4,5 Dyes are one of the most significant water pollutants that are generated by industrial activities. [6][7][8][9][10] Therefore, it is of critical importance to remove these dyes prior to discharge into wastewater streams. However, reducing dye content in wastewater streams is not an easy task as dyes are recalcitrant organic compounds and resistant to many treatment methods. 11 Numerous studies have used many methods for wastewater treatment, including physical methods, chemical processes and biological activities. These processes include degradation, coagulation, photocatalysis, oxidation, ion exchange and adsorption. [11][12][13] Adsorption is a more efficient treatment process for removing dyes due to its potential high efficiency, ease of access, 14 simplicity and wide range of adsorbent availability. Low-cost adsorbents, many of which are themselves waste products, [15][16][17][18] have been utilised previously in wastewater management. 19 Because of their relatively low cost of manufacture and high water absorption capacity, polymeric hydrogels are among the best candidates for the development of new adsorbent materials for the removal of heavy metals and dyes. 20 Hydrogels are three-dimensional crosslinked polymeric network structures capable of swelling to retain large volumes of water or aqueous fluids due to the inclusion of hydrophilic groups. [21][22][23] The specific properties of such gels vary depending upon the materials used in their synthesis and factors such as degree of crosslinking etc. but are typically characterised as possessing soft 'rubbery' characteristics when hydrated. 21,24 Their use is widespread in a variety of areas such as tissue engineering, drug delivery systems, agriculture, energy and wastewater treatment. [25][26][27][28] Two hydrogels were used in this study: H1 as a control gel and H2. 22 Methylene blue (MB) is a cationic dye that has several medicinal applications and uses in textile production 29,30 but has harmful corrosive and irritant effects. 31 Congo Red (CR) is an anionic dye. Its use has been banned in many countries, 32 mainly due to its carcinogenic properties; 33 however, its use persists in printing, paper, rubber and textile production. 15,34 Fuchsin acid (FA) is an acidic dye, commonly used in bacterial staining. 35 Treatment of wastewater containing FA has received significant attention due to its toxicity. 36 Xing et al. report a hydrogel formulated from gelatin, carboxymethyl cellulose and polypyrrole with exclusive FA adsorption at 94% removal efficiency. This removal efficiency remained unchanged over seven cycles of adsorption/desorption. 37 Jana et al. 19 used acrylamide, N,N-dimethylacrylamide and hydroxyl ethyl cellulose hydrogel for the removal of CR; the adsorption followed pseudo-second-order kinetics, and the Langmuir adsorption isotherm was well suited to the sorption data. The hydrogel showed a high adsorption capacity (102.4 mg g −1 ).
The isothermal studies of FA removal by modified zeolite showed that the adsorption was fitted well by the Langmuir isotherm with an adsorption capacity up to 31.0 mg g −1 while thermodynamic parameters indicated that the adsorption of FA using modified zeolites was exothermic. 38 The adsorption of MB onto poly(vinyl alcohol)-xanthan gum hydrogels displayed second-order kinetics and the Langmuir sorption isotherm. 39 A hydrogel that was formulated from 3-acrylamidopropyltrimethyl ammonium chloride and 2-acrylamido-2-methylpropanesulfonic acid was used for the removal of crystal violet (CV) and CR dyes in an aqueous medium and it was found to be more selective and efficient at pH 5.0 and 9.0 for CR and CV, respectively. 40 The formulations of new hydrogels derived from poly(acrylic acid), sodium humate and sodium alginate were investigated as efficient and selective adsorbents for MB and CV dyes. The binding capacities of these dyes were 367.0 and 359.0 mg g −1 , respectively, and the absorption was fitted well with the Langmuir sorption isotherm. 41 The current study involves the preparation of interpenetrating polymer network (IPN) hydrogel H1 from acrylamide with hydroxyethyl methacrylate (HEMA) as a control and IPN hydrogel H2 from the same components and perylene-5-ylpent-3-yne-2-methylprop-2-enoate-co-2-methyl-2-(prop-2-enoylamino)propane-1-sulfonic acid (PePnUMA-co-AMPS). The prepared hydrogel H2 was fully characterised and used in our previous work to sequester toxic, heavy metals from aqueous solutions. 42 In the present study, H2 is also used to remove organic dyes: methylene blue (MB), fuchsin acid (FA) and congo red (CR) . It was found that H2 exhibited good efficiency for removing these dyes.

Chemicals
All reactants, reagents and dry solvents were purchased from Merck (Watford, Hertfordshire, United Kingdom), Acros Organics and Fisher Scientific (Loughborough, United Kingdom) and were used without further purification except for 2,2 0 -azobis(2-methylpropionitrile) which was recrystallised from methanol and N-bromosuccinimide which was purified by recrystallisation from water prior to use. All glassware was heated in vacuum before use and all reactions were performed under nitrogen gas. Thin-layer chromatography (TLC) was carried out using DC-Fertigfolie POLY-GRAM® SIL G/UV254 precoated TLC sheets with substrate detection by UV light (254 and 365 nm).

Synthesis
Synthesis of the monomer and copolymer 5-(Perylen-3-yl)pent-4-yn-1-yl methacrylate (PePnUMA) (monomer) and PePnUMA-co-AMPS were synthesised and fully characterised according to our previously reported procedure. [42][43][44] Preparation of IPN hydrogels H1 and H2 H1 (control) was prepared by dissolving acrylamide (5.0 g), HEMA (2.0 g) and methylene N,N 0 -methylenebis(acrylamide) (0.72 g) as a crosslinker in 15 mL deionised (DI) water. To this homogeneous solution was then added ammonium persulphate (200 μL of a 10% w/v aqueous solution) as initiator and N,N,N 0 ,N 0tetramethylethylenediamine (25 μL) as an accelerator for initiator decomposition and the reaction mixture was stirred at room temperature for 10 min in order for the polymerisation reaction to go to completion. This process was repeated for the hydrogel H2 with the same components and PePnUMA-co-AMPS (0.2 g). The resulting hydrogels were dried and washed three times with DI water to remove any monomer residue, filtered and dried under vacuum (1 Torr) at 25°C for 24 h.

Preparation of aqueous dye solutions
Stock dye solutions were prepared by dissolving 1.0 g of each dye (MB, FA and CR) in DI water (1 L) to give a concentration of 1.0 g L −1 . Successive dilutions of these dye stock solutions in DI water were then used to obtain the working concentrations (40.0, 250.0 and 600.0 mg L −1 solution of MB, FA and CR respectively). The molecular structures and ⊗ max of these dyes and of PePnUMA-co-AMPS are shown in Table 1.

Batch adsorption experiments
Batch adsorption experiments of MB, FA and CR were carried out to evaluate the adsorption parameters and factors influencing the adsorption. A total of 25.0 mg of adsorbents H1 and H2 were placed in 100 mL solutions (40.0, 250.0 and 600.0 mg L −1 solution of MB, FA and CR respectively). The adsorption experiments were conducted on a thermostat shaker at 25°C, and the adsorption media was stirred magnetically at 225 rpm for a specific period of time (15-90 min). After adsorption, the adsorbent was separated from the solution by simple filtration. The concentration of dye was determined by UVvisible spectroscopy at the ⊗ max of each dye. UV-visible spectra were recorded using a T80 + spectrophotometer between 200 and 800 nm using a quartz cuvette with a path length of 1 cm.
Adsorption experiments were carried out at optimum conditions applying an agitation time of 60, 15 and 90 min at pH 9.0, 5.0 and 7.0 for MB, FA and CR respectively. The concentration of these dyes on H1 and H2 adsorbent hydrogels was calculated using the following equation: 45,46 where C 0 and Ce (mg L −1 ) are the initial and equilibrium concentration of dye in the solution, V (L) is the volume of solution, w (g) is the mass of hydrogel and q e is the adsorption capacity (mg dye per g hydrogel).
In order to study the adsorption kinetics and to calculate the kinetic parameters enthalpy (ΔH°), entropy (ΔS°) and free energy (ΔG°), experiments were conducted at 25, 45 and 65°C using 25 mg of H1 and H2 in 100 mL solutions of 40, 250 and 600 mg L −1 of MB, FA and CR respectively. By varying the pH of the dye solutions from 1 to 14, the maximum adsorption capacity was found at optimum pH. The pH of the dye solutions was adjusted using 0.1 mol L -1 solutions of HCl and NaOH.
The desorption of dyes from the hydrogels was carried out by applying five adsorption/desorption cycles of the same adsorbent. The maximum dye adsorption was accomplished by applying the optimum agitation time and pH for each dye-hydrogel combination. The desorption experiments were carried out by immersing dye-loaded adsorbent into a 1.0 mol L -1 solution of NaClO 4 , and the mixture was stirred at 25°C for 45 min. The desorbed dyes were then separated by centrifugation and filtration. The concentration of the dye in each solution was determined spectrophotometrically. [39][40][41] The removal efficiency of the MB, FA and CR dyes was calculated using 47 where S is the efficiency of dye desorption, C d is the dye concentration in solution after desorption (mg L −1 ) and V d is the volume of the eluent (L).

RESULTS AND DISCUSSION
The synthesis of the PePnUMA monomer, the corresponding PePnUMA-co-AMPS polymer network (1:30 monomer feed ratio) and the fabrication of semi-IPN hydrogels were attempted using our previously reported procedure. 48 From our previous work these systems have good chemical and thermal stability and show efficient detection and removal of heavy metal ions (e.g. Co2+, Cu2+, Ni2+, Hg2+ and Pb2+) from aqueous media at neutral pH at room temperature. The adsorption performance of H1 and H2 towards MB, FA and CR dyes was investigated by using batch system experiments at optimum pH, contact time, temperature and initial concentration (C 0 ) for each dye.
The important parameter influencing the hydrogel's adsorption capability is pH. The effect of pH on dye adsorption capacity using these hydrogels was measured between pH 1.0 and pH 14.0. Figures 1(A)-1(C) illustrate the influence of the variation of pH on the adsorption capacities of H1 and H2. Figure 1(A) shows that the adsorption capacities of H1 and H2 increase with increased pH for cationic dye MB and decrease with increased pH for anionic dyes. In the case of the cationic dye MB the electrostatic attraction between MB and the ionised sulfonic acid and phenolic groups present in the hydrogel at alkaline pH favour adsorption of the dye onto the hydrogel surface. [49][50][51] In the same way, the optimum pH value is 5.0 for FA adsorption; however, at low pH more H + ions are available, thereby increasing electrostatic attractions between negatively charged FA and CR dye anions and positively charged adsorption sites. This is due to the protonation of the -CONH 2 group which leads to an increase in the adsorption capacity. 19 Also, the abundance of OH − ions at high pH may cause an ionic repulsion between the negatively charged surface and the anionic dye molecules. The determination of optimum contact or agitation time 52 was carried out at diverse dye initial concentrations (20.0-700.0 mg L −1 ) and the best initial concentrations of MB, FA and CR dyes on H1 and H2 were found to be 40.0, 250.0 and 600.0 mg L −1 respectively.
show the plots for the effect of agitation time for the adsorption of MB, FA and CR dyes at pH 9.0, 5.0 and 7.0, respectively, onto adsorbent hydrogels at different temperatures. The adsorption of MB increased rapidly from 1 to 30 min; the adsorption capacity increases as the temperature increases (from 25 to 65°C) and the equilibrium attained within 60-150 min suggests an endothermic process. Figures 2(B) and 2(C) show the adsorption equilibrium times for FA and CR to be 15 and 90 min respectively. Since the adsorption of MB, FA and CR is endothermic, this tendency for the adsorption capacities of the adsorbent was expected. Thus, the optimum agitation times for all further experiments were chosen as 15, 60 and 90 min for the adsorption of MB, FA and CR dyes respectively.
A comparison of the maximum adsorption capacity of H2 with some reported adsorbents in the literature 37,53-59 for the removal of toxic dyes is summarised in Table 2. The results in this table show that the values of adsorption capacity of H2 are much better than some other reported adsorbents. This indicates that H2 is an excellent absorbent and can be used for the removal of dyes from aqueous media. Figure 3(A) shows the effect of pH on the adsorption capacity (q) of dyes onto hydrogels H1 and H2 while Fig. 3(B) shows the effect of temperature on the adsorption capacity (q) of dyes onto hydrogels H1 and H2.
These results indicate high adsorption efficiency of the hydrogel H1 due to the electrostatic interaction between the dyes and the hydrogel H1, while hydrogel H2 showed much higher adsorption efficiency. This observation can be rationalised by the presence of the large, polycyclic, aromatic residues which are embedded within the newly synthesised hydrogel which enables π-π stacking between the aromatic core of the polymer PePnUMA-co-AMPS and the dye molecules in a process that is augmented by H-bonding. [61][62][63] To further comprehend the adsorption and explore the adsorption process mechanism, the adsorption equilibrium isotherms between adsorbents and adsorbate were studied at a fixed temperature of 25°C. Adsorption data are generally described by adsorption models, such as the Langmuir and Freundlich isotherms. The Langmuir is illustrated by the following equation: 39 where C e (mg L −1 ) is the dye concentration remaining in solution at equilibrium, q e (mg g −1 ) is the equilibrium concentration of dye adsorbed by the hydrogel, K L (L mg −1 ) is the Langmuir adsorption constant and q max (mg g −1 ) is the monolayer adsorption capacity of the hydrogel. The slope of the plot of C e /q e versus C e is equal to 1/q max and the Y intercept is equal to 1/q max K L . The Langmuir equation is valid for monolayer adsorption of dye (adsorbate) onto the hydrogel surface (adsorbent), and it assumes that there are restricted and homogeneous adsorption sites. 64 The essential characteristics of a Langmuir isotherm can be expressed in terms of a dimensionless constant separation factor or equilibrium parameter R L which is defined as 65  wileyonlinelibrary.com/journal/pi where K L is the Langmuir constant and C 0 is the initial concentration.
The type of isotherm can be determined by R L values which indicate whether a sorption system is favourable (0 < R L < 1) or unfavourable (R L > 1). Figures 4(A)-4(C) give plots of the Langmuir adsorption isotherms of MB, FA and CR adsorbed onto hydrogels H1 and H2. Table 2,3 displays q max , K L , R 2 and the correlation  These results indicate that these hydrogels are favourable for adsorbing dyes from aqueous solutions under the optimised conditions applied in this study.
The Freundlich isotherm is based on a heterogeneous exponentially decaying distribution, which fits well to the tailing portion of the heterogeneous distribution of the adsorbent. The general Freundlich isotherm can be defined as 41 lnq e = ln K F + 1 n lnqC e ð5Þ   wileyonlinelibrary.com/journal/pi where K F (L mg −1 ) is a constant for the adsorption or distribution coefficient and represents the amount of dye adsorbed onto the hydrogel at equilibrium concentration. Figures 5(A)-5(C) show plots of the Freundlich adsorption isotherms of MB, FA and CR adsorbed onto the hydrogels. Table 3 displays also K F , 1/n and the correlation coefficients which were determined from the linear plot of ln q e versus ln C e . Table 3 shows the linearity of the Langmuir isotherm equation, which resulted in an improved fit compared to that of the Freundlich isotherm equation for MB, FA and CR dyes despite the complexity of the adsorption processes. However, this could suggest that there was a tendency for chemical adsorption to exist between functional groups of hydrogel H2 and the dyes. The values of 1/n were obtained from the slopes of the Freundlich isotherm plots in Fig. 5 for MB, FA and CR, and are below 1. These values are indicative of normal Langmuir isotherm behaviour of dye adsorption processes onto the H2 hydrogel. 66 Three kinetic models were tested to interpret the mechanism of adsorption of MB, FA and CR dyes onto the prepared hydrogels. The first model was pseudo-first-order; the mathematical expression of this model is given by 67 where q t and q 1 (mg g −1 ) are the amounts of dye adsorbed at time t and at equilibrium and K 1 (min −1 ) is the pseudo-first-order rate adsorption constant.
A plot of 1/q t versus 1/t gives the rate constants (K 1 ) and correlation coefficients. Figures 6(A)-6(C) represents the pseudo-firstorder equations for MB, FA and CR dyes at different temperatures and the calculated K 1 and q 1 values are shown in Table 3,4. The second kinetic model was pseudo-second-order, which can be represented by the following equation: [49][50][51][52] where q 2 is the maximum adsorption capacity (mg g −1 ) for pseudo-second-order adsorption and K 2 (g mg −1 min −1 ) is the equilibrium rate constant for pseudo-second-order adsorption. Values of q 2 and K 2 were calculated from the slope and Y intercept of the plot of t/q t versus t (Figs 7(A)-7(C)). The kinetic data for the adsorption of MB, FA and CR dyes onto the prepared hydrogels at various temperatures were calculated from the related plots and are summarised in Table 4. Intra-particle diffusion was the final model investigated in this study; the intra-particle diffusion model equation is shown as [49][50][51][52]61 where C is the Y intercept and K p is the intra-particle diffusion rate constant (mg −1 min -1/2 ) which were calculated from the slope of the plot of q t versus t -1/2 . Figure 8 displays the intra-particle  Table 4 shows the calculated K p and C values.
As is evident from the correlation coefficients in Table 4, the MB and FA dye adsorption systems follow the pseudo-first-order model and K 1 values increase with increase in temperature, while CR adsorption follows a pseudo-second-order model. However, the K 2 values were not always directly proportional to temperature.
Thermodynamic parameters can be determined from the thermodynamic equilibrium constant K L (or the thermodynamic distribution coefficient) and are expressed in the following equation: 67 where C a and C e are the equilibrium concentrations of MB, FA and CR dyes onto H1 and H2, as an adsorbent (mg g −1 ) and in the solution (mg L −1 ) respectively. The standard enthalpy change ΔH°(kJ mol −1 ) and the standard entropy change ΔS°(J mol −1 K −1 ) were calculated using the following equation: 39,67 where RAQ29 is the universal gas constant (8.314 J mol −1 K −1 ) and T is the absolute temperature. ΔH°was obtained from the slope of the plot of lnK L versus 1/T (K −1 ) and ΔS°was obtained from the Y intercept for the adsorption of MB, FA and CR dyes. The standard Gibbs free energy ΔG°(kJ mol −1 ) was calculated at different temperatures, as listed in Table 5, from the following equation: 39,49,67 ΔG°= ΔH°−TΔS°ð11Þ The activation energy E a (kJ mol −1 ), which is defined as the minimum amount of energy required in order for the adsorption process to proceed, was calculated from the Arrhenius equation: 39,49,67 where K is the rate constant of the pseudo-first-order kinetic model (min −1 ) for the adsorption system of MB and FA dyes whereas it is pseudo-second-order (g mg-1 min-1) for the adsorption system of CR, (this is based on R2 result). A is the Arrhenius factor, when ln K is plotted versus 1/T, a straight line with slope -E a /R is obtained. Table 5 lists the E a values for the adsorption of MB, FA and CR. The calculated thermodynamic parameters listed in Table 5 show negative values of enthalpy changes (ΔH°) for all dyes,  Pseudo-first-order Pseudo-second-order Intra-particle diffusion  indicating that the adsorption processes were endothermic. The positive values of the adsorption entropy (ΔS°) for the adsorption of MB, FA and CR onto the prepared hydrogels indicate an increase in the randomness at the adsorbent/solution interface and affinity of the prepared hydrogels towards these dyes. 60 The free energy change (ΔG°) values for the three dyes are negative and increase with increasing temperature, denoting that the adsorption processes proceed spontaneously and reversibly. The obtained ΔG°values are lower than 20 kJ mol −1 indicating that the process follows the physisorption mechanism. 68,69 The activation energy (E a ) values obtained are less than 40 kJ mol −1 and this further supports the fact that the adsorption process follows the physisorption mechanism. Desorption studies are essential to elucidate the reusability of an adsorbent, to understand the nature of the adsorption system and to evaluate the potential of recycling. Table 6 shows that the desorption percentages of H2 in the fourth cycle were 89.48%, 88.98% and 80.43% for MB, FA and CR respectively. In contrast, the percentage of desorption values of H1, the control, were 66.36%, 75.66% and 71.12% for MB, FA and CR respectively. This clearly demonstrates that the H2 hydrogel adsorbent can be used multiple times whilst retaining a high adsorption capacity.

CONCLUSION
In conclusion, the efficient removal of the dyes MB, FA and CR from aqueous solutions was observed when using the hydrogels H1 and H2 as adsorbents. The maximum adsorption of MB, FA and CR onto the prepared hydrogels was obtained at pH 5.0, 7.0 and 9.0 respectively. The adsorption systems of all dyes were found to increase with increase in temperature and fitted the Langmuir model, which allowed us to predict the chemical mechanism of the adsorption systems. From the linearised form (from the calculation of the Langmuir equation (q max )), values for the MB, FA and CR dyes were 769.20, 1666.70 and 2358.20 mg g −1 respectively, which exceed many of the reported values in the literature. 37,[53][54][55][56][57][58][59] Depending on the values of the correlation coefficient, the pseudo-first-order kinetic model indicated the adsorption kinetics of the MB and FA dyes, whereas the pseudosecond-order kinetic model accurately indicated the adsorption kinetics of the CR dye on the prepared hydrogel H2. The derived thermodynamic parameters from this study are indicative of a physisorption process; thus, the adsorption processes could have chemical and physical behaviour. Excellent regeneration efficiency is observed for hydrogels H1 and H2 over four successive