Quaternized triethanolamine-sebacoyl moieties in highly branched polymer architecture as a host for the entrapment of acid dyes in aqueous solutions

Wastewaters originating from the textile industry are polluted, as they contain residual color and other chemical substances (O'Neill et al. 1999). The non-biodegradable nature of the residual dye in the wastewaters may also obstruct light penetration, thus inhibiting aquatic life in the ecosystem (Walsh et al. 1980; Kuo 1992; Forgacs et al. 2004; Rai et al. 2005). In addition, many dyes are toxic and even carcinogenic and pose a serious threat to various microbiological or animal species (Willcock et al. 1992).

Apart from adsorption using low-cost adsorbents deriving from renewable resources or less expensive natural materials (Bouzaida & Rammah 2002; Aygun et al. 2003; Nakamura et al. 2003; Prado et al. 2004; Ozacar & Sengil 2005; Ferrero 2007; Amin 2009), numerous methods such as biological (Abadulla et al. 2000), electrochemical (Fernandez-Sanchez & Costa-Garcia 2000), photochemical (Hachem et al. 2001; Barka et al. 2008; Tahiri Alaoui et al. 2009; Benalioua et al. 2015), and membrane filtration (Ma et al. 2012) technologies have been successfully employed for the removal of dyes from wastewaters. Recently, several studies have shown the effectiveness of dendritic nanomaterials for water treatment due to their outstanding removal capacity, higher surface area and large number of active sites for interaction with pollutants (Qu et al. 2006; Saeed et al. 2008; Zhou et al. 2015).

Dendrimers are a class of highly branched three-dimensional polymers characterized by a compact shape that have numerous reactive functional end groups and room between branches for taking up guest molecules (Tomalia & Fréchet 2001). Dendrimers have been shown to be effective for dyeing fibers. For instance, Burkinshaw et al. (2000) used a dendrimer containing primary amino groups for the pretreatment of cotton in order to enhance the color strength of the fiber with reactive dyes. Polyamidoamine dendrimers have also been reported as promising candidates for different applications including water purification (Diallo et al. 1999; Xu & Zhao 2005). Extraction and encapsulation of dyes by dendrimers have been reported by several authors. At sufficiently low pH, dendrimers containing tertiary amine groups are able to give an acid–base interaction with acid dyes. The process is totally reversible since at alkaline pH, the number of positively charged sites on the surface of the dendrimer decreases, thus favoring the dyes' release. In this regard, Baars et al. (1997) investigated poly(propylene imine) dendrimers for the extraction of an acid dye from water by amine groups in an apolar solvent. The same principle of dye encapsulation has been applied by Cooper et al. (1997) for the extraction of an acid dye by a modified dendrimer in liquid CO₂.

One of the major factors that influences the performance of polymeric sorbents is the nature of the functional groups available on their surface for interactions with contaminants. These functional groups determine the removal capacity, stability, and reusability of the sorbent material. According to the literature (Wawrzkiewicz & Hubicki 2009a, 2009b, 2009c), the use of commercially available anion-exchanger resins to remove acid dyes from water has been thoroughly investigated. These resins showed high potential for adsorption of anionic dyes and excellent adsorption capacity due to their high content of positively charged functional groups. Moreover, complete regeneration without loss of their sorption capacity can be achieved in alkaline media (Karcher et al. 2001, 2002).

The current study was set to prepare a hyperbranched polymer as an efficient material for solid–liquid extraction of acid dyes. The hyperbranched polymer QTEAS was synthesized by a cost-effective one-step copolymerization of multifunctional A3 and B2 monomers, namely, triethanolamine and sebacoyl chloride, followed by methylation of amine groups. Quaternization of tertiary amine groups was necessary to lead to a material with plenty of positively charged sites on its surface, and this was also intended for rightly avoiding pH adjustment in our sorption studies. Sebacoyl monomer was chosen for its chain length (8 sp³ carbons) in order to get ample room between the branches in the QTEAS hyperbranched material, thus facilitating the encapsulation of guest molecules. Although its structure is not as perfect as that of dendrimers, QTEAS will still have many similar characteristics and properties of dendrimers, such as the three-dimensional globular architecture and abundant quaternary amine groups required for the removal of targeted dyes by electrostatic interactions. Two negatively charged dyes were chosen for the study: indigo carmine (IC), a divalent anion that is a real concern in textile wastewaters and a diazo dye with extended molecular structure; Evans blue (EB), a tetravalent anion (see Table 1). Furthermore, the structure of the prepared macromolecule is demonstrated by Fourier transform infrared spectroscopy (FTIR), cross polarization magic angle spinning (CPMAS) ¹³C NMR, thermogravimetric analysis (TGA), powder X-ray diffraction (DRX), and scanning electron microscopy (SEM). Sorption kinetics, isotherms, effect of pH, ionic strength, and effect of temperature have been investigated to identify a sorption mechanism of the dyes. Moreover, the data were analyzed using different well-known adsorption isotherms and kinetics models. The high performance of the QTEAS material after regeneration cycles has been carefully examined to ascertain its stability and reusability. Finally, the great efficiency of the hyperbranched polymer for dye removal from colored wastewater was clearly demonstrated.

Table 1

Molecular structures of targeted dyes

Dyes	Molecular structure	MW (g mol–1)	λmax (nm)	Charges on surface
Indigo carmine (IC)		466.35	610	2
Evans blue (EB)		960.81	610	4

Dyes	Molecular structure	MW (g mol–1)	λmax (nm)	Charges on surface
Indigo carmine (IC)		466.35	610	2
Evans blue (EB)		960.81	610	4

An infrared spectrum was obtained on a (2.5 wt %) sample in a KBr disk from 400 to 4,000 cm⁻¹ using a Nicolet Avatar 330 Fourier transform IR spectrometer. The CPMAS ¹³C NMR spectrum of QTEAS was recorded on a Bruker 300 (Digital NMR Avance) spectrometer. X-ray diffraction (XRD) patterns were recorded from 2θ = 3.5 to 70° on a Phillips X'Pert MPD diffractometer using monochromatic CuKα radiation (λ = 1.5418 Å) at 40 kV and 30 mA. The thermal stability of the sample was performed using TGA on a NETZSCH STA 409 PC/PG simultaneous thermal analyzer at a heating rate of 10 °C/min under nitrogen atmosphere. The morphology of the QTEAS hyperbranched polymer was examined at high magnification using a HITACHI S-4800 SEM.

Stock solutions (500 mg L^–1) of the dyes were prepared by dissolving IC and EB in distilled water, and test solutions of desired concentrations were obtained by further dilution with distilled water. Dye solutions of desired pH values were adjusted using HCl (0.1 N) and NaOH (0.1 N). The concentrations of the dyes were measured with a HACH DR4000 U UV-visible spectrophotometer at 610 nm for both dyes. The sorbed amounts of the dyes were determined from the difference between the initial and final concentrations by the following mass balance equation, q_e = (C_i − C_e)V/w, where q_e is the amount (mg g⁻¹) of dye sorbed, C_i and C_e are the initial and equilibrium dye concentrations (mg L⁻¹) in solution, respectively, V is the adsorbate volume (L) and w is the sorbent weight (g). All experiments described below were undertaken in either duplicate or triplicate.

The removal of IC and EB by QTEAS was investigated in batch experiments by stirring 50 mg of material with 50 mL of dye solution in 200 mL stoppered glass bottles at 25 °C for 3 and 6 h for IC and EB, respectively. Each isotherm consisted of 10 dye concentrations varying from 50 to 500 mg L^–1. The equilibrium concentrations of different combinations were measured by the spectrophotometer and referenced with the calibration curves. The kinetic measurements were carried out using similar equipment and conditions. The sample mass was 50 mg, and the volume of the dye solution was 50 mL (50 mg L⁻¹) in this series of tests. The mixtures were stirred at predetermined intervals of time, and were drawn for dye concentration analysis. Experiments with each dye were performed to determine the effect of pH on dye removal. The pH range studied was from 2 to 10. The sample mass was 50 mg, and the dye concentration was 50 mg L⁻¹ (50 mL) in this series of tests. The influence of temperature on the removal process was studied at three different temperatures (25, 35, and 45 °C) with QTEAS suspensions in IC and EB solutions (50 mg L⁻¹). The suspensions were stirred during 3 h and 6 h for IC and EB, respectively, and then the dye concentration was analyzed.

To assess the feasibility for consecutive reuse of QTEAS, sorption–desorption studies of IC and EB on the hyperbranched polymer were carried out at room temperature using 50 mg of material and 50 mL of dye solution at a concentration of 50 mg L⁻¹. Initially, the sorbent material was loaded with the dye following the general sorption procedure described above. The recovered material was washed with distilled water, air dried, and then suspended in 50 mL of aqueous NaOH solution (0.1 M) for desorption of the dye. The obtained suspensions were stirred for 15 min, then centrifuged and the regenerated material was thoroughly washed with distilled water and subsequently suspended in dye solutions under the same conditions as above. The sorption–desorption cycles were repeated three times.

The high degree of polymerization is evidenced by common features present in the FTIR spectrum. As shown in Figure 3(b), the spectrum of the QTEAS hyperbranched polymer features a strong carbonyl–carboxylic absorption band at 1,745 cm^–1, stemming from the large number of carbonyl bonds of both esters and carboxylic acid termini groups. The two absorption bands appearing at 2,932 and 2,847 cm⁻¹ are mainly due to the stretching vibration of sp³ carbons of alkyl groups. In addition, the band centered at 3,444 cm^–1 is characteristic of the OH stretching band of both the carboxylic acid and alcohol termini groups. The appearance of the absorption bands at 1,160 cm^–1 and 1,094 cm^–1, assigned for C–O antisymmetric stretching and C–O–C bond stretching, respectively, is an indication of an efficient esterification. Moreover, Figure 3(b) shows two distinguished absorption bands, one appearing at 1,462 cm^–1 assigned to the sp³ C–H of methylene substituent of quaternary amine groups and a shoulder around 1,523 cm^–1, belonging to the C–N⁺ groups (Colthup et al. 1990). Finally, the band centered at 1,642 cm^–1 in the spectrum could be assigned to a symmetric deformation of quaternary ammonium, which proves the successful quaternization of tertiary amine groups of the hyperbranched polymer (Jin et al. 2010).

The thermal stability of the synthesized QTEAS was studied by TGA (Figure 4(b)). The first derivative of the TG curve is proportional to the rate of decomposition, and represents the temperature corresponding to the maximum rate of weight loss (T_max). As shown in Figure 4(b), the mass loss occurs in multiple steps within the temperature range of 78–426 °C. The initial decomposition temperature (T_Di) is approximately 158 °C. Upon initial heating, the gradual loss of mass (1.71%) proceeded with a T_D of 78 °C, which is probably due to the loss of physically adsorbed water and/or residual solvent entrapped in the polymer network from the reaction workup. This was followed by two major mass losses at 178.3 °C and 310.5 °C corresponding to the degradation of the alcohol/carboxylic acids termini, which are subject to rapid thermal degradation and ester bond breaking in the branches of QTEAS, respectively. Further heating above 300 °C resulted in extensive chain scission and QTEAS decomposition, with a mass loss of about 73% at 455 °C. The TGA data indicate that 13% of the total weight of the sample remains after heating to the highest temperature (800 °C).

In order to examine the controlling mechanism of the sorption process, pseudo-first order model and pseudo-second order kinetic models were used to analyze the experimental data.

Figure 5(b) and Figure 5(b) inset show the linear fit plots of the pseudo-first order and pseudo-second order models, respectively. The equilibrium sorption capacity (q_e), the rate constants (k₁, k₂), and the coefficient (R²) values were calculated from the linear plots. The parameters obtained for the two models are presented in Table 2. Perfect correlation is however observed between experimental data and the pseudo-second order kinetic model with excellent correlation coefficients. Additionally, the calculated q_e values from the model were totally in agreement with experimental sorption capacities, emphasizing therefore the efficiency of the model. In the literature, many studies have shown that adsorption kinetics of numerous dyes onto various polymer-based adsorbents are well fitted by the pseudo-second order model. For example, Renault et al. (2008) have found that the pseudo-second order was the best model for describing the adsorption kinetics of AB 25 dye on cross-linked starch ion-exchanger.

Although the surface of the QTEAS adsorbent is tidy, with positively charged quaternary ammonium groups which make it a powerful reactant toward anionic dyes, the presence of alcohol/carboxylic acid termini groups within the polymer could however participate in the removal process via covalent, coulombic, hydrogen bonding or weak van der Waals forces with the dyes' functional groups (−OH, −NH₂, −SO₃Na, −N = N−). To shed light on the adsorption mechanism of the anionic dyes on the QTEAS material, a selective adsorption experiment was carried out using a mixture of cationic and anionic dyes in aqueous solutions. Figure 6(b) (photo in the left corner) shows that preliminary tests revealed that QTEAS was not able to remove the cationic dye Rhodamine B (RhB) at a concentration of 50 mg L⁻¹ from aqueous solutions after 16 h of contact time. Figure 6(b) shows also the UV-vis spectra of aqueous solutions of RhB/IC (3/2) (cationic dye RhB (100 mg L⁻¹); anionic dye IC (100 mg L⁻¹)) mixture before and after the adsorption process. As shown in Figure 6(b), after 16 h of contact time, IC was almost completely removed from the mixture solution by the QTEAS adsorbent, whereas RhB remained intact in the solution. These results suggest that the removal process of anionic dyes by QTEAS could be an ion-exchange mechanism and also confirm that the QTEAS adsorbent has a selective adsorption property for anionic dyes and could have potential for application in purification of cationic dyes from anionic dyes.

The experimental data were analyzed by using Langmuir and Freundlich isotherm models.

As shown in Figure 7(b), the analysis of the equilibrium experimental data of IC and EB sorption onto QTEAS by utilizing the Freundlich and Langmuir isotherm equations gave linear plots. The linearized forms of both Freundlich and Langmuir isotherms were found to be linear over a broad concentration range with good to excellent correlation coefficient values (see Table 3). These results clearly indicate that both models adequately fitted the experimental data of the dyes' removal by QTEAS material.

As shown in Table 3, both dyes are removed with the same order of magnitude by the hyperbranched polymer. The greater affinity of QTEAS sorbent for both dyes IC and EB is confirmed by the sorption capacities at saturation of the monolayer. Despite its extended molecular structure, EB is sorbed onto QTEAS with quite similar effectiveness (0.891 meq g^–1 for EB vs. 0.914 meq g^–1 for IC). These results suggest that EB has greater access to the surface of QTEAS without inducing a quick saturation of sorption sites because of steric hindrance and clearly show that the dyes' removal is closely linked to the total charge on the surface of the dyes' molecules. Nonetheless, the highest b value found for IC dye (0.42 vs. 0.32 for EB) indicates that QTEAS shows a little preference on binding divalent anion IC than the tetravalent anion EB. Finally, these results also confirm the presence of readily accessible sorption sites and strongly suggest the homogeneous distribution of active quaternary ammonium groups on the surface of the hyperbranched polymer. The QTEAS hyperbranched polymer showed better sorption capacities for IC and EB dyes compared to those of other adsorbents reported in the literature for the removal of these dyes (see Table 4).

The removal of IC and EB by the QTEAS sorbent was studied at three temperatures to determine the thermodynamic parameters, and the results are summarized in Table 5. The dyes' uptake decreased with an increase in temperature, indicating an exothermic process.

The vant'Hoff plot of log K_L versus 1/T gave straight lines. The calculated slope and intercept from the plot were used to determine ΔH and ΔS, respectively (Table 5). The negative value of ΔG (−31.75 to −26.11 kJ·mol⁻¹) at each temperature implies a favorable and spontaneous adsorption process and confirms the affinity of the QTEAS material for IC and EB dyes. In general, the change in free energy for physisorption is between −20 and 0 kJ·mol⁻¹, whereas chemisorption is in the range of −80 to 400 kJ·mol⁻¹ (Renault et al. 2008; Abdel Salam et al. 2012; Eskandarian et al. 2014). These values are in the intervals between physisorption and chemisorption, thus suggesting that the process is a physical adsorption enhanced by a chemical effect (Renault et al. 2008). The negative value of ΔH indicates that the adsorption is exothermic and also suggests that the sorption process is a physical adsorption enhanced by chemical interactions between the anionic dyes and the quaternary amine groups through ion-exchange.

In the present study, the sorption of two acid dyes onto QTEAS hyperbranched polymer was investigated. The results demonstrated that the hyperbranched polymer performs efficiently in a wide pH range of dye solutions. From the kinetic studies, it was found that the sorption process followed the pseudo-second order model. The sorption isotherms were adequately fitted by the Langmuir isotherm model and the removal process was found to take place by the electrostatic attraction between the positively charged hyperbranched polymer and the negatively charged dyes. Nevertheless, the ionic strength does not affect the dyes' removal efficiency by the hyperbranched polymer. Moreover, the removal capacity is closely connected to the total sulfonate groups on the surface of the dyes. The calculated thermodynamic parameters indicated the exothermic and spontaneous nature of the removal process. Finally, the third repeated use of QTEAS material for the dyes' removal showed no decrease of its capacity for the dyes' sorption. The overall results illustrated that QTEAS can be effectively used as an ion-exchanger for the removal of acid dyes from colored wastewaters.