Fabrication of Eco-Friendly Polyelectrolyte Membranes Based on Sulfonate Grafted Sodium Alginate for Drug Delivery, Toxic Metal Ion Removal and Fuel Cell Applications

Polyelectrolyte membranes (PEMs) are a novel type of material that is in high demand in health, energy and environmental sectors. If environmentally benign materials are created with biodegradable ones, PEMs can evolve into practical technology. In this work, we have fabricated environmentally safe and economic PEMs based on sulfonate grafted sodium alginate (SA) and poly(vinyl alcohol) (PVA). In the first step, 2-acrylamido-2-methyl-1-propanesulphonic acid (AMPS) and sodium 4-vinylbenzene sulfonate (SVBS) are grafted on to SA by utilizing the simple free radical polymerization technique. Graft copolymers (SA-g-AMPS and SA-g-SVBS) were characterized by 1H NMR, FTIR, XRD and DSC. In the second step, sulfonated SA was successfully blended with PVA to fabricate PEMs for the in vitro controlled release of 5-fluorouracil (anti-cancer drug) at pH 1.2 and 7.4 and to remove copper (II) ions from aqueous media. Moreover, phosphomolybdic acids (PMAs) incorporated with composite PEMs were developed to evaluate fuel cell characteristics, i.e., ion exchange capacity, oxidative stability, proton conductivity and methanol permeability. Fabricated PEMs are characterized by the FTIR, ATR-FTIR, XRD, SEM and EDAX. PMA was incorporated. PEMs demonstrated maximum encapsulation efficiency of 5FU, i.e., 78 ± 2.3%, and released the drug maximum in pH 7.4 buffer. The maximum Cu(II) removal was observed at 188.91 and 181.22 mg.g–1. PMA incorporated with PEMs exhibited significant proton conductivity (59.23 and 45.66 mS/cm) and low methanol permeability (2.19 and 2.04 × 10−6 cm2/s).


Fabrication of Polyelectrolyte Membranes (PEMs)
All membranes are made from blends of sulfonated SA (SA-g-AMPS/SA-g-SVBS) and PVA. Two grams of SA-g-AMPS/SA-g-SVBS was dissolved in 40 mL of distilled water in a separate beaker. Fixed concentrations of PVA (3 g) aqueous solution were made at 80 • C, and the resulting solution was cooled to room temperature after full dissolution of PVA. The two polymer solutions were mixed and stirred for 12 h. The resulting blend solution was allowed in rest mode for 2 h in order to eliminate any bubbles formed; finally, this solution was then casted on a clean glass plate so that the solvent can evaporate. The dried membranes were crosslinked with an acetone:water (50:50) mixture bath comprising 2.5 mL GA as the cross-linker and 2.5 mL HCl as the catalyst. These membranes are known as pristine PEMs, namely PSAAM and PSASB. Additionally, composite PEMs are made by adding 10% PMA to polymer mix solutions prior to casting, and these membranes are known as PSAAM-PMA and PSASB-PMA (Scheme 1).

Swelling Studies
Equilibrium water uptake capacities of PEMs were performed in DD water at 30 • C for 12 h. Dried PEM (W d ) of 0.10 g was immersed into 50 mL of DD water and removed after 12 h, and the surface of PEM wiped with tissue paper and weighed (W e ). The percentage of equilibrium swelling ratio (%S e ) was calculated by the following equation.

5-Fluorouracil (5FU) Encapsulation and Release Studies
5FU entrapment was performed using the equilibrium swelling method [13] by allowing PEMs to swell 12 h in alkaline 5FU solution, and then the 5FU entrapped PEMs were dried. These dried membranes were crushed in agate mortar by using 5 mL of DD water to extract trapped 5FUs. Then, 5FU was estimated by using a UV-vis spectrophotometer

Swelling Studies
Equilibrium water uptake capacities of PEMs were performed in DD water at 30 °C for 12 h. Dried PEM (Wd) of 0.10 g was immersed into 50 mL of DD water and removed after 12 h, and the surface of PEM wiped with tissue paper and weighed (We). The per centage of equilibrium swelling ratio (%Se) was calculated by the following equation.

5-Fluorouracil (5FU) Encapsulation and Release Studies
5FU entrapment was performed using the equilibrium swelling method [13] by al lowing PEMs to swell 12 h in alkaline 5FU solution, and then the 5FU entrapped PEM were dried. These dried membranes were crushed in agate mortar by using 5 mL of DD water to extract trapped 5FUs. Then, 5FU was estimated by using a UV-vis spectropho tometer (LABINDIA, UV-3092) at λmax 270 nm. The 5FU encapsulation efficiency (%5FU EE) was calculated by using the equations below. The release of 5FU from drug trapped PEMs was studied by using a dissolution tester apparatus (LABINDIA, DS-8000) at 37 • C in pH 1.2 and 7.4 buffer media with 100 rpm. Drug release was analyzed at appropriate intervals and estimated with UV-Vis spectrophotometer. The resulting data were fitted with drug release kinetics models [46], i.e., Korsmeyer-Peppas, zero order, first order, Higuchi and Hixson-Crowell models.
where C o and C e are the initial and residual concentrations of Cu(II) ion in liquid phase, respectively. V is the volume of the Cu(II) ion solution (L), and M is the amount of PEM (g).

Ion Exchange Capacity, Oxidative Stability, Proton Conductivity and Methanol Permeability Studies
The ion exchange capacity (IEC) of PEMs was determined by simple acid-base titration method. Briefly, the known amount of the PEM was soaked in 50 mL of aqueous NaCl solution (3 M) for 24 h. Subsequently, 10 mL of solution was titrated against the NaOH solution (0.01 N). IEC was calculated by using the following expression.
PEMs were analyzed for oxidative stability in 20 mL of Fenton's reagent at 30 • C for 6 h. The percentages of weight loss of PEMs were calculated by measuring before and after the reagent treatment.
The proton conductivity of PEMs was determined by the four-probe impedance technique using the electrochemical impedance analyzer (Biologic SP-200) [47]. Fully hydrated PEMs were sandwiched into Teflon blocks, which are equipped with Pt probes. Impedance was measured at 30 • C with 98% humidity. The conductivity of PEMs was calculated from PEM resistance (R) by using the following equation: where σ is the conductivity of PEM in S/cm, L is the thickness of the PEM and A is the cross sessional area of the PEM in cm 2 . Methanol permeability of PEMs was performed [48] by solution-diffusion mechanisms using a Teflon diaphragm diffusion cell, where two reservoirs (approximately 50 mL each) are separated by vertical membrane (effective area 3.14 cm). Permeate methanol concentration was measured by a refractometer AR4 (A. Kruss Optronic, Germany) and calculated using the below equation: where C a , C b and V b are concentrations of methanol in the receptor compartment, concentration of methanol in the donor compartment and receptor volume, respectively. A is the effective area of the membrane, and T is the time consumed when equilibrating the system.

Characterization
1 H Nuclear magnetic resonance spectroscopy (Brucker A vance 500 MHz) was used to analyze the chemical structure of sulfonate graft SA copolymers (SA-g-AMPS and SA-g-SVBS). Fourier transform infrared studies, FTIR (Perkin Elmer, Spectrum Two) and Polymers 2021, 13, 3293 6 of 17 attenuated total reflection-FTIR analysis (Bruker, Alpha-II, Eco-ATR) were used to characterize the graft copolymers and PEMs' chemical structure, encapsulation of 5FU in PEMs and sorption of Cu(II) ion by the PEMs. Scanning electron microscopy and energy dispersive X-ray study (JOEL, JSM IT500) were used to characterize the surface morphology of PEMs and elemental composition of the surface of the PEMs, respectively. Differential scanning calorimetry study of graft copolymer was performed with TA instruments (STA, Q600). X-ray diffraction studies (Rigaku, Miniflex 600) of PEMs and PMA incorporated with PEMs were characterized to analyze PMA dispersity in the polymer network.

Synthesis of Sulfonate Grafted Sodium Alginates
Sulfonate grafted sodium alginates were synthesized from sodium alginate (SA) with 2-acrylamide-2-methyl-1-propanesulphonic acid (AMPS) and sodium 4-vinylbenzene sulfonate (SVBS) by simple free radical polymerization using ceric ammonium nitrate (CAN) and ammonium persulphate (APS), respectively (Scheme S1). SA-g-AMPS and SA-g-SVBS were characterized by 1 H NMR, FTIR, XRD and DSC studies (Figures S1-S4). The plausible chemical structure of SA-g-AMPS and SA-g-SVBS is shown in Scheme S2. Graft copolymeric reaction parameters are optimized with respect to the concentration of monomer, initiator, temperature and time; the results are presented in (Tables S1 and S2; Figures S5 and S6). The maximum grating and grafting efficiency of SA-g-AMPS are 62.06% and 42.78%, respectively, at optimized reaction conditions, i.e., reaction temperature 60 • C, AMPS concentration 3.5 mmol, CAN concentration 0.11 mmol and reaction time 120 min. The maximum grafting and grafting efficiency of SA-g-AMPS are 84.38% and 65.44%, respectively, at optimized reaction conditions, i.e., reaction temperature 60 • C, SVBS concentration 3.5 mmol, APS concentration 0.036 mmol and reaction time 120 min.  [41,43]. From Figure 1E, the significant vibrational peaks observed at 1665, 3140, 3067 and 1262/861 cm −1 functional groups of 5FU are C=O, N-H, =C-H and C-F. However, in the case of 5FU encapsulated PEMs, they exhibited peaks at 1230 cm −1 for C-F, while the other functional groups of 5FU are overlapped with the PEM network. Compared to the spectra of pristine PEMs before and after adsorption of the Cu(II) ion, the adsorption peaks at 3465,1715, 1655, 1475 and 1061/1079/1040 cm −1 corresponding to -OH stretching, C=O stretching, C-N stretching, C=C stretching and S=O stretching either decreased in intensity or shifted the peaks from their respective adsorption. Hence, the presence of nitrogen, oxygen and sulphur atoms in the amide, vibrational peaks observed at 1665, 3140, 3067 and 1262/861 cm −1 functional groups of 5FU are C=O, N-H, =C-H and C-F. However, in the case of 5FU encapsulated PEMs, they exhibited peaks at 1230 cm −1 for C-F, while the other functional groups of 5FU are overlapped with the PEM network. Compared to the spectra of pristine PEMs before and after adsorption of the Cu(II) ion, the adsorption peaks at 3465,1715, 1655, 1475 and 1061/1079/1040 cm −1 corresponding to -OH stretching, C=O stretching, C-N stretching, C=C stretching and S=O stretching either decreased in intensity or shifted the peaks from their respective adsorption. Hence, the presence of nitrogen, oxygen and sulphur atoms in the amide, carboxyl/hydroxyl and sulfone groups are responsible for the sorption of Cu(II) ions attachment [30,31].   (321), respectively [50]. However, PMA incorporated PEMs are not shown such crysttaline peaks, which indicates that PMA is molecularly dispersed in PEMs.

SEM and EDAX Studies
SEM images of pristine PEMs, PMA incorporated PEMs, 5FU encapsulated PEMs and Cu(II) sorbed PEMs are presented in Figure 3. PSAAM and PSASB membranes displayed smooth surfaces with non porous structure. However, PMA incorporated PSAAM and PSASB membranes showed rough surfaces and homogeneous distribution. In general, aqueous Cu(II) ions are sorbed by PEMs due to the interactions between the

SEM and EDAX Studies
SEM images of pristine PEMs, PMA incorporated PEMs, 5FU encapsulated PEMs and Cu(II) sorbed PEMs are presented in Figure 3. PSAAM and PSASB membranes displayed smooth surfaces with non porous structure. However, PMA incorporated PSAAM and PSASB membranes showed rough surfaces and homogeneous distribution. In general, aqueous Cu(II) ions are sorbed by PEMs due to the interactions between the functional groups (-SO 3 − , -CONH, -OH and -COO − ) and sorbed Cu(II) ions. Figure 3E,F show the images of 5FU sorbed PEMs; the surface is very rough, and this may be because of the recrystalisation of encapsulated 5FU during the drying process of PEM. Figure 3G,H shows the images of Cu(II) ion sorbed PEMs, and many well-dispersed Cu(II) ion salts formed on the sufrace of the PEM, which resulted in the rougher surface. Polymers 2021, 13, x FOR PEER REVIEW 10 of 18

Equilibrium Swelling Studies
Swelling is one of the important characteristics of polymer membrane, which helps us understand the water holding capacity. It depends on the nature of functional groups and the type of dopant present in the matrix. Therefore, swelling studies were performed for both pristine PEMs and PMA incorporated membranes. The %Equilibrium swelling ratios (%Se) are presented in Table 1, i.e., 369, 258, 300 and 169 for PSAAM, PSASB, PSAAM-PMA and PSASB-PMA, respectively. The results indicate that %Se decreased with the incorporation of PMA. This may be due to the formation physico-chemical interaction between the polymer and inorganic polyacid [51].

Equilibrium Swelling Studies
Swelling is one of the important characteristics of polymer membrane, which helps us understand the water holding capacity. It depends on the nature of functional groups and the type of dopant present in the matrix. Therefore, swelling studies were performed for both pristine PEMs and PMA incorporated membranes. The %Equilibrium swelling ratios (%S e ) are presented in Table 1, i.e., 369, 258, 300 and 169 for PSAAM, PSASB, PSAAM-PMA and PSASB-PMA, respectively. The results indicate that %S e decreased with the incorporation of PMA. This may be due to the formation physico-chemical interaction between the polymer and inorganic polyacid [51].

5-Fluorouracil Drug Delivery
5FU is a potential chemotherapeutic agent and is extensively used for various types of cancers. In the present study, 5FU is physically encapsulated by the equilibrium swelling method; hence, it is essential to understand the release kinetics of 5FU. %Encapsulation efficiencies for PSAAM and PSASB are 78 ± 2.3 and 66 ± 4.7, respectively. The 5FU release profiles of PEMs at pH 1.2 and 7.4 at 37 • C are presented in Figure 5. Table S3 shows the various drug release models (zero order, first order, Higuchi, Hixson-Crowell and Korsmeyer-Peppas) fitted with the 5FU release kinetics data. The best R 2 observed for all the kinetic models, however, is maximum Hixson-Crowell and Korsmeyer-Peppas models at pH 1.2 with slope value (n) ranges from 0.309 to 0.94, i.e., 0.309 < n < 0.94. These results demonstrate that drug release followed the non-Fickian drug transport mechanism, i.e., the dissolution of water-soluble drug in PEM network [46]. Similar mechanisms are also observed in the case of pH 7.4.

5-Fluorouracil Drug Delivery
5FU is a potential chemotherapeutic agent and is extensively used for various types of cancers. In the present study, 5FU is physically encapsulated by the equilibrium swelling method; hence, it is essential to understand the release kinetics of 5FU. %Encapsulation efficiencies for PSAAM and PSASB are 78 ± 2.3 and 66 ± 4.7, respectively. The 5FU release profiles of PEMs at pH 1.2 and 7.4 at 37 °C are presented in Figure 5. Table S3 shows the various drug release models (zero order, first order, Higuchi, Hixson-Crowell and Korsmeyer-Peppas) fitted with the 5FU release kinetics data. The best R 2 observed for all the kinetic models, however, is maximum Hixson-Crowell and Korsmeyer-Peppas models at pH 1.2 with slope value (n) ranges from 0.309 to 0.94, i.e., 0.309 < n < 0.94. These results demonstrate that drug release followed the non-Fickian drug transport mechanism, i.e., the dissolution of water-soluble drug in PEM network [46]. Similar mechanisms are also observed in the case of pH 7.4.  Figure 6 shows the effect of Cu(II) ion concentration on PSAAM and PSASB PEMs in pH = 5.5 aqueous meadia at 30 °C. The presence of functional groups, such as croboxyl group of SA, amide and sulfone group of AMPS/SVBS and hydroxyl group of PVA, is a key factor in either chemisorption or physisorption of metal ions. In the present study, the maximum adsorption capacities were observed in pH 5.5 at 30 °C, and 188.91 and 181.22 mg.g -1 were achieved for PSAAM and PSASB, respectively. The present results compared with the literature with respect to adsorbant to adsorbate (mg.g −1 ) are shown in Table 2. Sudha Vani et al. synthesized novel sodium alginate-gelatin blend membranes and studied their toxic metal ion (Cu(II) and Ni(II)) removal capacities [4]. The Qe values are  Figure 6 shows the effect of Cu(II) ion concentration on PSAAM and PSASB PEMs in pH = 5.5 aqueous meadia at 30 • C. The presence of functional groups, such as croboxyl group of SA, amide and sulfone group of AMPS/SVBS and hydroxyl group of PVA, is a key factor in either chemisorption or physisorption of metal ions. In the present study, the maximum adsorption capacities were observed in pH 5.5 at 30 • C, and 188.91 and 181.22 mg.g −1 were achieved for PSAAM and PSASB, respectively. The present results compared with the literature with respect to adsorbant to adsorbate (mg.g −1 ) are shown in Table 2. Sudha Vani et al. synthesized novel sodium alginate-gelatin blend membranes and studied their toxic metal ion (Cu(II) and Ni(II)) removal capacities [4]. The Q e values are reported to be 43.51 (Cu(II)) and 240.64 (Ni(II)). Moreover, the development of membranes from polyaniline and poly(acrylic acid)-g-sodium alginate/gelatin was also reported [31]. The Q e values of this study were reported as 53.29 for Cu(II) and 129.28 for Ni(II). Zhao et al. developed sodium alginate-polyethylene glycol oxide-nano materials based composite gels for the removal of heavy metal ions such as Cu(II) and Cd 2+ from polluted water [52]. In this study, the fabricated materials showed the adsorption capacity (Q m ) for Cu(II) and Cd(II) as 6.78 and 3.43 with the removal rates 96.8% (Cu(II)) and 78% (Cd(II)). Li et al. synthesized calcium alginate immobilized kaolin using the sol-gel process, and it was used for the removal of Cu(II) from its aqueous solutions [53]. The Q e value related to the removal of Cu(II) using alginate immobilized kaolin was reported to be 53.63. Yang et al.

Ion Exchange Capacity, Oxidative Stability Proton Conductivity and Methanol Permeability Studies
In addition to drug delivery and sorption studies of metal ions, fabricated membranes with and without PMA are examined for the suitability to function as PEMs for fuel cells. In this context, %S e , ion exchange capacity (IEC), proton conductivity study and methanol permeability of the PEM were tested (Table 1). Figure 7 shows the proton conductivity study and methanol permeability of PSAAM, PSASB, PSAAM-PMA and PSASB-PMA at 30 • C under atmospheric pressure. IEC and %S e are the key factors influencing the proton conductivity study and methanol permeability. reported to be 43.51 (Cu(II)) and 240.64 (Ni(II)). Moreover, the development of membranes from polyaniline and poly(acrylic acid)-g-sodium alginate/gelatin was also reported [31].
The Qe values of this study were reported as 53.29 for Cu(II) and 129.28 for Ni(II). Zhao et al. developed sodium alginate-polyethylene glycol oxide-nano materials based composite gels for the removal of heavy metal ions such as Cu(II) and Cd 2+ from polluted water [52].
In this study, the fabricated materials showed the adsorption capacity (Qm) for Cu(II) and Cd(II) as 6.78 and 3.43 with the removal rates 96.8% (Cu(II)) and 78% (Cd(II)). Li et al. synthesized calcium alginate immobilized kaolin using the sol-gel process, and it was used for the removal of Cu(II) from its aqueous solutions [53].     In addition to drug delivery and sorption studies of metal ions, fabricated membranes with and without PMA are examined for the suitability to function as PEMs for fuel cells. In this context, %Se, ion exchange capacity (IEC), proton conductivity study and methanol permeability of the PEM were tested (Table 1). Figure 7 shows the proton conductivity study and methanol permeability of PSAAM, PSASB, PSAAM-PMA and PSASB-PMA at 30 °C under atmospheric pressure. IEC and %Se are the key factors influencing the proton conductivity study and methanol permeability.  In particular, PEMs are chemically treated with glutaraldehyde and results in the formation of acetal linkage and hydrophobic crosslinks, which protect polymer from dissolving in water and provides stable morphology. The oxidative stability of PEMs was evaluated in Fenton's reagent at 30 • C, and the results indicate that PMA incorporated PEMs shown significantly high stabilities owing to the dense structure of composite PEMs, which controls the penetration of radicals with polymer chains. This behavior also supports the good mechanical property of the PEMs.
Proton conductivity of PEMs are in the order of 10 −3 S/cm; however, it significantly increased with incorporation of 10% PMA into the PEMs. This may be due to proton transfer (i.e., in the form of H 3 O + , H 5 O 2 + and H 9 O 4 + ) across the membrane with the help of the functional groups -COO − , -CONH and -SO 3 H., where hydronium ion dissociates and forms hydrogen bonds (Grotthus Mechanism) [57]. This is also supported by high %S e (over the 100%), i.e., the more hydrophilic nature of PEMs enhances the ionic nature of PEMs [58]. On the contrary, PMA incorporated PEMs (2.19 × 10 −6 cm 2 /s and 2.04 × 10 −6 cm 2 /s) demonstrated lower methanol permeability when compared to pristine PEMs (1.36 × 10 −6 cm 2 /s and 1.61 × 10 −6 cm 2 /s). This is because of its polyacid pseudo liquid phase nature [59,60].

Conclusions
In summary, sulfonate functionalized sodium alginate-based PEMs were fabricated by the simple solution casting method. Composite PEMs were also fabricated by the incorporation of PMA in the polymer matrix. The fabricated PEMs were used for drug delivery applications of a chemotherapeutic agent (5FU) and adsorption of Cu(II) ions from the aqueous solutions. The drug release kinetics demonstrate that drug release followed the non-Fickian drug transport mechanism. From the adsorption studies, it was found that PEMs possess significant adsorption capacities (188.91 and 181.22 mg.g −1 for PSAAM and PSASB, respectively). Fabricated PEMs are well suited for fuel cells as their proton conductive membrane functions as an effective proton transporter with respect to their dual functionalities, i.e., -SO 3 − and -COO − . In addition, polyacid (PMA) incorporated PEMs accelerated more proton conductivity. PEMs showed relatively good water management, significant IEC, considerable proton conductivity and low methanol permeability. Therefore, PEMs are eco-friendly, low-cast and promising practical usage materials for delivery, adsorption and transport of molecules or ions for controlled release, precious metal extraction or organic dye removal and fuel cell applications.