In Situ Attachment of Acrylamido Sulfonic Acid-Based Monomer in Terpolymer Hydrogel Optimized by Response Surface Methodology for Individual and/or Simultaneous Removal(s) of M(III) and Cationic Dyes.

Herein, grafting of starch (STR) and in situ strategic inclusion of 2-(3-(acrylamido)propylamido)-2-methylpropane sulfonic acid (APMPS) via solution polymerization of 2-(acrylamido)-2-methylpropanesulfonic acid (AMPS) and acrylamide (AM) have resulted in the synthesis of smart STR-grafted-AMPS-co-APMPS-co-AM (i.e., STR-g-TerPol) interpenetrating terpolymer (TerPol) network hydrogels. For fabricating the optimum hydrogel showing excellent physicochemical properties and recyclability, amounts of ingredients and temperature of synthesis have been optimized using multistage response surface methodology. STR-g-TerPol bearing the maximum swelling ability, along with the retention of network integrity, has been employed for individual and/or simultaneous removal(s) of metal ions (i.e., M(III)), such as Bi(III) and Sb(III), and dyes, such as tris(4-(dimethylamino)phenyl)methylium chloride (i.e., crystal violet) and (7-amino-8-phenoxazin-3-ylidene)-diethylazanium dichlorozinc dichloride (i.e., brilliant cresyl blue). The in situ strategic protrusion of APMPS, grafting of STR into the TerPol matrix, variation of crystallinity, thermal stabilities, surface properties, mechanical properties, swellability, adsorption capacities (ACs), and ligand-selective superadsorption have been inferred via analyses of unadsorbed and/or adsorbed STR-g-TerPol using Fourier transform infrared (FTIR), 1H/13C NMR, UV-vis, thermogravimetric analysis, differential scanning calorimetry, X-ray diffraction, field emission scanning electron microscopy, energy-dispersive X-ray, dynamic light scattering, and rheological analyses and measuring the lower critical solution temperature, % gel content, pH at point of zero charge (pHPZC), and network parameters, such as ρc and M c. The prevalence of covalent, ionic (I), and variegated interactions between STR-g-TerPol and M(III) has been understood through FTIR analyses, fitting of kinetics data to the pseudosecond-order model, and by the measurement of activation energies of adsorption. The formation of H-aggregate type dimers and hypochromic and hypsochromic shifts has been explained via UV-vis analyses during individual and/or simultaneous removal(s) of cationic dyes. Several isotherm models were fitted to the equilibrium experimental data, of which Langmuir and combined Langmuir-Freundlich models have been best fitted for individual Bi(III)/Sb(III) and simultaneous Sb(III) + Bi(III) removals, respectively. Thermodynamically spontaneous chemisorption processes have shown the maximum ACs of 1047.39/282.39 and 932.08/137.85 mg g-1 for Bi(III) and Sb(III), respectively, at 303 K, adsorbent dose = 0.01 g, and initial concentration of M(III) = 1000/30 ppm. The maximum ACs have been changed to 173.09 and 136.02 mg g-1 for Bi(III) and Sb(III), respectively, for binary Sb(III) + Bi(III) removals at 303 K, adsorbent dose = 0.01 g, and initial concentration of Bi(III)/Sb(III) at 30/5 and 5/30 ppm.

Introduction

Hydrogels are hydrophilic polymeric networks formed via chemical or physical bonds that have the unique ability to withstand large deformations.[1] Alongside, high swellability, elasticity, flexibility, and permeability allow rapid sequestration of heavy metals via coordination of metal ions with various functional groups of hydrogels. Accordingly, hydrogels have been drawing attractive attention because of the various promising applications, such as adsorptive removals of dyes and/or metal ions,[2,3] tissue engineering,[4] drug delivery,[5] wound healing materials, biosensors,[6] and manufacturing contact lenses.[7] The conventional hydrogels composed of only natural polymer lack mechanical strength, whereas synthetic copolymer-based hydrogels show poor reusability and biodegradability, which limits extensive practical applications.[8] Thus, the optimum amount of natural polymer-to-synthetic monomer(s) is required to fabricate a heterogeneous hydrogel of desired physicochemical properties suitable for swelling and removals of contaminants.

Now-a-days, a growing global environmental concern is the water contamination resulted from inappropriate discharge of heavy metal(s)-laden wastewater from manufacturing industries into aquatic systems, leading to portable water scarcity, malnutrition, sickness, mass mortality of aquatic and terrestrial life, and even extinction of species. In fact, most of the heavy metals are immensely threatening to the living population because of their carcinogenic, nondegradable, persistent, and accumulative nature. Antimony (Sb), a toxic heavy metal enlisted as the priority pollutant, is usually generated from the effluents of diversified industries, such as hardening of alloys for lead–acid batteries, battery grids, small arms bullets, ammunition, flame retardants, clarification of glass products, mining, and cable sheathing. The major threat of Sb and Sb-based compounds to human body includes easy combination with sulfhydryl, which affects enzyme activity and destroys ion balance in cells, leading to cellular hypoxia. Notably, among various Sb-based components, Sb(III) is the most hazardous and 10 times more toxic than Sb(V). Moreover, bismuth (Bi) compounds are extensively used for the treatment of syphilis, gastritis, and ulcer, leading to increased exposure of living beings to Bi. Alongside, Bi-compounds pose numerous toxic effects, such as nephropathy, osteoarthropathy, hepatitis, and neuropathology. Thus, removal of Sb and Bi from wastewater is an essential task for reducing their harmful effects.

In recent years, more than 10 000 dissimilar toxic synthetic dyes/pigments in the effluents of textile, leather, paper, plastics, printing, electroplating, food, and cosmetics industries impart severe detrimental effect to the ecosystem and environment because of their carcinogenicity, genotoxicity, teratogenicity, and mutagenicity. Crystal violet (CV) causes tissue necrosis, cyanosis, jaundice, shock, and quadriplegia in humans.[9] Alongside, brilliant cresyl blue (BCB) is mainly used as a supravital stain for counting reticulocytes. However, the presence of these hazardous dyes, even in low concentration, can pose serious threats to human health. Thus, removal of dyes from industrial waste effluents is crucial for the betterment of the environment/ecosystem.

Accordingly, several cheap and eco-friendly techniques, such as precipitation,[10] ion exchange,[11] adsorption,[1−3] membrane filtration,[12,13] electrodeposition,[14] complexation,[15] and reverse osmosis,[16] have been employed for water purification and metal recovery operations from wastewater. Of these, adsorption is the universally accepted, promising, and widely implemented method for water treatment because of the ease of operation and maintenance, cost-effectiveness, high efficiency, flexibility, rapidness, simplicity of design, and availability of diversified adsorbents.

Starch (STR) is a natural, biodegradable, renewable, and abundantly available polysaccharide-based biopolymer obtained primarily from plants and is the major dietary source of carbohydrates. STR granules are made of mainly two kinds of α-glucan, that is, amylose and amylopectin, which are about 98–99% of the total weight. STR-based hydrogels modified by derivatization, grafting, network formation, and other polymer analogous reactions have drawn attractive attention in biomedical and pharmaceutical fields because of the natural abundance, low cost, biocompatibility/biodegradability, and ease of in vivo application(s).

Response surface methodology (RSM) is an empirical statistical technique adopted for analyzing the simultaneous relationship between a set of experimental factors and measuring effects of such factors on responses through the minimum number of runs. As compared to one-factor-at-a-time design, the experimental design and RSM can effectively reduce runs and facilitate the execution of runs required for construction of the response surface. However, RSM has been used for the evaluation of relative significance of several parameters in the presence of complex interactions for attaining the maximum adsorption capacity (AC).[17,18] In the present study, RSM has been employed for systematic multistage optimization of variables for the synthesis of optimum hydrogel possessing the maximum equilibrium swelling ratio (ESR).

Meanwhile, several homo-/co-/terpolymers and interpenetrating polymer network (IPN)-based polymers containing carboxylic-/sulfonic acid, amine, and imino acetate groups and natural polymer-grafted copolymers[19−21] have been used for the adsorptive removals of heavy metal ions [M(II/III/VI)] from aqueous solutions. In the last few years, sodium salt of 2-(acrylamido)-2-methylpropanesulfonic acid (AMPS) has received attractive attention as an ionic comonomer because of the prevalent ionizable sulfonate group, which completely dissociates in the entire pH range, and thus, the hydrogels derived from AMPS exhibit pH-independent swelling and superior ion-conducting ability. Several natural polymer [STR/pectin/guar gum (GG)/gumghatti (GGTI)]-grafted hydrogels, such as STR-g-poly(acrylic acid (AA)),[22] STR-g-poly(acrylamide (AM)),[23,24] STR-g-poly(N-isopropylacrylamide (NIPA)),[25] STR-g-poly(methyl methacrylate),[26] STR-g-poly(sulfobetaine methacrylate),[27] STR-g-polystyrene, GG-g-poly(AA-co-AM),[2] STR-g-poly(acrylonitrile-co-AMPS),[28] GGTI-g-(NIPA-co-3-(N-isopropylacrylamido) sodium propanoate-co-sodium acrylate),[1] and synthetic hydrogels, such as poly(AMPS-co-itaconic acid),[29] poly(AM-co-AMPS),[30] poly(AM-co-AMPS-co-4-vinylpyridine),[31] poly(acrylamidephenylboronic acid-co-AM),[32] poly(N-dodecylacrylamide-co-AMPS),[33] and poly(AMPS-co-3-acrylamidopropyltrimethylammonium chloride),[34] have been reported for drug delivery and adsorptive exclusion of dyes and M(II/III/VI). In this context, acrylamido functionality was introduced in the matrix using several AM derivatives, such as 2-acrylamido glycolic acid,[35] AMPS,[36,37] and 3-(acrylamido) phenylboronic acid,[38] as external monomers for synthesizing copolymer hydrogels. The in situ adjunct allocation of the third monomer, that is, 2-(3-(acrylamido)propylamido)-2-methylpropane sulfonic acid (APMPS), a novel/new monomer within the network of recyclable STR-g-TerPol via solution polymerization of AMPS and AM, using the RSM optimized AMPS/AM ratio, temperature, and other ingredients, extensive microstructural analyses of Sb(III)/Bi(III)–, Sb(III) + Bi(III)–STR-g-TerPol and/or unadsorbed hydrogels for individual and/or simultaneous decontamination of wastewater is reported for the first time.

Experimental Section

Materials

NaHCO3, Na2B4O7·10H2O, KCl, CH3COOH, CH3COONa, HCl, NaOH, AMPS, AM, STR, MBA, sodium bisulfite (SBS), and potassium persulfate (PPS) of analytical grades were purchased from Merck and used without any further modification. BCB, CV, SbCl3, and Bi(NO3)3 were purchased from Sigma-Aldrich.

Optimization of Synthesis Parameters

STR-g-TerPols were prepared using varied AMPS/AM ratios, STR (wt %), MBA (wt %), and SBS + PPS (wt %) within 2:1–20:1, 1.0–2.0, 0.2–0.8, and 0.6–1.4, respectively, to measure the effect of these ingredients on % swelling ratio (SR) (−) and ESR systematically. However, STR-g-TerPol composed of 20:1 mole ratio of AMPS/AM, 1.5 wt % STR, 0.4 wt % MBA, and 1.0 wt % SBS + PPS showed good balance between swelling, gelation time, and network integrity.

Synthesis of STR-g-TerPols

STR-g-TerPols were synthesized (Scheme ) through free-radical solution polymerization, in which STR was grafted into the TerPol network and APMPS was allocated in situ using AMPS and AM as ex situ monomers, SBS and PPS as the redox initiator, and MBA as the cross-linker in N2 atmosphere. The exact composition of ingredients and reaction temperature were optimized by RSM via ensuring the maximum swelling of the synthesized hydrogels, which were prepared by successive incorporation of varied dosages of AMPS/AM and MBA at different pHi. Initially, 1.5 g of STR was dissolved in 45 mL double distilled water at 323 K using an ultrasonicator to ensure complete dissolution of STR. In a three-neck reactor, the as-obtained homogeneously dispersed STR suspension was taken, followed by the dropwise addition of 0.093 mol AMPS solution prepared in 25 mL water at pHi = 10.5 and 0.010 mol AM solution prepared in 15 mL water. Thereafter, 0.518 mmol MBA solution prepared in 5 mL water was added at constant temperature (i.e., 295 K), and the resultant solution was allowed to homogenize under N2 atmosphere for 12 h. The polymerization reaction was initiated via gradual addition of SBS+PPS as redox initiators prepared in 10 mL water by dissolving 0.482 and 0.961 mmol in 10 mL water, respectively.

Scheme 1

Synthesis of STR-g-TerPol

The as-prepared STR-g-TerPols were allowed to swell in 1:3 methanol/water (v/v) solution and washed several times for the complete removal of unreacted monomers, water-soluble oligomers, and other ingredients. Finally, STR-g-TerPols were air-dried for 5 days, followed by drying under vacuum for another 2 days.

Characterization

Unadsorbed and/or adsorbed STR-g-TerPols were characterized by the techniques given in Table . In addition, STR-g-TerPol was characterized by measuring network parameters, such as cross-link density (CD, ρc), average molecular weight between cross-links (Mc), % gel content (% GC), % graft ratio (% GR), lower critical solution temperature (LCST), pHPZC, and ESR at different pHi and temperatures. Moreover, RSM, all graphic-based analyses, and drawing of chemical structures were carried out using Design-Expert 7.0.0, Origin 9.0 software, and ChemDraw Ultra 12.0 software, respectively.

Table 1

Characterization of TerPol and M(II)–STR-g-TerPol

characterization technique	model/make	operational conditions
FTIR	Spectrum-2, Singapore	performed using KBr pellet within 4000–400 cm–1
1H NMR	Bruker-Advance Digital 300 MHz	performed in CDCl3 solvent with TMS as the internal reference
13C NMR	JEOL ECX400	performed at a frequency of 100 MHz
TGA	Pyris6 TGA, The Netherlands	operated in N2 atmosphere with flow and scanning rates of 20.0 cm3 min–1 and 10 °C min–1, respectively, within 30–800 °C
DSC	Pyris6 DSC, The Netherlands	operated in N2 atmosphere with a flow rate of 20.0 cm3 min–1 within 30–445 °C
XRD	X’Pert PRO, made by PANalytical B.V., The Netherlands using Ni-filtered Cu Kα radiation (λ = 1.5418 Å)	operated at the scanning rate of 2θ = 0.005° s–1 and angle of diffraction from 2° to 72°
FESEM and EDX	JEOL JSM-7600F having resolution of 1 nm at 15 kV and 1.5 nm at 1 kV with the scanning voltage of 100 V to 30 kV	2.9 kV and 30k× magnification
Rheology	Anton Paar rheometer, MCR 102	isothermal frequency sweep from 0 to 100 rad s–1
DLS	Zetasizer Nano ZS90	performed in CHCl3 solvent at 48.30 kcps of count rate and 298 K for 50 s

Methodology

Solutions of Bi(NO3)3 and SbCl3 were prepared by dissolving 5000 mg each of the solid salt into 1000 mL distilled water. Solutions within 5–30 and 200–1000 ppm were prepared by exact dilution of the stock solutions. Dry STR-g-TerPols (0.01 g) were added to buffer solutions containing 25 mL Bi(III) and Sb(III) for individual adsorption, whereas 12.5 mL each of Bi(III) and Sb(III) was added to obtain 25 mL solution of Sb(III) + Bi(III) for simultaneous adsorption. Thereafter, 25 mL of buffer solution of pHi = 7.0 was added. In addition, 30 ppm dye solutions were prepared by the exact dilution of 1000 ppm stock solution. Dry STR-g-TerPol (0.01 g) was added to 50 mL buffer solutions containing 25 mL CV and BCB for individual adsorption, whereas 12.5 mL each of CV and BCB was added to obtain 25 mL solution of CV + BCB for simultaneous adsorption, followed by the addition of 25 mL of buffer solution of pHi = 9.0 at 300 rpm. After predetermined time intervals, the magnetic stirrer was stopped and the solution was allowed to settle for a couple of minutes. The progress of adsorption was monitored by withdrawing the supernatant solution, followed by measuring the absorbance (A) at the respective λmax values to determine the residual concentration (C, mg L–1) of metal ion(s) and dye(s) using an atomic absorption spectrometer (PerkinElmer A-ANALYST 100) and a UV–vis spectrophotometer (PerkinElmer Lambda 365), respectively. From the precalibrated equation, M(III)/dye(s) concentration (C) was calculated, from which AC (q) (mg g–1) was determined using eq .Here, C0/C (ppm), V (mL), and ms (g) are feed M(III)/dye(s) concentrations at t = 0/t, volume of adsorbate solutions, and mass of STR-g-TerPol, respectively. After the attainment of equilibrium, the residual concentration (Ce, mg L–1) was correlated with the equilibrium AC (qe, mg g–1) in eq .

Experimental Design and Model Development for the Synthesis of STR-g-TerPol

ESR of STR-g-TerPol was observed to depend on several synthesis parameters, such as amount of AMPS (wt %, A), SBS + PPS (wt %, B), STR (wt %, C), MBA (wt %, D), and pHi (−, E), individually or simultaneously. The hydrogel possessing the maximum ESR is anticipated to impart the highest AC for the removal of M(III) and dyes. However, investigation of such individual and/or simultaneous effect(s) via one-factor-at-a-time is unempirical, which does not consider two factor interaction(s) (2FIs) between variables. Therefore, design of experiment has been introduced to reduce the number of runs and measure the individual and/or simultaneous effect(s). At first, a fractional factorial design (resolution-IV) was executed to screen out the most significant parameters imparting considerable effect on ESR. Alongside, the simultaneous effects of such significant parameters on ESR were shown through central composite design (CCD) analysis.

Results and Discussion

RSM Optimized Hydrogel Synthesis

Phase-1: Selection of the Most Significant Variables

In phase-1, screening of variables was conducted through 11 experiments using resolution-IV design (Table S1). The maximum/minimum levels of A, B, C, D, and E were 96.77/83.33 wt %, 1.40/0.60 wt %, 2.00/1.00 wt %, 0.60/0.20 wt %, and 10.00/4.00, respectively. As A, B, and E crossed the Bonferroni limit of 4.38 in the Pareto chart (Figure S1), these three significant variables were considered for further CCD analysis. However, the adequacy of this model was evaluated through the following polynomial equation (eq ):

Indeed, such model was highly significant because of the close resemblance between Adj. and Pred. R2 (i.e., 0.9279/0.8479), very high R2 (i.e., 0.9599), and very low p-value (i.e., 0.0011) (Table S2).

Phase-2: CCD Optimization of the Significant Variables

The CCD was introduced for optimizing the three most significant process variables of synthesis, that is, A, B, and E, considering the individual and simultaneous effects on ESR. In fact, the ESR of such analyses was scrutinized and interlinked with input variables for optimization using the following empirical second-order polynomial equation (eq ):Here, Y, β0, β, β, and β represent the predicted response, constant, linear, quadratic, and interaction coefficients, respectively. The applicability of the predicted model was justified by analysis of variance (ANOVA) (Table S3), using A, B, and E within 83.33–96.77 wt %, 0.60–1.40 wt %, and 6–11, respectively. The process variables and software-generated responses are listed in Table . Additionally, the quadratic model was found to be the best among linear, 2FI, quadratic, and cubic models because of higher R2, that is, 0.9987, and close proximity between Adj. and Pred. R2 (i.e., 0.9975/0.9916). In this context, A, B, C, AB, AE, BE, A2, B2, and E2 were found to be significant with a p-value less than 0.05. The final equation generated by the model in terms of actual factors was reported using eq .

Table 2

CCD of the Experiment

run no.	amount of AMPS (wt %, A)	total amount of SBS + PPS (wt %, B)	pHi (−, E)	actual ESR (−)	predicted ESR (−)
1	83.33	0.60	6.00	86.00	86.74
2	96.77	0.60	6.00	165.00	172.46
3	83.33	1.40	6.00	130.00	127.99
4	96.77	1.40	6.00	226.00	228.21
5	83.33	0.60	11.00	128.00	129.28
6	96.77	0.60	11.00	182.00	185.50
7	83.33	1.40	11.00	128.00	126.03
8	96.77	1.40	11.00	194.00	196.75
9	78.75	1.00	8.50	186.00	190.05
10	99.01	1.00	8.50	340.00	330.70
11	90.05	0.33	8.50	95.00	91.96
12	90.05	1.67	8.50	135.00	136.10
13	90.05	1.00	4.30	120.00	122.87
14	90.05	1.00	12.70	135.00	132.19
15	90.05	1.00	8.50	315.00	317.03
16	90.05	1.00	8.50	315.00	317.03
17	90.05	1.00	8.50	315.00	317.03
18	90.05	1.00	8.50	315.00	317.03
19	90.05	1.00	8.50	315.00	317.03
20	90.05	1.00	8.50	315.00	317.03

The response surface plots (Figure a–c) showed the simultaneous effects among AB, AE, and BE. Finally, in the numerical optimization section, A, B, and E were considered in the range along with the maximum ESR. A, B, and E were found to be 96.34 wt %, 1.05 wt %, and 8.42, respectively.

Figure 1

Three-dimensional response surface plots of ESR (−) vs (a) amounts of SBS + PPS (wt %) and AMPS (wt %), (b) pHi (−) and amount of AMPS (wt %), and (c) pHi (−) and amount of SBS + PPS (wt %).

Swelling and pH Reversibility Studies of TerPol and STR-g-TerPol

The swelling properties of STR-g-TerPol and TerPol were studied at 30 °C and pHi = 2.0, 4.0, 7.0, 9.0, and 12.0 by immersing accurately weighed STR-g-TerPol and TerPol in 500 mL buffer solution. The weights of swollen STR-g-TerPol and TerPol were measured after withdrawing at preselected time intervals, followed by wiping out the excess solvent. This process was continued until the weight of the swollen STR-g-TerPol and TerPol attained the constant value. % SR of STR-g-TerPol and TerPol were calculated using eq .Here, Wi and W (g) are weights of STR-g-TerPol and TerPol at t = 0 and t (min), respectively. Alongside, ESR was determined replacing W by Wf in eq .

AMPS-based stimuli-responsive smart materials generally freeze out conformational flexibility via gem dimethyl buttressing effect because of the presence of the two −CH3 at C-1. The pHi-responsive swelling of STR-g-TerPol and TerPol varied with several synthesis parameters, such as the amounts of AMPS/AM, SBS + PPS, MBA, and STR, pHi, and temperature. Indeed, the swelling of hydrogel varies with intrinsic parameters, such as available free volume, chain relaxation, and ionizable hydrophilic functional groups, that is, −SO3H, −CONH2, −CH2OH, −OH, and −NH2, capillary effect, and osmotic pressure created because of the absorption of water.

Swelling kinetics of TerPol and STR-g-TerPol at different pHi are presented in Figure , in which the ESR increased from pHi = 3.0 to 9.0 and then decreased from pHi = 9.0 to 12.0. At pHi > pHPZC, anion–anion repulsion between −SO3– and a relatively higher number of free/mobile counterions in the hydrogel network caused extended chain relaxation. Thus, macromolecular expansion resulted in the high swelling via imbibing large quantity of water. However, at a very high pHi = 12.0, STR-g-TerPol ionized rapidly, causing higher counterion concentration inside the matrix. These counterions reduced the electrostatic repulsion between polymeric chains and hence reduced ESR. In contrast, reverse effects were produced at pHi < pHPZC because of the extensive protonation of −SO3–, leading to lack in electrostatic repulsion between the polymeric chains and decreased the extent of hydrogen bonding with water. As a result, ESR was found to be the maximum and minimum at pHi = 9.0 and 2.0, respectively. In this context, better relative population of hydrophilic functional groups in STR-g-TerPol resulted in relatively enhanced ESR as compared to that of TerPol.

Figure 2

Swelling study of (a) TerPol and (b) STR-g-TerPol at different pHi; (c) pHi reversibility; and (d) pHPZC of TerPol/STR-g-TerPol.

Calculation of % GC, % GR, pHPZC, and Network Parameters of STR-g-TerPol

The % GC, % GR, and pHPZC of STR-g-TerPol, estimated using methods reported elsewhere,[1−3] were calculated to be 75.10%, 9.17%, and 6.62 (Figure ), respectively. In addition, the network theory of Flory and Rehner was employed to determine the network parameters of STR-g-TerPols[2] bearing different amounts of STR, MBA, SBS + PPS, and AMPS/AM (Table ). STR-g-TerPols showed opposite variation between CDs and ESR with the increase in MBA from 0.2 to 0.8 wt %. However, the gradual increase in MBA resulted in moderate enhancement of ρc from 1.425 × 10–9 to 3.774 × 10–9 via formation of a closely spaced tighter network of STR-g-TerPols. In fact, such compact network of hydrogels showed the decrease in Mc from 1.306 × 1011 to 4.933 × 1010. In contrast, an enhancement of SBS + PPS from 0.6 to 1.0 wt % reduced Mc from 3.212 × 10–8 to 1.604 × 10–9, and finally, Mc became 1.246 × 10–8 at 1.4 wt % SBS + PPS. Indeed, a good balance between ESR and ρc was noted at 1.0 wt % SBS + PPS. Alongside, Mc of STR-g-TerPols was noted to increase from 6.694 × 1010 to 1.161 × 1011 as a result of increase in STR from 1.0 to 1.5 wt %, whereas any further enhancement restricted the polymerization reaction because of substantial lowering in solution fluidity. However, the maximum ESR was noted at 1.5 wt % STR. In fact, ESR and Mc of hydrogels were found to increase within 158.85–290.56 g g–1 and 2.299 × 1010–1.778 × 1011, respectively, via increase in AMPS/AM from 2:1 to 20:1, for which ρc inversely varied from 6.481 × 10–9 to 1.046 × 10–9.

Table 3

Variation of Physical Properties of Hydrogels

STR-g-TerPols (AMPS/AM/MBA/SBS + PPS/STR)	density (g mL–1)	SR in water (g g–1)	volume fraction of swollen STR-g-TerPols (φp)	polymer–water interaction parameter (χ)	average molar mass between cross-links (Mc)	CD (ρc)
2:1/0.4/1.0/1.5	1.2265	158.85	0.0050	0.5016	2.299 × 1010	6.481 × 10–9
10:1/0.4/1.0/1.5	1.2176	249.45	0.0032	0.5010	1.161 × 1011	1.486 × 10–9
15:1/0.4/1.0/1.5	1.2032	270.55	0.0030	0.5010	1.479 × 1011	1.258 × 10–9
20:1/0.4/1.0/1.5	1.1835	290.56	0.0028	0.5009	1.778 × 1011	1.046 × 10–9
10:1/0.2/1.0/1.5	1.2065	260.60	0.0031	0.5010	1.306 × 1011	1.425 × 10–9
10:1/0.4/1.0/1.5	1.2176	249.45	0.0032	0.5010	1.161 × 1011	1.603 × 10–9
10:1/0.6/1.0/1.5	1.2362	220.85	0.0036	0.5012	7.976 × 1010	2.334 × 10–9
10:1/0.8/1.0/1.5	1.2526	190.48	0.0041	0.5013	4.933 × 1010	3.774 × 10–9
10:1/0.4/0.6/1.5	1.1987	112.25	0.0073	0.5024	5.797 × 109	3.212 × 10–8
10:1/0.4/0.8/1.5	1.2022	150.32	0.0045	0.5018	1.711 × 1010	1.088 × 10–8
10:1/0.4/1.0/1.5	1.2176	249.45	0.0032	0.5010	1.161 × 1011	1.604 × 10–9
10:1/0.4/1.2/1.5	1.2279	160.45	0.0049	0.5017	2.420 × 1010	7.694 × 10–9
10:1/0.4/1.4/1.5	1.2322	121.38	0.0057	0.5019	1.494 × 1010	1.246 × 10–8
10:1/0.4/1.0/1.0	1.2372	210.32	0.0037	0.5012	6.694 × 1010	2.781 × 10–9
10:1/0.4/1.0/1.5	1.2176	249.45	0.0032	0.5010	1.161 × 1011	1.603 × 10–9
10:1/0.4/1.0/2.0	1.2249	180.54	0.0044	0.5014	3.652 × 1010	5.099 × 10–9

Fourier Transform Infrared Analyses

TerPol and STR-g-TerPol

In STR-g-TerPol, grafting of STR into a synthetic TerPol network could be realized from the arrival of new peaks at 1123 and 2855 (sh) cm–1 for −CH2–O–CH2– and −O–CH2–, respectively, indicating the formation of new ether linkages via reaction between −CH2–O• of STR and AMPS/AM. Indeed, such grafting via −CH2–OH was substantiated from the disappearance of several −C–OH def. peaks of STR at 1246, 1417, 1423, and 1428 cm–1 in STR-g-TerPol. Notably, the broad spectrum of STR at the higher-frequency region, consisting of broad peaks at 3151 and 2928 cm–1, along with a shoulder at 3617 cm–1, could be related to the simultaneous prevalence of unconventional C–H···O and conventional O–H···O H-bonds (Table S4).

Because STR is mostly populated with methine C–H carrying greater H-bond-forming ability than those of >CH2 and −CH3, methine C–H of pyranose rings acted as the preferred H-bond donor in C–H···O H-bond.[39] However, such reduced broadening in STR-g-TerPol and the appearance of considerably sharp and intense peaks at 2926, 1385, and 2874 cm–1 (sh) (Figure a) were related to the facilitated C–H vibrations of free −CH3 in STR-g-TerPol, via replacement of −CH3···O H-bonds with stronger methine C–H···O H-bonds, leading to altered mutual C–H/O–H/N–H H-bonding in STR-g-TerPol. Alongside, in both hydrogels, broad peaks within 2010–2210 cm–1 could be assigned to C=NH+ and >NH2+ moieties of AM and AMPS, respectively, which might be ascribed to protein and phosphate monoesters as impurities in STR (Table S4).

Figure 3

FTIR of (a) STR, TerPol, and STR-g-TerPol and (b) STR-g-TerPol, Sb(III)–STR-g-TerPol, Bi(III)–STR-g-TerPol, and Sb(III) + Bi(III)–STR-g-TerPol.

Bi(III)–, Sb(III)–, and Sb(III) + Bi(III)–STR-g-TerPol

In Bi(III)–STR-g-TerPol, preferential coordination of Bi(III) with the oxygen of sulfonates resulted in the arrival of the Bi–O peak at 580 cm–1 in Bi(III)–STR-g-TerPol and the disruption of strong O–H···O–H and unconventional C–H···O type H-bond, as realized from the disappearance of the peak at 2347 cm–1 and the scarcity of free −CH3 in Bi(III)–STR-g-TerPol (Table S5). In comparison to Bi(III), poor interaction between Sb(III) and STR-g-TerPol also resulted in a relatively smaller decrement in −CH2–O–CH2– specific frequency from 1123 cm–1 of STR-g-TerPol to 1115 cm–1 in Sb(III)–STR-g-TerPol. Moreover, superficial deposition of antimony oxides onto both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol was ascertained from the exclusive arrival of broad peaks centered at 508 and 506 cm–1 (Table S5). In fact, in Sb(III) + Bi(III)–STR-g-TerPol, a part of both Sb(III)–STR-g-TerPol and Bi(III)–STR-g-TerPol specific peaks was retained. However, it transpired that the coordination of Bi(III) with the oxygen donor ligands of STR-g-TerPol deteriorated in the presence of Sb(III), as the Bi–O specific peak at 580 cm–1 disappeared in Sb(III) + Bi(III)–STR-g-TerPol (Figure b), along with the reappearance of peaks at 2312 cm–1 corresponding to strong O–H···O–H H-bond. Alongside, the availability of free −CH3 in Sb(III) + Bi(III)–STR-g-TerPol was noted to register in between those of Bi(III)–STR-g-TerPol and Sb(III)–STR-g-TerPol, realized from the lack of C–H str. at 2932 and 2980 cm–1, while retaining the −CH3 doublet peaks at 1371 and 1395 cm–1. Thus, as compared to Bi(III)–STR-g-TerPol, partial disappearance of free −CH3 again substantiated the retaining of some C–H···O H-bond in consequence to the affected coordination of Bi(III) with the oxygen donor ligands in Sb(III) + Bi(III)–STR-g-TerPol. The average coordinating tendency of oxygen donor ligands in Sb(III) + Bi(III)–STR-g-TerPol was also envisaged from the intermediate values of −CH2–O–CH2– peak positioned in between those of Bi(III)–STR-g-TerPol and Sb(III)–STR-g-TerPol. Besides, in all the metal ion-adsorbed hydrogels, complete disappearance of broad peaks within 2010–2210 cm–1 could be because of deprotonation of C=NH+ and >NH2+ during adsorption at the neutral pHi.

NMR Analyses

STR

1H NMR (300 MHz, DMSO-d6, δ, ppm): 3.06, 3.58–3.60, 4.57, 4.90–5.06, 5.39–5.49 (Figure c); 13C NMR (75 MHz, DMSO-d6, δ, ppm): 60.56, 71.68, 72.05, 73.31, 78.84, 100.15 (Figure S2).

Figure 4

1H NMR of (a) TerPol and (b) STR-g-TerPol, (c) 1H NMR of STR, and (d) 13C NMR of STR-g-TerPol.

TerPol

1H NMR (300 MHz, CDCl3, δ, ppm): 1.05–1.78, 1.45, 2.84, 2.16–2.70, 3.60–3.70, 4.31, 7.30, 7.90 (Figure a).

STR-g-TerPol

1H NMR (300 MHz, CDCl3, δ, ppm): 0.84–1.57, 1.42, 2.84, 2.07–2.62, 3.49–3.54, 3.64, 4.32, 5.07, 5.34, 6.98, 7.09, 7.53, 7.72 (Figure b); 13C NMR (100 MHz, δ, ppm): 30.36, 54.79, 63.60, 38.98–45.05, 70.44–82.65, 106.75–111.79, 178.28 (Figure d).

1H NMR Analyses

The incorporation of AMPS within TerPol and STR-g-TerPol was inferred from the characteristic peaks at 1.45/2.84 and 1.42/2.84 ppm for −CH3/–CH2–SO3H of AMPS in TerPol and STR-g-TerPol, respectively. Moreover, the −CH2– and >CH– units resulted through free-radical polymerization of AMPS, AM, and MBA, which were confirmed from the characteristic peaks within 1.05–1.78/2.16–2.70 and 0.84–1.57/2.07–2.62 ppm (Figure a,b) in TerPol and STR-g-TerPol, respectively. Alongside, the complete disappearance of vinylic proton-specific peaks within 6.03–6.59, 5.70–6.30, and 5.59–6.13 ppm for AMPS, AM, and MBA, respectively, supported the formation of a saturated network via polymerization reaction. In this context, cross-linking by MBA was ascertained through the appearance of characteristic −CH2– peaks at 4.31 and 4.32 ppm in TerPol and STR-g-TerPol, respectively. However, the appearance of new peaks within 3.60–3.70 and 3.49–3.54 ppm in TerPol and STR-g-TerPol, respectively, inferred the formation of −CONH–CH2– fragment via in situ polymerization between AM and AMPS.[40,41] In this context, the −CONH– of AMPS, cross-linked MBA, and newly formed −CONH–CH2– units appeared at 7.30 and 7.90 ppm in TerPol and at 6.98, 7.09, 7.53, and 7.72 ppm in STR-g-TerPol.

STR is a natural polysaccharide consisting of a large number of glucopyranose units connected through α-(1 → 4) and α-(1 → 4)/α-(1 → 6) linkages of amylose and amylopectin, respectively. The characteristic peaks within 5.39–5.49 and 4.90–5.06 ppm which were related to H-1(1 → 4) and H-1(1 → 6), respectively, inferred the presence of (1 → 4)- and/or (1 → 6)-linked glucopyranose unit(s) in STR.[42,43] Moreover, the peaks within 3.58–3.60 ppm were assigned to H-2 and H-4 of glucopyranose (Figure c). However, incorporation of STR within STR-g-TerPol was confirmed from the arrival of several distinct STR-specific peaks at 5.34, 5.07, and 3.64 ppm, assigned to H-1(1 → 4), H-1(1 → 6), and ring protons, respectively.

13C NMR Analyses

In 13C NMR, the incorporation of AMPS within STR-g-TerPol was inferred from the appearance of characteristic peaks at 30.36, 54.79, and 63.60 ppm for −CH3, −CONHC(Me)2–, and −CH2–SO3H of AMPS, respectively. Moreover, −CH2– of MBA, −CH2–/>CH– of backbone, and −CONHCH2– of a strategically incorporated APMPS monomer appeared within 38.98–45.05 ppm. In addition, the in situ/ex situ allocation of pendant primary and secondary amides in backbone was inferred through the presence of characteristic −CONH–/–CONH2 specific peak at 178.28 ppm.

The 13C NMR of STR showed several characteristics peaks at 100.15, 78.84, 73.31, 72.05, 71.68, and 60.56 ppm (Figure S2) for C-1, C-2, C-3, C-4, C-5, and C-6 of glucopyranose, respectively.[44,45] In this context, modification via grafting into the TerPol network was ascertained from the shifting of several STR-specific peaks (Figure d) within 106.75–111.79 and 70.44–82.65 ppm for the anomeric carbon (i.e., C-1) and ring carbons (i.e., C-2–C-6), respectively.

Thermogravimetric Analyses of STR-g-TerPol and Sb(III)–/Bi(III)–/Sb(III) + Bi(III)–STR-g-TerPol

Grafting of STR into TerPol (Table S4) enhanced the thermal stability of STR in STR-g-TerPol, which were envisaged from the almost similar degradation pattern of both TerPol and STR-g-TerPol above 330 °C (Figure a). However, within 30–330 °C, relatively higher mass loss for STR-g-TerPol as compared to TerPol was attributed to the higher moisture loss from the STR moieties of STR-g-TerPol. From 330 to 345 °C, rapid mass losses of 30 wt % for both TerPol and STR-g-TerPol were attributed to the degradation of sulfonic groups. Importantly, delayed mass loss beyond 330 °C and thus, radically elevated thermal stability of Bi(III)–STR-g-TerPol over STR-g-TerPol could be ascribed to the restricted degradation of sulfonic groups coordinated with the adsorbed Bi(III). On contrary, relatively rapid mass loss for both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol could be attributed to the lack of complexation of sulfonic groups with the adsorbed Sb(III). Notably, similar nature of plots for both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol and diminished complex-forming ability of Bi(III) with the sulfonic groups in the presence of Sb(III) were also envisaged from the disappearance of the Bi–O peak at 580 cm–1 in Sb(III) + Bi(III)–STR-g-TerPol. In the presence of Sb(III), relatively reduced adsorption of Bi(III) in Sb(III) + Bi(III)–STR-g-TerPol was realized via similar residue content of 11.78 and 11.15 wt % for Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-(TerPol), respectively. However, in the absence of Sb(III), possibly elevated complexation between Bi(III) and sulfonic groups and associated huge adsorption of Bi(III) resulted in a substantially higher residue content of 44.08 wt % for Bi(III)–STR-g-(TerPol). Higher adsorption of Bi(III) over Sb(III) in STR-g-TerPol was also substantiated from the fact that the entire residue content of Bi(III)–STR-g-TerPol is predominantly constituted of metallic Bi and carbon, whereas for Sb(III)–STR-g-TerPol, the residue was composed of Sb2O3 and carbon.[1] Besides, sharp degradation of both Sb(III) + Bi(III)–STR-g-TerPol and Sb(III)–STR-g-TerPol beyond 220 °C was associated with the relative ease of amide to imide conversion. Altogether, as compared to Bi(III)–STR-g-TerPol, relatively poor thermal stabilities of Sb(III) adsorbed hydrogels could be related to the fewer population and weaker association of Sb(III) within the bulk of STR-g-TerPol, resulting in superficial deposition of antimony oxide-type particles onto Sb(III) + Bi(III)–STR-g-TerPol (Figure S3).

Figure 5

(a) TGA of STR, TerPol, STR-g-TerPol, and Sb(III)–/Bi(III)–/Sb(III) + Bi(III)–STR-g-TerPol; (b) LCST of TerPol and STR-g-TerPol; DSC of (c) STR, TerPol, and STR-g-TerPol and (d) STR-g-TerPol and Sb(III)–/Bi(III)–/Sb(III) + Bi(III)–STR-g-TerPol.

Differential Scanning Calorimetry Analyses

Grafting of STR into TerPol was substantiated from the relatively restricted thermal degradation of STR-g-TerPol than STR in the entire temperature range, as envisaged from the shift of all endothermic peaks of STR toward higher temperatures in STR-g-TerPol, along with the significant alteration in intensities. Indeed, slightly elevated thermal stability of STR-g-TerPol as compared to TerPol could also be realized from the conversion of a sharp peak at 340 °C of TerPol to a broad shoulder in STR-g-TerPol, suggesting grafting-assisted change in the nature of degradation of sulfonic groups (Figure c). In Bi(III)−STR-g-TerPol, relatively delayed thermal degradation was observed from the omission of STR-g-TerPol-specific endothermic peak at 379 °C, resulted from the possible stabilization of sulfonates via complexation, supported by thermogravimetric analysis (TGA). On the contrary, almost similar plots were resulted for Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol. In fact, almost identical endothermic peaks were originated within 200–205 and 300–307 °C, suggesting a pivotal role of Sb(III) in controlling the thermal degradation behavior of Sb(III) + Bi(III)–STR-g-TerPol (Figure d). Accordingly, in the presence of Sb(III), relatively poor adsorption of Bi(III) into STR-g-TerPol was substantiated from the similarity of differential scanning calorimetry (DSC) plots of both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol.

LCST Studies of TerPol and STR-g-TerPol

As compared to TerPol, the relative lowering of LCST in STR-g-TerPol was ascribed to the relatively deteriorated low-temperature stability of the STR-g-TerPol microstructure, facilitating easier temperature-dependent reversible collapsing through conformational changes, analogous to coil-to-globule transition (Figure b). Indeed, easier coil-to-globule transition and associated faster denaturation of STR-g-TerPol were assisted by the severe rupture of pre-existing strong O–H···O–H-bonds in TerPol, as realized from the disappearance of O–H str. peaks (Figure a) in STR-g-TerPol via grafting-assisted structural alteration and associated changes in H-bonding environment through intermixing of phases (Figure S3). Such low-temperature instability of STR-g-TerPol than TerPol could also be substantiated by a relatively greater mass loss from STR-g-TerPol as compared to TerPol (Figure a).

X-ray Diffraction Analyses

Grafting of STR into TerPol resulted in severe intermixing of phases via intermingling of macromolecular moieties, leading to the disappearance of several sharp STR-specific peaks in STR-g-TerPol (Figure S3a). In this context, relatively amorphous characteristics of STR-g-TerPol could be exemplified in consequence to the interpenetration of chains, resulting in the marked deterioration of the pre-existing short-range order in STR. Notably, crystalline nature was further deteriorated in the metal ion-adsorbed STR-g-TerPol, especially in Bi(III)–STR-g-TerPol (Figure S3b). However, few sharp peaks in Sb(III)–STR-g-TerPol were attributed to the superficial deposition of crystalline particulate matters based on antimony oxide (Figure S3b), detected from the peaks of variable intensities at 15.03, 28.96, 30.14, 38.03, 42.98, 45.92, and 58.82° for (111), (311), (222), (331), (422), (511), and (533) planes of antimony oxide, respectively. In this regard, small peaks between 45.92° and 45.95° could be assigned to the variable Sb–O bond distances of 1.9–2.2 Å.[46] In fact, all the characteristic peaks of Sb(III)–STR-g-TerPol were almost retained in Sb(III) + Bi(III)–STR-g-TerPol, suggesting superficial attachment tendency of Sb(III) onto STR-g-TerPol. In fact, such peaks were unaltered in Sb(III) + Bi(III)–STR-g-TerPol, as Bi(III) showed the tendency to penetrate deep inside the bulk of STR-g-TerPol rather than staying at the surface.

Field Emission Scanning Electron Microscopy and Energy-Dispersive X-ray Analyses

The photomicrograph of STR-g-TerPol demonstrated characteristic layered structures constituting of partly indistinct strata (Figure ). In this regard, multiple phases in STR-g-TerPol were resulted via interpenetration of macromolecular chains, leading to deterioration of the short-range order existing in STR (Figure S3).

Figure 6

Field emission SEM microphotographs of (a) STR-g-TerPol, (b) Bi(III)–STR-g-TerPol, (c) Sb(III)–STR-g-TerPol, and (d) Bi(III)–/Sb(III)–STR-g-TerPol; the inset of b–d shows EDX spectra of respective M(III).

In Bi(III)–STR-g-TerPol, the absence of superficial particulate depositions, along with rough topological features devoid of phase boundaries, was substantiated from the absence of peaks in diffractogram because of the intimate interactions among the bulk/surface of STR-g-TerPol and the adsorbed Bi(III) (Figure S3b). On the contrary, both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol showed depositions of antimony oxides (Figure S3), substantiating the relative abundance and scarcity of Sb(III) at the surface and bulk, respectively. Such depositions were responsible for the inferior thermal resistance of both Sb(III)–STR-g-TerPol and Sb(III) + Bi(III)–STR-g-TerPol (Figure a).

From energy-dispersive X-ray (EDX) spectrum (inset of Figure b–d), the appearance of different intense peaks supported variegated interactions between Sb(III)/Bi(III) and STR-g-TerPol through ionic/coordinate bonding. Indeed, the presence of such oxides of Bi(III) was substantiated by the characteristic peaks of Bi(III) and oxygen in the EDX spectrum, as apprehended earlier in Fourier transform infrared (FTIR) spectra (Figure b).

Rheological Analyses

The frequency sweep study of swollen TerPol and STR-g-TerPol, obtained through grafting of different amounts of STR into TerPol, was carried out to realize the relative alterations in rheological and viscoelastic properties. However, an incorporation of 0.5 wt % STR into TerPol resulted in the significant enhancement of storage modulus (G′) through the attainment of extra cross-linking and compact structure of H-bonded STR. Moreover, G′ increased further with the change in STR from 0.5 to 1.5 wt % (Figure ). Thereafter, a marginal change in G′ with further increase in STR above 1.5 wt % was ascribed to the attainment of an entangled complex network. However, G″ followed the similar kind of variation, in which G′ was always greater than G″. In fact, the obsolescence of crossover point within 0–100 rad s–1 indicated that the as-prepared hydrogels were more elastic than viscous.[47]

Figure 7

G′ and G″ of TerPol and STR-g-TerPols.

Adsorption Isotherm, Kinetics, and Thermodynamics Studies

Adsorption isotherm studies express the nature of interaction between the adsorbate and adsorbent. The equilibrium experimental data were fitted to Langmuir, Freundlich, and Brunauer–Emmett–Teller (BET) isotherms (eqs –8), of which the Langmuir model (Figure a–d) fitted the best (Table ). The ligand selectivity of STR-g-TerPol toward M(III) adsorption was corroborated from the qmax values in the following order: Bi(III) (1046.32 mg g–1) > Sb(III) (932.08 mg g–1). The higher binding affinity of Bi(III) with N-/O-donor elevated the AC for STR-g-TerPol. In this regard, separation factor (RL) (eq ) measuring the feasibility of adsorption, in which RL > 1, 0 < RL < 1, RL = 1, and RL = 0 indicates unfavorable, favorable, linear, and irreversible adsorptions, respectively, was found to vary within 0–1, indicating spontaneous adsorption.Here, kL/kF/k1/k2 and qmax/n/qBET are isotherm constant and isotherm parameter, respectively.

Figure 8

Nonlinear Langmuir fitting at (a) low ppm of Sb(III), (b) high ppm of Sb(III), (c) low ppm of Bi(III), and (d) high ppm of Bi(III).

Table 4

Adsorption Isotherm and Kinetics Parameters

	temperature (K)
models parameters	293	303	313	323
Bi(III)
Langmuir
qmax (mg g–1)/pHi/C0 (ppm)	288.25/7/5–30	282.39/7/5–30	262.03/7/5–30	254.22/7/5–30
kL (L mg–1)	0.8884	0.4803	0.3390	0.2168
R2/F	0.9974/3804.13	0.9987/7471.50	0.9979/4723.95	0.9997/42 051.37
Langmuir
qmax (mg g–1)/pHi/C0 (ppm)	1120.33/7/200–1000	1047.39/7/200–1000	931.64/7/200–1000	869.61/7/200–1000
kL (L mg–1)	0.0248	0.0130	0.0097	0.0066
R2/F	0.9979/3667.33	0.9975/3174.53	0.9975/3220.24	0.9995/16 858.74
Pseudosecond-Order
qe,cal (mg g–1)/pHi/C0 (ppm)	146.18/7/30	142.53/7/30	135.34/7/30	126.53/7/30
qe,exp (mg g–1)	146.27 ± 4.39	143.11 ± 4.29	134.00 ± 4.02	126.77 ± 3.80
k2 (g mg–1 min–1)	1.62 × 10–3	2.15 × 10–3	2.78 × 10–3	3.76 × 10–3
R2/F	0.9920/9936.71	0.9936/11 601.49	0.9919/7761.15	0.9921/8485.28
Pseudosecond-Order
qe,cal (mg g–1)/pHi/C0 (ppm)	484.99/7/1000	471.07/7/1000	454.06/7/1000	425.23/7/1000
qe,exp (mg g–1)	484.99 ± 14.55	468.23 ± 14.05	450.43 ± 13.51	426.33 ± 12.79
k2 (g mg–1 min–1)	6.05 × 10–4	6.11 × 10–4	7.40 × 10–4	10.0 × 10–4
R2/F	0.9934/8928.26	0.9956/13 061.09	0.9954/13 118.48	0.9926/8749.11
Sb(III)
Langmuir
qmax (mg g–1)/pHi/C0 (ppm)	141.00/7/5–30	137.85/7/5–30	125.92/7/5–30	122.08/7/5–30
kL (L mg–1)	1.4685	1.2344	1.2929	1.0960
R2/F	0.9915/1236.37	0.9878/868.46	0.9937/1743.48	0.9918/1358.49
Langmuir
qmax (mg g–1)/pHi/C0 (ppm)	952.09/7/200–1000	932.08/7/200–1000	918.60/7/200–1000	872.65/7/200–1000
kL (L mg–1)	0.0144	0.0123	0.0101	0.0090
R2/F	0.9986/6424.44	11 554.76/42 004.64	0.9998/42 004.64	0.9924/1174.65
Pseudosecond-Order
qe,cal (mg g–1)/pHi/C0 (ppm)	128.57/7/30	124.08/7/30	114.27/7/30	110.28/7/30
qe,exp (mg g–1)	126.49 ± 3.79	123.91 ± 3.71	115.30 ± 3.45	111.10 ± 3.33
k2 (g mg–1 min–1)	2.65 × 10–3	3.55 × 10–3	4.62 × 10–3	6.45 × 10–3
R2/F	0.9922/7186.62	0.9937/9390.50	0.9958/14 462.05	0.9971/21 416.45
Pseudosecond-Order
qe,cal (mg g–1)/pHi/C0 (ppm)	735.20/7/1000	708.00/7/1000	682.67/7/1000	642.92/7/1000
qe,exp (mg g–1)	740.71 ± 22.22	720.09 ± 21.60	693.29 ± 20.80	644.73 ± 19.34
k2 (g mg–1 min–1)	9.47 × 10–5	6.67 × 10–4	5.16 × 10–4	3.82 × 10–4
R2/F	0.9979/29 335.96	0.9964/16 423.51	0.9960/14 435.17	0.9967/16 097.83

The adsorption mechanism was studied through fitting of the kinetics data to pseudofirst-/-second-order kinetics models (Figures S4 and S5). Better fitting of the pseudosecond-order kinetics equation (eq ) than the pseudofirst-order equation (eq ) suggested the prevalence of chemisorption[48] (Table ) through ionic and variegated coordinative interactions between O-donor of −SO3–/N-donor of >N–/–NH2 in STR-g-TerPol and M(III). The activation energies (Ea) of adsorption, calculated using the Arrhenius-type equation (eq ), were 22.58 and 23.08 kJ mol–1 for Bi(III) and Sb(III), respectively, indicating the prevalence of chemisorption.[49,50] The spontaneity of the chemisorption was inferred from −ΔG0 equation (eq ) for all M(III).Here, distribution coefficient (kd) is defined as the ratio of M(III) concentrations in solid to liquid phases at equilibrium (eq ).

Again, the exothermic nature of adsorption was understood from negative ΔH0 (Table ), as calculated by van’t Hoff’s equation (eq ), whereas the positive ΔS0 suggested fair affinity between M(III) and STR-g-TerPol and the decrease in randomness at the solid–solution interface during adsorption (Figure S6).

Table 5

Adsorption Thermodynamics Parameters for Bi(III)–/Sb(III)–STR-g-TerPol

C0 (ppm)	T (K)	–ΔG0 (kJ mol–1)	–ΔH0 (kJ mol–1)	–ΔS0 (J mol–1 K–1)
5	293	14.04/13.39	47.87/17.87	116.66/11.67
	303	12.64/13.47
	313	11.24/13.62
	323	10.47/13.76
10	293	13.16/12.21	44.63/15.77	103.71/11.87
	303	12.18/12.18
	313	11.02/12.07
	323	10.18/11.75
15	293	12.73/11.04	43.26/12.86	102.84/6.31
	303	11.70/10.81
	313	10.86/10.85
	323	9.95/10.73
20	293	12.39/10.05	43.68/13.91	105.26/13.07
	303	11.41/9.90
	313	10.61/9.88
	323	9.65/9.57
25	293	12.23/8.83	42.26/13.66	101.78/16.04
	303	11.02/8.77
	313	10.31/8.54
	323	9.32/8.33
30	293	11.84/7.66	43.30/12.90	107.16/17.82
	303	10.61/7.54
	313	9.72/7.31
	323	8.88/7.14

Simultaneous Adsorption of Sb(III) and Bi(III) by STR-g-TerPol

For binary adsorption, that is, for simultaneous removal of Sb(III) + Bi(III), qe and Ce were fitted to nonlinear competitive Langmuir, noncompetitive Langmuir, and combined Langmuir–Freundlich models of adsorption expressed by eqs –18, respectively, with multiple regression for surface fitting.Here, i/j, qe, qmax′, KL,/KL,, KL,/KL,, and KLF,/KLF, are Bi(III)/Sb(III), equilibrium AC of i, maximum AC, competitive, noncompetitive Langmuir, and combined Langmuir–Freundlich model parameters, respectively.

In simultaneous adsorption of Sb(III) + Bi(III), restriction toward interaction, imposed during simultaneous adsorption, was inferred via substantial decrease in qe of both Bi(III) and Sb(III) as compared to the individual qe. In this context, qe of Bi(III) and Sb(III) were 143.11 and 123.91 mg g–1 at 30 ppm that reduced to 141.25 and 115.75 mg g–1, respectively, in simultaneous adsorption of Bi(III)/Sb(III) at 30/5 and 5/30 ppm (Figure a–d). However, the noncompetitive Langmuir model fitted the best as compared to the competitive model, realized through very high R2/F (i.e., 0.9813/2053.08) (Table ). In fact, FTIR, TGA, and scanning electron microscopy (SEM) analyses confirmed that Bi(III) penetrated deep inside the bulk to bind with O- and N-donor, whereas Sb(III) deposited onto the surface (Figure ) and preferentially coordinated with the O-donor. As a consequence, substantial deposition of Sb(III) onto the surface of STR-g-TerPol imposed restriction for the penetration of Bi(III) toward bulk and, resulting in reduced qe (Scheme ). It is tantamount to note that the combined Langmuir–Freundlich model fitted the best, reflecting mutual interaction between metal ions and surface heterogeneity of the adsorbents.

Figure 9

Fitting of (a) competitive Langmuir, (b) noncompetitive Langmuir for Bi(III), and combined Langmuir–Freundlich for (c) Bi(III) and (d) Sb(III).

Table 6

Binary Adsorption Isotherm Parameters of Bi(III)/Sb(III)

binary mixtures of Bi(III)/Sb(III)
Langmuir competitive		Langmuir noncompetitive		combined Langmuir–Freundlich
Bi(III)		Bi(III)			Bi(III)	Sb(III)
qmax′ (mg g–1)	260.85	qmax′ (mg g–1)	270.32	qmax′ (mg g–1)	173.09	136.02
KL,i (L mg–1)	0.8776	KL,i (L mg–1)	0.9607	KLF,i (L mg–1)	4.2368	2.4120
KL,j (L mg–1)	0.3016	KL,j (L mg–1)	0.5825	KLF,j (L mg–1)	1.4858	5.4652
R2	0.9760	KL,ii (L2 mg–2)	0.0602	ni	0.6851	0.9144
χ2	35.64	KL,jj (L2 mg–2)	0.0720	nj	1.1437	1.1805
F	2666.89	R2/χ2/F	0.9813/27.81/2053.08	R2/χ2/F	0.9935/9.60/5959.88	0.9880/9.69/3977.62

Scheme 2

Adsorption of Sb(III) and Bi(III) Individually and Simultaneously onto the Surface of STR-g-TerPol

Individual and/or Simultaneous Adsorption of BCB and CV at pHi = 9.0

The time-dependent UV–vis absorption spectra of the BCB solution constituted of BCB+ monomer-specific λmax at 624 nm, accompanied by a dimer-specific shoulder at 580 nm, suggesting the coexistence of BCB+ monomers and some H-aggregate type BCB dimers in aqueous solution during the entire period of absorption (Figure a).[51] In this regard, the aqueous solution of BCB+ was devoid of the peak corresponding to J-aggregate type dimers, as flat aromatic structures of cationic BCB dyes tend to form the stacking-type face-to-face aggregate in aqueous medium. Indeed, a time-dependent slow hypochromic effect was observed in both the peak and shoulder, with complete obsolescence of shoulder at equilibrium, suggesting breakdown and conversion of BCB+ dimers into monomers at the relatively higher dilution. On the contrary, relatively rapid adsorption of CV displayed an initial λmax at 587 nm and a shoulder at 546 nm, which was ascribed to the coexistence of a pyramidal and planar CV+ monomer, of which the planar CV+ monomer formed dimers or aggregates of higher order at lower dilution. Within first 30 min, the shoulder became increasingly prominent and eventually transformed into a broad peak shifted to 514 nm (Figure b). Such phenomenon was ascribed to the relatively rapid and preferential attachment of tiny pyramidal CV+ monomers with the STR-g-TerPol, whereas relatively bulkier dimers or aggregates of planer dimers were considerably reluctant to get adhered with the adsorbent. Notably, beyond 30 min, a considerable hypsochromic shift was observed from 546 to 514 nm, along with the consistent hypochromic effect for the CV+ monomer-specific peak. Such a bathochromic shift could be ascribed to the breakdown of higher-order aggregates into CV+ dimers and monomers because of increasing dilution. Accordingly, in the combined adsorption spectra, preferable CV+ absorption into STR-g-TerPol was noted to supersede the adsorption of BCB+, as the time-dependent combined adsorption spectra showed the faster hypochromic effect at 587 nm, in addition to the progressively prominent appearance of a BCB+-specific shoulder at 624 nm. In this regard, strong inter chromophoric interaction was less likely as the characteristic λmax of both BCB+ and CV+ were deviated marginally (Figure c). However, dimer or higher-order aggregate-specific peaks of both the dyes disappeared, as the initial concentration of both the individual dye components was reduced from 30 to 15 ppm. Nevertheless, in the presence of BCB+, the adsorbing tendency of CV+ onto STR-g-TerPol was retarded because of the common ion effect.

Figure 10

Full scan UV–vis spectra for the removal of (a) BCB, (b) CV, and (c) BCB + CV by STR-g-TerPol at pHi = 9.0, adsorbent dosage = 0.01 g, and temperature = 303 K; (d) reusability plot of STR-g-TerPol and DLS of STR-g-TerPol (e) before the first cycle and (f) after the completion of five cycles.

Desorption and Reusability

The recyclability of STR-g-TerPol was studied by carrying out repetitive adsorption/desorption studies at pHi = 7.0/2.0 (Figure d). The extent of desorption was approximately 90% for Sb(III)–STR-g-TerPol, whereas the extent of desorption remained within 80–85% for Bi(III)–STR-g-TerPol. The excellent reusability of STR-g-TerPol could be explained by measuring very high ACs even after the completion of five cycles. Dynamic light scattering (DLS) technique was employed for determining the particle size of STR-g-TerPol. The sizes of particles in solution were determined using the following Stokes–Einstein eq .[52]Here, D, k, η, T, and Rh are diffusion coefficient, Boltzmann’s constant, viscosity coefficient, temperature, and hydrodynamic radius of the particles in solution, respectively. DLS study was executed for hydrodynamic radius estimation of both unadsorbed STR-g-TerPol and desorbed STR-g-TerPol obtained after the attainment of five complete cycles of the recyclability experiment, in which the particle size varied within 131.5–121.2 nm without a significant loss of network (Figure e,f). Thus, STR-g-TerPol showed the excellent reusability.

Comparison of the Results

Several low-cost natural, physicochemically modified micro-/nanomaterials, blends, homo-/co-/ter-polymers, IPNs, and composite hydrogels have been employed for the adsorptive decontamination of Bi(III) and Sb(III) at varying initial concentrations (i.e., 0–20 000 ppm), temperatures (i.e., 293–323 K), and pHi (i.e., 2.8–11.0) (Table S6). From Table S6, the ACs of STR-g-TerPol were found to be excellent as compared to previously reported macromolecular adsorbents.

Conclusions

The present work reports the unorthodox fabrication of a series of reusable STR-g-TerPol superadsorbents through grafting of STR into the TerPol network formed via in situ allocation of the APMPS monomer by N–H activation. The synthesis of the TerPol network, via ex situ addition of only two monomers and in situ adjunct allocation of APMPS, was inferred from the arrival of −CONH–CH2– specific 1H NMR peaks within 3.49–3.54 ppm. The one-pot grafting of STR into the TerPol network resulted in −CH2–O–CH2– linkage in STR-g-TerPol, inferred from 1H/13C NMR and FTIR analyses. Alongside, the grafting of STR was confirmed through −CH2–O–CH2– str. at 1123 cm–1 and several STR-specific peaks within 106.75–111.79 and 70.44–82.65 ppm for the anomeric carbon and ring carbons, respectively. As compared to Sb(III), the preferential interaction of Bi(III) with O-donor resulted in the arrival of a Bi–O-specific peak at 580 cm–1 in Bi(III)–STR-g-TerPol. Consequently, the higher adsorption of Bi(III) over Sb(III) resulted in homogeneous bulk-/surface distribution of Bi(III) rendering higher residue in TGA thermogram, supported by DSC and SEM analyses. The replacement of −CH3···O type H-bonds with stronger nonconventional C–H···O H-bonds donated mostly by the methine of grafted STR significantly altered the C–H str. of STR-g-TerPol, producing considerably sharp and intense peaks at 2926, 1385, and 2874 cm–1. Such nonconventional H-bonds were affected significantly via adsorption of Bi(III) instead of Sb(III). In the simultaneous adsorption, the decrease in AC of Bi(III) was ascribed to the restricted penetration of Bi(III) deep inside the STR-g-TerPol that resulted via extensive surface deposition of Sb(III)-components. The introduction of this new pathway can be adopted for the synthesis of APMPS-based TerPols, without orthodox ex situ addition of APMPS or equivalent monomers. The systematic design of STR-g-TerPol bearing excellent recyclability, performance characteristics, outstanding superadsorption efficiency, and diversified application prospects for individual and/or simultaneous removal(s) of M(III) and cationic dyes has shown the novelty in a kinetically fast and thermodynamically spontaneous waste remediation process. STR-g-TerPol showing excellent physicochemical and thermomechanical properties may also be attempted for drug delivery, tissue engineering, membrane-based separation, and self-healing materials.