Alternate cyclopolymer of diallylglutamic acid and sulfur dioxide

Abstract

The monomer N,N-diallylglutamic acid hydrochloride [(CH₂=CH–CH₂)₂NH⁺CH(CO₂H)CH₂CH₂CO₂H·Cl⁻] (I)/SO₂, underwent cyclopolymerization to afford the alternate cationic polyelectrolyte (CPE) poly(I-alt-SO₂) (II) in very good yields. Upon depletion of HCl during soaking in water, water-soluble triprotic (+) II was transformed to water-insoluble diprotic zwitterionic (±) III containing residues of glutamic acid in each repeating unit. Upon treatment with 1 and 2 equivalents of NaOH, polyzwitterions (±) III, was converted into water-soluble monoprotic poly(zwitterion-anion) (±−) IV and the fully deprotonated polydianion (=) V, respectively. Basicity strengths of the chelation centers of CO₂⁻ and nitrogen in (=) V have been determined. The polymer demonstrated its effectiveness as an antiscalant in the inhibition of CaSO₄ scaling. Keeping in view the pH-dependent solubility of the polymers, a recyclable aqueous two-phase system (ATPS) comprising of (=) V and polyethylene glycol has been constructed for the potential purification of biomolecules. After its use, the ATPS can be recycled by precipitating the ionic polymer in the form of (±) III at a lower pH.

---

1. Introduction

Glutamic acid is a non-essential amino acid that is used in the biosynthesis of proteins (Scheme 1). Its side chain CO₂H with a pK_a of 4.1 exists almost entirely in the CO₂⁻ form in the physiological pH range 7.35–7.45. The carboxylate anion in glutamate is an important excitatory neurotransmitter in neural activation.^1,2 All meat, fish, dairy products, wheat, etc., are excellent sources of glutamic acid. Glutamate, an abundantly available molecule in the brain, is involved in learning and memory,³ it is also a key compound in cellular metabolism.⁴ A series of disodium sulphonamides of glutamic acid complexes with copper(II), nickel(II) and ruthenium(III) ions, have demonstrated good anticancer activities.⁵

Scheme 1 Glutamic acid, α- and γ-poly(glutamic acid) and polymer containing glutamic acid residues.

The biodegradable, non-toxic and non-immunogenic properties of naturally occurring poly-γ-glutamic acid (γ-PGA) have rendered its use in the food, medical and wastewater industries (Scheme 1).^6–8 α-PGA has been utilized in many medical applications including drug delivery⁶ and cancer treatment.⁹ γ-Glutamic acid enhances calcium absorption in the small intestine and increases the solubility of Ca²⁺, suggesting its chelating ability.^10,11 Glutamic acid-derived polymers have the potential to act as polychelatogenes to scavenge toxic metal ions.¹² The cross-linking of microbial γ-PGA with glucose has led to a hydrogel which has been shown to be a superadsorbent of water (3000 g g⁻¹).¹³ A biodegradable PGA/gadolinium chelate has been evaluated as a contrast agent for magnetic resonance imaging (MRI); this is an important development since most of the currently evaluated macromolecular contrast agents are not biodegradable.¹⁴ The recently developed Fe₃O₄–PGA nanoparticle has been shown to hold great promise to be used as a contrast agent for MRI of tumors.¹⁵ γ-PGA-functionalized alumina nanoparticles (γ-PAN) has been evaluated for cytotoxicity towards human prostate cancer cell PC-3; the study provides a basis for further screening of the promising material for future biomedical applications.¹⁶

The abundant availability of inexpensive glutamic acid has led us to synthesize and cyclopolymerize^17–19 diallylglutamic acid 4 to obtain poly(diallylglutamic acid) 5 which retained all three original functionalities of the amino acid (Scheme 1).²⁰ Note that the nitrogen in the peptide bond in PGA loses its basic character while retaining only one of the original carboxylate motifs of the amino acid. Glutamic acid and materials derived from it are of tremendous importance for biomedical and material research. We anticipated to use 5 as a polymer component in developing a recyclable aqueous two-phase system (ATPS) for its utilization in bioseparation.^21,22 However, failure to synthesize 5 having high molar mass and its solubility behavior (vide infra) impeded the development of a recyclable ATPS in which its pH-controlled solubility would permit precipitating out and hence reuse the polymer.

The current work envisages the application of copolymerization protocol²³ to obtain 4/SO₂ alternate copolymer poly(4-alt-SO₂) 8 (Scheme 2) with the anticipation that its high molar mass and pH-dependent solubility behavior would permit us to develop a recyclable ATPS. The study would also permit us to evaluate the effects of SO₂ spacer on the basicity constants of various ligand centers and antiscalant properties. The current study would pave the way to synthesize cross-linked adsorbents containing the metal chelating centers of glutamic acid for the removal of toxic metal ions.^23–25

Scheme 2 Synthesis of cyclopolymers containing glutamic acid residues.

2. Experimental

2.1. Materials

L-Glutamic acid from Fluka AG (Buchs, Switzerland) was used as received. Sodium 3-trimethylsilylpropionate-2,2,3,3-d₄ (TSP-deuterated), an NMR internal standard, was purchased from Merck Sharp & Dohme Canada Ltd, Montreal, Canada. 2,2′-Azoisobutyronitrile (AIBN) from Fluka AG (Buchs, Switzerland) was crystallized from C₂H₅OH/CHCl₃ mixture. Calcium hydride-dried dimethylsulfoxide (DMSO) was distilled at a boiling point of 64–65 °C (4 mmHg). Poly(ethylene glycol) (PEG) of number average molecular weight (M̄_n) of 35 000 was purchased from MERCK-Schuchardt.

Dimethyl N,N-diallylglutamate (7), synthesized from dimethyl glutamate (6), was hydrolyzed with NaOH and acidified with HCl to obtain N,N-diallylglutamic acid hydrochloride 4 as described.²⁰

2.2. Physical methods for structural characterization

A Perkin Elmer (Series II Model 2400) elemental analyzer and a Fourier transform infrared (FTIR) spectrometer (Perkin Elmer 16F PC) were utilized for elemental analyses and IR spectroscopy, respectively. The nuclear magnetic resonance (NMR) spectra were recorded using a 500 MHz JEOL LA spectrometer. Tetramethylsilane (TMS) in CDCl₃ and TSP-deuterated in D₂O were used as internal standards with ¹H signal at δ 0 ppm. The ¹³C chemical shifts in D₂O were referenced against ¹³C peak of external standard dioxane at δ 67.4 ppm. A pH meter (Sartorius PB 11) and an Orion Versa Star benchtop meter (Thermoscientific) were used to measure pH and conductivity, respectively. Viscosity was measured under N₂ in an Ubbelohde viscometer (constant = 0.005317 mm² s⁻²) using polymer solution prepared in CO₂-free water. An SDT analyzer (Q600: TA Instruments) was utilized to carry out thermogravimetric analysis (TGA) under N₂.

2.3. Procedure for 4/SO₂ copolymerization

The conditions of polymerizations are summarized in Table 1. For instance, initiator AIBN was added under N₂ to a homogeneous mixture of monomer 4 (3.96 g, 15 mmol) and SO₂ (960 mg, 15 mmol) in DMSO (3.75 g) in a 25 mL-RB flask. After stirring the contents in the closed flask at 60 °C for 48 h, the viscous polymer mixture was precipitated in water, filtered and dried at 60 °C under vacuum to a constant weight of copolymer polyzwitterion acid (PZA) 9. Thermal decomposition: stable up to 240 °C, slight phase change and expansion between 240 and 245 °C, turned tan at 280 °C with evolution of gas which continue up to 400 °C leaving behind dark particles (found: C, 45.0; H, 6.1; N, 4.7; S, 10.8. C₁₁H₁₇NO₆S requires C, 45.35; H, 5.88; N, 4.81; S, 11.01%); ν_max (KBr): 3400 (v br), 2978, 1724, 1625, 1405, 1306, 1126, 878, 812, 767, 640 and 512 cm⁻¹.

Table 1 Cyclocopolymerization of 4/SO₂ at 63 °C for 36 h

Entry	Monomer (mmol)	SO2 (g) (mmol)	DMSO (g)	AIBN^a (mg)	Yield (%)	[η]^b dL g^−1
a Azobisisobutyronitrile.b Intrinsic viscosity of 1–0.0625% polymer 9 treated with 2 equivalents NaOH in 0.1 M NaCl at 30 °C was measured with an Ubbelohde Viscometer (K = 0.005317 mm^2 s^−2).
1	15	15	3.75	45	48	0.873
2	2 × 15	2 × 15	2 × 3.75	2 × 75	82	1.03
3	15	15	3.75	105	87	1.22

---

2.4. Conversion of 9 into 11 by NaOH

A mixture of PZA 9 (1.46 g, 5.0 mmol) in water (10 cm³) at 0 °C was treated with NaOH (0.4 g, 10 mmol). Another portion of NaOH (0.2 g, 5 mmol) was added and the solution was quickly dropped onto acetone (75 cm³). The resultant white polydianion electrolyte (PDE) 11 was filtered and washed with MeOH/acetone 1 : 1 mixture. The polymer was dried under vacuum till constant weight (1.6 g, 91%). Thermal decomposition: stable up to 290 °C, slight phase change and expansion between 290 and 325 °C, turned tan at 330 °C with evolution of gas which continued up to 400 °C leaving behind dark particles. ν_max (KBr): 3400 (v br), 2962, 2829, 1579 (v br), 1448, 1409, 1300, 1123, 891, 827, 786, and 514 cm⁻¹ (found: C, 37.1; H, 4.9; N, 3.9; S, 8.8. C₁₁H₁₅NNa₂O₆S·1H₂O requires C, 37.40; H, 4.85; N, 3.96; S, 9.07%).

2.5. Solubility measurements

An aqueous solution of PZA 9 (1% w/w) containing NaBr or NaI or HCl were titrated salt-free water at 23 °C until turbidity appeared. Triplicate measurements gave critical (minimum) salt concentration (CSC) of 2.16 M NaBr, 0.581 M NaI, and 8.59 M HCl. PZA 9 was found to be practically insoluble in any possible concentration of NaCl.

2.6. Potentiometric titrations

The procedure for the determination of basicity constants (K) is described elsewhere.^26,27 A certain millimole (in terms of repeating unit (RU)) of water-insoluble PZA 9 (ZH₂^±) was dissolved by treating with 3 equivalents of NaOH using a 0.0978 M NaOH, then diluted with CO₂-free water to 200 mL, which thus contained a solution of PDE 11 in the presence of 1 equivalent excess NaOH. For water-soluble PDE 11, its solution was directly prepared in salt-free water (200 mL). The solutions were then titrated by gradual addition of 0.1222 M HCl (0.05–0.15 mL) as described in Table 2. The pH values, recorded after each addition of the titrant, were used to calculate log K_i by the Henderson–Hasselbalch eqn (2) (see eqn (1)–(3) embedded in Scheme 2). The insolubility of PZA 9 did not permit the determination of log K₃ associated with the equilibration: 9 (ZH₂^±) + H⁺ ⇋ (ZH₃⁺) 8.

Table 2 Protonation of polymer 11 (Z⁼) and 9 (ZH^±) at 23 °C in salt-free water

Run	ZH2^± (mmol)	CT^a (mol dm^−3)	α-Range	pH-Range	Points^b	log K^oi^c	n1^c	R^2^d
a (+)ve values describe titrations with HCl.b Number of data points.c Standard deviations in the last digit are given under the parentheses.d R = correlation coefficient.e log Ki = log K^oi + (ni − 1)log[(1 − α)/α].f Titration was carried out in the presence of 6.43 and 7.50 mL of added 0.0978 M NaOH, respectively, to solubilize and attain the required values of the α.g Titration was carried out in the presence of 6.43 and 7.50 mL of added 0.0978 M NaOH, respectively, to solubilize and attain the required values of the α.
Polymer 9 (ZH^±) or 11 (Z^=) in salt-free water
1	0.2007 (Z^=)	+0.1222	0.23–0.91	10.37–7.26	22	9.35	2.09	0.9942
2	0.2066^f (ZH2^±)	+0.1222	0.18–0.88	10.83–7.60	22	9.30	2.07	0.9945
3	0.2413^g (ZH2^±)	+0.1222	0.15–0.88	10.80–7.50	24	9.33	2.04	0.9950
Average						9.33(3)	2.07(3)
log K1^e = 9.33 + 1.07 log[(1 − α)/α]
1	0.2007 (Z^=)	+0.1222	0.16–0.66	6.36–4.93	17	5.40	1.45	0.9934
2	0.2066^f (ZH2^±)	+0.1222	0.12–0.75	6.60–4.74	20	5.45	1.40	0.9943
3	0.2413^g (ZH2^±)	+0.1222	0.09–0.87	6.85–4.20	23	5.49	1.46	0.9953
Average						5.45(5)	1.44(3)
log K1^e = 5.45 + 0.44 log[(1 − α)/α]

---

The log K₁ and log K₂ represent the basicity constants of the basic centers in PDE 11 (Z⁼) and polyzwitterion-anion (PZAN) 10 (ZH^±−), respectively. The degree of protonation (α) of 11 and 10 is calculated by [ZH^±−]_eq/[Z]_o, and [ZH₂^±]_eq/[Z]_o, respectively, where [ZH^±−]_eq and [ZH₂^±]_eq represent the corresponding equilibrium concentrations of the protonated species 10 and 9. [Z]_o represents the initial polymer concentration in terms of RUs.

Because of addition of 8–11 mL of 0.0978 M NaOH, which is more than two equivalents of the RUs in 9 (ZH₂^±), the polymer is converted to its dianionic form 11 (Z⁼) by neutralization with two equivalents of NaOH. By considering the excess NaOH as added OH⁻, the concentration of the protonated species 10 [ZH^±−] during the titration of 11 (Z⁼) with HCl to determine log K₁ was given by [ZH^±−]_eq = C_H⁺ − C_OH⁻ − [H⁺] + [OH⁻], where C_OH⁻ represents the concentration of the added ‘excess NaOH’. The equilibrium concentrations of [H⁺] and [OH⁻] were determined using the pH values, whereas C_H⁺ represents the concentration of added HCl during titrations. Continuing the titration, log K₂ was calculated using titrant volume after subtracting one-equivalent, from the total volume.

2.7. Evaluation of antiscalant behavior

The inhibition of CaSO₄ scale formation were evaluated using supersaturated solution of CaSO₄ containing Ca²⁺ (2600 ppm) and SO₄²⁻ (6300 ppm) in the presence of newly synthesized antiscalant 9 (x ppm) (Table 1, entry 2) (treated with minimum quantity of NaHCO₃ to dissolve it) at 40 ± 1 °C using the procedure described elsewhere.²⁸ The concentrations of the ions are three times the concentrations found in the reject brine of a reverse osmosis plant.²⁹ A rapid decrease in conductivity happened at the induction time indicating the beginning of precipitation of CaSO₄ (Table 3). Visual inspections for any turbidity were performed.

Table 3 Percent inhibition against precipitation at various times in the presence of various concentrations of the synthesized polymer 9 in a supersaturated CaSO₄ solution at 40 °C

Entry	Sample (ppm)	Percent inhibition at times (min) of							Induction time (min)
		30	60	90	120	150	500	1000
a No induction observed on the studied time range.
1	5	94	67	15	8.1	7.1	7.3	6.4	60
2	10	98	91	82	71	38	8.5	8.5	90
3	20	99	99	99	98	98	77	12	600
4	30	100	100	100	100	100	100	99	—^a

---

2.8. Phase compositions of PZA 9 (+2 equivs NaOH)–PEG–H₂O (NaCl) systems

2.8.1. The tie-lines by ¹H NMR spectroscopy. Several ATPSs of volume ≈ 7 cm³ having varying compositions of component polymers PZA 9 (+2.0 equivalents of NaOH) (Table 1, entry 2) and PEG were prepared in calibrated cylinders using their stock solutions (20 wt%) in 0.6 M NaCl and 0.6 M NaCl as the diluents. After centrifuging and equilibrating at 23 °C for 24 h, volume and density of the top and bottom layers were measured. The composition of the polymers in the phases were determined by ¹H NMR spectra of the D₂O exchanged phases in the presence of K₂CO₃ (vide infra).

2.8.2. Binodals by turbidity method. The binodal of the PZA 9 (+2 equivs NaOH)–PEG–H₂O (0.6 M NaCl) system was constructed by turbidimetric method at 23 °C as described elsewhere.³⁰

3. Results and discussions

3.1. Monomer and polymer synthesis

Dimethyl glutamate (6) upon alkylation with allyl bromide afforded dimethyl N,N-diallylglutamate (7) which upon alkaline hydrolysis followed by acidification gave monomer N,N-diallylglutamic acid hydrochloride (4).²⁵ Monomer 4 underwent AIBN-initiated cyclopolymerization with SO₂ to give CPE (+) 8 which upon depletion of HCl during soaking in water was transformed to its water-insoluble zwitterionic form: PZA (±) 9 in 87% yield (Table 1, entry 3). Increasing concentration of the initiator increases both the yields and viscosities of the polymer. The intrinsic viscosities of water-insoluble (±) 9 was measured in 0.1 M NaCl in the presence of two equivalents of NaOH which converts it to water-soluble PDE (=) 11 and are reported in Table 1. An analytical sample of 11 was prepared by treating PZA 9 (Table 1, entry 2) with excess NaOH in 91% yield. Note that both the monomer and polymers retain the unquenched nitrogen valency as well as the two carboxyl groups of the glutamic acid.

The TGA curve of PZA 9 is shown in Fig. 1; a minor loss of 6% up to 200 °C is attributed to the loss of trapped moisture. An accelerated loss of 24% in the range 200–250 °C could be accounted by the loss of the SO₂ units; note that the calculated mass of SO₂ amounts to 22.0%. The gradual loss of 48% in the range 250–800 °C may be associated with the removal of the glutamate pendant containing CO₂ units. At 800 °C, remaining mass of 22% belonged to some nitrogenated organic fraction. As can be seen from the TGA plot, the polymer remained stable up to 200 °C.

Fig. 1 TGA curve of PZA 9 (entry 2, Table 1).

Attempts to determine the molar masses of polymer 11 were unsuccessful owing to the strong adsorption aided by the chelating ligands amine and carboxy of the polymer to the materials of the GPC column.³¹

3.2. Infrared and NMR spectra

The presence of SO₂ in the backbone of polymers 9 and 11 was confirmed by the appearance of strong IR bands at ≈1305 and 1125 cm⁻¹ assigned to its asymmetric and symmetric stretching, respectively. For the dipolar motifs³² in (±) 9, the symmetric and asymmetric vibrations of COO⁻ and the C=O stretching of COOH were revealed at 1405, 1625 and 1724 cm⁻¹, respectively. The absorption bands at 1409 and 1579 cm⁻¹ were attributed to the symmetric and asymmetric stretching of COO⁻ in (=) 11, respectively.

The NMR spectra (Fig. 2 and 3) of monomer 4 and polymers 9 and 11 revealed the absence of residual double bonds in the macromolecules. The finding suggests that the chain termination happened via degradative chain transfer to the monomer³³ as well as by coupling process.³⁴ Integration of the relevant carbon signals revealed a 67 : 33 cis/trans ratio of the ring substituents at C_b,b (Scheme 1; Fig. 3c).^35,36

Fig. 2 ¹H NMR spectra of (a) 4, (b) 9 (+NaI) and (c) 11 in D₂O (referenced using signal of trimethylsilylpropionate-2,2,3,3-d₄ (TSP) at δ 0 ppm of as internal standard).

Fig. 3 ¹³NMR spectra of (a) 4, (b) 9 (+NaI) and (c) 11 in D₂O (referenced using δ 67.4 ppm of dioxane as external standard).

3.3. Solubility behavior

Like the majority of known polyzwitterions,^37–39 water-insoluble (±) 9 becomes soluble with the addition of various salts of small molar masses. For NaBr and NaI, the CSCs at 23 °C were determined to be 2.16 and 0.581 M, respectively. Iodide ions, being the most polarizable (soft), effectively neutralize the positive nitrogens so as to disrupt the intragroup, intra- and interchain ionic crosslinks and thereby, enhance solubility in water.¹⁸ Zwitterionic (±) 9 was found to be soluble in 8.59 M HCl as a result of shifting the mobile equilibrium: (±) 9 + H⁺ ⇋ (+) 8 toward right where the zwitterionic interactions required for water-insolubility vanishes. The presence of any concentration of NaCl was not able to disrupt the zwitterionic interactions and promote solubility of polymer (±) 9. This is an interesting solubility behavior which could be exploited in the development of recyclable ATPS (vide infra). Polyelectrolytes PZAN 10 and PDE 11 with charge imbalances are found to be water-soluble; anionic portion in (±−) 10 overcomes the zwitterionic interactions so as to impart water-solubility. It has been observed during the potentiometric titrations that the polymer solutions become cloudy at pH below ≈4.6, whereby the polymer backbone is calculated to have a (±) 9/(±−) 10 ratio of ≈80 : 20. Thus, increasing the zwitterionic portion to more than 80% leads to the polymer's insolubility in salt-free water.

3.4. Viscosity measurements

Polyzwitterion (±) 9 is expected to have the lowest viscosity; however it cannot be measured in the presence of usual salt NaCl since the polymer is insoluble (vide supra). Fig. 4a and b displays the viscosity plots for (±−) 10 and (=) 11 (derived from entry 2, Table 1). The viscosity plots for the polymers in salt-free water were concave upwards like any polyelectrolytes (Fig. 4a-i and ii). However, at lower concentration regime, the viscosity values fall off as a result of progressive shifting of the mobile equilibria: [(=) 11 or (±−) 10] + H₂O ⇋ [(±−) 10 or (±) 9] + OH⁻ towards right. As per general rule of hydrolysis, the degree of transformation of 11 to 10 (or 10 to 9) increases with decreasing concentration; overall decrease in the charge imbalance on the polymer chains decreases, thereby leading to lesser electrostatic repulsions hence lesser viscosity values.

Fig. 4 (a) The viscosity behavior of sample derived entry 2, Table 1 using an Ubbelohde viscometer at 30 °C: (i) ■ (±−) 10 and (ii) □ (=) 11 in salt-free water; (b) (i) □ (=) 11 and (ii) ■ (±−) 10 in 0.1 M NaCl; (c) plot for the apparent (i) log K₁ versus degree of protonation (α) (entry 3, Table 2) and (ii) log K₂ versus α for PDA 11 (derived from entry 2, Table 1) in salt-free water (entry 3, Table 2); (d) reduced viscosity (η_sp/C) at 30 °C of a 0.00858 M (i.e. 0.25 g dL⁻¹) solution of polymer PZA 9 (ZH₂^±) in salt-free water (•) versus equivalent of added NaOH. Distribution curves (dashed lines) of the various ionized species [■ 9 (ZH₂^±), □ 10 (ZH^±−), Δ 11 (Z⁼)] calculated using eqn (2) (Scheme 2) and pH of the solutions in salt-free water at 23 °C.

The higher viscosity values for zwitterionic/anionic (±−) 10 (Fig. 4a-i) than that of its dianionic counterpart (=) 11 (Fig. 4a-ii), as confirmed by careful triplicate measurements in salt-free water under N₂, is indeed unexpected. The greater repulsion among the (−)ve charges in PDE (=) 11 having the highest degree of charge asymmetry is supposed to impart higher viscosity values. A possible rationale for the higher viscosity of (±−) 10 could be attributed to the interchain H-bonding leading to the higher hydrodynamic volume as shown in Scheme 3. Note that the more distant negative oxygens of dianions (=) 11 may impart less effective repulsions. However, in 0.1 M NaCl, the interchain hydrogen bonding is discouraged because of the neutralization of the (+)ve charges in (±−) 10 by Cl⁻ ions. The viscosity plots in 0.1 M NaCl were found to be in line with the expectation (Fig. 4b-i versus Fig. 4b-ii), whereby the greater charge imbalances in (=) 11 dictating the viscosity values.

Scheme 3 Interchain H-bonding leading to increased hydrodynamic volume.

3.5. Basicity constants

The pH vs. log[(1 − α)/α] linear regression (eqn (2), Scheme 2) furnished the ‘n_i’ as the slope and log K^o_i as the intercept. The basicity constants (K_i) of the anionic centers in the polymers 9–11 is given by eqn (3) (Scheme 2) where log K^o_i = pH at α = 0.5 and n_i = 1 for sharp basicity constants. In salt-free water, log K₁ of the amine group in (=) 11, which is the pK₁ of its conjugate acid (±−) 10, and log K₂ of the terminal CO₂⁻ in (±−) 10 (i.e. pK₂ of its conjugate acid (±) 9) were determined to be 9.33 and 5.45, respectively, in salt-free water (Table 2). log K₃ (i.e. pK₃) involving the equilibrium: 9 (ZH₂^±) + H⁺ ⇋ (ZH₃⁺) 8 cannot be determined owing to the insolubility of zwitterionic (±) 9 (vide supra). Note that log K of a base is the pK_a of its conjugate acid.

The n_is, which reflect a measure of polyelectrolyte effect, are found to be greater than 1 thereby indicating a gradual decrease of basicity constants (K) with increasing degree of protonation (α) (Table 2). The n₁ and n₂, associated with K₁ and K₂, respectively, were determined to be 2.07 and 1.44, thereby reflecting a greater polyelectrolyte effect i.e. greater changes in K₁ values during the transformation of (=) 11 to (±−) 10. In salt-free water, charge centers in (=) 11 are expected to be more hydrated than in (±−) 10. The higher values of log K₁ and n₁ than log K₂ and n₂, respectively, are the consequences of the entropy driven protonation step.⁴⁰ During protonation, a repeating unit (RU) of (=) 11 would lose greater number of hydrated water molecules than that of (±−) 10. Note that polyzwitterion (±) 9, being the most compacted and least hydrated as confirmed by its insolubility in water. As a consequence, during protonation of (=) 11 and (±−) 10, the respective polymer backbones would consist of (=) 11/(±−) 10 and (±−) 10/(±−) 9 with the former having greater number of water molecules in the hydration shell of each RU. The variations of log K_is with α, shown in Fig. 4c, reflect their “apparent”^41,42 nature since instead of remaining constant, they decrease with the increase in α. It has been established⁴⁰ that the protonation process in similar cases is entropy-driven as a result of the release of water molecules from the hydration shell of the repeating unit that is being protonated. However, the exothermic enthalpy changes (ΔH^os) remain constant with increasing α.⁴⁰ Likewise, in both cases of protonation involving K₁ and K₂ for the current work, the exothermic enthalpy changes (ΔH^os) are expected to remain constant with increasing α, and the ΔG^os become less negative (i.e. less favorable) as a result of progressive decrease in the (+)ve ΔS^os since ΔG^o = ΔH^o − TΔS^o. With each protonation, the (−)ve charge density and the hydrated water molecules per RU decrease; as a consequence, a RU being protonated releases less water molecules from its hydration shell than that of the unit protonated in the previous step.⁴⁰ In other words, the continuous decrease of positive ΔS^o with α leads to lesser negative ΔG^os, hence a gradual decrease in log K values.

3.6. Viscometric titrations

The viscosity data of a 0.00858 M (i.e. 0.25 g dL⁻¹) PZA 9 in salt-free water in the presence of various concentration of added NaOH at 23 °C is reported using a solid line in Fig. 4d. The pH values and basicity constants (vide supra) were used to construct the distribution curves (dashed lines) of the specie ZH₂^± (PZA 9), ZH^±− (PZAN 10), and Z⁼ (PDE 11). The reduced viscosity, as expected, increases with the addition up to 1 equivalent of NaOH as a consequence of the transformation of ZH₂^± to ZH^±−. Maximum viscosity is achieved after the addition of 1 equivalent of NaOH whereby the concentration of ZH₂^± and ZH^±− reaches the minimum and maximum values, respectively. Further addition of NaOH, which initiates the transformation of ZH^±− to Z⁼, led to a gradual decrease in the reduced viscosities accompanied by an increase in the concentration of the ionic species Z⁼. This unusual viscosity behavior of dianionic 11 (Z⁼) having lower viscosity than its zwitterionic–anionic counterpart 10 (ZH^±−) has also been observed in Fig. 4a and a rationale for the finding is given (vide supra).

3.7. Scale inhibition by the synthesized polymer

Scaling of inorganic salts like CaSO₄ and CaCO₃ in reverse osmosis (RO) process damages membranes, thereby impeding their smooth functioning. Excessive presence of the relevant ions in the reject brine leads to their supersaturation and scaling. The percent scale inhibition (PSI) is calculated using eqn (4):where [Ca²⁺] represents the concentrations in the presence and absence of antiscalant 9 at time zero (t₀) and t.

In the current work, a supersaturated solutions of CaSO₄ containing Ca²⁺ (2600 ppm) and SO₄²⁻ (6300 ppm) in the presence of 0 (blank), 5, 10, 20, and 30 ppm of 9 as an antiscalant were examined by following the conductivity of the solutions versus time. The results of the investigation are given in Table 3. In the absence of 9, a sudden drop in conductivity indicates the precipitation of CaSO₄ (Fig. 5-iv: blank). To our satisfaction, the presence of 9 at a concentration of 30 ppm imparted a 99% scale inhibition for about 1000 min, while it was 100% at the time of 500 min as calculated using eqn (4) using conductivity values as proportional to the ionic concentrations. Note that the presence of 5, 10, 20, and 30 ppm of the antiscalant was able to register PSI of 94, 98, 99, and 100%, respectively, at a time of 30 min. These are indeed efficient PSI values since the feed water usually resides for ≈30 min in the osmosis chamber. The time, at which a sharp drop in conductivity happens, signaling the onset of quick precipitation, is known as an induction period (IP). The IPs of 60, 90 and 600 min were observed in the presence of 5, 10 and 20 ppm of 9, respectively. Note that at the concentration of 30 ppm and the time scale of 1000 min, no induction period was observed; it imparted 99% inhibition. The new antiscalant has demonstrated its efficacy in scavenging metal ions and disrupting the nucleation and crystallization processes,^43,44 hence it could be used as a potential antiscalant to minimize membrane fouling by CaSO₄ scale.

Fig. 5 Precipitation behavior of a supersaturated solution of CaSO₄ in the presence (5, 10, 15, 30 ppm) and absence of PZA 9 (treated with NaHCO₃). Inset showing the precipitation behavior in a shorter time scale of 100 min.

3.8. Phase diagrams: [PZA 9 + 2 equivs NaOH]–PEG–H₂O (0.6 M NaCl)

The polymer compositions in the phases, determined by ¹H NMR spectroscopy, were used to construct the tie-lines in the phase diagram (Table 4). A small portion of each phase was taken, and after exchanging H₂O with D₂O the NMR spectra were recorded in the presence of K₂CO₃ which helped to minimize the overlap of PEG signal at δ 3.65 with the signals of PDE 11 (i.e. 9 + 2 equivs NaOH). The proton spectrum of the bottom phase (system 1, Table 4) is displayed in Fig. 6a; the four Hs marked e, f of PDE 11 appeared in the range δ 1.6–2.2 ppm having an area integration of A, while the remaining eleven protons marked a–d in the range δ 2.4–3.6 ppm integrated for an area of B. The mole ratios of the polymers 11/PEG was determined as A/C where C is the area of the four-proton signals for PEG at δ 3.65 ppm. For the top phase (Fig. 6b), where the excessive presence of PEG led to overlapping of its signal with that of PDE 11, mole ratio of 11/PEG was calculated as [(A + C) − (11 × B/4)]/B, where (A + C) accounts for the combined areas of four-proton PEG and eleven-proton PDE 11 while C is equated to (11 × B/4). The respective molar masses of 291.3 and 44.05 g mol⁻¹ for each RU of 9 and PEG were used for calculation of the weight fractions (w) which were used to construct the tie lines as described³⁰ (Fig. 7a).

Table 4 Composition of the phases of the [POE^a + PZA^b] system (2.0 equiv. NaOH, 0.6 M NaCl) at 296 K shown in Fig. 7a

NMR method
System	Total system		Top phase		Bottom phase		Volume ratio^c
	PEG w × 100	PZA w × 100	PEG w × 100	PZA w × 100	PEG w × 100	PZA w × 100
a Poly(oxyethylene) of molar mass 35 000 g mol^−1.b 9.c Volume ratio of top and bottom phase.
1	3.07	4.46	7.70	0.101	0.128	7.14	0.634
2	3.68	3.15	6.76	0.182	0.254	6.43	1.06
3	2.48	3.63	5.56	0.282	0.354	5.83	0.683
4	2.67	2.65	4.84	0.282	0.454	5.02	1.01
5	1.94	2.81	3.66	0.557	0.831	4.13	0.619

---

Turbidity method
System	Binodal data		System	Binodal data
	PEG w × 100	PZA w × 100		PEG w × 100	PZA w × 100
a	0.332	4.25	h	2.48	0.823
b	0.457	3.55	i	2.92	0.657
c	0.603	2.97	j	3.32	0.553
d	0.897	2.37	k	3.88	0.451
e	1.38	1.74	l	4.58	0.380
f	1.70	1.40	m	6.06	0.254
g	2.05	1.09

---

Fig. 6 ¹H NMR spectrum of (a) bottom layer and (b) top layer of (System 5, Table 4) in D₂O (+K₂CO₃).

Fig. 7 (a) Phase diagram [■ and □ represent data obtained by respective NMR and turbidimetric method] at 296 K of 0.6 M NaCl containing PZA 9 (treated with 2 equivs NaOH)–PEG at 296 K; (b) correlation of phase diagram of PZA 9 (2 equiv. NaOH)–PEG–H₂O (0.6 M NaCl).

The binodal, constructed using turbidity method, is shown in Fig. 7a. The tie lines are drawn by connecting A_total, A_top and A_bottom which represents the composition of the total system, PEG-rich top phase and PZA-rich bottom phases, respectively. The ratio of the length of the tie-line segments (A_total − A_bot)/(A_total − A_top) is equated to V_top/V_bottom where V represents the volume of the phases.²¹ The composition on a tie-line therefore determines the ratio of the phase volumes.

A single- and two-phase region is demarcated by a binodal curve whose position with respect to the axes determines the economic viability for industrial application. As shown in Fig. 7a, the binodal for the ATPS is very much closer to the axes requiring only a total polymer concentrations of ≈5% for the separation of the phases. The most beneficial aspect of the environmentally friendly ATPS is the solubility of PDE (=) 11 at a higher pH, while it can be recycled²¹ at acidic pH by precipitating it in the form of (±) 9. The current ionic ATPS containing the polymer component having pH-triggerable functionalities (N and CO₂⁻) of glutamic acid is anticipated to have pH responsive behavior that would impart selectivity in bioseparation.

3.9. Correlation of the phase diagram

Diamond and Hsu⁴⁵ developed eqn (5) and (6) based on Flory–Huggins theory to check the consistency of the tie-lines.

ln K₁ = A₁(w′′₁ − w′₁) (5)

and

ln K₂ = A₂(w′′₂ − w′₂) (6)

where w′′ and w′ represent the wt% of polymer 1 (PZA 9 + NaOH) and polymer 2 (PEG), in the respective top and bottom phase. The ln K − (w′′_i − w′_i) plots with zero intercept gave the slopes A₁ and A₂ which reflect the effects of molar mass of the polymers and their interactions with water. The partition coefficients K₁ and K₂ of polymer 1 and polymer 2 were determined by their concentration ratio (C_t/C_b) in the top and bottom layer. The straight-line fits in Fig. 7b certifies the ability of the model to describe the phase behavior. Eqn (7)⁴⁶ describes the relationship between the root mean-square deviation (rmsd) with the K_exp and K_cal:where N represent the number of tie lines (Fig. 7a). The values of the parameters A_i and (rmsd)_i in Table 5 ascertain the usefulness of equations in correlating the experimental data. In this model, an extensive phase equilibrium determination is avoided, since a single phase composition would suffice to determine A₁ and A₂.

Table 5 Values for parameter A^a in eqn (5) and (6) along with the rmsd^b of the model from the experimental data of the partition coefficients

Entry	System	PZA^c		PEG^d
		A1^a	rmsd	A2^a	rmsd
a Ai is the slope of the linear regression of ln Ki versus w′′i − w′i, with zero intercept value (eqn (5) and (6)).b Root mean-square deviation.c PZA 9.d Poly(oxyethylene) of molar mass 35.0 kg mol^−1.
1	POE + PZA + 2.0 equiv. NaOH	−0.581	0.00587	0.537	3.28

---

3.10. A comparison of homopolymer 5 and copolymer 9

Several properties of poly(diallylglutamic acid) (5) and copolymer poly(diallylglutamic acid-alt-SO₂) (9) are compared in Table 6. The lower basicity constant K₁ of the copolymer 11 is attributed to the electron-withdrawing ability of the SO₂ in its RU. Since log K₁ = pK_NH⁺, the conjugate acid 10 has thus greater acidity than that of the corresponding homopolymer. Copolymer 11 is found to have much higher intrinsic viscosity (1.03 dL g⁻¹) than its corresponding homopolymer (0.160 dL g⁻¹) (Table 6). Homopolymer 5 is found to be a more effective antiscalant than copolymer 9; in the presence of 20 ppm each of 5 and 9, the scale inhibition was found to be 100 and 77%, respectively, after the elapse of 500 min. The better performance of the homopolymer 5 may be attributed to its lower molar mass; polymer with a smaller size may interfere more severely with the growth of scale by efficient poisoning of its active sites.⁴⁷

Table 6 Basicity constants K_i and scale inhibition efficiency of homo- and co-polymer

Polymer	pKNH^+^aor log K1	pKCO2H^a or log K2	[η]^b (dL g^−1)	Scale inhibition^c (%)	CSC^d
					NaCl (M)	NaBr (M)	NaI (M)	HCl (M)
a In salt-free water at 23 °C.b Intrinsic viscosity of 1–0.0625% polymer 5 (ref. 20) and 9 (Table 1, entry 2) treated with 2 equivalents NaOH in 0.1 M NaCl at 30 °C was measured with an Ubbelohde Viscometer (K = 0.005317 mm^2 s^−2).c After 500 min using 20 ppm polymer in a supersaturated solution of CaSO4 (aq.) at 40 °C.d Critical salt concentration required to promote solubility at 23 °C.e Insoluble in the presence of any concentration of NaCl.
	10.9	5.25	0.160	100	0.548	0.271	0.133	0.0104
	9.93	5.45	1.03	77	Insoluble^e	2.16	0.581	8.59

---

The solubility of the polymers differs greatly; the copolymer with much higher CSC values demonstrated stronger zwitterionic interactions (Table 6). The greater dispersion of the positive charges in the zwitterionic dipoles of 9 by electron-withdrawing SO₂ makes it less hydrated and thus more attracted to the negative end of the dipole. The solubility behavior of polymer 9 makes it an attractive component in a pH-controlled recyclable ATPS.

4. Conclusions

Monomer 4 containing residues of glutamic acid and SO₂ on alternate cyclopolymerization afforded CPE 8 [i.e. poly(4-alt-SO₂)] in excellent yields. CPE 8 on depletion of HCl during dialysis gave water-insoluble PZA 9 whose solubility is promoted by the addition of NaI, NaBr and HCl. PZA (±) 9 upon neutralization with 1 and 2 equivalents of NaOH was converted to water-soluble (±−) 10 and (=) 11, which has been used in developing an environmentally friendly recyclable ATPS. The basicity constants K_i for the ligand centres in PDE (=) 11 have been determined. The polymer has demonstrated antiscalant properties and imparted excellent CaSO₄-scale inhibition. Currently we are working on the synthesis of cross-linked resins containing residues of metal chelating glutamic acid for its use in the purification of wastewater.

Acknowledgements

The authors gratefully acknowledge the financial support provided by King Abdulaziz City for Science and Technology (KACST) through project No. AR-32-99 and the facilities provided by King Fahd University of Petroleum and Minerals.

References

1. B. S. Meldrum, J. Nutr., 2000, 130, 1007–1015.
2. R. Sapolsky, Biology and Human Behavior: The Neurological Origins of Individuality, The Teaching Company, See pages 19 and 20 of Guide Book, 2nd edn, 2005.
3. W. J. McEntee and T. H. Crook, Psychopharmacology, 1993, 111, 391–401.
4. M. G. Erlander and A. J. Tobin, Neurochem. Res., 1991, 16, 215–226.
5. I. Ali, W. A. Wani, K. Saleem and M. F. Hsieh, RSC Adv., 2014, 4, 29629–29641.
6. A. Ogunleye, A. Bhat, V. U. Irorere, D. Hill, C. Williams and I. Radecka, Microbiology, 2015, 161, 1–17.
7. I. Bajaj and R. Singhal, Bioresour. Technol., 2011, 102, 5551–5561.
8. I. L. Shih and Y. T. Van, Bioresour. Technol., 2001, 79, 207–225.
9. C. Li, D. F. Yu, A. Newman, F. Cabral, C. Stephens, N. R. Hunter, L. Milas and S. Wallace, Cancer Res., 1998, 58, 2404–2409.
10. T. Tsujimoto, J. Kimura, Y. Takeuchi, H. Uyama, C. Park and M. H. Sung, J. Microbiol. Biotechnol., 2010, 20, 1436–1439.
11. C. Park, Y. H. Choi, H. J. Shin, H. Poo, J. J. Song, C. J. Kim and M. H. Sung, J. Microbiol. Biotechnol., 2005, 15, 855–858.
12. M. R. Vakili, S. Zahmatkesh, M. V. Panahiyan and T. Jafarizadeh, Des. Monomers Polym., 2015, 18, 315–322.
13. S. Murakami and N. Aoki, Biomacromolecules, 2006, 7, 2122–2127.
14. X. Wen, E. F. Jackson, R. E. Price, E. Edmund Kim, Q. Wu, S. Wallace, C. Charnsangavej, J. G. Gelovani and C. Li, Bioconjugate Chem., 2004, 15, 1408–1415.
15. Z. Yu, C. Peng, Y. Luo, J. Zhu, C. Chen, M. Shen and X. Shi, RSC Adv., 2015, 5, 76700–76707.
16. Y. C. Rajan, B. S. Inbaraj and B. H. Chen, RSC Adv., 2015, 5, 15126–15139.
17. G. B. Butler, Cyclopolymerization and Cyclocopolymerization, Marcel Dekker, New York, 1992.
18. S. Kudaibergenov, W. Jaeger and A. Laschewsky, Adv. Polym. Sci., 2006, 201, 157–224.
19. P. K. Singh, V. K. Singh and M. Singh, e-Polym., 2007, 030, 1–34.
20. Z. A. Jamiu, H. A. Al-Muallem and S. A. Ali, Des. Monomers Polym., 2016, 19, 128–137.
21. P. Albertsson, Partition of Cell particles and Macromolecules, Wiley, N.Y., 3rd edn, 1986.
22. M. Waziri, B. F. Abu-Sharkh and S. A. Ali, Biotechnol. Prog., 2004, 20, 526–532.
23. Z. A. Jamiu, T. A. Saleh and S. A. Ali, RSC Adv., 2015, 5, 42222–42232.
24. S. A. Ali, O. C. S. Al-Hamouz and N. M. Hassan, J. Hazard. Mater., 2013, 248(249), 47–58.
25. S. A. Ali, I. W. Kazi and N. Ullah, Ind. Eng. Chem. Res., 2015, 54, 9689–9698.
26. S. A. Ali, S. A. Haladu and H. A. Al-Muallem, J. Appl. Polym. Sci., 2014, 131, 40615.
27. S. A. Ali, N. Y. Abu-Thabit and H. A. Al-Muallem, J. Polym. Sci., Part A: Polym. Chem., 2010, 48, 5693–5703.
28. S. A. Haladu and S. A. Ali, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 5130–5142.
29. F. H. Butt, F. Rahman and U. Baduruthamal, Desalination, 1995, 103, 189–198.
30. S. A. Ali, H. A. Al-Muallem and M. A. J. Mazumder, J. Chem. Eng. Data, 2013, 58, 2574–2585.
31. F. Rullens, M. Devillers and A. Laschewsky, Macromol. Chem. Phys., 2004, 205, 1155–1166.
32. J. F. Pearson and M. A. Slifkin, Spectrochim. Acta, Part A, 1972, 28, 2403–2413.
33. G. B. Butler and R. J. Angelo, J. Am. Chem. Soc., 1957, 79, 3128–3131.
34. R. M. Pike and R. A. Cohen, J. Polym. Sci., 1960, 44, 531–538.
35. W. Jaeger, J. Bohrisch and A. Laschewsky, Prog. Polym. Sci., 2010, 35, 511–577.
36. V. D. Vynck and E. J. Goethals, Macromol. Rapid Commun., 1997, 18, 149–156.
37. T. A. Wielema and J. B. F. N. Engberts, Eur. Polym. J., 1987, 23, 947–950.
38. J. C. Salamone, W. Volksen, A. P. Olson and S. C. Israel, Polymer, 1978, 19, 1157–1162.
39. M. Yoshizawa and H. Ohno, Chem. Lett., 1999, 28, 889–890.
40. R. Barbucci, M. Casolaro, P. Ferruti and M. Nocentini, Macromolecules, 1986, 19, 1856–1861.
41. R. Barbucci, M. Casolaro, N. Danzo, V. Barone, P. Ferruti and A. Angeloni, Macromolecules, 1983, 16, 456–462.
42. R. Barbucci, M. Casolaro, M. Nocentini, S. Correzzi, P. Ferruti and V. Barone, Macromolecules, 1986, 19, 37–42.
43. H. David, S. Hilla and S. Alexander, Ind. Eng. Chem. Res., 2011, 50, 7601–7607.
44. R. J. Davey, The Role of Additives in Precipitation Processes, Industrial Crystallization, 81, ed. S. J. Jancic and E. J. de Jong, North-Holland Publishing Co., 1982, pp. 123–135.
45. A. D. Diamond and J. T. Hsu, AIChE Symp. Ser., 1992, 290, 105–111.
46. S. Shahriari, S. G. Doozandeh and G. Pazuki, J. Chem. Eng. Data, 2012, 57, 256–262.
47. D. Hasson, H. Shemer and A. Sher, Ind. Eng. Chem. Res., 2011, 50, 7601–7607.

---