European Polymer Journal 47 (2011) 1113–1123

Contents lists available at ScienceDirect

European Polymer Journal

journal homepage: www.elsevier .com/locate /europol j

Phosphonobetaine/sulfur dioxide copolymer by Butler’scyclopolymerization process

N.Y. Abu-Thabit, Izzat W. Kazi, Hasan A. Al-Muallem, Shaikh A. Ali ⇑Center of Research Excellence in Petroleum Refining and Petrochemicals and Chemistry Department, King Fahd University of Petroleum and Minerals,Dhahran 31261, Saudi Arabia

a r t i c l e i n f o

Article history:Received 14 September 2010Received in revised form 31 January 2011Accepted 5 February 2011Available online 2 March 2011

Keywords:PolyelectrolytePolybetainePolyphosphonobetainepH-responsiveStimuli responsiveSalt effects

0014-3057/$ - see front matter � 2011 Elsevier Ltddoi:10.1016/j.eurpolymj.2011.02.004

⇑ Corresponding author. Tel.: +966 3 860 3830; faE-mail address: shaikh@kfupm.edu.sa (S.A. Ali).URL: http://faculty.kfupm.edu.sa/CHEM/shaikh/

a b s t r a c t

The zwitterionic monomer, ethyl 3-(N,N-diallylammonio)propanephosphonate and sulfurdioxide were cyclocopolymerized in DMSO using azobisisobutyronitrile or ammonium per-sulfate as initiators to afford a pH-responsive polyphosphonobetaine/SO2 (PPB/SO2)copolymer. The polymers, on treatment with HCl and NaOH, gave the aqueous solutionsof the corresponding cationic polyphosphononic acid (CPP) and anionic polyphosphonate(APP). The solution properties of the PPB having two pH-responsive functionalities wereinvestigated in detail by potentiometric and viscometric techniques. Basicity constants ofthe amine and phosphonate groups in APP were found to be ‘‘apparent’’ and as such followthe modified Henderson–Hasselbalch equation. The incorporation of SO2 moiety hasresulted in the decrease of basicity constant of the nitrogens in the copolymer by stagger-ing P2 units of log K in compare to that of the corresponding homopolymer. The basicitydifference is expected to have an effect on the chelating properties of the polymers. In con-trast to many polycarbo- and -sulfobetaines, the PPB was all found to be soluble in salt-freewater as well as in salt (including Ca2+ and Li+)-added solutions. The PPB demonstrated‘antipolyelectrolyte’ viscosity behavior and found to have higher viscosity values in LiClthan in NaCl or NaI.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

The polymerization of monomeric carbo-, phospho- andsulfo-betaines (zwitterions) (M±) having charges of bothalgebraic signs in the same molecular framework lead tothe corresponding polybetaines (polyzwitterions) [1–3].While biopolymers like proteins or DNA mediate life pro-cesses, commercial polyampholytes (having the presenceof both M+ and M� in the same polymer chain) and polyz-witterions, whose structure and behavior seem to mimicbiopolymers, have offered many new applications in med-icine, biotechnology, hydrometallurgy and oil industry[1,4,5]. The polyzwitterions are usually derived from free

. All rights reserved.

x: +966 3 860 4277.

(S.A. Ali).

radical polymerization of acrylamide and acrylate-basedcarbo- and sulfo-betaine monomers [6], while Butler’s cyc-lopolymerization reaction [7] of zwitterionic diallylammo-nium monomers or their copolymerizations with sulfurdioxide has been an attractive method for the synthesisof polycarbo- and sulfo-betaines [8–13]. Most of theknown polyphosphobetaines, (containing dialkylphos-phate motif: a derivative of phosphoric acid, H3PO4), arepolymeric phospholipid analogues [14]. To our knowledge,there are only three reports [15–17] that describe thehomocyclopolymerization of phosphorous acid (H3PO3)derivatives like diallylaminomethyl- and propyl-phos-phonic acids to polyphosponobetaines (PPB). A novel phos-phonic acid monomer containing methacrylate andmethylacrylamide motifs has been prepared for use in den-tal composites [18]. Proton conducting ionomers contain-ing phosphonic acid motif has been used for possibleapplication in polymer electrolyte fuel cells [19,20].

1114 N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123

Extraordinary chelating properties of compounds con-taining amine and phosphonate groups in the moleculehave attracted considerable attention to synthesize lowmolecular-weight chelating agents that should be ableto form polymer-heavy metal ion complexes from wastewater [21–23]. In pursuit of tailoring pH-responsive poly-mers, we continue to synthesize the cyclocopolymer 4 ofzwitterionic monomer ethyl 3-(N,N-diallylammonio)prop-anephosphonate (1) and SO2 (Scheme 1). To the best ofour knowledge, phosphonobetaine (PPB)/sulfur dioxide

Scheme

cyclocopolymer would represent the first example of thisclass of ionic polymers. The incorporation of electron-withdrawing SO2 moiety in the polymer backbone isexpected to deplete electron density at the nitrogen,hence decrease its basicity. This would indeed impartremarkable differences in the chelating properties of (1)/SO2 cyclocopolymer 4 versus the corresponding homocy-clopolymer of (1) [17]. We intend to measure and com-pare the apparent basicity constants of the co- andhomo-polymers.

1.

N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123 1115

2. Experimental

2.1. Materials

2,20-Azobisisobutyronitrile (AIBN) from Fluka ChemieAG (Buchs, Switzerland), was purified by crystallizationfrom a chloroform-ethanol mixture. Ammonium persulfate(APS) from BDH Chemical Co. (Poole, UK) was used as re-ceived. Ethyl 3-(N,N-diallylammonio)propanephosphonatewas prepared as described [17]. Dimethylsulfoxide (DMSO)was dried over calcium hydride overnight and thendistilled under reduced pressure at a boiling point of64–65 �C (4 mm Hg). All glassware were cleaned usingdeionized water. For dialysis, a Spectra/Por membranewith a molecular weight cut-off (MWCO) value of 6000–8000 was purchased from Spectrum Laboratories, Inc.

2.2. Physical methods

Melting points were recorded in a calibrated Electro-thermal-IA9100-Digital Melting Point Apparatus. Elemen-tal analysis was carried out on a Perkin Elmer ElementalAnalyzer Series 11 Model 2400. IR spectra were recordedon a Perkin Elmer 16F PC FTIR spectrometer. 1H, 13C and31P NMR spectra were measured in CDCl3 (using TMS asinternal standard) or D2O at +25 �C (using HOD peak at d4.65 and the 13C peak of dioxane at d 67.4 ppm as internalstandard) on a JEOL LA 500 MHz spectrometer. 31P spectrawere referenced with 85% H3PO4 in DMSO.

A gentle stream of N2 was passed through distilleddeionized water at 90 �C for 15 min in order to remove dis-solved gases. This water was used for viscosity measure-ments in salt-free and salt (NaCl, 99.9% purity) solutions.Viscosity measurements were made by Ubbelohde viscom-eter (having Viscometer Constant of 0.005718 cSt/s at alltemperatures) using CO2-free water under N2 in order toavoid CO2 absorption that may affect the viscosity data.Molecular weights of some synthesized polymers weredetermined by GPC analysis using Viscotek GPCmax VE2001. The system was calibrated with nine polyethyleneoxide monodispersed standards at 30 �C using two Viscotekcolumns G5000 and G6000 in series.

Table 1Cyclocopolymerization of the monomer 1 with sulfur dioxidea.

Entry no. Initiator (mg) Temp (�C) Time (h) Yield

1 AIBN (100) 65 24 Trace2 AIBN (75) 65 24 Trace3 AIBN (75) 63 24 275 AIBN (75) 57 48 626 AIBN (60) 57 48 424 AIBN (48) 57 48 507 AIBN (100) 50 48 668 AIBN (75) 45 48 Trace9 APS (130) 45 3 5410 APS (100) 45 48 5611 APS (60) 45 48 19

a Polymerization reactions were carried out using 15 mmol each of the mono(AIBN) or ammonium persulfate (APS).

b Viscosity of 1–0.25% polymer solution in salt-free and 0.1 N NaCl at 30 �C wc Polydispersity index.

2.3. General procedure for the copolymerization of themonomer 1 with SO2

All the polymerizations were carried out using condi-tions as described in the Table 1. In a typical experiment,SO2 (15 mmol) was absorbed in a solution of the monomer1 (15 mmol) in DMSO (3.75 g). The required amount of theinitiator AIBN or APS (as listed in Table 1) was then addedunder N2 and the closed flask was stirred using a magneticstir-bar at the specified temperature and for duration aslisted. The polymer was dissolved in water and dialyzedagainst deionized water for 48 h. The resulting solutionof PBB 2 was freeze dried to obtain the white polymer,which was stored in a desiccator.

2.4. Basification of PPB 2 to APP 3

To a solution of PPB 2 (entry 7, Table 1) (0.623 g,2 mmol) in water (7 cm3) in a dialysis bag was added asolution of NaOH (0.240 g, 6 mmol) in water (7 cm3). Thesolution in the dialysis bag was briefly shaken and thendialyzed against distilled deionized water for 24 h. Theresulting solution of APE 3 was freeze dried to obtain thewhite polymer (0.587 g, 88%), which was stored in a desic-cator. The onset of thermal decomposition (closed capil-lary) 245–255 �C (decomposed, turned black); (found: C,39.3; H, 6.5; N, 4.1; S, 9.4. C11H21NNaO5PS requires C,39.64; H, 6.35; N, 4.20; S, 9.62%); mmax (KBr) 3423, 2978,2921, 1647, 1458, 1417, 1307, 1173, 1129, 1039, 945,and 791 cm�1. dP (202 MHz, D2O): 25.45. 1H and 13C NMRspectra are shown in Figs. 1 and 2, respectively.

2.5. Thermal decomposition, elemental analyses, infraredspectra, 31P NMR, molecular weights (MW) and polydispesityindex (PDI) of polymer PPB 2

The onset of thermal decomposition (closed capillary)240–250 �C (decomposed, turned black); (found: C, 42.2;H, 7.2; N, 4.40; S, 10.0. C11H22NO5PS requires C, 42.44; H,7.12; N, 4.50; S, 10.30%); mmax (KBr) 3361, 2937, 1467,1296, 1158, 1125, 1026, 941, and 778 cm�1. dP (202 MHz,D2O): 25.79 and a minor (5%) signal at 25.03 ppm.

(%) Intrinsic Viscosityb (dl g�1) MW (PDI)c

Salt-free 0.1 N NaCl

– – – –– – – –0.0205 0.0243 – –0.0791 0.0957 – –0.0225 0.0268 – –0.0785 0.0973 67,300 2.30.0802 0.0992 72,100 2.2– – – –0.0761 0.0906 67,100 2.30.0733 0.0916 – –0.0801 0.0977 – –

mer and SO2 in DMSO (3.75 g) in the presence of azobisisobutyronitrile

as measured with a Ubbelohde Viscometer (K = 0.005718).

Fig. 1. 1H NMR spectrum of monomer 1, PBB 2 and APP 3 in D2O.

1116 N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123

For the determination of molecular weights, the polymerPPB 2 was analyzed using an aqueous solution of 0.1 NNaNO3 as the eluant. Refractive Index and viscometer detec-tors were used to detect polymers. The molecular weight(MW) and PDI of some of the polymers are given in Table 1.

2.6. Potentiometric titrations

The potentiometric titrations were carried out at 23 �Cas described elsewhere [24]. For each titration, 200 cm3

of salt-free or 0.1 N NaCl solution containing a weighedamount (usually around 0.15–0.27 mmol of the monomeror repeating unit) of PB 1 or PPB 2 was used. The solutionwas titrated at 23 �C with �0.1 M HCl (or �0.1 M NaOH)delivered by a buret under N2. After each addition of the ti-trant (in the range 0.10–0.25 cm3) the solution was stirredbriefly, using a magnetic stir bar, under N2. The pH of thesolution was recorded using a Corning pH Meter 220.

The protonation constant of the amine nitrogen, log K1, iscalculated at each pH value by the well-known Henderson–Hasselbalch Eq. (1) (Scheme 2) where degree of protonation(a) is the ratio [ZH±]eq/[Z]0. For the titration with NaOH, the[Z]0 is the initial analytical concentration of the monomericunits in the PPB 1 and [ZH±]eq is the concentration ofthe protonated species at the equilibrium given by

[ZH±]eq = [Z]0 � C�OH � [H+] + [OH�], where C�OH is theconcentration of the added NaOH; [H+] and [OH�] at equi-librium were calculated from the pH value [25].

The typical electrolytes having apparent basicity con-stants could be described by the Equation 2 (Scheme 2)where log K0 = pH at a = 0.5 and n = 1 in the case of sharpbasicity constants. The linear regression fit of pH vs. log[(1-a/a)] gave log K0 and ‘n’ as the intercept and slope,respectively. Simultaneous protonation of the two basicsites is least likely since the basicity constant for thePO(OEt)O� group is less than that of the amine group byat least seven orders of magnitude. The second step pro-tonation constant (log K2) involving the PO(OEt)O� is cal-culated by titration of the PB 1 or the monomeric units inthe PPB 2 with HCl. In this case, a represents the ratio[ZHþ2 ]eq/[Z]0 whereby [ZHþ2 ]eq equals CþH � [H+] + [OH�]and CþH is the concentration of the added HCl.

The experimental details of the potentiometric titra-tions are summarized in Tables 2 and 3. Inserting the valueof pH from Eq. (2) into Eq. (1) leads to modified Hender-son–Hasselbalch equation (Eq. (3)) [26,27] (Scheme 2)where (n – 1) gives a measure of the deviation of the stud-ied polymers from the behavior of small molecules show-ing sharp basicity constants (for molecules having sharpbasicity constants, n becomes 1).

Fig. 2. 13C NMR spectrum of monomer 1, PBB 2 and APP 3 in D2O.

Scheme 2.

N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123 1117

3. Results and discussion

3.1. Synthesis and physical characterization of the polymers

The results of the polymerization under various condi-tions are given in Table 1. As evident from the table,decreasing the temperature led to an increase in the yieldof the polymer 2. While no polymerization occurred at65 �C (entries 1 and 2), the highest yield and viscositywas obtained at 50 �C (entry 7). The results thus indicatethe presence of a ceiling temperature for the polymeriza-tion. Changing the initiator from AIBN to ammonium per-sulfate (APS) (entries 9–11) did not improve the yield.The low yield and molecular weight as indicated by lowerintrinsic viscosity values (Table 1) could be a result ofchain transfer between polymer and monomer throughthe alkoxy group attached to the phosphorous atom [28].

PPB 2 was found to be stable up to 240 �C and thereafterdecomposes as result of elimination of SO2. The PPB 2 wasfound to be readily soluble in water as well as in protic sol-vents having higher dielectric constants (Table 4); hydrogenbonding with the protic solvents is responsible for the solu-bility behavior. Nonprotic solvents like DMF and DMSO hav-ing high dielectric constants failed to dissolve the polymer.Polyzwitterions are in general expected to be insoluble inwater. However, there are quite a few exceptions. A consid-erable number of polycarbo- and sulfobetaine are reportedto be soluble in salt-free as well as salt (NaCl) added solu-tions [29]. It is reported that carbobetaines having carboxylpKa values of62.0 and >2 are found to be insoluble and sol-uble, respectively, in salt-free water [24,30,31]. It is ex-plained that the negative charges in carbobetaines, havingapparent carboxyl pKa values >2.0, are expected to be lessdispersed, hence more hydrated. The higher degree ofhydration thus led to exhibit weaker Coulombic interactionswith the cationic charges [32] thereby resulting in their sol-ubility in salt-free water. Steric encumbrance around thecharges also prevents effective intra- or intermolecular Cou-lombic interactions, thereby leading to the solubility ofpolyzwitterions in salt-free water [8,33]. It is worth men-tioning that the temperature and molecular weight alsomay dictate the solubility behavior of polybetaines [34].Since the phosphonate [O@P(OEt)O�] pKa in PPB 2 is deter-mined to be >2, the polyphosphobetaine is expected to besoluble in salt-free as well as salt-added water (vide infra).

Table 2Experimental details for the protonation of monomer 1 (ZH±), and polymer PPB 2 (ZH±) at 23 �C in salt-free water.

Run ZH± or Z�(mmol) CTa (mol dm�3) a�Range pH�range Pointsb

Log K0i

c nic R2d

Polymers PPB 2 (ZH±) in salt-free water1 0.1783 (ZH±) �0.1001 0.78–0.19 7.10–10.36 16 8.57 2.80 0.99752 0.2104 (ZH±) �0.1001 0.76–0.18 7.04–10.45 18 8.54 2.78 0.99533 0.2457 (ZH±) �0.1001 0.78–0.22 7.06–10.15 19 8.64 2.82 0.9978Average 8.58 (5) 2.80 (2)LogKe

1 ¼ 8:58þ 1:80 log½ð1� aÞ=a�; for the reaction : Z� þHþ� ZH�

Monomer 1 (ZH±) in salt-free waterLogKe

1 ¼ 8:99; for the reaction : Z� þHþ� ZH�

Polymers PPB 2 (ZH±) in salt-free water1 0.2524 (ZH±) +0.09930 0.18–0.47 2.98–2.56 21 2.53 0.68 0.99132 0.1908 (ZH±) +0.09930 0.19–0.51 2.98–2.54 18 2.57 0.66 0.99223 0.1626 (ZH±) +0.09930 0.14–0.42 3.08–2.61 15 2.51 0.74 0.9978Average 2.54 (3) 0.69 (4)LogKe

2 ¼ 2:54� 0:46 log½ð1� aÞ=a�; for the reaction : ZH� þ Hþ�ZHþ2Monomer 1 (ZH±) in salt-free waterLogKe

2 ¼ 2:39; for the reaction : ZH� þ Hþ�ZHþ2

a Titrant concentration (negative and positive values indicate titrations with NaOH and HCl, respectively).b Number of data points from titration curve.c Values in the parentheses are standard deviations in the last digit.d R = correlation coefficient. elog Ki = log K0

i + (n � 1) log [(1 � a)/a].

Table 3Experimental details for the protonation of monomer 1 (ZH±), and polymer PPB 2 (ZH±) at 23 �C in 0.1 M NaCl.

Run ZH± or Z�(mmol) CTa (mol dm�3) a�Range pH�range Pointsb

Log K0i

c nic R2d

Polymers PPB 2 (ZH±) in 0.1 N NaCl1 0.1905 (ZH±) �0.1001 0.83–0.14 6.18–9.01 15 7.63 1.94 0.99372 0.2361 (ZH±) �0.1001 0.79–0.18 6.52–8.85 15 7.66 1.91 0.99733 0.2923 (ZH±) �0.1001 0.79–0.15 6.51–9.02 20 7.65 1.87 0.9988Average 7.65 (2) 1.91 (4)LogKe

1 ¼ 7:65þ 0:91 log½ð1� aÞ=a�; for the reaction : Z� þHþ� ZH�

Monomer 1 (ZH±) in 0.1 N NaClLogKe

1 ¼ 8:74; for the reaction : Z� þHþ� ZH�

Polymers PPB 2 (ZH±) in 0.1 N NaCl1 0.3244 (ZH±) +0.0993 0.12–0.57 3.31–2.66 22 2.75 0.66 0.99462 0.2634 (ZH±) +0.0993 0.14–0.63 3.30–2.68 22 2.78 0.63 0.98933 0.1792 (ZH±) +0.0993 0.13–0.48 3.35–2.73 20 2.74 0.71 0.9914Average 2.76 (2) 0.67 (4)LogKe

2 ¼ 2:76� 0:33 log½ð1� aÞ=a�; for the reaction : ZH� þ Hþ�ZHþ2Monomer 1 (ZH±) in 0.1 N NaClLogKe

2 ¼ 2:52; for the reaction : ZH� þ Hþ�ZHþ2

a Titrant concentration (negative and positive values indicate titrations with NaOH and HCl, respectively).b Number of data points from titration curve.c Values in the parentheses are standard deviations in the last digit.d R = correlation coefficient. elog Ki = log K0

i + (n � 1) log [(1 � a)/a].

1118 N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123

3.2. Infrared and NMR spectra

The strong absorptions at 1158 and 1026 cm�1 in the IRspectrum of PPB 2 were attributed to the stretching fre-quency of P@O and PAOAC, respectively. 1H and 13CNMR spectra of monomer PB 1 and its polymers PPB 2and APP 3 are displayed in Figs. 2 and 3, respectively.The proton signals of 2 are shifted downwards in compar-ison with that of 3 as a result of increased electronegativityof the positive nitrogen in the former. The absence of anyresidual alkene proton or carbon signal in the spectra alsosuggested the chain transfer process [35] for the termina-tion reaction. Because of the width and overlap of the

peaks in the spectrum, the size of the ring can not be deter-mined from the 1H NMR spectrum.

The assignments of the 13C peaks are based on earlierworks, which confirmed the formation of kinetically fa-vored five-membered ring structure [31,36–39]. The sub-stituents at the C-b of the PPB 2 can either be in thesymmetrical cis (major) or unsymmetrical trans (minor)dispositions (Scheme 1). The cis/trans ratio of 75/25 wasdetermined by integration of the relevant peaks in the13C spectrum. The ratio was found to be similar to that ob-served for the related polymers derived from cyclopoly-merization of quaternary diallyl ammonium ammoniumsalts [31,40]. The cis substituents at C-b of the cis form

Table 4Solubilitya,b of PPB 2 (entry 7, Table 1).

Solvent Dielectric constant, e Solubility

Formamide 111.0 +Water 78.4 +Formic acid 58.5 +DMSO 47 �Ethylene glycol 37.3 �DMF 37 �Methanol 32.7 �Diethyleneglycol 31.7 �Ethanol 24.5 �Acetone 21 �Acetic acid 6.1 �

a 2%(w/w) of polymer solution was made after heating the mixture at70 �C for 1 h and then cooling to 23 �C.

b ‘+’ Indicates soluble; ‘�’ indicates insoluble.

N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123 1119

can exist in two structural forms; the -(CH2)3PO(OEt)group can either be trans- or cis- oriented with respect tothe cis substituents at C-b (Scheme 3). The presence ofthe two major cis forms of 2 in unequal proportions thusleads to the unequal splitting of the signals (acis, bcis,Fig. 2). The C-a as well as C-b signals of 2 also show twoweak lines of equal intensity in each case for the minortrans (atran, btrans, Fig. 2) which can exist solely in one formin which case C-a and C-a0 or C-b and C-b0 becomes mag-netically and chemically non equivalent. However, it isinteresting to note that the polymer 3 have trivalent nitro-gen that undergoes fast nitrogen lone pair inversion in theNMR time scale, and as such the A(CH2)3PO(OEt) grouplooses its stereochemical integrity, and the splitting ofthe carbon signals in the cis or trans forms is not observed(Fig. 2, Scheme 3). The correctness of the assignment ofcarbon signals was confirmed by DEPT experiment usingpolymer 3 at 135�; the carbon signals bcis and btrans were

Fig. 3. The intrinsic viscosity [g] behavior of PPB 2 (entry 7, Table 1) in the preseViscometer.

ascertained to be that of the methine (CH) carbons. TheDEPT experiment thus ruled out the possible existence ofthe highly unsymmetrical six-membered rings in structureA as the predominant repeating unit (Scheme 3). In such ascenario, the methine carbons C-b and C-e would have ap-peared at �d 35 and �65 ppm, respectively [41]. The sig-nal-to-noise ratio achieved in the 13C spectra ofpoly(diallyldimethylammonium chloride) set an upperlimit of 2–3% for possible existence of any six-memberedring piperidine structure [37]. The report [41] on the struc-ture of cyclocopolymers of dialkyldiallyl ammonium saltswith SO2 have established the existence of predominant,if not exclusive, five-membered pyrrolidine rings. Detec-tion of structure A and its quantification are further com-plicated by the possibility of cis- and trans- orientationsof the substituents at C-b and C-e. DEPT experiment of 3revealed the presence of a methylene and a methine (CH)carbon at d40.1 and 50.8 ppm, respectively, which couldpresumably be assigned to the C-b and C-a of the piperi-dine ring in structure A. The six-membered ring structuremay thus be present to the extent of 3% as approximatedby integration.

3.3. Viscosity measurements

Viscosity data were evaluated by the Huggins equation:gsp/C = [g] + k [g]2 C. Unlike in the case of a polyelectrolyte,the Huggins viscosity plot for PPB 2 in the absence orpresence of added salt (NaCl) remains linear. The intrinsicviscosity, slope and Huggins constant are included inTables 1 and 5. ‘‘Anti-polyelectrolyte’’ behavior of the PPB2 is manifested as a result of increased intrinsic viscosityin the presence of an added salt (Fig. 3, Table 5). Note thatfor a common Na+ cation, the intrinsic viscosity increasesas the anion is changed from Cl� to I� in the presence of

nce of varying concentrations of added salts at 30 �C using an Ubbelohde

Scheme 3.

1120 N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123

0.1 N salts (Fig. 3, Table 5). This is expected since the iodideanion is more polarizable (soft), hence it is particularlyeffective in neutralizing the cationic charges on the poly-mer backbone. A decrease in the value for the Huggins con-stant k in the presence of NaI in comparison to NaCl maybe associated with an increased polymer–solvent interac-tion. However, for a common anion Cl�, viscosity increasesas the cation is changed from Na+ to Li+. The hydrationshell of Li+, having a larger charge/radius ratio, is largerthan that of Na+, and as such makes it less effective thanNa+ in shielding the phosphonate anions. The increasedrepulsions among the more exposed anionic pendants inthe presence of LiCl lead to a higher viscosity value owingto an increase in the hydrodynamic volume. Contrary to

Table 5Effect of added salts on the intrinsic viscosity ([g]a and Huggins constant(k0) of PPB 2 (entry 7, Table 1).

Solvent Slope [g] k

Salt-free H2O 0.0175 0.0802 2.72

0.1 N NaCl 0.0052 0.0992 0.5280.5 N NaCl 0.0136 0.106 1.211.0 N NaCl 0.0119 0.0985 1.22

0.1 N NaI 0.0046 0.107 0.4020.5 N NaI 0.0084 0.0984 0.8671.0 N NaI 0.0158 0.0975 1.66

0.1 N LiCl 0.0215 0.108 1.840.5 N LiCl 0.0146 0.113 1.141.0 N LiCl 0.0069 0.114 0.531

0.1 N CaCl2 0.0096 0.103 0.905

0.1 N HCl 0.0050 0.0907 0.607

0.1 N CsCl 0.0156 0.103 1.47

a Viscosity of 1–0. 25% salt-added polymer solution at 30 �C wasmeasured with a Ubbelohde Viscometer (K = 0.005718).

earlier reports on polycarbo- and sulfobetaines [8,40,42],it is interesting to note that the viscosity values were foundto be higher in the presence of LiCl than NaCl or even NaI(Fig. 3). At higher salt concentrations, the increase in [g]either tends to level off or decreases a little (in the pres-ence of NaCl). Upon completion of the site and atmosphericbinding [42] of the cationic charges on the nitrogens byCl�, the exposed anionic phosphonate pendents in sea ofNa+ ions are also expected to be shielded to a certain de-gree, thereby leading to a decrease in [g] values at thehigher end of NaCl concentrations. The divalent cationCa2+, which is usually known to precipitate anionic poly-electrolytes from aqueous solution even at low concentra-tions, did not precipitate the PPB 2.

The changes in reduced viscosities of a 1.0 g/dL solutionof the polymer (entry 7, Table 1) in salt-free water at 30 �Cin the presence various amounts of HCl and NaOH demon-strates the pH-responsiveness of PPB 2 as displayed inFig. 4. Reduced viscosity does not change with the additionof HCl; the amount of added HCl is not sufficient enough toshift the equilibrium (PBB) 2 � (CPP) 4 towards the CPP.The presence of increasing proportion of the protonatedform 4 is expected to increase the viscosity as a result ofincreased repulsions among the unshielded cationiccharges on the nitrogens. A steep increase in the viscosityvalues is observed on the addition of NaOH which changesthe polymer PPB 2 to the zwitterionic/anionic polymer(PPB/APP) 5 and finally to APP 3. The anionic motifs in 5provoke an extension of the polymer backbone owing tothe increased repulsion among negative charges. A de-crease in the viscosity upon addition of greater than 1equivalent of NaOH could be attributed to the effect ofincreasing ionic strength that can shield the anionicpendants.

Fig. 4. Reduced viscosity (gsp/C) of a 1.0 g/dL solution of polymer PPB 2 (entry 7, Table 1) in salt-free water versus equivalent of added NaOH at 23 �C.Negative (�) values indicate the equivalent of added HCl.

N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123 1121

Viscosity data for PPB 2 (entry 7, Table 1) in neutralmedium and in the presence of 1 equivalent of HCl andNaOH are presented in Fig. 5. In the absence of added salt(NaCl), the plot for (PPB 2 + 1 equiv NaOH) i.e. APP 3 sys-tem is concave upwards, a typical case for a polyelectro-lyte; the reduced viscosity increases slightly withdecreasing concentration of the polymer. Note that the dy-namic equilibrium: APP 3 + H2O � PPB/APP 4 + OH� inaqueous solution via protonation of the weakly basic nitro-gens in 3 is expected to move more and more towards theright as the concentration of the polymer decreases. As aresult, at higher dilution, a decrease in the reduced viscos-ity values could be attributed to the increasing proportionof zwitterionic motifs in 4.

3.4. Basicity constants

In order to gain further information on the conforma-tional transitions, we have determined the basicity con-stants, K0

i , and the corresponding ni values relative to theprotonation of the tertiary amine and carboxylate in poly-mer APP 3 and monomer 1 (NaOH-treated) in salt-freewater and 0.1 N NaCl solution. Basicity constants and thecorresponding ni values as well as the experimental detailsof the potentiometric titrations are given in Tables 2 and 3.

The basicity constants log K1 for the protonation of thetrivalent nitrogen of this polymer are found to be ‘‘appar-ent’’ [26,27,43] in salt-free as well as 0.1 N NaCl solutionsince the n values are not equal to 1; n values were foundto be 2.80 and 1.91, respectively. The behavior of variationof log K1 with the degree of protonation (a) is known aspolyelectrolyte effect; polymer exhibiting a strong poly-electrolyte effect shows a large change in log K in goingfrom one end to the other end of the titration. The magni-tude of the n reflecting the polyelectrolyte effect is an in-dex of accessibility of the proton to the amine nitrogenduring the protonation reaction. The basicity constant(log K0

1) of the amine group was found to be 8.58 and7.65 in salt-free water and 0.1 N NaCl, respectively (Tables

2 and 3). Since the n1 values are greater than 1, theapproach of the incoming protons to the amine nitrogensbecomes more and more difficult with the increasingdegree of protonation (a). The decrease in the overallnegative charge density in the macromolecule decreasesthe electrostatic field force that attracts the proton therebydecreases basicity constant (log K) with the increase indegree of protonation (a).

Protonation of polymers is dictated by entropy effectsassociated with the release of water molecules from thehydration shell of the repeating unit that is being proton-ated [43]. Comparison of Fig. 5a vs. 5b reveals that APP 3(i.e. PPB 2 + 1 equiv NaOH) in salt-free water is more ex-tended as indicated by higher viscosity values. The APP isthus more hydrated (more water molecules in each hydra-tion shell) than in salt solution (0.1 N NaCl) where thepolymer chain adopts a compact conformation due toscreening of the negative charges by the sodium ions. Log-ically, the contraction of the macromolecule chain in salt-free water upon protonation of each repeating unit inAPP 3 is larger (Fig. 5a transforming into 5d) than in0.1 N NaCl (Fig. 5b transforming into 5c). This would leadto the loss of greater number of water molecules fromthe hydration shell of the repeating unit being protonatedin salt-free water. This is reflected by the higher n1 value(2.80), greater basicity constant (log K0

1: 8.58), and greaterchanges in the basicity constant in salt-free water (Fig. 6)than in 0.1 N NaCl.

The second basicity constants K02 and the n2 values for

the protonation of the P(@O)OEtO� group were deter-mined as log K0

2 = 2.54, n2 = 0.69 in salt-free water; logK0

2 = 2.76, n2 = 0.67 in 0.1 N NaCl. The slightly higher logK0

2 in 0.1 N NaCl could be attributed to the higher viscosityvalue in the saline media (Fig. 5c vs. 5d) in which the mac-romolecule is more hydrated. Upon each protonation,greater number of water molecules is thus released in0.1 N NaCl thereby leading to the higher basicity constant.

The basicity constant (log K01) of the amine group

in monomer 1, copolymer 4 and the corresponding

Fig. 5. The viscosity behavior of the PPB 2, APP 3 and CPP 4 (obtained via PPB 2 from entry 7, Table 1) in salt-free water and 0.1 N NaCl at 30 �C using anUbbelohde Viscometer.

Fig. 6. Plot for the apparent log K1 versus degree of protonation (a) for PPB 2-APE 3 system (entry 7, Table 1) in salt-free water and 0.1 N NaCl.

1122 N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123

homo-polymer [17] at 23 �C in salt-free water are deter-mined to be 8.99, 8.58 and 10.5, respectively; while therespective values in 0.1 N NaCl were found to be 8.74, 7.65and 10.0. The incorporation of electron-withdrawing SO2

moiety has thus resulted in the decrease of basicity constantof the nitrogens in copolymer 4 (in compare to homopoly-mer) by a staggering P2 units of log K. We intend to study

and compare the chelating properties, including ion-selectivity, of co-(4) and its corresponding homo-polymer[17] to form polymer-heavy metal ion complexes fromwaste water [21–23]. While the decrease in basicity of thenitrogens on the copolymer may have an adverse effect onthe chelation process, the presence of lone pairs on theSO2 units may be beneficial to chelation.

N.Y. Abu-Thabit et al. / European Polymer Journal 47 (2011) 1113–1123 1123

4. Conclusions

We have, for the first time, synthesized a pH-responsivephosphonobetaine/SO2 copolymer (PPB) 2 using Butler’scyclopolymerization process. The basicity constants ofthe pH-triggerable amine and phosphonate functionalitieshave been determined. The PPB 2 showed ‘antipolyelectro-lyte’ behavior in salt-added media. The study demon-strated a correlation between the basicity constants andviscosity values in salt-free and salt-added solutions. ThePPB was found to soluble in protic but insoluble in aproticsolvents. The polymer was found to be soluble in the pres-ence of Ca2+ ions thereby it may find potential application,for instance, in desalination plants in inhibiting CaCO3 scal-ing by making soluble polymer-Ca2+ metal complex. ThePPB is soluble in salt-free as well as salt-added water as aresult of the phosphonic acid pKa being >2. The natureand concentration of salt on the viscosity values have beendetermined; the polyphosphobetaine showed higherintrinsic viscosity values in LiCl than in NaCl. Tremendousdifference in the basicity constants of the nitrogens incopolymer 4 and its corresponding homopolymer [17]would allow us to study and compare their chelating prop-erties and selectivity with heavy metal ions.

Acknowledgements

The authors thankfully acknowledge the financial assis-tance provided by Center of Research Excellence in Petro-leum Refining and Petrochemicals (under Grant CoRE-PRP# 8) established by the Saudi Ministry of Higher Educationat King Fahd University of Petroleum and Minerals (theHost University), Dhahran, Saudi Arabia.

References

[1] Kudaibergenov S, Jaeger W, Laschewsky A. Adv Polym Sci2006;201:157–224.

[2] Singh PK, Singh VK, Singh M. e-Polymers 2007;030:1–34.[3] Lowe AB, McCormick CL. Chem Rev 2002;102:4177–89.[4] Armitage BA, Bennet DE, Lamparski HG, O’Brien DF. Adv Polym Sci

1996;126:54.[5] Waziri M, Abu-Sharkh BF, Ali SA. Biotechnol Progr 2004;20:526–32.

[6] Lee WF, Lee, CH. Polymer 1997;38:971-9 and references citedtherein.

[7] Butler GB. Cyclopolymerization and Cyclocopolymerization. NewYork: Marcel Dekker; 1992.

[8] Wielema TA, Engberts J. BFN Eur Polym J 1987;23:947–50.[9] Anton P, Laschewsky A. Makromol Chem 1993;194:601–24.

[10] Ali SA, Mazumder MAJ, Al-Muallem HA. J Polym Sci Part A: PolyChem 2003;41:172–84.

[11] Ali SA, Al-Muallem HA, Mazumder MAJ. Polymer 2003;44:1671–9.[12] Umar Y, Abu-Sharkh BF, Ali SA. Polymer 2005;46:10709–17.[13] Ali SA, Umar Y, Abu-Sharkh BF, Al-Muallem HA. J Pol Sci Part A: Poly

Chem 2006;44:5480–94.[14] Nakaya T, Li YL. Prog Polym Sci 1999;24:143.[15] Haruta M, Kageno K, Soeta M, Harada S. Jpn Kokai Tokkyo Koho

1975; JP 50072987 19750616.[16] Reidelsberger K, Jaeger W. Des Monomers Polym 1998;1:387–407.[17] Ali SA, Abu-Thabit NY, Al-Muallem HA. J Polym Sci Part A: Poly Chem

2010;48:5693–703.[18] Mou L, Singh G, Nicholson JW. Chem Commun 2000;5:345–6.[19] Perrin R, Elomaa M, Jannasch P. Macromolecules 2009;42:5146–54.[20] Lafitte B, Jannasch P. J Polym Sci Part A: Poly Chem 2005;43:273–86.[21] Rajan KS, Murase J, Martell AE. J Am Chem Soc 1969;91:4408.[22] Westerback S, Martell AE. Nature 1956;178:321.[23] Yamashoji Y, Matsushita T, Shono T. Technol Rep Osaka Univ

1985;35:331.[24] Al-Muallem HA, Wazeer MIM, Ali SA. Polymer 2002;43:4285–95.[25] Felty WK. J Chem Educ 1978;55:576.[26] Barbucci R, Casolaro M, Ferruti P, Barone V, Lelji F, Oliva L.

Macromolecules 1981;14:1203.[27] Barbucci R, Casolaro M, Nocentini M, Correzzi S, Ferruti P, Barone V.

Macromolecules 1986;19:37.[28] Pike RM, Cohen RA. J Polym Sci 1960;44:531–8.[29] Wilema TA, Engberts JBFN. Eur Polym J 1990;26:415–21.[30] Ali SA, Aal-e-Ali. Polymer 2001;42:7961–70.[31] Ali SA, Rasheed A, Wazeer MIM. Polymer 1999;40:2439–46.[32] Walsh DJ, Cheng GL. Polymer 1984;25:499.[33] Soto Monroy, Galin VM. J. C Polymer 1984;25:254–62.[34] Schultz DN, Peiffer DG, Agarwal PK, Larabee J, Kaladas JJ, Soni L, et al.

Polymer 1986;27:1734–42.[35] Butler GB, Angelo RJ. J Am Chem Soc 1957;79:3128–31.[36] Ali SA, Ahmed SZ, Wazeer MIM, Hamad EZ. Polymer

1997;38:3385–93.[37] Lancaster JE, Baccei L, Panzer HP. J Polym Sci Part C: Polym Lett

1976;14:549–54.[38] Butler GB. Acc Chem Res 1982;15:370.[39] Vynck VD, Goethals EJ. Macromol Rapid Commun 1997;18:149.[40] Ali SA, Rasheed A. Polymer 1999;40:6849–57.[41] Vorobeva AI, Vasileva EV, Gaisina KA, Puzin YI, Leplyanin GV. Polym

Sci. Ser A 1996;38:1077–80.[42] Salamone JC, Volksen W, Olson AP, Israel SC. Polymer

1978;19:1157–62.[43] Barbucci R, Casolaro M, Ferruti P, Nocentini M. Macromolecules

1986;19:1856.