Synthesis and reactivity of α,ω-homotelechelic polymers by Cu(0)-mediated living radical polymerization

European Polymer Journal xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

European Polymer Journal

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

Synthesis and reactivity of a,x-homotelechelic polymersby Cu(0)-mediated living radical polymerization

http://dx.doi.org/10.1016/j.eurpolymj.2014.07.0140014-3057/� 2014 Elsevier Ltd. All rights reserved.

⇑ Corresponding authors at: University of Warwick, Chemistry Depart-ment, Library road, CV4 7AL Coventry, United Kingdom.

E-mail address: [email protected] (D.M. Haddleton).

Please cite this article in press as: Simula A et al. Synthesis and reactivity of a,x-homotelechelic polymers by Cu(0)-mediated livingpolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpolymj.2014.07.014

Alexandre Simula a, Gabit Nurumbetov a, Athina Anastasaki a, Paul Wilson a,b,⇑,David M. Haddleton a,b,⇑a University of Warwick, Chemistry Department, Library road, CV4 7AL Coventry, United Kingdomb Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3052, Australia

a r t i c l e i n f o

Article history:Received 6 June 2014Received in revised form 9 July 2014Accepted 10 July 2014Available online xxxx

Keywords:Single electron transfer living radicalpolymerizationTelechelic polymersEnd-group modificationBlock copolymerSelf-assembly

a b s t r a c t

Telechelic poly(n-butyl acrylate) and poly[poly(ethylene glycol) methyl ether acrylate] areobtained with high monomer conversions and narrow molecular weight distributions(Ð < 1.2) by Cu(0)-mediated living radical polymerization (SET-LRP). The high end groupfidelity of the polymer is confirmed by a combination of 1H NMR and MALDI-ToF-MS anal-ysis. The reactivity of the telechelics is exploited to yield amphiphilic BAB triblocks, bysequential addition of a second monomer, which self-assemble in water. Furthermore,nucleophilic thiol-bromine substitution using 2-mercaptoethanol post-polymerizationenables the incorporation of primary alcohols on both the a- and x-polymer chain ends.The presence and reactivity of hydroxyl end-groups is confirmed upon reaction withisocyanates and further exploited to induce the ring opening polymerization (ROP) ofe-caprolactone.

� 2014 Elsevier Ltd. All rights reserved.

1. Introduction

The design and control over the synthesis of precisionmacromolecules remains an ongoing key challenge in poly-mer science [1]. The ability to tune macromolecular com-position and architecture, thus tailoring mechanical,physical and chemical properties has a major influenceon the final application of such polymeric materials rang-ing from therapeutic drug delivery [2,3] to personal careapplications [4].

Homo and hetero-telechelic polymers [5] provideaccess to a large variety of architectures via chainextension including simple BAB triblock copolymers,cross-linked co-network structures [6] and higher orderedself-assembled structures [7,8]. The reactivity of endgroups of telechelic polymer [9] has been exploited in

the preparation of functional polyesters [6,10] and poly-urethanes [6] via polycondensation chain extension proto-cols. Indeed, none of these supramolecular structures canbe achieved without a control over the morphology andchain end composition [11]. Recent advances in con-trolled/living radical polymerization (CRP) provide thenecessary control required to design increasingly complexarchitectures with good tolerance to side chain chemicalfunctionality and maximum retention of chain-end func-tionality [12,13]. Predictable molecular weights (Mw), nar-row distributions (Ð = Mw/Mn) and high end group fidelitycan be achieved by many techniques including reversibleaddition-fragmentation chain transfer polymerization(RAFT) [14–16], nitroxide-mediated polymerization(NMP) [17,18] and ring-opening polymerization (ROP)[19–21] processes. Transition metal-mediated living radi-cal polymerization (TMM-LRP) remains a widely usedtechnique to yield precision polymeric structures andmaterials [22–25]. Functionality can be introduced viathe initiator [26,27], the monomer through functional

radical

2 A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx

pendant groups, or by post-polymerization modifications[28–33] to yield functional scaffolds. Such techniques havebeen applied to introduce reactive hydroxyl, carboxylicacid and amine functional groups to the polymer chainends and along (meth)acrylic side chains [6]. Nevertheless,unavoidable termination reactions, present in all CRP pro-cesses have an unavoidable, deleterious effect upon endgroup fidelity [34,35].

More recently, Percec and coworkers introduced SingleElectron Transfer Living Radical Polymerization (SET-LRP)[36]. This Cu(0)-mediated CRP enables the polymerizationof acrylates [37,38], acrylamides [39], methacrylates[40,41,38], styrene [42] and less active vinyl halides [43]monomers in polar organic, aqueous [36,39,44], complex[45] and biologically [46] relevant media. This techniqueleads to relatively fast polymerization, exhibits very highinitiator efficiency, with minimal termination which isreciprocated by a unprecedented end group fidelity for aradical process (up to >99%) [47,48]. This has beenexploited to yield bifunctional hydrophobic polymers witha wide range of molecular weights [49]. Furthermore, usingincreasingly hydrophobic monomers has resulted in thedevelopment of self-generating biphasic systems with theretention of catalytic activity, polymerization control andin some cases, enhanced end-group fidelity [50–52]. Typi-cally, phase separation furnishes a catalyst-free polymer-rich layer ready to use for further modifications withoutthe need for additional purification steps.

Herein, the high a,x-chain end fidelity of hydrophilicand hydrophobic polymer chains is exploited to givefunctional polymeric diols. Poly(n-butyl acrylate) (PBA)and poly[poly(ethylene glycol) methyl ether acrylate] (av.Mn 480 g mol�1, PPEGA) are polymerized by SET-LRP usingthe bifunctional initiator ethylene bis(2-bromoisobutyrylbromide) in DMSO. Primary hydroxyl groups are then intro-duced post-polymerization by nucleophilic thio-bromosubstitution, using 2-mercaptoethanol. The combinationof 1H NMR spectroscopy, Gel Permeation Chromatography(GPC) and Matrix-Assisted Laser Desorption IonizationTime-of-Flight mass spectroscopy (MALDI-ToF-MS),together with chain extension and chain-end modificationdemonstrate the high end group fidelity and successfulfunctionalization of these polyacrylates. The reactivity ofthe hydroxyl groups is exploited by reaction withisocyanates and the ring opening polymerization ofe-caprolactone.

2. Experimental

2.1. Materials

n-Butyl acrylate (n-BA, 99%, Sigma–Aldrich), poly(ethyl-ene glycol) methyl ether acrylate (PEGA480, av. Mn

480 g mol�1, Sigma–Aldrich), dimethylsulfoxide (DMSO,reagent grade, Fisher scientific), dimethylformamide(DMF, reagent grade, Fisher scientific), tetrahydrofuran(THF, reagent grade, Fisher scientific), 2-mercaptoethanol(99%, Sigma–Aldrich), 1-dodecanethiol (98%, Sigma–Aldrich) and copper(II) bromide (98%, Sigma–Aldrich) were

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

used as received without further purification. Copper(0)wire (£ 0.25 mm) was pre-treated by washing in hydro-chloric acid (35%) for 10 min, then rinsed with acetone, driedunder nitrogen and used immediately. N,N,N0,N0,N00,N00-Hexamethyl-[tris(aminoethyl)amine] (Me6-TREN) was syn-thesized according to a reported procedure [53], degassedand stored at 4 �C under nitrogen prior to use. e-Caprolac-tone (97%, Sigma–Aldrich) was distilled under reduced pres-sure and stored under nitrogen at 4 �C prior to use. All otherreagents and solvents were obtained at the highest purityavailable from Sigma–Aldrich, Fisher Chemicals, or VWRand used without further purification unless specifiedotherwise.

2.2. Characterization

1H, 13C NMR spectra were recorded on Bruker ACF-250and DPX-400 spectrometers using deuterated solventsobtained from Sigma–Aldrich. IR spectra were collectedon a Bruker VECTOR-22 FT-IR spectrometer using a GoldenGate diamond attenuated reflection cell. ESI-MS data werecollected in positive mode, using a Bruker HCT Ultra ESIspectrometer. MALDI-ToF MS spectra were recorded inreflection mode on a Bruker Daltonics Ultraflex II MALDI-ToF mass spectrometer, equipped with a nitrogen laserdelivering 2 ns laser pulses at 337 nm with positive ionToF detection performed using an accelerating voltage of25 kV. The matrix solution was prepared by dissolvingtrans-2-[3-(4-tertbutylphenyl)-2-methyl-2-propenylidene]malononitrile (DCTB) in THF (200 mg/mL). Sodium iodidewas dissolved in THF (2 mg/mL). Polymer samples weredissolved in THF (1–5 mg/mL). Samples were prepared bymixing 5 lL of polymer solution, 5 lL of salt solution and20 lL of matrix solution. Calibration was performed witha poly(ethylene glycol) methyl ether acrylate Mn

1100 g mol�1 standard. GPC using a CHCl3 eluent was car-ried out at 30 �C on a Varian 390-LC system with aCHCl3 + 2% TEA eluent, equipped with 2 � PLgel 5 mmmixed D columns (300 � 7.5 mm), 1 � PLgel 5 mm guardcolumn (50 � 7.5 mm), autosampler and a refractive indexdetector. DMF GPC traces were obtained on a Varian 390-LCsystem using a DMF (5 mM NH4BH4) eluent at 50 �C,equipped with refractive index, UV and viscometry detec-tors, 2 � PLgel 5 mm mixed D columns (300 � 7.5 mm),1 � PLgel 5 mm guard column (50 � 7.5 mm) and autosam-pler. Narrow linear poly (methyl methacrylate) standardsin range of 200 to 1.0 � 106 g mol�1 were used to calibratethe system. All samples were passed through 0.45 lm PTFEfilter before analysis. TEM analysis was performed employ-ing a JEOL 1200 EX-II microscope with a Gatan 1 k � 1 kCCD camera. Samples were collected on 200-mesh Cu gridsand stained with 2% uranyl acetate prior to analysis. Meandiameter of the polymer micelles, coefficient of variationand normal distribution values were calculated by analys-ing TEM images in ImageJ� software. Optical microscopyanalysis was performed using a LEICA DM2500 microscopeequipped with a Nikon D5100 DSLR camera. Mean diameterof the polymer micelles, coefficient of variation and normaldistribution values were calculated by analysing TEMimages in ImageJ� software.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx 3

2.3. Synthesis of ethylene bis(2-bromoisobutyrate)

The synthesis was adapted from a reported procedure[52]. To a 2 L 3 neck RB flask equipped with a magneticstirring bar and a dropping funnel ethylene glycol (8 mL,0.14 mol) was added under nitrogen. Anhydrous dichlo-romethane (700 mL) was cannulated into the flask and tri-ethylamine (60 mL, 3 eq.) was added to the reactionmixture via a degassed syringe and allowed to cool to0 �C. 2-Bromoisobutyryl bromide (37.6 mL, 2.5 eq.) wasadded dropwise under nitrogen (over the course of onehour). Subsequently, the reaction mixture was allowed tostir at 0 �C for one hour and at ambient temperature over-night. The mixture was filtered and the volatiles removedby rotary evaporation. The resulting brown solution wasdissolved in chloroform (300 mL) and treated with 1 MHCl solution (250 mL), saturated NaHCO3 solution(250 mL) and deionised water (3 � 250 mL). The organiclayer were dried over anhydrous MgSO4 and passed twicethrough a basic activated alumina oxide column. The vola-tiles were removed in vacuo to give a light brown solid(40.66 g, 78% yield).

1H NMR (CDCl3, 400 MHz), d (ppm): 4.37 (s, 4H, RO-(CH2)2-OR) and 1.87 (s, 12H, –(CH3)2).13C NMR (CDCl3,100 MHz), d (ppm): 63.2 and 30.7. IR (m, cm�1): 3000 (C–H stretch) 1750 (O–CO–R), 1380 (–(CH3)2), 1300 (CHstretch), 1200 (CH stretch). HRMS (ESI; m/z, Da):[M + Na+] 382.9 (382.94 th).

2.4. Synthesis of bis[2-(2-bromoisobutyryloxy)ethyl] disulfide[DSDBr]

The synthesis of DSDBr was conducted according to areported procedure [54].

1H NMR (CDCl3, 400 MHz), d (ppm): 4.45 (t, 4H,J = 6.5 Hz, RO–CH2–), 2.99 (t, 4H, J = 6.5 Hz, –CH2–SR) and1.95 (s, 12H, –(CH3)2).13C NMR (CDCl3, 100 MHz), d(ppm): 63.5, 36.7 and 30.7. FT-IR (m, cm�1): 3000 (C–Hstretch) 1750 (O–CO–R), 1380 (–(CH3)2), 1300 (CH stretch),1200 (CH stretch), 510 (RS-SR). HRMS (ESI, m/z, Da):[M + Na+] 474.8 (474.9 Th).

2.5. Cu(0)-mediated polymerization of acrylates in organicsolvents

This procedure is generic for both the polymerization ofn-BA and PEGA480 in various solvents. The polymerizationof n-BA with ethylene bis(2-bromoisobutyrate) initiatorin DMSO is described. In an oven dried Schlenk tube ethyl-ene bis(2-bromoisobutyrate) (0.5 g, 1.39 mmol), copper(II)bromide (15.5 mg, 0.05 eq.), n-BA (7.96 mL, 40 eq.) andDMSO (8 mL) were added. The stirring bar with the pre-activated copper wire was subsequently added and theSchlenk tube sealed with a rubber septum. The reactionwas kept stirring with nitrogen sparging for 10 min priorto addition of Me6-TREN ligand (75 lL, 0.18 eq.) via adegassed microsyringe and the reaction was left to poly-merize at ambient temperature (25 �C). Samples of thereaction mixture were taken periodically for 1H NMR,GPC and MALDI-ToF MS analysis. Samples were passedthrough a basic alumina oxide column to remove metal

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

salts prior any analysis. Samples for 1H NMR were dilutedin deuterated chloroform while the samples for GPC werediluted in the corresponding eluent. Chain extension wasperformed by injecting a second monomer to the reactionmixture via a degassed syringe.

2.6. Nucleophilic substitution of polymer end groups with 2-mercaptoethanol

Following the polymerization, poly(n-BA) was usedwithout further modification which was possible due tothe phase separation of the polymer from the catalyst.Poly(n-BA) (0.5 g, 0.14 mmol, Mn 3500 g mol�1) wascharged to a vial fitted with a magnetic stirring bar and arubber septum with THF (5 mL), triethylamine (1 mL,50 eq.) and 2-mercaptoethanol (500 lL, 50 eq.). The reac-tion mixture was left to stir overnight at ambient temper-ature. After filtration, the volatiles were removed by rotaryevaporation and the remaining product analysed by 1HNMR and GPC analysis. The remaining thiols were removedby dialysis against 2-isopropanol (IPA, MWCO 1 kDa) andthe product isolated under vacuum prior to analysis by1H NMR and MALDI-ToF.

2.7. Nucleophilic substitution of the polymer end groups withdodecanethiol

Following polymerization poly(PEGA480) (10 g, 1.40mmol, Mn 7100 g mol�1) was diluted in THF (20 mL) andpassed through a basic activated alumina column toremove metal salts. The solution was placed in a 50 mLRB flask fitted with a magnetic stirring bar and a rubberseptum. Triethylamine (9.8 mL, 50 eq.) and dodecanethiol(4.9 mL, 50 eq.) were added to the solution and left to stirovernight at ambient temperature. After filtration, thesolution was dialyzed over water (MWCO 1 kDa) over-night. One part of the solution was analysed by DLS andTEM, while the remainder was freeze dried prior to analy-sis by 1H NMR and GPC.

2.8. Self-assembly of copolymers

The obtained copolymers were dissolved in THF at a10 mg/mL concentration and dialyzed over water (MWCO3.5 kDa) overnight. The resulting cloudy solutions wereanalysed by TEM. Samples were collected on 200-meshCu grids and stained with 2% uranyl acetate prior toanalysis.

2.9. Phenyl isocyanate modification

a,x-Hydroxyl terminated PBA (Mn 3600 g mol�1, 0.5 g,0.28 mmol) was added to a 50 mL RB flask equipped witha magnetic stirring bar and a rubber septum. The polymerwas isolated by removal of volatiles under vacuum at 70 �Covernight before anhydrous DMF (10 mL) being addedunder nitrogen flow. The mixture was degassed usingfreeze-pump thaw cycles. Phenyl isocyanate (604 lL,40 eq.) and dibutyltin dilaurate (3 drops) were added viaa degassed syringe. The reaction mixture was placed inan oil bath at 60 �C and left to stir for 9 h. Following the

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

4 A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx

reaction, the polymer was isolated by removal of volatilesunder vacuum at 70 �C overnight. The success of the func-tionalization of the hydroxyl end groups was assessed byFT-IR, GPC, 1H NMR and MALDI-ToF.

2.10. Ring-opening polymerization of e-caprolactone initiatedby a,x-hydroxyl terminated poly(n-BA)

This procedure is adapted from Zhang et al. [55]. In anoven dried Schlenk tube, OH-terminated PBA (Mn

3600 g mol�1, 0.5 g, 0.28 mmol) was dried at 70 �C undervacuum for one day. e-Caprolactone (308 lL, 20 eq.) wasadded to the polyol followed by three evacuation/N2 refill-ing cycles. Stannous octoate (36 mg, 0.65 eq.) was dis-solved in anhydrous toluene (2 mL) in a vial and themixture injected into the Schlenk tube via a degassed syr-inge and the evacuation/N2 refilling cycles repeated threetimes. The reaction mixture was left to stir at 120 �C undernitrogen for 12 h. Subsequently, the polymer mixture wasdialyzed (MWCO 3.5 kDa) against MeOH:H2O 9:1 and iso-lated by rotary evaporation. The copolymer was analysedby 1H NMR and GPC.

O

8-9

Scheme 1. Cu(0)-mediated polymerizat

Fig. 1. Cu(0)-mediated polymerization of n-butyl acrylate ([I]:[M]:[Me6-TREN]:[Ctime (mins) and linear fit (dash line). (B) Normalized dw/dlogM vs. M at didispersities (squares) at different monomer conversions and linear fit (dash line

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

3. Results and discussion

Telechelic homopolymerization of hydrophobic butylacrylate (n-BA) and hydrophilic poly(ethylene glycol)methyl ether acrylate (PEGA480) was carried out accordingto the general procedure using bifunctional initiator, ethyl-ene bis(2-bromoisobutyrate) (EbBiB), in DMSO (Scheme 1).

The Cu(0)-mediated, telechelic polymerization of n-BAexhibited living characteristics, exemplified by a linearevolution of Mn with respect to monomer conversion, inkeeping with previously reported results (Fig. 1C). Quanti-tative conversion (>99%) was reached in <3 h and theresulting polymer retained a narrow molecular weight dis-tribution (Ð < 1.12). The linear and telechelic polymeriza-tion of n-BA has previously been reported to proceedthrough a self-generating biphasic system [52]. Pleasingly,the expected phase separation was observed during thisinvestigation when the growing polymer chain reached amolecular weight of �2500 g mol�1 generating an unstableemulsion with stirring during the reaction giving a poly-mer rich top layer and a catalyst rich bottom layer imme-diately following cessation of agitation. The final polymer

ion of acrylates in DMSO at 25 �C.

u(0)]:[Cu(II)Br2] 1:40:0.4:0.18:0.05) in DMSO at 25 �C. (A) Ln([M]0/[M]) vs.fferent monomer conversions. (C) Experimental molecular weights and). (D) MALDI-ToF-MS spectrum obtained at 98% monomer conversion.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx 5

can be used without further purification for mass spec-trometry analysis or for end group modifications. TheMALDI-ToF-MS analysis of the unpurified polymer revealsa single mass distribution, corresponding to a n-BA unit(128 Da, Fig. 1D). Moreover, each peak follows the isotopicdistribution consistent of a,x-bromo chain ends, suggest-ing a high end group fidelity of telechelic poly(n-BA)(PBA, Fig. S8) which is further supported by symmetricaland monomodal GPC traces throughout the molecularweight evolution (Fig. 1C).

The telechelic nature, and ‘livingness’ of both a,x-chainends of the polymer was also assessed using a ‘‘cleavable’’

Fig. 2. Reduction of poly(n-butyl acrylate) initiated by bis[2-(2-bromo-isobutyryloxy)ethyl] disulphide upon addition of tributylphosphine(CHCl3 eluent).

Fig. 3. Cu(0)-mediated polymerization of poly(ethylene glycol) methyl ether a1:40:0.4:0.18:0.05) in DMSO at 25 �C. (A) Ln([M]0/[M]) vs. time and linear fit (dashconversion and linear fit (dash line). (C) dw/dlogM vs. M at different monomer

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

initiator, bis[2-(2-bromoisobutyryloxy)ethyl] disulphide.For comparison, n-BA was polymerized under the sameconditions as described above. Upon completion of thepolymerization, an excess of tributylphosphine was addedin order to reduce the disulphide bond. Within an hour, fullreduction was observed by GPC as illustrated by completeshift in the molecular weight distribution to approximately50% of the original value, with retention of monomodalityand narrow dispersity (Fig. 2). This is indicative of thepropagation proceeding from both chain ends with mini-mal termination at either end. Additional evidence fordisulfide bond cleavage was observed in 1H NMR(Fig. S12), whereby the –S–CH2 peaks were found to shiftfrom 4.0 ppm to 3.8 ppm upon addition of the reducingagent.

The polymerization of water soluble PEGA480 was con-ducted in DMSO using bi-functional initiator EbBiB. The‘living’ nature of the polymerization was again confirmedby the pseudo first order kinetics (Fig. 3A) observed follow-ing a short induction period (�90 min) and linear evolu-tion of molecular weight with conversion (Fig. 3B). Ashoulder at high molecular weight is observed on themolecular weight distribution which is common when per-forming CRP of PEGA based monomers and possibly resultsfrom the presence of diacrylate species present in the start-ing material [56]. Despite the induction period and thepresence of undesired diacrylates, very high monomer con-versions are obtained, observed molecular weights corre-late well with theoretical values and the molecularweight distribution remains narrow (Ð < 1.16, Fig. 3C).

The high end group fidelity of the PPEGA480 was exper-imentally verified by chain extension with varying ratios ofn-BA to furnish amphiphilic BAB tri-block copolymers

crylate (average Mw 480 g mol�1, [I]:[M]:[Me6-TREN]:[Cu(0)]:[Cu(II)Br2]line). (B) Experimental molecular weights and distributions vs. monomer

conversions.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

Table 1PBA-PPEGA-PBA ABA triblocks and statistical copolymers prepared by SET-LRP.

Entry PEGA480:nBA Time (h) Conv. (PEGA480/nBA)a (%) MnGPCb (g mol�1) Ð (Mw/Mn)

1 10 5 95/n.a. 7200 1.092 10:20 17 >99/83 9900 1.123 10:40 17 >99/>99 11,300 1.114 10:80 17 >99/>99 17,300 1.125 40:10 17 >99/>99 18,300 1.146 80:10 17 98/86 29,000 1.187 80:10c 17 >99/94 29,300 1.17

a Determined by 1H NMR (CDCls).b Determined by CHCl3 GPC.c Random copolymers.

Fig. 5. Optical microscopy image of self-assembled dodecanethiol mod-ified poly[poly(ethylene glycol) methyl ether acrylate] [M]:[I] 1:10. Zoom�40, scale bar 25 lm.

6 A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx

(Table 1). Chain extension was realised upon addition of adeoxygenated aliquot of comonomer (n-BA) when thehomopolymerization had reached high conversion (98–99%, by 1H NMR, Fig. S13). The resulting BAB triblocks wereobtained in high conversion, with respect to the comono-mer (>80%), and with narrow molecular weight distribu-tions (Ð < 1.18, Fig. 4). The complete shift to highermolecular weight is indicative of minimal loss of the a,x-Br end groups, even at high conversion, during homopoly-merization. The high molecular weight shoulder can againbe attributed to impurities in the PEGA480 monomer.

Further evidence for the high chain-end fidelity wasprovided by thio-bromine nucleophilic substitution at thea,x-chain ends of a low molecular weight (DPn = 10, Mn

�5,000 g.mol�1) PPEGA480. PPEGA480 was synthesized in5 h exhibiting quantitative monomer conversion (> 99%by 1H NMR) and narrow molecular weights distributions(Mn = 7100 g.mol�1, Ð = 1.10). Thio-bromine substitutionwas achieved upon addition of an excess of the nucleo-philic dodecanethiol in the presence of triethylamine. Thesolution was then dialyzed against water (MWCO3.5 kDa) to remove the reaction solvent and by-products.During dialysis, the product solution became cloudy imply-ing aggregation into higher order structures. Examinationof a sample of the cloudy solution by optical microscopyrevealed the presence of microparticles with a diameter

Fig. 4. Chain extension of telechelic poly[poly(ethylene glycol) methylether acrylate] (average Mw 480 g mol�1) with different amounts of n-butyl acrylate in DMSO at 25 �C. Numbers correspond to the entrynumbers in Table 1.

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

between 2–4 lm. This indicates that substitution of thea,x-chain ends of the hydrophilic PPEGA480 polymer witha hydrophobic molecule such as dodecanthiol confers sur-

Fig. 6. 1H NMR (CDCl3, 300 MHz) of dodecanethiol modifiedpoly[poly(ethylene glycol) methyl ether acrylate] [M]:[I] 1:10.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx 7

factant like properties on the resulting polymers leading tothe observed self-assembly in aqueous media (see Fig. 5).

The dialyzed product was collected by lyophilizationand the resulting viscous solution was analysed by 1HNMR (Fig. 6) and GPC (Fig. S20) to reveal the successfulmodification of the halogen chain ends. The GPC chromato-gram shifts to lower molecular weights (Mn = 5900g mol�1, Ð = 1.11), as a result of an enhanced hydrophobic-ity and solvability in the eluent, without affecting themolecular weight distribution (Fig. S21). In 1H NMR, theappearance of new peaks between 1.5 and 2.5 ppmsuggests the successful nucleophilic substitution of thehalogen chain ends with dodecanethiol (Fig. 6).

Integration of the peaks corresponding to the a,x-end–CH3 (t, 0.88 ppm) of the triblock copolymer has an inte-gration of 6, using the –(CH3)2 initiator peaks as a reference(1.14 ppm, s, 12H, Fig. S22). This confirms the high end

Fig. 7. 1H NMR (CDCl3, 400 MHz) of a,x-hydroxyl terminated poly(n-butyl acrylate) (Mn = 3600 g mol�1, Ð = 1.09) after purification.

Fig. 8. MALDI-ToF spectra of poly(n-butyl acrylate) before (blue) and after (red) ain this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

group fidelity of the telechelic PPEGA480 polymer as twothiol moieties are incorporated onto the polymer a,x-chain ends.

The reactivity of the halogen chain ends of hydrophobicPBA was also investigated by nucleophilic substitution. Thetelechelic n-BA homopolymer was synthesized by Cu(0)-mediated polymerization in DMSO reaching high monomerconversion (>99% by 1H NMR) within 3 h whilst retainingnarrow dispersities (Mn = 3500 g mol�1, Ð = 1.11). In thiscase, 2-mercaptoethanol was selected as the nucleophileto introduce a,x-hydroxyl functionality to the polymerchains. Successful substitution was achieved followingaddition of 2-mercaptoethanol, in the presence of triethyl-amine to yield the a,x-hydroxyl terminated PBA as charac-terized by 1H NMR, COSY and MALDI-ToF-MS analysis.Though a small molar excess of mercaptoethanol isreported to be sufficient to achieve substitution, a vastexcess (50 eq.) of mercaptoethanol and triethylamine wasemployed to ensure complete substitution and to drivethe reaction to completion. The shift in 1H NMR of the –CH2 peaks of 2-mercaptoethanol from 3.7 ppm to 3.9 ppmand 2.7 ppm to 2.9 ppm respectively, suggest a significantincorporation of the nucleophile into the polymer endgroups (Fig. 7). The excess reagents employed wereremoved by dialysis prior to characterization.

1H NMR spectra of both unpurified (Fig. S16) and puri-fied (Fig. 7) polymer were recorded in order to confirm thepresence of hydroxyl groups on the chain ends and toascertain the purity of the final material. The remainingthiols were removed by dialysis against IPA (MWCO1 kDa) overnight and the resulting product was isolatedby removal of volatiles under reduced pressure. The nucle-ophilic substitution was also monitored by MALDI-ToF-MSanalysis which confirmed successful incorporation of pri-mary alcohols on both ends of the polymer chain (Fig. 8).

The major distribution corresponding to the end-func-tional telechelic polymer (blue) was observed to shift upon

ddition of 2-mercaptoethanol. (For interpretation of the references to color

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

8 A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx

thio-bromo substitution to furnish a macromolecular diol(red). A minor distribution, attributed to hydrogen termi-nation taking place during the initial homopolymerizationcannot be ignored. Nevertheless, the loss of the bromidechain end before modification is low and has limited effecton the incorporation of hydroxyl groups.

The reactivity of the terminal alcohols was examined bythe reaction with isocyanates to further modify the termi-nal functionalities of the polymer chains. The reactivitybetween isocyanate and hydroxyl groups is well-studied[57–60]. Thus, hydroxyl-terminated PBA was reacted withan excess (40 eq.) of phenyl isocyanate, in the presence of atin(II) catalyst under anhydrous conditions, as monitoredby 1H NMR, FT-IR, GPC and MALDI-ToF-MS analysis. Theshift of the thioether –CH2 peaks and the appearance of

Fig. 10. FT-IR of a,x-hydroxyl terminated poly(n-butyl acrylate) befor

Fig. 9. 1H NMR (CDCl3, 400 MHz) of ‘end-capped’ a,x-hydroxyl termi-nated poly(n-butyl acrylate) (Mn = 3600 g mol�1, Ð = 1.09) with phenylisocyanate.

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

the aromatic signals at 7.3 ppm corresponding to the phe-nyl ring confirms the incorporation of the urethane group(Fig. 9).

Additional evidence for the chain end modification waselicited from FT-IR analysis. The isocyanate band (2275–2250 cm�1) was found to gradually decrease throughoutthe reaction, coinciding with the evolution of a urethanestretch (1650 cm�1) and appearance of the aromatic sig-nals (1540 cm�1). The absence of remaining isocyanatesor hydroxyl signals confirms successful modification ofthe a,x-terminal hydroxyl groups (Fig. 10).

The incorporation of the isocyanate onto the polymerchain ends was also verified by GPC equipped with sequen-tial RI and UV detectors (k = 280 nm). The coincidence ofthe peaks obtained from both detectors confirms the addi-

e [bottom] and after ‘end-capping’ with phenyl isocyanate [top].

Fig. 11. 1H NMR (CDCl3, 400 MHz) of poly(e-caprolactone)-block-poly(n-butyl acrylate)-block-poly(e-caprolactone) (Mn 5800 g mol�1, Ð = 1.4).

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx 9

tion of the phenyl ring throughout the molecular weightdistribution (Fig. S25).

The reactivity of the hydroxyl group was also used toinitiate a ring-opening polymerization of cyclic lactone e-caprolactone (e-CL) as an alternative route to synthesizingBAB triblock copolymers. ROP is considerably more sensi-tive than Cu(0)-mediated polymerization. Therefore, allreagents and solvents were stringently dried prior to reac-tion to minimize the chance of side reactions.

The polymerization was monitored by 1H NMR, usingthe appearance of the peaks at 3.9 and 2.2 ppm corre-sponding to the a,x–CH2 of the ring-opened e-caprolac-tone (Fig. 11). The complementary GPC analysis revealeda shift to higher molecular weight values and the finalobtained molecular weight was close to the calculatedvalue by 1H NMR. Nevertheless, the dispersity of the PCL-b-PBA-b-PCL triblock copolymer was found to increase

Fig. 13. Transmission Electron Microscopy of self-assembled PBA-b-PPEGA-b-PBA (degrees of polymerization [40]-[10]-[40]) in water. Zoom�40,000, scale bar 50 nm.

Fig. 12. GPC traces (CHCl3 eluent) of a,x-hydroxyl terminated poly(n-butyl acrylate) (Mn 3600 g mol�1, Ð = 1.09) before [blue] and after ring-opening of e-caprolactone (Mn 5800 g mol�1, Ð = 1.4) [orange]. (Forinterpretation of the references to color in this figure legend, the readeris referred to the web version of this article.)

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

(Ð > 1.2, Fig. 12) reflecting a loss of control during the ringopening process. This could be attributed to the presenceof moisture, as water can competitively initiate polymeri-zation, or inaccuracies in the calculated stoichiometriesboth of which result in a loss of integrity of the targeted tri-blocks and highlight the importance of maximum reten-tion of the a,x-hydroxyl end groups.

The self-assembly of the PBA-b-PPEGA480-b-PBA tri-blocks in water to form particles was also investigated.The assembly of widely used copolymers of PBA-b-PPEGAhas not been investigated in BAB triblock morphology.Consequently, we attempted to assemble a PBA-b-PPEGA-b-PBA triblock copolymer with corresponding DPs of 40-10-40. Following polymerization, the triblock copolymerwas dissolved in THF at a concentration of 10 mg/mL anddialyzed against water overnight, with regular changes ofthe water bath. The resulting particles were visualized byTransmission Electron Microscopy (TEM, Fig. 13). TheTEM image reveals successful self-assembly of the amphi-philic triblock copolymer in water, with formation ofmicelles within the range of 13–26 nm. Despite the poordispersity of the sample this result serves as an indicationof the amphiphilic nature character of the telechelic mac-romolecules prepared.

4. Conclusions

Cu(0)-mediated controlled/living radical polymeriza-tion has been exploited to prepare hydrophobic, hydro-philic and amphiphilic telechelic polymers with excellenta,x-end groups functionality. The high end group fidelityhas been confirmed both through characterization (NMR,GPC, MALDI-ToF-MS) and experimentally by sequentialmonomer addition and thio-bromine substitutions at thea,x-bromide chain ends. Subsequent incorporation ofhydroxyl end groups furnishes prepolymers capable of ini-tiating ring-opening polymerization of e-caprolactone andalso possess the expected reactivity towards isocyanates.The combination of all of the results enhances the use oftelechelics as prepolymers to generate functional materials.

Acknowledgments

The authors gratefully acknowledge financial supportfrom Lubrizol (A.S., A.A.). Equipment used in this researchwas funded in part through Advantage West Midlands(AWM) Science City Initiative and in part by the ERDF.DMH is a Royal Society/Wolfson Fellow.

Appendix A. Supplementary material

Supplementary material associated with this article canbe found, in the online version, at http://dx.doi.org/10.1016/j.eurpolymj.2014.07.014.

References

[1] Hawker CJ, Wooley KL. Science 2005;309(5738):1200–5.[2] Klok HA, Lecommandoux S. Adv Mater 2001;13(16):1217–29.[3] Lehn J-M. Molecular and supramolecular devices. In: Supramol.

chem.. Wiley-VCH Verlag GmbH & Co. KGaA; 2006. p. 89–138.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014

10 A. Simula et al. / European Polymer Journal xxx (2014) xxx–xxx

[4] Goddard ED, Gruber JV. Principles of polymer science and technologyin cosmetics and personal care. Marcel Dekker; 1999.

[5] Boutevin B. Telechelic oligomers by radical reactions. New polymermaterials, vol. 94. Berlin Heidelberg: Springer; 1990. p. 69–105.

[6] Tasdelen MA, Kahveci MU, Yagci Y. Prog Polym Sci2011;36(4):455–567.

[7] Grumelard J, Taubert A, Meier W. Chem Commun 2004;13:1462–3.[8] Blanazs A, Armes SP, Ryan AJ. Macromol Rapid Commun 2009;30(4–

5):267–77.[9] Horie K, Báron M, Fox RB, He J, Hess M, Kahovec J, et al. Pure Appl

Chem 2004;76(4):889–906.[10] Guillaume SM. Eur. Polym. J. 2013;49(4):768–79.[11] Matyjaszewski K, Gnanou Y, Leibler L. Macromolecular Engineering.

In: Macromolecular Engineering. Wiley-VCH Verlag GmbH & Co.KGaA; 2007. p. 1–6.

[12] Quirk RP, Lee B. Polym Int 1992;27(4):359–67.[13] Otsu T. J Polym Sci, Part A: Polym Chem 2000;38(12):2121–36.[14] Chiefari J, Chong YK, Ercole F, Krstina J, Jeffery J, Le TPT, et al.

Macromolecules 1998;31(16):5559–62.[15] Benaglia M, Chiefari J, Chong YK, Moad G, Rizzardo E, Thang SH. J Am

Chem Soc 2009;131(20):6914–5.[16] Barner-Kowollik C, Davis TP, Heuts JPA, Stenzel MH, Vana P,

Whittaker M. J Polym Sci, Part A: Polym Chem 2003;41(3):365–75.[17] Georges MK, Veregin RPN, Kazmaier PM, Hamer GK.

Macromolecules 1993;26(11):2987–8.[18] Hawker CJ, Bosman AW, Harth E. Chem Rev 2001;101(12):3661–88.[19] Labet M, Thielemans W. Chem Soc Rev 2009;38(12):3484–504.[20] Stridsberg K, Ryner M, Albertsson A-C. Controlled ring-opening

polymerization: polymers with designed macromoleculararchitecture. Degradable aliphatic polyesters, vol. 157. BerlinHeidelberg: Springer; 2002. p. 41–65.

[21] Khanna A, Sudha Y, Pillai S, Rath S. J Mol Model 2008;14(5):367–74.[22] Kamigaito M, Ando T, Sawamoto M. Chem Rev

2001;101(12):3689–746.[23] Matyjaszewski K. Macromolecules 2012;45(10):4015–39.[24] Wang J-S, Matyjaszewski K. J Am Chem Soc 1995;117(20):5614–5.[25] Matyjaszewski K, Xia J. Chem Rev 2001;101(9):2921–90.[26] Haddleton DM, Waterson C, Derrick PJ, Jasieczek CB, Shooter AJ.

Chem Commun 1997(7):683–4.[27] Zhang X, Matyjaszewski K. Macromolecules 1999;32(22):7349–53.[28] Coessens V, Matyjaszewski K. Macromol Rapid Commun

1999;20(3):127–34.[29] Ando T, Kamigaito M, Sawamoto M. Macromolecules

1998;31(19):6708–11.[30] Coessens V, Matyjaszewski K. Macromol Sci, Part A 1999;36(5-

6):811–26.[31] Coessens V, Matyjaszewski K. Macromol Sci, Part A 1999;36(5-

6):667–79.[32] Opsteen JA, van Hest JCM. Chem Commun 2005(1):57–9.[33] Lutz J-F. Angew Chem Int Ed 2007;46(7):1018–25.

Please cite this article in press as: Simula A et al. Synthesis and reactivitypolymerization. Eur Polym J (2014), http://dx.doi.org/10.1016/j.eurpoly

[34] Fischer H. Chem Rev 2001;101(12):3581–610.[35] Fischer H. J Polym Sci, Part A: Polym Chem 1999;37(13):1885–901.[36] Percec V, Guliashvili T, Ladislaw JS, Wistrand A, Stjerndahl A,

Sienkowska MJ, et al. J Am Chem Soc 2006;128(43):14156–65.[37] Rosen BM, Percec V. Chem Rev 2009;109(11):5069–119.[38] Samanta SR, Anastasaki A, Waldron C, Haddleton DM, Percec V.

Polym Chem 2013.[39] Zhang Q, Wilson P, Li Z, McHale R, Godfrey J, Anastasaki A, et al. J Am

Chem Soc 2013.[40] Lligadas G, Percec V. J Polym Sci, Part A: Polym Chem

2008;46(14):4917–26.[41] Fleischmann S, Percec V. J Polym Sci, Part A: Polym Chem

2010;48(21):4884–8.[42] Tom J, Hornby B, West A, Harrisson S, Perrier S. Polym Chem

2010;1(4):420–2.[43] Hatano T, Rosen BM, Percec V. J Polym Sci, Part A: Polym Chem

2010;48(1):164–72.[44] West AG, Hornby B, Tom J, Ladmiral V, Harrisson S, Perrier S.

Macromolecules 2011;44(20):8034–41.[45] Waldron C, Zhang Q, Li Z, Nikolaou V, Nurumbetov G, Godfrey J, et al.

Polym Chem 2014;5(1):57.[46] Zhang Q, Li Z, Wilson P, Haddleton DM. Chem Commun

2013;49(59):6608–10.[47] Soeriyadi AH, Boyer C, Nyström F, Zetterlund PB, Whittaker MR. J Am

Chem Soc 2011;133(29):11128–31.[48] Nyström F, Soeriyadi AH, Boyer C, Zetterlund PB, Whittaker MR. J

Polym Sci, Part A: Polym Chem 2011;49(24):5313–21.[49] Lligadas G, Percec V. J Polym Sci, Part A: Polym Chem

2007;45(20):4684–95.[50] Waldron C, Anastasaki A, McHale R, Wilson P, Li Z, Smith T, et al.

Polym Chem 2014;5(3):892.[51] Anastasaki A, Waldron C, Nikolaou V, Wilson P, McHale R, Smith T,

et al. Polym Chem 2013;4(15):4113–9.[52] Boyer C, Atme A, Waldron C, Anastasaki A, Wilson P, Zetterlund PB,

et al. Polym Chem 2013;4(1):106–12.[53] Ciampolini M, Nardi N. Inorg Chem 1966;5(1):41–4.[54] Syrett JA, Jones MW, Haddleton DM. Chem Commun

2010;46(38):7181–3.[55] Zhang Q, Li G-Z, Becer CR, Haddleton DM. Chem Commun

2012;48(65):8063–5.[56] Grice AJ. Investigation of single electron transfer radical

polymerization (SET-LRP). University of Warwick; 2010.[57] Flores JD, Treat NJ, York AW, McCormick CL. Polym Chem

2011;2(9):1976–85.[58] Gody G, Rossner C, Moraes J, Vana P, Maschmeyer T, Perrier S. J Am

Chem Soc 2012;134(30):12596–603.[59] Biedermann F, Appel EA, del Barrio Js, Gruendling T, Barner-Kowollik

C, Scherman OA. Macromolecules 2011;44(12):4828–35.[60] Mansur AAP, Nascimento OLd, Oréfice RL, Mansur HS. Surf Interface

Anal 2011;43(3):738–43.

of a,x-homotelechelic polymers by Cu(0)-mediated living radicalmj.2014.07.014