One-step synthesis of α-p-vinylphenylalkyl-ω-hydroxy poly(ethylene oxide) macromonomers by anionic polymerization initiated from p-vinylphenylalkanols

Polymer 44 (2003) 3221–3228 www.elsevier.com/locate/polymer One-step synthesis of a-p-vinylphenylalkyl-v-hydroxy poly(ethylene oxide) macromonomers by anionic polymerization initiated from p-vinylphenylalkanols Renhua Shen, Takamichi Senyo, Chinami Akiyama, Yuji Atago, Koichi Ito* Department of Materials Science, Toyohashi University of Technology, Tempaku-cho, Toyohashi 441-8580, Japan Received 5 February 2003; received in revised form 6 March 2003; accepted 6 March 2003 Abstract v-( p-Vinylphenyl)alkanols, including methanol, ethanol, propanol, pentanol, and hexanol, have been partially alkoxidated with potassium naphthalene to initiate anionic polymerization of ethylene oxide (EO) in order to directly prepare the corresponding a-p-vinylphenylalkyl-vhydroxy poly(ethylene oxide) (PEO) macromonomers. p-Vinylphenylmethanol, i.e. p-vinylbenzyl alcohol (VBA) afforded the expected well-defined macromonomer via living polymerization mechanism and the kinetics have been examined as a function of extent of potassiumalkoxidation. Other alcohols such as p-vinylphenylpropanol (VPP), -pentanol (VPPT), and -hexanol (VPH) were also successful to afford the corresponding PEO macromonomers, while p-vinylphenylethanol (VPE) alkoxide polymerized EO to give p-divinylbenzene and poly(ethylene glycol) without p-vinylphenylethoxy end group, which were supposed to form by a very facile intramolecular chain transfer of the activated oligomeric alkoxide chain end to abstract a benzylic proton of the initiating fragment. q 2003 Elsevier Science Ltd. All rights reserved. Keywords: Ethylene oxide; Anionic polymerization; Macromonomers 1. Introduction Reactive or polymerizable amphiphiles have been of increasing concern because of their organizing properties to construct well-defined polymeric architecture [1]. Among others, so-called macromonomers have been useful in design of branched polymers by homo- and co-polymerization [2]. We have been particularly interested in poly(ethylene oxide) macromonomers carrying a hydrophilic poly(ethylene oxide) (PEO) chain and a hydrophobic polymerizable end group. They were found to organize into micelles in water and polymerize very rapidly to afford comb or brush polymers [3 –10], copolymerize with a small amount of styrene solubilized in the micelles to give unimolecular nanoparticles [11], and copolymerize with excessive amounts of styrene in emulsion or dispersion system to monodisperse polymeric microspheres of submicron to micron size [12 – 14]. So far conventional syntheses of macromonomers have * Corresponding author. Tel.: þ 81-532-44-6814; fax: þ81-532-48-5833. E-mail address: itoh@tutms.tut.ac.jp (K. Ito). involved introduction of polymerizable functions onto living polymer chain ends, called termination method [2]. Styryl-ended PEO macromonomers have also been successfully prepared by this method by polymerizing ethylene oxide (EO) followed by termination with corresponding pvinylphenylalkyl halides (Scheme 1(a)). Terminating agents, however, are needed to be used in considerable excess on molar basis, even more than 4 times excess in case of the bromides of m ¼ 4 or 7 in order to overcome the consumption due to a side reaction such as elimination [4]. So here comes an idea of ‘initiation’ method (Scheme 1(b)) in which p-vinylphenylalkanols are used as the initiator for polymerization of EO. If favorable, the initiation method utilizes all the initiator functions effectively incorporated as the chain ends to afford the expected macromonomers in one-step. Moreover, the PEO macromonomers obtained after work-up should have hydroxy groups as the other chain ends, which were introduced in the termination method in some money- and time-consuming procedure, for example by starting with silyl-protected alkoxide as an initiator to polymerize EO followed by termination and subsequent deprotection [5]. A 0032-3861/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0032-3861(03)00231-3 3222 R. Shen et al. / Polymer 44 (2003) 3221–3228 Scheme 1. problem involved here in the initiation method is just the requirement of no reaction between the propagating chain end and the initiator fragment. Since the oxy anions are usually believed to react very hardly with styryl functions, the situation appears very favorable [15 – 18]. In fact, Rempp and co-workers used potassium p-isopropenylbenzylate successfully to polymerize EO [19] and very recently Soula and Guyot used potassium vinylbenzylate to polymerize butylene oxide and EO successively to obtain the block macromonomers [20]. In this paper we present the results of using various v-( pvinylphenyl)alkanols, 2, which are partially alkoxidated with potassium naphthalene to polymerize EO to obtain well-defined PEO macromonomers, 3, carrying styryl end groups with varying hydrophobic alkylene spacers (m ¼ 1; 3, 5, 6) and controlled hydrophilic PEO chain lengths. Detailed kinetics of EO polymerization with p-vinylbenzyl alcohol will be also discussed as a function of degree of alkoxidation. To our knowledge, such information under extremely dry and high vacuum condition has not been available but will provide fundamental understanding of anionic polymerization of EO. Also some unexpected but interesting result with 2-( p-vinylphenyl)ethanol as an initiator (2, m ¼ 2) will be included. 2. Experimental 2.1. Materials p-Vinylbenzyl alcohol (VBA) (2, m ¼ 1) was prepared from p-vinylbenzyl chloride (VBC) by reaction with sodium acetate followed by alkaline hydrolysis, according to the procedure described [20,21]. 1H NMR in CDCl3 (Fig. 2(A)): d 1.8 (br, H; –CH2OH); d 4.65 (br., 2H; –CH2OH); d 5.25 (dd, 1H; yCH2); d 5.75 (dd, 1H; yCH2); d 6.75 (dd, 1H; – CHy), and d 7.37 (q, 4H; C6H4). 2-( p-Vinylphenyl)ethanol (VPE) (2, m ¼ 2) was prepared from p-chlorostyrene, via Grignard reagent followed by reaction with EO [22]. 1H NMR: d 2.33 (t, 1H; – CH2OH); d 2.84 (t, 2H; – ArCH2CH2OH); d 3.80 (m, 2H; – CH2CH2OH); d 5.24 (dd, 1H; yCH2); d 5.75 (dd, 1H; yCH2); d 6.72 (dd, 1H; – CHy), and d 7.35 (q, 4H; C6H4). 3-( p-Vinylphenyl)propanol (VPP) (2, m ¼ 3) was prepared from VBC, via Grignard reagent followed by reaction with EO as follows. A solution of VBC (0.2 mol, 30.5 g) in dry ether (80 mL) was dropped over 1 h under vigorous stirring into finely crushed magnesium turnings (0.22 mol, 5.3 g) in ether (120 mL) with a small amount of iodine. Temperature was kept at 0 – 10 8C. The reaction was continued for 1 h further without cooling. Then, cooled EO (0.4 mol, 20 mL) was added into the flask chilled at 2 78 8C, stirred for 1 h after the temperature was allowed to rise to ambient, followed by hydrolysis 2N aq. HCl. The organic layer was washed with water, dried over MgSO4, and filtered. Ether was evaporated and the residue was distilled under a reduced pressure. Bp 90 – 95 8C/6 –8 Torr. Yd. 70%. 1H NMR: d 1.38 (br, 1H; – CH2OH); d 1.9 (m, 2H; –ArCH2CH2CH2OH); d 2.7 (t, 2H; – ArCH2CH2 –); d 3.68 (t, 2H; –CH2OH); d 5.20 (dd, 1H; yCH2); d 5.71 (dd, 1H; yCH2); d 6.70 (dd, 1H; – CHy), and d 7.26 (q, 4H; C6H4). 5-( p-Vinylphenyl)pentanol (VPPT) (2, m ¼ 5) was prepared from p-(3-bromopropyl)styrene [23] via Grignard reagent followed by reaction with EO as above, except that the reaction with EO was conducted at 40 8C for 24 h in a closed system under vacuum with breakable seal technique, in a similar procedure for EO polymerization (see below). Ether extract was evaporated and freeze-dried from benzene. Yd. 40%. 1H NMR: d 1.27 (br, 1H; –CH2OH); d 1.4 (m, 2H; – CH2CH2CH2CH2OH); d 1.6 (m, 4H; – CH2CH2CH2CH2CH2OH); d 2.62 (t, 2H; –ArCH2CH2 – ); d 3.54 (t, 2H; – CH2 –CH2OH); d 5.19 (dd, 1H; yCH2); d 5.70 (dd, 1H; yCH2); d 6.70 (dd, 1H; – CHy), and d 7.22 (q, 4H; C6H4). 6-( p-Vinylphenyl)hexanol (VPH) (2, m ¼ 6) was prepared from p-(5-bromopentyl)styrene [24] via Grignard reagent followed by reaction with formaldehyde [25], worked up as above, and purified by column chromatography over silica gel with cyclohexane/ethyl acetate (80/20 R. Shen et al. / Polymer 44 (2003) 3221–3228 v/v) as an eluent. Yd. 38%. 1H NMR: d 1.19 (br, 1H; – CH2OH); d 1.4 (m, 4H; – CH2CH2CH2CH2CH2CH2OH); d 1.6 (m, 4H; – CH2CH2CH2CH2CH2CH2OH); d 2.60 (t, 2H; – ArCH2CH2 – ); d 3.63 (t, 2H; – CH2 – CH2OH); d 5.18 (dd, 1H; yCH2); d 5.70 (dd, 1H; yCH2); d 6.70 (dd, 1H; – CHy), and d 7.23 (q, 4H; C6H4). VBA, VPE, VPP, and VPPT were finally distilled over CaH2 under high-vacuum line and sealed into calibrated tubes with a breakable seal. VPH was evacuated under high vacuum, dissolved in tetrahydrofuran (THF), and sealed into calibrated tubes with a breakable seal. THF, distilled from a blue solution with sodium benzophenone, was dried and purified under vacuum by distillation over LiAlH4 and then over sodium anthracene, and finally from a red solution with disodium salt of amethylstyrene tetramer (Na2MS4) into calibrated flasks with a breakable seal. A solution of Na2MS4 in THF was prepared by reaction of a-methylstyrene with sodium mirror at room temperature, filtered, and stocked as dilute solutions in ampoules with a breakable seal. EO was distilled trap-totrap twice over KOH pellets, three times over CaH2 powder, and finally over Na mirror into calibrated tubes with a breakable seal. Potassium naphthalene (KC10H8) was prepared under high vacuum by reacting naphthalene with excess potassium mirror in THF. Naphthalene was purified by sublimation and dissolved in THF. Potassium mirror was prepared on the wall of a flask after careful trap-to-trap distillations over a small oxygen-free flame. The dark green solution obtained was filtered and divided into calibrated tubes with a breakable seal. The concentration was usually 0.2– 0.5N, as determined by titration of an aliquot in water with a potassium hydrogen phthalate solution. 2.2. Polymerization of EO Polymerization was conducted under high vacuum (5 £ 1025 Torr or 3.7 £ 1023 Pa) with all the reagents sealed into appropriate, calibrated ampoules which were also prepared under the vacuum with breakable seal technique. Kinetics of EO polymerization with VBA was followed in a procedure as follows. Ampoules including a washing solution (Na2MS4 in THF), VBA as an initiator, THF as a solvent, potassium naphthalene solution (KC10H8/THF), and EO were, respectively, attached into an apparatus with the reaction flask and several tubes for sampling as shown in Fig. 1. The apparatus was attached upside-down to a vacuum line, evacuated, baked over an oxygen-free flame, and sealed off from the line. The breakable seal of the ampoule of the washing solution (a) was broken with a magnetic bar to rinse all the inner walls. The walls were then completely washed and cleaned by fresh THF, which comes on distillation by cooling on the outer walls with cotton tips wetted with chilled isopropanol by dry ice, until the red color of the Na2MS4disappeared from the wall. The washing 3223 Fig. 1. Apparatus for kinetics measurement of EO polymerization. (a) Washing solution (Na2MS4/THF); (a0 ) ampoule for recovering a washing solution; (b) potassium naphthalene solution (KC10H8/THF); (c) THF; (d) initiator alcohol (VBA); (e) EO; (f) sampling tubes; (g) reaction flask. solution was recovered into the flask (a0 ) and sealed off. Initiator (VBA) and solvent (THF) were introduced into the reaction flask. Then the KC10H8/THF solution was introduced drop by drop into the flask under vigorous magnetic stirring, so that the dark green color immediately disappeared upon mixing, indicating the reaction with VBA to the alkoxide. The lower half of the apparatus was sealed off above the reaction flask. The breakable seal of the chilled ampoule of EO was finally broken to introduce the monomer into the reaction flask. The flask was then placed in a bath of 40 8C to start the polymerization. From time to time, the aliquots were transferred by inverting the apparatus into sampling tubes (f) and sealed off to check for conversion or degree of polymerization. Thus the content was terminated with small amounts of methanol and poured into a large amount of hexane to precipitate out the polymers, which were collected by filtration or by decantation, washed with hexane, and finally freeze-dried from benzene, and characterized by 1H NMR and size exclusion chromatography (SEC). The preparative syntheses of the macromonomers were similarly carried out starting from partially (about 40%) alkoxidated VBA, VPP, VPPT, and VPH with the apparatus as in Fig. 1 but without sampling tubes. The polymerization was conducted at 40 8C for more than 2 days to achieve almost quantitative conversion, and the polymers were purified by re-precipitation from THF into hexane and finally freeze-dried from benzene. The polymerization of EO with partially alkoxidated VPE was similarly conducted to almost quantitative conversion. The polymers isolated as the hexane-insoluble part, however, were found to be just poly(ethylene glycol) without any p-vinylphenylethyl groups as judged by 1H NMR. So the hexane soluble part was evaporated and the residue was analyzed by 1H NMR to be identified as p-divinylbenzene: d 5.2 (dd, 2H; yCH2), d 5.7(dd, 2H; yCH2), d 6.7 (q, 2H; yCH –), and d 7.37 (s, 4H; C6H4). The amount was almost comparative to that expected from the original VPE used. Independent experiment starting with 3224 R. Shen et al. / Polymer 44 (2003) 3221–3228 2-phenylethanol instead of VPE produced poly(ethylene glycol) and styrene just as expected. The attempted reaction just between 2-phenylethanol and KC10H8 in THF without EO polymerization, however, resulted in recovery of the alcohol after work-up. Also the reaction between 2phenylethanol and K-alkoxide of poly(ethylene glycol) monomethyl ether, KO(CH2CH2O)nCH3 ðn ¼ 15Þ; resulted in recovery of the alcohol and poly(ethylene glycol) monomethyl ether, as the hexane-soluble and -insoluble parts, respectively. These experiments show that the polymerization of EO with potassium VPE must have produced p-divinylbenzene and poly(ethylene glycol) by some intramolecular transfer involving hydrogen abstraction of the propagating alkoxide anion from the initiator fragment as will be discussed with Scheme 3. 2.3. Characterization 1 H NMR spectra were measured on Mercury Varian 300 with deutero-chloroform (CDCl3) solutions, with tetramethylsilane as an internal standard. Pulse width and delay were 7.25 ms and 1.5 s, respectively, to allow complete relaxation of the protons. Number of accumulation was 16 times. SEC was recorded on JASCO PU980 as a pump, with JASCO RI980 as an RI detector, and Shodex GPC KF-802 and -803 as columns. The eluent was THF with the flow rate of 1 mL/min at 40 8C. The standard poly(ethylene glycol)s were used for calibration of the molecular weights. 3. Results and discussion solution (KC10H8/THF) to convert the alcohols fractionally to alkoxides because the solution is easy to handle in a vacuum system. A problem is to avoid any reaction with the styryl groups which could occur easily via charge transfer as is well known since the discovery of the living polymerization of styrene [27]. This was accomplished by slowly adding (drop by drop) the KC10H8 solution to the excess alcohols in THF under vigorous stirring as described in Section 2. Detailed study was conducted first on kinetics of EO polymerization with VBA. 3.2. Kinetics of EO polymerization with VBA EO was polymerized in THF at 40 8C with VBA ðm ¼ 1Þ partially alkoxidated by potassium naphthalene (KC10H8). Polymerization was followed by 1H NMR with the samples isolated from time to time. Typical spectra are shown in Fig. 2 for the polymerization at 32% alkoxidation ðx ¼ 0:32Þ: Polymerization or incorporation of EO units can be seen in the appearance and increase of the oxyethylene peak around d 3.7, while the upfield shift of the benzyl methylene protons from d 4.65 for VBA (A) to d 4.55 after polymerization (B and C) indicates that initiation incorporated all the pvinylbenzyloxy groups as the initiator fragments in the polymer chains. It also appears that the vinylphenyl groups remain intact during polymerization of EO. So the increase in the peak of oxyethylene protons around d 3.7 relative to that of benzylmethylene or vinyl protons was taken as the measure of conversion or degree of polymerization by assuming no reaction of the vinylbenzyl groups. Fig. 3 shows the conversion vs time plots for various degree of alkoxidation ðxÞ: Clearly, the rate becomes increasingly higher with x: Table 1 summarizes the data 3.1. General scheme Since the propagating species in anionic polymerization of EO is an oxy anion via ring-opening of EO, the polymerization can be initiated by alkoxide and the propagation will continue without termination even in the presence of free alcohols because any proton exchange will reproduce the same oxy anion: 2 þ 2 RO K þ RO – H Y RO – H þ RO K RO2 Kþ þ EO ! RO – CH2 CH2 O2 Kþ þ ð1Þ ð2Þ Here RO can be any alcohol residue or poly(ethylene oxide) chain. Therefore, so long as the equilibrium in Eq. (1) is much faster than the propagation in Eq. (2), the system looks like ‘living’ polymerization with all the initial alcohol residues as the initiator fragments, just as observed in ‘immortal’ polymerization of epoxides by aluminum porphyrin complexes [26]. Thus in practical view of synthesis of PEO macromonomers, we thought the styrylalkanols as convenient initiators as given in Scheme 2, since the styryl double bonds are known to be inactive to oxy anions [15 – 20]. We used potassium naphthalene Table 1 Characterization of PEO macromonomers obtained at complete conversion of EO polymerization with VBA VBA xa (mmol) EO Time Mn;calc b Mn;NMR c Mn;SEC d Mw =Mn;SEC d (mmol) (h) 13.5 11.3 11.9 10.5 11.6 12.6 9.7 10.4 5.1 3.6 250 308 299 353 214 226 223 587 552 467 0.09 0.18 0.32 0.43 0.63 0.73 0.94 0.45 0.42 0.59 600 300 150 48 24 6 3 168 168 168 950 1320 1440 1610 940 920 1120 2630 4900 5890 970 1410 1320 1700 1090 1010 1100 2660 4820 5950 1060 950 1220 1400 990 1020 1270 2500 4100 5700 1.09 1.20 1.09 1.16 1.17 1.10 1.17 1.03 1.05 1.08 THF ¼ ca. 80 mL, 40 8C, conversion ¼ nearly quantitative. Degree of alkoxidation, x ¼ ½KC10 H8 Š=½VBAŠ0 : b Mn;calc ¼ MA þ 44½MŠ0 =½AŠ0 ; where MA ¼ molecular weight of initiator alcohol, here 134 for VBA. c Mn;NMR ¼ MA þ 44ðIEO =4Þ=ðIVBA =2Þ; where IEO ¼ peak intensity of the oxyethylene protons at d 3.7 and IVBA ¼ peak intensity of the VBA benzylic methylene protons at d 4.55. d Determined by SEC calibrated with standard poly(ethylene glycol)s. a R. Shen et al. / Polymer 44 (2003) 3221–3228 3225 Scheme 2. of characterization of the polymers obtained under various conditions after almost complete conversion. The numberaverage molecular weights as determined from 1H NMR ðMn;NMR Þ and those from SEC calibrated with standard poly(ethylene glycol)s ðMn;SEC Þ; and those calculated from the molar ratio of EO to VBA charged ðMn;calc Þ are in fair accord with each other, strongly supporting the living polymerization mechanism just as shown in Scheme 2. The chromatograms in SEC are unimodal in each case with nearly monodisperse distribution in the molecular weight ðMw =Mn # 1:2Þ: Thus all the alcohol molecules charged in the feed can be initiator fragments to afford the PEO macromonomers with the number-average degree of polymerization (DPn ¼ n; in Scheme 2) given as follows. DPn ¼ n ¼ ½MŠ0 u=½AŠ0 ð3Þ where [M]0 and [A]0 are the initial molar concentrations of EO and alcohol, respectively, and u is the conversion of EO polymerized and u ¼ 1 in Table 1. The data in Fig. 3 were re-plotted in Fig. 4 to follow the first-order kinetics: ln½MŠ0 =½MŠ ¼ 2lnð1 2 uÞ ¼ kp;app ½Pp Št ð4Þ where [M] is the monomer concentration after time t; with u ¼ ð½MŠ0 2 ½MŠÞ=½MŠ0 ; [Pp] is the concentration of active Fig. 2. Typical 1 H NMR spectra of VBA and products of EO polymerization at x ¼ 0:32 : (A) original VBA; (B) after 10 h ðn ¼ 4:5Þ; and (C) after 25 h ðn ¼ 13:9Þ: Peak with an arrow due to impurity (CHCl3). Fig. 3. Time-conversion plots of EO polymerization at various degree of alkoxidation: (a, W) x ¼ 0:94; (b, X) x ¼ 0:73; (c, L) x ¼ 0:63; (d, O) x ¼ 0:43; (e, A) x ¼ 0:32; (f, B) x ¼ 0:18; (g, K) x ¼ 0:09: See upper seven rows in Table 1 for the feed composition of VBA and EO. 3226 R. Shen et al. / Polymer 44 (2003) 3221–3228 3.3. Polymerization of with VPE to poly(ethylene glycol) and p-divinylbenzene Fig. 4. First-order plots of the data in Fig. 3: (a, W) x ¼ 0:94; (b, X) x ¼ 0:73; (c, L) x ¼ 0:63; (d, O) x ¼ 0:43; (e, A) x ¼ 0:32; (f, B) x ¼ 0:18; (g, K) x ¼ 0:09: chain ends, and kp;app is the corresponding apparent propagation constant. Since the polymerization is accelerated with x, we took the total potassium alkoxide concentration as [Pp], i.e. ½Pp Š ¼ x½AŠ0 ; to calculate even roughly the value of kp;app ; which will provide an idea of activity of each potassium alkoxide species in ring-opening polymerization of EO. Since the living nature of the present polymerization is evident by the data in Table 1, the scattering in the first-order plots in Fig. 4 may be due to sampling procedure in such a closed vacuum system (Fig. 1) which may change the concentrations of the species involved to some extent. Nevertheless, the results in Fig. 5 clearly shows as a fact that the kp;app values are not constant but increases with x; indicating that the free alcohols interfere the propagation reaction of the alkoxides as the active chain ends. We suppose that the exchange equilibrium and/or the complex formation among the free alcohols and the potassium alkoxides may apparently reduce the reactivity of the alkoxide moiety in ring-opening of EO. Further discussion, however, should be made after more detailed examination of the kinetics and some spectroscopic investigation of the possible complexes. Fig. 5. Change of kp;app as a function of degree of alkoxidation, x: VPE (10 mmol) was partially alkoxidated with potassium naphthalene ðx ¼ 0:44Þ to polymerize EO (250 mmol) under a condition similar to the legend to Table 1. The polymers were isolated as usual in quantitative yield by precipitation into hexane but identified just as poly(ethylene glycol) (PEG) with Mn;SEC around 103 without any pvinylphenylethoxy fragments. Instead, p-divinylbenzene was isolated from hexane-soluble part, with no indication of the initial VPE residue in the 1H NMR spectrum. Similar polymerization starting with 2-phenylethanol instead of VPE produced styrene and PEG as the hexane-soluble and insoluble fractions, respectively. On the other hand, no apparent reaction occurred and just the original alcohols were recovered after work-up either when VPE was just alkoxidated by potassium naphthalene or when 2-phenylethanol was reacted with potassium alkoxide of PEG monomethylether (see Section 2). These results strongly suggest an intramolecular hydrogen-transfer reaction after some degree of normal polymerization to release p-divinylbenzene and potassium alkoxide of oligo(ethylene glycol), which will continue to propagate to PEG. Thus we propose Scheme 3 as a mechanism. Activation of the oxy anion by crown etherlike complexation of the counter ion (Kþ), say, after normal addition of about 5 or 6 EO units, appears to be a driving force for intramolecular abstraction of the benzylic proton, together with formation of elongated conjugated phenylalkenes, i.e. p-divinylbenzene from VPE and styrene from 2phenylethanol here. 3.4. Polymerization with VPP, VPPT, and VPH for syntheses of hydrophobically enhanced styryl-ended PEO macromonomers PEO macromonomers carrying hydrophobicallyenhanced polymerizing end groups are particularly inter- Scheme 3. R. Shen et al. / Polymer 44 (2003) 3221–3228 3227 esting in view of so enhanced organization to micelles and (co)polymerizability [4,6,13,28,29]. Therefore successful use of p-styrylalkanols as initiators for EO polymerization is valuable for application. The results of preparation of the PEO macromonomers by use of VPP ðm ¼ 3Þ; VPPT ðm ¼ 5Þ; and VPH ðm ¼ 6Þ are summarized in Table 2 together with typical 1H NMR spectra in Fig. 6. The agreements in the number-average molecular weights by 1H NMR ðMn;NMR Þ; SEC ðMn;SEC Þ; and calculation ðMn;calc Þ are usually satisfactory to support the living polymerization mechanism in Scheme 2. Some difference observed in the values of Mn appears to be due to probable errors involved in calibration of very small amounts of the alcohols charged and calibration of SEC with poly(ethylene glycols). Thus we conclude that all the alcohols charged are effectively incorporated as the initiating fragments of the PEO macromonomers to initiate polymerization of EO in living fashion. 4. Conclusions Partially alkoxidated alcohols, including VBA, VPP, VPPT, and VPH (m ¼ 1; 3, 5, 6) successfully initiated polymerization of EO to afford the expected a-styrylalkyland v-hydroxy-ended PEO macromonomers, just as shown in Scheme 2, with the degree of polymerization controlled by initial ratio of EO/alcohol. VPE ðm ¼ 2Þ; however, gave p-divinylbenzene and PEG very probably as a result of intramolecular chain transfer as given in Scheme 3. The initiation method proposed appears applicable to design of various kinds of hetero-telechelic PEO macromonomers and polymers in general. The study along this line as well as application of the macromonomers to emulsion and dispersion polymerization are to be published in due course. Fig. 6. Typical 1H NMR spectra of PEO macromonomers prepared from (A) VPP (m ¼ 3; n ¼ 50); (B) VPPT (m ¼ 5; n ¼ 63), and (C) VPH (m ¼ 6; n ¼ 50). Peaks with an arrow are due to impurities (CHCl3, C6H6, and H2O from low to upfield). Table 2 Characterization of PEO macromonomers obtained at complete conversion of EO polymerization with VPP ðm ¼ 3Þ; VPPT ðm ¼ 5Þ; and VPH ðm ¼ 6Þ Alcohol (mmol) xa EO (mmol) Time (h) Mn;calc b Mn;NMR c Mn;SEC d Mw =Mn;NMR d VPP 6.3 VPP 4.3 VPP 3.8 VPP 3.7 VPPT 8.9 VPPT 4.3 VPH 5.7 VPH 8.1 0.50 0.27 0.40 0.43 0.63 0.49 0.61 0.36 240 208 343 522 298 247 278 262 48 72 64 54 72 72 72 168 1820 2110 4150 6390 1660 2720 2340 1640 2340 2080 3660 6740 2130 2990 2650 2430 1970 2000 2830 4390 2300 2930 2320 2000 1.07 1.09 1.16 1.12 1.12 1.10 1.10 1.24 THF ¼ ca. 90 mL, 40 8C, conversion ¼ nearly quantitative. Degree of alkoxidation, x ¼ ½KC10 H8 Š=½VBAŠ0 : b Mn;calc ¼ MA þ 44½MŠ0 =½AŠ0 ; where MA ¼ molecular weight of the initiator alcohol used. c Mn;NMR ¼ MA þ 44ðIEO =4Þ=ðIA =2Þ; where IEO ¼ peak intensity of the oxyethylene protons and IA ¼ peak intensity of the benzylic methylene protons of the initiator alcohol fragnents. d Determined by SEC calibrated with standard poly(ethylene glycol)s. a 3228 R. Shen et al. / Polymer 44 (2003) 3221–3228 References [1] (a) Guyot A. In: Arshady R, Guyot A, editors. Functional polymer colloids & microparticles. London: Citus Books; 2002. p. 77. (b) Ito K, Cao J, Kawaguchi S. In: Arshady R, Guyot A, editors. Functional polymer colloids & microparticles. London: Citus Books; 2002. p. 109. [2] (a) Ito K. Prog Polym Sci 1998;23:581. (b) Ito K, Kawaguchi S. Adv Polym Sci 1999;142:129. [3] Ito K, Tanaka K, Tanaka H, Imai G, Kawaguchi S, Itsuno S. Macromolecules 1991;24:2348. [4] Chao D, Itsuno S, Ito K. Polym J 1991;23:1045. [5] Ito K, Hashimura K, Itsuno S, Yamada E. Macromolecules 1991;24: 3997. [6] Nomura E, Ito K, Kajiwara A, Kamachi M. Macromolecules 1997;30: 2811. [7] Kawaguchi S, Akaike K, Zhang Z-M, Matsumoto H, Ito K. Polym J 1998;12:1004. [8] Kawaguchi S, Ito K. Colloids Surf A: Physicochem Engng Aspects 1999;153:173. [9] Kawaguchi S, Maniruzzaman M, Katsuragi K, Matsumoto H, Iriany, Ito K, Hugenberg N, Schmidt M. Polym J 2002;34:253. [10] Maruyama R, Maniruzzaman M, Kawaguchi S, Ito K. Macromol Symp 2002;179:133. [11] Maniruzzaman M, Kawaguchi S, Ito K. Macromolecules 2000;33: 1583. [12] Kawaguchi S, Tano K, Maniruzzaman M, Ito K. Macromol Symp 2000;150:101. [13] Kawaguchi S, Winnik MA, Ito K. Macromolecules 1995;28:1159. [14] Nugroho MB, Kawaguchi S, Ito K, Winnik MA. Macromol Rep 1995; A32:593. [15] Sheikh MRK, Imae I, Tharanikkarasu K, LeStarat VM-J, Kawakami Y. Polym J 2000;32:527. [16] Shen R, Kaneta M, Nakao W, Ito K. Asia Polymer Symposium, Preprints, Toyohashi, Japan; 8–12 July 2001. p. 52. We tried polymerization of styrene with K-alkoxide of methoxy-poly(ethylene [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] glycol) in THF or toluene under vacuum but could not find any sign of polymerization against the report in Ref. [17] even after treatment with 1,1-diphenylethylene at 278 to 40 8C. Upon addition of excess cryptand [222] according to Ref. [18] together with excess 1,1diphenylethylene in toluene, yellow–red color appeared due to carbanions, which could initiate to polymerize styrene quantitatively to afford PEO-b-polystyrene, but the initiation efficiency was as low as 1.7% as judged from the too high degree of polymerization observed for polystyrene block. Stolarzewicz A, Czaja M, Neugebauer D. Polymer 2000;41:7865. Boileau S, Hemery B, Keampf B, Schue F, Viguier M. J Polym Sci: Polym Lett Ed 1974;12:217. Mason P, Beinert G, Franta E, Rempp P. Polym Bull 1982;7:17. (a) Soula O, Guyot A. Langmuir 1999;15:7956. (b) Soula O, Petiaud R, Liauro M-F, Guyot A. Macromolecules 1999;32:6938. (a) Abramo JG, Chapin EC. J Org Chem 1961;26:2671. (b) Emerson WS, Herd JW, Lucas VE, Owens GR, Shortridge RW. J Am Chem Soc 1947;69:1905. Tanimoto S, Oda R. Kogyo Kagaku Zasshi 1961;69:932. Tomoi M, Ogawa E, Yokoyama Y, Kakiuchi H. J Polym Sci, Polym Chem Ed 1982;20:3015. (a) Tamura M, Kochi J. Synthesis 1971;303. (b) Friedman L, Shani A. J Am Chem Soc 1973;96:7101. Gilman H, Gatlin WE, Organic Syntheses, Collect. Vol. 1.; 1956. p. 188. (b) Reutenauer S, Hurtrez G, Dumas P. Macromolecules 2001; 34:755. (a) Endo M, Aida T, Inoue S. Macromolecules 1987;20:2982. (b) Aida T, Maekawa Y, Asano S, Inoue S. Macromolecules 1988;21:1195. Szwarc M. Adv Polym Sci 1983;49:1. Furuhashi H, Kawaguchi S, Itsuno S, Ito K. Colloid Polym Sci 1997; 275:227. (a) Liu J, Chew CH, Gan LM. J Macromol Sci-Pure Appl Chem 1996; A33:337. (b) Liu J, Chew CH, Wong SY, Gan LM. J Macromol SciPure Appl Chem 1996;A33:1181.