9946
Langmuir 2005, 21, 9946-9954
Biomimetic Stimulus-Responsive Star Diblock Gelators† Yuting Li,‡ Yiqing Tang,‡ Ravin Narain,‡ Andrew L. Lewis,§ and Steven P. Armes*,‡ Department of Chemistry, Dainton Building, University of Sheffield, Brook Hill, Sheffield S3 7HF, U.K., and Biocompatibles, Chapman House, Farnham Business Park, Weydon Lane, Farnham, Surrey GU9 8QL, U.K. Received February 8, 2005. In Final Form: June 13, 2005 Novel biomimetic gelators with star diblock copolymer architectures have been synthesized by atomtransfer radical polymerization (ATRP). Two types of trifunctional ATRP initiator were used to polymerize 2-(methacryloyloxy)ethyl phosphorylcholine [MPC] at 20 °C, followed by sequential monomer addition of various tertiary amine methacrylates or mixtures thereof. Poor living character was achieved using an amide-based trifunctional initiator, but the analogous triester initiator gave reasonably well-defined thermoresponsive and pH-responsive star diblock copolymers. The most effective thermo-responsive gelators were obtained by the statistical terpolymerization of 2-(dimethylamino)ethyl methacrylate [DMA], 2-(diethylamino)ethyl methacrylate [DEA], and a monomethoxy-capped poly(propylene oxide) methacrylate [PPOMA], whereas pH-responsive gelators were prepared using 2-(diisopropylamino)ethyl methacrylate [DPA] as the second monomer. Star diblock copolymer gelators that were both thermo-responsive and pH-responsive were obtained by the statistical copolymerization of DMA with DPA. Copolymer compositions were assessed by 1H NMR spectroscopy, and the molecular weight distributions of the three-arm star MPC homopolymer precursors were assessed by aqueous gel permeation chromatography. Static light scattering was used to obtain weight-average molecular weights of selected star diblock copolymers and rheological measurements and variable-temperature 1H NMR were used to probe the onset of gelation.
Introduction 1-3
Hydrogels that can change from free-flowing liquids to free-standing gels on application of an external stimulus have recently attracted increasing attention in the context of drug delivery4-7 and tissue engineering.8 Typical gelator molecules normally comprise either amphiphilic block or graft copolymers that can self-assemble in water to form organized structures on the nanometer length scale. Gelation can be triggered by changes in solution temperature, solution pH, ionic strength, electric field, and light or in the presence of specific analytes.6,9,10 Many water-soluble polymers exhibit a lower critical solution temperature (LCST) at around room temperature. In principle, such polymers can be used to produce injectable biomaterials that are capable of in situ gelation at physiologically relevant temperatures (around 37 °C). Indeed, there are a few preliminary reports concerning thermo-responsive polymeric gelators, but usually either †
Part of the Bob Rowell Festschrift special issue. * To whom correspondence should be addressed. E-mail:
[email protected]. ‡ University of Sheffield. § Biocompatibles. (1) Yamamoto, K.; Serizawa, T.; Muraoka, Y.; Akashi, M. Macromolecules 2001, 34, 8014. (2) Wang, C.; Kopecek, J.; Stewart, R. J. Biomacromolecules 2001, 2, 912. (3) Zhang, J. T.; Cheng, S. X.; Huang, S. W.; Zhuo, R. X. Macromol. Rapid Commun. 2003, 24, 447. (4) Davis, K. A.; Anseth, K. S. Crit. Rev. Ther. Drug Carrier Syst. 2002, 19, 385. (5) Nam, K. W.; Watanabe, J.; Ishihara, K. J. Biomater. Sci., Polym. Ed. 2002, 13, 1259. (6) Soppimath, K. S.; Aminabhavi, T. M.; Dave, A. M.; Kumbar, S. G.; Rudzinski, W. E. Drug Dev. Ind. Pharm. 2002, 28, 957. (7) Hatefi, A.; Amsden, B. J. Controlled Release 2002, 80, 9. (8) Luo, Y.; Shoichet, M. S. Nat. Mater. 2004, 3, 249. (9) Nagarsekar, A.; Crissman, J.; Crissman, M.; Ferrari, F.; Cappello, J.; Ghandehari, H. Biomacromolecules 2003, 4, 602. (10) Stayton, P. S.; Shimoboji, T.; Dong, C.; Chilkoti, A.; Chen, G. H.; Harris, J. M.; Hoffman, A. S. Nature 1995, 378, 472.
their biocompatibility is insufficient or else the minimum copolymer concentration required for gelation is relatively high.11-13 Deming and co-workers have reported the synthesis of efficient polypeptide-based gelators, but these examples are not stimulus-responsive. Moreover, the N-carboxyanhydride monomers used in these studies are not commercially available.14 2-(Methacryloyloxy)ethyl phosphorylcholine (MPC) is a zwitterionic methacrylic monomer that has received increasing attention because its copolymerization confers clinically proven biocompatibility to a wide range of copolymer coatings.15-17 Recently we showed that MPC can be polymerized with reasonably good control using atom-transfer radical polymerization (ATRP).18,19 This synthetic advance allowed a wide range of new biocompatible diblock copolymers to be prepared, with selected examples being used (i) to form pH-sensitive micelles for the delivery of hydrophobic drugs or (ii) as new synthetic vectors for DNA condensation, respectively.20 Subsequently, we reported the synthesis of novel, biocompatible pH-responsive gelators based on ABA triblock copolymers, (11) Oh, K. S.; Han, S. K.; Choi, Y. W.; Lee, J. H.; Lee, J. Y.; Yuk, S. H. Biomaterials 2004, 25, 2393. (12) Matthew, J. E.; Nazario, Y. L.; Roberts, S. C.; Bhatia, S. R. Biomaterials 2002, 23, 4615. (13) Pisal, S. S.; Paradkar, A. R.; Mahadik, K. R.; Kadam S. S. Int. J. Pharm. 2004, 270, 37. (14) Nowak, A. P.; Breedveld, V.; Pine, D. J.; Deming, T. J. J. Am. Chem. Soc. 2003, 125, 15666. (15) Moro, T.; Takatori, Y.; Ishihara, K.; Konno, T.; Takigawa, Y.; Matsushita, T.; Chung, U. I.; Nakamura, K.; Kawaguchi, H. Nat. Mater. 2004, 3, 829. (16) Lewis, A. L. Colloids Surf., B 2000, 18, 261. (17) Uchiyama, T.; Watanabe, J.; Ishihara, K. J. Membr. Sci. 2002, 208, 39. (18) Lobb, E. J.; Ma, Y.; Billingham, N. C.; Armes, S. P.; Lewis, A. L. J. Am. Chem. Soc. 2001, 123, 7913. (19) Ma, Y.; Lobb, E. J.; Billingham, N. C.; Armes, S. P.; Lewis, A. L.; Lloyd, A. W.; Salvage, J. P. Macromolecules 2002, 35, 9306. (20) Ma, Y.; Tang, Y.; Billingham, N. C.; Armes, S. P.; Lewis, A. L.; Lloyd, A. W.; Salvage, J. P. Macromolecules 2003, 36, 3475.
10.1021/la050356u CCC: $30.25 © 2005 American Chemical Society Published on Web 07/14/2005
Stimulus-Responsive Star Diblock Gelators
where the central B block comprises MPC and the outer A blocks are composed of 2-(diisopropylamino)ethyl methacrylate (DPA).21 These ABA triblocks were readily synthesized via sequential monomer addition using a commercially available bifunctional ATRP initiator, diethyl meso-2,5 dibromoadipate. Free-standing gels were obtained at neutral pH by adding sufficient base to freeflowing acidic solutions to ensure deprotonation of the outer DPA blocks. Our hypothesis of a micellar gel network was subsequently confirmed by small-angle neutron scattering studies, which revealed hydrophobic domains with length scales comparable to those expected for DPAcore micelles.22 Thus, some of the triblock copolymer chains act as bridges between micelles, leading to physical cross linking. This gelation is completely reversible: addition of acid leads to immediate dissolution of the gel network. Optimization of the triblock composition allowed us to prepare free-standing gels from 10% triblock copolymer solutions, and preliminary drug release studies were also conducted using dipyridamole as a model hydrophobic drug.21 In view of these results, we wished to explore the synthesis of stimulus-responsive gelators based on star diblock copolymer architectures, with the aim being to improve the gelator efficiency. This is an important objective given the relatively high cost of the MPC monomer. Our initial results were reported in a recent communication.23 Herein we report in detail the synthesis and characterization of these new biomimetic gelators. In all cases, the inner blocks are composed of MPC, whereas the outer blocks are either homopolymers or statistical copolymers comprising one or more of 2-(dimethylamino)ethyl methacrylate (DMA), 2-(diethylamino)ethyl methacrylate (DEA), 2-(diisopropylamino)ethyl methacrylate (DPA), and monomethoxy-capped poly(propylene oxide) (PPOMA) macromonomer. Preliminary rheological data of the gelation behavior of selected copolymers in aqueous solution is also presented. Experimental Section Materials. 2-(Methacryloyloxy)ethyl phosphorylcholine [MPC] (>99.9% purity) was a gift from Biocompatibles (Farnham, Surrey, U.K.). 2-(Dimethylamino)ethyl methacrylate (DMA) and 2-(diethylamino)ethyl methacrylate (DEA) were purchased from Aldrich. 2-Hydroxyethyl methacrylate (HEMA) was donated by Cognis Performance Chemicals (Hythe, U.K.). 2-(Diisopropylamino)ethyl methacrylate (DPA) was purchased from Scientific Polymer Products. Cu(I)Br, 2,2′-bipyridine (bpy), and methanol were purchased from Aldrich and were used as received. The water used in all experiments was deionized and doubly distilled prior to use. The silica used for removal of the ATRP copper catalyst was column chromatography-grade silica gel 60 (0.0630.200 mm) purchased from E. Merck (Darmstadt, Germany). Monohydroxy-capped poly(propylene oxide) (PPO) (Mn ) 1940 by 1H NMR and Mw/Mn ) 1.18 by THF GPC) was donated by Cognis Performance Chemicals (Hythe, U.K.). Synthesis of the PPOMA Macromonomer. Methacryloyl chloride (6.27 g, 60 mmol, 6.0 equiv) was added dropwise to a toluene solution (100 mL) of PPO (20.00 g, 10 mmol, 1.0 equiv) and triethylamine (6.06 g, 60 mmol, 6.0 equiv) under nitrogen. This mixture was stirred for 7 days at 20 °C and then filtered to remove the triethylamine hydrochloride byproduct. The solution was then washed three times with an aqueous solution of 0.1 M Na2CO3, followed by three washings with water. This solution was dried over anhydrous MgSO4, and the solvent was (21) Ma, Y.; Tang, Y.; Billingham, N. C.; Armes, S. P.; Lewis, A. L. Biomacromolecules 2003, 4, 864. (22) Castelletto, V.; Hamley, I. W.; Ma, Y.; Bories-Azeau, X.; Armes, S. P.; Lewis, A. L. Langmuir 2004, 20, 4306. (23) Li, Y.; Narain, R.; Ma, Y.; Lewis, A. L.; Armes, S. P. Chem. Commun. 2004, 23, 2746.
Langmuir, Vol. 21, No. 22, 2005 9947 removed under reduced pressure. The final macromonomer product was obtained as a slightly yellow liquid (17.8 g, 85%) and was stored in a freezer in the absence of light prior to use. 1H NMR spectroscopy indicated a mean degree of esterification of at least 99%. Synthesis of the TrisA Initiator. Tris-(2-aminoethyl)amine (10.0 mL, 9.77 g, 0.066 mol) was added to dry THF (200 mL). Excess triethylamine (60.0 mL, 0.43 mol) was then added, and the mixture was stirred under a nitrogen atmosphere. The solution was cooled in an ice bath, and 2-bromoisobutyryl bromide (26 mL, 46 g, 0.20 mol) was added dropwise from a dropping funnel over a period of 1 h. The mixture was stirred for another 2 h. The white precipitate was removed by filtration, and the pale-yellow solution was concentrated under vacuum at 30 °C. The resulting viscous yellowish syrup was cooled in an ice bath. The solid that was formed was stirred in distilled water, filtered, and washed copiously with water. A pale-yellow powder was isolated after drying under vacuum for 24 h at ambient temperature (yield 85%). 1H NMR: 3.2 ppm (6 H, triplet, N-(CH2CH2N)); 2.6 ppm (6 H, triplet, N-(CH2CH2N)); 1.83 ppm (18 H, singlet, NCOC(CH3)2Br). Synthesis of the TrisE Initiator. Triethanolamine (5.00 g, 33.5 mmol) was added to dry THF (200 mL). Excess triethylamine (60 mL, 0.43 mol) was added, and the mixture was stirred under a nitrogen atmosphere. After cooling the solution in an ice bath, 2-bromoisobutyryl bromide (24.7 mL, 46.0 g, 0.20 mol) was added dropwise to the mixture using a dropping funnel over a period of 1 h. The solution slowly became reddish brown in color. After stirring the mixture for another 2 h at 20 °C, the triethylamine hydrochloride salt was removed by filtration. The resulting clear solution was concentrated under vacuum at 30 °C and stirred with 0.1 M Na2CO3 so as to hydrolyze any residual acid bromide. The product was then extracted three times with dichloromethane using a separating funnel. The combined dichloromethane extracts were first dried with anhydrous MgSO4 and then concentrated to give a dark reddish-brown oil (16.0 g, 80%) that was stored at 4 °C prior to use. 1H NMR: 4.2-4.3 ppm (6 H, triplet, N-(CH2CH2O)); 2.9-3.0 ppm (6 H, triplet, N-(CH2CH2O)); 1.9 ppm (18 H, singlet, OCOC(CH3)2Br). General Polymerization Protocols. 1. Homopolymerization of MPC. A typical protocol for the homopolymerization of MPC via ATRP in methanol using the trifunctional ester-based initiator was as follows: The TrisE initiator (19.8 mg, 0.0333 mmol, 1 equiv) and MPC (3.7 g, 12.5 mmol, 375 equiv) were placed together in a two-necked Schlenk flask. After purging with nitrogen for 30 min, 5.0 mL of degassed methanol was added, and the monomer and initiator dissolved quickly within 2 min. The Cu(I)Br catalyst (14.3 mg, 0.1 mmol, 3 equiv) and bpy ligand (31.2 mg, 0.2 mmol, 6 equiv) were added to this stirred solution under nitrogen. The reaction mixture immediately became dark brown and progressively more viscous, indicating the onset of polymerization. After approximately 90 min, 1H NMR analysis (relative attenuation of vinyl signals between δ 5.5 and 6.1) indicated that more than 95% of the MPC had been polymerized. The reaction solution turned blue on exposure to air, indicating aerial oxidation of the Cu(I) catalyst. The resulting MPC homopolymer was diluted with methanol and passed through a silica column to remove the spent ATRP catalyst. The polymer solution was dried under vacuum to remove the solvent. 2. Self-Blocking (Chain Extension) Experiments. MPC (1.85 g, 6.23 mmol) was polymerized in methanol (2.5 mL) using an [MPC]/[I]/[Cu(I)Br]/[bpy] relative molar ratio of 180:1:3:6 (where I refers to either TrisA or TrisE). After 50 min, the monomer conversion had reached about 95%, as judged by 1H NMR spectroscopy. At this point, an aliquot of the polymerization solution was extracted for subsequent characterization, and a second batch of degassed MPC (1.85 g, 6.23 mmol; dissolved in 2.5 mL of methanol) was added to the polymerizing solution. 1H NMR analysis after 20 h indicated the absence of any vinyl signals (formerly at δ 5.5 and 6.1), suggesting that the overall conversion was approximately 100%. 3. Block Copolymerization of MPC with Other Methacrylic Monomers. The following examples illustrate the general synthetic protocols employed. MPC was polymerized first (3.7 g, 12.5 mmol; dissolved in 5.0 mL methanol) with the TrisE initiator using an [MPC]/[TrisE]/[CuBr]/[bpy] relative molar ratio of 375:
9948
Langmuir, Vol. 21, No. 22, 2005
Li et al.
Figure 1. Reaction scheme for the synthesis of the thermo-responsive star diblock copolymers via atom-transfer radical polymerization (ATRP) using a trifunctional TrisE ATRP initiator. 1:3:6. After 80 min, the monomer conversion was about 95%. A comonomer mixture containing DMA monomer (1.10 g, 7.0 mmol, 70 equiv), DEA monomer (0.556 g, 3.0 mmol, 30 equiv), and PPOMA (1.035 g, 0.50 mmol, 5 equiv; overall target degree of polymerization (Dp) for this terpolymer block ) 105) dissolved in 4.0 mL of methanol was then added to this reaction solution. The reaction mixture was maintained under a dry nitrogen purge for the duration of the polymerization. On exposure to air after 24 h, the reaction solution turned blue, indicating aerial oxidation of the ATRP catalyst. 1H NMR studies indicated that the overall conversion of the three comonomers was >95%. The reaction solution was passed through a silica gel column to remove the spent catalyst. Most of the solvent was then evaporated to form a 20-30% solution, and excess n-hexane was added to this stirred solution to precipitate the copolymer. This precipitation cleanup was repeated several times to ensure that all of the unreacted monomers (DMA, DEA, PPOMA) were completely removed. The copolymer was then dried in a vacuum oven at room temperature to produce a white solid (5.37 g). Polymer Characterization. 1H NMR Spectroscopy. All 1H NMR spectra were recorded using a 300 MHz Bruker Avance DPX300 spectrometer. The kinetics of polymerization were determined for reactions carried out in either D2O or CD3OD by comparing the peak integrals due to the monomer vinyl signals at δ 5.5 and 6.1 to those of the methacrylate backbone at δ 0.51.5. GPC Protocols. The molecular weights and molecular weight distributions of the MPC homopolymer precursors were determined by aqueous gel permeation chromatography (GPC). The standard GPC protocol involved using a Pharmacia Biotech Superose 6 column connected to a Polymer Labs ERC-7517A refractive index detector. The eluent was a 0.20 M NaNO3 solution with 50 mM Trizma buffer at pH 7. Calibration was based on poly(ethylene oxide) standards ranging from 440 to 288 000 g mol-1. Determination of the Residual ATRP Catalyst Level. After silica treatment to remove the ATRP catalyst, aqueous solutions of selected, purified MPC copolymers were analyzed for their copper contents using a Perkin-Elmer Plasma 400 inductively-coupled plasma atomic emission spectrometer (ICP-AES). Static Light Scattering (SLS) Studies. Static light scattering (SLS) studies were performed using a DAWN DSP laser photometer equipped with a 5 mW He-Ne laser (λ ) 633 nm) and 18 photodiode detectors at scattering angles ranging from 22.5 to 147°. The dn/dc values of the final copolymer solutions were determined using an Optilab DSP interferometric refractometer (λ ) 633 nm). Rheology Measurements. The flow and viscoelastic properties of aqueous solutions of the star block copolymers were measured
using a Rheometrics SR-5000 stress-controlled rheometer. A 6 cm cone with an angle of 2° was used in these measurements. Selected star copolymers were dissolved in a pH 7.4 buffer and stored at 5 °C overnight prior to analysis. In the flow measurements, samples were presheared for 3 min at an applied shear stress of 5 Pa. Measurements of the shear storage modulus G′ and loss modulus G′′ involved temperature ramps at a frequency of 1 Hz and dynamic frequency sweeps at a constant stress of 5 Pa at 20 and 37 °C, respectively.
Results and Discussion Previously we reported that reasonably well-defined DPA-MPC-DPA triblock copolymers could be readily prepared using sequential monomer addition in conjunction with a bifunctional ATRP initiator.21 These copolymers were demonstrated to be pH-responsive gelators, with the best examples forming free-standing gels at around neutral pH at copolymer concentrations of approximately 10 w/v %.21,22 However, our attempts to prepare the analogous thermo-responsive triblock copolymer gelators using DMA24 or 2-hydroxyethyl methacrylate25 in place of DPA were much less successful. The best examples produced only weak gels at relatively high temperatures (80-90 °C), and in many cases, gelation did not occur at all. In view of these negative results, we decided to examine alternative copolymer architectures. More specifically, we anticipated that three-arm-star diblock copolymer architectures might be more effective because the extra functionality should promote more efficient formation of the required micellar gel network. Figure 1 shows the reaction scheme for the synthesis of the target star diblock copolymers via atom-transfer radical polymerization (ATRP) using a trifunctional ATRP initiator. It is noteworthy that, provided that gelation can be induced within a physiologically relevant temperature range, thermo-responsive gelators offer several advantages over pH-responsive gelators for biomedical applications. This is because the latter gelators must necessarily exist at nonphysiological conditions either before or after the pH switch. Moreover, such a pH change is usually (24) Lowe, A. B.; Billingham, N. C.; Armes, S. P. Macromolecules 1998, 31, 5991. (25) Weaver, J. V. M.; Bannister, I.; Robinson, K. L.; Bories-Azeau, X.; Armes, S. P. Macromolecules 2004, 37, 2395.
Stimulus-Responsive Star Diblock Gelators
rather exothermic and involves a buildup of background salt during pH cycling, especially for semiconcentrated copolymer solutions. These disadvantages either do not apply or are less problematic in the case of thermoresponsive gelators. Trisamide (TrisA) and triester (TrisE) trifunctional ATRP initiators were prepared by reacting either tris(2-aminoethyl)amine or triethanolamine, respectively, with excess 2-bromoisobutyryl bromide. The 1H NMR spectra indicated high degrees of purity for the two purified initiators. In our preliminary experiments, we evaluated the homopolymerization of MPC using TrisA and TrisE: the molecular weight distributions of the resulting threearm star MPC homopolymers were characterized by aqueous GPC, and 1H NMR spectroscopy was used to monitor the kinetics of polymerization. A shorthand notation is used to refer to these star copolymers. Thus, an MPC three-arm-star homopolymer prepared using the TrisE initiator with a target degree of polymerization of 50 is referred to simply as E-[MPC50]3, whereas the same copolymer prepared with the TrisA initiator is referred to as A-[MPC50]3. Similarly, E-[(MPC125)-(DPA)50]3 denotes a star diblock copolymer prepared using the TrisE initiator that has an MPC inner block with a target degree of polymerization of 125 per arm and a DPA outer block with a target degree of polymerization of 50 per arm. Finally, E-[(MPC125)(DMA50/DEA50/PPOMA3)]3 refers to a star diblock copolymer in which the outer block comprises a statistical terpolymer based on DMA, DEA, and PPOMA that has an overall target degree of polymerization of 103 per arm. Ma et al. reported19 that the ATRP of MPC in methanol using a monofunctional initiator has reasonably good living character (i.e., linear evolution of Mn with conversion and relatively low final polydispersities) when conducted in methanol at 20 °C. However, poorer results were obtained for target degrees of polymerization greater than 150, which is a typical observation for ATRP syntheses. For efficient gelator performance, it is clear that the degree of polymerization of the inner MPC block should be relatively high, so less than ideal living character was expected for the star diblock copolymer syntheses. Moreover, a survey of the many literature examples of ATRPsynthesized star polymers reported confirms that, in almost all cases, polymerizations are deliberately stopped at relatively low conversions in order to minimize termination by combination. In the present study, we are synthesizing star diblock copolymers via sequential monomer addition, thus the very high conversions required for the first monomer (MPC) coupled with the relatively high target degrees of polymerization are likely to compromise the degree of living character that can be achieved. The kinetic results obtained for the trifunctional amidebased initiator (TrisA) are shown in Figure 2. The target Dp of each arm is 125, which gives a total Dp of 375. Although the semilogarithmic plot was linear up to 95%, the molecular weight versus conversion plot was nonlinear, and there was no evidence for a reduction in polydispersity with conversion, with final polydispersities being around 1.55. A self-blocking experiment (Figure 3) shows that, although the final target Dp of the chain-extended polymer was more than double the initial target Dp, aqueous GPC analyses indicated that the molecular weight increased from only 41 500 to 59 500, with almost no change in the polydispersity. Overall, these results suggest relatively poor living character for this TrisA initiator. These observations are consistent with work by Heming, who has shown that amide-based ATRP initiators usually have
Langmuir, Vol. 21, No. 22, 2005 9949
Figure 2. Homopolymerization of MPC via methanol ATRP at 20 °C using the trisamine-based (TrisA) initiator: (a) conversion vs time data and semilogarithmic plot of monomer concentration vs time (conditions: 42% monomer concentration; target Dp ) 375; TrisA/CuBr/bpy ) 1:3:6) and (b) evolution of molecular weight and polydispersity vs conversion; conditions as stated in part a.
Figure 3. Self-blocking experiments in methanol at 20 °C performed using the trisamine-based (TrisA) initiator. Conversion of the first batch of MPC prior to chain extension was 95%. Overall final MPC conversion was 100%. Conditions: first batch of MPC (1.85 g, 6.2 mmol), 2.5 mL of methanol, target Dpn ) 187, molar ratio of [TrisA]/[Cu(I)]/[L] ) 1:3:6; second batch of MPC (1.85 g, 6.2 mmol).
poor initiator efficiencies, lead to lower monomer conversions, and produce polymers with relatively high polydispersities.26 To attempt to improve the living character of these polymerizations, the trifunctional ester-based initiator, TrisE, was evaluated. Kinetic results obtained with the TrisE initiator are shown in Figure 4. The semilogarithmic plot was linear up to 95%, and the evolution of molecular weight with conversion plot was linear up to 95%. The final polydispersity remained relatively high at 1.48, but given the high target Dp, the high conversion, and the star architecture, these results were considered quite encouraging. Moreover, the selfblocking experiment (Figure 5) confirmed that when the target Dp was increased from 185 to 375 the GPC molecular weight approximately doubled from 24 000 to 51 800 and the polydispersity decreased from 1.67 to 1.37. Thus, the ATRP of MPC with the TrisE initiator under these conditions has useful, if not ideal, living character. 2-(Dimethylamino)ethyl methacrylate (DMA) is a hydrophilic monomer with applications in textiles, pigment (26) Heming, A. M. Ph.D. Thesis, University of Warwick, October 1999.
9950
Langmuir, Vol. 21, No. 22, 2005
Figure 4. Homopolymerization of MPC via methanol ATRP at 20 °C using the triester-based (TrisE) initiator: (a) conversion vs time data and semilogarithmic plot of monomer concentration vs time (conditions: 42% monomer concentration; target Dp ) 375; TrisE/CuBr/bpy ) 1:3:6) and (b) evolution of molecular weight and polydispersity vs conversion; conditions as stated in part a.
Figure 5. Self-blocking experiments in methanol at 20 °C with the triester-based TrisE initiator. Conversion of the first batch of MPC prior to chain extension was 95%. Overall final MPC conversion was 100%. Conditions: first batch of MPC (1.85 g, 6.2 mmol), 2.5 mL of methanol, target Dpn ) 187, molar ratio of [TrisE]/[Cu(I)]/[L] ) 1:3:6; second batch of MPC (1.85 g, 6.2 mmol).
dispersion, production of hydrogels, and as the cationic component of synthetic vectors for DNA complexation.27-30 DMA homopolymer is soluble in aqueous media at low pH as a weak cationic polyelectrolyte (pKa ) 7.0 ( 0.531-33) because of protonation of the tertiary amine groups. At pH 8 or above, DMA homopolymer has low or zero charge density, thus hydrogen bonding is solely responsible for its water solubility. Under these conditions, the neutral DMA chains exhibit inverse temperature solubility, and the observed LCST (or cloud point) varies from around 32 to 46 °C depending on its degree of polymerization, with shorter chains being more hydrophilic.33 Thus, in principle (27) Creutz, S.; Jerome, R.; Kaptijn, G. M. P.; van der Werf, A. W.; Akkerman, J. M. J. Coat. Technol. 1998, 70, 41. (28) Creutz, S.; Jerome, R. Prog. Org. Coat. 2000, 40, 21. (29) Sassi, A. P.; Blanch, H. W.; Prausnitz, J. M. J. Appl. Polym. Sci. 1996, 59, 1337. (30) Rungsardthong, U.; Deshpande, M.; Bailey, L.; Vamvavaki, M.; Armes, S. P.; Garnett, M. C.; Stolnik, S. J. Controlled Release 2001, 73, 359. (31) Pradny, M.; Seveik, S. Makromol. Chem. 1985, 186, 111. (32) Merle, Y. J. Phys. Chem. 1987, 91, 3092.
Li et al.
it is possible to prepare technologically useful thermoresponsive star diblock copolymers using just MPC and DMA as the monomer building blocks. Unfortunately, star diblock copolymers such as E-[MPC125-DMA100]3 (see entry 1 in Table 2) formed only weak gels at relatively high temperatures (approximately 80 °C), even at 20% copolymer concentration. Similar results were obtained by Li et al. for DMA-MPC-DMA triblock copolymers prepared using a bifunctional ester-based ATRP initiator.34 It is known that the cloud points of thermo-responsive polymers such as polyDMA are often elevated if they are conjugated to more hydrophilic polymers such as polyMPC.20 Thus, DMA-based blocks alone do not allow the design of thermoresponsive gelators that operate at physiologically relevant temperatures. Poly(propylene oxide) (PPO) is a thermo-responsive polymer whose cloud point is lower than that of the DMA homopolymer. For example, Liu and Armes et al. reported that PPO-DMA-OEGMA triblocks undergo micellar selfassembly above approximately 15 °C.35 In the present study, monohydroxy-capped PPO (Mn ) 2000) was first converted into a PPOMA macromonomer by reacting with excess methacryloyl chloride. Copolymerization of PPOMA with DMA in a second-stage polymerization after homopolymerization of MPC using the TrisE initiator led to the formation of an E-[MPC-(DMA/PPOMA)]3 star diblock copolymer in which the outer block comprised a thermoresponsive statistical copolymer with pendent PPO chains. Such copolymers can form free-standing gels at 37 °C at a copolymer concentration of 6 w/v % (entry 7, Table 2). Perhaps surprisingly, increasing the PPOMA content of the copolymer did not lead to lower critical gelation temperatures or copolymer concentrations (entries 2-6, Table 2). At present, we are unable to account for this unexpected observation. 2-(Diethylamino)ethyl methacrylate (DEA) is structurally very similar to DMA, but it is significantly more hydrophobic.36 Hence, it was anticipated that replacing some of the DMA with DEA might improve the gelator performance (entries 8 and 9, Table 2). Indeed, a 5.0 w/v % PBS buffer solution of a star diblock copolymer in which the outer blocks comprised a statistical terpolymer of DMA, DEA, and PPOMA (entry 10, Table 2) formed a freestanding gel at 37 °C. Although DEA-based copolymers are typically pH-responsive,37 it is noteworthy that the E-[MPC-(DMA/DEA/PPOMA)]3 star diblock copolymers do not exhibit pH-responsive behavior: simply increasing the solution pH does not lead to gelation. Presumably, the DEA-based outer blocks are not sufficiently hydrophobic at higher pH to form a micellar gel network. Ma et al. previously reported that DPA50-MPC250-DPA50 triblock copolymers could act as pH-responsive gelators at a copolymer concentration of approximately 9-10 w/w %.21,22 In the present study, we found that a 5.0 w/v % aqueous solution of the analogous E-(MPC125-DPA50)3 star diblock copolymer can form a free-standing gel at around neutral pH. Clearly, the star diblock architecture leads to significantly better gelator efficiency than the simple ABA triblock architecture. Aqueous GPC was used to assess the molecular weights and molecular weight distributions of the central MPC (33) Bu¨tu¨n, V.; Armes, S. P.; Billingham, N. C. Polymer 2001, 42, 5993. (34) Li, C.; Tang, Y.; Armes, S. P.; Morris, C. J.; Rose, S. F.; Lloyd, A. W.; Lewis, A. L. Biomacromolecules 2005, 6, 994. (35) Liu, S.; Armes, S. P. J. Am. Chem. Soc. 2001, 123, 9910. (36) Lee, A. S.; Gast, A. P.; Butun, V.; Armes, S. P. Macromolecules 1999, 32, 4302. (37) Liu, S.; Tang, Y.; Weaver, J. V. M.; Billingham, N. C.; Armes, S. P.; Tribe, K. Macromolecules 2002, 35, 6121.
Stimulus-Responsive Star Diblock Gelators
Langmuir, Vol. 21, No. 22, 2005 9951
Table 1. Formulation Details for the ATRP Synthesis of MPC-Based Triblock Copolymers entry initiator no. type 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
TrisE TrisA TrisA TrisA TrisA TrisE TrisE TrisE TrisE TrisA TrisE TrisE TrisE TrisE TrisE
target copolymer structure
solvent
E-(MPC125-DMA100)3 A-[MPC125-(DMA93/PPOMA7)]3 A-[MPC125-(DMA90/PPOMA10)]3 A-[MPC125-(DMA93/PPOMA7)]3 A-[MPC125-(DMA93/PPOMA7)]3 E-[MPC125-(DMA93/PPOMA7)]3 E-[MPC125-(DMA95/PPOMA5)]3 E-[MPC125-(DMA97/PPOMA3)]3 E-[MPC125-(DMA50/DEA50)]3 A-[MPC125-(DMA50/DEA50)]3 E-[MPC125-(DMA50/DEA50/PPOMA3)]3 E-[MPC125-(DMA70/DEA30/PPOMA5)]3 E-[MPC125/PPOMA10]3 E-[MPC125-(DMA70/DPA30)]3 E-(MPC125-DPA100)3
methanol methanol/water methanol/water methanol methanol methanol methanol methanol methanol/water methanol methanol methanol methanol methanol methanol
overall MPC DMA DEA PPOMA DPA Cu content conversion (mmol) (mmol) (mmol) (mmol) (mmol) (ppm) % 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5
10.0 9.3 9.0 9.3 9.3 9.3 9.5 9.7 5.0 5.0 5.0 7.0