Article pubs.acs.org/Macromolecules
4‑Vinylphenyl Glycidyl Ether: Synthesis, RAFT Polymerization, and Postpolymerization Modifications with Alcohols David C. McLeod and Nicolay V. Tsarevsky* Department of Chemistry and Center for Drug Discovery, Design, and Delivery at Dedman College, Southern Methodist University, 3215 Daniel Avenue, Dallas, Texas 75275, United States S Supporting Information *
ABSTRACT: 4-Vinylphenyl glycidyl ether (4VPGE), an epoxidecontaining styrenic monomer, was synthesized and then polymerized in a controlled fashion under reversible addition−fragmentation chaintransfer (RAFT) polymerization conditions using butyl 1-phenylethyl trithiocarbonate as the chain-transfer agent. The high degree of chainend functionalization of the produced polymers was confirmed by chain extension reactions with styrene that afforded well-defined block copolymers. Phenyl glycidyl ether was utilized as a model compound to identify the optimal reaction conditions for alcoholysis of the glycidyl moiety using BF3 as a Lewis acid catalyst, and postpolymerization modifications were subsequently carried out on the epoxide groups of poly4VPGE with a library of structurally diverse alcohols to yield a number of β-hydroxy ether-functionalized polymers.
■
INTRODUCTION Epoxides (oxiranes) are potent electrophiles that are readily capable of undergoing ring-opening reactions with a multitude of nucleophiles.1,2 The tremendous versatility of the epoxide group has made low- and high-molecular-weight epoxides indispensable components in many coating and adhesive applications, where reactions with polyfunctional amines, thiols, phenols, carboxylic acids, or anhydrides are employed to form cross-linked resins.3−5 Polymeric materials derived from the radical polymerization of epoxide-containing vinyl monomers are especially interesting because, in addition to the above uses, the polymer-bound epoxide groups can be modified with a wide variety of reactants to produce functional materials with useful properties that might otherwise be inaccessible through direct polymerization of analogous monomers due to incompatibilities between the functional groups and the polymerization process.6−8 The postpolymerization modification approach also makes possible the rapid preparation of libraries of compositionally diverse polymers with identical degrees of polymerization and macromolecular architectures, which greatly facilitates structure−property relationship studies.9 Glycidyl methacrylate (GMA) has been utilized in automotive coating formulations and dental resin composites for decades and is the only epoxide-containing vinyl monomer that is currently produced on an industrial scale. The polymerization of this monomer has been extensively studied, including under controlled/“living” radical polymerization (CRP) conditions, such as atom transfer radical polymerization (ATRP),10−13 reversible addition−fragmentation chain-transfer (RAFT),14,15 and nitroxide-mediated polymerizations.16,17 Although acrylic polymers have excellent weatherability © XXXX American Chemical Society
characteristics, the susceptibility of their ester groups to hydrolysis and transesterification reactions makes them less than ideal for applications where they are routinely exposed to strong nucleophiles, acids, bases, or esterase enzymes. The deterioration of ester-containing dental resin composites as a result of salivary and bacterial enzymatic action18 is a notable example of the deficiency of acrylic polymers in certain environments. Styrenic polymers do not necessarily have hydrolytically labile pendant groups such as esters, so epoxide-containing styrenic polymers could potentially serve as resilient alternatives to GMA-based materials for applications where chemical degradation is an undesirable side effect. Only a few dozen articles concerning the synthesis and polymerization of epoxide-containing styrenic monomers have been published over the past 60 years, likely due to the difficulties associated with their preparation, but improved methodologies for the synthesis and the well-controlled low-catalyst-concentration ATRP19 and RAFT polymerization20 of 4-vinylphenyloxirane (4VPO) have recently been reported. It has also been demonstrated20 that the aryl-oxirane groups of poly4VPO are capable of undergoing efficient postpolymerization modification reactions with various alcohols in the presence of acid catalysts (CBr4 or BF3) to the corresponding β-hydroxy ethers. Such modifications cannot be carried out on polymers bearing glycidyl ester groups, e.g., polyGMA, due to the prevalence of transesterification side reactions. 4-Vinylphenyl glycidyl ether Received: November 9, 2015 Revised: January 29, 2016
A
DOI: 10.1021/acs.macromol.5b02437 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules
Polymer samples were withdrawn from the reaction mixtures, diluted with THF, filtered using Acrodisc 0.2 μm PTFE filters, and injected into the SEC system without any further purification. Prior to spectral analysis, the β-hydroxy ether-containing polymers were carefully purified of unreacted alcohol and catalyst either by dialysis against acetone using Spectra/Por dialysis membrane with MWCO = 2000 Da (wet in 0.05% aqueous NaN3) or by precipitation in either hexane or diethyl ether. Infrared (IR) spectra of thin films of the polymers cast on NaCl plates from chloroform solutions (ca. 20 mg/mL) were collected on a Thermo Scientific Nicolet iS10 FT-IR spectrometer. Synthetic Procedures. Synthesis of 4-Vinylphenol. 4-Acetoxystyrene (10.00 g, 60.12 mmol) was diluted with THF (100 mL) in a round-bottom flask equipped with a magnetic stir bar and then chilled in an ice bath. NaOH (6.02 g, 150 mmol, 2.5 equiv) was dissolved in 30 mL of water and then added dropwise over 5 min to the vigorously stirred solution of 4-acetoxystyrene. After 4 h, 100 mL of 1.5 M HCl chilled in an ice bath was added dropwise over 15 min to the cold, yellow reaction mixture, which was then further diluted with 200 mL of cold water. The mixture was extracted with diethyl ether (2 × 200 mL) using a separatory funnel. The organic phase was collected and dried with MgSO4, and the majority of the solvent was removed under vacuum by rotary evaporation at 25 °C. Ethanol (100 mL, anhydrous) was then added to the product solution, and the majority of the solvent was again evaporated to remove all remaining THF and acetic acid. Note: some ethanol (ca. 10 mL) is allowed to remain to prevent self-initiated cationic polymerization of the monomer.26 The yield, as determined by NMR spectroscopy, was quantitative (7.22 g), and the product was immediately used in the next step without any further purification or storage. 1H NMR (500 MHz, DMSO-d6, δ [ppm]): 9.52 (s, 1H), 7.31−7.25 (m, 2H), 6.76−6.70 (m, 2H), 6.61 (dd, J = 17.7, 10.9 Hz, 1H), 5.58 (dd, J = 17.7, 0.9 Hz, 1H), 5.04 (dd, J = 10.9, 1.0 Hz, 1H). Synthesis of 4VPGE. The solution of 4-vinylphenol (7.22 g, 60.1 mmol) in ethanol (ca. 10 mL) from the previous step was further diluted with anhydrous ethanol (50 mL) in a round-bottom flask equipped with a magnetic stir bar, and NaOH (3.12 g, 78.0 mmol, 1.3 equiv) was added. The mixture was stirred for 30 min to completely dissolve the NaOH, and then epichlorohydrin (14.1 mL, 180 mmol, 3 equiv) was rapidly added to the mixture using a dropping funnel. NaCl slowly precipitated, and the mixture became cloudy. After 12 h, water (100 mL) was added, and the mixture was extracted with hexane (2 × 100 mL) using a separatory funnel. The organic phase was dried with MgSO4, and the solvent was removed under vacuum by rotary evaporation at 35 °C to yield a crude yellow oil (10.71 g), which was composed of 4VPGE, unreacted epichlorohydrin, and an unidentified impurity. Hydroquinone (10 mg) was added to the crude product, and high-vacuum (0.015 Torr (2 Pa)) distillation at 83 °C rendered 4VPGE as a yellow-tinged oil with a small amount of impurity. Alternatively, the majority of the relatively polar impurities could be separated from the product by flash chromatography using hexane/ methylene chloride as an eluent. All traces of impurities were removed by crystallization of 4VPGE in hexane (200 mL) at −20 °C. Final product yield was 5.730 g (32.52 mmol, 54%). 1H NMR (400 MHz, CDCl3, δ [ppm]): 7.38−7.30 (m, 2H), 6.92−6.83 (m, 2H), 6.66 (dd, J = 17.6, 10.9 Hz, 1H), 5.62 (d, J = 17.6 Hz, 1H), 5.14 (d, J = 10.9 Hz, 1H), 4.23 (dd, J = 11.0, 3.1 Hz, 1H), 3.95 (dd, J = 11.0, 5.7 Hz, 1H), 3.36 (ddt, J = 5.8, 4.1, 2.9 Hz, 1H), 2.91 (t, J = 4.5 Hz, 1H), 2.76 (dd, J = 4.9, 2.6 Hz, 1H) (Figure S3). RAFT Polymerization of 4VPGE. In the following procedure, the targeted degree of polymerization at complete conversion of monomer (DPn,targ = [4VPGE]0/[CTA]0) was 200 with 30 mol % radical initiator vs the CTA. 4VPGE (0.500 g, 2.84 mmol) was weighed out in a 5 mL glass vial equipped with a stir bar, and dry anisole (0.3 mL) was added. A 10× stock solution of CTA and radical initiator was prepared by dissolving BPT (38.4 mg, 0.142 mmol) and VAZO88 (10.4 mg, 0.0426 mmol) in anisole (2.0 mL), and a 1/10 portion of the stock solution (0.2 mL) was added to the reaction vial. The vial was capped with a rubber septum and sealed with electrical tape. The reaction mixture was sparged for 30 min with a steady flow of nitrogen and then immersed in an oil bath at 90 °C. Samples (ca. 0.2 mL) were
(4VPGE) is another interesting epoxide-containing styrenic monomer that has been neglected owing to the lack of easy and efficient synthetic procedures. Unlike 4VPO, the epoxide moiety of 4VPGE is not stabilized by an adjacent aryl ring and is known to be more susceptible to ring-opening reactions with nucleophiles under basic or neutral conditions and less reactive under acidic conditions.21,22 Given the similarity in structure, and therefore reactivity, of 4VPGE to GMA and especially bisphenol A diglycidyl ether, two epoxy−resin precursors with well-understood chemistries that are already widely used on an industrial scale, this glycidyl-containing monomer may be more valuable than 4VPO as a nondegradable substitute in some cases. Herein, it is demonstrated that 4VPGE can be synthesized from 4-acetoxystyrene using a two-step method in much higher yield than has been previously reported in a one-step procedure.23,24 To the best of our knowledge, this reactive monomer has not been polymerized under CRP conditions, and we report on the utility of RAFT polymerization to prepare well-defined homopolymers and block copolymers derived from 4VPGE. Additionally, it is shown that poly4VPGE, despite being less reactive than poly4VPO toward nucleophiles under acidic conditions, can also be efficiently modified with alcohols using BF3 as a Lewis acid catalyst and is in some instances actually superior to poly4VPO as a precursor to well-defined βhydroxy ether-functionalized polymers.
■
EXPERIMENTAL SECTION
Materials. The chain transfer agent (CTA), butyl 1-phenylethyl trithiocarbonate (BPT), was synthesized according to the literature.25 Styrene (99%, stabilized with 10−15 ppm 4-tert-butylcatechol, Alfa Aesar) was purified by passing the neat liquid through a short column filled with basic alumina (Brockmann grade I, 58 Å, Alfa Aesar). All other reagents and solvents were used as received: 4-acetoxystyrene (96%, stabilized with hydroquinone monomethyl ether, Aldrich), NaOH (95+%, pellets, Fisher), aqueous HCl (11.6 M, Fisher), epichlorohydrin (99%, ACROS), hydroquinone (99%, Alfa Aesar), MgSO4 (97+%, anhydrous powder, Fisher), azobiscyclohexanecarbonitrile (VAZO88, 98%, Aldrich), phenyl glycidyl ether (PGE, 99%, Aldrich), allyl alcohol (98+%, Alfa Aesar), phenol (99+%, Fisher), propargyl alcohol (99%, ACROS), 1-butanol (BuOH, 99.4+%, Baker), cyclohexanol (99%, Alfa Aesar), benzyl alcohol (99.8%, Aldrich), tert-butanol (t-BuOH, 99+%, Alfa Aesar), L-(−)-menthol (99.5%, Acros), 4-nitrobenzyl alcohol (NBA, 99%, Acros), BF3−Et2O (98+%, Alfa Aesar), anisole (99%, ACROS), acetone (99+%, Fisher), CH2Cl2 (99+%, Fisher), diethyl ether (99+%, Fisher), ethanol (99+%, anhydrous, Koptec), hexanes (99%, Fisher), and tetrahydrofuran (THF; 99%, VWR). The deuterated solvents, DMSO-d6 (99.9% D), CD2Cl2 (99.9%), and CDCl3 (99.8% D), were purchased from Cambridge Isotope Laboratories; either the solvent peak or a small amount of added tetramethylsilane (TMS) was used as a chemical shift reference. Analyses. The structures of all synthetic intermediates, 4VPGE, model compounds, and β-hydroxy ether-functionalized polymers were confirmed by 1H NMR spectroscopy on either a Bruker Avance DRX 400 or JEOL ECA-500 spectrometer operating at 400 or 500 MHz, respectively, using samples diluted in DMSO-d6, CD2Cl2, or CDCl3. Monomer conversions were determined by periodically removing samples from the polymerization reaction mixtures and diluting them with DMSO-d6 or CDCl3 for NMR analysis. Apparent number-average molecular weights (Mn) and molecular weight distribution (MWD) dispersities (Đ = Mw/Mn) of the polymer samples were determined by size exclusion chromatography (SEC) on a Tosoh EcoSEC HLC-8320 system equipped with a series of four columns (TSK gel guard Super HZ-L, Super HZM-M, Super HZM-N and Super HZ2000) using THF as the eluent with a flow rate of 0.35 mL min−1 at 40 °C. The SEC calibration was based on linear narrow-MWD polystyrene standards. B
DOI: 10.1021/acs.macromol.5b02437 Macromolecules XXXX, XXX, XXX−XXX
Article
Macromolecules periodically withdrawn using a nitrogen-purged syringe to determine monomer conversion and molecular weight characteristics. For similar experiments with different DPn,targ, the amount of stock solution added to the monomer was altered accordingly: 0.4 mL stock solution containing 0.0284 mmol of BPT for DPn,targ = 100 and 0.1 mL stock solution containing 0.00710 mmol of BPT for DPn,targ = 400; in both cases the amount of dry anisole added to the monomer was changed so that the total concentration of monomer remained constant at 1:1 (w/v) with respect to solvent. When varying percentages of radical initiator to CTA were desired, the amount of VAZO88 in the stock solution was changed to 3.5 mg (0.0142 mmol for 10 mol % vs BPT), 6.9 mg (0.0284 mmol for 20 mol % vs BPT), or 17.3 mg (0.0710 mmol for 50 mol % vs BPT). Poly4VPGE Macro-CTA Synthesis. In the following procedure, DPn, targ = 100 with 30 mol % radical initiator vs the CTA. 4VPGE (2.000 g, 11.35 mmol) was weighed out in a 10 mL round-bottom flask equipped with a stir bar. A solution of CTA and radical initiator was prepared in a separate vial by dissolving BPT (30.7 mg, 0.114 mmol) and VAZO88 (8.3 mg, 0.034 mmol) in anisole (1.0 mL). The CTA solution was then added to the reaction flask, and an additional 1.0 mL of anisole was used to ensure complete transfer of all reagents. The 10 mL round-bottom flask was capped, sealed, and then sparged for 1 h with a steady flow of nitrogen. The reaction mixture was immersed in an oil bath set at 90 °C for 6 h (63% monomer conversion by NMR) and then quenched by removal from heat and by opening the flask to air. The reaction mixture was transferred to a dialysis bag and diluted with acetone (4 mL). The mixture was then placed in a 500 mL beaker equipped with a stir bar and dialyzed against acetone (300 mL, replaced after 12 h) over 24 h to remove unreacted monomer and anisole. The polymer solution was then transferred from the dialysis bag to a 50 mL round-bottom flask and the acetone was removed by rotary evaporation under vacuum at 40 °C to afford poly4VPGE macro-CTA as a light yellow solid (1.2 g). In cases where the poly4VPGE was intended for use in modification reactions with alcohols, the polymer was precipitated in diethyl ether to avoid any complications due to acetone impurities, as explained in the Results and Discussion section. SEC: Mn = 8600 g mol−1, Đ = 1.09. 1H NMR (400 MHz, CDCl3, δ [ppm]): 6.8−6.2 (4H), 4.3−4.0 (1H), 4.0−3.7 (1H), 3.4−3.3 (1H), 3.0−2.8 (1H), 2.8−2.7 (1H), 2.1−1.5 (1H), 1.5−0.9 (2H) (Figure 5a). Chain-Extension of Poly4VPGE Macro-CTA with Styrene. In the following procedure, DPn,targ = 300 with 30 mol % VAZO88 vs the poly4VPGE macro-CTA. Styrene (0.250 g, 2.40 mmol) was injected into a 5 mL glass vial containing poly4VPGE macro-CTA (0.089 g, corresponding to 0.008 mmol of CTA) and a stir bar. A 10× stock solution of radical initiator was prepared by dissolving VAZO88 (5.9 mg, 0.024 mmol) in anisole (2.0 mL), and a 1/10 portion of the stock solution (0.2 mL) was added to the reaction vial. The vial was capped, sealed, and sparged for 50 min with a steady flow of nitrogen. Samples were periodically withdrawn and analyzed in the manner previously described. Alcoholysis of PGE. To a 2.5 mL glass vial equipped with a screw-on cap with PTFE-faced silicone septum were added BuOH (74.1 mg, 1.00 mmol, 5 equiv vs epoxide), CH2Cl2 (0.9 mL), and PGE (27 μL, 0.20 mmol). A 100× stock solution of catalyst was prepared by dissolving BF3−Et2O (142 mg, 1.00 mmol) in CH2Cl2 (10.00 mL), and a 1/100 portion (0.1 mL, containing 1.42 mg (0.0100 mmol, 5 mol % vs epoxide) of BF3−Et2O) was added to the reaction mixture. The reaction was carried out at room temperature, and samples were periodically withdrawn over a 45 h timespan and diluted with CDCl3 for NMR analysis. Similar reactions were carried out using 1.00 mmol of cyclohexanol, t-BuOH, or benzyl alcohol instead of BuOH. NMR spectra of these reactions are included in the Supporting Information (Scheme S1, Figures S4−S7). Alcoholysis of Poly4VPGE. A solution of poly4VPGE (35.2 mg, 0.200 mmol) in 0.5 mL of CH2Cl2 was injected into a 2.5 mL glass vial, diluted with CH2Cl2 (0.4 mL), and then sealed with a screw-on cap with PTFE-faced silicone septum. BuOH (91.5 μL, 1.00 mmol, 5 equiv vs epoxide) was added to the reaction vial, which was then vigorously shaken for a few seconds to ensure even mixing of the
reagents. A 0.1 mL aliquot of the catalyst stock solution, containing 1.42 mg (0.0100 mmol, 5 mol % vs epoxide) of BF3-Et2O, was added. After 72 h, a small sample of the modified polymer was analyzed by SEC, and the remainder was purified by dialysis as previously described for NMR analysis. Similar reactions were carried out using 10, 20, 40, or 80 equiv of BuOH or 20 equiv of cyclohexanol, t-BuOH, benzyl alcohol, propargyl alcohol, or allyl alcohol. Similar reactions were also carried out using 5 equiv of either menthol or NBA. Menthol-modified polymer was isolated by precipitation in rapidly stirred hexane (100 mL), and NBA-modified polymer was likewise precipitated in diethyl ether.
■
RESULTS AND DISCUSSION Synthesis of 4VPGE. The synthesis of 4VPGE was first reported as a one-step procedure in which 4-acetoxystyrene was deacetylated by a 3-fold excess of aqueous NaOH and simultaneously reacted with epichlorohydrin to give the product in 11% yield after purification.23,24 The poor yield was likely due to epoxide ring-opening by excess hydroxide, acetate, or unreacted 4-vinylphenoxide anions. In the improved procedure reported here, side reactions were minimized by isolating the 4-vinylphenol intermediate produced from the deacetylation reaction, thereby removing the excess hydroxide anions and acetate byproducts, and then in the second step deprotonating the 4-vinylphenol with only a slight excess of NaOH before reacting it with 3 equiv of epichlorohydrin (Scheme 1). Vacuum distillation or flash chromatography Scheme 1. Synthesis of 4VPGE from 4-Acetoxystyrene
followed by crystallization in cold hexane of 4VPGE was necessary to remove all traces of impurities, which were found to have a detrimental effect on the polymerization control. 4VPGE was obtained in 54% yield. RAFT Polymerization of 4VPGE. RAFT polymerization27−29 has been proven to tolerate the presence of many different functional groups, including highly reactive, electrophilic moieties such as epoxides.14 The successful RAFT polymerization of 4VPO was recently reported,20 primarily using BPT as the CTA, VAZO88 as the radical initiator, and anisole as the solvent (50% (w/v)) at 90 °C. Given the recent success with 4VPO and its similarity in structure to 4VPGE, these same conditions were used for the polymerization of 4VPGE. When the RAFT polymerization of 4VPGE was first attempted (Scheme 2) with DPn,targ = [4VPGE]0/[BPT]0 = 200, [VAZO88]0 = 0.2 × [BPT]0, and [4VPGE]0 = 5.68 mol L−1 in anisole at 90 °C, the control exhibited by the system was satisfactory, but the MWDs were somewhat broader than expected for an ideal CRP with Đ = 1.2−1.3 up to 63% monomer conversion, which was reached in 20 h. These MWD dispersities are similar to those reported for the RAFT polymerization of GMA, which mostly ranged from 1.2 to 1.6 and only approached 1.1 when the DPn,targ were low (
