Imidazolium Bromide Derivatives of Brominated Poly(isobutylene-co

Oct 16, 2014 - Jackson M. Dakin†, Ralph A. Whitney‡, and J. Scott Parent†. †Department of Chemical Engineering and ‡Department of Chemistry,...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/IECR

Imidazolium Bromide Derivatives of Brominated Poly(isobutyleneco-para-methylstyrene): Synthesis of Peroxide-Curable Ionomeric Elastomers Jackson M. Dakin,† Ralph A. Whitney,‡ and J. Scott Parent*,† †

Department of Chemical Engineering and ‡Department of Chemistry, Queen’s University, Kingston, Ontario, Canada K7L 3N6 ABSTRACT: A series of new elastomeric ionomers is prepared through halide displacement from brominated poly(isobutyleneco-para-methylstyrene) (BIMS) by a range of N-substituted imidazoles. Studies of the dynamics of solvent-borne and solvent-free reactions demonstrate the influence of N-substituent structure on nucleophilicity, with alkyl and allyl groups providing greater reactivity than vinyl functionality. Ionomers bearing pendant styrenic, methacrylic, vinylic, and allylic functionality are crosslinked by standard peroxide initiators to give thermoset derivatives, with reaction rates and yields depending markedly on functional group structure. The discovery of a benzylic N-allylimidazolium ionomer is particularly important, as this material is amenable to solvent-free preparations, and provides a unique combination of moderate cross-linking rates and high cure extents, owing to a balance of radical addition and H-atom transfer reactivity. Physical properties of the resulting thermosets are a product of their hybrid ionic/covalent network, as the material response to deformation is affected by a stable covalent network as well as a labile network of aggregated ion pairs. leads to relatively slow rates of ionomer formation.6 The most economical and environmentally responsible method of preparing these ionomers is a solvent-free process where a halogenated elastomer is reacted with the required imidazole nucleophile in a conventional polymer compounding device.11 These processes seek full conversion of the electrophile to imidazolium bromide functionality in the shortest possible reaction time, using the least amount of nucleophile. The BIIRbased process illustrated in Scheme I is ill-suited to solvent-free

1. INTRODUCTION Isobutylene-rich elastomers are used in applications that require oxidative stability and gas impermeability.1 Commercial grades of these materials are typically sulfur-cured, but interest in peroxide-curable grades has increased, as their thermosets contain fewer reaction byproducts, and their carbon−carbon cross-links are more stable than sulfidic networks. The challenge is to overcome the susceptibility of isobutylene-rich polymers toward radical degradation, as poly(isobutylene) and poly(isobutylene-co-isoprene) containing small amounts of residual unsaturation degrade when treated with peroxide initiators at standard cure temperatures.2,3 One approach to this problem is to prepare macromonomer derivatives of these materials bearing multiple acrylate, styrenic, or maleimide groups per polymer chain, which can then be cross-linked by radical oligomerization of their pendant functionality, yielding thermosets whose cross-link density scales linearly with functional group content.4,5 We have recently described a new class of macromonomer that contains a small amount (0.10−0.25 mmol/g) of 1vinylimidazolium bromide functionality.6 These reactive ionomers cure to high cross-link density when activated by standard peroxide initiators to give thermosets whose polymer network is comprised of carbon−carbon cross-links as well as aggregates of imidazolium bromide ion pairs. This hybrid ionic/ covalent network provides unique physical properties, owing to differences in the stress relaxation properties of the two network components. Furthermore, the ionic functionality within these materials confers antimicrobial properties to their thermosets,7 enhances adhesion to high-energy surfaces, and improves their dispersion of siliceous fillers during elastomer compounding operations.8,9 While this clean-curing ionomer technology has considerable promise, two deficiencies compromise its commercial viability. In the first place, the low nucleophilicity of vinylimidazole10 © 2014 American Chemical Society

Scheme I. Synthesis and Peroxide Cross-Linking of an NVinylimidazolium Bromide Ionomer Derivative of BIIR

processes such as reactive extrusion, due to excessive residence time requirements, and to the reaction temperature restrictions imposed by the susceptibility of BIIR to dehydrobomination.12 In the second place, vinylimidazolium bromide ionomers crosslink so rapidly at standard cure temperatures that the material can be rendered thermoset before assuming the shape of the compression mold, a phenomenon known as scorch. Alternate Received: Revised: Accepted: Published: 17527

July 20, 2014 October 14, 2014 October 16, 2014 October 16, 2014 dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

tetrabutylammonium hydroxide in methanol (Bu4NOH), tetrabutylammonium bromide (TBAB, 98%), 11-bromo-1undecene (95%), 1-bromododecane (97%), allyl bromide (97%), 2-ethylimidazole (98%), 1-bromobutane (99%), 1,2dichloroethane (anhydrous, 99.8%), hydroquinone (99.5%), benzyl bromide (98%), and vinylbenzyl chloride (VBCl, 97%) were used as received from Sigma-Aldrich. 2,6-Di-tert-butyl-4methylphenol (BHT, 99%) and 1-allylimidazole (AlIm, 99%) were used as received from Alfa Aesar. Brominated poly(isobutylene-co-para-methylstyrene) (BIMS) (Exxpro 3745, 0.23 mmol benzylic bromide functionality/g BIMS) was used as supplied by Exxon Mobil Chemical. 2.2. Solution Synthesis of IMS-VImBr. BIMS (10 g, 2.3 mmol benzylic bromide) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. VIm 2 (0.433 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 30 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1 H NMR. The N-alkylation product, IMS-VImBr 12, was obtained by precipitation in excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 10.41 (s, 1H, −N+−CH−N−), δ 5.73 (dd, 1H, −N−CHCH−Htrans), δ 5.51 (s, 2H, Ph−CH2−N+−), δ 5.41 (dd, 1H, −N−CHCH− Hcis), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.3. Solid-State Synthesis of IMS-VImBr. BIMS (40 g, 9.2 mmol benzylic bromide) was mixed with VIm 2 (1.732 g, 18.4 mmol, 2 equiv) and BHT (0.008 g, 200 ppm) at 100 °C and 30 rpm using a Haake Polylab R600 internal batch mixer. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 10.41 (s, 1H, −N+−CH−N−), δ 5.73 (dd, 1H, −N−CHCH−Htrans), δ 5.51 (s, 2H, Ph−CH2−N+−), δ 5.41 (dd, 1H, −N−CHCH−Hcis), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.4. Synthesis of IMS-BuImBr. BIMS (10 g, 2.3 mmol benzylic bromide) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. BuIm 1 (0.571 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 7 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR. The N-alkylation product, IMS-BuImBr 11, was obtained by precipitation in excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 9.99 (s,1H, −N+−CH−N−), δ 5.43 (s, 2H, Ph−CH2−N+−), δ 4.27 (t, 2H, −N−CH2−CH2−CH2−CH3), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.5. Solution Synthesis of IMS-AlImBr. BIMS (10 g, 2.3 mmol benzylic bromide) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. AlIm 5 (0.497 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 7 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and analyzed by 1H NMR. The N-alkylation product, IMS-AlImBr 15, was obtained by precipitation from excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 9.86 (s, 1H, −N+−CH−N−), δ 6.00 (ddt, 1H, −N−CH2−CHCH2), δ 5.43 (s, 2H, Ph−CH2−N+−), δ 5.43 (dd, 2H, −N−CH2− CHCH2), δ 4.93 (d, 2H, −N−CH2−CHCH2), found for

oligomerizable functionality is required that can provide moderate cure rates as well as high cure extents. The objective of the present work was to develop new technology that allows thermoset ionomers to be prepared under commercially acceptable reaction conditions, yielding a product that cures with appropriate rate and extent. To overcome reactivity limits imposed by BIIR, we have used brominated poly(isobutylene-co-para-methylstyrene) (BIMS), whose benzylic bromide functionality is more electrophilic and less susceptible to dehydrohalogenation.13 We have also prepared and studied alternate imidazole nucleophiles (Scheme II), whose oligomerizable group is remote from the Scheme II. Imidazolium Bromide Ionomers Examined in This Study

imidazolium moiety, thereby avoiding electronic effects of an N-vinyl substituent (nucleophiles 3−7). In a further attempt to improve reaction rates, we have prepared imidazoles bearing 2ethyl substituents (nucleophiles 8−10), which according to a report of Salamone et al.,14 may enhance nucleophilicity by electron donation to the heteroaromatic ring. Our report begins with a description of the alkylation rates for N-substituted imidazoles by BIMS under solvent-borne and solvent-free conditions, before progressing to studies of peroxide-initiated cross-linking dynamics and yields. Particular emphasis is placed on allylimidazolium ionomer cross-linking chemistry, for which model compound experiments are used to elucidate potential cure pathways. We conclude with a brief examination of the physical properties of thermoset ionomers, in which new materials are compared with established benchmarks.

2. EXPERIMENTAL SECTION 2.1. Materials. 1-Butylimidazole (BuIm, 98%), 1-vinylimidazole (VIm, 99+%), dicumyl peroxide (DCP, 98%) methacryloyl chloride (97.0%), 1-(2-hydroxyethyl)imidazole (97%), oxalic acid (98%), 10-undecenoic acid (98%), 1 M 17528

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.6. Solvent-Free Synthesis of IMS-AlImBr. BIMS (40 g, 9.2 mmol benzylic bromide) was mixed with AlIm 5 (1.998 g, 18.4 mmol, 2 equiv) at 100 °C and 30 rpm using a Haake Polylab R600 internal batch mixer equipped with Banbury rubber-mixing blades. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR as described above. 2.7. Synthesis of IMS-ImEMABr. 2-(1H-Imidazol-1-yl)ethyl methacrylate 3 was synthesized as per literature precedent.15 1-(2-Hydroxyethyl)imidazole (0.121 g, 1.076 mmol) was dissolved in DCM (5 mL) in a dry flask under N2. A solution of methacryloyl chloride (0.169 g, 1.614 mmol) in DCM (5 mL) was added dropwise at 25 °C, and the mixture was stirred under N2 overnight at 25 °C before removing DCM by rotary evaporation. The resulting imidazolium chloride salt was dissolved in water and acidified with oxalic acid. The aqueous phase was then washed with ethyl acetate and basified with sodium carbonate. The pale yellow oil, ImEMA 3, was extracted with excess toluene and recovered in 42% yield after removing the solvent by vacuum. 1H NMR (CDCl3): δ 7.50 (s, 1H, −N−CH−N−), δ 7.07 (s, 1H, −N−CHCH−N−), δ 6.95 (s, 1H, −N−CHCH−N−), δ 6.08 (s, 1H, −OOC− C(CH3)CH−Htrans), δ 5.60 (s, 1H, −OOC−C(CH3) CH−Hcis), δ 4.38 (t, 2H, −N−CH2−CH2−COO−), δ 4.23 (t, 2H, −N−CH2−CH2−COO−), δ 1.92 (s, 3H, −OOC− C(CH3)CH2). BIMS (10 g, 2.3 mmol benzylic bromide) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. ImEMA 3 (0.829 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 7 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR. The N-alkylation product, IMS-ImEMABr 13, was obtained by precipitation in excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 9.94 (s, 1H, −N+−CH−N−), δ 6.08 (s, 1H, −OOC−C(CH3)CH−Htrans), δ 5.60 (s, 1H, −OOC− C(CH3)CH−Hcis), δ 5.39 (s, 2H, Ph−CH2−N+−), δ 4.70 (t, 2H, −N−CH2−CH2−COO−), δ 4.54 (t, 2H, −N−CH2− CH2−COO−), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.8. Synthesis of IMS-VBImBr. 1-(4-Vinylbenzyl)-1Himidazole (VBIm 4) was synthesized as per literature precedent.16 NaHCO3 (5.25 g, 62.4 mmol) and imidazole (13.61 g, 0.199 mol) were added to 100 mL of a binary mixture of water/acetone (1:1 v:v) and stirred until completely dissolved. To this mixture was added 4-vinylbenzyl chloride (7.61 g, 49.8 mmol) dropwise at room temperature after which the reaction mixture was heated to 50 °C and stirred for 22 h. Following the reaction, solids were filtered and discarded, and the acetone was distilled under reduced pressure. The remaining filtrate was diluted with 500 mL of diethyl ether and washed with 50 mL of deionized water six times. The organic phase was then washed with 100 mL of 2.0 M HCl three times, saving the aqueous washes. Then, 200 mL of 4.0 M NaOH was added to the mixture and extracted with 50 mL of diethyl ether three times. The organic phase was dried over anhydrous magnesium sulfate, and the ether was removed under reduced pressure to yield a pale orange oil in 58% yield.1H NMR (CDCl3): δ 6.85−7.71 (mult, 7H, aromatic), δ 6.65 (dd, 1H, Ph−CH−CH2), δ 5.71 (dd, 1H, Ph−CH−CH−

Htrans), δ 5.24 (d, 1H, Ph−CH−CH−Hcis), δ 5.04 (s, 2H, N− CH2−Ph). BIMS (10 g) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. VBIm 4 (0.847 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 7 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR. The N-alkylation product, IMS-VBImBr 14, was obtained by precipitation in excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 10.15 (s, 1H, −N+−CH−N−), δ 5.80 (dd, 1H, Ph−CH−CH−Htrans), δ 5.50 (s, 2H, Ph−CH2−N+−), δ 5.42 (s, 2H, N+−CH2−Ph), δ 5.30 (d, 1H, Ph−CH−CH−Hcis), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.9. Synthesis of IMS-UDImBr. 1-(10-Undecen-1-yl)-1Himidazole 7 (UDIm) was prepared as per literature precedent.17 NaOH (12.50 g, 312.5 mmol) was dissolved in water (12.50 g) and mixed with THF (60 mL). Imidazole (1.31 g, 19.8 mmol) and TBAB (0.606 g, 1.98 mmol) were added to the solution at 25 °C until all the solids were dissolved, prior to adding 11bromo-1-undecene (4.35 g, 18.64 mmol) and heating to 60 °C for 16 h. Upon cooling, the THF was removed by rotary evaporation, and the product was extracted three times with DCM. The organic layer was dried with magnesium sulfate, and the filtrate was concentrated by rotary evaporation. The resulting pale yellow oil, 1-(10-undecen-1-yl)-1H-imidazole, was obtained in 98% yield. 1H NMR (CDCl3): δ 7.40 (s, 1H, −N−CH−N−), δ 7.00 (s, 1H, −N−CHCH−N−), δ 6.85 (s, 1H, −N−CHCH−N−), δ 5.75 (ddt, 1H, −CH2−CH CH2), δ 4.90 (dd, 2H, −CH2−CHCH2), δ 3.86 (t, 2H), δ 1.98 (dd, 2H), δ 1.71 (t, 2H), δ 1.22−1.32 (m, 12H). BIMS (40 g, 9.2 mmol benzylic bromide) was mixed with UDIm 7 (2.230 g, 1.1 equiv), at 100 °C and 30 rpm using a Haake Polylab R600 internal batch mixer for 1 h. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 10.22 (s, 1H, −N+−CH−N−), δ 5.80 (ddt, 1H, −CH2−CHCH2), δ 5.48 (s, 2H, Ph−CH2− N+−), δ 4.95 (dd, 2H, −CH2−CHCH2), δ 4.30 (dd, 2H, −N−CH2−CH2−CH2−). 2.10. Synthesis of IMS-DDImBr. NaOH (12.50 g, 312.5 mmol) was dissolved in water (12.50 g) and mixed with THF (60 mL). Imidazole (1.31 g, 19.8 mmol) and TBAB (0.606 g, 1.98 mmol) were added at 25 °C until all the solids were dissolved, prior to adding 1-bromododecane (4.65 g, 18.64 mmol) and heating to 60 °C for 16 h. Upon cooling, THF was removed by rotary evaporation, and the product was extracted three times with DCM. The organic layer was dried with magnesium sulfate, and the filtrate was concentrated by rotary evaporation. The resulting oil, 2-dodecyl-1H-imidazole 6, was obtained in quantitative yield. 1H NMR (CDCl3): δ 7.44 (s, 1H, −N−CH−N−), δ 7.03 (s, 1H, −N−CHCH−N−), δ 6.88 (s, 1H, −N−CHCH−N−), δ 3.90 (t, 2H, N−CH2− CH2−), δ 1.75 (t, 2H, N−CH2−CH2−), δ 1.24 (m, 18H), δ 0.86 (t, 3H, −CH2−CH2−CH3).18 BIMS (40 g, 9.2 mmol benzylic bromide) was mixed with DDIm (3.720 g, 1.5 equiv), at 100 °C and 30 rpm using a Haake Polylab R600 internal batch mixer for 1 h. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 10.93 (s, 1H, −N+−CH−N−), δ 5.46 (s, 2H, Ph−CH2−N+−), δ 4.25 (t, 2H, −N−CH2−CH2− CH2−). 2.11. Synthesis of IMS-EtBuImBr. 1-Bromobutane (1.350 g, 10.0 mmol), 2-ethylimidazole (0.96 g, 11.0 mmol), and NaOH (0.48 g, 12.0 mmol) were dissolved in acetonitrile (2 17529

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

mL) and heated to reflux overnight, cooled, filtered, and concentrated by rotary evaporation. The clear liquid product, EtBuIm 8, was recovered in 68% yield by vacuum distillation. 1 H NMR: (CDCl3): δ 6.76 (d, 1H, −N−CHCH−N−), δ 6.65 (d, 1H, −N+−CHCH−N−), δ 3.66 (t, 2H, −N−CH2− CH2−), δ 2.51 (q, 2H, N+−C−CH2−CH3), δ 1.55 (quint, 2H, −N−CH2−CH2−), δ 1.20 (quint, 2H, −N−CH2−CH2− CH2−), δ 1.18 (t, 3H, N+−CN−CH2−CH3), δ 0.79 (t, 3H, −CH2−CH2−CH2−CH3).19 BIMS (10 g, 2.3 mmol benzylic bromide) was dissolved in chlorobenzene (100 mL) at 25 °C, and the resulting cement was heated to 100 °C. EtBuIm 8 (0.700 g, 4.6 mmol, 2 equiv) was added to the heated solution and stirred for 7 h at 100 °C. Aliquots of the mixture (∼0.3 g) were removed at regular intervals and characterized by 1H NMR. The N-alkylation product, IMS-EtBuImBr 18, was obtained by precipitation in excess acetone and purified by dissolution/precipitation using THF/acetone and dried in vacuo. 1H NMR (CDCl3 + 5 wt % CD3OD): δ 5.49 (s, 2H, Ph−CH2−N+−), δ 4.23 (t, 2H, −N− CH2−CH2−CH2−CH3), δ 3.24 (t, 2H, −C−CH2−CH3), found for residual brominated para-methylstyrene mer: δ 4.49 (s, 2H), δ 4.45 (s, 2H). 2.12. Synthesis of 1-Allyl-2-ethyl-1H-imidazole. In 2 mL of acetonitrile was dissolved allyl bromide (0.665 g, 5.5 mmol) and 2-ethylimidazole (0.480 g, 5.0 mmol). NaOH (0.24 g, 6.0 mmol) was added, and the mixture was stirred at 25 °C for 5 h. The reaction mixture was then filtered, and the filtrate evaporated by rotary evaporation. The clear liquid product, EtAlIm 10, was recovered in 73% yield by Kugelrohr distillation. 1H NMR (CDCl3): δ 6.93 (d, 1H, −N−CH CH−N−), δ 6.78 (d, 1H, −N−CHCH−N−), δ 5.88 (ddt, 1H, −N−CH2−CHCH2), δ 5.19 (dd, 1H, −N−CH2−CH CH−Htrans), δ 4.98 (dd, 1H, −N−CH2−CHCH−Hcis), δ 4.43 (dt, 2H, −N−CH2−CHCH2), δ 2.62 (q, 2H, −C− CH2−CH3), δ 1.30 (t, 3H, −C−CH2−CH3). 13C NMR (CDCl3): δ 149.26, 132.95, 127.12, 119.25, 117.40, 47.98, 19.98, 12.03. HRMS (EI): calc for C8H12N2 136.1000, found 136.1005. 2.13. Synthesis of 1-Vinyl-2-ethyl-1H-imidazole. EtVIm 9 was prepared as per literature precedent.20 2-Ethylimidazole (1.442 g, 15.0 mmol) was added to a stirring mixture of 1,2dichloroethane (40 mL), TBAB (0.104 g, 0.32 mmol), KOH (5.6 g, 100 mmol), and K2CO3 (4.42 g, 32 mmol). The mixture was stirred at 50 °C for 5 h, cooled, and filtered, and the organic solution washed with water (2 × 10 mL). The organic phase was then dried with magnesium sulfate, and the solvent was removed by rotary evaporation. The clear liquid product, 2ethyl-1-chloroethyl-1H-imidazole, was purified and isolated in 41% yield by vacuum distillation. 1H NMR (CDCl3): δ 6.90 (d, 1H, −N−CHCH−N−), δ 6.82 (d, 1H, −N−CHCH− N−), δ 4.12 (d, 2H, −N−CH2−CH2−Cl), δ 3.65 (d, 2H, −N− CH2−CH2−Cl), δ 2.64 (q, 2H, −C−CH2−CH3), δ 1.29 (t, 3H, −C−CH2−CH3). A mixture of 2-ethyl-1-chloroethyl-1H-imidazole (0.98 g, 6.18 mmol), potassium hydroxide (1.38 g, 24.6 mmol), and hydroquinone (30 mg) was added to isopropyl alcohol (40 mL) and refluxed for 2 h. Upon cooling and filtering, the alcohol was removed by rotary evaporation, water (20 mL) was added to the reaction mixture, and the organic material was extracted with DCM (3 × 20 mL). The organic phase was dried with magnesium sulfate and filtered, and the solvent was removed by rotary evaporation. The dehydrohalogenation product, EtVIm 9, was purified and isolated in 53% yield by

vacuum distillation. 1H NMR (CDCl3): δ 7.02 (d, 1H, −N+− CHCH−N−), δ 6.84 (d, 1H, −N+−CHCH−N−), δ 6.77 (dd, 1H, −N−CHCH2), δ 5.01 (dd, 1H, −N−CHCH− Htrans), δ 4.75 (dd, 1H, −N−CHCH−Hcis), δ 2.63 (q, 2H, −C−CH2−CH3), δ 1.21 (t, 3H, −C−CH2−CH3). 2.14. General Procedure for the Synthesis of Imidazolium Bromide Ionic Liquids. The appropriate 3alkyl-1H-imidazole (8 mmol) and benzyl bromide (1.368 g, 8 mmol) were dissolved in THF (12 mL) and stirred at 25 °C overnight. A viscous ionic liquid settled to the bottom of the stirred flask, and the organic phase was decanted. The ionic liquid was washed and decanted three times with additional THF, and any residual solvent was removed by rotary evaporation. 3-Benzyl-1-allyl-1H-imidazol-3-ium Bromide. DSC: Tg = −48 °C. High resolution MS analysis; required for C13H15N2+ m/e 199.1227; found m/e 199.1235. IR (film) ν = 3427(s), 3130(s), 3068(s), 1625(m), 1560(s), 1497(m), 1454(s), 1341(m), 1155(s), 994(m), 947(m), 757(s), 713(s) cm−1. 1H NMR (CDCl3): δ 10.43 (s, 1H, N+−CH−N), δ 7.29−7.46 (m, 7H, aromatics), δ 5.95 (ddt, 1H, −N−CH2−CHCH2), δ 5.56 (s, 2H, Ph−CH2−N+−), δ 5.38 (dd, 2H, −N−CH2− CHCH2), δ 4.93 (d, 2H, −N−CH2−CHCH2). 13C NMR (CDCl3): δ 136.5 (−N+−CH−N−), δ 129.8 (−N−CH2− CHCH2), δ 129−133 (phenylic), δ 122.7 (−N−CH2− CHCH2), δ 122.3 (−N+−CHCH−N−), δ 122.2 (−N+− CHCH−N−), δ 53.2 (Ph−CH2−N+−), δ 52.1 (−N−CH2− CHCH2). 3-Benzyl-1-vinyl-1H-imidazol-3-ium Bromide. DSC: Tg = N/A. IR (film) ν = 3429(s), 3127(s), 3082(s), 1651(s) 1550(s), 1497(w), 1455(m), 1369(m), 1163(s), 958(m), 922(m), 719(s) cm−1. 1H NMR (CDCl3): δ 10.68 (s, 1H, −N+−CH−N−), δ 7.17−7.91 (m, 7H, aromatics), δ 7.22 (dd, 1H, −N−CHCH2), δ 5.89 (dd, 1H, −N−CHCH−Htrans), δ 5.52 (s, 2H, Ph−CH2−N+−), δ 5.16 (dd, 1H, −N−CH CH−Hcis). 3-Benzyl-2-ethyl-1-allyl-1H-imidazol-3-ium Bromide. 1H NMR (CDCl3): δ 7.66 (d, 1H, −N+−CHCH−N−), δ 7.59 (d, 1H, −N+−CHCH−N−), δ 7.34 (m, 5H, phenylic), δ 5.98 (ddt, 1H, −N−CH2−CHCH2), δ 5.57 (s, 2H, Ph− CH2−N+−), δ 5.38 (dd, 1H, −N−CH2−CHCH−Htrans), δ 5.29 (dd, 1H, −N−CH2−CHCH−Hcis), δ 4.93 (dt, 2H, −N−CH2−CHCH2), δ 3.19 (q, 2H, −C−CH2−CH3), δ 1.01 (t, 3H, −C−CH2−CH3). 2.15. General Procedure for the Peroxide Initiated Polymerization of N-Functional Imidazolium Bromide Ionic Liquids. A mixture of imidazolium bromide ionic liquid (2.4 mmol) and dicumyl peroxide (52.1 mg, 0.19 mmol) was stirred in a 1 mL Wheaton vial and heated in a heating block at 160 °C for 1 h. The resulting mixture was taken up in CDCl3, and to it was added TBAB (0.193 g, 0.60 mmol) as an external standard for 1H NMR analysis. To isolate the microparticles, the mixture was taken up in water (5 mL), centrifuged, and decanted. The washing process was repeated 3 times, and the purity of the insoluble particles was verified by 1H NMR. PhAlImBr-XL. DSC: Tg = N/A. IR (film) ν = 3395(m), 3962(m), 2929(m), 1627(m), 1496(w), 1454(m), 1357(w), 1261(m), 1157(s), 1096(m), 1027(m), 800(m), 703(m) cm−1. PhVImBr-XL. DSC: Tg = N/A. IR (ATR-ZnSe) ν = 3402(s), 3124(m), 3060(s), 1632(m), 1549(s), 1497(w), 1454(m), 1209(w), 1151(s), 1109(w), 711(s). 2.16. Preparation of Cured Macrosheets. A 35 g batch of purified ionomer was coated with the required amount of a 17530

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

Substantial differences in the observed rates of BuIm and VIm N-alkylation demonstrate the extent to which substituents can affect the nucleophilicity of a functional imidazole. Salamone et al. reported improved nucleophilicity for imidazoles bearing alkyl substituents at the 2-position,15 leading us to prepare and study 2-ethyl-1-butyl imidazole (EtBuIm 8). The conversion data showed no improvement when compared to BuIm, suggesting that this strategy may not apply to the system of interest. Furthermore, it was found that introducing an ethyl substituent to the otherwise robust VIm system increased the monomer’s susceptibility to autoinitiated crosslinking. Even in the presence of antioxidants, 2-ethyl-1vinylimidazole (EtVIm 9) produced insoluble gel early in the BIMS alkylation, and, as a result, samples of high-conversion materials could not be analyzed by 1H NMR. We therefore examined model reactions, wherein benzyl bromide was reacted with EtVIm and with VIm in acetone. These experiments revealed no significant differences in N-alkylation rate, indicating that a 2-ethyl substituent does not enhance vinylimidazole reactivity toward our electrophile. Unfortunately, it only compromises the ionomer’s storage stability. The effect of distancing the polymerizable functional group from the heteroaromatic ring was studied by preparing 2-(1Himidazol-1-yl)ethyl methacrylate (ImEMA 3) and studying its N-alkylation dynamics. The data plotted in Figure 1 show significant gains in nucleophilicity but not sufficient to match the reactivity of BuIm. Similarly, 1-(4-vinylbenzyl)-1Himidazole (VBIm 4) was more reactive toward BIMS than VIm. However, this improvement was gained at the expense of product stability, as the resulting ionomer was sparingly soluble in organic solvents. Allyl monomers are not usually amenable to conventional radical polymerization, since allylic hydrogen atom donation promotes chain transfer to monomer.24 Nevertheless, the conditions used in elastomer cross-linking are sufficiently different from those of standard polymerizations to warrant a study of 1-allylimidazole (AlIm). The data plotted in Figure 1 show that the rate of AlIm alkylation by BIMS was indistinguishable from that of BuIm, and the resulting allyl imidazolium bromide ionomer was stable for prolonged periods at room temperature. Interestingly, the 2-ethyl substituted analogue (10) yielded an insoluble gel, similar to the vinyl imidazolium system (9) described above. 3.2. Peroxide-Initiated Cross-Linking Dynamics. The standard method for studying polymer cross-linking dynamics is to the monitor the dynamic storage modulus (G′) of the compound at a fixed temperature, oscillation frequency, and strain amplitude.25 Figure 2 provides rheological data for the macromonomers of interest as well as a control material, IMSBuImBr (11). This ionomer contains no CC functionality, and it degrades when mixed with 18 μmol of dicumyl peroxide (DCP) per gram of elastomer and heated to 160 °C. Note that the benzylic site adjacent to the imidazolium ion pair has a relatively low bond dissociation energy and could support cross-linking by H-atom abstraction followed by benzylic radical combination. However, the observed loss of G′ shows that this mechanism is not competitive with polymer chain scission under these reaction conditions. We have shown previously that vinylimidazolium ionomers derived from butyl rubber (Scheme I) cross-link rapidly and to high extent,6 reaching their maximum storage modulus within one-half-life of the peroxide initiator (5.4 min at 160 °C).26 The BIMS analogue, IMS-VImBr (12), behaves similarly, reaching

solution of DCP in acetone and allowed to dry before being passed through a two roll mill ten times. The compounded sample was then sheeted in the mill and compression molded at 160 °C and 20 MPa for 60 min.21 The sheeted product had a thickness of 2.00 ± 0.2 mm. 2.17. Analysis. NMR spectra were acquired with a Bruker AM400 instrument with chemical shifts (δ) reported in ppm relative to tetramethylsilane. Rheological characterization was performed with an Advanced Polymer Analyzer 2000 from Alpha Technologies operating in a biconical disk configuration. A 5.0 g sample of elastomer was coated with 0.5 wt % DCP dissolved in acetone and allowed to dry before being passed through a two-roll mill ten times. The resulting compound was cured in the rheometer cavity at 160 °C for 60 min, at a 3° oscillation arc and a frequency of 1 Hz. Stress relaxation measurements were conducted at 100 °C with a strain of 2° arc for 5 min. Temperature sweeps were conducted from 100 to 200 °C at a frequency of 1 Hz and a 3° oscillation arc. Tensile data were acquired using an INSTRON Series 3360 universal testing instrument, operating at a crosshead speed of 500 mm/min at 23 ± 1 °C.22 The 5% modulus was found by calculating the slope of the stress (MPa) vs strain (mm/mm) curve from 0 mm to 0.2 mm extension. Specimens were cut from cured macrosheets as described in ASTM D4482,23 with values reported as the average of five replicates.

3. RESULTS AND DISCUSSION 3.1. Dynamics of Solvent-Borne N-Alkylations. Studies of solvent-free reaction dynamics are complicated by the speed of the reaction as well as the difficulty in controlling temperature in conventional polymer processing devices. More precise data are obtained from slower, solvent-borne reactions such as those illustrated in Figure 1. These results

Figure 1. N-alkylation dynamics of imidazole nucleophiles by BIMS (10 wt % polymer in monochlorobenzene, 100 °C, 2 equiv of nucleophile).

were acquired using 10 wt % solutions of BIMS in chlorobenzene that contained 2 molar equiv of nucleophile relative to the benzylic bromide functionality within the elastomer. 1H NMR analysis of samples withdrawn at intervals showed no evidence of side-reactions that might compromise reaction yields, as all N-alkylations went to full conversion if given sufficient time. 17531

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

peroxide concentration. Macromonomers whose functionality oligomerizes with high kinetic chain length require relatively little initiator to reach full CC conversion. This is the case for IMS-VImBr, which needed only 3.6 μmole DCP/g to reach a storage modulus plateau of 660 kPa (Figure 3a). In contrast, IMS-AlImBr is much less reactive, as it required several times more DCP to achieve a comparable state of cure (Figure 3b). It is important to note, however, that the extent of the IMSAlImBr cure approached a definite upper limit. Peroxide cures of saturated polymers such as polyethylene are purely stoichiometric processes involving H-atom abstraction from the polymer followed by combination of the resulting macroradicals.29 In these simple cure systems, cross-link density increases proportionally with initiator loading. This is clearly not the case for IMS-AlImBr, whose response to DCP concentration is consistent with a macromonomer whose limited reactivity requires higher initiator loadings to convert its pendant CC functionality. What distinguishes IMS-AlImBr from other macromonomers that suffer from low polymerization rates is its ability to reach high cross-linking extents. Consider the rheology data plotted in Figure 4, in which IMS-AlImBr (15) is compared with two analogues. IMS-UDImBr (17) is an allyl-functionalized imidazolium ionomer whose reactive functional group is remote from the imidazolium group, and IMS-DDImBr (16) is a completely saturated ionomer. The data confirm that the allyl group within IMS-AlImBr is activated by the heteroaromatic ring. Whether this activation involves improving the rate of CC group oligomerization and/or the rate of allylic hydrogen atom donation cannot be discerned by rheological data. At present, literature regarding allylimidazolium monomers is limited to a study of the copolymerization of 1-allyl-3methyl imidazolium chloride with acrylonitrile at mild reaction temperatures, wherein reactivity ratios confirmed the relatively low polymerization activity of this monomer.30 Given a lack of information regarding the radical chemistry of allylimidazolium bromides, we studied PhAlImBr (21) as a model for IMS-AlImBr to determine the main reactions that underlie ionomer cross-linking. Since imidazolium bromide salts are insoluble in nonpolar solvents (leading to ion-pair aggregation in the ionomer), neat PhAlImBr was heated with DCP to 160 °C for 1 h. The 18 mmol DCP/g polymer used in

Figure 2. Peroxide cross-linking dynamics of various imidazolium ionomers of BIMS ([DCP] = 18 μmol/g, 160 °C, 1 Hz, 3° arc).

G′ = 660 kPa at the 3 min mark, before reverting slightly as vinylimidazolium conversion to cross-links is overcome by peroxide-initiated degradation of the polymer backbone.5 The acrylate functionalized ionomer 13, and its styrenic analogue 14, also cured rapidly to high G′ plateaus. It should be noted, however, that highly reactive CC functionality can lead to scorch issues as well as storage stability problems. This was observed for IMS-VBImBr (14), which cross-linked slightly during its preparation and while stored at room temperature. Surprisingly, the allylic ionomer 15 produced a final storage modulus on par with that observed for its more reactive counterparts but at a significantly slower rate. In general, allylic monomers are not considered to be highly polymerizable.27 In the context of an isobutylene-rich macromonomer cure, low functional group reactivity usually correlates with low cure extent, since polymer backbone degradation competes with cross-linking through oligomerization of pendant CC functionality.28 The observed combination of moderate cure rate and high cure yield makes IMS-AlImBr of considerable fundamental and technological interest. Insight into the differences between IMS-VImBr (12) and IMS-AlImBr (15) was gained by studying their sensitivity to

Figure 3. Peroxide cure dynamics of (a) IMS-VImBr and (b) IMS-AlImBr with increasing peroxide loading. (160 °C, 1 Hz, 3° arc). 17532

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

Figure 5. SEM image of solids derived from PhAlImBr + DCP.

could not identify a glass transition, unlike the monomer that exhibited a distinct Tg at −48 °C. FT-IR analysis of the solids confirmed the presence of imidazolium bromide functionality and the conversion of CC groups. Similar results were observed for 2-ethyl-1-allyl imidazolium bromide (PhEtAlImBr, 22), with 67% of the 2-alkyl functionalized monomer converted to solids, as opposed to just 31% for compound 21. This higher radical reactivity may explain the cross-linking we observed when trying to prepare its BIMS-derived ionomer. The chemistry of PhVImBr (23) differs significantly from its allylic analogues, in that the vinyl monomer reached 100% conversion to give a leathery polymer that could not be dissolved in water or organic solvents. Note that radical polymerization of vinylimidazolium halides to give linear polyionic liquids is well-known,32−34 with typical conditions involving dilute solutions heated to moderate temperature with 1−2 mol % initiator.15,35 Clearly, the more severe conditions used in this work led to cross-linking of the polyionic liquid, likely through a conventional mechanism involving hydrogen atom abstraction from the polymer, followed by combination of the resulting macroradicals. Taken together, our cure rheometry and model compound studies can be summarized as follows: a. IMS-AlImBr provides an unusual combination of moderate cure rate and high ultimate cross-link density; b. the allyl group is activated by adjacent imidazolium bromide functionality; c. the kinetic reactivity of the allyl group is less than that of vinyl functionality, but its propensity for branching leads to crosslinked architectures. These observations are explained by the susceptibility of allylic monomers to allylic H-atom abstraction. Scheme IV illustrates two pathways for DCP-initiated cross-linking of IMSAlImBr. Oligomerization by repeated radical addition to pendant CC bonds is the conventional process through which macromonomers cure, and the potential exists for the adjacent imidazolium group to improve the reactivity of allyl functionality in this respect. Allylic H-atom donation will also be favorable, yielding a resonance-stabilized allyl macroradical that prefers termination by combination to further propagation reactions. In conventional polymerization systems, this H-atom abstraction is referred to as degradative chain transfer,36in that the polymerization rate and degree of polymerization are reduced. In the present context, allylic H-atom transfer may also lower the kinetic chain length of CC oligomerization, thereby reducing cure rates. However, the termination products derived from allyl macroradicals will contain residual unsaturation that can engage in further cross-linking. Moreover, H-atom donation can shift the balance between polymer crosslinking and chain scission in a favorable direction.

Figure 4. Peroxide cure dynamics of allylic, terminally unsaturated, and saturated imidazolium bromide ionomers ([DCP] = 18 μmol/g, 160 °C, 1 Hz, 3° arc).

the ionomer cure studies corresponds to 13 mol % DCP relative to the imidazolium bromide content of IMS-AlImBr, leading us to use this peroxide concentration throughout our model compound studies (Scheme III). 1H NMR analysis of Scheme III. Imidazolium Model Compound Reactions

the reaction mixture in acetone-d6 and CDCl3 found 69% of the starting material (21) charged to the reaction as well as acetophenone, the latter resulting from β-scission of cumyloxyl radicals derived from DCP thermolysis.31 Taking up the PhAlImBr (21) + DCP reaction product in water revealed a fine suspension, from which solids were isolated by sonicating, centrifuging, and decanting the supernatant. Scanning electron microscopy of the material revealed submicron spheres (Figure 5). The insolubility of these particles is consistent with a highly cross-linked architecture, as opposed to a linear or slightly branched polyionic liquid, which would be water-soluble. DSC analysis of the particles 17533

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

Scheme IV. Radical Reactions of IMS-AlImBr Functionality

Figure 6. Solvent-free N-alkylation dynamics of VIm and AlIm by BIMS (100 °C, 30 rpm, 2 equiv).

20 min for AlIm, compared to 60 min for VIm. The alkylation of VBIm under solvent-free conditions (not shown) proceeded even more rapidly than AlIm, reaching full conversion after just 5 min, but the product had high gel content and poor processing characteristics. Considering that IMS-AlImBr is more amenable to solvent-free preparation and it cures at a more moderate rate, it may be the preferred reactive ionomer for commercial purposes. Note that residual monomer had a detrimental effect on the peroxide cure of IMS-AlImBr, presumably due to excessive allylic hydrogen atom donation. As such, devolatilization of solvent-free reaction products to remove residual monomer is recommended. 3.4. Physical Properties of Ionomer Thermosets. Previous studies of BIIR-derived imidazolium thermosets demonstrated unique material properties that are provided by a hybrid ionic/covalent network. 8 The ionic network established by ion-pair aggregation is labile, and uncured ionomers such as IMS-BuImBr are susceptible to stress relaxation when subjected to a constant applied strain. Figure 7 shows that the static modulus of BIMS and IMS-BuImBr declined rapidly and extensively when the material was strained at 100 °C for 5 min. By comparison, stress relaxation incurred

The influence of H-atom donor functionality on the peroxide modification of isobutylene-rich copolymers was first reported by Loan, who noted that 3 mol % isoprene gave an elastomer that did not degrade when heated with DCP at 153 °C.3 He proposed that polyisobutylene degrades by H-atom abstraction from the methyl groups of isobutylene mers, which has recently been confirmed by independent study,37 followed by β-scission of the resulting primary macroradical. Therefore, a H-atom donor can reduce polymer backbone degradation by lowering the population of primary macroradicals in favor of allylic radical intermediates that do not cleave. In the case of butyl rubber, the allylic H-atom donor is provided by residual unsaturation within isoprene mers. In the case of IMS-AlImBr, there are benzylic donors derived from unbrominated paramethylstyrene mers, benzylic donors that are adjacent to imidazolium functionality, and allylic donors adjacent to imidazolium groups. All three are expected to mitigate chain scission, thereby boosting cross-link densities. Note, however, that IMS-UDImBr did not cure as effectively as IMS-AlImBr (Figure 4), despite containing the same number and distribution of benzylic and allylic H-atom donors. Clearly, the inductive effects of imidazolium functionality have a positive effect on allyl group oligomerization, while resonance effects impact positively on allylic H-atom donation. The net result is a macromonomer with a unique relationship between cure rate and cure yield. 3.3. Solvent-Free N-Alkylations. The most cost-effective and environmentally responsible method to synthesize ionomers is a one-step, solvent-free process. Since industrialscale operations generally involve reactive extrusion with residence times on the order of minutes, reaction rates can dictate the commercial viability of this technology. AlIm was more nucleophilic than VIm in solvent-borne alkylations and proved to be more reactive under solvent-free conditions, as illustrated in Figure 6. These experiments involved mixing BIMS with 2 equiv of nucleophile at 100 °C in a Haake Polylab mixing bowl. Benzylic bromide conversion reached 100% in just

Figure 7. Stress relaxation of cross-linked imidazolium bromide ionomer derivatives of BIMS (100 °C, 2° strain). 17534

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

for 48 days. The high cross-link density of this material, identifiable in both the static modulus data (Figure 7) and dynamic storage modulus data (Figure 8), resulted in an elongation at break that is too low for many applications. However, the peroxide cure extent of isobutylene-rich macromonomers, of which IMS-AllmBr is one, scales linearly with CC content.5 Therefore, mixed alkylations of AlIm and BuIm, for example, provide a means of tailoring the reactive ionomer’s composition to meet a desired cross-link density target. Table 1 also provides tensile data for samples heated to 110 °C in air for 48 days. This accelerated aging test showed no differences in elongation at break and minimal changes in maximum tensile stress, suggesting that the cured ionomer can provide good oxygen and heat resistance, even without filler, antioxidants, or other stabilizers. However, more detailed analysis of long-term stability is required to assess oxidation of our thermoset materials.

by the thermoset ionomers was relatively small, with the static modulus declining by just 10% for IMS-ImEmABr-XL, and on the order of 30% for IMS-VImBr-XL and IMS-AlImBr-XL. Clearly, the stability of the covalent network within the thermosets dominates the relaxation response. Author: The online 6 has been changed to ref 6 in the previous paragraph.et derivatives of isobutylene-rich elastomers provide a purely elastic response, with the static modulus remaining unchanged with time.6 That the thermoset ionomers demonstrate some relaxation is consistent with their hybrid network character, with covalent cross-links contributing a constant equilibrium modulus, and ion-pair aggregates providing time dependent relaxation. Further insight into the strength and stability of the ionomers of interest was gained by measuring dynamic storage modulus as a function of temperature (Figure 8). For clarity, only data

4. CONCLUSIONS Substituents at the 1-position of imidazole nucleophiles have a significant effect on reactivity toward the benzylic bromide electrophile within BIMS, with alkyl and allyl functionality providing substantially greater nucleophilicity than a vinyl group. Of the ten N-substituted imidazoles examined, the allylic functionalized ionomer, IMS-AlImBr, is the most amenable to solvent-free preparation and provides high cure yields at moderate cross-linking rates. This unique relationship is attributed to activation of the allylic functionality by the adjacent imidazolium moiety, giving a pendant −CH2−CH CH2 group that engages in both radical addition and allylic Hatom transfer to give highly branched architectures. The resulting thermoset ionomer provides physical properties that are consistent with a hybrid ionic/covalent network and exhibits good intrinsic thermal stability.

Figure 8. Temperature sweep of cross-linked imidazolium bromide ionomer derivatives of BIMS between 100 and 200 °C (1 Hz, 3° arc).



collected while heating from 100 to 200 °C is plotted. However, measurements acquired during a subsequent cooling sequence coincided with those observed during heating, indicating that none of the ionomers suffered from thermal degradation over the time scale of the experiment. The data further demonstrate the temperature insensitivity of the thermoset’s storage modulus. Whereas uncured ionomers such as IMS-BuImBr respond to increasing temperature with a reduction in ionic network strength,38 and nonionic thermosets respond with an increase in G′ due to entropydriven elasticity,39 the thermoset ionomers do not change appreciably; strengthening of the covalent network is offset by weakening of the ionic network. We conclude with a brief examination of the tensile properties of IMS-AlImBr-XL, since this is the most promising thermosetting formulation produced in this study. Table 1 provides 5% moduli and failure properties of the peroxidecured ionomer as initially prepared, and after heating to 110 °C

Corresponding Author

*Phone: 613 533-6266. Fax: 613 533-6637. E-mail: parent@ queensu.ca. Notes

The authors declare no competing financial interest.



after 48 days @ 110 °C

physical property

mean

SD

mean

SD

5% modulus (MPa) maximum tensile stress (MPa) elongation at break (%)

1.15 1.48 142

0.06 0.29 12.8

1.32 1.74 142

0.04 0.16 10.7

REFERENCES

(1) Shaffer, T.; Tsou, A.; Webb, R. In Kirk-Othmer Encyclopedia of Chemical Technology; Kirk-Othmer; John Wiley and Sons: 2007; Vol. 4. (2) Oxley, C. E.; Wilson, G. J. A peroxide curing butyl rubber. Rubber Chem. Technol. 1969, 42, 1147−1154. (3) Loan, L. D. The reaction between dicumyl peroxide and butyl rubbers. J. Polym. Sci., Part A: Polym. Chem. 1964, 2, 2127−2134. (4) Wang, H. C.; Fusco, J. V.; Hous, P. Acrylate ester modifications of isobutylene/p-methylstyrene copolymer. Rubber World 1994, 37− 41. (5) Xiao, S.; Parent, J. S.; Whitney, R. A.; Knight, L. K. Synthesis and characterization of poly(isobutylene-co-isoprene)-derived macromonomers. J. Polym. Sci., Part A: Polym. Chem. 2010, 48, 4691−4696. (6) Ozvald, A.; Parent, J. S.; Whitney, R. A. Hybrid ionic/covalent polymer networks derived from functional imidazolium ionomers. J. Polym. Sci., Part A: Polym. Chem. 2013, 51, 2438−2444. (7) Kleczek, M.R. Imidazolium ionomer derivatives of isobutylene-rich elastomers: Thermosets, emulsions, filler composites and clay nanocomposites, Ph.D. Thesis, Queen’s University, 2013.

Table 1. Tensile Properties of Cross-Linked IMS-AlImBr initial properties

AUTHOR INFORMATION

17535

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536

Industrial & Engineering Chemistry Research

Article

(8) Parent, J. S.; Liskova, A.; Resendes, R. Isobutylene-based ionomer composites: siliceous filler reinforcement. Polymer 2004, 45, 8091− 8096. (9) Parent, J. S.; Penciu, A.; Guillen-Castellanos, S. A.; Liskova, A.; Whitney, R. A. Synthesis and characterization of isobutylene-based ammonium and phosphonium bromide ionomers. Macromolecules 2004, 37, 7477−7483. (10) Mori, H.; Yahagi, M.; Endo, T. RAFT polymerization of Nvinylimidazolium salts and synthesis of thermoresponsive ionic liquid block copolymers. Macromolecules 2009, 42, 8082−8092. (11) Parent, J. S.; Malmberg, S. M.; Whitney, R. A. Auto-catalytic chemistry for the solvent-free synthesis of isobutylene-rich ionomers. Green Chem. 2011, 13, 2818−2824. (12) Parent, J. S.; Thom, D.; White, G.; Whitney, R. A.; Hopkins, W. Thermal stability of brominated poly(isobutylene-co-isoprene). J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 2019−2026. (13) Malmberg, S. M.; Parent, J. S.; Pratt, D. A.; Whitney, R. A. Isomerization and elimination reactions of brominated poly(isobutylene-co-isoprene). Macromolecules 2010, 43, 8456−8461. (14) Salamone, J. C.; Israel, S. C.; Taylor, P.; Snider, B. Synthesis and homopolymerization studies of vinylimidazolium salts. Polymer 1973, 14, 639−644. (15) Masahiro, Y. Japan Patent 2010-260850, November 18, 2012. (16) Green, M. D.; Wang, D.; Hemp, S. T.; Choi, J.-H.; Winey, K. I.; Heflin, J. R.; Long, T. E. Synthesis of imidazolium ABA triblock copolymers for electromechanical transducers. Polymer 2012, 53, 3677−3686. (17) Aitken, B. S.; Lee, M.; Hunley, M. T.; Gibson, H. W.; Wagener, K. B. Synthesis of precision ionic polyolefins derived from ionic liquids. Macromolecules 2010, 43, 1699−1701. (18) Ringstrand, B.; Seifert, S.; Firestone, M. A. Preparation of a solution-processable, nanostructured ionic polyacetylene. J. Polym. Sci., Part B: Polym. Phys. 2013, 51, 1215−1227. (19) Hindman, M. S.; Stanton, A. D.; Irvin, C. A.; Wallace, D. A.; Moon, J. D.; Reclusado, K. R.; Liu, H.; Belmore, K. A.; Liang, Q.; Shannon, M. S.; Turner, C. H.; Bara, J. E. Synthesis of 1,2-dialkyl-, 1,4(5)-dialkyl-, and 1,2,4(5)-trialkylimidazoles via a one-pot method. Ind. Eng. Chem. Res. 2013, 52, 11880−11887. (20) Bogdal, D.; Jaskot, K. Synthesis of vinyl monomers with active azaaromatic groups. Phase-transfer catalytic approach. Synth. Commun. 2000, 30, 3341−3352. (21) ASTM International, Designation: D3182 - 07. Standard Practice for Rubber - Materials, Equipment, and Procedures for Mixing Standard Compounds and Preparing Standard Vulcanized Sheets. http://enterprise2.astm.org/DOWNLOAD/D3182.144734-1. pdf (accessed 24 July 2013). (22) ASTM International, Designation: D412 - 06a. Standard Test Methods for Vulcanized Rubber and Thermoplastic Elastomers − Tension. (23) ASTM International, Designation: D4482 - 11. Standard Test Method for Rubber Property - Extension Cycling Fatigue. (24) Bartlett, P. D.; Altschul, R. The Polymerization of allyl compounds. II. Preliminary kinetic study of the peroxide-induced polymerization of allyl acetate. J. Am. Chem. Soc. 1945, 67, 816−822. (25) Auluck, F. C.; Kothari, D. S. The Kinetic Theory of Rubber. J. Chem. Phys. 1943, 11, 387−392. (26) Technical Bulletin, Atofina Chemicals, 2001. (27) Schildknecht, C. E. In Allyl Compounds and Their Polymers (Including Polyolefins); John Wiley & Sons: New York, 1973. (28) Vikram Siva Shanmugam, K. Peroxide curable butyl rubber derivatives. PhD. Thesis, Queen’s University, 2012. (29) Hyslop, D. K.; Parent, J. S. Functional nitroxyls for use in delayed-onset polyolefin cross-linking. Macromolecules 2012, 45, 8147−8154. (30) Zhang, G.; Liu, X.; Li, B.; Bai, Y. Solution copolymerization of acrylonitrile with 1-allyl-3-methyl-imidazolium. Acta Polym. Sin. 2009, 3, 216−221. (31) Avila, D. V.; Brown, C. E.; Ingold, K. U.; Lusztyk, J. Solvent effects on the competitive β-scission and hydrogen atom abstraction

reactions of the cumyloxyl radical. Resolution of a long-standing problem. J. Am. Chem. Soc. 1993, 115, 466−470. (32) Hirao, M.; Ito, K.; Ohno, H. Preparation and polymerization of new organic molten salts; N-alkylimidazolium salt derivatives. Electochim. Acta 2000, 45, 1291−1294. (33) Yuan, J.; Antonietti, M. Poly(ionic liquid) latexes prepared by dispersion polymerization of ionic liquid monomers. Macromolecules 2011, 44, 744−750. (34) Petrak, K. L. Reactivity of some vinylimidazoles towards other vinyl monomers in radical copolymerization. J. Polym. Sci., Polym. Lett. Ed. 1978, 16, 393−399. (35) Damas, C.; Baggio, S.; Brembilla, A.; Lochon, P. Microstructure study of new amphiphilic copolymers from 3-alkyl-1-vinylimidazolium salts. Eur. Polym. J. 1997, 33, 1219−1224. (36) Bartlett, P. D.; Tate, F. A. The Polymerization of Allyl Compounds. VI. The polymerization of allyl-1-d2 acetate and the mechanism of its chain termination. J. Am. Chem. Soc. 1953, 75, 91− 95. (37) Garrett, G. E.; Mueller, E.; Pratt, D. A.; Parent, J. S. Reactivity of polyolefins toward cumyloxy radical: Yields and regioselectivity of hydrogen atom transfer. Macromolecules 2014, 47, 544−551. (38) Eisenberg, A.; Kim, J. S. In Introduction to Ionomers; Wiley: New York, 1998. (39) Gao, J.; Weiner, J. H. Range of validity of the entropic spring concept in polymer melt relaxation. Macromolecules 1992, 25, 3462− 3467.

17536

dx.doi.org/10.1021/ie502853z | Ind. Eng. Chem. Res. 2014, 53, 17527−17536