Polymerized Ionic Liquid Block and Random Copolymers: Effect of

Article pubs.acs.org/Macromolecules

Polymerized Ionic Liquid Block and Random Copolymers: Effect of Weak Microphase Separation on Ion Transport Yuesheng Ye,† Jae-Hong Choi,‡ Karen I. Winey,*,‡ and Yossef A. Elabd*,† †

Department of Chemical and Biological Engineering, Drexel University, Philadelphia, Pennsylvania 19104, United States Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, Pennsylvania 19104, United States

‡

S Supporting Information *

ABSTRACT: A series of polymerized ionic liquid (PIL) block and random copolymers were synthesized from an ionic liquid monomer, 1-[(2-methacryloyloxy)ethyl]-3-butylimidazolium bis(trifluoromethanesulfonyl)imide (MEBIm-TFSI), and a nonionic monomer, methyl methacrylate (MMA), at various PIL compositions with the goal of understanding the influence of morphology on ion transport. For the diblock copolymers, the partial affinity between the PIL and PMMA blocks resulted in a weakly microphase-separated morphology with no evident long-range periodic structure across the PIL composition range studied, while the random copolymers revealed no microphase separation. These morphologies were identified with a combination of techniques, including differential scanning calorimetry, small-angle X-ray scattering, and transmission electron microscopy. Surprisingly, at similar PIL compositions, the ionic conductivity of the block copolymers were ca. 2 orders of magnitude higher than the random copolymers despite the weak microphase-separated morphology evidenced in the block copolymers. We attribute the higher conductivity in the block copolymers to its microphase-separated morphology, since significant differences in conductivity are still observed even when differences in glass transition temperature are considered. This work demonstrates that local confinement and connectivity of conducting ions in nanoscale ionic domains in PIL block copolymers can accelerate ion transport significantly.

■

INTRODUCTION

The self-assembled nanostructured morphologies of block copolymers are of great interest as it pertains to the utilization of block copolymers as solid-state polymer electrolytes. Specifically, in regards to tuning these nanostructured morphologies to benefit ion transport, e.g., by forming nanostructured ion-conducting channels. Recently, due to the unique properties of ILs, ILs have been introduced in neutral block copolymers to investigate how the addition of these liquid salts impacts nanostructured morphology and ionic conductivity. For example, anisotropic ionic conductivity was observed in IL-doped block copolymers with lamellae morphologies,15,16 while the ionic conductivity was direction independent when the ion-conducting paths were well connected in a three-dimensional network morphology.17 Additionally, recent results have shown an ion diffusion enhancement in an IL-doped block copolymer compared to its IL-doped homopolymer analogue.18 In comparison to ILdoped block copolymers, lessons can also be learned from previous work on lithium (Li) salt-doped block copolymers as the physics of ion transport are similar in both systems.19−22 Despite the number of reports on the morphologies of block copolymers and their effect on the solid-state ion transport, it is

Polymerized ionic liquid (PIL) block copolymers, a new type of solid-state polymer electrolyte, have been recently synthesized and are of interest for energy conversion and storage devices, such as fuel cells, batteries, supercapacitors, and solar cells.1 PIL block copolymers conjoin the advantages of block copolymers and PILs. Specifically, the self-assembly of block copolymers can result in a range of nanostructures (e.g., body-centeredcubic spheres, hexagonal cylinders, bicontinuous gyroid, lamellae), where morphology and domain size are tunable,2 while PILs possess unique physiochemical properties, such as high solid-state ionic conductivity, chemical and thermal stability, and widely tunable physical properties (e.g., via anion exchange).3 Different from block copolymers doped with ionic liquid (IL), the introduction of IL moieties through covalent attachment to the polymer backbone or side chain mitigates the leakage of liquid electrolytes during extended use.4 So far, only a handful of PIL block copolymers have been reported in the literature, and these studies have mainly focused on micellar structures,5,6 self-assembly in solutions,7−9 magnetic materials,10 carbon dioxide capture,11 stimuli-responsive materials,12 and morphological properties.13 Fundamental studies on nanostructured morphology and its influence on electrochemical properties, such as solid-state ion transport in PIL block copolymer films, remain relatively unexplored.1,14 © 2012 American Chemical Society

Received: May 21, 2012 Revised: August 16, 2012 Published: August 27, 2012 7027

dx.doi.org/10.1021/ma301036b | Macromolecules 2012, 45, 7027−7035

Macromolecules

Article

Scheme 1. Synthesis of (a) Poly(MMA-b-MEBIm-TFSI)) Block Copolymers and (b) Poly(MMA-r-MEBIm-TFSI)) Random Copolymersa

a (a) Synthesis of PIL diblock copolymers, poly(MMA-b-MEBIm-TFSI): (1a) 4-cyano-4-(phenylcarbonothioylthio)pentanoic acid (CTA), AIBN, THF, 70 °C, 5 h; (2a) MEBIm-Br, AIBN, DMF, 70 °C, 5 h; (3a) LiTFSI, DMF, 50 °C, 24 h. (b) Synthesis of PIL random copolymers, poly(MMAr-MEBIm-TFSI): (1b) MEBIm-Br, AIBN, DMF, 70 °C, 5 h; (2b) LiTFSI, DMF, 50 °C, 24 h.

transport in PILs. A significant increase (2 orders of magnitude) in ionic conductivity from the random copolymers to the block copolymers was observed at similar PIL compositions and is attributed to the weak microphase separation in the block copolymer morphology. These results suggest that the local confinement of conducting ions in nanoscale ionic domains can accelerate ion transport, where strong microphase separation is not required for significant enhancements in ion transport.

worthwhile to note that these previous studies involve the mixture of either a solid salt (e.g., Li salt) or a molten salt (e.g., IL) with a neutral block copolymer. In other words, both cations and anions migrate simultaneously in these systems to provide ion conduction. In contrast, PIL block copolymers are single-ion conductors, where only one type of ion (e.g., anion) is mobile, while the counterion (e.g., cation) is covalently attached to the polymer chain. Thus, a PIL block copolymer provides a unique model system to investigate morphology− ion transport relationships in single-ion conductors with ILbased chemistry. One recent report on PIL block copolymers demonstrates that morphology is dependent on both PIL composition and film processing conditions, and these differences in morphology have a significant impact on ionic conductivity.1 However, there are only a few studies on the solid-state ionic conductivity of PIL block copolymers.1,14 In order to obtain a better understanding of the impact the PIL block copolymer microphase-separated morphology on ionic conductivity, it would be worthwhile to investigate this in relation to a better control system, e.g., its random copolymer analogue. To the best of our knowledge, such a comparison between PIL block and random copolymers has not been conducted, and here we provide such a study to benefit the understanding of morphology−ion transport relationships and to guide the design of new PIL block copolymers. In this study, a series of PIL diblock copolymers were synthesized from 1-[(2-methacryloyloxy)ethyl]-3-butylimidazolium bis(trifluoromethanesulfonyl)imide (MEBIm-TFSI) IL monomer and methyl methacrylate (MMA) nonionic monomer at various PIL compositions using the reversible addition− fragmentation chain transfer (RAFT) polymerization technique. An analogous series of PIL random copolymers at similar PIL compositions were synthesized using conventional free radical polymerization. The comparison of PIL block and random copolymers at similar PIL compositions allows for a clear understanding of the impact of morphology on ion

■

EXPERIMENTAL SECTION

Materials. 4-Cyano-4-(phenylcarbonothioylthio)pentanoic acid (chain transfer agent (CTA), >97%, HPLC), tetrahydrofuran (THF, ≥99.9%), N, N-dimethylformamide (DMF, 99.9%, HPLC), methanol (99.9%, HPLC), diethyl ether (≥98%), acetonitrile (anhydrous, 99.8%), calcium hydride (CaH2, 95%), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, 97%), lithium bromide (LiBr, ≥99%), and dimethyl-d6 sulfoxide (DMSO-d6, 99.9 atom % D, contains 0.03% v/v TMS) were used as received from SigmaAldrich. Azobis(isobutyronitrile) (AIBN, 98%, Sigma-Aldrich) was purified by recrystallization twice from methanol. Methyl methacrylate (MMA, 99%, Sigma-Aldrich) was purified by distillation over CaH2 at a reduced pressure. Ionic liquid monomer, 1-[(2-methacryloyloxy)ethyl]-3-butylimidazolium bromide (MEBIm-Br), was prepared similarly according to the literature.3 The quaternization reaction was carried out at room temperature for 30 h without using an inhibitor. Dialysis tubing (Spectra/Por biotech membrane, molecular weight cutoff (MWCO) = 500) was purchased from Fisher Scientific. Ultrapure deionized (DI) water with resistivity ca. 16 MΩ cm was used as appropriate. Synthesis of PMMA Macro-CTA. The preparation of PMMA macro-chain-transfer agent (macro-CTA) is shown in Scheme 1a(1a). 25.168 g of MMA (251.378 mmol), 141.5 mg of CTA (0.506 mmol), and 20.8 mg of AIBN (0.127 mmol) were mixed with 9 mL of THF in a 250 mL single-neck Schlenk flask. The flask was subjected to four freeze−pump−thaw degassing cycles followed by sealing the reactor and carrying out the reaction under static vacuum at 70 °C for 5 h. The resulting polymer was twice precipitated in methanol and dried under vacuum in an oven at room temperature for 24 h. Yield: 9.62 g 7028

dx.doi.org/10.1021/ma301036b | Macromolecules 2012, 45, 7027−7035

Macromolecules

Article

of solid particles (38.2%). 1H NMR (500 MHz, DMSO-d6, 23 °C) δ (ppm): 7.87−7.50 (m, C6H5), 3.57 (s, 3H, O−CH3), 1.84−1.76 (d, 2H, CH2−C(CH3)), 0.94−0.74 (d, 3H, CH2−C(CH3)). Mn = 13.1 kg mol−1 (NMR). SEC (DMF, 40 °C): Mn = 12.53 kg mol−1, Mw/Mn = 1.19 (against PEG/PEO standards). Synthesis of Diblock Copolymer Poly(MMA-b-MEBIm-Br). The synthesis of the PIL block copolymer precursor (poly(MMA-bMEBIm-Br-13.3)) is shown in Scheme 1a(2a). A typical example is given as follows. 3.006 g of IL monomer (MEBIm-Br) in DMF (MEBIm-Br/DMF = 1/1 w/w, MEBIm-Br = 4.738 mmol), 3.653 g of PMMA macro-CTA in DMF (PMMA/DMF = 1/2 w/w, PMMA = 0.097 mmol), and 1.6 mg of AIBN (0.010 mmol) were mixed with 5 mL of DMF solvent in a 50 mL Schlenk flask and subjected to four freeze−pump−thaw degassing cycles. After degassing, the reactor was sealed and the reaction was then carried out under static vacuum at 70 °C for 5 h. The resulting polymer was twice precipitated in DI water and subsequently washed extensively with DI water. The block copolymer was filtered and then dried under vacuum in an oven at 40 °C for 24 h. Yield: 1.365 g of solid particles (50.2%). 1H NMR (500 MHz, DMSO-d6, 23 °C) δ (ppm): 9.82 (s, 1H, N−CHN), 8.02 (d, 2H, N−CHCH−N), 4.64−4.29 (d, 6H, N−CH2−CH2−O, N− CH2−CH2−CH2), 3.57 (s, 3H, OCH3), 1.88 (s, 4H, CH2−C(CH3), N−CH2−CH2−CH2−CH3), 1.32 (s, 5H, N−CH2−CH2−CH2−CH3, CH2−C(CH3)), 0.93 (s, 6H, N−CH2−CH2−CH2−CH3, CH2− C(CH3)), 0.77 (s, 3H, CH2−C(CH3)). SEC (DMF, 40 °C): Mn = 22.93 kg mol−1, Mw/Mn = 1.31 (against PEG/PEO standards). Synthesis of Diblock Copolymer Poly(MMA-b-MEBIm-TFSI). The anion exchange from Br to TFSI neutralized form is shown in Scheme 1a(3a). Poly(MMA-b-MEBIm-Br-13.3) (0.317 g, 0.014 mmol) and LiTFSI (0.653 g, 2.275 mmol) were mixed with DMF (5 mL) and then stirred at 50 °C for 24 h. The reaction mixture was twice precipitated into methanol/water (1/1 v/v) and washed extensively with DI water. The resulting polymer was filtered and dried under vacuum in an oven at 40 °C for 24 h. Yield: 0.277 g of solid particles (72.5%). 1H NMR (500 MHz, DMSO-d6, 23 °C) δ (ppm): 9.24 (s, 1H, N−CHN), 7.83−7.75 (d, 2H, N−CHCH− N), 4.48−4.20 (d, 6H, N−CH2−CH2−O, N−CH2−CH2−CH2), 3.56 (s, 3H, OCH3), 1.81 (s, 4H, CH2−C(CH3), N−CH2−CH2−CH2− CH3), 1.32 (s, 5H, N−CH2−CH2−CH2−CH3, CH2−C(CH3)), 0.94 (s, 6H, N−CH2−CH2−CH2−CH3, CH2−C(CH3)), 0.74 (s, 3H, CH2−C(CH3)). Elemental Anal. Calcd: C, 48.78; H, 6.30; N, 3.61; F, 9.79; S, 5.51; Br, 0.00. Found: C, 47.09; H, 5.96; N, 4.01; F, 10.66; S, 5.97; Br,