Cross-Linked Block Copolymer/Ionic Liquid Self-Assembled Blends for Polymer Gel Electrolytes with High Ionic Conductivity and Mechanical Strength

Poly(propylene oxide)−poly(ethylene oxide)−poly(propylene oxide) (PPO−PEO−PPO) block copolymers (BCPs) with cross-linkable end groups were synthesized, blended with an ionic liquid (IL) diluent, and cross-linked to form polymer gel electrolytes. The IL prevented crystallization of PEO at high concentrations, enabling fast ion transport. In addition, the IL was selective for the PEO block, inducing strong microphase separation in what are otherwise disordered or weakly ordered BCP melts. Cross-linking the BCPs in the presence of the IL resulted in the formation of solid, elastic gels with high ionic conductivitiesgreater than 1.0 mS/cm at 25 °C for some compositions. However, it was found that neither the presence or absence of microphase separation nor the BCP composition of the microphase separated gels substantially influenced ionic conductivity. Increasing the cross-link density through the use of phaseselective PEOand PPO-based cross-linking reagents was also evaluated. It was revealed that confinement of cross-links to the PPO rich domains through the use of PPO-based diacrylates enhanced the mechanical strength of the gels without detriment to the ionic conductivity. Conversely, cross-linking in the PEO-rich domains through the use of PEO-based acrylates significantly reduced conductivity. Isolation of cross-links within a minor nonconducting domain in a microphase separated gel is a viable strategy for mechanical property enhancement without a large sacrifice in conductivity, effectively decoupling ionic conductivity and mechanical strength. This approach yielded solid-like gel electrolytes fabricated from BCPs that can be produced inexpensively, with ionic conductivities of 0.64 mS/cm at 25 °C and a frequency independent storage modulus of approximately 400 kPa. ■ INTRODUCTION Solid polymer electrolytes are ion-conducting media composed of polymers that are capable of coordinating with and transporting charged molecules. These materials have many desirable characteristics for applications such as anion and proton exchange membranes in fuel cells, electrolyte layers in dye-sensitized solar cells, as separators in capacitors, and electrolyte layers in lithium batteries. In particular, there are many benefits to using solid polymer electrolytes in place of the conventional liquid electrolytes in lithium ion batteries. These advantages include reduced flammability by the elimination of flammable organic solvents used in liquid electrolytes and also the capacity to construct electrolytes as thin films, leading to greater energy density in batteries. Unfortunately, the ionic conductivity of solid polymer electrolytes is generally quite poor, well below the 1.0 mS/ cm (at 25 °C) minimum required for a practical battery electrolyte. One reason is the high crystallinity of most ioncoordinating polymers. It was established very early in the development of polymer electrolytes that polymer crystallization suppresses ionic conductivity. More recent work has established a clear correlation between polymer segmental dynamics and ion transport. In addition, stoichiometric polymer/salt complexes can form in mixtures of lithium salts and ion-coordinating polymers, decreasing ionic conductivity due to the increased physical restraints imparted by these complexes. Therefore, to achieve fast ion transport, a diluent is required to plasticize the polymer and ensure complete lithium salt disassociation. Such polymer gel electrolytes should have superior performance to all-solid polymer electrolytes. Room temperature ionic liquids (ILs) are a very promising class of diluents for the fabrication of polymer gel electrolytes. Room temperature ILs are salts of organic cations and anions with low melting temperatures (<25 °C), negligible vapor pressure, and have thermal and electrochemical stability superior to most organic solvents. As such, replacing flammable organic solvents used in liquid electrolytes or polymer gel electrolytes with low-flammability ILs would allow for the construction of batteries with reduced fire hazards. ILs also have the capacity to plasticize ion-coordinating polymers and to dissolve lithium salts, promoting fast Received: June 29, 2013 Revised: October 6, 2013 Published: November 26, 2013 Article pubs.acs.org/Macromolecules © 2013 American Chemical Society 9313 dx.doi.org/10.1021/ma401302r | Macromolecules 2013, 46, 9313−9323 ion transport. ILs have been used on many occasions as diluents for polymer gel electrolytes. However, if the polymer is fully plasticized by an IL diluent, it will behave as a polymer solution or melt and cannot form a mechanically stable film as required for a thin film electrochemical cell. To counter this, several methods have been explored to provide mechanical support to IL/polymer gel electrolytes. These methods include polymerizing a polymer in an IL solution along with a cross-linking comonomer, swelling poly(vinylidenefluoride-co-hexafluoropropylene) random copolymers in an IL to make a physical gel, and chemical cross-linking of poly(ethylene oxide) (PEO) blended with an IL. Many studies have also investigated the use of block copolymers (BCPs) for polymer electrolytes, as well as mixtures of BCPs with IL diluents or BCPs incorporating polymerized ionic liquids. BCPs consist of two dissimilar polymer chains (blocks) linked together by a covalent bond. When the repulsive interactions between the component blocks are sufficiently large, the blocks phase separate on a nanometer scale (historically termed microphase separation) into welldefined, periodic nanostructures. It has been suggested that BCP microphase separation creates well-defined ion conducting channels, and indeed several reports have shown that control over BCP morphology and microdomain orientation can be used to enhance ionic conductivity. Strong microphase separation and large microdomain width also favor high conductivity. In addition, by using a BCP consisting of both an ion-conducting block and a rigid block, it may be possible to obtain both high conductivity and high mechanical strength. This would be a significant development, as there is typically a trade-off in polymer electrolytes between ionic conductivity and mechanical strength due to the correlation between slow polymer dynamics and reduced ionic conductivity. Microphase separation of an ionconducting block and a reinforcing block reportedly allows ionic conductivity and mechanical strength to be decoupled, allowing this trade-off to be avoided. However, it is not universally reported that BCP microphase separation is beneficial for ion transport and polymer electrolyte applications in general. A few reports have concluded that controlling the BCP morphology has no effect on ionic conductivity. In addition, transitioning from an ordered BCP morphology to a disordered system was also found to have little measurable effect on conductivity. Further work is therefore required to determine the optimal function of BCPs in electrolyte applications. In this work, the potential benefits of using BCPs for IL containing polymer gel electrolytes were assessed by evaluating the performance of cross-linked IL/polymer gels with varying phase behavior. Gels fabricated from homopolymers with only a single phase, from low molecular weight BCPs having weak microphase separation, and from higher molecular weight BCPs with strong microphase separation were investigated. The effects of BCP phase behavior on the mechanical properties and ionic conductivity were studied to contribute to the understanding of the role of BCP microphase separation in polymer electrolytes. In addition, the feasibility of using PPO−PEO−PPO BCPs as polymer electrolytes has been investigated. These polymers can be produced inexpensively on an industrial scale, as demonstrated by BASF’s catalog of reverse Pluronic surfactants. We used a sample of a reverse Pluronic (25R4), as well as labmade analogues that can be made by the same chemistry. The low cost of producing these polymers may prove beneficial for practical implementation as battery electrolytes. The crosslinking strategy used may prove beneficial as well, as covalent cross-links persist up to thermal decomposition, whereas physical cross-links such as those from glassy polymers relax out when the glass transition temperature (Tg) is exceeded. 56 ■ EXPERIMENTAL SECTION Materials. The IL used in this work is 1-butyl-3-methylimidazolium hexafluorophosphate ([BMI][PF6]), obtained from SigmaAldrich. PEO having Mn equal to 4600 g/mol (PEG 4600) and 8000 g/mol (PEG 8000) were also purchased from Sigma-Aldrich. Pluronic 25R4 was donated from BASF. Synthesis and distillation reagents were all obtained from Sigma-Aldrich: propylene oxide, calcium hydride (CaH2), potassium hydride (KH, in mineral oil), 12.0 M hydrochloric acid (HCl) 18-crown-6, methacryloyl chloride, triethylamine (TEA), sodium metal (in mineral oil), and benzophenone. All solvents were purchased from Fisher Scientific. A radical inhibitor 4-hydroxy-2,2,6,6-tetramethyl-1-piperidine-1-oxyl (4-OH− TEMPO) was also purchased from Fisher. Cross-linking oligomers, poly(propylene glycol) diacrylate (Mn = 800 g/mol) and poly(ethylene glycol) diacrylate (Mn = 700 g/mol) were purchased from Sigma-Aldrich. A free-radical generating photoinitiator, 2,2-dimethoxy2-phenylacetophenone was also purchased from Sigma-Aldrich. Tetrahydrofuran (THF) was distilled over a sodium/benzophenone ketyl still prior to use. Propylene oxide was vacuum distilled over calcium hydride. Prior to use, the mineral oil was removed from KH by mixing the slurry with distilled THF, followed by gravity filtration. 18crown-6 was dried under vacuum at 70 °C for 20 h. TEA was distilled over CaH2. The cross-linkable poly(ethylene oxide) and poly(propylene oxide) oligomers were stripped of inhibitor by passage through a plug of activated basic alumina. The purified oligomers were stored in a 4 °C refrigerator until required. All other reagents were used as received. Caution must be used when handling sodium metal, CaH2, and KH as these reagents are all highly reactive with water. BCP Synthesis. In this work, IL/polymer blends were made from three different PPO-b-PEO-b-PPO triblock copolymers and from the PEG 8000 homopolymer. Low molecular weight BCPs are commercially available, in the form of Pluronic 25R4 (Mn = 3600 g/ mol, 0.40 PEO). Higher molecular weight PPO-b-PEO-b-PPO copolymers are not commercially available, and had to be synthesized. These BCPs were prepared from PEO macroinitiators with terminal hydroxyl end groups, using KH to activate the end groups and 18crown-6 as a phase transfer catalyst. Polymerization of propylene oxide was then carried out by anionic ring-opening polymerization from these end groups. Detailed procedures are contained in the Supporting Information section. The characteristics of the synthesized and the purchased polymers are summarized in Table 1. For the BCPs, BASF’s naming convention for Pluronic was employed: the first two digits multiplied by 100 is the combined Mn of the PPO blocks, “R” indicates “reverse” as PPO comprises the end blocks, and the last digit multiplied by 10 is the weight fraction of PEO. The molecular weight and composition of the BCPs was determined by H NMR. The dispersity of all the polymers was measured by GPC (THF, polystyrene standard). Table 1. Characteristics of the Polymers Used in This Work polymer total Mn (g/mol) weight fraction PEO dispersity