High-Modulus, High-Conductivity Nanostructured Polymer Electrolyte

Dec 13, 2013 - This strategy holds tremendous potential to advance lithium-ion battery technology by enabling the use of lithium metal anodes or to se...
11 downloads 26 Views 1MB Size
Download PDF
Letter pubs.acs.org/NanoLett

High-Modulus, High-Conductivity Nanostructured Polymer Electrolyte Membranes via Polymerization-Induced Phase Separation Morgan W. Schulze,† Lucas D. McIntosh,† Marc A. Hillmyer,*,‡ and Timothy P. Lodge*,†,‡ †

Department of Chemical Engineering and Materials Science and ‡Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455-0431, United States S Supporting Information *

ABSTRACT: The primary challenge in solid-state polymer electrolyte membranes (PEMs) is to enhance properties, such as modulus, toughness, and high temperature stability, without sacrificing ionic conductivity. We report a remarkably facile one-pot synthetic strategy based on polymerization-induced phase separation (PIPS) to generate nanostructured PEMs that exhibit an unprecedented combination of high modulus and ionic conductivity. Simple heating of a poly(ethylene oxide) macromolecular chain transfer agent dissolved in a mixture of ionic liquid, styrene and divinylbenzene, leads to a bicontinuous PEM comprising interpenetrating nanodomains of highly cross-linked polystyrene and poly(ethylene oxide)/ionic liquid. Ionic conductivities higher than the 1 mS/cm benchmark were achieved in samples with an elastic modulus approaching 1 GPa at room temperature. Crucially, these samples are robust solids above 100 °C, where the conductivity is significantly higher. This strategy holds tremendous potential to advance lithium-ion battery technology by enabling the use of lithium metal anodes or to serve as membranes in high-temperature fuel cells. KEYWORDS: Polymer electrolyte, ionic liquid, bicontinuous, polymerization-induced phase separation, lithium-ion battery, high-temperature fuel cell GPa for T < 125 °C), and (iii) ionic conductivity in excess of 1 mS/cm can be produced by this facile and scalable process. To the best of our knowledge, no other polymer electrolyte reported achieves such a combination of modulus (≥108 Pa) and ionic conductivity (≥103 S/cm), nor the high temperature robustness. Seo and Hillmyer recently reported that arresting microphase separation during the growth of a diblock copolymer is a simple route to mechanically robust nanostructured materials with a bicontinuous morphology.22 Using a poly(lactide) macro-chain transfer agent (PLA-CTA), they prepared nanoporous crosslinked polystyrene monoliths via sequential reversible-addition fragmentation chain transfer (RAFT) polymerization and hydrolysis of the PLA domain. We have extended this polymerization-induced phase separation (PIPS) approach to the preparation of mechanically rigid block polymer electrolytes via the polymerization of a styrene/divinylbenzene mixture from macromolecular PEO-CTA in the presence of an ionic liquid (Scheme 1). The controlled RAFT polymerization process induces partitioning of growing block polymer chains into nanoscale domains, and concerted chemical cross-linking by divinylbenzene during polymerization of the styrene restrains the coarsening of the resultant bicontinuous morphology. The direct integration of the ionic liquid and/or

P

olymer electrolyte membranes (PEMs) are promising alternatives to conventional liquid electrolytes in energy storage applications such as lithium batteries and high temperature fuel cells.1−9 Effective PEMs must maintain a combination of high modulus, toughness, environmental resistance, and high ionic conductivity during use.2,10−12 The key challenge is to achieve high mechanical and thermal performance without sacrificing the requisite ionic conductivity. The most successful strategies to date leverage nanostructured AB diblock copolymers, where block A is a glassy, rigid insulator and block B is a low glass transition ion conductor, to independently tune the mechanical and conductivity properties of the membrane.9,13−17 Previous studies have suggested that domain alignment is necessary to render pathways continuous for ion transport, and the continuity of the mechanically robust phase leads to superior mechanical properties.18−21 In this report, we demonstrate that nanostructured block polymers with a disordered, bicontinuous morphology are particularly attractive as highly conductive, thermally stable, and mechanically robust PEMs due to the long-range, isotropic continuity of high modulus and ion conducting domains. Our simple, onestep protocol exploits simultaneous in situ block copolymer formation and chemical cross-linking such that local segregation of a growing poly(styrene/divinylbenzene) segment from an ionic liquid-swollen poly(ethylene oxide) (PEO) domain is preserved. PEMs with (i) room temperature moduli near 1 GPa, (ii) significantly improved mechanical stability over previously reported systems at high temperature (E′ > 0.1 © 2013 American Chemical Society

Received: September 18, 2013 Revised: November 25, 2013 Published: December 13, 2013 122

dx.doi.org/10.1021/nl4034818 | Nano Lett. 2014, 14, 122−126

Nano Letters

Letter

Small-angle X-ray scattering (SAXS) profiles indicate that nanoscopic structural heterogeneities exist in PEMs prepared by this method (Figure S1). The broad principal scattering peak at low wavevector q, typical of microphase-separated but disordered block polymer structures, is accompanied by a secondary shoulder that becomes more pronounced as the ionic liquid concentration is increased. The development of diffuse higher order reflections indicates that the ionic liquid induces greater structural coherence on the length scale of two or more domains (ca. 5−15 nm). The characteristic length scale increases with increasing PEO molar mass, from 10−15 nm for samples prepared with 5 kg mol−1 PEO-CTA to 25−35 nm for samples prepared with 28 kg mol−1 PEO-CTA. Also, the position of the primary peak shifts to a lower wavevector and becomes more intense with increasing ionic liquid content, which is consistent with the introduction of BMITFSI as a selective solvent for the PEO phase and a corresponding increase in the effective degree of segregation between the two domains.24 Exposure of a monolithic PIPS PEM sample to 57 wt % aqueous HI solution at 60 °C resulted in quantitative etching of the PEO/IL composite domain, as confirmed gravimetrically and by IR analysis. As shown in Figure 1, a scanning electron microscopy (SEM) image of the resulting nanoporous structure mirrors the corresponding transmission electron microscopy (TEM) image of the unetched sample. In addition, analyses of Fourier transform data generated from TEM images are in agreement with SAXS intensity profiles, confirming that the TEM images are representative of the bulk morphology (Figure S2). Combined, the SAXS, SEM, DSC (differential scanning calorimetry, Figure S3), and TEM analysis of the cross-linked PEM samples point to a bicontinuous, nanostructured material with interpenetrating and percolating domains of cross-linked PS and IL-swollen PEO. The ionic conductivity of the resulting PEMs was obtained from measurements of the bulk impedance from 30 to 150 °C (Figure 2). The data were fit with excellent agreement to the Vogel−Fulcher−Tammann (VFT) equation,

Scheme 1. Reaction Scheme Used To Prepare Polymerization-Induced Phase Separation Polymer Electrolyte Membranes

Li-salt in the liquid precursor obviates post-polymerization saltdoping steps that are common to other dry diblock copolymerbased electrolytes. The principal ionic liquid (IL) utilized, 1butyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (BMITFSI), is immiscible with polystyrene with M ≥ 3 kg mol−1, thus leading to its partitioning into the PEO domains.23,24 The result is the one-step formation of a bicontinuous and nanostructured membrane consisting of a mechanically robust phase (cross-linked PS) and an ion conductive phase (PEO/BMITFSI). As a representative example, homogeneous solutions of 30 vol % PEO-CTA were prepared in a 4/1 molar mixture of styrene/divinylbenzene (S/D). BMITFSI (and mixtures with LiTFSI) could be added to this reaction mixture while retaining homogeneity over a wide compositional window (5−40 overall vol % BMITFSI). Table S1 summarizes the resulting composition of the PEMs in terms of the volume percent of conducting phase (PEO + BMITFSI), as well as the concentration of ionic liquid within the conducting phase. The presence of BMITFSI with the introduction of LiTFSI facilitates the dissolution of the lithium salt in the mixture. Heating the quaternary solution to 120 °C in the presence of radical initiator (AIBN) for at least 20 h resulted in a transparent, insoluble solid PEM. Additionally, other molar ratios of S/D have been shown to retain bicontinuity, while modulation of the cross-link density does not affect conductivity because mechanical and transport properties are decoupled via microphase separation.

⎛ −B ⎞ σ = σ0 exp⎜ ⎟ ⎝ T − T0 ⎠

(1)

which describes relaxation processes above the Tg in glassforming liquids.25 Here, σ0 is the conductivity at infinite temperature, B is a pseudoactivation energy related to the entropic barrier to ion motion, and T0 is the temperature at which molecular motion is frozen. The increase in conductivity as ionic liquid concentration is increased, as shown in Figure 2, is due primarily to an increase in the number of carrier ions. Conductivity is high in these PEMs and exceeds the critical benchmark of 1 mS/cm at 80 °C for the samples containing 21 vol % BMITFSI and 1 M LiTFSI in BMITFSI.15 The conductivity can be expressed as a fraction of the conductivity achieved in a homogeneous electrolyte of the same PEO/ionic liquid composition, σh, by26−28 σ = σh

fc τ

(2)

where fc and τ are the volume fraction and the tortuosity of the conducting phase, respectively. Given various experimental studies of small molecule transport in cocontinuous, heterogeneous media (e.g., gas and water filtration through disordered pores,28−30 and tracer molecules in ordered block polymers in 123

dx.doi.org/10.1021/nl4034818 | Nano Lett. 2014, 14, 122−126

Nano Letters

Letter

Figure 2. Ionic conductivity as a function of temperature for PEMs prepared with 5 kg mol−1 PEO-CTA. The inset is a photograph of a typical sample used for conductivity experiments. Open symbols: Samples prepared with BMITFSI. Filled symbols: Samples prepared with a 1 M mixture of LiTFSI in BMITFSI. Error bars (in some cases under the data points) are one standard deviation based on at least three samples. Overall salt concentrations are 5 (△), 7 (▲), 21 (□ and ■), and 40 vol % (○). Parameters of the VFT fits are provided in Table S2 of the Supporting Information.

Figure 1. Morphology of PIPS PEM samples prepared with 28 kg mol−1 PEO-CTA and 21 vol % BMITFSI. Upper panel: Scanning electron micrograph of the sample after etching of PEO and BMITFSI with 57 wt % aqueous hydroiodic acid. The remaining structure is cross-linked polystyrene. The sample was coated with 1−2 nm of Pt prior to imaging. Lower panel: Transmission electron micrograph of the same sample prior to etching. The PEO/ionic liquid domain appears dark after staining with RuO4. Both scale bars represent 100 nm.

the gyroid phase),26,27 one expects τ to attain values between 1.5 and 3. As shown in Figure S4, the conductivities observed are consistent with eq 2 and our anticipated range of τ, indicating that the conductive channels are continuous over macroscopic distances. The dynamic moduli, E′ and E″, of a representative PEM were determined as a function of frequency at temperatures up to 200 °C, and time−temperature superposition was used to generate master curves (Figure S5). Figure 3 compares E′ at a fixed frequency (10 rad/s) of this representative PEM to a sample without ionic liquid. The samples are glassy at room temperature and perform as high modulus solids (tan(δ) = E″/ E′ < 0.1) over the entire temperature range studied. The supporting cross-linked polystyrene phase allows for operation

Figure 3. Temperature-dependent linear elastic response of PIPS PEMs prepared with 28 kg mol−1 PEO-CTA and no ionic liquid (○) and 21 vol% BMITFSI (●). The inset is a photograph of a typical tensile bar used. The data points are the elastic modulus, E′, at 10 rad/ s and were extracted from isothermal frequency sweeps.

at temperatures above the glass transition temperature of a linear polystyrene-block analogue, where the conductivity can be above 10 mS/cm (Figure 2). Consistent with the decrease in the volume fraction of the high modulus phase, the addition of ionic liquid to the PEO domain reduces E′ by a factor of 2−3. 124

dx.doi.org/10.1021/nl4034818 | Nano Lett. 2014, 14, 122−126

Nano Letters

■

The preparation of polymeric networks swollen with a high fraction of ionic liquid (“ion gels”) by ABA copolymer selfassembly23,31,32 or by direct polymerization of monomer and cross-linker33 affords some tunability of the modulus, but due to the lack of continuity of the high modulus domain, these are soft solids (E′ ≤ 104 Pa). Although ion gels can be easily processed in the liquid state and solidified in situ, increasing the modulus typically results in reduction of the ionic conductivity, illustrating the nontrivial dependence of both properties on the segmental relaxation of polymer chains. On the other hand, PEMs composed predominantly of polymer, such as the isotropically oriented lamellae from PS-b-PEO and lithium bis(trifluoromethylsulfonyl)imide (LiTFSI), achieve sufficient continuity of both the conductive and glassy domains such that the modulus and conductivity can be decoupled.14 However, both modulus and conductivity fall short of the target region (E′ ≈ 1 GPa, σ ≥ 1 mS/cm) because (i) grain boundaries and lack of domain continuity reduce ion mobility, and (ii) the glassy modulus is reduced with progressive increases in temperature and suffers a precipitous drop above the glass transition of polystyrene (Tg ≈ 100 °C). Furthermore, diblockbased electrolytes are prepared from multistep processes involving synthesis, isolation, and solution-casting, and generally exhibit rather low toughness.34−36 In this work, the important contribution to the generation and application of solid-state electrolytes is a process that marries the processability and synthetic simplicity of a liquid precursor with the independent tunability of mechanical and mass transport properties offered by a nanostructured block polymer. The result is a material that achieves record combinations of modulus and conductivity at high temperature (Figure 4). This approach is very promising for the scalable production of PEMs for critical energy applications such as solid-state lithium ion batteries and high temperature fuel cells.

Letter

ASSOCIATED CONTENT

S Supporting Information *

Detailed synthesis information, viscoelastic master curves, and SAXS, TEM, and DSC data for all samples. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Author Contributions

M.W.S. and L.D.M. contributed equally. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Science Foundation (DMR-1006370 and DMR-1206459). Parts of this work were carried out in the Characterization Facility, University of Minnesota, which receives partial support from NSF through the MRSEC program. SAXS data was obtained at the DuPontNorthwestern-Dow Collaborative Access Team (DND-CAT) located at Sector 5 of the Advanced Photon Source (APS). DND-CAT is supported by The Dow Chemical Company, E.I. DuPont de Nemours & Co., and Northwestern University. Use of the APS, an Office of Science User Facility operated for the U.S. Department of Energy (DOE) Office of Science by Argonne National Laboratory, was supported by the U.S. DOE under contract no. DE-AC02-06CH11357. The authors thank Prof. Philippe Bü hlmann for access to his impedance spectroscopy equipment and Prof. Dean Waldow (Pacific Lutheran University) and Nicholas Smith for assistance. The authors are declared to be inventors on a provisional patent filed by the University of Minnesota related to the work presented here.

■

REFERENCES

(1) Hallinan, D. T.; Balsara, N. P. Annu. Rev. Mater. Res. 2013, 43, 503−525. (2) Steele, B. C.; Heinzel, A. Nature 2001, 414, 345−352. (3) Tarascon, J. M.; Armand, M. Nature 2001, 414, 359−367. (4) Young, W.-S.; Kuan, W.-F.; Epps, T. H. J. Polym. Sci., Part B: Polym. Phys. 2014, 42, 1−16. (5) Goodenough, J. B.; Kim, Y. Chem. Mater. 2010, 22, 587−603. (6) Goodenough, J. B.; Park, K.-S. J. Am. Chem. Soc. 2013, 135, 1167−1176. (7) Noda, A.; Susan, M. A. B. H.; Kudo, K.; Mitsushima, S.; Hayamizu, K.; Watanabe, M. J. Phys. Chem. B 2003, 107, 4024−4033. (8) Nakamoto, H.; Watanabe, M. Chem. Commun. 2007, 2539−2541. (9) Hoarfrost, M. L.; Tyagi, M. S.; Segalman, R. A.; Reimer, J. A. Macromolecules 2012, 45, 3112−3120. (10) Monroe, C.; Newman, J. J. Electrochem. Soc. 2003, 150, A1377. (11) Monroe, C.; Newman, J. J. Electrochem. Soc. 2004, 151, A880. (12) Monroe, C.; Newman, J. J. Electrochem. Soc. 2005, 152, A396. (13) Hallinan, D. T.; Mullin, S. A.; Stone, G. M.; Balsara, N. P. J. Electrochem. Soc. 2013, 160, A464−A470. (14) Singh, M.; Odusanya, O.; Wilmes, G. M.; Eitouni, H. B.; Gomez, E. D.; Patel, A. J.; Chen, V. L.; Park, M. J.; Fragouli, P.; Iatrou, H.; Hadjichristidis, N.; Cookson, D.; Balsara, N. P. Macromolecules 2007, 40, 4578−4585. (15) Choi, I.; Ahn, H.; Park, M. J. Macromolecules 2011, 44, 7327− 7334. (16) Choi, J.-H.; Ye, Y.; Elabd, Y. A.; Winey, K. I. Macromolecules 2013, 46, 5290−5300.

Figure 4. Trade-off relationship between elastic modulus and ionic conductivity in a solid electrolyte illustrated using data from refs 14 (◇), 23 (○), and 33 (□). Data represented by (●) are from this study. All data are reported at 90 °C, unless otherwise noted. Only chemically cross-linked electrolytes remain solid at elevated temperature (>90 °C), where the conductivity is maximized. 125

dx.doi.org/10.1021/nl4034818 | Nano Lett. 2014, 14, 122−126

Nano Letters

Letter

(17) Virgili, J. M.; Hoarfrost, M. L.; Segalman, R. A. Macromolecules 2010, 43, 5417−5423. (18) Young, W.-S.; Epps, T. H. Macromolecules 2012, 45, 4689− 4697. (19) Weber, R. L.; Ye, Y.; Schmitt, A. L.; Banik, S. M.; Elabd, Y. A.; Mahanthappa, M. K. Macromolecules 2011, 44, 5727−5735. (20) Scalfani, V. F.; Wiesenauer, E. F.; Ekblad, J. R.; Edwards, J. P.; Gin, D. L.; Bailey, T. S. Macromolecules 2012, 45, 4262−4276. (21) Majewski, P. W.; Gopinadhan, M.; Jang, W.-S.; Lutkenhaus, J. L.; Osuji, C. O. J. Am. Chem. Soc. 2010, 132, 17516−17522. (22) Seo, M.; Hillmyer, M. A. Science 2012, 336, 1422−1425. (23) Zhang, S.; Lee, K. H.; Frisbie, C. D.; Lodge, T. P. Macromolecules 2011, 44, 940−949. (24) Simone, P. M.; Lodge, T. P. ACS Appl. Mater. Interfaces 2009, 1, 2812−2820. (25) Broadband Dielectric Spectroscopy; Kremer, F., Schönhals, A., Eds.; Springer: Berlin, 2003. (26) Hamersky, M. W.; Hillmyer, M. A.; Tirrell, M.; Bates, F. S.; Lodge, T. P.; von Meerwall, E. D. Macromolecules 1998, 31, 5363− 5370. (27) Milhaupt, J. M.; Lodge, T. P. J. Polym. Sci., Part B: Polym. Phys. 2001, 39, 843−859. (28) Cussler, E. L. Diffusion: Mass Transfer in Fluid Systems, 3rd ed.; Cambridge University Press: Cambridge, UK, 2009. (29) Phillip, W. A.; Amendt, M.; O’Neill, B.; Chen, L.; Hillmyer, M. A.; Cussler, E. L. ACS Appl. Mater. Interfaces 2009, 1, 472−480. (30) Chen, L.; Phillip, W. A.; Cussler, E. L.; Hillmyer, M. A. J. Am. Chem. Soc. 2007, 129, 13786−13787. (31) Zhang, S.; Lee, K. H.; Sun, J.; Frisbie, C. D.; Lodge, T. P. Macromolecules 2011, 44, 8981−8989. (32) Gu, Y.; Zhang, S.; Martinetti, L.; Lee, K. H.; McIntosh, L. D.; Frisbie, C. D.; Lodge, T. P. J. Am. Chem. Soc. 2013, 135, 9652−9655. (33) Susan, M. A. B. H.; Kaneko, T.; Noda, A.; Watanabe, M. J. Am. Chem. Soc. 2005, 127, 4976−4983. (34) Bates, F. S.; Hillmyer, M. A.; Lodge, T. P.; Bates, C. M.; Delaney, K. T.; Fredrickson, G. H. Science 2012, 336, 434−440. (35) Hermel, T. J.; Hahn, S. F.; Chaffin, K. A.; Gerberich, W. W.; Bates, F. S. Macromolecules 2003, 36, 2190−2193. (36) Hermel, T. J.; Wu, L.; Hahn, S. F.; Lodge, T. P.; Bates, F. S. Macromolecules 2002, 35, 4685−4689.

126

dx.doi.org/10.1021/nl4034818 | Nano Lett. 2014, 14, 122−126