(30i) Relationship between Ionic Conductivity and Polymer Properties in Electrolytes with Neutral Polymer Hosts: A Combined Simulation/Experimental Investigation. | AIChE

(30i) Relationship between Ionic Conductivity and Polymer Properties in Electrolytes with Neutral Polymer Hosts: A Combined Simulation/Experimental Investigation.


Lynd, N. - Presenter, University of Texas at Austin
Ganesan, V., The University of Texas at Austin
Aoshima, S., Osaka University
Freeman, B. D., University of Texas at Austin
Maruyama, K., Osaka University
Imbrogno, J., University of Texas at Austin
Zhu, C., University of Texas at Austin
Baltzegar, J. R., University of South Carolina
Zhang, Z., The University of Texas at Austin
Meyer, P. W., National Renewable Energy Laboratory (NREL)
Commercial lithium-ion battery electrolytes typically contain solvent blends with two components that impart both high-polarity/high-viscosity and low-viscosity/low-polarity to the electrolyte solvent mixture. Lithium salt (e.g., LiPF6) is added to provide ionic conductivity. The composition of the solvent blend is optimized with respect to ionic conductivity by balancing ion solubility and mobility. Polymer-based electrolytes are expected to be subject to a similar balance of polarity versus dynamics. Understanding structure-property relationships in polymer electrolytes could potentially advance the design of new electrolytes and expand available battery chemistries, reduce device complexity, and increase longevity. We report the results of a combined simulation/experimental investigation of the relationship between polarity, segmental dynamics, and ionic conductivity. Molecular dynamics simulations predicted that polymer electrolytes with neutral polymer hosts exhibited two regimes of behavior: At the low-dielectric (low polarity) regime, ion dissociation limited ionic conductivity whereas in the high-dielectric regime polymer segmental-dynamics limited ion motion. The crossover of these regimes represents an optimum balance of rapid segmental dynamics and sufficient polarity for ion dissociation. We experimentally investigated both the low- and high-dielectric regimes through homologous polymer synthesis and characterization. At low-dielectric constant, a series of poly(vinyl ether)s were synthesized using living cationic polymerization, which provided materials with dielectric constants (ε) of 1.3 to 9.0 and glass-transition-temperatures (Tg) of –3.9 to –74.5 ºC. Ionic conductivities ranged from 10–10 to 10–3 S/cm with lithium bis(trifluoromethylsulfonyl)imide salt (LiTFSI) and the dielectric constant dependence of ionic conductivity was described well by a simple model for electrostatically induced ion pairing effects in continuum media based on equilibrium association considerations and an ideal solution model. In the high-dielectric regime, we synthesized a series of polyethers with polar functional groups such as allyl, methoxy, epoxy, sulfite, and nitrile substituents. Dielectric constants ranged from 7–35 at 60 ºC and the ionic conductivity varied from 10–7 to 10–3 S/cm with LiTFSI salt. As predicted by molecular dynamics, in this higher dielectric regime, the glass transition temperature increased with dielectric constant while ionic conductivity decreased. Stronger polymer-polymer and polymer-ion interactions slowed segmental dynamics, resulting in decreased ionic conductivity and an associated increase in neat polymer glass transition temperature. The disparate chemical structures of the polymers of our study, combined with the results of coarse-grained molecular dynamics simulations, support the generality of our conclusions and highlight to the difficulty of the design of high-conductivity polymer electrolytes. Widely-used poly(ethylene oxide) represents a near-optimal balance between the low- and high-dielectric constant regimes. To improve upon the ionic conductivity of polymer electrolytes, single-component neutral polymer hosts are unlikely to resolve the trade-off between the need for ion dissociation while retaining rapid segmental dynamics.

This research was supported by the National Science Foundation (CBET-1706968), the Robert A. Welch Foundation under Grant Nos. F-1599 and F-1904, the International Joint Research Promotion Program of Osaka University, and the Center for Materials for Water and Energy Systems (M-WET), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award # DE-SC0019272.