Toward a quantitative interfacial description of solvation for Li metal battery operation under extreme conditions

John Holoubek, Kunpeng Yu, Junlin Wu, Shen Wang, Mingqian Li, Hongpeng Gao, Zeyu Hui, Gayea Hyun, Yijie Yin, Anthony U. Mu, Kangwoon Kim, Alex X. Liu, Sicen Yu, Tod A. Pascal, Ping Liu, Zheng ChenProceedings of the National Academy of Sciences 2023 · DOI

Abstract

The future application of Li metal batteries (LMBs) at scale demands electrolytes that endow improved performance under fast-charging and low-temperature operating conditions. Recent works indicate that desolvation kinetics of Li<sup>+</sup> plays a crucial role in enabling such behavior. However, the modulation of this process has typically been achieved through inducing qualitative degrees of ion pairing into the system. In this work, we find that a more quantitative control of the ion pairing is crucial to minimizing the desolvation penalty at the electrified interface and thus the reversibility of the Li metal anode under kinetic strain. This effect is demonstrated in localized electrolytes based on strongly and weakly bound ether solvents that allow for the deconvolution of solvation chemistry and structure. Unexpectedly, we find that maximum degrees of ion pairing are suboptimal for ultralow temperature and high-rate operation and that reversibility is substantially improved via slight local dilution away from the saturation point. Further, we find that at the optimum degree of ion pairing for each system, weakly bound solvents still produce superior behavior. The impact of these structure and chemistry effects on charge transfer are then explicitly resolved via experimental and computational analyses. Lastly, we demonstrate that the locally optimized diethyl ether-based localized-high-concentration electrolytes supports kinetic strained operating conditions, including cycling down to -60 °C and 20-min fast charging in LMB full cells. This work demonstrates that explicit, quantitative optimization of the Li<sup>+</sup> solvation state is necessary for developing LMB electrolytes capable of low-temperature and high-rate operation.

Introduction: The Desolvation Bottleneck

Lithium-metal batteries (LMBs) are often considered the "holy grail" of energy storage due to their high theoretical capacity. However, their practical application is hindered by poor performance under "kinetic strain"—extreme cold or rapid charging. The primary bottleneck is the desolvation penalty: the energy required for a $Li^+$ ion to shed its "shell" of solvent molecules before it can move through the interface and into the electrode.

Conceptual overview of how solvation structure and chemistry influence the interfacial energy barrier and resulting lithium morphology.

While researchers have long recognized that "ion pairing" (where anions assist $Li^+$ in shedding solvents) is beneficial, the approach has remained largely qualitative. This study, published in PNAS by researchers at UC San Diego, provides a quantitative, interfacial description of how specific solvation microstates dictate battery kinetics.

Methodology: Tuning Solvation Microstates

The team utilized Localized High-Concentration Electrolytes (LHCEs), which use a non-solvating "diluent" (BTFE) to maintain the benefits of high salt concentration while keeping viscosity low. They compared two ether solvents: DME (strongly binding to $Li^+$) and DEE (weakly binding).

Key novel approaches included:

  • Local vs. Global Dilution: Distinguishing between adding more diluent (global) and adding just enough solvating solvent to change the $Li^+$ microstates (local).
  • Distribution of Relaxation Times (DRT): A sophisticated analysis of Electrochemical Impedance Spectroscopy (EIS) data used to decouple the resistance of the Solid-Electrolyte Interphase (SEI) from the desolvation-related charge-transfer resistance.
  • Metadynamics (MTD) Simulations: Accelerated molecular dynamics were used to map the free-energy landscape of $Li^+$ as it approaches a polarized interface, allowing them to "see" the energy barriers of different solvation states.

Key Insights: The "Goldilocks" Zone of Ion Pairing

A central finding of the paper is that more ion pairing is not always better. The authors discovered that maximum ion pairing (at the saturation point) is actually suboptimal. Instead, a slight "local dilution" creates a distribution of microstates that are easier to desolvate at the interface.

Furthermore, the study emphasizes that chemistry still rules. Even when both systems are optimized for structure, the weakly-binding solvent (DEE) outperformed the strongly-binding one (DME). The inherent chemistry of the solvent sets the baseline for the desolvation penalty, which is then further refined by the solvation structure. By using a pre-cycling and reassembly method, the team proved that performance gains were due to improved desolvation kinetics of the liquid electrolyte itself, rather than just a "better" SEI layer.

Results: Record-Breaking Performance

The optimized Localized DEE (LDEE) electrolyte achieved several performance milestones:

  • Ultra-Low Temperature Operation: A Coulombic efficiency of 98.9% at -60°C, significantly higher than the saturated version (96.6%).
  • Fast Charging: Full cells (NCA||Li) supported 20-minute fast charging at room temperature with minimal capacity fade over 100 cycles.
  • Pouch Cell Validation: A single-layer pouch cell demonstrated exceptional reversibility, retaining 93% capacity after 100 cycles that included segments at -20°C, -40°C, and -60°C.
Coulombic efficiency as a function of solvating solvent content at -60°C, demonstrating the optimal performance peak achieved through local dilution.

Conclusion and Future Directions

This work demonstrates that explicit, quantitative optimization of the $Li^+$ solvation state is necessary for developing LMB electrolytes capable of low-temperature and high-rate operation. While these results are promising, challenges remain in scaling these optimized microstates to high-pressure, multi-layer industrial cell formats and ensuring oxidative stability at very high voltages (>4.5V). Nevertheless, the authors project that this approach could support cell-level energy densities of >300 Wh/kg even under extreme operating conditions.

Summary updated

Abstract

The future application of Li metal batteries (LMBs) at scale demands electrolytes that endow improved performance under fast-charging and low-temperature operating conditions. Recent works indicate that desolvation kinetics of Li<sup>+</sup> plays a crucial role in enabling such behavior. However, the modulation of this process has typically been achieved through inducing qualitative degrees of ion pairing into the system. In this work, we find that a more quantitative control of the ion pairing is crucial to minimizing the desolvation penalty at the electrified interface and thus the reversibility of the Li metal anode under kinetic strain. This effect is demonstrated in localized electrolytes based on strongly and weakly bound ether solvents that allow for the deconvolution of solvation chemistry and structure. Unexpectedly, we find that maximum degrees of ion pairing are suboptimal for ultralow temperature and high-rate operation and that reversibility is substantially improved via slight local dilution away from the saturation point. Further, we find that at the optimum degree of ion pairing for each system, weakly bound solvents still produce superior behavior. The impact of these structure and chemistry effects on charge transfer are then explicitly resolved via experimental and computational analyses. Lastly, we demonstrate that the locally optimized diethyl ether-based localized-high-concentration electrolytes supports kinetic strained operating conditions, including cycling down to -60 °C and 20-min fast charging in LMB full cells. This work demonstrates that explicit, quantitative optimization of the Li<sup>+</sup> solvation state is necessary for developing LMB electrolytes capable of low-temperature and high-rate operation.
BibTeX
@inproceedings{holoubek2023toward,
  author = {John Holoubek and Kunpeng Yu and Junlin Wu and Shen Wang and Mingqian Li and Hongpeng Gao and Zeyu Hui and Gayea Hyun and Yijie Yin and Anthony U. Mu and Kangwoon Kim and Alex X. Liu and Sicen Yu and Tod A. Pascal and Ping Liu and Zheng Chen},
  title = {{Toward a quantitative interfacial description of solvation for Li metal battery operation under extreme conditions}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2023},
  doi = {10.1073/pnas.2310714120},
}
Figure