Large-scale energy storage is poised to play a critical role in enhancing the stability, security, and reliability of tomorrow’s electrical power grid, including the support of intermittent renewable resources. Batteries are appealing because of their small footprint and flexible siting; however, conventional battery technologies are unable to meet the demanding low-cost and long-lifespan requirements of this application.

—Bradwell et al.

Earlier work with high-temperature liquid batteries demonstrated impressive current density capabilities, but generally used prohibitively expensive metalloids such as bismuth (Bi) and tellurium (Te) as the cathode, the authors note. Antimony is less costly and more earth-abundant than Bi and Te, but until now had not been demonstrated in a liquid metal battery.

The team selected magnesium as the anode material on the basis of its low cost, high earth abundance, low electronegativity, and overlapping liquid range with both Sb and candidate electrolytes. The electrolyte was selected on the basis of its sufficiently low melting point (396 °C) and the greater electrochemical stability of NaCl and KCl in comparison with MgCl₂.

Because of density differences and immiscibility, the salt and metal phases stratify into three distinct layers:

Schematic of a Mg||Sb liquid metal battery comprising three liquid layers that operates at 700 °C. During charging, Mg is electrochemically extracted from the Mg−Sb alloy electrode and deposited as liquid Mg on the top (negative) electrode. During discharging, the Mg electrode is consumed, and Mg is deposited into the Mg−Sb liquid bottom (positive) electrode. During charging, the battery consumes energy; upon discharge, the battery supplies energy. Credit: ACS, Bradwell et al. Click to enlarge.

The team assembled Mg||Sb single cell batteries and electrochemically characterized them by cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) using a two-electrode electrochemical setup with the negative electrode (Mg) as the counter electrode/reference electrode and the positive electrode (Sb) as the working electrode. Among their findings were:

• Cells cycled at 50 mA/cm² for a predefined discharge period of 10 h to a cutoff charging voltage limit of 0.85 V achieved a round-trip Coulombic efficiency of 97% and a voltage efficiency of 71%, resulting in an overall energy efficiency of 69%.

• Cells were fully discharged at various rates ranging from 50 to 200 mA/cm² with 0.05 V as the discharge cutoff limit. Operation at higher current density resulted in increased IR voltage loss and decreased capacity.

• Cells were cycled more than 30 times for periods of up to 2 weeks and did not exhibit obvious signs of corrosion of the solid-state cell components (current collectors and walls), as determined through optical imaging and scanning electron microscopy (SEM)/energy-dispersive spectroscopy (EDS) analysis.

• After several weeks of cycling, the cells ceased to operate. The observed cause of failure was evaporation of the molten salt electrolyte into the surrounding containment vessel, a mechanism that could be mitigated by alternative cell designs with reduced head space.

In summary, an all-liquid battery with Mg and Sb liquid metal electrodes has been proposed and its performance capability demonstrated. The use of Sb as the positive electrode and the self-segregating nature of the liquid components may enable a low-cost energy storage solution. Cells were cycled under constant-current conditions, demonstrating high current density capabilities and negligible corrosion of the solid-state cell components over the testing period.

Further work is required for evaluation of the long-term performance of the proposed cells, which may require an alternative cell design. At some larger scale, the action of electric current flowing through the electrolyte could generate enough Joule heat to keep the components molten, thereby obviating the need for external heaters, as is the case with electrolytic cells producing aluminum on a commercial scale.

Future work will include long-term corrosion testing of solid-state components, current collector optimization, and investigation of alternative sheath materials. While the initial cell performance results are promising, exploration of other metal−metalloid couples with still greater cell voltages and lower operating temperatures is warranted. If a low-cost, high-voltage system with sufficiently low levels of corrosion were discovered, it would find utility in a wide array of stationary storage applications.

—Bradwell et al.

Resources

• David J. Bradwell, Hojong Kim, Aislinn H. C. Sirk, and Donald R. Sadoway (2012) Magnesium–Antimony Liquid Metal Battery for Stationary Energy Storage. Journal of the American Chemical Society doi: 10.1021/ja209759s