Researchers led by the Department of Energy’s Pacific Northwest National Laboratory (PNNL) have extended the capacity and duration of sodium-aluminum batteries. The new battery design could help ease integration of renewable energy into the electrical grid at lower cost, using Earth-abundant metals, according to a study just published in Energy Storage Materials.

The new sodium-based molten salt battery uses two distinct reactions. The team previously reported a neutral molten salt reaction. The new discovery shows that this neutral molten salt can undergo a further reaction into an acidic molten salt. Crucially, this second acidic reaction mechanism increases the battery’s capacity. Specifically, after 345 charge/discharge cycles at high current, this acidic reaction mechanism retained 82.8% of peak charge capacity.

Sodium and aluminum are a natural combination of inexpensive, abundant elements as a redox pair for battery energy storage. Recent explorations pairing a sodium anode and aluminum cathode have demonstrated reversible, energy dense Na-Al cells with excellent rate capability using the electrochemical reaction between a molten Na anode and a NaAlCl₄/Al cathode. In this work, the NaAlCl₄/Al cathode is extended beyond the neutral NaAlCl₄ composition by unlocking the NaCl-AlCl₃ phase diagram to explore the extra accessible capacity hidden in acidic chloroaluminate melts up to and beyond the composition NaAl₂Cl₇.

This enables higher specific capacity and average discharge voltages than previous Na-Al batteries, utilizing two distinct cell reaction mechanisms in one battery. Fundamental aspects of the NaAlCl₄-NaAl₂Cl₇ reaction chemistry are investigated, and Na-metal/chloroaluminate batteries with excellent reversibility and areal capacity are demonstrated.

Increasing the voltage window of the chloroaluminate Na-Al battery takes advantage of the higher voltage (∼ 2 V vs ∼1.6 V for neutral NaAlCl₄) contributed by the acidic chloroaluminate cathode reaction, unlocking an additional specific energy of ∼119 Wh kg⁻¹ by utilizing the conversion of NaAlCl₄ to NaAl₂Cl₇, which adds to the neutral melt reaction between NaAlCl₄/Al and Na (∼493 Wh kg⁻¹ theoretical). By significantly increasing the cathode thickness and therefore accessible areal capacity up to 131.7 mAh cm⁻², a discharge duration of 28.2 h is achieved with an estimated raw active materials cost of $7.02 kWh⁻¹.

These metrics show the great potential of this unlocked chloroaluminate battery for future low-cost, long-duration electrochemical energy storage.

—Weller et al.

Weller et al.

Although the battery is in early-stage testing, the researchers speculate that it could result in a practical energy density of up to 100 Wh/kg. In comparison, the energy density for lithium-ion batteries used in commercial electronics and electric vehicles is around 170–250 Wh/kg. However, the new sodium-aluminum battery design has the advantage of being inexpensive and easy to produce in the United States from much more abundant materials.

The new sodium-aluminum battery design allows only sodium (depicted as yellow balls) to move through the solid-state electrolyte to charge the battery. Being constructed of inexpensive Earth-abundant materials such as sodium salts and aluminum wool, a scrap product of aluminum manufacturing, is an advantage. (Diagram by Sara Levine | Pacific Northwest National Laboratory)

PNNL scientists collaborated with colleagues at the US-based renewable energy pioneer Nexceris to assemble and test the battery. Nexceris, through their new business Adena Power, supplied their patented solid-state, sodium-based electrolyte to PNNL to test the battery’s performance.

Our primary goal for this technology is to enable low-cost, daily shifting of solar energy into the electrical grid over a 10- to 24-hour period. This is a sweet spot where we can start to think about integrating higher levels of renewables into the electrical grid to provide true grid resiliency from renewable resources such as wind and solar power.

—Vince Sprenkle, co-author

Sprenkle was part of the team that developed this battery’s new flexible design, which also shifted the battery from a traditional tubular shape to a flat, scalable one that can more easily be stacked and expanded as the technology develops from coin-sized batteries to a larger grid-scale demonstration size. More importantly, this flat cell design allows the cell capacity to be increased by simply using a thicker cathode, which the researchers leveraged in this work to demonstrate a triple capacity cell with sustained discharge of 28.2-hours under laboratory conditions.

Compared with a seasonal battery, this new design is especially adept at short- to medium-term grid energy storage over 12 to 24 hours. It is a variation of a sodium-metal halide battery. A similar design employing a nickel cathode as part of the system has been shown effective at commercial scale and is already commercially available.

Because it operates at a lower temperature, it can be manufactured with inexpensive battery materials, instead of requiring more complex and expensive components and processes as in conventional high-temperature sodium batteries, said David Reed, a PNNL battery expert and study co-author.

The research was supported by the DOE Office of Electricity and the International Collaborative Energy Technology R&D Program of the Korea Institute of Energy Technology Evaluation and Planning. The electrolyte development was supported by a DOE Small Business Innovation Research program. The nuclear magnetic resonance measurements were made in EMSL, Environmental Molecular Sciences Laboratory, a DOE Office of Science User Facility sponsored by the Biological and Environmental Research program.

Resources

• J. Mark Weller, Minyuan M. Li, Evgueni Polikarpov, Kee Sung Han, Neil Kidner, Anant Patel, Mai Nguyen, Meghan Stout, Michael Gossett, Keeyoung Jung, David M. Reed, Vincent L. Sprenkle, Guosheng Li (2023) “Unlocking the NaCl-AlCl₃ phase diagram for low-cost, long-duration Na-Al batteries,” Energy Storage Materials, Volume 56, Pages 108-120 doi: 10.1016/j.ensm.2023.01.009