The team focused on the availability of the elements for making active materials. They considered five main classes of batteries—aqueous electrolyte; Li-based (Li-ion and Li metal); high-temperature; redox-flow; and metal-air—in the context of three main questions:

• Based on the specific energy, is the couple suitable for electric vehicles or only for grid-scale batteries? Vehicle batteries should have high specific energy and energy density.

• What is the energy storage potential (in TWh) based on annual production (“flow”) and reserve base (“stock”)?

• What is the cost of the elements in the couple (in $/kWh)?

Albertus, then at Berkeley and now with Bosch Research and Technology Center, presented an overview of the findings at the recent 4^th Symposium on Energy Storage: Beyond Lithium-ion.

The researchers found that Li-based couples aer best suited for vehicles, but that only a few meet specific energy targets. Source: Paul Albertus. Click to enlarge.

Suitability. Wadia et al. started by noting that batteries for grid-scale and electric vehicle energy storage have significantly different performance requirements. While all of the 27 couples could be used in a grid-storage application, only a subset are appropriate for EVs, and those have limitations as well. System-level energy is one performance attribute to consider when choosing a battery, they noted; others include power capability, reversibility (the ability to cycle hundreds or thousands of times), operating temperature range, flexibility in size and shape, and round-trip energy efficiency.

Of the 27 couples, a few achieved the specific energy required to meet the US Department of ENergy’s goal (200 Wh/kg) for electric vehicle packs. However, the team noted, none of the couples have sufficient reversibility and other performance characteristics to be used in vehicles.

Neither the Li/CoO₂ nor the Li/S couple can safely cycle the hundreds to thousands of cycles required by an EV application, and the Zn/O₂ couple has a limited reversibility, low power capability, and a low round-trip energy efficiency (ca. 60%, compared to ca. 90% for Li-ion couples). The DOE specific energy goals for plug-in hybrid (PHEV) and hybrid (HEV) electric vehicles are significantly lower, explaining the forthcoming use of couples such as C₆/LiFePO₄ and C₆/Li_0.8Co_0.15Al_0.05O₂, and the present use of the Rare Earth Element (REE) REE-Ni₅H₅/NiOOH couple, in these applications. In general, negative- and positive-electrode materials are interchangeable in lithium-ion systems (“lithium-ion” refers to a class of couples); this means many more combinations (such as Si/_0.3LiMn₂O₃·0.7LiMn_0.5Ni_0.5O₂) are possible than we show in Fig. 1.

The metal air cells, especially the Li/O₂ cell, may have very high specific energy, although the kinetics and reversibility of the O₂ electrode are poor and much development remains. The performance of the Li/S system has improved significantly in recent years, and is an area of active research. While the theoretical specific energy of Na/NiCl₂ and Na/S are high, they both operate at elevated temperatures (several hundred °C), and are therefore not suitable for EVs. The Zn/Cl₂ system also has a high specific energy but has not been built with a cell design suitable for EVs.

From our discussion...we draw the conclusion that some couples will never meet the EV specific energy target, while those with the potential still need more development in terms of practical specific energy and other performance characteristics. Our discussion also shows that in the near term, the couples with the best potential to meet EV performance requirements are all based on Li. Thus, Li will be the most important element for the scale-up of EVs in the short and possibly long-term.

—Wadia et al.

Resource potential. They obtained the energy storage potential from the annual production and reserve base, and basic electrochemistry. Their analysis on the availability of the elements builds on the United States Geological Survey (USGS) Mineral Commodity summaries, and draws on annual production (representative of a “flow”), reserve base (representative of a “stock”), and cost numbers.

They calculated the number of TWh that can be produced based on both the annual production and reserve base, calling it the “battery energy storage potential”, based on the limiting element in each couple—i.e., the element that would run out first during production. They also calculate the $/kWh cost of the elements in each couple, taking all the elements into account.

In the short-term (10–15 years) and long-term (40–50 years) there is sufficient availability of the elements for battery deployment in grid-scale applications. For the EV application, scale-up of Li production will be needed to meet short-term goals, but will be especially necessary to meet long-term goals. Eventually, on the order of 1 billion 40kWh Li-based EV batteries can be built with the currently estimated reserve base of Li. Achieving aggressive cost reductions will continue to be a challenge for grid-scale and electric vehicle energy storage markets but the cost targets for battery storage should not be hampered by the costs of the elements in the active materials in most cases. Expansion of battery research into alternative materials may accelerate our ability to work through both the scaling and cost challenges inherent in long-term planning for battery energy storage.

—Wadia et al.

In their paper, the team noted that their analysis can also inform the search for EV battery couples based on non-Li couples.

For example, among the couples in our present analysis, those containing Cl, Cr, Fe, Mn, Na, S, and Zn are found in couples with a particularly high ESP and have a low extraction cost; new couples utilizing these elements may be of particular interest. Mg-based couples have also been discussed as having the potential to achieve a high specific energy and eventually replace Li-based couples...Mg, for example, has an annual production nearly two orders higher than Li, as well as a lower extraction cost, warranting further investigation providing active materials with only other abundant elements are used.

—Wadia et al.

Resources

• Cyrus Wadia, Paul Albertus, and Venkat Srinivasan (2011) Resource Constraints on the Battery Energy Storage Potential for Grid and Transportation Applications, Journal of Power Sources, Volume 196, Issue 3, p. 1593-1598 doi: 10.1016/j.jpowsour.2010.08.056