New Ways To Improve Batteries

Researchers around the world are racing to develop more efficient, denser, and safer battery technology, and they are reaching far beyond where research has gone before.

Much of this is being driven by concern over exhaust from internal combustion engines, which are responsible for a significant portion of global CO₂ emissions. Nearly all carmakers today have announced plans to develop battery electric vehicles, and some are already selling them. But to be competitive with gasoline-powered cars and trucks, they will need significant improvements in a number of areas.

Fig. 1: Estimated U.S. energy consumption in 2020 (92.9 Quads) and where it came from. Source: Lawrence Livermore National Laboratory

Among the challenges, batteries will need to deliver increased energy density with less expensive and more environmentally friendly materials. Today, only about half of the lithium ions contained in the cathode in current battery designs are available for transfer to the anode. In cathode materials like LiCoO₂, overcharging can lead to mechanical collapse.

As a general rule, battery designers seek to eliminate or minimize all components other than the ion source itself. They have several options to increase energy density. They can reduce the weight of other battery components, such as electrolytes, separators, and casings. They can use materials in which more of the total lithium is available, such as Li₂S or lithium metal. Or they can use alternatives to lithium, such as sodium, magnesium or zinc.

Lithium metal electrodes are an obvious solution to the problem of lithium availability. They offer very high capacity — up to 500 Wh/kg — and do not need carbon to improve ionic conduction. Indeed, the first lithium-ion batteries used lithium metal electrodes. Unfortunately, this design was soon abandoned because lithium dendrite growth in the presence of liquid electrolytes caused short circuits. More recently, though, battery manufacturers have been investigating solid state electrolytes for safety reasons, as they do not contain flammable solvents. Because solid state electrolytes are heavier, designers would like to use lithium metal electrodes to offset the weight penalty they impose. The battery industry is therefore revisiting the problem of lithium dendrite growth.

Liquid electrolytes are compliant, flowing around electrode particles as the battery flexes while charging and discharging. Solid electrolytes are not. It’s reasonable to hope that their more rigid structure can impede dendrite formation. Accordingly, researcher Jung Hwi Cho and colleagues at Brown University examined cracking at the interface between a garnet-based lithium lanthanum zirconium oxide (LLZO) electrolyte and a lithium metal electrode. In research presented at December’s Materials Research Society fall meeting, they found short circuit failures that apparently were due to grain boundary lithium plating in the electrolyte. Surface defects from the electrolyte manufacturing process allowed lithium metal to intrude, creating stress that allowed expansion of the crack and further lithium intrusion.

One possible solution, proposed by researchers Shizhao Xiong and Aleksandar Matic at Chalmers University of Technology, uses a quasi-solid electrolyte paste between the metal electrodes and the bulk electrolyte. The paste, composed of lithium aluminum germanium phosphate (LAGP, under the trade name NASICON) powder mixed with an ionic liquid, stabilizes the interface chemistry and can provide some degree of flexibility to accommodate mechanical strain.

Another approach, suggested by researcher Qisheng Wu and colleagues at Brown University and the University of Maryland, used an electrolyte based on cellulose fiber rather than the polyethylene oxide (PEO) typically used. In PEO, ions are transported by movement of the polymer chains. The fiber-based electrolyte appears to allow ion transport by hopping between sites, instead. This hopping transport facilitates improved ion conduction.

In work reported by Wu Xu, researchers at Pacific Northwest National Laboratory observed that most solid-state electrolytes are produced by dissolving the electrolyte in a suitable solvent and allowing it to permeate the electrode structure, then evaporating the solvent. Complete evaporation of solvents from porous structures is difficult, though. They suggest that residual solvent may be responsible for the inconsistent ionic conduction measurements seen in experimental results. As an alternative, they used a dry ball-milling process to produce a PEO-lithium salt mixture. The ratio of the two determined the properties of the electrolyte, with a 1:1 ratio producing the best ionic conductivity.

Batteries beyond lithium
Other emerging battery concepts depend on materials other than lithium. Sodium, for instance, is the sixth most abundant element in the earth’s crust. However, Syracuse University researcher Francielli Gender pointed out that sodium ions are larger than lithium ions, and will not fit between the graphite layers used by electrodes in lithium-based batteries. Hard carbon compounds offer larger spacing and are usually produced by carbonization of biomass. Most established methods require chemical activation and high carbonization temperatures, but the Syracuse group produced suitable materials by washing avocado peels in water and then carbonizing them at between 900 and 1,100°C.

Avocado is one of the most commonly consumed fruits in the United States. With its thick, inedible peel and large pit, up to 30% of each avocado is wasted. Sodium-ion batteries using hard carbon electrodes derived from avocado waste are potentially a more sustainable alternative to the incumbent lithium technology.

Both lithium and sodium are monovalent. Each ion carries only one charge. Multivalent materials like zinc and magnesium have the potential to store more charge with the same number of ions. So far, however, no suitable electrolytes have been found. Diffusion of multivalent ions is slow, in part due to electrostatic repulsion. Arumugam Manthiram, a researcher at the University of Texas at Austin, found high energy barriers prevented insertion of zinc ions into non-aqueous electrolytes. In aqueous electrolytes, though, free charge-compensating protons (hydrogen nuclei) liberated from water molecules can easily diffuse into solid state electrodes, releasing equivalent numbers of zinc ions into the electrolyte.

Another multivalent metal, magnesium, is globally abundant and less reactive than alkali metals like sodium and lithium. However, reactions with conventional electrolytes passivate the surface and prevent ion diffusion. Magnesium alloys are less reactive, but also substantially reduce the energy density. Instead, Clément Pechberty and colleagues at the University of Montpellier used a gallium cap layer to protect the magnesium surface while preserving the storage capacity of bulk magnesium metal.

Conclusion
As with many conferences, results presented at the Materials Research Society meeting were preliminary proof-of-concept results. Solid-state electrolytes and metals other than lithium still need to prove they are reliable and manufacturable enough to displace incumbent technologies. Taken as a group, though, it’s clear that battery technology is leading the way to a post-petroleum transportation future.

Related
Batteries Have Moving Parts
Work is underway to reduce the risk of fires and to improve the utilization of energy stored in batteries.
Making Batteries Denser And Safer
Creating better batteries requires more than just chemistry.
Batteries Take Center Stage
No matter how efficient a system design, it only works when there is enough battery power.