By Deborah Borfitz
January 14, 2019 | Scientists from Honda Research Institute, together with researchers at the California Institute of Technology (Caltech) and NASA’s Jet Propulsion Laboratory (JPL), recently achieved a milestone in fluoride-ion battery (FIB) technology—the ability to run energy cells at room temperature rather than heating them to at least 150 degrees Celsius. The temperature limitation is “inconvenient, if not impossible, in many consumer-relevant applications,” according to Simon Jones, principal investigator and JPL technical staff member. Certainly no one wants to carry around a smart phone that’s frying-pan hot.
Details of the breakthrough were published in the Dec. 7, 2018 issue of Science.
“Fluoride-ion batteries offer high energy density but research to date has shown their operation only at high temperatures because the electrolyte is a solid material,” Jones explains. But making fluoride-conducting liquid electrolytes that more readily conduct ions isn’t exactly a straightforward proposition. Fluorides don’t dissolve in much and, if they do, “react rather than stick around to carry charge,” he says. Regardless, the solvents tried to date “can’t support a high concentration of fluoride ions or a useful operating voltage for high energies.” There has additionally been a problem of electrode materials dissolving in the electrolyte, expediting battery failure.
“All of these issues have been addressed with a liquid electrolyte made useful by its salt design [dry tetraalkylammonium fluoride salts], choice of solvent [organic fluorinated ether], and by putting a protective shell [nanostructure of lanthanum fluoride] around the copper electrode material that lets fluoride in but keeps the solvent away,” Jones says.
The materials currently being tested are cerium, lanthanum, and copper-based, Jones adds, and those elements are more plentiful than cobalt and lithium more popularly used in batteries today. The research team is still trying to get naturally-occurring calcium fluoride to work as an anode, which would additionally give their FIB a “significantly lower eco-footprint.”
Liquid electrolyte quest
Until now, FIBs are made in a laboratory cell that operates at high temperatures. This is because “fluoride ions are only known to be usefully conductive [i.e., move toward a polarized electrode] when a solid-state electrolyte is heated to high temperatures,” says Victoria Davis, the study’s lead author who is now a graduate student at the University of North Carolina.
“The big challenge with making a room-temperature fluoride-ion battery is that fluoride ions aren’t stable in common liquid solutions that are known,” says Davis. “Fluoride ions either pair so strongly to their counter ions in salt [thus, don’t dissolve in solution], or if they do dissolve, they easily react with any acidic hydrogen available in solution to form bifluoride ions.” That’s undesirable because bifluoride electrolytes have small “voltage windows,” meaning low useful energy storage capacity. Bifluoride ions are also not natively good at promoting conversion reactions at an electrode surface.
“To avoid bifluoride ions, we need our salts to be dry and compatible with the solution we want to dissolve the salt into,” Davis explains. “‘Like dissolves like,’ so an organic salt should be made to dissolve into an organic solution. Our research succeeded in dissolving fluoride salt into an organic solution without forming bifluoride ions. We succeeded in making an electrolyte with stable fluoride ions in a liquid solution at room temperature and made a room-temperature laboratory cell.”
Many questions remain unanswered
Market expectations tend to move faster than the frustratingly slow pace of battery progress, as gauged by dialogue among battery enthusiasts on social media site Reddit. Jones is quick to stress that the work of his research team is by nature highly conceptual. “At a materials level—considering the metal/metal fluoride reaction alone and no additional components—materials exist that offer very high energy densities compared to their lithium ion (Li-ion) equivalents [e.g., graphite and metal oxides], assuming 100% reaction. This is our basis for comparison when nothing else is really known about the other components, how they fit together in a real cell, and how this compares to a typical Li-ion cell design.”
The comparison is also “volumetric,” in terms of energy or capacity per unit volume rather than “gravimetric” and per unit mass, he adds. “In a gravimetric comparison, the fluoride ion batteries are not as attractive as the materials can be heavier than Li-ion equivalents [e.g., cerium] although they win on a charge-per-volume basis.”
The published study indicates that batteries created with the new chemistry “may have more than five times the energy density of lithium-ion batteries,” but that’s not a promise and comparative performance is definitely a “moving target,” says Jones. Even as development hurdles with FIBs are overcome, Li-ion batteries still have headroom for improvement. “My experience is that the Li-ion battery engineers are extremely skilled at cramming more material into tighter and tighter spaces, so it would not surprise me if Li-ion systems continue to increase [their capacity] by a few percent a year for the next few years at least.”
It’s impossible to speculate on the safety aspects of a full cell design that doesn’t yet exist, Jones says. “The final reckoning will involve a combination of dryness of the electrolyte [risk of hydrofluoric acid formation is related to water content], and the reactivity of the electrode materials with themselves and the electrolyte solvent upon short-circuit/puncture [which would lead to thermal runaway].” The latter issue could possibly be mitigated by the core-shell nature of the electrode material.
‘Long way from ubiquity’
Academic research groups are intrigued by the fluoride battery breakthrough, says Jones, and collaborative studies have already begun around the novel electrolyte. It helps that the patent situation is often “less opaque” in academic than in industrial labs, which can slow progress, he adds. “With academic patent licenses, which are often non-exclusive and tied to key commercialization milestones, the opportunity to hold back an important technology is much less prevalent.” His fluoride ion research is covered by patents and applications owned in whole or part by Caltech, which manages JPL and encourages the transfer of the technology to the public “to get performance where it needs to be for real-world applications.”
“There are many electrochemical researchers out there with an abundance of good ideas, and we want to make them aware of another part of the electrochemical toolbox they can work with,” says Jones. “Given the intensive worldwide studies on lithium-sulfur, lithium-oxygen, magnesium and other ‘next-generation’ battery chemistries, here’s something quite different and potentially high-energy that should be worth their research efforts.”
While a “long way from ubiquity,” Jones continues, it’s reasonable to expect current challenges with FIBs to be resolved and the technology critically evaluated in a “closer-to-commercial format” within the next few years. He bases his prediction in part on the development of Li-ion batteries: “Whittingham described the first prototype in 1975 [using lithium metal] with a liquid electrolyte, Goodenough demonstrated the lithium-cobalt oxide cathode in 1980, Basu and Yazami the graphite anode in 1977-1982, and Yoshino the first full cell in 1985 leading to Sony releasing the first commercial battery in 1991. So it’s a slow process, and we are currently at the electrolyte/cathode/anode development stages for fluoride ion batteries [e.g., around 1980 in the Li-ion timeline]. However, many more people are working on battery technology worldwide compared to the early days of Li-ion.”
Among the remaining challenges, according to Jones:
• “Achieving performance at the high energies promised by the materials.” In the published study, the copper material was cycled only to around 15% of its theoretical value. More recent work being finalized for publication shows “considerable improvement,” Jones notes. “Watch this space.”
• “Achieving the cycle life necessary for a particular application—e.g., hundreds of cycles for consumer electronics and thousands of cycles for automotive and stationary energy storage.”
• “Figuring out the engineering to get a full cell working.” This is “becoming possible as the fundamental materials work gets done,” he says.
No battery is likely to have enough energy to support space travel, Jones says, but an early-stage fluoride battery may be well suited to mission scenarios that need high energy density but only a limited number of cycles. One possibility: a Rover mission involving “a small number of trips where the volume available for energy storage is extremely tight.”
Most major automotive and electronic/industrial companies invest in at least some next-generation battery research, Jones says. “They’re trying to figure out the limits of their technology five to 10 years from now and plan their roadmap accordingly.” Since their timelines are fairly long, fundamental researchers like Jones will have at least five years to put a working prototype through “exhaustive testing” before the market starts clamoring for it.
