A piece of aluminum has more free energy content than the same amount of either methanol or ethanol. That, succinctly, explains why scientists are now trying to perfect batteries based on chemistry that involves aluminum. Even better, aluminum-ion batteries promise to squeeze more energy into a given space than the lithium-ion batteries that seem to be in the headlines these days for all the wrong reasons.
Aluminum is the most abundant metallic element and the third most abundant of all elements on the planet. Most of the countries that mine the bauxite ore from whence it comes are friendly to the U. S. And the metal’s abundance brings the promise of aluminum-based batteries that are relatively cheap.
The key quality of aluminum that makes it advantageous in battery chemistry is the fact that it is trivalent, having three valence electrons involved in forming valence bonds. Lithium, in contrast, has one valence electron. The result is that battery charge/discharge reactions (per formula unit) involving aluminum transfer three electrons compared with only one with lithium. So an aluminumbased battery can be smaller than a lithium-ion cell packing the same power: Aluminum-based batteries have a potential energy density of 1,060 W-hr/kg compared with 406 W-hr/kg for those based on lithium ion.
Problem is, there are a number of obstacles that have made it tough to devise a practical aluminum-battery chemistry. For one thing, aluminum batteries typically use an aqueous electrolyte and consume water during their discharge cycle. They also give off hydrogen. This behavior has made it impossible to come up with an aluminumbased battery that could be sealed. And aluminum used in the battery anode typically corrodes or forms an oxide film. Both factors reduce the efficiency of battery reactions over time. Unfortunately, these effects can take place so quickly that commercial batteries based on aluminum have been impractical for most applications.
But there are ways to overcome these difficulties. Researchers have pursued one tactic in particular that involves replacing the aqueous electrolyte with one that is ionic, filled with ions and ionic pairs. Ionic electrolytes don’t evaporate as quickly, and by eliminating the use of water, they also avoid the problem of aluminum batteries discharging hydrogen.
A lot of original work on ionic electrolytes and aluminum took place at the Oak Ridge National Laboratory. There, a research team headed up by M. Parans Paranthaman and Gilbert Brown came up with an ionic electrolyte comprised of 3-ethyl-1-methylimidazolium chloride containing excess aluminum trichloride. It both prevents the formation of oxides on the aluminum anode as well as eliminates the formation of hydrogen gas, they say.
One difficulty with this approach, though, is that ionic electrolytes are less conductive than their aqueous counterparts. In that regard, they have some of the same drawbacks as lithium-ion batteries, which also use various ionic concoctions for their electrolytes.
For example, aluminumion batteries will likely heat up just like their lithium-ion counterparts. “All high-performance, high-voltage batteries use a nonaqueous electrolyte having a conductivity that is a couple orders of magnitude lower than that of the aqueous sulfuric acid you find in lead-acid cells,” explains ORNL Senior Research Staff Scientist Gilbert Brown. “You can mitigate the problem somewhat by making the current path as short as possible, which is why lithium-ion cells are packaged as pouches. But when you discharge in a hurry through that resistive electrolyte, things are still going to get warm.”
Building a better cathode
The ORNL researchers devised an aluminum-ion coin cell that used aluminum in the anode and spinel manganese- oxide as the cathode, a material that reacts reversibly with aluminum. Another research group at Cornell University has devised a similar aluminum-ion coin cell, but with a different cathode material. Researchers, directed by Cornell Professor of Chemical and Biomolecular Engineering Lynden Archer, used the same ionic electrolyte as the ORNL team but substituted vanadium-oxide nanowires as the cathode. The aluminum wets and permeates the pores of the metal-oxide cathode. The resulting battery is electrochemically stable over a relatively wide range of voltages and currents, says the team.
“Vanadium oxide is one of those workhorse materials that has an open-crystal structure,” says Cornell’s Archer. “The thought is that a relatively large aluminum complex can fit in its structure. Use of nanowires gives a high surface area and short transport distances when you extract electrons.”
The Cornell team characterized their coin cells using what’s called cyclic voltammetry and galvanostatic cycling. The results, says Archer, were promising. The voltammeter tests showed repeatable batterylike electrochemical processes taking place as voltage across the battery ramped up and down. Galvanostatic tests of battery current showed the cells gave up and liberated electrons repeatably and that they retained a lot of their capacity after numerous charge/discharge cycles. But Cornell’s aluminum-ion battery design isn’t quite ready for prime time. For one thing, its coulombic efficiency isn’t where it needs to be for long-term practical use as a secondary battery. A perfectly coulombic-efficient battery would preserve all its ions in every charge/discharge cycle. Anything less than 100% coulombic efficiency denotes a steady loss of active material that eventually kills the battery.
“We don’t want to create too much excitement right now because we are actively pursuing this area,” explains Archer. “Our aluminum-ion cells already produce an energy density higher than that for most lithium-ion batteries in use today. That’s pretty good, but not as good as we want it to be. Our goal is to create a battery that can compete with an internal combustion engine, which has an energy density of around 5,000 W-hr/kg.”
If developments progress as planned, the Cornell group could have even more-interesting results to show in about a year. “The nice thing about batteries is that scaling up to a bigger cell involves just scaling up the amount of material, once you get something that works reliably at a small scale,” says Archer. “We are actively pursuing several chemistries, and we are having some success. In about a year, we should have pouch-sized batteries with enough capacity to drive a laptop.”
However, the process of producing an aluminum-ion battery with enough oomph to power an EV goes beyond what’s practical in an academic setting, Archer says. Cornell just doesn’t have the infrastructure to produce such large cells. So if the Cornell group gets to a point where large cells seem practical, the university will be looking for a commercial partner to actually make them happen.
As promising as the aluminum-ion developments sound, a lot of work remains to be done. “There are going to be an enormous number of problems with aluminumanode batteries. We just don’t know what they all are yet,” says ORNL’s Gilbert Brown.
For one thing, the ionic electrolyte is pricey, though it probably can be recycled. Nor is the question of aluminum- ion anode material completely settled. “In aluminum- ion batteries, you can get the same kind of dendritic formations on the electrodes that cause problems in lithium-ion batteries,” says ORNL distinguished research staff member and group leader M. Parans Paranthaman. “There are things we need to look at taking place at the interfaces between the electrodes and the electrolyte.”
And a switch from lithium-ion to aluminum-ion chemistry wouldn’t eliminate problems arising from a complete discharge of an EV battery. These difficulties made headlines recently when it came to light that a few owners of Tesla roadsters had let the batteries in their $100,000 EVs discharge to zero and were stuck paying for replacements at costs running well into five figures.
“We don’t completely understand why a totally depleted lithium-ion battery can’t be recharged,” says Gilbert Brown. “It would be premature to say an aluminum-ion battery would solve the problem. The two types of batteries have two completely different anode structures. The lithium anode is graphite intercalated with lithium. An aluminum battery has an aluminum anode onto which aluminum gets electrodeposited. All you can say definitively is that an aluminum-ion battery would have a different problem with total discharge than would a lithium-ion battery.”
Moreover, aluminum-ion batteries that get smashed in a crash could cause the same concerns as lithium-ion cells. These came into national prominence when a Chevy Volt’s lithium-ion battery pack caught fire three weeks after a severe, side-impact rollover crash test. But initial indications are that aluminum-ion batteries would have no more crash concerns than their lithiumion cousins, though there are a lot of unknowns. “Both lithium and aluminum are reactive with oxygen and water,” says ORNL’s Parans Paranthaman. “In a breech of the battery package, aluminum has one additional factor going for it because aluminum oxide will form on the surface of the anode and quench the reaction. Of course, this will also wreck the battery.”
Finally, it may turn out to be more advantageous to power EVs with unrechargeable primary batteries made with aluminum-ion chemistry rather than struggle to devise cells able to last the life of a car. That’s one scenario that ORNL researchers won’t totally discount.
“Think of a battery you could pick up at a service station that would be good for 500 miles,” says Gilbert Brown. “People might be willing to drive in for a regular swap-out if the cost was reasonable. Aluminum costs roughly $2.50/kg. Given its free-energy content, buying an aluminum-based battery might be like paying $5/gallon for gasoline. It would be even more compelling if we could learn to produce aluminum metal with a lower-temperature process.”