Aluminium is more plentiful and cheaper that lithium, making it potentially attractive for large-scale energy storage even if the technology never became light enough for automotive or phone use.
Aluminium also has the advantage over lithium that it can be used in metallic form without the safety concerns that metallic lithium arouses.
The Cornell team set out to discover why aluminium batteries develop short-circuits and die after only a few charge-discharge cycles – studying a simple cell where a metallic aluminium cathode faces a stainless steel anode across a glass fibre separator soaked in electrolyte.
Tall sparsely-spaced aluminium peaks grew on the stainless steel in a few charge-discharge cycles, pushing through the glass fibre and shorting to the opposite aluminium electrode. The same was true using non-planar nickel foams instead of flat stainless steel.
These mountainous deposits locked up aluminium, resulting in a cycle-by-cycle reduction in stored energy in the cell’s short life, according to Cornell.
None of this was new knowledge, nor was the next thing that Cornell confirmed: that a significant part of the problem is aluminium’s desire to react with the glass fibre separator, and its lack of enthusiasm for the stainless steel or nickel anode surface.
The question became: is there an electrode surface that aluminium wants to bond with much more than glass fibres.
And Cornell’s answer was ‘yes’: carbon fibres with an oxidised surface – one that allows a C-O-Al bond to form, linking fibre to metal.
Oxidisation came free with exposure to the electrolyte, which is mix of imidazolium chloride and aluminium chloride that naturally sticks oxygen to plentiful defects on the carbon surface.
Instead of forming peaks, ions of aluminium landing on the oxidised carbon surface accumulated sideways rather than upwards, forming an even mat of nanoscale crystals (see photo). Even when larger crystals grew once the surface was completely covered, they stayed low and didn’t climb into peaks – and this remained true at higher currents that would normally encourage uneven growth.
The anodes have return well above 99% of deposited aluminium through hundreds of charge-discharge cycles over thousands of hours – or even thousands of cycles, depending on current and how much of the last 1% is left each time.
Subsequently a carbon-based coating was found for stainless steel that had a similar effect, and a modified version of the technique shows promise for metallic zinc batteries.
Cornell University worked with Brookhaven National Laboratory and the State University of New York at Stony Brook.
This research is covered in the Nature Energy paper ‘Regulating electrodeposition morphology in high-capacity aluminium and zinc battery anodes using interfacial metal–substrate bonding‘.