The all-solid-state battery : the state of the art, recent developments, why the sudden craze ?

Listen to audio

The all-solid-state battery continues to be the subject of media announcements by electric vehicle manufacturers (Toyota, BMW, Renault-Nissan...), who predict that by 2022, the all-solid-state battery could be competing with Tesla vehicles. We recall, however, that we experienced the same craze five years ago, with similarly optimistic announcements about Li-air technology, which is now forgotten. So what's happening with solid state?

That's what this latest lecture will attempt to answer, by reviewing recent scientific advances. We begin by noting that the all-solid-state battery is a story that is resurfacing due to a period of intense activity in the 1980s, centred around sulphide fluoride glasses, with J.R. Akridge's 1986 achievement of an all-solid-state thin-film battery that exceeded 1,000 cycles.

However, these efforts were not sustained due to difficulties in controlling interfaces and finding better ionic conductors. Thanks to the perseverance of certain groups over the years, new families of electrolytes (NASICON, GRENATS, PEROVSKITE, THIO-LISICON) have been identified, with the biggest breakthrough coming in 2011, when a Japanese group announced a new _Li10GeP2S12 conductor (known as LGPS) that boasted a conductivity of 10 mS - ^cm-1.

This conductivity compares favorably with that of liquid electrolytes. Recent chemical substitutions, replacing germanium with silicon and tin, and sulfur with chlorine or oxygen, have resulted in ionic conductivities of 30 mS - ^cm-1 to 40 mS - ^cm-1.

These advances were matched by progress in all-solid battery assembly through the use of hot sintering (SPS) and even strip casting. Thanks to DFT calculations and in line with experimental results, it has been shown that most solid electrolytes are not stable. As with liquid electrolytes, their stability results from the slow kinetics of decomposition reactions at the interfaces. We have shown that the use of _LixSiy or _LixAlalloys enables us to switch from lithiophobic to lithiophilic behavior at the Li level, thereby achieving a better interface.

Finally, in terms of safety, DSC measurements revealed an exothermic peak associated with the reaction (2Li + ½ _O2 → _Li2O), making these batteries also flammable.

In summary, the all-solid-state battery has recently undergone significant progress, raising many hopes. However, these must be tempered by the complexity of the interfaces, which has yet to be mastered. Hence the unrealistic aspect of certain announcements.