Dopant synergy vs. competition in codoped Li7La3Zr2O12 garnet solid electrolytes

Abstract

LLZO is typically singly doped with Ga or Ta to reach ionic conductivities over 1 mS/cm, and some recent studies have examined the potential conductivity benefits of codoping the material. However, to date codoped LLZO fails to out-perform singly doped LLZO in terms of conductivity, and no study has sufficiently explored the consequences on other necessary properties. We have previously screened 59 possible single dopants to determine their impact on these properties. No single dopant addresses all requirements, thus, here we explore triple doping. We choose Ga for high ionic conductivity and low electronic conductivity, Dy for high voltage stability, and Ti for low voltage stability. The benefits and tradeoffs of codoping are determined with 64 samples spanning the triple doping space. We find fundamental limitations to codoping. The high ionic conductivity of Ga-doped LLZO is lowered by the addition of the other dopants, though this is mitigated partly by increased Ga levels. Electronic conductivity, by contrast, shows that a critical level of Ga is needed to obtain low conductivities, and this is resilient to the other dopants such that codoping is effective here. By contrast, the high voltage stability is systematically limited by the poor performance of Ga, and Dy is not effective in overcoming this, but Ti alone does help the low voltage stability of the Ga containing materials. These fundamental limitations suggest that multilayer designs will likely be required for viable solid batteries.