TU Graz pulled oxygen out of a battery anode on purpose, and lithium moved faster

Where the defect sits in the charge curve matters more than the fact that a defect helps. Pristine lithium titanate conducts lithium badly and normally only becomes a good conductor once charging forces extra lithium and electrons into its structure ^[2]. That puts its worst transport state at the empty end, which is exactly where a fast charge begins ^[1]. An anode marketed on rate has its weakest ion pathway at the moment it is asked to accept the most current, and the 300 C treatment attacks that state rather than the charged one ^[3].

The intervention is cheap in the specific sense that it consumes nothing new. The oxygen vacancies open room for lithium without altering the material's basic formula ^[4], and the approach leans on thermal history rather than composition or a lithium pre-insertion step ^[11]. So the delta is a furnace recipe, not a bill of materials ^[2]. The temperature choice also protects the reason anyone buys LTO in the first place: it is a zero-strain host whose lattice barely expands or contracts on insertion and extraction, and generating vacancies at low temperature preserves that stability instead of trading it away ^[8].

What the account does not contain is a magnitude. Martin Wilkening's description of the outcome, that vacancies transform an originally poor ionic conductor into a significantly better one ^[5], is the only sizing on offer; there is no conductivity ratio, no activation energy, and no cell-level result in this write-up ^[3]. The paired measurement, conductivity spectroscopy for the macroscopic current and nuclear magnetic resonance to follow individual lithium cations along the opened route ^[7], is good evidence about mechanism. Mechanism and rate performance are different claims, and only the first one is settled here.

The narrower statement in the work is the more useful one. Wilkening says the diffusion pathway is already pre-formed in the LTO structure and is only activated by the defect structure ^[6]. That is a fact about this oxide's architecture, not a general licence to under-oxidise ceramics and hope. Bernhard Gadermaier and Wilkening skipped the charging step and built the defects in from the start ^[10]; the reason that worked is that LTO had a channel waiting.

For anyone specifying anodes, the relevant comparison is against the material's own datasheet. LTO already displaces graphite on the strength of ultra-fast charging and lifetimes in the thousands of cycles ^[9]. A conductivity gain in the discharged state, if it survives into electrodes, shows up as headroom at the start of charge rather than as a new capability. The paper is in Science Advances ^[12], and the number that decides whether the extra anneal earns its place in a line is the one that has to come out of it.