Ionic and Electronic Conduction in TiNb₂O₇

Figure: TiNb₂O₇ is a battery anode material, which allows lithium ions to rapidly pass through it and is a good electrical conductor when lithium is inserted into its structure.

This work was carried out in collaboration with the groups of Prof Clare Grey and Dr Siân Dutton at the University of Cambridge, Dr Andrew Morris at the University of Birmingham and Prof Graham Henkelman at the University of Texas.

Summary by Dr Kent Griffith

Lithium-ion batteries are the most popular energy storage technology for cell phones and laptops as well as a growing variety of cordless power tools and home appliances. Increasingly, electric vehicles with large lithium-ion battery packs are being adopted to reduce tailpipe emissions and our reliance on fossil fuels. As we move toward intermittent renewable energy sources such as solar and wind, we also need to change the grid to include more energy storage. For all of these applications, we will need better battery performance.

A lithium-ion battery comprises a cathode (positive electrode), an anode (negative electrode), and an ion-conducting electrolyte. Nearly all lithium-ion batteries use a graphite anode because it is cheap and lightweight. However, graphite can be unsafe, particularly if it is charged too fast or at extreme temperatures. In these conditions the graphite can degrade, which generates heat. Even worse, graphite-based cells can be short-circuited by spindly needles of lithium (dendrites) which form on repeated use, risking fires. The use of graphite is therefore the main reason that batteries cannot charge rapidly. In this work, we investigated a metal oxide-based anode material, TiNb₂O₇, that is inherently much safer. TiNb₂O₇ is also a good lithium-ion conductor, unlike other safer anodes, and so it is being developed as a candidate for fast charging batteries capable of being completely recharged in 10 minutes. When designing a battery, there are always tradeoffs: TiNb₂O₇ will have a lower energy density than graphite-based cells but it operates safely at higher power and in a wider range of environmental conditions.

We used a wide range of experimental and computational methods to study how TiNb₂O₇ behaves as lithium is inserted into its crystal structure. Battery electrodes must be good ion and electron conductors, but pure TiNb₂O₇ is an electrical insulator. However, for every lithium ion inserted into TiNb₂O₇, an electron is also injected. We found that when only a few electrons per metal are added TiNb₂O₇ becomes metallic: the very act of using it as a battery electrode improves its properties! We also used studied how the lithium moves through the crystal structure using variable-temperature NMR spectroscopy and quantum chemical calculations. In most materials, the sites that lithium hops through are known and so the activation energy need for Li to make the jump can be calculated. However, the complex structure of TiNb₂O₇ meant we needed to first carry out an open-ended search to find the stable hopping sites. TiNb₂O₇ is known as a crystallographic shear structure and it contains ‘blocks’ of corner-shared octahedra, separated by planes of edge-shared octahedra. We discovered that lithium diffusion within the blocks is rapid while diffusion between the blocks is nearly impossible. Our calculations also allowed us to predict the barriers to motion for other ions. We found that Na⁺ and K⁺ have rather high energy barriers and so TiNb₂O₇ is unlikely to be a suitable electrode for sodium- or potassium-ion batteries, but that Mg²⁺, although it has larger barrier than Li⁺, could be interesting to study, particularly as few Mg-ion conducting materials are known.