Electrochemical Characterization of Lithiation and De-Lithiation of Ultra-Thin Amorphous TiO2 Films

Research on TiO₂ as an electrode for lithium-ion batteries is mainly focused on the anatase polymorph, as high capacities can be achieved in bulk, i.e. ~200 mAh/g^1–3. However, due to limited diffusion of lithium inside anatase, these capacities are only achieved either by nanostructuring the material, or simply using slow charging rates (<<1C). In contrast, amorphous TiO₂ has only recently gained attention as a high-power insertion electrode^4–6; it was shown that next to its ease of fabrication, it shows a high capacity, good rate capability and a stable cyclic performance. Amorphous TiO₂ undergoes an electrochemically driven phase transformation into a face-centered-cubic structure upon lithiation after the first cycle⁵. It was furthermore shown that this cubic phase has a capacity of 225 mAh/g at 2C which was stable up to 600 cycles —better than any other known nanostructured TiO₂ polymorph (i.e. anatase and TiO₂(B))⁵.

In order to create a well-defined model system, we deposited ultra-thin (35 nm) amorphous TiO₂ films by atomic layer deposition. Such thin films are interesting since they possess the same length scale as nanostructured particles or other nanostructures, such as nanotubes, but they offer a better defined geometry allowing more precise determination of the underlying lithium insertion/extraction kinetics. Furthermore, such ultra-thin electrodes might find application in 3D micro-batteries⁷.

In order to "activate" amorphous TiO₂ we performed a few potentiostatic cycling steps between 3.0 V and 0.1 V vs Li⁺/Li, which significantly increased the lithium insertion kinetics of 35 nm amorphous TiO₂films, as probed by galvanostatic charge/discharge, cyclic voltammetry and impedance spectroscopy.

Cyclic voltammetry performed at 10 mV/s showed a decrease in the peak potential difference between Li⁺ insertion and extraction of 0.36 V to 0.2 V after potentiostatic activation. Furthermore, galvanostatic charge/discharge experiments showed 106% increase in the specific capacity from 1.8 µAh cm^-2 to 3.7 µAh cm^-2 at a charging rate of 1C after activation.

Impedance spectroscopy as function of lithium content before and after potentiostatic cycling showed that the overall impedance is greatly diminished, reflecting improved kinetics which lead to the increased rate-performance. Figure 1 shows complex plane plots of the potential dependent impedance measurements. The impedance response was fitted using a small-signal equivalent circuit incorporating, amongst others, a finite length Warburg impedance and charge transfer resistance.

These results offer new electrochemical insights on the effect of the electrochemically driven phase transition of amorphous TiO₂ showing its promises as a potential high-rate and high capacity electrode for lithium-ion batteries.

References
1.         M. Wagemaker, D. Lützenkirchen-Hecht, A. A. van Well, and R. Frahm, J. Phys. Chem. B, 108, 12456–12464 (2004).
2.         J. Wang, J. Polleux, J. Lim, and B. Dunn, J. Phys. Chem. C, 111, 14925–14931 (2007).
3.         J. Zhu, K. Zeng, and L. Lu, J. Solid State Electrochem., 16, 1877–1881 (2011).
4.         H. Fang et al., Nanotechnology, 20, 225701 (2009).
5.         H. Xiong et al., J. Phys. Chem. C, 116, 3181–3187 (2012).
6.         H. Furukawa, M. Hibino, and I. Honma, J. Electrochem. Soc., 151, A527 (2004).
7.         M. Valvo, et al., J. Mater. Chem. A, 32, 9281 (2013).