FULL PAPER © 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim (1 of 11) 1401142 wileyonlinelibrary.com Unfolding the Mechanism of Sodium Insertion in Anatase TiO 2 Nanoparticles Liming Wu, Dominic Bresser, Daniel Buchholz, Guinevere Giffin, Claudia Ramirez Castro, Anders Ochel, and Stefano Passerini* L. Wu, D. Bresser, D. Buchholz, Dr. G. Giffin, Dr. C. Ramirez Castro, A. Ochel, Prof. S. Passerini Institute of Physical Chemistry and MEET Battery Research Centre University of Muenster Corrensstrasse 28/30 & 46, 48149, Muenster, Germany E-mail: stefano.passerini@kit.edu L. Wu, D. Bresser, D. Buchholz, Dr. G. Giffin, A. Ochel, Prof. S. Passerini Helmholtz-Institute Ulm, Karlsruher Institute of Technology Albert-Einstein-Allee 11, 89081, Ulm, Germany DOI: 10.1002/aenm.201401142 Hard carbons, on the contrary, show good sodium storage capability, [7,8] and recently sodium titanates [9,10] as well as alloying materials such as tin, [11–13] ger- manium, [14] or antimony [15] were reported as promising sodium-ion anodes. Amor- phous TiO 2 , [16–18] TiO 2 -B, [19] and anatase TiO 2 [20–24] were also recently studied in terms of their reversible sodium uptake and release. For the latter, an excellent cycling stability and high rate performance was reported. [21,23] However, the under- lying reaction mechanism, i.e., the process of reversible sodium storage, remained unclear. Although early studies presented contradictory results on the possibility of reversible sodium ion (de-)insertion into the anatase lattice, [25–28] Cha et al. [22] very recently claimed reversible sodium ion insertion into the anatase lattice based on ex situ XRD analysis of cycled electrodes. Nevertheless, the conclusions presented appear debatable, as the lattice expansion was calculated for an almost amorphous phase and the reappearance of the anatase reflections were observed after rinsing the charged electrode with water. Similarly Kim et al. [24] suggested that sodium ions would be reversibly (de-)inserted into the anatase lattice, showing that the anatase reflections reversibly shifted upon sodium uptake and release while the crystallinity was essen- tially maintained. In addition, they reported the reduction of Ti 4+ to Ti 3+ upon sodium ion insertion, indicating a charge bal- ancing caused by the sodium ion insertion. González et al., [23] on the contrary, reported that the anatase-related X-ray diffrac- tion (XRD) reflections would remain unaffected even after full discharge of the electrode down to 0 V vs. Na/Na + . Based on additional NMR and electrochemical studies, they concluded that the majority of the capacity observed for an operational potential window of 2.6 to 0.5 V would be pseudocapacitive in nature, while the irreversible faradaic processes at potentials below 0.3 V would be basically related to electrolyte decomposi- tion. In fact, the absence of the characteristic potential profile as for the reversible lithium ion (de-)insertion following a two- phase transition process [29–34] indicates that there are great dif- ferences between the electrochemical reaction with lithium and sodium. The results obtained for a commercial sample of nano- particulate anatase TiO 2 , and presented herein, lead us to another conclusion on the faradaic processes taking place at low potentials. Rather unexpectedly, it appears that sodium is It is frequently assumed that sodium-ion battery chemistry exhibits a behavior that is similar to the more frequently investigated lithium-ion chemistry. However, in this work it is shown that there are great, and rather surprising, differences, at least in the case of anatase TiO 2 . While the gener- ally more reducing lithium ion is reversibly inserted in the anatase TiO 2 lat- tice, sodium ions appear to partially reduce the rather stable oxide and form metallic titanium, sodium oxide, and amorphous sodium titanate, as revealed by means of in situ X-ray diffraction, ex situ X-ray photoelectron spectroscopy, scanning electron microscopy, and Raman spectroscopy. Nevertheless, once the electrochemical transformation of anatase TiO 2 is completed, the newly formed material presents a very stable long-term cycling performance, excel- lent high rate capability, and superior coulombic efficiency, highlighting it as a very promising anode material for sodium-ion battery applications. 1. Introduction Research in sodium-ion battery chemistry is currently attracting a steadily increasing number of scientists all around the world and targets the replacement of state-of-the-art lithium-ion bat- teries for large-scale applications such as stationary energy storage, where weight and size of the battery are of less impor- tance than cost. Fortunately, sodium-ion chemistry follows sim- ilar mechanisms as lithium-ion chemistry, enabling rather fast progress in identifying suitable electrode materials, particularly with respect to the cathode side. Accordingly, very promising candidates, such as transition metal oxides, [1,2] poly-anionic compounds, [3] and NASICON-type (NASICON is defined as natrium super ionic conductor) sodium metal phosphates [4,5] were recently reported. With respect to the anode side, never- theless, it turns out that the state-of-the-art lithium-ion anode material graphite does not reversibly intercalate sodium ions. [6] Adv. Energy Mater. 2014, 1401142 www.MaterialsViews.com www.advenergymat.de