Li-excess metal oxide having high charge-discharge capacity has been expected to realize higher energy density in lithium-ion batteries. Several lithium-excess metal oxides have been studied extensively in order to use effectively the charge compensation by redox reaction of oxide ions. [1-5] In 4d and 5d metal oxides such as Li₂RuO₃ [1] and Li₂IrO₃-based systems [2] , high covalent bond between 4d orbital of transition-metal and 2p orbital of oxygen enables charge compensation by the oxidation of oxide ions to peroxo-like dimer. On the other hand, in a 3d transition-metal oxide, Li_1.2Ni_0.13Co_0.13Mn_0.54O₂ exhibits formation of isolated holes on oxide ions because highly ionized 3d transition-metal ions stabilize the isolated holes on the oxide ions surrounded by lithium ions. [3, 4] Other 3d transition-metal oxides such as Li_1.2Ti_0.4Mn_0.4O₂, also exhibit stable oxide ions redox. [5] According to the previous studies, covalent character or ionic character between transition-metal ions and oxide ions in Li-excess metal oxides is a key factor to stabilize oxide ions redox. However, behavior of oxide ions redox in Li-excess metal oxides with different covalent character or ionic character has not been systematically examined. In this study, we focus on four different Li-excess metal oxides, Li₂MeO₃ (Me = Mn and Ru) and Li_1.2Ti_0.4Me_0.4O₂ (Me = Mn and Fe), as models with different covalent character or ionic character. Developing a new operando soft X-ray absorption spectroscopy, we observed directly the electronic state of the oxide ions in these materials during charge process via the technique. [6]

Experimental

operando soft X-ray absorption spectroscopy measurements were performed at BL27SU in SPring-8, Japan. The absorption spectra were collected with partial fluorescence mode using home-made cell. Composite electrodes were pasted on aluminum sputtered-Si₃N₄ window. For Li_1.2Ti_0.4Mn_0.4O₂, half charged-LiFePO₄ was used as a counter electrode, and 4.2 mol dm^-3 lithium bis(trifluoromethanesulfonyl)amide/ acetonitrile was used as an electrolyte solution. Two electrode cells were assembled in the Ar-filled grovebox. XAS spectra were collected during galvanostatic measurement at rate of 1/20 C rate at 55°C.

Results and Discussion

The charge and discharge profiles of the Li_1.2Ti_0.4Mn_0.4O₂ showed that the voltage gradually increased at initial stage under charging process, then a plateau was observed at approximately 4.2 V. The first discharge profile was similar to the first charge one. The first discharge capacity was 270 mAh/g, which was comparable to the first charge capacity of 300 mAh/g. operando Mn L-edge XAS spectra of the Li_1.2Ti_0.4Mn_0.4O₂ showed that a board peak was observed at 642 eV before charge process. The peak separated to two peaks and shifted higher energy in the slope region of charge process, and the two peaks unchanged with subsequent charge process. The peak separation and shift toward higher energy in the slope region were caused by spin multiplicity and crystal field effect with oxidation Mn ions. No change of the spectra in the subsequently charge process means that the Mn ions did not contribute to the charge compensation. operando O K-edge XAS spectra of Li_1.2Ti_0.4Mn_0.4O2 showed a board peak was observed around 531 eV before charge process. The peak was divided into two peaks during charge process. These peaks reflect transition from O 1s orbitals to hybridized states between 3d orbital of transition-metal and 2p orbital of oxygen. The integrated intensity of the peak at the lower energy increased in the slope region of charge process, whereas the integrated intensity of the peak at the higher energy increased with subsequent charge process. The change of the spectrum in the slope region was caused by oxidation Mn ions (Mn³⁺ to Mn⁴⁺) as observed in the Mn-L XANES spectra. The increase of the integrated intensity at the higher energy in the subsequently charge process was caused by oxidization of oxygen ions.

References:

[1] Sathiya, M. et al., and Tarascon, J.-M. Nat. Mater. 2013, 12, 827. [2] McCalla, E. et al., and Tarascon, J.-M. Science 2015, 350, 1516. [3] Luo, K. et al., and Bruce, P. G. Nat. Chem. 2016, 8, 684. [4] Seo, D.-H. et al., and Ceder, G. Nat. Chem. 2016, 8, 692. [5] Yabuuchi, N. et al., and Ohta, T. Nat. Commun. 2016. 7, 13814. [6] Yamamoto, K. et al., and Uchimoto, Y. submitted.

Acknowledgement

This research was financially supported by the Japan Science and Technology Agency (JST), Advanced Low Carbon Technology Research and Development Program (ALCA), Specially Promoted Research for Innovative Next Generation Batteries (SPRING) Project.