Disentangling the Ligand and Electronic Structure in KVPO₄F1-XOx Positive Electrode Materials by Valence-to-Core X-Ray Emission Spectroscopy

Among all positive electrode materials for K-ion batteries (KIBs), KVPO 4 F offers a high theoretical capacity of 131 mAh·g -1 and an average working potential reaching above 4.3 V vs K + /K, resulting in a theoretical energy density of up to 520 Wh·kg -1 . [1] The anionic substitution of fluorine by oxygen in KVPO 4 F 1-x O x (x = 0, 0.25, 0.5, 0.75, 1) induces an increasing working potential by activating the V 3+/4+ and V 4+/5+ redox couples and replacing the VO 4 F 2 “ionic” entity by {V=O}O 5 unit with a highly “covalent” vanadyl-type bond. [2] The accurate prediction of the electrode potential as well as the understanding of the mechanisms involved upon cycling requires the determination of how the coordination environment of the transition metal ion influences the ionicity/covalency of the metal-ligand bonds and thus the electronic structure. Nevertheless, discriminating these ligands remains a challenge due to the limited sensitivity to light elements of common characterization techniques such as X-ray diffraction and extended X-ray absorption fine structure. Despite the valuable information from these techniques, fundamental concerns still need to be addressed which include: i) the implication of the covalent V=O vanadyl bond in the electronic structure; ii) the difference of V cis and V trans , the two different sites for V, in the electronic configuration of the material; and finally, (iii) the impact of F/O ligands on the electrochemical mechanism upon battery cycling. To address this issue, we employed valence-to-core Kβ X-ray emission spectroscopy (VtC XES) combined with ab initio modelling and conducted a systematic investigation of KVPO 4 F 1-x O x : two distinct regions were identified in the spectra, Kβ" and Kβ 2,5 , which are critical to probe the electronic structure close to the Fermi level and to discriminate different ligands coordinated with the vanadium atoms. [3] Our approach allows distinguishing in KVPO 4 F 1-x O x the contributions of V-F, V-O, and V=O bonds, with intensities at the Kβ" region highly correlated to the F - and O 2- anionic composition. Additionally, the evolution of the features at the Kβ 2,5 region is highly associated to the presence of short V=O bonds, strongly influencing the electrode potential of the material. Overall, we present a detailed and reliable approach for understanding the occupied electronic states of the electrode material, proving valuable for a thorough comprehension of the structural and redox mechanisms involved in batteries. Acknowledgement This work was supported by the DESTINY Marie Skłodowska-Curie Actions COFUND PhD Programme (Grant Agreement #945357) co-funded by the European Union's Horizon2020 research and innovation program and the Synchrotron SOLEIL. ANR is also acknowledged for funding the RS2E network through the STORE-EX Labex Project ANR-10-LABX-76-01, and the ANR TROPIC project ANR-CE05-0026. Alistore-ERI network is also acknowledged. References [1] K. Chihara et al, Chem. Commun. 2017, 53, 5208-5211 [2] R. Wernert et al, Chem. Mater. 2022, 34, 4523-4535 [3] E. Gallo and P. Glatzel, Adv. Mater. 2014, 26, 7730-7746