Understanding Local Defects in Li-Ion Battery Electrodes through Combined DFT/NMR Studies: Application to LiVPO₄F

Abstract

In a recent study, we showed by solid-state NMR that LiVPO4F, which is a promising material as positive electrode for Li-ion batteries, often exhibits some defects that may affect its electrochemical behavior. In this paper, we use DFT calculations based on the projector augmented-wave (PAW) method in order to model possible defects in this (paramagnetic) material and to compute the Fermi contact shifts expected for Li nuclei located in their proximity. The advantage of the PAW approach versus FP-LAPW we have been previously using is that it allows considering large supercells suitable to model a diluted defect. In the first part of this paper, we aim to validate the Fermi contact shifts calculation using the PAW 2 approach within the VASP code. Then we apply this strategy for modeling possible defects in LiVPO4F. By analogy with the already existing homeotypic LiVOPO4 phase, we first replace one fluoride ion, along the VO2F4chains, by an oxygen one and consider, in a second step, an association with a lithium vacancy. As a result, the agreement between the calculated NMR spectra and the experimental one is satisfying. In both cases, the local electronic structure and the spin transfer mechanisms from V 3+ or V 4+ ions to the Li nuclei are analyzed.

LiVPO4F1–yOy Tavorite-Type Compositions: Influence of the Concentration of Vanadyl-Type Defects on the Structure and Electrochemical Performance

“…the formation of vanadyl-type defects) could be at the origin of these discrepancies. This hypothesis has already been investigated by Nuclear Magnetic Resonance (NMR) spectroscopy and Density Functional Theory (DFT) calculations 7,8 but its experimental validation was still challenging due to the low concentration of defects in LiVPO4F. Therefore, we synthetized LiVPO4F1-yOy compositions controlling the substitution ratio of oxygen for fluorine, in order to study the influence of vanadyl-type defects on the average structure as well as on the local environments around vanadium.…”

Local atomic and electronic structure in the LiVPO4(F,O) tavorite‐type materials from solid‐state NMR combined with DFT calculations

“…[2] Nevertheless, among the numerous studies reporting on apparently pure LiVPO 4 F, discrepancies in the average crystal structure and electrochemical properties are observed. [2][3][4][5] Using 2D 7 Li magic-angle spinning (MAS) nuclear magnetic resonance (NMR), we evidenced the presence of defects in LiVPO 4 F [6] whose amounts may vary according to the synthesis conditions. Following this work, we identified the nature of the defects using 7 Li MAS NMR combined with density functional theory (DFT) calculations [7] : F − ions are substituted locally by O 2− , thus forming V 4+ ions involved in a vanadyl bond with a very specific electronic structure that modifies the Fermi contact shifts observed for adjacent Li + ions.…”

“…[2][3][4][5] Using 2D 7 Li magic-angle spinning (MAS) nuclear magnetic resonance (NMR), we evidenced the presence of defects in LiVPO 4 F [6] whose amounts may vary according to the synthesis conditions. Following this work, we identified the nature of the defects using 7 Li MAS NMR combined with density functional theory (DFT) calculations [7] : F − ions are substituted locally by O 2− , thus forming V 4+ ions involved in a vanadyl bond with a very specific electronic structure that modifies the Fermi contact shifts observed for adjacent Li + ions. Recently, Kang and co-workers reported a LiVPO 4 F 0.25 O 0.75 phase, whose average structure is similar to that of LiVPO 4 F (despite the large O 2− /F − substitution ratio) showing the possibility of extended substitution of fluorine by oxygen.…”

“…Extra 7 Li NMR experiments were performed at the French national platform IR-RMN-THC FR3050 CNRS (Orléans, France), on a Bruker Avance III spectrometer with a 17.6-T magnet (291.5-MHz resonance frequency for 7 Li) and a wide-bore Bruker 1.3-mm MAS probe with a spinning rate of 64 kHz. The chemical shifts were referenced to an aqueous 1 mol L −1 LiCl solution.…”

“…The chemical shifts were referenced to an aqueous 1 mol L −1 LiCl solution. Two-dimensional 7 Li- 7 Li EXchange SpectroscopY (EXSY) and finite pulse Radio Frequency-Driven Recoupling (fp-RFDR) correlation spectra were recorded with a repetition time of 50 ms (sufficient for relaxation of the signals of interest-peaks 2 and 4 are saturated) and a RF power of 98.3 W (208 kHz). The other conditions were similar to those used for the study published earlier.…”