First-Principles Investigation of the ⁵⁷Fe Mössbauer Parameters of LiFePO₄ and FePO₄

Abstract

The 57Fe Mössbauer isomer shift and quadrupole splitting of LiFePO4 (LFP) and FePO4 (FP) were evaluated from density functional theory. The effects of structural optimization, Fe 3d electron correlation, and spin configuration are found to be rather small for the isomer shift but not for the quadrupole splitting. The observed decrease of the isomer shift from LFP to FP is explained from the variations of both Fe 4s and Fe 3d electrons and is related to differences in the Fe–O chemical bond properties caused by oxidation of high-spin Fe2+ into high-spin Fe3+. The values of the quadrupole splitting of the two compounds strongly increase from nonspin-polarized to spin-polarized calculations and then slightly increase by taking into account the Fe 3d electron correlation, providing a good agreement with experimental data. The observed decrease of the quadrupole splitting from LFP to FP is quantitatively correlated to the decrease of the magnitude of the highest principal value V ZZ of the electric field gradient (EFG) tensor arising from the anisotropy of Fe p-type electrons for FP and Fe 3d↓ electrons for LFP. The EFG of FP reflects the distortion of the FeO6 octahedra mainly arising from edge-sharing PO4 tetrahedra, while it is due to high-spin Fe2+ for LFP. Finally, the calculations predict that the sign of V ZZ changes from positive for LFP to negative for FP, reflecting electron depletion and accumulation, respectively, along the principal axis Z due to the rotation of the principal axis system. As a consequence, the charge–discharge cycles of LFP-based cathode for Li-ion batteries produce strong periodic variations of V ZZ through almost the overall EFG scale for iron compounds.

Best Practices and Protocols in Mössbauer Spectroscopy

“…With the recent improvements in density functional and first-principles calculations, the best value appears to be Q = 0.16(1) × 10 –28 m 2 or 0.16(1) b . In agreement with this value, one also finds Q = 0.1549(2) b, Q = 0.173 b, and Q = 0.17 b, in the recent literature. The value of Q = 0.16(1) b has been recently updated at in agreement with the most recent review of nuclear quadrupole moments.…”

Mössbauer Spectroscopy of Triphylite (LiFePO4) at Low Temperatures

“…Table 2 collects results of these calculations (see the second column). The AF2 order is observed in reality (see, e.g., Reference [10]) and was also confirmed in other computations [8,18]. Table 2 also contains results of hyperfine parameter calculations.…”

“…AF1 and AF2 show very similar hyperfine parameters, whereas AF3 differs slightly from them. Nevertheless, this indicates a weak sensitivity of the EFG and Fermi contact contribution to B hf to various antiferromagnetic arrangements (Reference [18] reports similar observation).…”

“…The main subject of the present study is thus low temperature magnetic properties as seen by Mössbauer spectroscopy and the way how MS can help in understanding aforementioned characteristics. Previous MS experimental investigations (see, e.g., References [16][17][18]) indicate that above T N there is no magnetic ordering (only quadrupolar splitting is observed), whereas below T N magnetic order is seen (magnetic hyperfine splitting is combined with the quadrupolar one). The question now arises: What happens if an external magnetic field is applied?…”

Electrochemical Performance and Mechanisms of NaSn2(PO4)3/C Composites as Anode Materials for Li-Ion Batteries

“…The theoretical values of the isomer shift were evaluated from the electronic density at the nucleus, ρ(0), and a calibration constant. This constant was obtained from the linear correlation between δ exp and ρ(0) for a series of tin compounds by considering the same computational parameters such as NaSn 2 (PO 4 ) 3 , following the procedure described previously. − The values of the isomer shift are in the range from −0.49 to −0.24 mm s –1 depending on the Sn crystallographic site, Sn(1) or Sn(2), the structural model: R 3̅ or R 3̅ c and experimental or optimized structures (Table S4). For R 3̅, the difference between Sn(1) and Sn(2) is lower than 0.1 mm s –1 for both optimized and experimental structures, which explains why they cannot be distinguished experimentally.…”