Abstract
Transition metal dissolution is an important contributor to capacity fade in lithium-ion cells. NMR relaxation rates are proportional to the concentration of paramagnetic species, making them suitable to quantify dissolved transition metals in battery electrolytes. In this work, (7)Li, (31)P, (19)F, and (1)H longitudinal and transverse relaxation rates were measured to study LiPF(6) electrolyte solutions containing Ni(2+), Mn(2+), Co(2+), or Cu(2+) salts and Mn dissolved from LiMn(2)O(4). Sensitivities were found to vary by nuclide and by transition metal. (19)F (PF(6)(-)) and (1)H (solvent) measurements were more sensitive than (7)Li and (31)P measurements due to the higher likelihood that the observed species are in closer proximity to the metal center. Mn(2+) induced the greatest relaxation enhancement, yielding a limit of detection of ∼0.005 mM for (19)F and (1)H measurements. Relaxometric analysis of a sample containing Mn dissolved from LiMn(2)O(4) at ∼20 °C showed good sensitivity and accuracy (suggesting dissolution of Mn(2+)), but analysis of a sample stored at 60 °C showed that the relaxometric quantification is less accurate for heat-degraded LiPF(6) electrolytes. This is attributed to degradation processes causing changes to the metal solvation shell (changing the fractions of PF(6)(-), EC, and EMC coordinated to Mn(2+)), such that calibration measurements performed with pristine electrolyte solutions are not applicable to degraded solutions-a potential complication for efforts to quantify metal dissolution during operando NMR studies of batteries employing widely-used LiPF(6) electrolytes. Ex situ nondestructive quantification of transition metals in lithium-ion battery electrolytes is shown to be possible by NMR relaxometry; further, the method's sensitivity to the metal solvation shell also suggests potential use in assessing the coordination spheres of dissolved transition metals.