Mechanism of Action of the Tungsten Dopant in LiNiO₂ Positive Electrode Materials

Abstract

The addition of tungsten has been reported to greatly improve the capacity retention of Ni‐rich layered oxide cathode materials in lithium‐ion batteries. In this work, Ni(OH)2 precursors, coated with WO3 and also W‐containing precursors prepared by co‐precipitation followed by heat treatment with LiOH·H2O, are studied. Structural analysi s and electron microscopy show that W is incorporated as amorphous LixWyOz phases concentrated in all the grain boundaries between the primary particles of LiNiO2 (LNO) and on the surface of the secondary particles. Tungsten does not substitute for Ni or Li in the LNO lattice no matter how W is added at the precursor synthesis stage. Scanning electron microscopy (SEM) images show that adding W greatly suppresses primary particle growth during synthesis. In agreement with previous literature reports, cycling test results show that 1% W added to LNO can greatly improve charge–discharge capacity retention while also delivering a high specific capacity. The LixWyOz amorphous phases act as coating layer on both the primary and secondary particles, restrict primary particle growth during synthesis and increase the resistance of the secondary particles to microcracking.

Oxygen-redox activity in non-Li-excess W-doped LiNiO2 cathode

“…1b inset) as the broad bump near the 003 reflection. [35] Nevertheless, the reported effects of W doping such as increased lattice parameters and peak broadening are clearly seen in the WLNO XRD data when compared to undoped LiNiO2 (Fig. S1).…”

“…1b) of pristine WLNO can be fit with a rhombohedral R3 ̅ m unit cell with the refined lattice parameters comparable to those in literature (Table S1). [33,35,36] No traces of unreacted precursor are seen. However, there is compelling evidence that true bulk W doping (i.e., W occupying Li/Ni sites) does not occur in WLNO, and that amorphous LixWyOz (x/y > 1) phases are formed along grain boundaries and particle surfaces.…”

“…However, there is compelling evidence that true bulk W doping (i.e., W occupying Li/Ni sites) does not occur in WLNO, and that amorphous LixWyOz (x/y > 1) phases are formed along grain boundaries and particle surfaces. [34,35] This makes the bulk Li-deficient, leading to the formation of off-stoichiometric phases. Evidence for amorphous LixWyOz phases can be seen in the XRD data (Fig.…”

“…S1). [33,35] Structural analysis of the pristine WLNO was done through Rietveld refinement [43,44] against the XRD data (SI section 2.3). LiNiO2 often exists as off-stoichiometric Li-deficient Li1−zNi1+zO2 (z < 1) phases.…”

“…To probe the nature of oxidized O 2− species in non-Li-excess cathodes, we performed an O K-edge hr-RIXS study using 2% tungsten doped LiNiO2 (LiNi0.98W0.02O2) as a model system. In addition to its superior electrochemical performance compared to undoped LiNiO2 [33][34][35][36], it is a Mn-free system, which removes the possibility of the theoretically predicted [37] but experimentally unverified [38,39] Mn 4+ to Mn 7+ oxidation and the stabilization of orphaned O 2p orbitals by Mn. [40][41][42] The crystallographic and electronic structure changes in the LiNi0.98W0.02O2 cathode were investigated as a function of delithiation using X-ray diffraction and spectroscopy methods.…”