Lack of Cation Clustering in Li[Ni_xLi_1/3-2x/3Mn_2/3-x/3]O₂ (0 < x ≤ ¹/₂) and Li[Cr_xLi_(1-x)/3Mn_(2-2x)/3]O₂ (0 < x < 1)

Abstract

Recent papers by Ammundsen et al. and Pan et al. give evidence for the formation of local regions high in Mn content and other local regions high in Cr or Ni content in Li[Li0.2Cr0.4Mn0.4]O2 and Li[Ni0.5Mn0.5]O2 by EXAFS and NMR methods, respectively. These observations are surprising for the following reasons: (1) each of these materials is a part of a solid solution series, Li[CrxLi(1-x)/3Mn(2-2x)/3]O2 (0 < x < 1) or Li[NixLi1/3-2x/3Mn2/3-x/3]O2 (0 < x < 1/2); (2) the materials are made at high temperature, and entropy considerations suggest that like transition-metal atoms should not cluster; and (3) the electrochemical and structural properties of the materials vary smoothly with composition. Here, using careful X-ray diffraction on many samples from each solid solution, we show that it is very unlikely that such local regions high in Mn, Ni, or Cr exist. We show that long-ranged lithium ordering on the 31/2 a × 31/2 a superstructure occurs as expected based on the work of Schick et al., however, this does not imply local regions of Li2MnO3. Instead, the diffraction angles of the superstructure peaks shift with composition suggesting that the Mn, Cr, or Ni are uniformly mixed on the transition-metal sites. In addition, we show that the electrochemical behavior of Li[NixLi1/3-2x/3Mn2/3-x/3]O2 heated to 1000 °C is improved compared to that of samples made at 900 °C.