The Effects of Small Amounts of Cobalt in LiNi1-XCoxO2 on Lithium Diffusion

Abstract

Lithium-ion batteries with Co-free Ni-rich cathodes have attracted interest due to lower cost and supply chain issues compared to cobalt-containing cathodes1. However, cobalt is thought to provide some benefits in cycling stability and rate performance. In this work, cobalt substitution for nickel in the positive electrode material LiNi1-xCoxO2 at 0 ≤ x ≤ 0.10 is systematically investigated to determine the impact of Co and material synthesis conditions on Li diffusivity, measured using the Atlung Method for Intercalant Diffusion (AMID)2,3 in coin cells vs Li metal negative electrodes. Cobalt was found to have no impact on Li diffusivity in the intermediate voltage range (4.2 V to 3.7 V), while cation mixing (%Ni in Li layer) was found to have a strong correlation to slow Li diffusivity. At high voltage (4.3 V to 4.2 V), 0 to 10% cobalt incrementally suppresses the H2-H3 phase transition4,5 enabling significantly faster lithium diffusion. Cation mixing can be minimized through synthesis conditions, improving Li diffusivity for the low and intermediate voltage regions, without using Co. In summary, with optimal positive electrode material synthesis conditions and limiting cell upper cutoff voltage to 4.2 V, Co-free Ni-rich materials can be made with similar rate performance as 10% Cobalt-containing materials. Additionally, cobalt was found to have minimal impact on the following material properties: crystallinity of LiNi1-xCoxO2, surface impurities, particle size, and electronic conductivity. Cobalt substituted for nickel from 0% to 10% was found to decrease first cycle discharge capacity in the voltage range between 3.0 V and 4.3 V and improve capacity retention in coin cell cycling vs Li metal negative electrodes. The capacity retention improvement is most likely due to the suppression of the H2-H3 phase transition as Co is added5. Figure 1. (a), (b) shows %Ni in Li layer vs Cobalt content, obtained from powder XRD Rietveld refinement from a series of LiNi1-xCoxO2 materials. (a), (b) were synthesized under two different O2 flow velocities, both at 700oC for 20 hours, with a Li:TM ratio of 1.02:1. (c), (d) Show the lithium chemical diffusion coefficient vs cell voltage for the materials from (a), (b), respectively, measured using the Atlung Method for Intercalant Diffusion (AMID), in coin cells at 30oC. The protocol consists of 2C to C/160-rate discharge steps per voltage interval (0.1V), with OCV relaxation in between discharge steps. Figure 1