Effects of Doping on XLi2MnO3•(1-x)LiNi1/3Mn1/3Co1/3O2 Prepared Via Spray Pyrolysis

Abstract

LixMO2 (M= Mn, Ni, Co) materials with layered structures have received attention as high-capacity, low cost and safer cathode materials for lithium-ion batteries [1]. Various synthesis methods have been developed for the production of these materials, including co-precipitation, solid state synthesis and ball milling [2-4]. Recently a spray pyrolysis synthesis method has been developed by this group for producing cathode materials with porous morphologies [5]. Spray pyrolysis has certain advantages compared to other methods: the short residence time in the reactor allows large throughput; the process is scalable, no further post-synthesis purification steps are required, the product is inherently purified throughout the synthesis. Most importantly, each particle is synthesized with precise stoichiometry because each droplet behaves as a mini-reactor. This allows for accurate control of dopant levels in the product. The layered lithium-nickel-manganese-cobalt oxides, such as 0.5Li2MnO3•0.5LiNi1/3Mn1/3Co1/3O2 are known to demonstrate a voltage fade over cycling. This effect has been attributed to a layered-spinel transformation [6-7]. Many studies have suggested that voltage fade can potentially be eliminated via doping. Therefore, this study will focus on doping xLi2MnO3•(1-x)LiNi1/3Mn1/3Co1/3O2using trace level dopants and exploring their effects on voltage fade and cycling stability.Figure 1 shows a typical SEM result of lithium excess material prepared via spray pyrolysis, demonstrating the uniform and spherical morphology. Figure 2 compares the charge and discharge profile for xLi2MnO3•(1-x)LiNi1/3Mn1/3Co1/3O2 materials annealed at 900°C for 2 hours when x is 0.5 and less than 0.5. Voltage fade is typically observed as a shift in the charge and discharge curves as a result of the structural transformation [6]. It is clearly present for x=0.5 (black line in Fig.2), however the effect is almost overcome for lower x values. Initial results indicated a significant drop in capacity values, which are expected to be overcome by optimization of synthesis conditions. Figure 3 compares the cycling results of the materials with and without Al doping. The material doped with Al clearly demonstrates a better cycling stability after 100 cycles. The Al doped sample retains 99.4% of it’s capacity between cycle 5 and 100, compared to 91.5% for the undoped sample. Acknowledgements: The authors are grateful to the NSF and X-tend Energy LLC for support. RLA and Washington University may receive income based on a license of related technology by the University to X-tend Energy LLC.