Revisiting Tin Oxide: A Systematic Investigation of Performance As a Function of Particle Size, Dopant and Electrolyte

Abstract

The reliable application of conversion and alloying materials as high capacity electrodes in Li-ion batteries has been a mostly elusive ambition for nearly 30 years. The most notable successes being the commercialisation of the Tin based Nexelion™ chemistry by Sony in 2005, and the current commercialisation of Si and SiOx materials as anode additives today. Despite the relatively narrow scope of commercialised systems however, the scientific literature remains vibrant on this topic, spurred on by the allure of specific capacities in excess of 1000mAh/g.The limitations of conversion systems are well known: Extremely poor first cycle efficiencies, large voltage hysteresis, wide voltage windows and generally poor cyclability arising from the massive morphological changes during cycling. Nanostructuring is widely regarded as the best route to mitigating these problems, and the literature is accordingly rich with complex and “photogenic” nanostructures and composites demonstrating greatly improved performance. These are however often produced by routes which are difficult to scale.Within the SMICE-Li project we have been revisiting conversion and conversion-alloy materials with an inherently scalable synthesis route: Flame Spray Pyrolysis. This involves the continuous combustion of an atomised spray of combustible precursors and yields an inherently nanoscaled product which can be controllably grown and modified with post-synthesis thermal annealing.We will present the results of a systematic study into doped SnO2 as a function of dopant, particle size, cycling voltage window and electrolyte formulation. Clear differences in cyclability are observed for SnO2 just by altering particle size from ~7nm to ~25nm, but these may be suppressed by the addition of dopants. Clear performance differences between dopants (Ni, Fe, Co, Mn; Ti) are also found, with results replicated across a range of organic-carbonate and ionic liquid electrolyte compositions. For all compositions the Sn-SnO2 Conversion reaction is seen to be the major source of capacity fade, but some dopants stabilise the Sn-LixSn alloy reaction whilst others do not. Restricting the voltage cycling window to limit cycling to only the alloy reaction greatly improves cyclabilty in all compositions, and for the best performing compositions, stable cycling with capacity retention in excess of 450mAh/g is achieved after 300 cycles.