Sooting tendencies: Combustion science for designing sustainable fuels with improved properties

The transition from fossil fuels to sustainable fuels offers a unique opportunity to select new fuel compositions that will not only reduce net carbon dioxide emissions, but also improve combustor performance and reduce emissions of other pollutants. A particularly valuable goal is finding fuels that reduce soot emissions. These emissions cause significant global warming, especially from aviation since soot particles are the nucleation site of contrails. Furthermore, soot contributes to ambient fine particulates, which are responsible for millions of deaths worldwide each year. Fortunately, soot formation rates depend sensitively on the molecular structure of the fuel, so fuel composition provides a strong lever for reducing emissions. Sooting tendencies measured in laboratory-scale flames provide a scientific basis for selecting fuels that will maximize this benefit. Recent work has developed new techniques that expand the range of compounds that can be tested by reducing the required sample volume and increasing the dynamic range. This has many benefits, but it is particularly essential for the development of structure-property relationships using machine learning algorithms: the accuracy and predictive ability of these relationships depends strongly on the number of compounds in the training set and the coverage of structural features. This paper reviews: (1) these new techniques; (2) trends in sooting tendency versus molecular structure; (3) structure-property relationships for sooting tendency; and (4) interpretation of the observed trends based on first-principle chemical kinetic and molecular dynamic simulations.