Sensitivity analysis and uncertainty propagation in numerical simulations of yield sooting tendencies

The sooting tendencies of various fuels have been historically quantified via the Threshold Sooting Index (TSI), which is based on spoke point measurement. More recently, a Yield-based Sooting Index (YSI) has been proposed with lower measurement uncertainties, high throughput, and applicability to low-volatility fuels. YSI of a test species is experimentally measured in an atmospheric-pressure, laminar, non-premixed co-flow methane/air flame generated with a Yale Co-flow Burner by doping the baseline methane flame with a small concentration of the test species. The YSI concept interfaces well with computations since it is a perturbationbased approach with nearly identical temperature and velocity fields in the YSI flames compared to those in the undoped methane/air flame. In this work, we use a flamelet-based model to computationally simulate the YSI of fuels relevant to diesel and gasoline combustion. The numerically predicted YSI values are compared against experimental measurements. For fuels with large discrepancies in their simulated YSI, sensitivity analysis is performed to highlight deficiencies in the chemical reaction pathways. Numerical uncertainties are then introduced in the rate constants of these reactions and propagated through YSI simulations to quantify the impact of these reactions on the YSI predictions.