Recently, our group performed a set of direct numerical simulations (DNS) of soot formation and growth in a n-heptane three dimensional non-premixed jet flame [Attili et al., Proc. Comb. Inst, 35, 2015], [Attili et al., Comb. Flame, 161, 2014], [Bisetti et al.,Trans of the Royal Soc, 372, 2014]. The evolution of species relevant to soot formation and growth have been sampled along a large number of Lagrangian trajectories in the DNS. In this work, the DNS results are post-processed to compute the soot evolution along selected Lagrangian trajectories using a Monte Carlo method. An operator splitting approach is adopted to split the deterministic processes (nucleation, surface growth and oxidation) from coagulation, which is treated stochastically. The morphological properties of soot and the particlesize distribution are investigated. For trajectories that experience an early strong nucleation event, the particle size distribution is found to be bimodal, as the soot particles have enough time to coagulate and grow while it is unimodal for trajectories characterized by only late nucleation events. As a results, the average size distribution at two different crosswise positions in the flame is unimodal.
展开▼
