Onset and progression of soot in high-pressure n-dodecane sprays under diesel engine conditions. (August 2017)

• Record Type: Journal Article
• Authors: Pandurangi, Sushant S; Bolla, Michele; Wright, Yuri M; Boulouchos, Konstantinos; Skeen, Scott A; Manin, Julien; Pickett, Lyle M
• Is Part Of: International journal of engine research. Volume 18:Number 5/6(2017)
• Page Start: 436
• Page End: 452
• Publication Date: 2017-08
• Subjects: Engine Combustion Network -- soot formation -- soot onset -- Spray A -- high-pressure diesel sprays; Engines -- Periodicals; 629.25
• DOI: 10.1177/1468087416661041
• Languages: English
• ISSNs: 1468-0874

Abstract

Soot onset in n - dodecane sprays is investigated both experimentally, by means of high-speed imaging data from the Sandia spray combustion vessel, and numerically, using the conditional moment closure combustion model and an integrated two-equation soot model in a Reynolds-averaged Navier–Stokes framework. Five operating conditions representative of modern diesel engines are studied at constant density (22.8 kg/m 3 ) with variations in ambient oxygen concentration and temperature. The reference case at 15% O2 and 900 K is compared with measurements in terms of the evolving soot mass distribution and spatiotemporal distributions of formaldehyde and polycyclic aromatic hydrocarbons obtained by 355-nm laser-induced fluorescence (polycyclic aromatic hydrocarbons represented by C2 H2 in simulation) and soot optical thickness ( KL ) signal obtained by diffused back-illumination extinction imaging. All operating points are validated in terms of ignition delay and lift-off length, soot onset time and location, soot mass evolution, and peak location. Measurements show that time lag between ignition and soot onset is considerably increased by a reduction in ambient oxygen or temperature. The trend of this time lag is captured very well by the simulations, as is the evolving axial distribution of soot, despite the simple soot model employed. Building on the good agreement between spatiotemporal distributions in experiment and simulation, further results from the latter are extracted to provide insight into relevant processes. The advancing soot tip lags behind the fuel–vapor spray tip due to soot oxidation. Tracking the Lagrangian time history of notional fluid particles from the soot onset location back to the injector orifice reveals that their trajectories evolve along rich conditions (φ > 1.5) throughout the entire path. Overall, novel insights obtained from experiments with respect to soot and soot precursor evolutions are complemented by simulations using the integrated conditional moment closure/soot modeling approach, showing encouraging results for prediction and understanding of transient soot processes in high-pressure diesel sprays.