Ignition features of methane and ethylene fuel-blends in hot and diluted coflows. (1st September 2017)

• Record Type: Journal Article
• Authors: Evans, M.J.; Chinnici, A.; Medwell, P.R.; Ye, J.
• Is Part Of: Fuel. Volume 203(2017)
• Page Start: 279
• Page End: 289
• Publication Date: 2017-09-01
• Subjects: MILD combustion -- Non-premixed flames -- Lifted flames -- Autoignition -- Fuel blends; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2017.04.113
• Languages: English
• ISSNs: 0016-2361

Abstract

Abstract: Turbulent flames in the moderate or intense low oxygen dilution (MILD) combustion regime have previously exhibited less susceptibility to lift-off than conventional autoignitive flames in a jet-in-hot-coflow (JHC) burner. This has been demonstrated through laser-based diagnostics and examination of CH * chemiluminescence. New experimental observations are presented of turbulent flames of natural gas, ethylene and blends of the two fuels, in coflows with temperatures from 1250–1385 K and oxygen concentrations from 3–11% (vol./vol.). Zero- and one-dimensional simulations, as well as turbulent flame modelling, are used to explain the trends seen experimentally with different coflows and fuels. Numerical simulations using simplified batch reactors and opposed-flow flames demonstrate that blending of methane and ethylene fuels is most significant near 1100 K. Near this temperature, pure ethylene exhibits a transition between high and low temperature ignition pathways. Further analyses show that a 1:1 methane/ethylene blend behaves more like ethylene near MILD combustion conditions, and more like methane in conventional autoignition conditions. Two-dimensional modelling results of the turbulent flames are then discussed and explained in the context of the simplified reactor results. The flames confined by the lean flammability-limit in the coflow and high strain-rates in jet shear layer, in agreement with previous work using a semi-empirical jet model. The two-dimensional modelling is additionally able to qualitatively replicate the trends in lift-off height, with normalised heat release rate profiles reproducing, and serving to explain, the effects seen in experimental campaigns.