Kinetic effects of non-equilibrium plasma-assisted methane oxidation on diffusion flame extinction limits
Wenting Sun, Mruthunjaya Uddi, Sang Hee Won, Timothy Ombrello, Campbell Carter, Yiguang Ju
Published 16 August 2011
The kinetic effects of low temperature non-equilibrium plasma assisted CH4 oxidation on the extinction of partially premixed methane flames was studied at 60 Torr by blending 2% CH4 by volume into the oxidizer stream of a counterflow system. The experiments showed that non-equilibrium plasma can dramatically accelerate the CH4 oxidation at low temperature. The rapid CH4 oxidation via plasma assisted combustion resulted in fast chemical heat release and extended the extinction limits significantly. Furthermore, experimental results showed that partial fuel mixing in the oxidizer stream led to a dramatic decrease of O concentration due to its rapid consumption by CH4 oxidation at low temperature. The products of plasma assisted CH4 oxidation were measured using the Two-photon Absorption Laser-Induced Fluorescence (TALIF) method (for atomic oxygen, O), Fourier Transform Infrared (FTIR) spectroscopy, and Gas Chromatography (GC). The product concentrations were used to validate the plasma assisted combustion kinetic model. The comparisons showed the kinetic model over-predicted the CO, H2O and H2 concentrations and under-predicted CO2 concentration. A path flux analysis showed that O generated by the plasma was the critical species for extinction enhancement. In addition, the results showed that O was produced mainly by direct electron impact dissociation reactions and the collisional dissociation reactions of electronically excited molecules with O2. Moreover, these reactions involving electron impact and excited species collisional dissociation of CH4 contributed approximately a mole fraction of 0.1 of total radical production. The present experiments produced quantitative species and extinction data of low temperature plasma assisted combustion to constrain the uncertainties in plasma/flame kinetic models.