Turbulent burning velocity of hydrogen/n-heptane/air propagating spherical flames: Effects of hydrogen content.

This study investigated the flame propagation and turbulent burning velocity of hydrogen/n-heptane/air mixtures subjected to differential diffusion and flame stretch induced by turbulence at different hydrogen molar fractions (X H2 = 0–100%), equivalence ratios (ϕ = 0.5–1.0) and turbulence intensities (u' = 0–3.62 m/s). The experiments were carried out on a medium-scale, fan-stirred cylindrical combustion chamber generating near-isotropic turbulence. Results show that as hydrogen content increases, the turbulent flame propagation speed tends to be affected by stronger differential diffusion and weakened turbulence, characterized by decreasing Lewis number (Le) and Karlovitz number (Ka) within a similar range of turbulence intensity. Among them, the effect of differential diffusion on the normalized turbulent propagation speed (d < r > / d t σ S L ) appears to be well correlated with Re T,flame Le -2, where Re T,flame represents the turbulent flame Reynolds number. However, inconsistencies in the fitting parameters for this scale are observed at different X H2 , highlighting the impact of substantial variations in flame stretch induced by turbulence. Therefore, the normalized turbulent flame propagation velocity under varying hydrogen contents is proposed to be accurately predicted by d < r > / d t σ S L -1 = 0.07 K a 0.32 (R e T , flame L e − 2) 0.5 . The synergistic effects of Le and Ka can also account for the peak phenomenon in the normalized turbulent burning velocity (S T / S L) observed at 85% hydrogen content. Furthermore, the quantitative characterization of turbulence and differential diffusion under extensive experimental data (Ka = 0.03–90) is studied. As evidenced by the Damköhler theoretical hypothesis, the two-stage trend between S T / S L and (u' / S L) Le -1 is identified with a transition near unity Ka. The modified correlations for turbulent burning velocity in terms of (u' / S L) Le -1 and Ka 0.2 are obtained with consideration of the differential diffusion and flame stretch rate induced by turbulence, which show good predictive performance for data at different turbulence regime diagrams regardless of fuel type and equipment size. [ABSTRACT FROM AUTHOR]