Numerical analysis of laminar burning velocity of hydrogen and carbon monoxide-enriched natural gas (HyCONG) blends

The laminar burning velocities of various HyCONG blends, consisting of hydrogen, carbon monoxide, and natural gas, were measured experimentally and numerically using the outwardly spherical expanding flame technique. Fifteen different combinations of H2, CH4, and CO were analyzed, with five CO blends (0%, 20%, 40%, 60%, and 80%) and three HCNG blends (0%, 20%, and 40%) by volume selected. To evaluate the uncertainty associated with extrapolation models, three models LS (linear model based on stretch rate), NQ (quasi-study nonlinear model), and N3P (nonlinear model with 3 fitting parameters) were used. Four chemical kinetic mechanisms (Aramco 2.0, San-Diego, GRI-Mech 3.0, and USC-Mech 2.0) were used for numerical analysis. The laminar burning velocity increased with increasing CO fraction for a constant H2, CNG (CH4) ratio, and equivalence ratio. The GRI-Mech 3.0 mechanism closely matched the LS-measured values, but the NQ and N3P models also predicted LBV well for some specific blends and equivalence ratios. CO enrichment in the fuel had a significant effect on the laminar burning velocity of the blend and adiabatic temperature, while the H2 fraction also played a noticeable role in the chemical kinetic study. The laminar burning velocity of the HyCONG blends increased with increasing equivalence ratio, which could be explained by the decreasing unburned blend density and increasing H and O radical concentration. The chemical reaction of CO and OH radical had the highest flowrate sensitivity at a high CO fraction HyCONG blend. Individual heat release rate analysis was performed for each type of chemical kinetic mechanism. The highest HRR is found as 14.7*108, 13.7*108, and 18.57*108 J/m3-s, respectively, for 0–80, 20–80, and 40–60 HyCONG at ϕ = 1.0. It is observed that the highest HRR is given by reaction s-R49 and a-R92 which is a similar kind of reaction and not participating in GRI-Mech 3.0 and USC-Mech 2.0. It was found that the best extrapolation model and kinetic mechanism are LS and GRI-Mech 3.0, but there is a possibility to reduce the mechanism for more accurate predictions.