Laminar flame speed of ammonia-oxygen and ammonia-oxygen enriched air mixtures near saturation conditions

As ammonia appears as one of the most promising renewable energy carriers of the future, it is of critical importance to fill the gaps in the understanding of its combustion properties in the perspective of developing real industrial applications. With the objective to contribute to the complete understanding of ammonia combustion properties, this works analyzes the effect on laminar flame speed of modeling ammonia-oxygen and ammonia-enriched air as a mixtures of real gas. Results show quantitative and qualitative discrepancies with ideal gas modeling when increasing the mixtures temperature and pressure along the saturation curve of ammonia. Results indicate that, for near-ambient temperature, real gas effects should be included in the uncertainty calculation for pressure as low as approximately 350 kPa for ammonia-oxygen mixtures and 700 kPa for ammonia-oxygen enriched air mixtures. Given high pressure and temperature conditions, discrepancies up to 29% in the laminar flame speed are obtained. Moreover, the laminar flame speed demonstrates different qualitative evolution when changing the gas models; it decreases/increases when P > 1 . 062 MPa and T > 275 K for real gas/ideal gas models. The main driver is the inclusion of the real gas equation of state and thermodynamics, while high-pressure transport models and kinetics are less important. Real gas effects should therefore be included in future analyses for the correct assessment of ammonia combustion properties, mostly for high-pressure industrial applications. Novelty and significance statement The novelty of this research is to include real gas (RG) modeling to compute ammonia-based mixtures laminar flame speed (LFS). It is significant because (i) ammonia is a promising energy carrier of renewable energy, and (ii) modeling ammonia-based mixtures as a RG induces major changes in LFS predictions along the saturation curve of ammonia. Given high pressure and temperature, RG mixtures yield a LFS up to 29% lower than obtained when considering the ideal gas model. Moreover, the qualitative trends of LFS along the saturation curve diverge between the two models. The main RG effects are shown to be related to the modification of the equation of state and associated thermodynamics. We conclude that RG modeling should be in general included for simulating the combustion of ammonia-based mixtures, which is an outcome of primary importance for the community.