The influence of radiation modeling on flame characteristics and emissions prediction in microcombustors with Ammonia/Hydrogen blends

This study advances numerical methodologies for designing microcombustor-based thermophotovoltaic systems by incorporating comprehensive analyses of radiation models and absorption coefficients. It examines the impact of non-premixed hydrogen-ammonia-air fuel blends on the micro combustion process using three-dimensional CFD simulations on a Y-shaped microcombustor under atmospheric conditions. Key to this research is modeling radiative heat transfer from multiple species, including ammonia (NH3), nitric oxide (NO), and water vapor, ensuring accurate predictions and optimization of thermal behavior and efficiency. Three modeling approaches were compared: a multispecies Planck-mean approximation radiation model (PMAC), a narrowband model for water vapor, and a scenario excluding radiation effects. The PMAC model, incorporating the absorption coefficients of ammonia and its products, consistently produced lower errors in flame length compared to the NB method across all equivalence ratios, with errors ranging from 0% to 7% for PMAC and 10 % to 30 % for NB, with the lowest error observed at Phi =1.0. Additionally, the prediction of laminar burning velocity (LBV) differed by approximately 4 % between the two models. Considering different fuel blends, two scenarios were compared: constant input thermal power and constant fuel flow rate. Efficiency in the constant input thermal power scenario maintained higher radiation efficiency, in the range of 32-38%. Additionally, radiation efficiency initially increased with increasing H2 content but subsequently decreased for high H2 values due to effects on flame area, flame position, and temperature.