A Direct Numerical Simulation Study on Combustion and NO Formation of Coal/ammonia Co-Firing Flames

Jiangkuan Xing,Kun Luo,Ryoichi Kurose
DOI: https://doi.org/10.1016/j.apt.2024.104484
IF: 4.969
2024-01-01
Advanced Powder Technology
Abstract:Coal/ammonia co-firing (CACF) is thought to be a promising economic and clean combustion technology. This paper reports a three-dimensional (3D) point-particle direct numerical simulation (PP-DNS) work of CACF flames in a temporally evolving mixing layer using detailed chemistry. Two co-firing ratios of ammonia, 10% and 20%, are considered, and a pure coal flame is also calculated for comparison to highlight the effect of ammonia co-firing on the flame ignition and NO formation pathway. Zero-dimensional (0D) auto-ignition calculations are also conducted to access auto-ignition behavior of ammonia and coal volatile under different mixing and co-firing levels. The 0D calculations indicate that co-firing ammonia significantly increases the ignition delay by several orders. This increase trend is still valid in the 3D PPDNS flames but much milder, since the devolatilization delay dominates the CACF flame ignition. Three stages, inert mixing stage (IMS), (ii) pre-ignition stage (PIS), and (iii) ignition/propagation stage (IPS), are found during the flame ignition. The devolatilization delay plays a significant role in the IHS and PIS, and the auto-ignition delay dominated the PIS. As ammonia fraction increases, those two delays obviously increase, synthetically leading to the delay of flame ignition. Compared with the pure coal flame, NO formation is enhanced in the CACF flames, and the peak NO mass fraction forms at regions with intermediate A values and mixture fraction between the stoichiometric Zvol and ZNH3, caused by a combined effect of co-firing ammonia on flame temperature and nitrogen source. In addition, the reaction pathway analyses also highlight the difference of conversion pathways of nitrogen in the pure coal and CACF flames. (c) 2024 The Society of Powder Technology Japan. Published by Elsevier B.V. and The Society of Powder Technology Japan. All rights are reserved, including those for text and data mining, AI training, and similar technologies.
What problem does this paper attempt to address?