Quantitative laser-induced fluorescence of NO in ammonia-hydrogen-nitrogen turbulent jet flames at elevated pressure

Abstract

Ammonia is a very promising carbon-free fuel, but its combustion is prone to generate a large amount of harmful nitric oxide (NO). Designing NO reduction strategies for ammonia flames requires computational fluid dynamics and accurate kinetic mechanisms. However, there are currently no available experimental data that can be used to validate models describing chemistry-turbulence interactions in ammonia flames. This study introduces two non-premixed turbulent jet flames that emulate some features of the cracked ammonia combustion at 5 bar, relevant to micro gas turbines. These ammonia-hydrogen-nitrogen jet flames feature well-controlled boundary conditions and are particularly amenable to modeling. A one-dimensional NO laser-induced fluorescence (NO-LIF) method was implemented and combined with 1-D Raman spectroscopy to measure the NO mole fraction quantitatively. To avoid laser absorption by ammonia, excitation in the A2 + − X 2 (0–1) band near 236 nm was chosen instead of the more conventional (0–0) band near 226 nm. Due to the unsteady nature of turbulent jet flames, offline NO-LIF measurements are not useful, and undesirable interferences from Rayleigh scattering and oxygen LIF were instead quantified and removed using the temperature and major species mole fractions measured with Raman spectroscopy. NO quantification algorithms were optimized and validated first with a laminar NH3–H2–N2 counterflow flame, and then applied to the turbulent jet flames. Results show that a large amount of NO (∼1500 ppm) is produced in the NH3–H2–N2 jet flames. Increasing the ammonia cracking ratio from 14% to 28% reduces the NO concentration in laminar and turbulent NH3–H2–N2 flames. Data also shows that turbulence smear effects of differential diffusion. To the best of our knowledge, this is the first available database featuring quantitative measurements of spatially-resolved NO mole fraction in turbulent NH3–H2–N2 flames at a practically-relevant pressure. This unique database can be used in the future to validate turbulent combustion models for such flames.