Direct Numerical Simulation of Premixed Turbulent Ammonia Flames

  • Handle-Kesselring, Paula-Maria (Graz University of Technology)
  • Gößnitzer, Clemens (Large Engines Competence Center GmbH)
  • Posch, Stefan (Graz University of Technology)
  • Steiner, Helfried (Graz University of Technology)

Please login to view abstract download link

Ammonia is a promising carbon-free energy carrier for large-bore engines. However, its turbulent pre-mixed flame behavior is not sufficiently understood, particularly during ignition and early stages of flame propagation. Direct Numerical Simulations (DNS) are therefore performed to investigate premixed ammonia flames propagated in large-scale turbulence representative of large-bore engines, where turbulence is characterized by correspondingly large integral length scales of Lt ≈ 10mm and higher and moderate to high turbulent Reynolds numbers Ret ≈ 100 ÷ 1000. Most recently, Hoflack et al. [1] investigated a similar premixed ammonia flame, however considering the regime of small scale turbulence structures and relatively low turbulent Reynolds numbers. The present simulations are conducted with the finite-volume, low Mach number code PeleLMeX [2] and the POLIMI2023 reaction mechanism [3]. The turbulent flow inside a three-dimensional (3D) cubic domain representative of a real engine ignition region, is initialized with a homogeneous, isotropic turbulent (HIT) velocity field with a turbulent Reynolds number Ret = 630. The HIT field is generated using the von K´arm´an–Pao energy spectrum and sustained throughout the simulation by an ABC forcing approach [4]. Ignition is modeled by a volumetric energy source depositing energy at the center of the domain to mimic realistic ignition characteristics. Based on the obtained predictions, the present work assesses the effect of turbulence eddies on the initial ammonia flame propagation. The findings provide fundamental insights into the flame–turbulence interactions of ammonia and support the development, calibration, and validation of predictive combustion models for carbon-free large-bore engine combustion applications.