Carbon-Free Fuels for Large Engines: Flame Propagation of Ammonia and Ammonia/Hydrogen Blends

Beschreibung

Abstract

Ammonia is a promising future fuel for reducing greenhouse gas emissions from large engines due to its carbon-free nature and higher volumetric energy density than hydrogen. However, the combustion behavior of ammonia appears challenging because of its low reactivity. Blending with hydrogen or partial cracking of ammonia is a favorable approach to enhance the reactivity of ammonia. As the flame propagation of ammonia and hydrogen differs fundamentally from that of conventional fuels, this thesis aims to provide basic research to gain a comprehensive understanding of the phenomena governing ammonia-based combustion.

A methodology for the systematic characterization of flame propagation is developed. It consists of experiments and simulations on laminar flames, turbulent flames, and engine combustion, which build on each other. For this, optical measurements are performed on a constant volume combustion chamber and an optical engine test rig, and pre-existing measurement data from a single-cylinder engine are used. The experimental data are evaluated for their respective flame propagation measures and examined for their main influence parameters. Numerical zero-/one-dimensional models for laminar burning velocity, turbulent burning velocity, and the heat release rate in engine combustion are preselected and assessed for ammonia and ammonia/hydrogen blends based on theoretical analysis, literature data, and the present experimental results.

The experimental results show a very slow flame propagation under laminar and close to ambient conditions. Under engine conditions, ammonia shows a favorable response to high temperature, high pressure, and high turbulence. Internal combustion engines are thus an ideal application for ammonia as a fuel. Hydrogen blending further enhances the reactivity and turbulence response. The strong turbulence-flame interaction of ammonia and ammonia/hydrogen blends is discussed in detail. Firstly, high turbulent Karlovitz number values enable turbulent eddies to interact with and enter the flame front. Secondly, differential diffusion causes thermodiffusive instabilities for lean fuel/air mixtures.

Recent reaction mechanisms can adequately reproduce the laminar burning velocity of ammonia and ammonia/hydrogen blends. Recent turbulent flame speed models for ammonia capture the dominant influences on turbulent burning velocity. Well-established models fail to do so, therefore simulation models built on these must be reassessed for ammonia and ammonia/hydrogen blends.