This graduation thesis studies the CI combustion of ammonia and hydrogen in an ICE. It contains a review of the literature and a modeling study of the ignition and cylinder performance of the AmmoniaDrive test engine. AmmoniaDrive is a NWO-funded research project that aims to decarbonize shipping by introducing an ammonia-fuelled SOFC-ICE power plant for ships and other heavy-duty applications.
Ammonia (NH3) has unfavorable properties for combustion, such as a high heat of vaporization, narrow flammability limits, a high flame quenching distance, low flame speed, and most important: a high resistance to autoignition. Those properties have to be overcome and hydrogen (H2) is known to be capable of playing a role. As the properties of hydrogen are extremely high combustion speed, very wide flammability limits and a very short flame quenching distance. Unfortunately, both ammonia and hydrogen have a high resistance to autoignition, while CI engines need a fuel with low resistance to auto-ignition. For this reason, a carbon-based fuel, like DME or HVO, is considered necessary to achieve ignition.
The state-of-the-art experimentally achieved combustion concepts are an homogeneous charge compression ignition (HCCI) combustion concept of Pochet et al. [2020a] and an reactivity controlled compression ignition (RCCI) combustion concept of Chiera et al. [2022]. The HCCI combustion concept is fueled by NH3 and H2, without a carbon-based fuel. However, it requires a compression ratio (CR) of 22, a high intake temperature, and is limited by the maximum pressure rise rate (MPRR). The RCCI combustion concept is fueled by NH3 and diesel. This concept can achieve up to 81%e NH3, but still requires 19%e diesel.
A modeling study is done to investigate how to improve the CI combustion strategies, taking into consideration the context of AmmoniaDrive. The modeling study consists of two closed volume single-zone thermodynamic reactor models: the ignition model and the engine cylinder model. The ignition model is a constant volume model, resembling top dead center (TDC) conditions. The engine cylinder model simulates a closed volume from bottom dead center (BDC) to 90 CAD after TDC and incorporates volume change and the heat loss. Bothmodels make use of the chemical kinetic mechanism of Shrestha et al. [2018] to incorporate the combustion reaction. Due to the limitation imposed by the available species in chemical kinetic mechanisms, the carbon-based fuel in the modeling study is DME. For the future, hydrotreated vegetable oil (HVO) seems a more favorable carbon-based fuel, based on experimental results in a constant volume combustion chamber (CVCC) of Hernandez et al. [2023].
The results of the two models indicate that an HCCI combustion concept of ammonia and hydrogen, without a carbon-based fuel, will not ignite within the engine limits of the AmmoniaDrive test engine. An HCCI combustion concept of ammonia, hydrogen, with DME will ignite, but has a limited power output due to the MPRR. An RCCI combustion concept with stratification of DME throughout the cylinder looks promising based on the engine cylinder model results. Stratifying DME concentration, and with that the fuel reactivity, is likely to reduce the MPRR. This would allow for a higher power output due to the possibility to injectmore fuel energy without exceeding the engine limits.
Combining the literature and modeling results, it is likely that an RCCI combustion concept with ammonia, hydrogen, and HVO will lead to a higher power output and a decreased required amount of carbon-based fuel, whilst staying within engine limits.