Nitrogen-based emissions in a spark ignition engine fueled with neat NH3, Part I: phenomenological model

Decarbonization of internal combustion engines (ICEs) intensified the interest in ammonia (NH 3 ). This work extends prior works that focused on just giving possible explanations for observed experimental results. Part I presents the development of a novel 0D modeling framework calibrated on thermodynamic parameters (i.e., in-cylinder pressure, volume, heat transfer, etc.) suitable for predicting emissions from a SI engine operating with neat NH 3 , with the focus on investigating the contributions of thermal and fuel bound NO mechanisms to the NH 3 , NO x , and N 2 O engine-out emissions. The model is well suited for the systematic exploration of various engine operating strategies (e.g., the effect of equivalence ratio, spark timing, engine speed, etc.) and, once the root cause of the observed behavior is identified, the model’s sub‑systems can capture key emissions formation metrics by resolving the influence of individual reaction pathways, as will be shown in Part II of this study that focused on the model’s predictive capability. The model provided useful insights into the formation of nitrogen-based emissions when the SI engine is fueled with NH 3 . The sample results presented in this paper showed that the use of numerical simulations developed almost exclusively for and validated with hydrocarbon fuels will provide inaccurate results when modeling NH 3 combustion inside a SI engine. The addition of flame chemistry via a sequence of 0D homogeneous reactors, a reactor that modeled the chemistry inside the pockets of unburned mixture initially trapped in crevices, and a reactor that modeled the chemistry during the exhaust blowdown allowed the detailed tracking of NH 3 chemistry, explained N 2 O formation, and predicted DeNO x effects on the final exhaust composition. Under engine relevant conditions, the model showed that NO production was primarily governed by HNO chemistry and that the conventional assumption of Zeldovich reactions always producing NO was incorrect for NH 3 combustion.