NH3 oxidation by NO2 in a jet-stirred reactor: The effect of significant uncertainties in H2NO kinetics

Abstract

Understanding the kinetics of ammonia (NH3) is becoming increasingly important to a growing variety of applications — ranging from its role as a NOx reduction agent, a key intermediate during combustion of biomass and energetic materials (especially green propellants), and a potential carbon-free energy carrier and storage medium. This wide variety of applications calls for comprehensive NH3 kinetic models that are reliable over wide ranges of temperatures, pressures, and mixtures. Yet, many still consider the present understanding of its kinetics to be incomplete. For example, there are few experimental studies of NH3 oxidation by nitrogen-containing species, which offer the opportunity to probe relatively untested reactions (or combinations thereof) to enable a more comprehensive understanding of NH3 kinetics. To address this gap, we perform jet-stirred reactor experiments of NH3 oxidation by NO2 over an intermediate temperature range (700–1100 K). The mole fractions of NH3, NO2, NO, and O2 are measured through a combination of gas chromatography, chemiluminescence, and infrared absorption. Agreement among different diagnostics (≤4% for NH3 and ≤7% for NO2) and excellent experimental repeatability ensure high confidence in all species measurements. Comparisons of species measurements to model predictions revealed deficiencies in recent kinetic models, particularly for NH3 consumption and NO formation at elevated temperatures (≥900 K). Uncertainty-weighted kinetic analyses point to the importance of reactions that form (NH2 + NO2) and consume (H2NO + NO2, H2NO + OH) H2NO, both of which are uncertain and influential in this system (and many other NH3 oxidation systems). These and other reactions accentuated in the present dataset are also key reactions in NH3/air ignition and N2O formation, both of which remain outstanding challenges for NH3 combustion in engines. Consequently, resolving the modeling deficiencies observed for the present dataset appears especially important to predictive models to enable the use of NH3 as a fuel.