Computational fluid dynamics investigation of alternative nitric oxide emission mechanisms in a hydrogen-fueled spark-ignition engine

• CFD investigation in a hydrogen-fueled, SI engine operating in variable load/equivalence ratio.
• Investigation of three NO emissions mechanisms: thermal NO, NNH route, and N2O path.
• Thermal NO mechanism is dominant at medium/high engine loads.
• The NNH route becomes important at low loads (lean and low-temperature flame).
• The N2O route has negligible effects on calculated NO emissions.

In the present work, the reaction mechanisms of nitric oxide (NO) are investigated in a hydrogen-fueled, spark-ignition engine during load variation by mixture leaning. As these exhaust emissions are practically the only ones emitted from such engines, there is an obvious vivid interest for their reliable prediction under different operating conditions and modes, in order to investigate the potential future use of hydrogen in engines. For that purpose, a 3D-CFD (three dimensional – computational fluid dynamics) code is applied, which has been developed by the authors and validated extensively in the past under both motoring and firing conditions, being capable to simulate in high detail the in–cylinder processes. The CFD simulation has been accomplished using different sets of constants of the thermal NO mechanism, widely known as “Zeldovich mechanism”, to provide insight into the NO production pattern at each location in the cylinder during the combustion and expansion periods. The NO emission modeling has been enhanced here with the addition of alternative production paths, such as via the NNH and N2O species formation. The NNH path has been shown to be favored at lean and low-temperature combustion conditions for hydrogen flames, whereas the N2O path becomes important for lean flames irrespectively of the fuel used. The calculations are compared with available measurements, in order to quantify the applicability of the different sets of constants and the use of these alternative production paths at such conditions and applications. For the high load cases the NO predictions have good accuracy, since the constants used have been mainly derived at such high temperature conditions where additionally the thermal NO is almost the exclusive production path. On the contrary at mid and low engine loads, when the combustion temperature is lower and leaner mixtures are used, a significant discrepancy exists between the calculations and the measured data, which is however improved especially when the NNH route is taken into consideration.