A computational and experimental study of the effects of thermal boundary layers and negative coefficient chemistry on propane ignition delay times

Abstract

Experimental measurements of low-temperature ignition often exhibit higher uncertainties and scatter compared with ignition measurements at higher temperatures. One source of experimental variability and uncertainty is the effect of negative temperature coefficient (NTC) chemistry in thermal boundary layers at low-temperature conditions. The objective of the current work was to evaluate the effects of NTC chemistry and thermal boundary layers on propane ignition delay times using a combined modeling and experimental approach. A two-zone “balloon” model was developed where the core zone represented the high-temperature region of an experimental facility and an annulus zone represented the cooler thermal boundary-layer region. Independent initial conditions and elementary reaction chemistry was used in each zone. The effects of boundary-layer size were evaluated using different volume fractions (Vcore/Vtotal). At high initial temperatures (Tcore > 1000 K), there was negligible effect of NTC chemistry on the predicted ignition delay time for propane compared with a single-zone model, and the core region dominated the ignition characteristics for all sizes of thermal boundary layer. However, for lower initial temperature conditions (e.g., Tcore = 800 K), the thermal boundary layer was predicted to ignite first and all ignition delay times, regardless of the size of the boundary layer, were faster than the single-zone prediction for 800 K by up to 25%. Based on the results of the modeling study, propane ignition experiments were conducted using two rapid compression facilities (Vcore/Vtotal ≅ 0.9). at core temperatures of 744–1044 K and pressures of 9.8–25.4 atm for stoichiometric propane-air mixtures. While the measured ignition delay times were within the uncertainties of the both the single- and two-zone model predictions, high-speed imaging showed clear transition from ignition initially in the annulus region at lower temperatures to ignition in the core region at higher temperatures. The results of the study provide important new insights into the effects of NTC chemistry on ignition studies and support the use of diagnostics that can resolve core and boundary layer ignition behavior, particularly for experiments conducted in the NTC region. The two-zone model is also a valuable new tool for assessing the impact of NTC chemistry on previous and planned ignition delay time measurements.