Spectroscopic study of CO formation from CO2-enriched pyrolysis of C2H6 and C3H8 under engine-relevant conditions

Abstract

To address the challenges of climate change, technologies and processes that contribute to reducing net CO2 emissions are key. Herein, engine-based dry reforming combines the possibility of storing excess energy and converting unwanted CO2 into syngas. To fundamentally investigate this process at simpler, but engine-like conditions, the pyrolysis of CO2/C2H6 and CO2/C3H8 mixtures behind reflected shock waves was studied. The targeted conditions were especially temperatures of 1830 – 2590 K, as they are encountered in piston engines, and atmospheric pressure. The time-resolved CO formation was measured using a quantum cascade laser, providing a unique experimental dataset. In addition, laser absorption data at temperatures above 2500 K reveal physically unattainable CO mole fractions, so these experiments are discussed separately. This phenomenon is shown and briefly discussed.A detailed chemical kinetics analysis reveals the interaction of linear alkanes and CO2, and the influence of the respective linear alkane on CO formation. The decomposition of all linear alkanes leads to radical formation, initiating CO2 decomposition via, e.g., CO2 + H ⇌ CO + OH and CH2(S) + CO2 ⇌ CH2O + CO. Overall, the C2H6/CO2 blends exhibit smaller τ20 (the time at which 20% of maximum CO mole fraction is reached based on the atom balance) since the C-C chain cleavage of C2H6 decomposes to CH3, enhancing CH2(S) formation and also the subsequent enhanced C1-oxidation path. In contrast, the C-C-C chain of C3H8 leads to C2H5 and CH3 and to faster H-radical formation via C2H5 (+M) ⇌ H + C2H4 (+M). As C2H2 is one of the key species with respect to the high-temperature pyrolysis of alkanes, its formation and decomposition has a great influence on the whole process because its respective reactions compete for H-radicals, also needed for CO2 decomposition. A comparison was made with results predicted from literature reaction mechanisms. It was reported that all tested models need improvement, underlining the limitation of chemical kinetics mechanisms and validation data at untypical conditions.