A data-driven approach for deducing process parameter influences on N2O selectivity for the Ostwald process

Abstract

Nitrous oxide (N2O) represents one of the key chemicals regarding climate change due to the high global warming potential and significant annual emissions. The largest industrial contribution results from the catalytic oxidation of ammonia to nitric acid, known as the Ostwald process. Successful efforts to reduce the emissions have been made for decades, but mainly focus on secondary and tertiary reduction. However, due to the severe process parameters, only a few publications were able to study the primary reduction – minimizing the formation in the first place – under industrial conditions. Here we present a laboratory setup with an extended range of parameters reflecting and exceeding industrial conditions, while allowing for reproducibility of over 99.9 %. The reactor was successfully employed for process parameter analysis and optimization via a data-driven approach through statistical analysis within design of experiments. Increasing preheating temperature by 200 °C and ammonia volume fraction by 4 % L/L decreased nitrous oxide selectivity by a factor of 2.7 and 3.1 respectively, while an increase in pressure from 2.5 to 4.3 bar promoted laughing gas formation by 38 %. Additionally, the volume flow has been found to be statically insignificant. Furthermore, for the first time the non-linear and interaction effects of all process parameters have been deduced. The interaction between pressure and ammonia concentration, as well as pressure and preheating temperature decreases nitrous oxide formation and can even reverse the primary effect of pressure. These findings pave a path for an alternative way of optimization of the Ostwald process alongside the more common kinetic-based optimizations.