The recently proposed RapreNOx process involves the injection of HNCO into combustion effluents in order to reduce their NO content via what was initially assumed to be a homogeneous gas-phase reaction. Although the process is thus similar to the Thermal DeNOx process it had the apparent advantage of operating at temperatures as low as 450°C, as contrasted with the 700°C minimum temperature of Thermal DeNOx. Further, the Thermal DeNOx process had the disadvantage that downstream of the reaction zone unreacted ammonia could react with SO3 to form NH4HSO4 and cause fouling problems, a difficulty thart using HNCO would seem to avoid. In this reexamination of the RapreNOx process we show that the process involves three different modes of NO reduction. The first is catalytic; NO reductions at temperatures significantly below 700°C are found not to occur in the absence of catalytic surfaces. While noncatalytic NO reductions were found to occur at 700 °C in the presence of wet CO, NH3 will also reduce NO at these conditions. Modeling calculations indicate that for this mode of reaction HNCO reduces NO via a complex chain reaction mechanism very similar to that involved in the Thermal DeNOx process. Within this reaction mechanism the interconversion of HNCO and NH3 is sufficiently rapid that the amounts of HNCO and NH3 left at the end of reaction are roughly independent of whether one starts with HNCO or NH3. The modeling calculations also indicate the existence of a third mode of NO reduction by HNCO. In this third mode OH formed by thermal dissociation of H2O reacts with HNCO to form the NCO radical, which reduces NO to N2O. This third mode of NO reduction is predicted to occur at temperatures substantially higher than those used in the Thermal DeNOx process and it is thus potentially suitable for limiting NOx emissions from gas turbines, but the N2O produced in this mode may be a significant disadvantage.
