A methodology comprising careful consideration of sample preparation, reactor design, experimental procedures and data evaluation routines for precise analysis of the kinetics of gas-solid reactions, specifically the oxidation of carbonaceous materials, has been developed and validated. The well-controlled solvent-free deposition of the carbonaceous material onto cordierite monolith substrates ensures experimental studies in the absence of diffusion limitations, temperature gradients and hot zones. These critical aspects are supported also by theoretical considerations. Temperature-programmed oxidation and isothermal oxygen stepresponse experiments in continuous gas-flow reactor using a homogeneous synthetic carbonblack material demonstrate excellent reproducibility and the conversion profiles agree well with previously reported data. An independent set of global kinetic parameters was estimated ⇤ To whom correspondence should be addressed 1 for each 5% sub-conversion interval using linear regression such that the conversion dependence of each parameter could be analyzed separately and compared to previously published data. The results show that the evolution of reactive carbons cannot be described with a single global reaction order. This is supported by intermittent ex situ measurements of the specific surface area of the carbon-black material during the course of isothermal oxidation, which reveals a developing microporous structure at high conversions. Physically the changes in carbon reaction order are interpreted as changes in fraction of accessible reactive carbon atoms during progressing oxidation. Moreover, at high conversions, the carbon reaction order approaches 0.7 implying that the evolution of the concentration of reactive carbon atoms is not only proportional to the external surface area of shrinking spheres but also that these spheres have approximately the same size.

In this study, an SCR-coated particulate filter with soot loaded from engine bench experiments was evaluated. Prior to soot loading, the sample was hydro-thermally aged at 850 °C. Flow reactor measurements were performed under various reaction conditions to examine the impact of soot in an SCR-coated diesel particulate filter (DPF) on standard SCR, fast SCR, NH 3 oxidation, and NO oxidation. In the presence of soot, NO x conversion was slightly lower at 200−300 °C due to the blocking of active sites. However, at higher temperatures, the NO x conversion was somewhat higher with soot, indicating that soot more strongly inhibited the oxidation of NH 3 than the SCR reaction. When feeding equal amounts of NO and NO 2 together with NH 3 , ammonium nitrates on the sample were formed. The presence of soot significantly decreased the formation or the stability of ammonium nitrate, resulting in higher conversion with soot. To further understand this interesting aspect, ammonium nitrate formation and decomposition experiments using model Cu/BEA catalyst with and without soot were performed. It was observed that less N 2 O was formed in the presence of soot, and this in combination with CO x formation during the ammonium nitrate formation step suggested that soot reacted with the ammonium nitrates, and we propose that this occurs on CuO species on the outside of the zeolite particles.

No abstract

This study demonstrates the application of two different methods to evaluate the kinetics of O 2 -and NO 2 -based oxidation of carbonaceous matter. The influence of reactor setup and experiment execution on the accuracy and interpretation of the kinetic parameters is also discussed. The fundamental difference between the two methods lies in how rate changes during the progress of oxidation are interpreted: by changes in the fraction of atoms available for reaction (traditional approach) or by changes in the activation energy (unconventional approach). Using the traditional approach, it was found that two parallel reactions with a first-order reaction model and a third-order Avrami− Erofeyev model could accurately reproduce the O 2 -based oxidation curves. The same model (with a different set of parameter values) could also describe the NO 2 -based kinetics. The unconventional method also provided an accurate representation of the data and can thus function as a complementary evaluation method.

A laboratory reactor for kinetic studies has been simulated using computational fluid dynamics (CFD). Analysis of temperature distribution within the system shows that adding an inert monolith upstream the catalyst enhances heat conduction and therefore significantly reduces radial temperature gradients. Adjustment of the heating coil plays an important role as well by allowing the gas phase to smooth out the radial temperature profile. Radiative heat transfer and its effects on both the heat losses from the catalyst and on the measurements with an unprotected thermocouple have been particularly investigated. In order to prevent both the falsifying effect of radiation and the influences from the ongoing reaction, thermocouples should be placed and shielded inside a clogged channel. An inert monolith that is placed downstream the catalyst effectively serves as a radiation shield and drastically reduces both axial and radial gradients. Studies of the dispersion in the system reveal that the FTIR's gas cell is the most important source of the overall broadening in the concentration signal, with the reactor tube being the next major source of the distortion. An algorithm based on the Tikhonov regularization method has been developed for calculating the deconvolution of the concentration data, which has refined the time‐resolution in transient experiments from 20 to 2 s.