(39h) An Experimental and CFD Based Study of the Pt-Catalyzed NH3-Oxidation in Different Scales

After 100 years the Ostwald process is still the most common way for the industrial manufacturing of nitric acid, which represents the most important nitrogen source for the fertilizer and explosives production. The process is divided into three steps, whereat the first step is the catalytic combustion of ammonia with air towards nitrogen monoxide. Byproducts are N₂ and N₂O whereby the latter is a well-known greenhouse gas. 400.000 tons N₂O are emitted due to the nitric acid production yearly, which is the second largest synthetic source of N₂O after the agriculture.^[1]

As catalyst mainly Pt-alloy gauzes are applied with different compositions and structures. Although the process is known for a long time and is highly optimized nowadays, the knowledge about influencing factors and kinetic details under realistic conditions is small. Reasons for that are the harsh conditions (up to 1000 °C, corrosive atmosphere) and the mass transport limitation of the reaction.^[2]

The main focus of this contribution will be the investigation of the N₂O-selectivity in a lab scale reactor and the prediction of the results by CFD-simulation in consideration of the wall effects caused by heat transfer phenomena.

Methods

The investigations were carried out by using CFD-simulations and a lab scaled setup. For the CFD-simulations the Ansys^® Fluent^® software package was used, which allows to compute the surface reaction kinetics and the fluid-dynamics simultaneously. The use of look-up tables for the surface kinetics reduces the calculating time significantly.^[3] The implemented surface kinetics are based on Krähnert.^[4] The complete lab scale reactor was simulated including heat transfer phenomena.

The lab scaled setup is engineered to work under nearly realistic process conditions (tubular reactors of 1 and 3 cm diameter, woven Pt/Rh-gauzes as catalyst, 4 atm pressure, feed: 10.5 % (L L^-1) NH₃ in air). The temperature reaches up to 1000 K behind the catalyst. To estimate the influence of the wall effects the reaction-zone can be external heated up to 973 K. The gas phase was analysed by a special equipped Online-FTIR.

Results and discussions

The experimental results of the lab scale reactor obtain in all cases higher N₂O-selectivities than the known ones from industrial plants at similar conversions of NH₃.

In this context we introduced CFD-simulations combined with the surface kinetics. It can be shown that the N₂O-selectivity depends on the individual spatial temperature of the gauze. The formation of N₂O predominantly takes place in colder regions. An inhomogeneous radial temperature and concentration distribution along the catalyst gauze can be shown for different reactor diameters.

Accordingly in lab scale setups with small diameters (cm) the portion of cold regions is higher compared to industrial scales (m). Therefore, the area-averaged selectivity depends on the chosen scaling factor.

Further experimental results reveal a decreasing N₂O-selectivity with increasing external heating in the reaction zone. This confirms that the higher formation rate of N₂O in lab scale reactors is an effect of heat transfer phenomena.

Provided CFD-simulation and experiments pointed out that there is a limit in minimizing reactors for the investigation of the ammonia oxidation on Pt-catalysts due to the heat loss through the reactor wall. The diameter of an experimental lab scale reactor has not to be below this limit in order to avoid wall effects and provide feasible N₂O-selectivities.

[1] J. Pérez-Ramirez, F. Kaptejin, K. Schöffel, J. Moulijn, Applied Catalysis B: Environmental, 2003, 44, 117-151.

[2] R. Imbihl, A. Scheibe, Y. F. Zeng, S. Gunther, R. Kraehnert, V. A. Kondratenko, M. Baerns, W. K. Offermans, A. P. J. Janse, R. A. van Santen, Phys. Chem. Chem. Phys., 2007, 9, 3522-3540.

[3] M. Votsmeier, Chem. Eng. Sci., 2009, 64, 1384.

[4] R. Kraehnert, M. Baerns, Chem. Eng. J., 2007, 137, 361-375.