(119c) Concentration and Temperature Uniformity during Catalytic Experiments: A CFD Analysis of the Berty Reactor

Scott D. Anderson¹, Bjarne Kreitz², Prof. Dr.-Ing. Thomas Turek¹

and Prof. Dr.-Ing. Gregor D. Wehinger¹

¹Institute of Chemical and Electrochemical Process Engineering,

Clausthal University of Technology, Clausthal-Zellerfeld, Germany

²School of Engineering, Brown University, Providence, RI, USA

Approximately 90 % of processes in the chemical industries rely on heterogeneous catalysis [1]. Consequentially, knowledge about the transport processes and reaction kinetics involved is of crucial importance for chemical engineers. Kinetic measurements are commonly conducted under differential conditions, allowing for easier extraction of reaction rates from end-of-pipe measurements. At integral conditions, axial profiles are recorded and detailed reactor models are necessary.

To overcome the drawbacks of the classical packed-bed reactor (PBR) continuously stirred tank reactors (CSTR) are used for catalytic measurements, as non-uniform temperature and concentration within laboratory reactors pose a challenge for derivation of kinetic models from catalytic measurements [2]. Two basic types of CSTR can be distinguished, where either the catalyst is placed in a rotating basket (Carberry reactor) or where the catalyst is fixed and a turbine circulates the gas-phase internally (Berty reactor) [3,4]. The Berty reactor has been characterized via experimental methods [5-7]. While these studies have confirmed its concentration uniformity via the residence time distribution (RTD) experimentally, little is known about the temperature distribution inside of the Berty reactor and very few studies have investigated it via Computational Fluid Dynamics (CFD) simulations [8].

The aim of this work is to bridge this gap of knowledge by investigating the temperature distribution inside a state-of-the-art Berty reactor by usage of CFD simulations to resolve flow, concentration, and temperature fields, respectively.

Therefore, a CFD model is established using Siemens SimCenter STAR-CCM+. Simulations of residence time behavior are performed using the Passive Scalar model and compared against experimental RTD data for model validation purposes. Further, global kinetics for the methanation of CO₂ over nickel-alumina catalysts are introduced into the CFD model [9]. Simulations of the strongly exothermic methanation of CO₂ are performed and the influence of operating parameters on temperature and concentration uniformity are assessed. The influence of relevant operating parameters (pressure, temperature, catalyst dilution, feed rate, stoichiometric ratio, rotation rate) is investigated via parameter variation. From an experimentalists point of view, uniform conditions in the catalyst bed that resemble the average reactor conditions are desirable. Therefore, the mean and maximum deviation of temperature and concentration in the catalyst packed bed with respect to the gas phase were computed to assess the reactors homogeneity.

Results have shown that the conversion per pass of the catalyst bed plays a determining role in the formation of temperature and concentration deviations. High conversion per pass can lead to the release of large amounts of reaction heat, causing pronounced non-isothermal temperature profiles. Furthermore, heterogeneous conditions for the concentration profile may arise, as well. It is therefore paramount for catalytic experiments to keep the conversion per pass low, which can be achieved either via dilution of the catalyst bed or the reactor recycle ratio.

The recycle ratio is a function of turbine rotation speed, operating temperature and pressure. Increasing the turbine speed leads to higher recycle ratios. Low temperature and high pressure result in high gas density. The Berty reactor turbine can transfer more energy to the gas phase at higher densities, hence increasing the recycle ratio. Consequentially, it is especially important to ensure isothermicity when conducting experiments at low pressures.

Comparing these results with findings from experimental investigations as well as the residence time simulations reveals disparities. Even at operating conditions where the RTD confirmed ideal CSTR behavior, local hot spots of up to 20 K were found, indicating that RTD experiments alone are insufficient to ensure homogeneous conditions for catalytic experiments.

This study provides the first investigation of temperature and concentration profiles inside a Berty reactor under reaction conditions. It was demonstrated how common practice of using RTD experiments for confirmation of ideal CSTR behavior can lead to erroneous conclusions and subsequently false predictions of kinetic models derived under heterogeneous reaction conditions. CFD simulations can however assist the chemical engineer with valuable insights.

References

[1] National Research Council. Catalysis Looks to the Future. Washington DC., The National Academies Press (1992).

[2] Kapteijn, F. et al. in: “Handbook of Heterogeneous Catalysis” pp 1189–1261. John Wiley & Sons, Ltd, 1997.

[3] Berty, J. Catal. Rev. Sci. Eng. 20, 75–96, (1979).

[4] Carberry, J. J. Ind. Eng. Chem. 56(11), 39-46, (1964).

[5] Haumann, M. et al. J. Catal. 263, 321–327, (2009).

[6] Al-Saleh, M. et al. Chem. Eng. J. 37(1), 35-41, (1988).

[7] Ountaksinkul, K. et al. RSC Adv. 10(60), 36667-36677, (2020).

[8] Ountaksinkul, K. et al. React. Chem. Eng.
https://doi.org/10.1039/D1RE00390A (2021).

[9] Koschany, F et al. Appl. Catal. B, 181, 504-516, (2016).