(349e) Efficient Coupling Between Microkinetic Modeling and CFD: Towards a Fully First-Principles Catalytic Chemical Reaction Engineering

Central to the quest towards an atomic-scale understanding of a catalytic process is the identification of the dominant reaction mechanism that establishes under a particular set of operating conditions. In essence, the dominant reaction mechanism consists of the chemical pathways and intermediate species through which the reactants convert into the products at specific conditions. These dominant chemical pathways are the result of the interplay among a much larger number of chemical events that can potentially occur at the catalyst surface at specific conditions of temperature, pressure and composition. These operating conditions are dictated at the reactor scale by the transport phenomena of mass, energy and momentum. Therefore, the dominant reaction mechanism turns out to be the result of different phenomena occurring at different time and length scales (from the catalyst to the reactor). In this view, it is crucial to adopt a first principles approach, i.e., based on the fundamental governing equations at each scale (e.g., electronic structure theories calculations at the microscale and Navier-Stokes equations at the macroscale). Given the length and time scales involved, along with the potential huge number of chemical events that can occur at the catalyst surface and in the gas-phase, the identification of the dominant reaction mechanism is a very challenging problem, especially for processes of real technological interest. In particular, it is of particular relevance to develop efficient methodologies to link insights across all the relevant time and length scales in one multiscale simulation. Unfortunately, the numerical simulation of multidimensional systems with realistic kinetics places severe demands on computational resources. Beside the dimension of the system, the main difficulty arises from the stiffness of the problem (especially when the energy balance is solved) and the strong coupling between the scales. Conventional CFD methods cannot be efficiently applied in this context: (i) segregated algorithms have serious difficulties to treat the stiffness and the high non-linearities of the equations; (ii) fully coupled methods can be applied only to relatively small systems. In order to overcome these problems, in this contribution we have proposed and implemented a technique based on the operator-splitting approach [1, 2]. The advantage is that it usually avoids many costly matrix operations (typical of fully coupled algorithms) and allows the best numerical method to be used for each type of term or process. The resulting new solver CatalyticFOAM [3] – implemented in the OpenFOAM® framework [4] –  is then able to perform CFD simulations for very general geometries (both in laminar and turbulent conditions) along with a microkinetic description of the surface reactivity. This tool allows for a fully first-principles approach to catalytic chemical reaction engineering, from the microscale (microkinetic models based on electronic structure theories calculations) to the macroscale (computational solution of the Navier-Stokes and transport equations). CatalyticFOAM allows for the solution of reacting flows – based on microkinetic descriptions – in complex geometries. In particular, CHEMKIN and UBI format [5], along with classical rate equations, can be read and coupled with gas-phase reactions (CHEMKIN format). Moreover, tools for post-processing analysis, such as identification of Rate Determining Steps and reaction path analysis, have been implemented. Show-cases for structured and random packing will be presented in the context of the CH₄ Partial oxidation on Rh. Overall, our approach attempts to be first-principles at each scale, from the microscale to macroscale. Such an approach represents an essential and crucial step for the multiscale analysis of catalytic processes and the identification of the dominant mechanisms underlying the observed macroscopic phenomena. This paves the way towards the rational understanding and development of new reaction and reactor concepts.

References

[1] Strang G., SIAM Journal of Numerical Analysis 5 (1968) 506

[2] Ren Z., Pope S. B., Journal of Computational Physics, 227 (2008) 8165

[3] website: www.catalyticfoam.polimi.it

[4] website: OpenFOAM®: Open Source CFD www.openfoam.com

[5] M. Maestri, D.G. Vlachos, A. Beretta, G. Groppi, E. Tronconi, AIChE J., 55 (2009) 993