Catalysis Science & Engineering, Sponsor Lecture
CE-021

Revealing the structural secrets of random catalyst packings

J. von Seckendorff1,3, O. Hinrichsen2
1Clariant Produkte (Deutschland) GmbH, Heufeld, Germany, 2Technical University of Munich, Garching, Germany, 3

Spheres, tablets, rings, double alphas, three-holed trilobes or flower shapes, extruded or tablet pressed – the list of readily available catalyst shapes is long and diverse. More precisely, catalyst geometries became uniquely prevalent for certain reactions as each shape promotes different aspects in fluid dynamics, heat and mass transfer. Still, catalyst shape development has long been a matter of instinct and experience while being significantly constrained by the limits of catalyst manufacturing techniques and experimental characterization capabilities.
But with on and on raising computing capabilities including numerical packing generation tools and computational fluid dynamics, an in-depth and highly resolved virtual representation of catalytic reactors is facilitated. Using these reactor models in virtual screenings, reaction systems can be optimized at highly reduced cost, time and manpower while simultaneously allowing a detailed understanding of specific phenomena.

Numerical Screening Methodology
The catalyst shapes are numerically packed into tubes using an automated python script in Blender™ (procedure adopted from [1]) or the Discrete Element Method tool DigiDEM™ (procedure as of [2]). Subsequently, the random packings are meshed and steady state flow simulations are performed using OpenFOAM®. If desired, dynamic flow simulations (e.g. for residence time distribution evaluation) can be added. The existing script is currently enlarged to allow heat and mass transfer evaluations. In a final step, extensive post-processing is performed allowing an in-depth evaluation of the flow profiles including the discrimination of bulk flow and flow through the holes of holed particles.

Validation strategy
The numerical packing procedure is validated with highly resolved x-ray tomography scans, both for simple spheres [4] and more complex shapes (see fig. 2). Pressure drop experiments have been performed [5] to confirm the fluid dynamic simulations. 

Results
Several studies have been performed all targeting the evaluation of catalyst shape effects on packing geometry especially porosity (distribution), pressure drop, flow profiles and residence time distribution. These comprise the evaluation of the influence of the tube-to-particle diameter ratio, especially in terms of spheres [4], the influence of shape aspect ratios, e.g. in terms of cylinders [5], or the development of (novel) promising catalyst geometries such as the yo-yo shape (fig. 1) [3]. Currently, the transfer from ideality (in terms of catalyst shape and material) to reality is the main focus of our research.

[1]  B. Partopour, A. G. Dixon, Powder Technology 2017, 322, 258–272.
[2] J. Fernengel, J. von Seckendorff, O. Hinrichsen, Proceedings of the 28th European Symposium on Computer Aided Process Engineering 2018, 97–102.
[3] J. von Seckendorff, M. Tonigold, P. Scheck, R. Fischer, O. Hinrichsen; submitted.
[4] J. von Seckendorff, N. Szesni, R. Fischer, K. Achterhold, F. Pfeiffer, O. Hinrichsen; submitted.
[5] J. von Seckendorff, N. Szesni, R. Fischer, O. Hinrichsen; Chemical Engineering Science 2020, 222, 115644.