2015 AIChE Annual Meeting Proceedings
(470a) Dry Reforming of Methane on Ni in a Fixed-Bed Reactor: Spatial Reactor Profiles and Detailed CFD Simulations
Authors
Dry
Reforming of Methane on Ni in a Fixed-Bed Reactor: Spatial Reactor Profiles and
Detailed CFD Simulations
Gregor
D. Wehinger1, Matthias Kraume1, Viktor
Berg2, Katharina Mette3,
Malte Behrens4, Robert Schlögl3, Oliver Korup2,
Raimund Horn2
Current design of fixed-bed reactors is
based on pseudo-homogeneous or heterogeneous reactor models in which the
structure of the solid phase is not explicitly accounted for. This approach
leads to unsatisfactory results in particular for small tube-to-particle-diameter
ratios.
The discrete element method[1]
allows generating a random packing of catalyst particles in a single tube. Industrial
multi-tubular reactors contain thousands of tubes, hence this randomly filled
tube can be considered representative. For the geometry of the packing, it is
possible to simulate the flow field and the transport processes of mass and
heat between the bulk fluid and each individual catalyst particle. In this way
the usage of lumped transport correlations can be avoided. The catalytic
chemistry is implemented by detailed microkinetics.
In the present contribution we demonstrate
this simulation approach for dry reforming of methane on nickel. To validate
the simulations spatial reactor profiles of species concentrations and
temperature were measured in a dedicated reactor[2]. The nickel
catalyst is applied as a washcoat of a Ni, Mg, Al hydrotalcite-like precursor[3]
on α-Al2O3 spheres of 1 mm diameter.
The conservation equations of total mass,
momentum, energy, and the species balances were solved in 3D assuming laminar
flow. The local temperature of the catalyst particles was set to the
experimental value and the particles were considered to be isothermal. Further
details on the simulation procedure can be found in Ref. [1].
For heat transfer only, simulation and
experiment are in excellent agreement. If chemistry is included, the experimental
species profiles are qualitatively reproduced by the model, but some
quantitative deviations occur. This can be due to catalyst deactivation during
the measurements or to inaccuracies in the microkinetics. The presented approach
gives more insight into catalytic fixed-bed reactors than conventional reactor
simulations treating the phases as a continuum.
Fig.
1: Top left: Reactor tube with catalyst packing. Top right:
Velocity field in the bed. Bottom: Experimental species and temperature
profiles.
[1] G. D. Wehinger, T. Eppinger, M. Kraume Chem. Eng. Sci. 122
(2015) 197.
[2] R. Horn, O. Korup, M. Geske, et al., Rev. Sci. Inst. 81
(2010) 064102.
[3] K. Mette, S. Kühl, H. Düdder, K. Kähler, A. Tarasov, M. Muhler,
M. Behrens ChemCatChem 6 (2014) 100.