B.3 Modelling and simulation of fixed-bed reactors

  • Jørgensen, Sten Bay (Project Manager)
  • Recke, Bodil (Project Participant)
  • Keldebæk, Morten (Project Participant)

Project Details


Gas phase fixed bed reactors constitute one of the most common reactor classes in industry. However fixed bed reactors are generally not included as standard modules in commercial simulation packages. The goal of this project is to develop models for fixed bed reactors with the purpose of understanding non-linear dynamic behaviour. Three reactor types are considered. Hereof two types are continuously operated fixed bed reactors which in one case is indirectly cooled and contain two or more catalyst beds with one or more inter-bed heat exchangers and in the other case is directly cooled by heat exchange through the catalyst tube wall. The third type is a periodically operated fixed bed reactor containing multiple beds. This reactor type is used for heat regenerative oxidation of a dilute mixture of combustible gases.
With the given purpose simple models should be developed. For an indirectly cooled ammonia reactor a dynamic pseudo-homogeneous model of the catalyst bed with only 3 partial differential equations has been developed. Simulations with the model show very good agreement with a heterogeneous model with 6 PDEs. One inter-bed heat exchanger is modelled with 4 ordinary differential equations and good agreement has been achieved when compared with more complex models also for varying flow rates. Models have also been derived on a similar basis for a directly cooled methanol reactor and for a periodically operated multi-bed regenerative reactor with a single oxidation reaction.
Future work include performing bifurcation analysis for a part of an ammonia reactor where the first bed is coupled to an inter-bed heat exchanger which effectively acts as a feed-effluent heat exchanger. This system thus includes energy recycle.
Effective start/end date01/05/1997 → …


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.