David Robbins
| Position | PhD student |
|---|---|
| College | Robinson |
| Qualifications/honours | MEng (Hons) Chemical Engineering, University of Sheffield, 2010, AMIChemE |
| Address | Magnetic Resonance Research Centre Department of Chemical Engineering and Biotechnology c/o Cavendish Stores JJ Thomson Avenue Cambridge CB3 0HE |
| djr57@cam.ac.uk | |
| Research group | Magnetic Resonance, Catalysis |
| Research project title | Developing CFD Codes to Predict Two-Phase Flows and Reaction in Heterogeneous Catalytic Reactors |
| Supervisor | Prof Lynn Gladden |
| Advisor | Prof Stewart Cant |
Research description
Trickle-bed reactors are broadly defined as reactors in which the liquid and gas phases flow co-currently downwards through a packed bed. They find use in a wide range of industries, predominantly in the oil and speciality chemicals sector. The low flow rates used in trickle- or pulsing-flow regimes are however inherently difficult to model due to the coupling of reaction kinetics, mass transfer, global- and local-scale hydrodynamics and multiphase interactions.
Computational Fluid Dynamics (CFD) simulations numerically solve the governing equations of fluid dynamics over control volumes in a domain to predict how the fluid will flow in a given geometry with prescribed initial and boundary conditions. However, commercial codes (such as FLUENT and CFX) have difficulty in the low-Mach number range. Numerical issues are further amplified by the disparity in scales presented by multiphase flow, existence of variable discontinuities and closure of the multiphase equations with relevant physical terms. Current CFD modelling techniques in this area are physically unsatisfactory and we are thus interested in several numerical methods (flux-vector splitting, gradient limiting and preconditioning) to allow for accurate simulations of single and multiphase flow in trickle bed reactors.
The CFD codes used in this study are in-house codes used at the Department of Engineering. MRI is an extremely valuable tool for the computational engineer in that it can provide data on a fine spatial resolution for direct comparison of the simulated flowfields and validation of the numerical methods used. The combination of these two tools will hopefully enable the development of accurate, physically-based multiphase models which have been validated with the revelent experimental data.
Key publications
D.J. Robbins, R.S. Cant, L.F. Gladden, E. von Harbou, M.S. El-Bachir. "CFD Modelling of Single-Phase Flow in a Packed Bed with MRI Validation" AIChE Journal (accepted for publication).
D.J. Robbins, R.S. Cant, L.F. Gladden. "Simulation of multiphase flows using a modified upwind-splitting scheme". To be presented at the XXXIII International Conference on Computational Fluid Dynamics, Aug. 22-23 2012, Paris.
D.J. Robbins, R.S. Cant, L.F. Gladden. "Development of accurate and robust liquid equations of state for multi-phase CFD modelling" (submitted for review to present at the 2012 AIChE Annual Meeting, Oct 28 - Nov 2 2012, Pittsburgh PA).
D.J. Robbins, R.S. Cant, L.F. Gladden. "Development of accurate, robust liquid equations of state for multi-phase CFD simulations with a modified AUSM+-up scheme" Computers & Fluids (submitted for review).