Rigopoulos, Stylianos; (2003) Modelling of gas-liquid precipitation systems. Doctoral thesis (Ph.D), UCL (University College London).

Text Modelling_of_gas-liquid_precip.pdf Download (13MB)

Abstract

The performance of gas-liquid precipitation in industrial reactors, such as bubble columns, is determined by the interplay between multiphase fluid dynamics, gas-liquid reaction engineering and crystallisation mechanisms such as nucleation, growth and agglomeration. The aim of this work is to develop efficient computational methods for dealing with these issues. The coupling of turbulent mixing and chemical phenomena lies at the heart of multiphase reaction engineering, but direct CFD approaches are usually confronted with excessive computational demands. A hybrid approach is proposed here, where the quantification of mixing is accomplished through averaging the flow and concentration profiles resulting from a CFD flow field calculation and a computational ("virtual") tracer experiment. These results form the basis for a mapping of the CFD grid into a generalized compartmental model, where the chemistry calculations can be carried out efficiently. A phenomenological model of mass transfer and chemical reaction is employed to simulate the interfacial phenomena in gas-liquid reactors, but the main concept lends itself to a variety of chemical reaction engineering problems. A first-order finite element scheme is proposed for solving the one-dimensional dynamic population balance equation with nucleation, growth, aggregation and breakage. The numerical scheme features improved stability by taking into account the convective nature of the growth term, and enhanced conservation of moments at non-uniform grids by identifying and mitigating the sources of conservation error in the evaluation of the aggregation source term. Exhaustive benchmarking is carried out against analytical solutions, and in all cases tested the method proves capable of producing accurate results at a reasonable CPU time. Finally, the theoretical and computational methods developed in this work are applied to simulate the precipitation of CaCO3 in a bubble column reactor. Dynamic measurements of the crystal size distribution (CSD) are carried out, and the evolution of the process is adequately reproduced by the model.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item