Effect of coal particle size distribution on packed bed pressure drop and gas flow distribution

Abstract

This paper focuses on the role of coal particle size distribution on pressure drop and gas flow distribution through packed coal beds. This fundamental knowledge is helpful in better understanding the operational behaviour of fixed bed dry bottom gasifiers. The Sasol synfuels plants in South Africa use 80 such gasifiers to convert more than 26 million tons of coal per annum to synthesis gas, and ensuring stable operation is of primary importance to ensure high synthesis gas production rates and gasifier availability. Pressure drop measurements on laboratory scale equipment were conducted to investigate the effect of particle size distribution on packed bed pressure drop. The well-known Ergun equation for pressure drop does not accommodate the effect of size distribution on pressure drop. A novel approach was followed to model pressure drop through simulated coal bed structures using Computational Fluid Dynamics (CFD). The coal bed structures were simulated by assuming that the coal particles are represented by randomised convex polyhedra in three-dimensional space. The computational space was divided into polyhedra using statistical Voronoi tessellation technique, which have been shown to be versatile in modelling problems in many fields, e.g. filtration, molecular physics, metallurgy, geology, forestry and astrophysics. This approach of flow modelling through packed coal beds is able to accommodate size distribution effects on pressure drop and gas flow distribution. The modelling work shows large deviations from plug flow with broad size distributions. The lowest bed pressure drop with the closest approximation to plug flow is obtained with the narrowest particle size distribution. Low gas flow rates are also beneficial for reducing excessive channel flow. Combustion profiles for different particle size distributions were studied using a pilot scale combustor. The combustion profiles provide confirmation of the CFD modelling results, namely that narrow particle size distributions and low gas flow rates reduce channel burning. Excessive channel burning was observed for broad particle size distributions, and is enhanced by high gas flow rates. The experimental and modelling work which was conducted, clearly indicate that narrow coal particle size distributions are desirable for optimum gas flow distribution and lowest packed bed pressure drop.