Displacement Washing in the Sulfate Pulp Mill

Displacement washing is an operation in which it is possible to wash solid-phase particles with a minimum of water. In a sulfate pulp mill, the main application is the pulp washing, but lime mud washing also occurs in the chemical recovery. The washing consumes a great deal of water. For environmental reasons there is an interest in increasing closure of the liquor systems. This means that excess water is a ballast which has to be pumped around or evaporated, and the water consumption should be kept as low as possible. It also means that non-process elements (NPE), will accumulate in the system and reach levels which may be harmful, e.g. in the bleaching operation. In order to handle the NPE problem, one important factor is the knowledge of how to make an effective separation in the pulp washing operation. Another crucial washing operation is that of lime mud, since it also influences the lime kiln operation and, later on, the quality of the white liquor produced. Hence, a knowledge of the phenomena that govern the displacement washing operation is crucial for the pulping industry to be able to meet the demands in an economically feasible way. Better knowledge of the phenomena also means more accurate models. The pulp fibre and the lime mud are examples of completely different solids, which lead to different conditions in the displacement washing operation. They are therefore suitable both to illustrate aspects of the displacement washing, and for examining conditions in general. In this thesis, studies on both lime mud and pulp are presented. Due to the great differences between the two solid phases studied, there are different phenomena that dominate the displacement washing operation. While lime mud particles form relatively homogeneous beds having small to moderate interactions with the stationary phase, the pulp fibres form more heterogeneous beds having complex stationary phase interactions, regarding both stagnant liquor and the actual fibre. In the lime mud study the dispersion model was found to give good agreement with experimental data, except for a small tailing at the end of the breakthrough curve. This could be improved by adding a disturbance term to the inlet boundary condition. No effect of pore diffusion could be seen, due to the small size of the particles, whereas there was a noticeable sorption effect. The dispersion coefficient was higher than for more ideally formed particles, and it depended on the type of salt in the solution. This is probably due to pH variations causing changes in the bed structure. The linear adsorption coefficient results were not quite satisfactory, but still showed some dependence on lime mud type and salt type. For pulp fibres, the study of the sorption of metal ions showed that the general sorption behaviour for mono- and divalent ions can be described with a Langmuir isotherm. The Langmuir isotherm could also be used for multi-ion solutions, when adding the influence of the charges on the fibre and the second power of the charges in the solution. The sorption was found to be slightly temperature-dependent. The sorption of ion pairs could be described by separation factors, showing a preference for divalent ions over monovalent ions on the fibre. The separation factors showed that the Mn2+ ions, which are harmful in modern bleaching, were more strongly bound than Mg2+, which is more preferable for the bleaching operation. This study of lignin diffusion out of the fibre wall showed that it could be divided into a short time diffusion and a long time diffusion. The difference is probably due to that the outer layers of the fibre wall have a more open structure than the inner layers, yielding a higher diffusivity during the time these layers are emptied. The short time diffusion is the dominant part for the times used in pulp washing. The increase in diffusivity due to temperature is more rapid above 80-90 \ub0C