Abstract
Plug flow pattern is normally assumed in order to simplify the biofilter models. This assumption could result in the discrepancies between the predicted and experimental results since the axial dispersion in the gas phase could easily occur. Therefore, a steady state model taking into account dispersion effects was developed in this study. The model predictions were compared to experimental data from a downward flow biofilter using compost for removing methanol. The dispersion coefficient used in the model was determined by curve fitting with a set of experiment and was found to be a function of methanol concentration. Comparing to the plug flow model, the predicted results of dispersion model appear to fit well with the experimental data.Water movement within the bed was also predicted. The leachate from the biofilter was collected and compared to the model predictions. The amount of collected water increased much more rapidly with inlet methanol concentration than predicted by the model. In practice, water is continuously removed from the bed by convection, resulting in the bed drying. Therefore, the steady state model is no longer valid unless the optimum amount of water is added to the bed to keep the solid phase in equilibrium. However, this model can be used to optimize the amount of additional water to ensure the maximum microbial activity during operation.
