Untitled

In Fig. 1, experimental and predicted results for a crossflow velocity of 1 ms–1 and four TMPs are shown. For long-time scales, the permeate fluxes predicted by the model are lower than those experimentally observed. However, the shapes of the curves for flux decline with time estimated by the model are similar to those experimentally obtained. For all the crossflow velocities tested, the model is able to predict the change in permeate flux as a consequence of increasing the TMP and moving from a pressure-controlled process to a mass-transfer-controlled process. It can be observed that the difference between permeate fluxes for different TMPs is reduced as the TMP increases. According to the model, for short-time The best model predictions were obtained for high TMPs and low crossflow velocities, when the cake layer is more likely to form. Under certain experimental conditions, pore blocking may not occur. Moreover, the gel layer could be formed for very short-time scales, and for certain conditions, it may not be developed. The model also considers that the gel layer is only formed on the membrane surface where the pores have been previously blocked. However, the gel layer might be formed all over the membrane surface. These could be the reasons for the discrepancy between experimental and predicted results under certain operating conditions.