Some observations on the chemical cleaning of fouled membranes

Multi-photon microscopy, a fluorescence based technique, was used to image in-situ the fouling and cleaning of a 0.22 μm microfiltration membrane fouled by bovine serum albumin and ovalbumin. Imaging of the cleaning was possible for two of the three cleaning regimes used. The free chlorine cleaner bleached the fluorescence labelling on the proteins and so visualisation of the process, whilst using this cleaner, could not be undertaken. However, during the use of solutions of either sodium hydroxide or Ultrasil 53, appropriate fluorescent labelling of the proteins allows the foulant species to be clearly identified. Cleaning with sodium hydroxide was less effective resulting in only 50% flux recovery. The cleaning mechanism was different and large aggregates remained on the membrane surface even after 1 h of cleaning. Images correlated well with filtration data and clearly show re-fouling of the membrane during the second half of the cleaning cycle by ovalbumin alone. The discussion includes observations on the need for one to be aware of the potential for this to occur and on the need for some studies to involve repeated cycles of fouling and cleaning.