Membrane autopsy helps to provide solutions to operational problems

Within the lifetime of most membranes some fouling will adversely affect membrane performance and cleaning will be necessary. The cleaning will depend on the type of deposit such as organic fouling, inorganic scaling or colloidal deposition. However strong the circumstantial evidence as to the identity of the foulant, the only reliable method of determining the true identity of the foulant with confidence, is by membrane autopsy. Different deposits affect the performance of the plant in different ways. Deposits such as organic fouling cause reduced output and increased differential pressure along the membranes and often require higher operating pressure to maintain output. Inorganic deposits can be a cause of irreversibly increase in salt passage. All forms of membrane fouling cause a significant increase in plant downtime and maintenance costs. Following an extensive study of membrane fouling during the autopsy procedure, we can usually recommend a protocol for the most suitable cleaning procedure. Cleaning procedures can be supported by proposals for antiscalant treatments to avoid the formation of deposits and in some cases retard membrane fouling. It is common to find deterioration in the salt rejection layer leading to a significant increase in salt passage. In some cases this is associated with an increase in product flow and may be caused by oxidation of the membrane, deposition of inorganic salts, abrasion or even the tearing of the membrane leaves themselves. If the cause of the fouling can be identified, recurrence of the problem can often be prevented, although in some cases, salt rejection performance may be irrecoverable. This paper considers the results of some autopsies from plants with different problems and presents a general opinion of current operating practice. Photographic evidence is included showing some of the different forms of scale and fouling. Conclusions are given based on the general experiences of the staff carrying out these autopsies.

• Membrane autopsies remain a useful diag• • • • • nostic tool and give a good insight into membrane operation and best cleaning solutions. Cleaning programmes tend to be applied, too little and too late. Cleaners may be ineffective when changes in flow patterns take place at the membrane surface due to fouling on the spacer material. Cleaning solutions will follow the path of least resistance and may have minimal contact with the foulant. Ignoring problems of microbiological growth at the membrane surface, there is strong evidence to indicate that low Silt Density Indexes (SDI) are not a sure indication of low membrane fouling. Care with pre-treatment design to ensure the elimination of colloidal ‘silica’ and ‘aluminosilicates’ may give better fouling protection than the use of chlorination/dechlorination. After all, every polyamide membrane has an active biofilm. Autopsy indicates that the use of chlorination and dechlorination does not have any statistical impact on membranes suffering from biological fouling. The best way of controlling biofilm growth is to limit the amount of nutrient available.