Report No DWI0754

Jan 1996


The objectives of this contract were to assess the ability of three point-of-use filters to remove Cryptosporidium oocysts from source water. The filters evaluated were an in-line ceramic cartridge filter, a gravity filter containing four ceramic candles and a microfiltration system. The first two are true point-of-use filters which produce purified water on demand. The microfiltration system is designed for larger-scale production in which storage capacity may be used and so it is not a true point-of-use filter.

The common testing protocol used water of three different turbidities (high, medium and low), to which sufficient oocysts had been added to allow a 5 log removal to be measured. The filters were operated under conditions of continuous dosing at normal flow rates and under conditions of shock challenge on an intermittent basis. The experience gained from this work allowed recommendations for the cleaning and replacement of the filters and an assessment of the health risks associated with the handling and disposal of used filter cartridges to be made.

The continuous dosing trials were run over a period of 7 days. It was found that the inline ceramic cartridge filter excluded oocysts during the high and medium turbidity trials but that low numbers were detected in the effluent during the trial with water of low turbidity. The gravity filter excluded oocysts during the three trials although low numbers of oocysts occurred during a low turbidity trial when the filter elements were cleaned rather than replaced. A repeat series of trials with the gravity filter using new candles and water of low turbidity failed to demonstrate breakthrough of oocysts. Throughout the trials with the microfiltration system, Cryptosporidium oocysts were not detected in the effluent. During the second series of trials, in which the filter candles were challenged intermittently, oocysts were only detected in the effluent from the gravity filter, using cleaned filter candles, during tests with water with a low turbidity. However, only low numbers of oocysts were detected and removal was greater than 99.9 per cent.

To prevent exposure to Cryptosporidium oocysts trapped by the ceramic filters the cleaning protocols recommended by the manufacturers must be adhered to strictly. Whilst the protocol for cleaning the candles is effective for the majority of contaminants, it is recommended that, where contamination by Cryptosporidium is possible, disposal rather than cleaning is the safest action. It is recommended that contaminated filters are sealed in plastic bags and disposed of through the local authority waste disposal system.

Currently there are no guidelines regarding the disposal of the membrane concentrates which would be generated periodically during the operation of the microfiltration system. Until such guidelines are available the best option would appear to be disposal to the public sewer, if a consent to discharge can be obtained, or where main sewerage is not available, discharge to a septic tank.

This contract was funded by the Department of the Environment and supervised by the Drinking Water Inspectorate.

Copies of this report may be available as an Acrobat pdf download under the 'Pre 2000 Reports' heading on the DWI website.