CONSULTANCY: REMOVAL OF PESTICIDES BY GAC AND GAC/OXIDANTS FINAL REPORT TO THE DRINKING WATER INSPECTORATE, DEPARTMENT OF THE ENVIRONMENT
Report No DWI0256

Sept 1991

SUMMARY

Water companies are required to comply with the standards for pesticides set in the Water Supply (Water Quality) Regulations 1989. Companies have given undertakings to the Secretary of State, under Section 20 of the Water Act 1989, which require them to carry out improvement programmes to secure or facilitate compliance with these standards. The Drinking Water Inspectorate (DWI) is currently reviewing these undertakings and needs to be satisfied that the improvement programmes, and those in any new undertakings, represent the most practical and appropriate steps to take and the timescales are appropriate. Therefore, the DWI, on behalf of the Department of the Environment, has commissioned WRc to produce a report which sets out the current state of knowledge on the effectiveness and use of granular activated carbon (GAC) or of GAC/oxidants (particularly ozone) for the removal of triazine, uron and phenoxyalkanoic acid pesticides from drinking water.

Properly designed and operated granular activated carbon processes can be effective at removing triazines, urons and phenoxyalkanoic acids to below the maximum permissible concentration of 0.1ęg/l in drinking water. However, the effectiveness of GAC filters, i.e. the magnitude of the regeneration interval, depends on the type of pesticide, the influent pesticide concentration, the type of GAC and empty bed contact time. Coal and wood based GACs appear to give longer bed lives than other types of GAC. The type of water treated would seem to have no effect on the order of effectiveness of the different types of GAC studied. However, treatment of groundwaters may lead to longer bed lives than for surface waters because groundwaters, in general, have a lower total organic carbon content.

GAC may be used to replace the sand in the rapid gravity filters. Rapid gravity beds are usually designed to operate with short contact times, which give high volumetric throughputs, and this may lead to a short GAC bed life. The bed life may be so short that the regeneration frequency becomes operationally impractical in which case purpose built GAC adsorbers can be installed as post-filter adsorbers. In general, each treatment plant needs to be looked at individually to determine the position of the GAC bed since it depends on several factors: the pesticides to be removed and their level of occurrence, the type of treatment works and what is an operationally acceptable regeneration interval.

For the pesticides covered in this report, ozonation has been shown to reduce the concentrations of pesticides in water. The evidence so far indicates that higher percentage removals are obtained for urons and phenoxyalkanoic acids than for triazines. In general, the effectiveness of ozonation is dependent on the type of pesticide, the pesticide concentration and ozone dose.

Work is showing that atrazine is more readily removed by a combination of ozone and hydrogen peroxide, than by ozone on its own.

The position in the process stream where ozonation is installed may not be important for the removal of the pesticides discussed in this report. What appears to be more important than the position of ozonation in the process stream are the ozone dose and the type and concentration of pesticide present in the water. However, the position of the ozonation stage may be important for other uses of ozone.

For the pesticides covered in this report, combining GAC filtration with ozonation may extend the bed life of the GAC. If ozonation and GAC filtration are to be combined, ozonation should precede GAC filtration.

Breakthrough of bacteria populated carbon fines may be reduced by minimising changes in flow rate during normal operation. High influent turbidities to the GAC bed seem to increase the amount of bacteria on the carbon fines.

The formation of brominated trihalomethanes is strongly dependent on the ozone dose, bromide concentration, bicarbonate concentration, pH and type of organic substances present. Owing to this complex chemistry, it is difficult to predict the extent of the formation of brominated trihalomethanes at a particular site. It is also known that ozone will react with bromide in water to give bromate.

Ozone will react with organic matter in water to produce assimilable organic carbon (AOC) which could lead to biological aftergrowth in distribution. GAC (both as a sand replacement and a post-filter adsorber) or sand filters (both rapid gravity and slow sand filters) will reduce the AOC concentration but may not reduce the levels to those found before ozonation.

Desorption of adsorbed substances once a GAC bed is exhausted has not been observed to date for pesticides.

Limited work has been undertaken on assessing the extent and control of any potentially deleterious side effects from the use of GAC or GAC/ozone for the removal of pesticides and further work is needed to clarify the situation.

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