MICROCYSTIN-LR REMOVAL FROM WATER
Report No FR0367

J Hart and P Stott

Mar 1993

SUMMARY

I BENEFITS

Identification of processes capable of removing algal toxins from drinking water.

Optimisation of processes for removing algal toxins from drinking water.

II OBJECTIVES

To identify and develop processes suitable for the removal of algal toxins and to provide a basis for estimating potential water treatment costs.

III REASONS

Blooms of blue-green algae can give rise to the production of toxins that may contaminate freshwaters used as sources for potable water. There is concern that these materials are not effectively removed by conventional water treatment processes and that they may enter supply.

IV CONCLUSIONS

The laboratory tests have shown that advanced water treatment processes can be used to remove microcystin-LR from lowland and upland derived reservoir waters. The following conclusions can be made about each of the treatment processes examined.

• PAC dosing is effective at removing microcystin, but the degree of removal is strongly dependent on the dose applied.
• GAC is effective, but the bed lives for toxin removal are short.
• Chlorine and hydrogen peroxide are ineffective.
• Chlorine dioxide is effective at removing the toxin when applied to clarified/sand filtered (treated) water, but not when applied to raw water.
• Ozonation is effective. Greater degrees of removal are achieved when ozone is applied to treated water, than to raw water.
• Potassium permanganate is effective when applied to treated water.
• Nanofiltration is effective.

Several different microcystins (congeners) exist, all cyclic heptapeptides, with some variation in amino acid residues(2) Because of the similar nature of the microcystins, their removal during water treatment is likely to be similar.

V RECOMMENDATIONS

The most effective processes for removing microcystin from water are ozonation and potassium permanganate oxidation of clarified/sand filtered water. Both process offer the advantage that they can be applied when needed, i.e. because of seasonal occurrences of algae. Before either of these process could be installed at a treatment works, careful consideration should be given to their impact on existing treatment practices and water quality. In addition, to use the processes to deal with seasonal occurrences, monitoring of the raw water source would be needed to ensure that the presence of toxin was detected and that the processes were applied accordingly.

PAC dosing is suitable if toxin removals of less than 90% are required, for raw water concentrations of 10µg/l. If higher degrees of removal are required, the process becomes expensive as high PAC doses are required. In addition, the PAC doses required may be impractical to dose at a treatment works.

For GAC adsorption to be effective, contact times of more than 15 minutes should be considered, assuming that microcystin does not biodegrade on the GAC.

Unless the toxin degrades, contact times of up to 40 minutes may have to be considered for a persistent toxin problem, to avoid significant breakthrough from the GAC. Using GAC as a sand replacement media is, therefore, not recommended unless there is sufficient contact time available.

For new works, nanofiltration could be considered as an effective process for toxin removal. However, significant pre-treatment would be required to ensure that fouling of the membrane did not occur.

Vl RESUME OF CONTENTS

The report contains the results of laboratory scale studies, used to investigate the removal of microcystin-LR from lowland and upland derived reservoir waters, by advanced water treatment processes. It also contains cost estimates for removing microcystin-LR from water by various methods of treatment.

Copies of the Report are available from FWR, price £25 less 20% to FWR Members.