BIOLOGICAL DENITRIFICATION OF POTABLE WATER Final Report to the Department of the Environment - 319-S
Report No DWI0082

Nov 1984

SUMMARY

The objectives of the work on biological denitrification of potable water were:

  1. To test fully the process and engineering design of the prototype biological denitrification plant and to monitor its process performance.

  2. To identify scale-up problems.

  3. To investigate start-up techniques for biological denitrification units while monitoring critically the quality of the product water.

  4. To evaluate different carbon sources.

This report summarises the main findings of the project.

The prototype biological fluidised sand bed denitrification plant was constructed at the Bucklesham treatment works of Anglian Water and commissioned on 9 June 1982. The plant had a capacity of 2.8 Ml/d and treated a quarter of the total site flow of 11.2 Ml/d. The plant was designed to remove 12.8 mg DO/l and 15 mg NO3 N/l at a temperature of 5ºC.

The denitrification plant was operated for two years Using methanol as carbon source and for short periods with ethanol (11 weeks) and acetic acid (8 weeks) as carbon sources. For long periods of operation, totalling over 1 year, the denitrified water was blended into supply with no adverse effect on the final water quality.

Using methanol as carbon source the plant removed up to 14 mg N03 N/l at temperatures as low as 2ºC confirming the validity of the design data used. No problems were experienced starting-up the biological reactor which took approximately 30 days to achieve a nitrate removal of 6 mg N03 N/l. Additional pilot plant work carried out on site indicated that this could probably be reduced to about 20 days.

As this was a spring-fed source the phosphate concentration in the raw water was very low, and phosphate dosing was required for optimum operation of the plant.

Elevated nitrite residuals of up to 2 mg N02 N/l were at times produced in the denitrified water. In most cases this could be controlled by reducing the carbon source dose. No reason for this occasionally high nitrite production has been found despite extensive pilot plant trials.

The sludge thickening and storage plant was effective in handling the sludge and reduced the sludge volume requiring disposal by at least 80%.

The reliability and accuracy of the methanol and nitrate monitors were improved significantly during the duration of the project, however, the nitrate monitor was not sufficiently to be accurate used for process control.

The direct operating costs based on the total flow of 11 Ml/d at Bucklesham has been estimated at 0.3 p/m3/mg NO3 N/1 removed. The use of ethanol and acetic acid as carbon sources would increase the direct operating costs by approximately 25% and 50% respectively.

The project was carried out jointly by Anglian Water and the Water Research Centre.

Copies of this report may be available as an Acrobat pdf download under the 'Find Completed Research' heading on the DWI website.