Summary and conclusions, Miljøstyrelsen

Fjernelse af MTBE i danske vandværker

Summary and conclusions

MTBE (Methyl tertiary-Butyl Ether) has been detected in the inlet water at Grubbemølle Waterworks since 1999, and the content of MTBE has in the same period of time been reduced throughout the treatment processes at the waterworks, resulting in a reduced content of MTBE in the drinking water. The main objective of this project has been to identify the treatment processes at a typical Danish groundwater based waterworks, which ensures removal of MTBE to below the Danish drinking water standard of 5 mg MTBE/l.

In order to fulfil the main objective, the effect of some of the operational procedures at Grubbemølle Waterworks were analysed with the aim of achieving a better knowledge of the most optimal way of running the waterworks in relation to the MTBE reduction. It was chosen to focus on the MTBE removal in relation to: Stripping of MTBE during aeration, the oxygen content in the water during filtration, the retention time in the sand filter, periods of standstill and the backwash intensity and frequency. Furthermore, the removal of MTBE at another waterworks was analysed in order to compare with the removal of MTBE at Grubbemølle Waterworks.

The performed investigations were primarily performed at Grubbemølle Waterworks, where the reduction of MTBE was followed under different operational conditions of the waterworks during aeration and filtration. In order to support a few of the investigations performed at the waterworks, investigations of the degradation of MTBE were conducted in batch systems in the laboratory. In addition to the investigations performed at Grubbemølle Waterworks, the MTBE content at Fåborg Waterworks in the County of Ribe was investigated. The analyses of MTBE were performed by an approved laboratory on a GC-MS, except from the degradation experiments performed in the laboratory where MTBE was analysed at DTU using a GC-Purge & Trap.

The concentration of MTBE in the inlet water at Grubbemølle Waterworks varies between approx. 10-65 mg/l, depending on the actual production pattern. MTBE is reduced up till 65 m g/l during the treatment at the waterworks. The removal of MTBE occurs during aeration due to stripping (approx. 20-30% of the MTBE in the groundwater) and during filtration due to biological degradation (approx. 60-80% of the MTBE in the groundwater). The hydrocarbons BTEX and degradation products of the MTBE removal were not detected at the waterworks.

Stripping of MTBE during the aeration at Grubbemølle Waterworks is dependent on the air-water rate. Aeration of oxygen and stripping of carbon dioxide is dependent on both the air-to-water ratio and mass transfer coefficient K_La. The K_La-value for oxygen and carbon dioxide was found to be in the range of 2 – 16 h^-1 depending on the air-water ratio. The air-to water ratio at Grubbemølle Waterworks was found to be in the range of 5:1 – 9:1. The retention time was determined to be 12, 15 or 47 minutes depending on the actual production pattern, and the aeration system was found to be a completely mixed system with no dead volumes.

Batch experiments conducted with filter material from the sand filter showed no effect of variations of the oxygen content of 7, 13 and 15 mg O₂/l on the MTBE degradation.

Short time changes of the retention time in the sand filter were found to have a significant effect on the degradation of MTBE. A lower flow rate, i.e. longer retention time, resulted in a high efficiency of the MTBE-degradation compared to a short retention time. However, when a short retention time (10 min.) was maintained for 4 weeks, the MTBE-degrading microorganisms had the ability to adapt to the new conditions, and a high removal efficiency was obtained eventually.

Short time stand-still periods of 1-4 hours were found to have no significant effect on the removal of MTBE, since the removal efficiency stayed constant at around 95-100%. A long time stand-still period of 4 weeks showed, on the other hand, to have a great effect on the removal efficiency of MTBE. After 4 weeks of stand-still no degradation of MTBE in the filter occurred; though after a further 4 weeks of normal production it seemed like the efficiency of MTBE-removal was re-established. The difference between continuos and varying operation was found to be insignificant at a short-term influence.

The backwash strategy currently used at Grubbemølle Waterworks is optimal in relation to removal of iron- and manganese precipitates. The effect of different changes of the backwash strategy was investigated. The investigated changes resulted in a reduction of the out-wash of active biomass, and thereby a more efficient removal of MTBE in the filter after backwash. Further, it was found that a lower frequency between the back wash is beneficial for the degradation of MTBE in the sand filter.

The results from Fåborg Waterworks show that no reduction of MTBE occurs during the treatment, despite from a small reduction during the aeration. However, the inlet concentration was only a few mg/l, which is much lower than at the Grubbemølle water works.