Miljøprojekt, 970 – Metalafgivelse til drikkevand, DEL 2 - Videreførelse af Rig-tests af materialer til husinstallationer

Release of Metal to Drinking Water, Part 2

Summary and conclusions

In this project metal release has been investigated from materials commonly used in domestic water installations. The materials tested were hot dip galvanized steel, copper and various copper alloys. The tested copper alloys were gun metal, aß-brass, dezincification resistant brass, a.o. A-metal, which is a special type of dezincification resistant brass.

In the experiments performed at Danish waterworks a part of the test pieces tested were identical to test pieces tested in Germany. In this way it has been possible to compare the results from Denmark and Germany and to reveal the causes for differences in test results.

All experiments were carried out as 12-hour stagnation tests, and the test pieces were exposed for 1-4 years. Some pieces were continuously exposed from the first test run in 1999, whereas others were mounted in 2002.

The test-rigs were situated at waterworks with 4 different water types representing soft, medium hard and very hard water. The 4 water works are Astrup and V. Gjesing with soft water, Lysholt with medium hard water and Regnemark with very hard water.

The experiments performed are reasonably representative for the conditions in drinking water installations. In practice there is much variation.

The new investigations of metal release with one-year exposure time in drinking water (2002) show the same tendencies as the previous experiments (1999). The metal release depends strongly on water quality and it increases with increasing concentrations of hydrogen carbonate, salt content and conductivity. The level of metal release in the medium hard water in Lysholt, TRE-FOR thus was higher than in the softer waters in Esbjerg (Astrup and Vester Gjesing) and lower than in the hard water in Regnemark, Københavns Energi.

The nickel release seems to be diverting from this pattern, as it is almost not affected by the water composition of the 4 water types. There is a tendency for highest Ni release in the soft water in Astrup.

Copper release from copper pipes and zinc release from hot dip galvanized steel pipes tested in Lysholt were below the limits in the regulation from the Danish Ministry of the Environment (Bekendtgørelse no. 871, 21. September 2001), which was not the case in previous test on the same materials in harder waters.

The release of lead, copper and arsenic was suitably low (below limits in the regulation from the Danish Ministry of the Environment) from the copper alloys brass, DZR-brass and gun metal produced as standard test pieces and mounted in stainless steel pipes. Yet, lead release is too high the first three month of exposure, unless the copper alloys are washed in a special cleaning process. These results are in good agreement with previously reported German results. In the test rig the test piece area accounted for 15 % of the test pipe area. The nickel release from brass and DZR-brass was low. The gunmetal tested had a high nickel content of 1,1 %. The nickel release from this alloy was high and above the limits in the Danish regulation.

The improved properties of the recently tested DZR-brass is due to a number of factors such as heat treatment in the production process, geometry of the piece and test conditions.

A combined test piece with water meter and block valve, containing both brass and DZR-brass in all water types released less lead, copper and arsenic than DZR-brass.

The nickel-plated parts after 1 year of exposure gave rise to a nickel release below the limits in the regulation from the Danish Ministry of the Environment. This is true when the water from the water work is free of nickel. It should, however, be taken into consideration that some Danish water types contain nickel.

Another type of DZR-brass (A-metal) could fulfill the demands in the regulation from the Danish Ministry of the Environment in the medium hard water of Lysholt. The material has only been tested in the form of manifolds.

An adjusted fabrication method for 2 types of mixer taps lowered the nickel release in Regnemark water considerably. The mixer taps could fulfill the demands in the Danish regulation with the new production method.