Precautions advised for tar-lined supply pipes
A research team led by Dr Matthias Maier of Surrey University has investigated the contamination of water supplies with polycyclic aromatic hydrocarbons. The results show that if there are tar-lined pipes in your network, you should avoid chlorination, stagnation and sudden changes in pressure.Polycyclic aromatic hydro-carbons (PAHs) are sometimes found in water supplies, although usually at low concentrations. Some PAHs, such as benzo (a) pyrene (BaP), are known to be carcinogenic and have endocrine-disrupting properties. The Drinking Water Inspectorate (DWI) in the UK has therefore decided on a very strict standard for BaP of 0.01æg/l and follows EC standards for several others.
According to the WHO: "The main source of PAH contamination in drinking-water is usually the coal-tar coating of drinking-water distribution pipes." Coal tar has been used since the 19th century as a lining to protect cast iron and steel water pipes from corrosion.
Because many water companies still have tar-lined pipes, researchers have been trying to find out how PAHs are released and whether they are released in quantities likely to endanger human health. One research team, led by Dr Matthias Maier and Professor Barry Lloyd of Surrey University's Centre for Environmental Health Engineering and Professor Dietrich Maier of Karlsruhe University in Germany has found some interesting results.
Dr Maier's team suspected that a common cause of increased PAH levels in water from tarred pipes would be damage to a natural biofilm which normally grows on top of the tar. Dr Maier predicted damage to the film would be accelerated by a number of chemical and physical factors.
A survey of water from taps showed PAH concentrations were significantly raised in areas with chlorinated supplies. When chlorine or chlorine dioxide was added to a previously PAH-free tar-lined network, PAHs were then detected. Chlorine dioxide was found to have a greater effect than chlorine. When chlorination stopped, the PAH levels decreased.
The highest total PAH level resulting from chlorination (with chlorine dioxide) was 0.69 æg/l. This is not in breach of WHO guidelines but clearly breaches the current EC and UK standard of 0.2æg/l for total PAH.
The predominant PAH released from the tar was, as expected, the less toxic fluoranthene. BaP levels reached 0.05æg/l following extended chlorination, which is also in breach of the current UK standard of 0.01æg/l. Chlorin-ation was found to have similar effects on water quality in a specially-built test rig.
Tests were also carried out in the laboratory which showed PAH levels to increase following reductions in the water's oxygen levels, and following marked decreases in pH caused by adding hydrochloric acid. Trihalomethanes (THMs) were also added because they can form in supply by chlorine reacting with natural organic compounds. Chloroform was used as the test chemical as it is known to be a good solvent for coal tar, but even at levels of up to 3,700æg/l it was found to have no effect on PAH levels.
Chemical changes were not the only cause of increased PAH levels. Physical changes also had a significant effect. A test of the effect of changes in pressure in supply mains from 1-7bar resulted in total PAH levels exceeding EC guidelines and BaP levels of up to 0.07æg/l. The water from the pipes was examined and found to contain iron-rich particles. In sections of old pipe rust and tar particles from unprotected areas are found trapped in the biofilm. The particles are easily dislodged by changes in pressure, resulting in a dramatic increase in PAH levels as the particles can contain up to 4mg/l of PAHs. Dr Maier's group tested the particles and found that 75% of the PAHs measured, including BaP, were adsorbed onto the particles and would be removed by a simple filtration process.
At the opposite extreme to changes in pressure and flow, stagnation was also found to result in PAH contamination. When anaerobic water was kept in the same section of tar-lined pipe for seven hours, PAH levels were seen to increase. After a two-day period a total PAH level of 0.135mg/l was reached, way in excess of acceptable limits. The cessation of flow and lack of oxygen also resulted in increased turbidity, probabbly as a result of the biofilm breaking down and rust particles being released.
Analysis of the turbid water showed particles to consist of iron, magnesium and calcium, lending strength to this idea. Dr Maier suggests that the rust forming and dissolution process may be aided by the presence of iron-oxidising bacteria, such as Gallionella ferruginea, which was found living in the rust layer of the pipes using electron microscopy. Iron-oxidising bacteria can thrive on energy gained from oxidising ferrous iron (Fe2+) to rust (Fe3+).
Supply network managers, should, on the basis of these results, take a number of precautions if a supply network is known to include tar-lined pipes. Firstly, avoid excessive chlorination, especially with chlorine dioxide. Dr Maier suggests: "Alternative disinfection methods without a free residual, like UV-radiation, should be considered."
Secondly, try to avoid sudden changes in flow and pressure as these can not only cause discoloration as iron deposits are dislodged, but can also result in contamination with PAHs.
Thirdly, try and prevent flow stagnation as this can have a similar effect. If such conditions are thought likely, perform regular analysis of water samples and if necessary, filter or isolate the supply.
Information in this article is derived from Water Research Vol. 34 March 2000, No. 3, pg 773-786: "Factors influencing the mobilisation of polycyclic aromatic hydro-carbons (PAHs) from the coal-tar lining of water mains." By M. Maier, D. Maier and B. J. Lloyd.
At the works, granular activated carbon filters remove hydrocarbons - but PAHs can still turn up in supply development of PAH standards for drinking water The development of PAH standards for drinking water supplies has been a fairly tortuous process as the results of various experiments have shown different types of PAH to vary in toxicity. The WHO set a standard of 0.7æg/l for one of the most toxic PAHs, benzo (a) pyrene (BaP) in 1996. But BaP is not the most prevalent PAH to be leached from coal-tar linings and the WHO has now recognised this fact by introducing a standard of 4æg/l for the more common but less toxic PAH known as fluoranthene.
The EC has adopted a different standard which takes into account the range of PAHs to be found in coal-tar pipe linings. The current standard observed by member states, including the UK, is a total of 0.2æg/l for a combination of six PAHs, including fluoranthene and BaP. However, this standard will soon be replaced by the EC's latest Drinking Water Directive, with which member states now have until 2004 to comply. The new standard will be for a total of 0.1æg/l for a combination of four PAHs, not including fluoranthene or BaP. Owen Hydes, deputy chief inspector of the DWI said: "The new EC directive will have separate standards for fluoranthene and BaP. The fluoranthene limit will be 4æg/l, while that for BaP will be the same as that already enforced in the UK, which is just 0.01æg/l." The DWI prosecutes any water companies which fail to comply with UK and EC drinking water standards.