Water-pitcher filters miss lead particles
Water-pitcher filters may not reduce lead concentrations to acceptable standards in homes that have elevated levels of the metal in their tap water, because the filters do not remove lead particles, according to the Ontario Ministry of the Environment Canada. The ministry's declaration was issued just days after it received new advice from NSF International, the organization that certifies water filters.
Recently, high levels of lead have been found in tap-water samples taken throughout the Canadian province (Environ. Sci. Technol. 2007, 41, 5174). Government officials had been recommending that water-pitcher filters, such as Brita and PUR, be used at home to remove the lead. But now this recommendation has been withdrawn. Such filters are used widely. According to representatives from Brita and PUR, about 35% of American households use pitcher-type filters.
On July 16, NSF International updated its website listing for filters that meet the new, more stringent NSF-53 standard for reducing lead in drinking water. No pour-through filters meet the revised standard, but other products listed on the website do, such as some faucet-mounted filters and plumbed-in filters.
To meet the new requirements, the devices must be able to filter out not only soluble lead, which was covered by the previous standard, but also lead particles, according to Richard Andrew at NSF International. "We determined that lead particles can be a concern in drinking water and developed a new test, in which 30% of the lead exists as particles," he says. Filters that pass the test reduce overall lead levels of 150 parts per billion (ppb) to below 10 ppb, the Canadian water-quality standard.
Pitcher filter manufacturers haven't had enough time to comply with the revised standard, says Brita brand manager Margaret Quan, who adds that companies had just 2 months to prepare their products. "We support the new standard," she says. "Our products have to evolve with the science," she says, adding that the company is exploring its options.
The occurrence of particulate lead in drinking water has received increased scrutiny. In the past 2 years, cases of childhood lead poisoning have been tied to solder particles in tap water in both Greenville and Durham, N.C. After these episodes, the U.S. EPA modified water sampling guidance (PDF size: 156 KB) to account for the effect of particles trapped in faucet aerators by requiring sampling with aerators switched on.
Yet current models used by EPA and other groups to estimate human exposure to lead, as well as sampling protocols, analytical methods, and environmental assessments, are often based on the presumed dominance of soluble lead in drinking water.
As a result, the agency's sampling methods for lead in drinking water can, in some cases, underestimate the amount of lead present. Recent research by environmental engineer Simoni Triantafyllidou and colleagues at Virginia Polytechnic Institute and State University (J. Am. Water Works Assoc. 2007, 99[6], 107–117) used EPA's method and found that the amount of lead in drinking water was underestimated by a factor of five when particles are involved.
The team also compared the EPA method with a rough estimate of bioavailability—how much particulate lead dissolves in warmed and agitated simulated gastric fluid. They found that the simulation dissolves as much particulate lead as the EPA method or more. This suggests that the agency's tests can also underestimate the bioavailability of lead particles in drinking water.
Scientists and engineers are still divided over the prevalence, and hence the significance, of lead particles in drinking water, says Rob Herman, laboratory manager at NSF. "The problem exists in the real world, but it is hard to quantify," he adds. "Existing monitoring data don't help because it has not been measuring particles. But people want to make sure when they buy a product that it works in all circumstances. That's why we developed the new standard," he says.
