Lead in Seattle school district’s drinking water prompts major review
There are many ways drinking water changes on its path to consumers’ taps. Corrosion of metal piping materials is one of the most significant causes of degradation. The more noticeable aesthetic degradation occurs as a result of oxidation of iron piping producing red, yellow or “rusty” water.
A few years ago, rusty water from Seattle Public Schools drinking fountains alarmed parents, and they initiated water quality testing. This, in turn, prompted a comprehensive review of drinking water quality throughout the district. In addition to high iron levels, lead in the water from a significant number of bubbler drinking fountains was above the 20-ppb guidance level used by the U.S. EPA as an action level in schools and daycare facilities. The immediate action taken by the district was to provide bottled water to all of the schools.
Next, HDR/EES helped the district initiate a comprehensive drinking water monitoring program throughout the district’s 102 school and administrative buildings for iron, lead and other parameters.
Out of approximately 2,170 first-draw 250-mL samples, 525 samples, or 24%, exceeded the 20-ppb guidance level for lead. For iron, out of a total of approximately 1,800 two-minute flushed samples, almost 26% exceeded the aesthetic threshold of 0.3 mg/L. The school board set a 10-ppb maximum lead level at all drinking water sources in their schools. This challenging standard — one half of the EPA guidance level of 20 ppb — caused significant compliance difficulties.
Identifying Sources & Mitigation
Several potential sources of lead were identified in school piping systems. These sources included galvanized steel pipe, the 50:50 (lead to tin ratio) solder installed before the lead ban, and brass components such as bubbler heads, valves, ferrules and flexible connectors. These components may contain up to 8% lead (by weight) and still be considered lead-free per the Safe Drinking Water Act regulations definition. Several mitigation measures were investigated. Depending on the specific circumstances in a building, recommendations included one or more of the following measures:
One of the more challenging aspects of the investigations included identifying end-use plumbing components that contained little or no lead and that would not cause lead levels to exceed the school board limit of 10 ppb. End-use components such as faucets, fixtures and bubblers are designed to meet the federal Lead and Copper Rule (LCR) applying to water utilities, but not to schools’ limits unless they have their own source. (The LCR has an action level of 15 ppb in a 1,000-mL sample collected after standing for a minimum of six hours. As part of the NSF testing protocol, components are exposed to test waters several times in a fill-and-dump procedure, and an averaging method is used to determine if the component passes or fails the NSF criteria.)
As part of the mitigation efforts at many of the Seattle schools, NSF-certified components (including brass bubbler heads with lead less than 0.3% by weight) were specified, installed and tested at drinking fountains and sink faucets. The water quality test results, however, revealed that many of the fountains and sink faucets with new NSF-certified end-use components still exceeded the 10-ppb level in first-draw 250-mL samples.
This prompted more investigation. A testing protocol was established to determine which of the end-use components were contributing to the higher-than-expected lead concentrations. The testing was conducted in HDR’s Water Services Laboratory in Bellevue, Wash. A component, such as a bubbler head, was filled with tap water and allowed to stand for two hours. A sample of that water was drawn and analyzed for lead; the water in the component was dumped, and then the component was refilled with test water. The entire process was repeated several times to passivate the interior surfaces of the component.
The components tested were a bubbler head with a stainless steel threaded nipple; a stainless steel bubbler head; an 18-in.-long plastic-lined flexible connector with brass ferrules at each end; a brass elbow that was used in the filter installation; and a 1/4-in. brass shutoff valve. This typical plot shows the lead concentration versus time for flexible connectors and illustrates how the lead levels drop with multiple exposures. To evaluate the potential impact of an individual component’s contribution to lead levels at the tap, Table 1 was prepared to illustrate the potential contribution in two scenarios — typical and worst case — in a first-draw 250-mL sample.
The components with the highest potential to contribute high levels of lead were the plastic-lined flexible connector with brass ferrules at the ends and the brass elbow. Either of these components could have caused a first-draw 250-mL sample to exceed 10 ppb of lead. In addition, the laboratory results indicate that the lead contribution from several of these end-use plumbing components could result in elevated lead levels, particularly when these components have not been previously exposed to water. It is interesting that small amounts of lead were contributed by the new stainless steel bubbler head. These were attributed to surface impurities. Laboratory and field testing indicate they could be easily removed with flushing.
This applied research effort helps the Seattle Public Schools in several ways. It helps identify the likely causes of high lead levels in water in new installations so that solutions can be implemented immediately. Different components that do not contribute lead, or contribute minimally, were specified. Bubbler heads are being preconditioned by a thorough flushing program in the shop before they are taken to the field for installation. A major flushing program also takes place in schools after the installation of plumbing component replacements and before sampling.