The recent identification of elevated levels of lead in Flint, Michigan’s drinking water have drawn attention to the use of lead in the drinking water infrastructure of the United States. However, plumbing used for the conveyance of drinking water has been constructed of lead for millennia. Indeed, “plumbing” derives its name from the Latin word for lead, “plumbum.” This article is intended to provide an introduction to the long history of lead in drinking water infrastructure, the complexities of its potential release into water, and associated management considerations.
Lead, a bluish metal common in the Earth’s crust, is among the first metals utilized by humans. Lead is believed to have been used by ancient Egyptians as early as 5,000 B.C. Several unique properties prompted lead’s early and continued use in engineering applications, including drinking water infrastructure. In addition to its low melting point, lead is cast easily into molds and is malleable at room temperature, providing for ease of manipulation and fabrication. Lead also occurs in accessible ore deposits and has a high resistance to corrosion, such that it is relatively easy to obtain and is durable. For these reasons, lead has a long history of use in the construction of pipes and associated apparatus for the conveyance, storage, and drainage of water.
The Roman aqueducts are famous for conveying water to ancient Roman cities, but the Romans’ use of lead piping to distribute and drain that water was equally prominent in many Roman cities. Ancient Romans are believed to have had the world’s most advanced plumbing systems at the height of their empire. Romans formed water pipes by hammering lead into sheets, rolling the sheets into pipes, and filling pipe joints with molten lead to form a standardized water infrastructure. When Mt. Vesuvius erupted in 79 A.D. and quickly engulfed the Roman cities of Pompeii and Herculaneum in ash, it essentially preserved the towns—including the lead pipes in their plumbing systems, drainages, and ancillary components—for centuries until these artifacts were later unearthed by archaeologists.
Beyond the Roman Empire, lead pipes were historically used in distribution systems throughout much of the modern world. The use of lead compounds for pipes, plumbing fittings, and as solder in water distribution systems spans centuries and has been described by the World Health Organization (WHO) as having been “almost universal” in practice. World Health Organization, Lead in Drinking-Water 1 (2011). Lead pipes were the standard for service connections in many cities of the United States through at least the 1950s and were installed in some areas through 1986.
Lead pipes were often chosen to transmit water in those areas in which water sources were naturally corrosive. Today, with the benefit of modern science regarding lead-water release mechanisms, this appears to be a paradox. However, many decades or centuries ago the science of lead had yet to develop. Lacking detailed knowledge of the release of lead into drinking water, early engineers primarily chose lead as the most durable metal for conveying water. Other pipes, such as various grades of steel or galvanized pipe, would rapidly degrade if used for the conveyance of corrosive water, necessitating costly repairs and replacements.
As the science of lead and its health effects emerged over time, the decision to curtail the use of lead for drinking water applications in the United States ultimately took many years. Lead pipes and lead-containing components continued to be used in some drinking water systems until the U.S. Environmental Protection Agency (EPA) enacted regulations restricting their use. In the 1970s, the EPA evaluated lead exposures in response to growing concerns regarding lead’s health effects. Notably, lead had not been used solely in plumbing applications, but had gained widespread use. Due to its unique and “outstanding” properties, lead was considered “an essential commodity in the modern industrial world” at this time. Michael King et al., and updated by staff, Lead and Lead Alloys, in Kirk Othmer Encyclopedia of Chemical Technology (2005). As of 1978, the world’s consumption of lead exceeded 4.4 million metric tons, of which the United States consumed approximately 22 percent. Prominent uses of lead included paints, gasolines, batteries, ammunition, bearings, and caulking. Because of the widespread use of lead in so many applications, EPA expressed concerns regarding cumulative lead exposures, but noted that there was “still controversy surrounding tolerable blood-lead levels and the significance of environmental sources of lead exposure.” EPA, The Environmental Lead Problem: An Assessment of Lead in Drinking Water from a Multi-Media Perspective 2 (May 1979). Still, EPA ultimately enacted regulations to remove two significant uses of lead. EPA banned leaded paints in 1978 and phased out the use of leaded gasoline from approximately 1974 to 1996.
While the major uses of lead paint and leaded gasoline may have initially overshadowed the use of lead in drinking water infrastructure, EPA ultimately banned the installation of lead pipes for drinking water infrastructure in 1986 via amendments to the Safe Drinking Water Act (SDWA). These amendments also limited the allowable lead content of brass alloys and solders used in drinking water applications, requiring them to be “lead free” as defined by regulation. However, the SDWA still allowed brass containing up to 8 percent lead to be considered “lead free” and, therefore, used in drinking water plumbing for over 25 additional years. In 2011, the Reduction of Lead in Drinking Water Act redefined “lead free” brass as containing less than 0.25 percent lead, effective January 2014.
What Is the Extent of Lead Drinking Water Infrastructure in the United States?
The installation of leaded pipes and fixtures in many cities and towns of the United States over the greater part of a century has resulted in a legacy of lead-containing drinking water infrastructure. Although lead was not commonly used to construct large, rigid water mains, lead pipes were used commonly as smaller water distribution pipes, such as residential service lines and internal plumbing, because of lead’s malleability and machinability properties.
Lead service lines are relatively concentrated in the upper Midwest and Northeastern sections of the United States. In 1991, an estimated 10.2 million lead service lines existed in approximately 15,000 community water supplies located throughout the United States. Following removals over the two succeeding decades, national surveys conducted in 2011 and 2013 identified approximately 6.1 million lead service lines spanning 11,200 community water supplies (30 percent of the total). These lead pipes serve an estimated 15 to 22 million people located throughout the United States.
As noted, in addition to lead pipes, appreciable lead was used in plumbing throughout the United States as a component in brass fixtures and solders. Lead was added historically to brass alloys to enhance the seal of threaded joints. Various grades of brass components have been used in many plumbing applications, including valves and connectors. Brass also has been used commonly in the internal apparatus of larger plumbing fixtures and appliances such as faucets, water coolers, refrigerators, and coffee makers. The “yellow brass” alloys used in faucets and other internal applications historically have contained higher lead concentrations than the “red brass” alloys used in plumbing components such as joints and valves.
Plumbing solders commonly consisted of tin-lead alloys that were heated and applied to piping connections as a filler metal to prevent leaks. Notably, lead-containing solders and brass fixtures were used commonly to connect metal pipes that do not contain lead, such as copper piping.
How Is Lead Released into Drinking Water?
The release of lead from plumbing components into drinking water is more than a simple function of lead–water solubility. Rather, the release of lead is a complex function of metallurgy and water chemistry that depends on the type and connections of plumbing components, as well as the chemical properties and flow of water through those components. Lead can enter drinking water from several processes:
- The leaching of dissolved lead from plumbing components such as lead pipes, brass fixtures, and solder connections;
- The abrasion and detachment of particulate lead from corrosion of pipe plumbing components;
- Lead interactions with scales and films that precipitate along the inside of the pipe walls and are released through dissolution or abrasion; and
- Electrochemical reactions between metal connections and deposition points.
The following graphic illustrates these processes.
Id. at 3.
Certain aspects of plumbing metallurgy alone can release lead into drinking water. Joints of lead and copper plumbing can form an electrochemical reaction called “galvanic corrosion,” in which lead electrochemically corrodes (as the less noble metal). This phenomenon can accelerate the corrosion of otherwise relatively stable lead pipes near the copper connection points and can be enhanced by water chemistry, such as chloride and sulfate content. Galvanic corrosion can also occur when dissolved or particulate copper reattaches to the walls of downstream lead pipes, forming a micro-galvanic cell at the deposition point.
Released lead has been demonstrated to interact with iron corrosion scales in downgradient steel or galvanized plumbing, resulting in the reattachment or absorption of lead particles to the pipe walls, followed by the potential for eventual release into drinking water. Further, the interaction of lead with pipe scales and biological films also can affect whether lead is released into drinking water. In particular, manganese and iron-rich scales have been identified as affecting the release of lead into drinking water. Accordingly, both the metallurgy of the piping system and geochemical aspects of scales formed therein can complicate the release of lead from piping.
The properties of the water conveyed through lead plumbing are equally important—if not more important—than the lead content of the plumbing itself. Numerous reactions occur in the drinking water system that are governed primarily by the chemistry of the water. Despite lead’s high resistance to corrosion, oxidants in the water (e.g., dissolved oxygen, chlorine, or chloramines) inevitably can react with the metallic lead in the pipes to form a more soluble form of lead. This soluble lead enters the water and can undergo further reactions. The presence of carbonates in the water is important to these reactions and the resulting solubility of the lead complexes that are formed. Carbonates are influenced by the natural mineral content of the water, as well as by the level of dissolved organic carbon present.
When required, water quality engineers apply treatment chemicals to drinking water at the treatment plants to inhibit the accumulation of the soluble forms of lead. One such application is the use of passivation, in which chemicals are added to form a film on the inner walls of lead piping. These chemicals, typically phosphate inhibitors, encourage the formation of lead solids as a scale to discourage the entry of dissolved lead into the passing water.
Despite attempts to manage drinking water parameters to control lead, chemical changes can occur that alter the delicate balance of lead passivation and associated corrosion. In addition to controlling lead, treatment plant operators also must address other important water quality considerations, such as the application of disinfectants (typically chlorine) to control microorganisms. For example, the application of chemicals for water disinfection purposes, as well as seasonal variations in the source water itself, can alter the chemistry of the water, resulting in changes that can mobilize dissolved lead. The concentration of lead in drinking water also can vary with seasonal temperature fluctuations. Accordingly, many considerations must be balanced to provide for disinfected water with a low lead content. These same considerations affect the leaching of lead from brass fixtures and solder connections located within buildings as an indoor environment consideration for end users.
In the United States, lead in drinking water is regulated primarily under the Lead and Copper Rule (LCR), which was enacted in 1991. 40 C.F.R. Part 141 Subpart I. Recognizing the complex interactions associated with the release of lead from drinking water components, EPA devised the LCR as a treatment technique rule to control lead through the processes noted above, rather than a risk-based health standard. Specifically, EPA has described the thought process behind this unique regulation as follows:
USEPA decided to use a regulatory framework with a lead action level rather than a maximum contaminant level (MCL) so that potential exposure from lead service lines, lead solder, and lead plumbing fixtures could be taken into account when determining if corrosion control was effective. Therefore, the LCR established a set of trigger conditions or action levels, which if the water system does not meet, it must take additional actions. For lead, if the 90th percentile lead concentration observation in a water system’s compliance monitoring is greater than 15 parts per billion then the water system must engage in active public education, evaluate the need for or success of existing corrosion treatment, and remove lead service lines.
Leah Harnisch, Adam T. Carpenter, & Shawn Moran, Comparing Water Source Knowledge in Cities that Exceed the Lead Action Level, Journal American Water Works Association (AWWA), 2–3 (unedited manuscript submitted Aug. 24, 2016).
Because of the controls implemented by the LCR, serious problems associated with lead in drinking water were believed to be largely historical circa the early 2000s. At that time, blood lead levels in the United States had declined 80 percent due largely to the ban of lead in paint and the phase out of lead from gasoline. Lead levels in drinking water were generally much lower than had been historically experienced due to the LCR controls, and drinking water generally was considered a minor contributor to blood lead levels. However, in approximately 2001, a change in water treatment processes in Washington, D.C., inadvertently resulted in escalating levels of lead leaching into the city’s drinking water through approximately 2004. The concentrations of lead identified in certain drinking water samples during this event spurred a 90th percentile LCR result several times the lead action level of 15 parts per billion (ppb). The source of the issue was identified through subsequent research to have been associated with a treatment process change to chloramine disinfection (from prior use of chlorine) that the city had implemented. The change was conducted to prepare for the requirements of the EPA’s Disinfection Byproduct Rule, but inadvertently resulted in complex chemical reactions with the particular types of scales present in the drinking water infrastructure. While the exact mechanism of the chemical reactions still was being debated years later, the switch from chlorine to chloramine disinfection was identified as the cause.
The well-publicized crisis identified in Flint, Michigan, in 2015 was also the result of a change in water chemistry. In April 2014, the bankrupt city of Flint switched its water source from the city of Detroit to the Flint River—which is more corrosive. In addition to this change, the city of Flint did not implement corrosion control measures required by the LCR, exposing the protective passivation layers that had accumulated in the system to corrosive waters. As a result, significant quantities of lead—up to 13,200 ppb—were leached ultimately into the water supply of certain residences. This change not only released significant levels of lead into the drinking water of many homes, but also drastically aged the city’s water infrastructure, resulting in water main breaks. Following scientific studies and agency involvement, EPA issued an emergency administrative order in January 2016 to enact additional measures to correct the situation. Several criminal cases have been filed in association with the Flint matter alleging misconduct in the city’s water treatment operations during this period.
With a dearth of information on legacy plumbing and recent changes to the definitions of “lead free” components, as well as a broad range of available drinking water sampling methods, municipalities and particularly drinking water end users are facing difficult questions regarding how to assess, manage, and communicate risk associated with drinking water quality.
Because of the characteristics discussed above, the presence of non-lead piping alone in a building is insufficient to eliminate lead-bearing plumbing components from consideration. Neither is the age of the building, unless it was constructed very recently. As noted, brass containing up to 8 percent lead could still be used in water distribution components until as recently as 2014, when the lead content of such “lead free” components was reduced to 0.25 percent lead by the 2011 Reduction of Lead in Drinking Water Act. As a result, the overwhelming majority of buildings in the United States likely have some number of brass components in association with water services that contain appreciable lead above the current “lead free” standard. In addition, buildings constructed before 1986 likely contain plumbing comprising lead-containing solders and other components associated with legacy construction practices. Studies have shown that both leaded brass and solder can contribute to appreciable lead in drinking water under certain conditions.
The National Sanitation Foundation (NSF) and the American National Standards Institute (ANSI) developed certification protocols to verify that commercial and consumer products on the market meet the current “lead free” content requirements as well as protocols to certify that water filtration systems are capable of treating water to meet the current drinking water standard for lead in the United States. As the allowable lead content has changed over time, so have the standards and certification symbols used by the NSF. Faucets and plumbing fixtures that contact potable water must meet one of two current standards: NSF/ANSI 61 or NSF/ANSI 372. Although the methodology of the two standards varies slightly (e.g., leachate concentration versus physical lead content of the fixture), both standards provide certification intended to meet the current lead content requirements. Other standards developed by the NSF apply to lead content in food-contact scenarios or specifically to beverage services.
In the United States there are eight third-party certification bodies that provide product certification for manufacturers of drinking water system components and plumbing fixtures. Each of these bodies uses a registered trademark to identify that the certified component meets the requirements listed in the NSF standards. The table below provides an overview of the various certification marks that identify components that meet requirements.
Source: EPA, How to Identify Lead Free Certification Marks for Drinking Water System & Plumbing Products, EPA/600/F-13/153e, at 2 (revised Mar. 2015)
In addition to the recent changes regarding certification of “lead free” components, the regulatory landscape continues to change, creating additional uncertainty and complexity. EPA recently announced its intention to revise and strengthen the LCR, stating that “the regulation and its implementation are in urgent need of an overhaul.” U.S. Envyl. Prot. Agency, Office of Water, Lead and Copper Rule Revisions White Paper 3 (Oct. 2016). EPA announced that it anticipates that the LCR revisions “will include both technology-driven and health-based elements that focus on proactive, preventative actions to avoid high lead levels and health risks.” Id. at 8. EPA also stated that it plans to incorporate a health-based benchmark in the revised LCR to strengthen protection, noting that the public often mistakes the 15 ppb action level specified in the LCR as having significance in terms of health effects. Rather, EPA devised the 15 ppb action level for use as a screening level to assess when certain treatment actions are needed in a water system. Although the outcome is currently uncertain, EPA’s stance on this matter has suggested that its health-based benchmark would likely be lower than the current action level. However, recent administration changes following the 2016 election—including restrictions on new regulations recently announced as of this writing—have created uncertainty in the future of the LCR revisions.
One solution that EPA has considered for revision of the LCR is a potential role for the use of filters to address the risks of lead at a household level. In addition to providing certification protocols for drinking water system components and plumbing fixtures, the NSF also has developed standards for the treatment and filtration of drinking water. NSF/ANSI Standard 53 for Drinking Water Treatment Units is the primary standard for evaluating and certifying drinking water treatment systems for the reduction of contaminants from drinking water. NSF/ANSI Standard 58 applies to water treatment systems that use reverse osmosis technology. Treatment systems and filters are tested and certified to either of these standards to ensure that the treatment systems reduce contaminant concentrations to meet the requirements of the standards. Although other NSF/ANSI standards have been developed for water filtration (e.g., NSF/ASNI Standard 42), only NSF/ANSI Standards 53 and 58 provide certification for lead reduction. While NSF/ANSI standards are recognized internationally, numerous other international requirements and certification standards are available.
NSF-certified water filters for lead reduction are evaluated to provide a 90 percent reduction in lead concentration, meaning in order to reduce levels to below the current EPA action level, the maximum concentration of lead that filters are certified to treat is 150 ppb. Multiple types of filters have been certified by the NSF including pour-through pitcher or carafes, faucet mounted units, countertop units connected to sink faucets, plumbed-in units to separate taps, or in-line refrigerator units.
Although engineering controls such as the removal and replacement of lead-containing fixtures or the installation of filters may reduce the potential for exposure, administrative controls, including posting signs stating that certain fixtures are to be used only for non-potable purposes, also may be used to reduce exposures.
While sampling is conducted under the LCR, it remains a treatment technique rule as previously described. It provides a measure of the overall effectiveness of the corrosion control measures undertaken by a municipality by sampling a small subset of homes that have lead service lines. Because of site-specific variations in building fixtures and the complexities of how lead is released into drinking water, the LCR does not provide sampling data that can be extrapolated accurately to understand lead concentrations in drinking water of an individual building or from a given fixture or appliance within that building. This nature of lead in drinking water—that it emanates from localized plumbing and fixtures closest to the end user—makes it difficult to regulate and monitor lead from individual receptor locations. Accordingly, identifying the sources of lead requires careful study design and more complex sampling strategies than are currently required by the LCR.
Another complexity is that the source of lead often consists of privately owned pipes and fixtures within a building. This results in regulatory complexities when addressing lead in drinking water. Whose responsibility is it to address the issue? The total lead content in water at the tap can result from a combination of sources including municipal water lines and private plumbing/fixtures located within a building. Attempts to identify the location of lead sources in a water distribution system may require a detailed inspection, review, and an assessment protocol consisting of a series of samples collected from various points in the system. Assessment of lead content in water is further complicated by variables that can result in higher or lower lead levels. Water that has been stagnant has longer contact with lead-containing materials than water from a tap that has been used recently. Temperature fluctuations also can affect lead concentrations in drinking water. These, and other variables, need to be considered when assessing lead in a water distribution system.
Lead had been used in drinking water infrastructure for millennia because of its simplicity. It was easy to get, malleable, and durable. After thousands of years of use, the legacy of lead in plumbing is an extraordinarily complex combination of chemistry, health risk assessment, and public policy.