By Lory Littlefield
Plumbing components such as pipes, valves and fittings are tested according to the requirements of ANSI/NSF Standard 61 (“Drinking Water System Components—Health Effects”) to determine that unsafe levels of contaminants don’t leach into drinking water. Lead is a major contaminant leaching from brass components and this analyte was used to evaluate two methods of testing internally threaded valves and fittings.
Components and materials contacting potable water are potential sources of contamination. Manufacturers of drinking waters system components are required by the Safe Drinking Water Act (SDWA) and plumbing codes to have their products certified by conformity testing agencies to show that they’re in compliance with requirements of Standard 61. This standard covers the health effects in testing of pipe, valves and fittings together with other products contacting drinking water. The standard was developed jointly between NSF International and the U.S. Environmental Protection Agency (USEPA).
‘Real world’ situation
Recently, Underwriters Laboratories (UL) carried out a series of experiments designed to identify the most representative method for the evaluation of threaded fittings using a leachate test protocol. This research was done to develop a Lab Process Guide, which is used internally at UL. The intent of Standard 61 is to provide a “real world” evaluation of the product. Therefore, exposure testing of the interior of the product approximates the real world whereas full immersion of the fitting does not. In considering the real world in-product exposures, one must consider the fact that threaded surfaces on fittings tend to leach chemicals at higher levels than the body of the fitting due to machining of the threaded surface. However, since a majority of the threaded portion is actually enclosed during the installation of the fitting into a piping system, the “real world” evaluation of the fitting shouldn’t include exposure of the entire threaded surface.
The only guidance Standard 61 provides for the exposure of these fitting products is in Section B.2.6.1, which states: “When practical, products/devices shall be evaluated such that only the (exposed) wetted surface is exposed to the extraction medium.” Thus, UL has completed preliminary testing to identify the closest approximation to “real world” application. Several lots of brass valves were exposed with different treatments of the machined, threaded areas. Lead was a component of all of these brass fittings and a readily identified leachate. Therefore, lead was used as the test analyte during these evaluations.
Lots of exposure
Four lots of three valves were exposed according to the requirements of the standard using differing techniques of protecting the normally unwetted surfaces.
One lot of three valves was exposed using RTV silicone rubber (Dow Corning 732) spread over 75 percent of the threads on both ends of the valve to simulate normal field exposure of the threaded portion of the valve. After the RTV silicone rubber on the threads cured, the valves were cemented using RTV silicone rubber to a polyethylene plate, blocking off the lower end. The RTV silicone was allowed to cure over night. The valves were then rinsed and conditioned according to the standard and exposed to extraction using cold water at pH 5 and pH 10.
The exposure was accomplished by filling the valve completely with extraction water and collection was achieved by pouring the extraction water out of the valve body. The three extracts were analyzed for lead using USEPA Method 200.9 and the results normalized to one liter. Section B.10.4 of the standard stipulates a field volume of one liter and a dispersion factor of 0.33. Normalization is a procedure, specified by the standard, which adjusts for differences between the laboratory exposure and actual field use and calculates the projected “at the tap” concentration of the contaminant.
PVC pipe plugs
A second lot of three valves was exposed using a PVC pipe plug screwed into the bottom of the valve until it was watertight. Plugs were used after they were found to have no effect on the concentration of lead. The valves were rinsed and conditioned in the same parameters as the previously tested lot. The number of threads left exposed by the threaded plugs on the bottom end of the valve was noted. A volume of extraction water was added to each valve such that the same number of threads were wetted at the top of the valve as were exposed at the bottom of it. After exposure, the water was collected by pipetting it from the valve body while the volume was noted. In both of these scenarios the normally unwetted threads didn’t contact the exposure water. The extracts were analyzed for lead using the same analytical method, and the raw data were normalized in the same manner described above. Two additional lots of valves were also exposed and analyzed using the same procedures.
The third lot of valves was tested with the threads protected with silicone rubber but the exposure water was pipetted rather than poured from the valve. This was done to determine if pipetting had an effect on the accuracy of the testing.
The fourth lot of three valves was exposed in the same manner as the first three lots but the internal threads weren’t protected from the exposure water in any way. The analysis of the exposure water and the normalization of the data for these lots were also identical to the earlier lots.
Lead and leaching
The data are summarized in Figure 1. For both pH 5 and pH 10 exposure waters, the valves exposed using the plastic plugs had a lower concentration of leached lead. Specifically, for pH 5, the valves with the plugs averaged 1.7 parts per billion (ppb) compared to 51.5 ppb lead for the valves with silicone over the threads. For pH 10 exposure water, the plug-ged valves averaged 13.1 ppb while the valves with silicone rubber on the threads averaged 32.9 ppb lead. Interestingly, valves exposed with no protection on the threads actually had lower levels of lead than valves exposed using silicone rubber. These valves averaged 9.2 ppb lead for pH 5 and 17.6 ppb lead for pH 10.
The increased lead cannot be explained simply by the presence of the silicone rubber because exposure blanks were carefully prepared and analyzed along with the samples. Exposure blanks are part of a series of blank samples analyzed along with the actual samples. Blanks, by definition, don’t contain the sample that’s undergoing analysis. It appears the presence of the silicone rubber actually causes more lead to leach into the exposure water. It may be theorized that the acetic acid generated as the RTV silicone rubber cures causes the lead on the surface of the brass to convert to the soluble lead acetate. There are no plans to test this theory further. Additionally, it’s apparent the layer of silicone rubber wasn’t effective in isolating the brass threads from the exposure water. Because of the large standard deviation in the lead values for valves using silicone rubber, it wasn’t possible to determine if pipetting or pouring the exposure water from the valve had any significant effect upon the measured levels of lead.
This series of exploratory experiments was used to decide the proper procedure for the exposure testing of internally threaded components used in potable water systems that most closely approximates actual field conditions. Based upon this testing, UL developed and uses a test protocol detailing the use of PVC pipe plugs to achieve the conditions specified in Section B.2.6.1 of the standard in the testing of internally threaded fittings and components. After exposure, the extraction water is pipetted from the valves—rather than poured—to avoid contact with normally unwetted surfaces. These experiments also illustrate some of the complexities of the factors that affect the amount of lead leaching into drinking water from lead-bearing brass components.
About the author
Dr. Lory Littlefield is senior chemist with the Environmental and Public Health Group at Underwriters Laboratories Inc. in Northbrook, Ill. His three years with UL was preceded by 20 years of experience in the environmental field, first as a chemist in the Office of Research and Development (ORD) of the USEPA and later as a contractor providing laboratory services to ORD. During this time, he specialized in analytical methods development, quality assurance and laboratory management. His doctorate is in organic chemistry.