Environ. Sci. Technol. 2008, 42, 760–765
The Release of Lead from the Reduction of Lead Oxide (PbO2) by Natural Organic Matter YI-PIN LIN* AND RICHARD L. VALENTINE Department of Civil and Environmental and Engineering, University of Iowa, Iowa City, Iowa 52242-1527
Received August 9, 2007. Revised manuscript received November 14, 2007. Accepted November 15, 2007.
PbO2 has been identified as an important scale in some distribution systems that historically use lead service lines and free chlorine for maintaining a disinfectant residual. The stability of this highly insoluble scale with respect to its reductive dissolution may play an important role in lead release into drinking water. In this study, we investigated the release of lead from a commercially available PbO2 in the presence of natural organic matter (NOM) using a hydrophobic acid extracted from the Iowa River. Experiments were conducted using synthetic solutions with different NOM concentrations, solution pH, and NOM samples with different levels of prechlorination. It was found that release of lead from PbO2 occurred both in solutions with and without NOM, and the extent of lead release increased with increasing NOM concentration and decreasing pH value. Furthermore, the released lead was Pb(II) and not particulate PbO2 conclusively showing that reductive dissolution occurred. Prechlorination of NOM reduced the rate of lead release. Our results indicate that PbO2 can be reduced both by water and NOM. Characterization of final solid phasesbyscanningelectronmicroscopyandX-rayphotoelectron spectroscopy are also presented.
Introduction Corrosion of lead bearing materials in the distribution system and the subsequent release of lead into drinking water pose a concern to public health. A prevailing paradigm for lead corrosion considers lead release as a result of the electrochemical oxidation of the zerovalent lead in the material. This can occur by reaction of oxidants such as oxygen or free chlorine if one considers the standard state potentials (1, 2). Pb(II) species have long been thought to regulate the release of lead from lead bearing materials into drinking water. The reactions of Pb2+ with water constituents may result in the formation of a large variety of both soluble species and solid phases. Formation of Pb(II) solid phases, such as PbO (litharge), PbCO3 (cerrussite), Pb3(OH)2(CO3)2 (hydrocerrussite), and Pb5(PO4)3OH (hydroxypyromorphite), can reduce the concentration of soluble Pb(II) species by acting as a sink and by forming a passivating layer (3–7). The passivating layer may reduce the rate of electrochemical oxidation by creating a resistance to mass transfer of the oxidant as well as to the transport of soluble Pb(II) species to the aqueous phase. In fact, mass transfer resistance may * Corresponding author phone: (319) 594-1103; e-mail: yipinlin@ engineering.uiowa.edu (
[email protected], effective 2008;
[email protected] (R.L.)). 760
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 3, 2008
be so large as to entirely govern lead release, which can then be modeled as a mass transfer controlled process (8). Recent attention has been given to the dramatic increase of lead concentrations in Washington D.C. drinking water in 2003 (9). Examination of the pipes collected in that system revealed that PbO2, a Pb(IV) solid phase, was formed inside of the pipes over the period while the system was using free chlorine as the disinfectant (9). Field evidence supports the idea that switching from free chlorine to monochloramine for disinfection was a critical reason for the elevated lead concentrations (10). The change of disinfectant was believed to have changed the type of lead solid phases normally formed on lead bearing materials (11). The presence of free chlorine was hypothesized to reduce lead release by maintaining a layer of insoluble PbO2 that passivated the pipe surface (12). Monochloramine, on the other hand, results in a reduced oxidative environment of the water and a condition where the formation of the critical PbO2 is eliminated (10, 11) or the dissolution of pre-existing PbO2 is possible (9). It is hypothesized that common reducing agents in water, such as NOM, may lead to the reductive dissolution of PbO2 and its reduced moieties may survive in the presence of monochloramine. In fact, it has been shown that chlorinated organic byproduct from chloramination of fulvic acid solutions remain tied up in high molecular weight compounds, while those from chlorination are much smaller fragments (13), indicating the preservation of main structure and possibly the reductive potential of NOM in chloraminated solutions. A reduction in the formation of many disinfection byproducts, which is a major reason why many utilities have switched from free chlorine to monochloramine, is also a consequenceoftheweakeroxidativepropertyofmonochloramine (14). Dryer and Korshin (15) recently provided evidence of lead release from PbO2 in the presence of NOM, although the nature of the released lead was not ascertained. Dissaggregation of PbO2 aggregates induced by the presence NOM could not be ruled out. While difficult to precisely define, NOM can be oxidized and reduced by many redox active substances (16–19). In addition, NOM is a well-known inhibitor for the crystal growth of minerals. It has been shown that NOM can inhibit the formation of cerrussite and hydrocerrussite on lead coupons and lead-containing plumbing materials and result in high soluble lead concentration in synthetic drinking water (20, 21). The main objective of this study was to characterize the stability of PbO2 in water and to resolve the potentially competing process of NOM induced disaggregation with release of small lead oxide particles from the reductive dissolution of PbO2 which forms soluble Pb(II). Lead release was not studied extensively in the presence of a disinfectant residual because we wanted to evaluate the role of NOM and its true reductive potential in the absence of competing reactions involving any disinfectant. We present experimental results that show the influence of pH, NOM concentration, and the effects of NOM prechlorination on lead release. Prechlorination was evaluated as a way to investigate changes in the reductive potential of NOM because it is practiced as part of the chloramination process. Final solid phases were also examined by scanning electron microscopy (SEM) and X-ray photoelectron spectroscopy (XPS).
Material and Methods Chemicals. All solutions were prepared by reagent grade chemicals and deionized water obtained from a Barnstead ULTRO pure water system (Barnstead-Thermolyne Corp.). Commercially reagent grade PbO2 (Fisher Scientific) was used 10.1021/es071984w CCC: $40.75
2008 American Chemical Society
Published on Web 01/05/2008
in this study. XRD analysis indicated that this PbO2 is plattnerite. Both plattnerite (tetragonal polymorph) and scrutinyite (orthorhombic polymorph) can be formed from the oxidation of Pb2+ by free chlorine (12) and present in the distribution system (22). These particles have a specific surface area of 4.14 m2/g determined by the 7-point N2-BET method. Primary particle size was as large as 5 µm. However, these large particles appeared to be aggregates of much smaller particles with size ranging from 100 to 300 nm as determined by SEM. NaHCO3 was used as the source for the inorganic carbon and buffer in our solutions. NOM Sample. Iowa River hydrophobic acid (IRHPOA) was used as the NOM source. IRHPOA was obtained by first concentrating Iowa River NOM using a RealSoft PROS/2S reverse osmosis unit (Stone Mountain, GA) followed by conventional XAD 8 resin extraction. Detailed description of the extraction procedures can be found elsewhere (23). The IRHPOA dry sample was obtained by freeze-drying the XAD 8 extract. This IRHPOA sample has a 40% carbon content and a specific UV absorbance at 254 nm (SUVA254) of 2.14 L/mg-m. Lead Release Experiments. Experiments were conducted using 250 mL amber glass bottles at 25 °C. Reagent grade PbO2 was used and 1 mM of NaHCO3 was added to simulate the alkalinity normally encountered in the finished water. Variable examined include NOM concentration (0–20 mg/L as NOM), solution pH (6.0–9.0) and NOM oxidation state. Solution pH values were adjusted to desired values by 0.1 N HCl and NaOH. After filling 250 mL amber bottles with the solutions, they were covered with aluminum foil to avoid any possible light-induced reactions. The bottles were placed on a shaker rotating at 200 rpm. Five mL samples were removed for analysis of released lead over a period of up to 28 day. Solution pH values were also measured and were maintained at the target pH ( 0.1 by addition of acid or base. Due to the sampling for lead analysis, the experiments were conducted in the presence of head space. The pH generally did not vary by more than 0.2 pH unit from the desired value. Selected experiments were run in duplicate with relative error within 20%. Detailed procedures for the preparation of prechlorinated NOM solutions can be found in the Supporting Information. UV absorbance of these solutions and the original stock (without prechlorination) were scanned from 200 to 600 nm to provide UV absorbance information. Chlorination of NOM solution has been shown to significantly oxidize functional groups in NOM macromolecules and decrease the UV absorbance (24), which is believed to correlate to NOM aromatic content and reactivity (18, 25, 26). Lead release experiments were conducted using the procedures described above. The following solution composition was employed: 1 mM of NaHCO3, pH 7.0 ( 0.1, and NOM ) 10 mg/L of the prechlorinated NOM samples. Analytical Methods. Lead release from PbO2 was measured using two approaches: lead that can pass through membrane with 0.2 µm pores was determined by graphite furnace atomic absorption spectroscopy (GFAA) and Pb(II) was determined by anodic stripping voltammetry (ASV). Detailed procedures for the GFAA measurement can be found in the Supporting Information. In controlled experiment using D.I. water spiked with 4 mg/L of PbO2, no measurable lead was detected by the GFAA method indicating that small PbO2 particles did not pass through the 0.2 µm membrane pores, probably due to their aggregation. However, release of tiny PbO2 particles by dissagregation of larger particles under long-term experiments in the presence of NOM cannot be ruled out based solely on data obtained by GFAA. To resolve the speciation of the released lead, additional unfiltered samples were collected in some experiments and analyzed for Pb(II) by ASV (Nano-Band Explorer II, Trace
FIGURE 1. Effects of NOM concentration on lead release from PbO2. Released Pb was measured as a function of time. pH 7.0 ( 0.1, PbO2 ) 4 mg/L, total inorganic carbon ) 1 mM NaHCO3. Iowa River hydrophobic acid was used as the source of NOM. (carbon content ) 40%). The Pb concentrations shown by solid triangle (2, in the presence of 20 mg/L NOM) were measured by ASV. The rest of the Pb concentrations (
