Environ. Sci. Technol. 2009, 43, 3278–3284
Interactions of Pb(II)/Pb(IV) Solid Phases with Chlorine and Their Effects on Lead Release HAIZHOU LIU,* GREGORY V. KORSHIN, AND JOHN F. FERGUSON Department of Civil and Environmental Engineering, University of Washington, Box 352700, Seattle, Washington 98195-2700
Received November 10, 2008. Revised manuscript received March 04, 2009. Accepted March 06, 2009.
This study examined changes of colloidal properties and lead release from representative solid phases of lead (IV) (PbO2) and lead (II) (hydrocerussite, cerussite) and during the oxidation of the lead (II) solids by chlorine. Chlorine is determined to cause the ζ-potential of lead (II) solids to undergo significant changes apparently associated with the generation of Pb(III) intermediates that are formed before the PbO2 phase becomes abundant enough to be morphologically distinct. Simultaneously with the changes of the ζ-potential, a pronounced decrease of lead release from hydrocerussite takes place. In contrast with that, lead release from cerussite undergoes a transient increase during the oxidation of that solid by chlorine. The existence of differences in processes governing lead release from these Pb(II) solids is supported by SEM data showing different patterns of morphological changes of the cerussite and hydrocerussite crystal surfaces.
Introduction In 2002, lead levels in the Washington, DC drinking water system were found to consistently exceed EPA regulatory levels (1). The association of this phenomenon with the reduction of lead (IV) species (2-4) necessitated research to determine mechanisms that govern lead release from and the stability of corrosion scales formed on the surface of lead-containing materials present in drinking water distribution networks. Several studies have shown that these scales are predominated by coexisting Pb(II) and Pb(IV) phases (3-6). In the presence of chlorine and the absence of phosphate corrosion inhibitors, these phases include hydrocerussite Pb3(CO3)2(OH)2, cerussite PbCO3, sometimes litharge PbO (7, 8), and lead dioxide PbO2 (9-11). Recent research has shown that the oxidation of hydrocerussite and cerussite by chlorine is accompanied by the formation of PbO2 on the surface of corroding lead (3, 12). The solubility of PbO2 is very low, and it is accordingly deemed to be beneficial for lead control. However, given the practically universal coexistence of PbO2 and Pb(II) minerals in corrosion scales (11, 12), it is unclear to what extent Pb(II) minerals need to be oxidized to effectively control lead release. On the other hand, a significant portion of Pb release, notably that during the Washington, DC event, is associated with colloidal particles with varying sizes (13-16). Relative contributions of Pb(II) and Pb(IV) species in the structures of submicron * Corresponding author phone: (206)543-0785; fax: (206)685-9185; e-mail
[email protected]. 3278
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 9, 2009
particles that constitute particulate lead have not been ascertained. However, it has been shown that similarly to Pb(II) solids (17-19), PbO2 can be reduced in the absence of chlorine by NOM and other reductants, undergo colloidal mobilization (2, 4, 20), and contribute to a possibly significant fraction of lead exposures (13). While plumbosolvency models (21, 22) rigorously describe the behavior of Pb solutes, they do not model the mobilization of lead-containing particles. In addition, “soluble” lead, which is operationally defined as its fraction passing through a 0.45 µm filter, can contain a considerable part of colloidal particles that are not true solutes. Little information is currently available to determine how chemical conditions (notably, pH, concentrations of carbonate, chlorine and/or chloramine) affect the release of these particles and the colloidal behavior of lead solids in general. The goal of this work was to examine effects of chlorine, pH, and carbonate on colloidal lead release in these systems, with specific attention to the release of colloidal particles during the oxidation of hydrocerussite and cerussite by chlorine. The oxidation of these solids is a necessary step to form PbO2. This process takes place in drinking water distribution systems where lead-containing materials are exposed to water containing free chlorine residual and also during maintenance of systems that use chloramine as the secondary disinfectant. Experimental results presented below will help understand the impact of chlorine on the colloidal mobilization from both Pb(II) and Pb(IV) solids.
Materials and Methods A detailed description of the experimental setup was presented in a previous published paper (23) and the Supporting Information (SI). Interactions of chlorine with and measurements of lead release from hydrocerussite and cerussite were examined in the ranges of pH and carbonate concentrations (TOTCO3) in which these solids were thermodynamically stable. The only exception was the measurements of ζ-potentials of hydrocerussite and cerussite. In order to determine their points of zero charge, their ζ-potentials were measured at pHs beyond the range in which hydrocerussite or cerussite was thermodynamically stable. Such measurements were done rapidly ( 7.3 for comparison with hydrocerussite.) (A) Impact of total carbonate. PbCO3 6.36 × 10-4 M, ionic strength 0.01 M. (B) Impact of ionic strength. PbCO3 6.36 × 10-4 M, total carbonate 0.001 M.
Synthesized PbO2 was prepared via the oxidation of 1.5 g/L hydrocerussite by 7 × 10-3 M NaOCl in a total volume of 500 mL at a controlled pH of 7.5 (Pb(II)/chlorine molar ratio 0.83). As the reaction progressed, the white-color hydrocerussite particles became darker. After agitation for approximately 24 h, the suspension had a dark brown color and no further decrease of chlorine concentration. 1 M HClO4 was then added to the suspension to decrease its pH to 2.5 and dissolve residual Pb(II) solids. The suspended PbO2 solids were filtered to deposit them onto a 0.45 µm Millipore cellulose membrane filter and rinsed with 1 L DI water to remove possibly remaining Pb(II) species or acid. The retained PbO2 particles were lifted from the filter and freezedried for 24 h at -40 °C and 0.133 mBar using a Labconco FreeZone 4.5 freeze-dry unit. The structural identification of various solid phases was performed by XRD using a Philips 1820 diffractometer. XRD examination showed that the dried particles were R-PbO2 (scrutinyite).
50 mV within 2 units of pH suggest that protonation-active Pb(II) hydroxo species exist on the surface of hydrocerussite particles. The negative shift of the ζ-potential of hydrocerussite at increasing TOTCO3 values suggests the formation of negatively charged Pb(II) carbonate surface species whose existence was postulated (14, 19). The increase of the ζ-potential of hydrocerussite at increasing ionic strength is consistent with the compression of the electric double layer (24). For cerussite, there was relatively little change of the ζ-potential with pH (Figure 2A), although some effects of the ionic strength were observed. In general, the results for cerussite were similar to those reported in preceding studies (19). The differences between the surface properties of hydrocerussite and cerussite are evidently due to a higher prominence of protonation-active surface sites on the surface of hydrocerussite and very different structural organization of these Pb(II) solids. ζ-Potential and Colloidal Behavior of PbO2 Phase. The behavior of PbO2 particles was examined at chlorine concentrations ranging from 0 mg/L to 50 mg/L. The ζ-potential of suspended PbO2 was initially decreasing, but it stabilized after approximately 1 day of exposure both in the presence and absence of chlorine (Figure S1 in the SI). Following the stabilization period, the ζ-potential of suspended PbO2 became increasingly negative at higher chlorine concentration and relatively less sensitive to pH variations (Figure 3A). This behavior possibly reflects the sorption of HClO/OCl-
Results and Discussion ζ-Potential of Pb(II) Solid Phases. To quantify colloidal properties of Pb(II) phases, electrophoretic mobility of hydrocerussite and cerussite was measured at different pHs in the absence of chlorine. Results presented in Figure 1 show that, for hydrocerussite, there was a significant decrease of the ζ-potential with increasing pH, increasing total carbonate concentration (Figure 1A) and decreasing ionic strength (Figure 1B). Changes of the ζ-potential that exceeded
VOL. 43, NO. 9, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3279
FIGURE 3. Colloidal behavior of PbO2 in the presence of chlorine. (A) Impact of chlorine concentration on ζ-potential of PbO2 after 1 day. PbO2 5.8 × 10-4 M, ionic strength 0.01 M, TOTCO3 0.001 M. (B) Impact of chlorine concentration on colloidal lead release from PbO2. PbO2 5.8 × 10-4 M, ionic strength 0.01 M, TOTCO3 0.001 M, pH 7.3. on the surface of PbO2 and possibly other changes of PbO2 surface properties. The nature of these changes will be discussed elsewhere. The increase of the negative surface charge on PbO2 particles in the presence of chlorine indicates that if these particles are agglomerates of even smaller structures, they can undergo colloidal dispersion. This phenomenon was observed experimentally (Figure 3B). As the contact time and chlorine dose increased, the concentrations of lead passing through a 0.45 µm filter rapidly increased as well. After 100 h of exposure, PbO2 suspension also became noticeably more opaque visually at chlorine concentrations g13 mg/L. These results indicate that PbO2 tends to undergo colloidal mobilization in the presence of chlorine when its ζ-potential becomes increasingly negative. This observation is in accord with the prior study by Lin and Valentine (2). The development of this effect in real drinking water distribution systems may require longer exposure times. It can also be impacted by the presence of NOM and other reducing species (2, 4, 20). The observed effects of chlorine on the formation of operationally defined soluble lead released from suspended PbO2 particles necessitated comparing the morphology of the PbO2 phase prior to and after its exposure to chlorine. SEM examination (Figure 4) showed that, as was observed in refs 2, 4, and 25, PbO2 particles had sizes about 100 nm, but they were agglomerates of smaller structures, which initially had distinct edges (Figure 4A). Ten days of exposure of these particles to water without chlorine did not induce 3280
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 9, 2009
any distinguishable morphological change. Exposure to 50 mg/L chlorine for 10 days caused little morphological change, but the edges of the PbO2 particles became blunter and the presence of smaller aggregates with sizes
