Reduction of Lead Oxide (PbO2) and Release of Pb(II) in Mixtures of

Apr 14, 2009 - Department of Civil and Environmental Engineering, University of ... Faculty of Engineering, National University of Singapore, Singapor...
0 downloads 0 Views 1MB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 3872–3877

Reduction of Lead Oxide (PbO2) and Release of Pb(II) in Mixtures of Natural Organic Matter, Free Chlorine and Monochloramine Y I - P I N L I N * ,†,‡ A N D R I C H A R D L . V A L E N T I N E * ,† Department of Civil and Environmental Engineering, University of Iowa, Iowa City, Iowa 52242-1527

Received February 5, 2009. Revised manuscript received March 27, 2009. Accepted March 30, 2009.

The primary focus of this paper is to elucidate the influence of mixtures of natural organic matter (NOM) and free chlorine and NOM and monochloramine on the reduction of PbO2 in drinking water. Parallel experiments using PbO2 particles of two different sizes (∼20 and ∼200 nm) were conducted to explore the effects of particle size on this process. In the absence of NOM, reduction of PbO2 was observed in monochloramine solutions but not in free chlorine solutions. In the presence of NOM, significant Pb(II) formation was observed in disinfectant-free solutions. The release of Pb(II) was suppressed by the additional presence of free chlorine until the point in time when free chlorine was exhausted. Monochloramine also repressed Pb(II) formation in the presence of NOM but not as significantly as free chlorine. The presence of NOM and monochloramine does not necessarily act additively or synergistically due to complex interactions including reduction of PbO2 by NOM, monochloramine mediated reduction of PbO2, and oxidation of NOM by monochloramine. Higher surface area-normalized Pb(II) formation was found in experiments using larger PbO2 particles. The high reactivity generally associated with nanoparticles was not observed in our study.

Introduction Exposure to lead can cause adverse health effects in humans, notably nerve damage and impaired mental development in children and high blood pressure and urinary tract problems in adults (1-3). To prevent human exposure from lead in drinking water, an 90th percentile action level of 15 µg/L was set by the U.S. Environmental Protection Agency in the 1991 Lead and Copper Rule (4). In 2003, hazardous levels of lead were found in Washington, DC.drinking water. The switch from free chlorine to monochloramine to control the formation of disinfection byproduct (DBPs) was believed to trigger this incident (5, 6). Examination of pipes collected from that system and elsewhere indicated that PbO2, a Pb(IV) solid phase with a strong oxidative potential, has been formed on inside surfaces * Address correspondence to either author. Phone: (65) 6516-4729 (Y. P. L.), (319) 335-5653 (R. L.V.); e-mail: [email protected] (Y.P.L.), [email protected] (R. L.V.). † University of Iowa. ‡ Current Address: Division of Environmental Science and Engineering, Faculty of Engineering, National University of Singapore, Singapore 117576. 3872

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

of lead service lines over the period when free chlorine residual was maintained in the finished water (5, 7-9). PbO2 is practically insoluble (6, 10) and may serve as a passivating layer to protect underlying lead pipes and maintain a low level of soluble lead. On the other hand, studies have shown that PbO2 does not form when metallic lead (Pb(0)) and lead ion (Pb(II)) are exposed to monochloramine under normal drinking water conditions (6, 11). Several recent studies have reported on the influence of the individual constituent including NOM, free chlorine, and monochloramine on the stability of PbO2. The influence of mixtures of these constitutes, however, has not been studied. It has been shown that monochloramine can cause the reductive dissolution of PbO2 which is consistent with a mechanism involving an unidentified intermediate formed during monochloramine autodecomposition (12). Reductive dissolution of PbO2 can also occur in the presence of common reducing agents, such as NOM (9, 13). It has been shown that a reduction in the specific UV absorbance at 254 nm (SUVA254), typically seen as a consequence of NOM oxidation, correlates well with a reduction in Pb(II) formation (9). Preoxidation of NOM by free chlorine can oxidize the reductive moieties in NOM and decrease the formation of Pb(II) (13). These studies, however, were conducted in the absence of disinfectant. It is not clear if the overall stability of PbO2 can be deduced from knowledge of the influence of individual components. Possible interactions between NOM and disinfectants on the reduction of PbO2 which represent more practical drinking water conditions have not been evaluated. In this paper we report studies of PbO2 reduction in systems containing NOM and free chlorine, and NOM and monochloramine using two different sized PbO2 at pH 7.0 to fill this knowledge gap.

Material and Methods Lead Oxide. Reagent grade PbO2 (Fisher Scientific) and PbO2 prepared in our laboratory were used in this study. Reagent grade PbO2 was used as received from the manufacturer without pretreatment. Laboratory prepared PbO2 was synthesized by the procedures described in Lin and Valentine (12). X-ray diffraction (XRD) analysis and scanning electron microscope (SEM) images of the two particles are shown in Supporting Information Figure S1. XRD patterns revealed that both PbO2 particles are plattnerite, and SEM images indicated that the individual particle size of reagent grade and laboratory prepared PbO2 particles are about 200 and 20 nm, respectively. Primary particles of both PbO2 appeared to be comprised of smaller particles that had aggregated. The BET specific surface area of reagent grade and laboratory prepared PbO2 were 4.14 and 24.29 m2/g, respectively. NOM Sample. Iowa River RO concentrate was used as the NOM source. This RO concentrate was obtained by using a RealSoft PROS/2S reverse osmosis unit (Stone Mountain, GA). Detailed description of the extraction procedures can be found elsewhere (14). This concentrated NOM solution has a DOC concentration of 50.5 mg/L and a specific UV absorbance at 254 nm (SUVA254) of 2.21 L/mg-m. Lead Release Experiments. Experiments were conducted at 25 °C using separate 125 mL polypropylene bottles without headspace to avoid the exchange of CO2 between the solution and headspace. Solutions were buffered by 1 mM of NaHCO3 to simulate the alkalinity normally found in the finished water. Solution pH values were adjusted by 0.1 N HCl and NaOH. Background solution composition was: pH 7.0, NaHCO3 ) 1 mM and PbO2 ) 10 mg/L. Disinfectants were dosed at the beginning of the experiments without redosing during the 10.1021/es900375a CCC: $40.75

 2009 American Chemical Society

Published on Web 04/14/2009

FIGURE 2. Pb(II) formation vs monochloramine decay in the absence of NOM. ∆NH2Cl ) NH2Cl0 - NH2Clt. NH2Cl0 is the initial monochloramine concentration and NH2Clt is the monochloramine at different times. Initial NH2Cl ) 2 mg/L as Cl2, pH ) 7.0, NaHCO3 ) 1 mM and PbO2 ) 10 mg/L.

FIGURE 1. Formation of Pb(II) from the reduction of PbO2 as a function of time in the presence of 2 mg/L as Cl2 free chlorine and monochloramine (a) Reagent grade PbO2 (b) Laboratory prepared PbO2. Background solution composition: pH 7.0, NaHCO3 ) 1 mM and PbO2 ) 10 mg/L. No NOM was added in these experiments. course of the experimental period. Monochloramine was added from a stock solution (280 mg/L as Cl2) freshly prepared before each experiment according to published procedures (15). Free chlorine was added from a stock solution (600 mg/L as Cl2) prepared by diluting the concentrated NaOCl solution (∼1 M, Fisher Scientific). Detailed experimental conditions employed in this study are shown in Supporting Information Table S1. Parallel experiments using reagent grade and laboratory prepared PbO2 were conducted to explore the effects of particle size on this process. Polypropylene bottles were covered by aluminum foil to avoid lightinduced free chlorine and monochloramine decay. The bottles were placed on a shaking table rotating at 200 rpm. Pb(II) concentration, disinfectant residual, UV254 and solution pH were measured over a period of 21 days with samples taken on day 1, 3, 7, 14, and 21. For all experiments, the pH values were within 7.0 ( 0.2 throughout the 21 day period. Analytical Methods. Pb(II) was measured by an anodic stripping voltammetry (ASV) using a Nano-Band Explorer II (TraceDetect, Seattle) (12, 13). A standard addition approach was employed and 0.1 M acetate was employed as the background electrolyte. In preliminary experiments, the presence of PbO2 particles did not interfere with the analysis of Pb(II) by this method. Free chlorine and monochloramine were measured by a modified iodometric method. In this method, 5 mL solution was withdrawn from the reactor and filtered by a nylon syringe filter with a 0.2 µm pore size (Fisher Scientific). 50 µL of phosphate buffer (for free chlorine) or acetate buffer (for monochloramine) was added to the filtrate followed by the addition of sufficient iodide to react with

free chlorine or monochloramine to form triiodate (I3-). The concentration of I3- was determined by the UV absorbance at 351nm (molar absorption coefficient ) 23325 cm-1M-1 (12)). This method showed less than 10% difference when compared to the widely used DPD-FAS method in preliminary experiments. UV absorbance was measured by a UV-vis spectrometer (UV-1601, Shimadzu). For experiments in the presence of monochloramine and NOM, the absorbance of NOM at UV254 was corrected for that contributed by monochloramine. Total organic carbon (TOC) was measured by a Shimadzu TOC5000 analyzer. SEM images of PbO2 particles were acquired by a Hitachi S-4000 scanning electron microscope. XRD pattern was determined by a MiniFlex II desktop X-ray diffractometer (Rigaku Americas). Solution pH values were measured by an Accument pH meter (A15, Fisher Scientific) coupled with a calomel combination electrode.

Results and Discussion Influence of Free Chlorine and Monochloramine on the Reduction of Lead Oxide in the Absence of NOM. The formation of Pb(II) from the reduction of PbO2 as a function of time in the presence of 2 mg/L as Cl2 free chlorine and monochloramine is shown in Figure 1. No NOM was added in these experiments in order to serve as a baseline for Pb(II) release in the presence of only free chlorine or monochloramine. In control experiments without disinfectant, Pb(II) slowly increased to 51 and 88 µg/L in the end of the experiments for reagent grade and laboratory prepared PbO2, respectively. These results are consistent with our previous study showing that PbO2 reduction occurs in water (13). In the presence of monochloramine, increasing levels of Pb(II) as a function of time were observed and the rates were higher than those in control experiments. On day 21, measured Pb(II) concentrations for reagent grade and laboratory prepared PbO2 were 94 and 205 µg/L, respectively. It has been demonstrated that the reduction of PbO2 to form Pb(II) can occur in monochloramine solutions and the formation of Pb(II) is linearly proportional to the amount of monochloramine that decomposes via a complex set of reactions (autodecomposition) (12). Water parameters that can accelerate the autodecomposition of monochloramine, including a higher monochloramine concentration, lower pH value, higher total carbonate concentration and lower Cl/N molar ration, can therefore enhance the reduction of PbO2. Linear relationships between Pb(II) formation and monochloramine autodecomposition were found for both particles as shown in Figure 2. These linear relationships VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3873

FIGURE 3. Effects of NOM on the formation of Pb(II) from the reduction of PbO2 as a function of time in the presence of free chlorine and monochloramine (a) Reagent grade PbO2 (b) Laboratory prepared PbO2. Background solution composition: pH 7.0, NaHCO3 1 mM, and PbO2 10 mg/L. indicate that given the PbO2 concentration (10 mg/L) employed in this study, decomposition of 0.1 mg/L as Cl2 monochloramine can induce the formation of approximate 15.8 and 59.8 µg/L of Pb(II) in solutions of reagent grade and laboratory prepared PbO2, respectively. In contrast, a maximum of 8 µg/L Pb(II) was measured in the presence of free chlorine in studies using both particle sizes. The low levels of Pb(II) formed in the presence of free chlorine were consistent with the strong oxidative nature of free chlorine which has been shown capable of oxidizing Pb(II) to PbO2 (6, 10, 11). Influence of Free Chlorine and Monochloramine on the Reduction of Lead Oxide in the Presence of NOM. The highest Pb(II) concentration was observed in experiments in the presence of NOM alone without the additional presence of disinfectant (NOM ) 2.5 mg/L as C, Figure 3). In the additional presence of 2 mg/L as Cl2 of monochloramine, the overall rate of Pb(II) formation was somewhat reduced compared to that of the system containing only NOM. This shows that monochloramine and NOM do not act additively or synergistically for PbO2 reduction. The same trend was observed for both PbO2 particle sizes. In the additional presence of 2 mg/L as Cl2 of free chlorine, a greater reduction of Pb(II) formation was observed. Compared to the NOMfree condition (Figure 1), however, a much higher Pb(II) concentration was detected in the presence of NOM although an equally initial concentration of free chlorine was dosed in both conditions. To further investigate the relationships among PbO2 reduction, UV254 and the presence of free chlorine or monochloramine during the course of the experiments, we measured residual UV254 and disinfectant residual as a 3874

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

function of time (see Supporting Information Figures S2 and S3). The trends of the time-dependent Pb(II) formation and corresponding residual UV254 and disinfectant residual for laboratory prepared PbO2 are shown Figure 4 and 5, respectively. The Pb(II) levels reflect well with the trends of residual UV254 and disinfectant residual, i.e., for the same reaction time, higher Pb(II) was observed in solutions with higher residual UV254 and lower disinfectant residual. Similar trends were also observed for reagent grade PbO2 (see Supporting Information Figures S4 and S5). Free chlorine is a stronger oxidant than monochloramine. It reacts much faster with NOM than monochloramine as shown by the fast decrease of both UV254 and free chlorine residual. As shown in Figure 5, for example, free chlorine residual in the 2 mg/L as Cl2 experiments almost completely dissipated within 1 d (NOM ) 2.5 mg/L as C, free chlorine ) 2 mg/L as Cl2, Figure 5); while the monochloramine residual in the 2 mg/L as Cl2 experiments slowly decayed and maintained at a level of about 1.0 mg/L as Cl2 after 21 days (NOM ) 2.5 mg/L as C, NH2Cl ) 2 mg/L as Cl2, Figure 5). Free chlorine can oxidize NOM to suppress the formation of Pb(II), as well as oxidize Pb(II) resulted from the water and NOM induced PbO2 reduction to reform PbO2 (6, 10, 11). Thus, as long as a sufficient free chlorine residual was maintained in the solution (approximately greater than 1 mg/L as Cl2 in the present study with 2.5 mg/L NOM as C), Pb(II) levels were lower than or approximately equal to those in control experiments (Figure 3, see first 7 day data for NOM ) 2.5 mg/L as C, free chlorine ) 5 mg/L as Cl2). As the free chlorine residual gradually dissipated to approach zero primarily due to NOM oxidation, Pb(II) levels resulting from the reduction of PbO2 by the remaining reductive capacity

FIGURE 4. Pb(II) formation and corresponding residual UV254 as a function of time. Background solution composition: pH 7.0, NaHCO3 ) 1 mM, and laboratory prepared PbO2 ) 10 mg/L.

FIGURE 5. Pb(II) formation and corresponding disinfectant residual as a function of time. Background solution composition: pH 7.0, NaHCO3 ) 1 mM, and laboratory prepared PbO2 ) 10 mg/L. of NOM started to increase (Figure 3, see 7-21 day data for NOM ) 2.5 mg/L, free chlorine ) 5 mg/L as Cl2). The effects of monochloramine on the reduction of PbO2 insolutionscontainingNOMaremorecomplex.Monochloramine can oxidize NOM as well, but much less rapidly than free chlorine as shown by the slower rate and smaller degree of UV254 decrease (Figure 4). Monochloramine is not able to oxidize Pb(II) to form PbO2 (6, 11). Instead, PbO2 reduction occurs in monochloramine solutions. Therefore, both monochloramine and NOM can cause PbO2 reduction. The oxidation of NOM by monochloramine, however, slowly consumes the capacity of NOM to reduce PbO2. The resulting Pb(II) formation (NOM ) 2.5 mg/L as C, NH2Cl ) 2 mg/L as

Cl2, Figure 4) was found to be slightly lower when compared to that in the presence of only NOM due to these combinative effects. The same intermediate of monochloramine autodecomposition that is capable of reducing PbO2 is not likely to regenerate the reactive functional groups in NOM oxidized by PbO2 because a consistent loss in UV254 was observed. Effects of Particle Size on the Reduction of PbO2. The trends of Pb(II) release, change in UV254 and disinfectant decay were similar for the two different sized PbO2 particles except that the Pb(II) released from laboratory prepared PbO2 (∼20 nm) were higher than those from reagent grade PbO2 (∼200 nm). Because of the different surface areas of the two PbO2 particles, it would be noteworthy to normalize the Pb(II) VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3875

FIGURE 6. Ratio of surface area normalized Pb(II) concentration (reagent grade/laboratory prepared) as a function of time. Background solution composition: Initial pH 7.0, NaHCO3 ) 1 mM and PbO2 ) 10 mg/L.

FIGURE 7. Schematic depiction of the roles of monochloramine, free chlorine and NOM in the formation of Pb(II) from the reduction of PbO2 in the distribution system. I*: intermediate formed in monochloramine autodecomposition. Note that the oxidation of NOM by monochloramine can deplete the reductive capacity of NOM. Thus, they do not act additively or synergistically for PbO2 reduction. concentration by surface area (SA) when comparing the Pb(II) release. Figure 6 shows the ratios of normalized Pb(II) concentration at different periods of time. The y axis of each point in Figure 6 was calculated by the following equation: Ratio of SA-normalized Pb(II)concentration ) Pb(II)t mass × SSA reagent grade (1) Pb(II)t mass × SSA laboratory prepared

(

(

)

)

where Pb(II)t represents Pb(II) concentration at time t, the mass employed for both PbO2 was 1.5 mg and the specific surface areas (SSA) for reagent grade and laboratory prepared PbO2 are 4.14 and 24.29 m2/g, respectively. The data shown in Figure 6 indicates that the amount of Pb(II) released from reagent grade PbO2 (∼200 nm) was generally higher than that from laboratory prepared PbO2 (∼20 nm) under the same experimental condition after normalization of surface area. The higher reactivity that is generally thought to be associated with nanoparticles, surprisingly, was not observed in our experiments. This phenomenon was also recently reported by Cwiertny et al. for the reduction of nitrobenzene by goethite (16). They found 3876

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009

that goethite nanoparticles showed lower reactivity than larger particles toward the reduction of nitrobenzene when reaction rates were normalized by the BET surface area. They argued that the lower reactivity of nanoparticles may be attributed to their more extensive particle aggregation compared to larger ones and that the BET surface area measured from dry particles should not be used to access the surface reactivity of wet nanoparticle suspensions. We suspect that particle aggregation may also contribute to the lower reactivity of nanosized PbO2 in our experiments. Environmental Significance. PbO2 is a strong oxidant and has been found in distribution systems with historical use of free chlorine and lead-containing plumbing materials. Its presence indicates that the distribution system can create a “memory” of the high oxidative environment induced by free chlorine. Monochloramine is a weaker oxidant than free chlorine. The decomposition of monochloramine has been demonstrated to be able to cause the reductive dissolution of PbO2. NOM is a powerful reductant toward the reduction of PbO2. Its reductive capacity may be largely preserved in chloraminated waters due to slow NOM oxidation compared to that occurring in the presence of free chlorine. Therefore, the formation of Pb(II) can be significantly reduced

by a more exhaustive oxidation of NOM and by reoxidation of Pb(II) to PbO2 in the presence of free chlorine. The roles of monochloramine, free chlorine and NOM in the formation of Pb(II) from the reduction of PbO2 in the distribution system is graphically depicted in Figure 7. The processes are complex and largely involving competing oxidation and reduction reactions. Water utilities that have adopted monochloramine for maintaining a disinfectant residual in the distribution system generally suffer from high concentrations of disinfection byproduct (DBPs) resulted from the reaction between NOM and free chlorine. The adverse role of NOM both in the DBPs formation and lead release suggests that enhanced removal of NOM in the water treatment plant may be the first step to solve the dilemma for some water utilities to simultaneously comply the regulations of DBPs and lead. Consideration of the use of monochloramine should consider both its influence on the potential release of lead from historically formed PbO2 as well as its impact on DBPs formation.

Acknowledgments We are grateful for the assistance of Michael Washburn on the preparation of PbO2 particles. This work was supported by a grant from the American Water Works Association Research Foundation (Project No. 3172). The assistance of ASV measurements from TraceDetect was highly appreciated.

Supporting Information Available Additional table and figures for detailed experimental conditions, SEM and XRD patterns of PbO2 particles, residual UV254 as a function of time and residual disinfectant as a function of time. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Finkelstein, Y.; Markowitz, M. E.; Rosen, J. F. Low-level leadinduced neurotoxicity in children: an update on central nervous system effects. Brain Res. Rev. 1998, 27, 168–176. (2) Papanikolaou, N. C.; Hatzidaki, E. G.; Belivanis, S.; Tzanakakis, G. N.; Tsatsakis, A. M. Lead toxicity update. A brief review. Med. Sci. Monit. 2005, 11, RA329–RA336.

(3) Dossnantos, A. C.; Colacciopo, S.; Dalbo, C. M. R.; Dossantos, N. A. G. Occupational exposure to lead, kidney-function tests, and blood-pressure. Am. J. Ind. Med 1994, 26, 635–643. (4) U.S. Environmental Protection Agency. Maximum contamination level goals and national primary drinking water regulations for lead and copper. Final rule. Fed. Regist. 1991, 56, 2646026564. (5) Renner, R. Plumbing the depths of D.C.’s drinking water crisis. Environ. Sci. Technol. 2004, 38, 224A–227A. (6) Edwards, M.; Dudi, A. Role of chlorine and chloramine in corrosion of lead-bearing plumbing materials. J. Am. Water Works Assoc. 2004, 96, 69–81. (7) Schock, M. R.; Giani, R. Oxidant/Disinfectant Chemistry and Impacts on lead corrosion; AWWA Water Quality Technology Conference: San Antonio, TX, 2004. (8) Schock, M. R.; Scheckel, K. G.; DeSantis, M.; Gerke, T. L. Mode of Occurrence, Treatment and Monitoring Significance of Tetravalent Lead; AWWA Water Quality Technology Conference: Quebec, Canada, 2005. (9) Dryer, D. J.; Korshin, G. V. Investigation of the reduction of lead dioxide by natural organic matter. Environ. Sci. Technol. 2007, 41, 5510–5514. (10) Lytle, D. A.; Schock, M. R. Formation of Pb(IV) oxides in chlorinated water. J. Am. Water Works Assoc. 2005, 97, 102–114. (11) Switzer, J. A.; Rajasekharan, V. V.; Boonsalee, S.; Kulp, E. A.; Bohannan, E. W. Evidence that monochloramine disinfectant could lead to elevated Pb levels in drinking water. Environ. Sci. Technol. 2006, 40, 3384–3387. (12) Lin, Y. P.; Valentine, R. L. Release of Pb(II) from monochloramine mediated dissolution of lead oxide (PbO2). Environ. Sci. Technol. 2008, 42, 9137–9143. (13) Lin, Y. P.; Valentine, R. L. The release of lead from the reduction of lead oxide (PbO2) by natural organic matter. Environ. Sci. Technol. 2008, 42, 760–765. (14) Chen, Z. The formation of N-Nitrosodimethylamine (NDMA) from reaction of monochloramine with natural organic matter. Ph.D. Thesis. University of Iowa, Iowa City, 2006. (15) Vikesland, P. J.; Ozekin, K.; Valentine, R. L. Monochloramine decay in model and distribution system waters. Water Res. 2001, 35, 1766–1776. (16) Cwiertny, D. M.; Handler, R. M.; Schaefer, M. V.; Grassian, V. H.; Scherer, M. M. Interpreting nanoscale size-effects in aggregated Fe-oxide suspensions: reaction of Fe(II) with goethite. Geochim. Cosmochim. Acta 2008, 72, 1365–1380.

ES900375A

VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

3877