Environ. Sci. Technol. 2009, 43, 3604–3611
PbO2(s, Plattnerite) Reductive Dissolution by Natural Organic Matter: Reductant and Inhibitory Subfractions ZHI SHI* AND ALAN T. STONE Department of Geography and Environmental Engineering The Johns Hopkins University, 3400 North Charles Street, Baltimore, Maryland 21218
Received December 5, 2008. Revised manuscript received March 17, 2009. Accepted March 31, 2009.
Natural organic matter (NOM) is a diverse collection of molecules, each possessing its own reductant, complexant, and adsorption properties. Here, we are interested in the ability of NOM to bring about the reductive dissolution of PbIVO2(s). Adding the coagulants FeCl3 or Al2(SO4)3 followed by membrane filtration is one way to remove a subset of NOM molecules from surface water samples. Another is to pass water samples through a granular activated carbon (GAC) column. Results from applying these treatments to Great Dismal Swamp water (DSW) and Nequasset Bog Water (NBW) can best be explained as follows: (i) GAC column treatment is more efficient at removing the NOM fraction most responsible for reductive dissolution. (ii) Coagulation/filtration, with either coagulant, is most efficient at removing a second, inhibitory fraction. Inhibition may arise from (i) adsorption at the mineral/water interface, which blocks approach of reductant molecules and (ii) a micellelike aggregate nature, which provides hydrophobic pockets that capture reductant molecules, again keeping them away from the mineral/water interface. Hypotheses regarding reductant and inhibitory fractions are further evaluated using representative low-molecular-weight compounds. Substituted hydroquinones are used as mimics of the reductant fraction, and malonic acid, quinic acid, trehalose, alginic acid, and polygalacturonic acid are used as mimics of the inhibitory fraction.
Introduction It has recently been established that chlorine in water supply systems oxidizes PbII to PbIV, which is found as (hydr)oxide coatings on pipe walls (1) and as suspended (hydr)oxide particles (2). The redox interconversion of PbII and PbIV therefore has public health consequences. PbIV (hydr)oxide reductive dissolution by Fe2+(aq) and by Mn2+(aq) (3) and by natural organic matter (4, 5) has recently been investigated. Dryer and Korshin collected water from three Washington, DC sources, concentrated the dissolved natural organic matter (NOM) using reverse osmosis, and removed associated cations via ion exchange (4). PbIV (hydr)oxide particles prepared from PbIICl2 chlorination followed by HCl/NaHCO3 washing were collected on cellulose filters and subjected to reductive dissolution by the concentrated NOM solutions. Approximately 10% of Pb was solubilized during 53 days of experiments. Surface area loading (in m2/L) is an important * Corresponding author phone: 1-410-516-8476; e-mail: zshi4@ jhu.edu. 3604
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
determinant of dissolution rates, but was not reported. Lin and Valentine studied NOM samples from the Iowa River after reverse osmosis and XAD 8 resin extraction (5). Less than 5% of Pb was released during 28 days of experiment using 20 mg C/L NOM. The present work seeks to clarify our understanding of PbIV (hydr)oxide reductive dissolution by NOM. PbIV (hydr)oxide particles prepared by chlorination of PbII in aqueous solution yield reproducible and readily quantified dissolution rates. These particles were contacted with NOM-rich surface waters before and after coagulation/filtration and granular activated carbon (GAC) column treatment. Low-molecularweight organic compound amendments possessing some of the properties of NOM were used to explore reaction mechanisms in greater detail. Pertinent NOM Characteristics. Each organic molecule comprising NOM exhibits its own reductant reactivity, and ability to complex dissolved metal ions, adsorb to mineral surfaces, and form aggregates with other organic molecules (6). Pretreatments that remove subfractions or chemically alter constituent molecules change the collective behavior of the organic molecules in the water sample. Coagulation and filtration effectively remove >1 kDa fractions of NOM, but are less effective in removing II ≈ III > IV. Adding DSW suppressed reaction with hydroquinone (41% decline in rate) and tert-butylhydroquinone (44% decline in rate) to similar extents. Compound III was most sensitive to the presence of DSW (65% decline) while Compound IV was the least sensitive (21% decline). PbIIaq time course plots for these experiments are shown in Figures 6 and S10. After 1 h of reaction, all four substituted hydroquinones yielded PbIIaq values that fell 4-7 µM short of the value expected for two-equivalent reductants, 20 µM. Untreated DSW (45 mg C/L) had virtually no effect on the plateauing seen with Compound IV, and a modest effect on the plateauing seen with hydroquinone. In contrast, untreated DSW lowered the plateau for tert-butylhydroquinone from 14 to 5 µM, and lowered the plateau for Compound III from 15 to 3 µM. Several months after the experiments just described were performed, one final experiment was conducted. tertButylhydroquinone was contacted with 45 mg C/L DSW for 1, 30, and 60 min prior to the addition of 20 µM PbO2(s). As shown in Figure 7, the longer the contact time, the lower the plateau. The drop in the plateau going from 1 to 30 min was considerably larger than the drop in going from 30 to 60 min. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3607
FIGURE 3. Reductive dissolution of 20 µM PbO2(s) by 20 µM hydroquinone in the presence and absence of untreated and Al2(SO4)3-treated DSW. Mass balance is explored using the sum [hydroquinone] + [benzoquinone].
Discussion Much of what we have observed regarding NOM-containing samples can be explained by postulating the existence of a reductant NOM fraction, responsible for PbO2(s) reductive dissolution, and an inhibitory NOM fraction, which prevents reaction from occurring. Plateauing tied to diminished reductant capacity was expected with FeCl3-treated NBW, since the DOC was only 1.5 mg C/L. Plateauing observed with GAC-treated NBW (6.7 mg C/L) but not with Al2(SO4)3-treated NBW (3.8 mg C/L) indicates the lack of a 1:1 correspondence between DOC lost and reductant capacity lost. We can conclude that GAC treatment is more efficient than Al2(SO4)3 treatment in removing the most reactive reductant molecules comprising the NBW sample. The increase in SUVA upon GAC treatment suggests that the most reactive reductant molecules are not significant contributors to SUVA. With DSW, DOC values both before and after treatment were higher than those observed with NBW, hence plateauing arising from loss of reductant capacity was not as pronounced. With NBW, FeCl3 treatment raised PbIIaq values during the first 4 h of reaction. With DSW, FeCl3 and Al2(SO4)3 treatment raised PbIIaq values, and also raised initial rates of PbO2(s) reductive dissolution. Treatments that lowered DOC yet promoted reductive dissolution are either especially inefficient at removing the reductant fraction or especially efficient at removing the inhibitory fraction. Adding 20 µM hydroquinone ensures that reductant capacity is sufficient for complete PbO2(s) dissolution to occur. Experiments of this kind provide direct evidence of inhibitory effects. Untreated DSW (Figure 3) and untreated NBW (Figure S6) interfered with the production of dissolved PbII, while Al2(SO4)3-treated samples did not. Similarly, 3608
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
FIGURE 4. (A) DSW (45 mg C/L) and the amendments malonate, quinate, trehalose, alginate, and polygalacturonate, all added to a concentration of 45 mg C/L, affect the reductive dissolution of 20 µM PbO2(s) by 20 µM hydroquinone. (B) The sum [hydroquinone] + [benzoquinone] decreases when NOM is present, but not when the other additives are present. untreated DSW (Figure S5) and untreated NBW (Figure S7) yielded a considerable discrepancy in the mass balance [hydroquinone] + [p-benzoquinone], while Al2(SO4)3-treated samples did not. When Al2(SO4)3 was added, an AlIII (hydr)oxide floc was generated which adsorbed NOM molecules, which were subsequently removed by filtration. Adsorption onto this floc might bear some resemblance to adsorption onto PbO2(s). One hypothesis is that Al2(SO4)3 treatment removes the inhibitory fraction because adsorption onto PbO2(s) causes the inhibitory effect. It would follow that NOM occupation of PbO2(s) surface sites blocks reductant access. An alternative hypothesis is that NOM molecules possess a distinctive property (e.g., high hydrophobicity) relevant to two otherwise disparate phenomena. NOM aggregates might serve as a phase into which hydroquinone and other reductant molecules might partition, thereby interfering with PbO2(s) reductive dissolution. NOM molecules might form covalent bonds with p-benzoquinone and semiquinone radical via nucleophilic substitution and radical coupling reactions, causing the observed decreases in [hydroquinone] + [p-benzoquinone] mass balance. As far as the two inhibitory mechanisms just described are concerned, whether or not NOM aggregates are associated with PbO2(s) surfaces might not matter. As we have seen (Figure 7), increasing the contact time from 1 to 30 min and finally to 60 min lowered the amount of reductive dissolution achieved once PbO2(s) had been added. We can
FIGURE 5. Initial rates of 20 µM PbO2(s) reductive dissolution by four substituted hydroquinones (20 µM) in the presence and absenceof45mgC/LuntreatedDSW:(I)hydroquinone,(II)tert-butylhydroquinone,(III)2,5-dihydroxy-1,4-benzenediacetate,(IV)2,5-bis[(dimethylammonium ion)methyl]hydroquinone.
FIGURE 6. Reductive dissolution of 20 µM PbO2(s) by four substituted hydroquinones (20 µM) in the presence (filled symbols) and absence (open symbols) of 45 mg C/L untreated DSW. The slope of the bold solid lines (created by linear regression of PbIIaq and time during the first 5 min of data) corresponds to the initial reaction rate (dPbIIaq/dt)i. VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3609
None of the NOM surrogates examined, including alginate and polygalacturonate, yielded the loss of [hydroquinone] + [p-benzoquinone] mass balance observed with untreated NOM. This loss of mass balance was even greater during reaction with PbO2(s). Identifying the NOM moiety responsible for the loss in mass balance is an important goal for future research. Water treatment plant operators seeking to minimize PbO2(s) reductive dissolution need to be cautious when addressing the effects of NOM. FeCl3 and Al2(SO4)3 treatments, efficient at removing inhibitory NOM fractions, can accelerate PbO2(s) reductive dissolution by the remaining reductant NOM fraction. GAC, on the other hand, more efficiently removes reductant capacity. Our work with substituted hydroquinones indicates that it is not enough to identify possible reductant moieties; neighboring functional groups may alter physicochemical properties in ways that significantly alter interactions with NOM aggregates and PbO2(s) surfaces. FIGURE 7. Effect of bringing 20 µM tert-Butylhydroquinone into contact with 45 mg C/L untreated DSW 1, 30, and 60 min prior to addition of 20 µM PbO2(s). conclude that significant partitioning into aggregates was taking place. Rather than being instantaneous, partitioning occurred on time scales of a few minutes to an hour. In NOM-free suspensions, reactivity differences exhibited by compounds I-IV were small: less than a factor of 2. Responses to NOM addition reveal important differences. Because tert-butylhydroquinone is more hydrophobic than hydroquinone, it should partition more into NOM aggregates. The plateau for PbIIaq was considerably lower for tertbutylhydroquinone than for hydroquinone, leading us to conclude that partitioning into aggregates keeps reductant molecules away from PbO2(s) surfaces. Because compound III is a dianion, it should be electrostatically repulsed from negatively charged NOM aggregates. Compound IV is a dication, and therefore should be attracted into NOM aggregates. Results with these two compounds, however, lead to the conclusion that partitioning into aggregates brings reductant molecules into contact with the PbO2(s) surfaces. There is a way to resolve the apparent conundrum. Hydrophobic partitioning and electrostatic attraction may “park” reductant molecules at different places within aggregates. When aggregates make contact with PbO2(s) surfaces, hydrophobically bound reductants may be held too far away for electron transfer, while electrostatically bound reductants (compound IV) may be close enough for electron transfer to occur. Our experiments offer some insight into the nature of NOM. The hydroquinone concentration we employed, 20 µM, is equivalent to 1.4 mg C/L. On a mg carbon basis, initial reductive dissolution rates by hydroquinone were 1600 times faster than untreated NBW, and 5200 times faster than untreated DSW. Amounts of PbO2(s) solubilized (Figures 1 and 2) and p-benzoquinone converted into hydroquinone (Figure S4) indicate, however, that both NOM samples contain significant reductant capacity, in excess of 20 microequivalents per liter. One possibility is that NOM reductant capacity comes primarily from less reactive, nonaromatic reductant moieties, such as the ones investigated by Wang and Stone (20, 21). A second possibility is that quinoid groups are significant but exist primarily in the oxidized, benzoquinone form. This possibility is in disagreement with the observed ability of untreated DSW to reduce p-benzoquinone to hydroquinone (Figure S4). A third possibility is that quinoid groups are present, but their reactivity is sterically suppressed by neighboring bulky substituents or electronically suppressed by neighboring electron-withdrawing substituents. This possibility is consistent will all of our experimental observations. 3610
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 10, 2009
Acknowledgments This work was supported by Agreement R828771-0-01 from the U.S. Environmental Protection Agency’s Science to Achieve Results (STAR) program, through the Center for Hazardous Substances in Urban Environments at Johns Hopkins. Although the research described in this article has been funded wholly or in part by the Environmental Protection Agency, it has not been subjected to the Agency’s required peer and policy review and therefore, does not necessarily reflect the views of the Agency and no official endorsement should be inferred. Acknowledgment is also made to the Donors of the American Chemical Society Petroleum Research Fund for partial support of this research. E. J. Bouwer provided access to ICP-MS. D. R. Veblen and Shouliang Zhang assisted with high-resolution transmission electron microscopy. Ed Thier of Dresden, ME provided the NBW sample. Marc Edwards (Virginia Tech) and Chad Jafvert (Purdue) provided very useful comments on this work.
Supporting Information Available Five additional tables and ten additional figures, plus additional details about the experimental results and their interpretation. This material is available free of charge via the Internet at http://pubs.acs.org.
Literature Cited (1) Lytle, D. A.; Schock, M. R. Formation of Pb(IV) oxides a in chlorinated water. J. Am. Water Works Assoc. 2005, 97 (11), 102– 114. (2) Triantafyllidou, S.; Parks, J.; Edwards, M. Lead particles in potable water. J. Am. Water Works Assoc. 2007, 99 (6), 107–117. (3) Shi, Z.; Stone, A. T. PbO2(s, plattnerite) reductive dissolution by aqueous manganous and ferrous ions. Environ. Sci. Technol. 2009, 43, DOI 10.1021/es8034686; In press. (4) Dryer, D. J.; Korshin, G. V. Investigation of the reduction of lead dioxide by natural organic matter. Environ. Sci. Technol. 2007, 41, 5510–5514. (5) 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. (6) Godtfredsen, K. L.; Stone, A. T. Solubilization of manganese dioxide-bound copper by naturally-occurring organic-compounds. Environ. Sci. Technol. 1994, 28, 1450–1458. (7) Owen, D. M.; Amy, G. L.; Chowdhury, Z. K.; Paode, R.; Mccoy, G.; Viscosil, K. NOM - characterization and treatability. J. Am. Water Works Assoc. 1995, 87 (1), 46–63. (8) Summers, R. S.; Roberts, P. V. Activated carbon adsorption of humic substances. 2. Size exclusion and electrostatic interactions. J. Colloid Interface Sci. 1988, 122, 382–397. (9) Schreiber, B.; Brinkmann, T.; Schmalz, V.; Worch, E. Adsorption of dissolved organic matter onto activated carbon - the influence of temperature, absorption wavelength, and molecular size. Water Res. 2005, 39, 3449–3456. (10) Traina, S. J.; Novak, J.; Smeck, N. E. An ultraviolet absorbance method of estimating the percent aromatic carbon content of humic acids. J. Environ. Qual. 1990, 19, 151–153.
(11) Weishaar, J. L.; Aiken, G. R.; Bergamaschi, B. A.; Fram, M. S.; Fujii, R.; Mopper, K. Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ. Sci. Technol. 2003, 37, 4702– 4708. (12) Edzwald, J. K.; Tobiason, J. E. Enhanced coagulation: US requirements and a broader view. Water Sci. Technol. 1999, 40 (9), 63–70. (13) Caldwell, W. T.; Thompson, T. R. Nuclear methylation of phenols. A new synthesis of intermediates in the preparation of antisterility factors. J. Am. Chem. Soc. 1939, 61, 765–767. (14) Black, A. P.; Christman, R. F. Chemical characteristics of fulvic acids. J. Am. Water Works Assoc. 1963, 55, 897–912. (15) O’Melia, C. R.; Becker, W. C.; Au, K. K. Removal of humic substances by coagulation. Water Sci. Technol. 1999, 40 (9), 47–54. (16) Johannesson, K. H.; Tang, J. W.; Daniels, J. M.; Bounds, W. J.; Burdige, D. J. Rare earth element concentrations and speciation in organic-rich blackwaters of the Great Dismal Swamp, Virginia, USA. Chem. Geol. 2004, 209 (3-4), 271–294.
(17) Cory, R. M.; McKnight, D. M. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. Technol. 2005, 39, 8142– 8149. (18) Leo, A. J.; Hansch, C.; Elkins, D. Partition coefficients and their uses. Chem. Rev. 1971, 71, 525–616. (19) Gunaseelan, K.; Romsted, L. S.; Gonzalez-Romero, E.; BravoDiaz, C. Determining partition constants of polar organic molecules between the oil/interfacial and water/interfacial regions in emulsions: a combined electrochemical and spectrometric method. Langmuir 2004, 20, 3047–3055. (20) Wang, Y.; Stone, A. T. The citric acid-MnIII, IVO2(s, birnessite) reaction. Electron transfer, complex formation, and autocatalytic feedback. Geochim. Cosmochim. Acta 2006, 70, 4463–4476. (21) Wang, Y.; Stone, A. T. Reaction of MnIII, IV (hydr)oxides with oxalic acid, glyoxylic acid, phosphonoformic acid, and structurally-related organic compounds. Geochim. Cosmochim. Acta 2006, 70, 4477–4490.
ES802441G
VOL. 43, NO. 10, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3611
