Competitive Sorption of Pb (II) and Zn (II) on Polyacrylic Acid-Coated

Sep 11, 2013 - ... (1–102) single-crystal 5 cm diameter wafers (Saint-Gobain Crystals ...... Kitts , K.; Choi , Y.; Eng , P. J.; Ghose , S. K.; Sutt...
0 downloads 0 Views 2MB Size
Article pubs.acs.org/est

Competitive Sorption of Pb(II) and Zn(II) on Polyacrylic Acid-Coated Hydrated Aluminum-Oxide Surfaces Yingge Wang,† F. Marc Michel,†,‡,▽ Clement Levard,†,○ Yong Choi,§ Peter J. Eng,∥ and Gordon E. Brown, Jr.†,‡,⊥,#,* †

Surface & Aqueous Geochemistry Group, Department of Geological & Environmental Sciences, Stanford University, Stanford, California 94305-2115, United States ‡ Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, MS 69, 2575 Sand Hill Road, Menlo Park, California 94025, United States § Advanced Photon Source, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, Illinois 60439, United States ∥ Consortium for Advanced Radiation Sources, University of Chicago, Chicago, Illinois 60637, United States ⊥ Department of Photon Science, SLAC National Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, California 94025, United States # Department of Chemical Engineering, Stauffer III, Stanford University, 381 North-South Mall, Stanford, California 94305-5025, United States S Supporting Information *

ABSTRACT: Natural organic matter (NOM) often forms coatings on minerals. Such coatings are expected to affect metal−ion sorption due to abundant sorption sites in NOM and potential modifications to mineral surfaces, but such effects are poorly understood in complex multicomponent systems. Using poly(acrylic acid) (PAA), a simplified analog of NOM containing only carboxylic groups, Pb(II) and Zn(II) partitioning between PAA coatings and α-Al2O3 (1−102) and (0001) surfaces was investigated using long-period X-ray standing wave-florescence yield spectroscopy. In the singlemetal−ion systems, PAA was the dominant sink for Pb(II) and Zn(II) for α-Al2O3(1−102) (63% and 69%, respectively, at 0.5 μM metal ions and pH 6.0). In equi-molar mixed-Pb(II)−Zn(II) systems, partitioning of both ions onto α-Al2O3(1−102) decreased compared with the single-metal−ion systems; however, Zn(II) decreased Pb(II) sorption to a greater extent than vice versa, suggesting that Zn(II) outcompeted Pb(II) for α-Al2O3(1− 102) sorption sites. In contrast, >99% of both metal ions sorbed to PAA when equi-molar Pb(II) and Zn(II) were added simultaneously to PAA/α-Al2O3(0001). PAA outcompeted both α-Al2O3 surfaces for metal sorption but did not alter their intrinsic order of reactivity. This study suggests that single-metal−ion sorption results cannot be used to predict multimetal−ion sorption at NOM/metal−oxide interfaces when NOM is dominated by carboxylic groups.



INTRODUCTION Minerals and humic substances (often referred to as natural organic matter (NOM)) are ubiquitous in soils and aquatic systems and are often spatially associated due to the formation of NOM coatings on mineral surfaces.1−4 Such coatings potentially induce significant modifications to mineral surface electrostatic properties, such as reversing surface charge from positive to negative, and provide abundant additional sorption sites for metal ions.3,5−8 As a result, NOM coatings are generally assumed to play an important role in the biogeochemical cycling of heavy metals in natural waters, soils, and sediments.3,5−8 Humic substances are natural biomacromolecules produced from the breakdown of plants, animals, fungi, and bacteria.5,9 These natural organic macromolecules are weak polyelectrolytes and have various compositions, sizes, and conformations © 2013 American Chemical Society

and a number of different types of functional groups, including carboxylic, amino, phenolic, and aromatic groups.5,9 As a result of this complexity, many studies have used chemically and structurally simple molecules as analogs of NOM. Polycarboxylic acids such as poly(acrylic acid) (PAA), a polymer containing carboxylic functional groups in linear CH2−CH2 chains, are often selected as simple surrogates for humic substances because of the general similarity of their polyelectrolyte properties and functional groups to those of humic substances.10−13 For example, PAA has been used as a model compound for humic substances to study the environReceived: Revised: Accepted: Published: 12131

March 27, 2013 July 30, 2013 September 11, 2013 September 11, 2013 dx.doi.org/10.1021/es401353y | Environ. Sci. Technol. 2013, 47, 12131−12139

Environmental Science & Technology

Article

gibbsite and boehmite in soils and aquatic systems. In addition, α-Al2O3 can be obtained in oriented single-crystal forms required for the grazing-incidence X-ray standing wave studies reported below, whereas the other more common Al(oxyhydr)oxide phases are not available as single crystals. Finally, the structures of the hydrated α-Al2O3 (0001) and (1− 102) surfaces have been determined by crystal truncation rod diffraction studies27,28 and their sorption properties have been well-studied.3,6,29,30 We used the long period X-ray standing wave fluorescent yield (LP-XSW-FY) method to measure Pb(II) and Zn(II) partitioning on PAA-coated single-crystal α-Al2O3 (0001) and (1−102) surfaces. LP-XSW-FY is a grazing-incidence spectroscopic technique for characterizing the spatial and chemical distribution of elements in single- or multilayered samples.3,6,31−35 It is element-specific, nondestructive, and has high sensitivity to elements at low concentrations (≥10−8 M) within a particular layer of material as well as at the buried interfaces between layers.3,6,31−35 As a result, this technique has become an effective tool for probing element distributions at various types of interfaces, including electrochemical interfaces,31,32 biological membranes,31 mineral/water interfaces,32 and mineral surfaces coated with microbial biofilms or thin organic films.3,6,31,32,34,35

mental behavior of lanthanide and actinide ions in natural media and to evaluate the long-term performance and safety of nuclear waste repositories.11 PAA is also widely used in industry as a scale inhibitor, as a dispersant in papermaking, and as a stabilizer and flocculant.12,13 Due to its very low environmental impact and high sorption capacity for metal ions, PAA is also considered as a promising sorbent in toxic heavy-metal removal from industrial effluents.12,14 For example, PAA is often used as a chelating agent in a technique known as polymer-assisted ultrafiltration (PAUF) to help remove trace metals from wastewater effluents. The strong effective binding of PAA to trace metal ions results in high removal efficiency and the desired quality of treated water.14 There have been extensive studies of the interaction of humic substances with metal ions and mineral surfaces,5,10,15−17 and a number of thermodynamic models have been developed to predict metal−ion binding by NOM although the general applicability of such models requires further improvements.5,15,16 Depending on the metal−ion affinities of NOM and mineral surfaces, and experimental conditions such as pH and the types and concentrations of metal ions, NOM can either enhance metal−ion uptake by increasing the negative charge on mineral surfaces and/or forming strong complexes with metal ions,7,8,18,19 or decrease metal−ion sorption by physically blocking the sorption sites on mineral surfaces.20,21 Moreover, metal−ion interactions at more complex NOM/ mineral/water interfaces cannot be described simply as the sum of the behavior observed in individual binary systems.10,17 Therefore, metal−ion partitioning in multicomponent systems approaching the complexity of real systems is still poorly understood. In addition, natural soil and aquatic systems contain a variety of metals at concentrations at and above trace levels.22,23 It is reasonable to expect that different types of metal ions may compete for available sorption sites and potentially interfere with each other in terms of uptake. Such competitive sorption effects could, in turn, impact the bioaccumulation and toxicity of metal ions in the environment. Therefore, understanding these processes and determining if the findings from single-metal−ion sorption studies are valid in multimetal− ion systems are essential for predicting the fate and transport of metal ions in natural environments.22−26 Only a few studies of competitive sorption among multiple metal ions at NOM/ mineral interfaces have been carried out to date.22−26 Competitive metal−ion sorption effects at mineral surfaces have been found to range from none or weak to fairly strong, depending on the experimental conditions including the types of metal ions and mineral surfaces, the initial metal concentrations used, and reaction kinetics.23−25 Although surface complexation models (SCM) have been used to predict metal−ion uptake in single-metal−ion systems, SCM models based on simple systems cannot correctly predict multiple metal−ion uptake in natural soil mixtures.26 Therefore, competitive sorption effects are largely unknown in multimetal−ion, multisubstrate systems under realistic conditions. This is due in part to the general lack of appropriate analytical tools capable of determining both the chemical speciation and spatial distributions of elements in these complex systems.3,6−8 Here we studied the competition between aqueous Pb(II) and Zn(II) ions for sorption sites on poly(acrylic acid) (PAA)coated, hydrated single-crystal α-Al2O3 (0001), and (1−102) surfaces. These mineral substrates were chosen as models because the aluminol sites present on these surfaces are representative of those of Al-(oxyhydr)oxide phases such as



EXPERIMENTAL METHODS Preparation of Metal−Oxide Surfaces and PAA Thin Films. The α-Al2O3 substrates used in this study are commercially available, highly polished α-Al2O3 (0001) and (1−102) single-crystal 5 cm diameter wafers (Saint-Gobain Crystals & Detectors Co.). These surfaces were prepared using a chemical cleaning procedure described in our previous studies.3,6,27,28 In brief, all substrates were cleaned with acetone, then washed in 10−3.5 M sodium hydroxide for 20 min, and subsequently washed in 10−2 M nitric acid for an hour. Each chemical washing step was followed by multiple rinses with Milli-Q water. The washed crystals were then baked at 350 °C for 4 h to minimize excess carbon on the surfaces. The cleaning procedure was repeated as necessary until the concentrations of metal surface impurities was undetectable (