Differences in Soil Mobility and Degradability between Water

The potential of QDs to release toxic Cd2+ and/or Se2-/SeO32- ions upon ... in the top soil after passing 10 column volumes of solution through the so...
0 downloads 0 Views 3MB Size
Download PDF
ARTICLE pubs.acs.org/est

Differences in Soil Mobility and Degradability between Water-Dispersible CdSe and CdSe/ZnS Quantum Dots Divina A. Navarro,† Sarbajit Banerjee,† David F. Watson,† and Diana S. Aga*,† †

Department of Chemistry, University at Buffalo, State University of New York, Buffalo, New York 14260-3000, United States

bS Supporting Information ABSTRACT: The relative leaching potential and degradation of water-dispersible CdSe and CdSe/ZnS quantum dots (QDs) were evaluated using small-scale soil columns. The potential of QDs to release toxic Cd2+ and/or Se2-/SeO32- ions upon degradation is of environmental concern and warrants investigation. Both classes of QDs exhibited limited soil mobility in CaCl2, with more than 70% of the total Cd and Se species from QDs retained in the top soil after passing 10 column volumes of solution through the soil column. However, mobilization of Cd- and Se-species was observed when EDTA was used as the leaching solution. Approximately 98% of the total Cd2+ loaded leached out from the Cd2+-spiked soil, while only 30% and 60% leached out from the CdSe and CdSe/ZnS QD-spiked soils, respectively. Soil column profiles and analysis of leachates suggest that intact QDs leached through the soil. Longer incubation (15 days) in soil prior to leaching indicated some degradation and/or surface modification of both QDs. These results suggest that chelating agents in the environment can enhance the soil mobility of intact and degraded QDs. It is apparent that QDs in soil, including the polymer-coated CdSe/ZnS QDs that are generally assumed to possess a higher degree of environmental stability, can undergo chemical transformations, which subsequently dictate their overall mobility.

’ INTRODUCTION Quantum dots (QDs) are semiconductor nanocrystals with diameters in the 2 100 nm size range known to exhibit remarkable size-tunable optical properties. QDs have found widespread interest across diverse research communities because of their potential for applications such as solar energy conversion,1,2 display and lighting technologies,3 and biomedical imaging.4 QDs consist of an inorganic crystalline core, oftentimes composed of a cadmium chalcogenide, with or without an outer shell of zinc chalcogenide, surrounded by a passivating layer of organic ligands. Although QDs have yet to penetrate the consumer market, clear pathways now exist for their commercialization in areas such as photovoltaics and display technologies. Consequently, several groups have raised concerns with regard to the health and environmental risks posed by these emerging contaminants.5,6 The projected incorporation of QDs into commercial products7 would indicate a potential for their release into different environmental compartments, from point and diffuse sources, e.g., manufacturing waste, intentional release in applications, disposal of consumer products or medical waste, and from biosolid land application. Degradation of QD-containing photovoltaic panels mounted on rooftops represents another possible route for these engineered nanomaterials to enter the soil column. To date, many countries and organizations, such as those from the European Union (European Commission), United Kingdom (Natural Environment Research Council) and United States (U.S. Environmental Protection Agency National Center for Environmental Research), have launched research programs and cooperative initiatives to understand the r 2011 American Chemical Society

potential health and environmental implications of nanotechnology, including QDs.5,8 10 The complete assessment of the environmental impact of QDs requires investigating QD exposure routes, environmental fate, and transport pathways. Whether QDs are dispersed in the aqueous phase, sorbed onto soil, or transported at sediment/ water interfaces, they can potentially undergo a variety of transformations including partial/complete degradation with concomitant leaching of component ions. The precise fate of QDs will likely depend on the stability of the QDs, their interactions with different species, and the specific environmental conditions. Fundamental studies have shown that the stabilities of QDs under realistic environmental conditions are generally dictated by the weakest (most easily disrupted) interactions in these material systems, which tend to be the coordinate covalent bonds between the capping ligands and the surface atoms of the inorganic core.11 13 For example, thiol-coated CdSe QDs are known to be photochemically unstable in suspension.11 Displacement or removal of capping ligands on QD surfaces as a result of changes in pH, UV light, or oxidation have been central to many of the published QD toxicity studies, where toxicity is closely associated with the release of toxic Cd2+, Se2-, and/or SeO32- ions. Stability of QDs may also depend on the stoichiometry and composition gradient of the inorganic Received: March 25, 2011 Accepted: June 22, 2011 Revised: June 20, 2011 Published: June 22, 2011 6343

dx.doi.org/10.1021/es201010f | Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology core, with those that have additional inorganic shells for surface passivation (i.e., CdSe/ZnS, CdSe/ZnS/ZnS QDs) purportedly endowing better resistance to degradation.14,15 Indeed, various structural aspects of the passivating ligands influence the stability of the QDs including the thickness of the afforded diffusion barrier, packing of ligand molecules on the nanocrystal surface, and strength of the metal ligand bond. Analogously, for intact QDs, their aqueous dispersibility will also be dictated by the surface ligands, which in turn determine the extent of electrostatic and steric stabilization. For instance, in an environmental setting, humic substances (HS) can modify the surfaces of QDs and alter the dispersibility and aggregation state of QDs in water,16 even inducing the phase transfer of originally hydrophobic QDs to the aqueous phase.17,18 Furthermore, changes in ionic strength and ionic composition can also affect the dispersibility of the QDs.19 Hence, with agglomeration of QDs potentially induced in aquatic systems, sufficiently large and unstable agglomerates of QDs could eventually be formed, and the transport of QDs may reasonably be expected to be dominated by sedimentation. Soil will likely be an important sink for nanomaterials that enter the environment. Surfactants and other organic acids present in soil can facilitate the mobility of materials that would otherwise have limited transport. Studies relating to the fate and transport behavior of engineered nanomaterials in solid matrices are very limited. A common observation has been that the mobilities of nanomaterials (particularly titania, alumina, carbon nanotubes, fullerenes, etc.) in porous media are dependent on their surface properties and especially the extent of surface modification upon environmental exposure.20,21 However, these studies do not report whether the nanomaterials remained intact, underwent structural modification, or were transformed in any manner. In our present study, we demonstrate the soil transport behavior and transformation of water-dispersible CdSe and CdSe/ZnS QDs using packed columns. Our objective was to examine the leaching potential of carboxylic acid-functionalized CdSe and CdSe/ZnS QDs in natural soil using either a 10-mM calcium chloride (CaCl2) solution or a 10-mM disodium salt of ethylenediammine tetraacetic acid (EDTA) solution for leaching. Since much of the concern associated with QDs is due to the likely release of their core metal ions, the mobility of the QDs was compared to that of aqueous Cd2+ and SeO32- solutions. Our results suggest that while QDs may exhibit limited mobility in soil, the presence of chelating agents in soil can enhance QD transport in the environment. Degradation observed for the CdSe and CdSe/ZnS QDs further emphasizes the importance of producing QDs that can resist decomposition and suggests design principles for enhancing environmental stability.

’ MATERIALS AND METHODS General. Water-dispersible CdSe QDs, coated with mercaptopropionic acid (MPA), were synthesized by ligand exchanging tetradecyl phosphonic acid (TDPA)-coated QDs prepared using CdO as a precursor by a hot colloidal synthesis approach.22 The CdSe/ZnS QDs were functionalized with a polymer shell bearing carboxylic acid ( CO2H) pendant groups. The core/shell QDs were purchased from Invitrogen (Eugene, OR). Prior to use of the QDs, the QDs were dialyzed in deionized (DI) water using a Spectra/Por Biotech cellulose ester dialysis membrane (MWCO of 0.5 1 kDa) (Spectrum Laboratories, Inc., Rancho Dominguez, CA) to remove free ions and unbound ligands. Other

ARTICLE

details on the leaching solutions and soil used are described in the Supporting Information (SI). Preparation of Soil Columns. A 15-mL polypropylene cartridge (internal diameter = 1.6 cm, length = 7.5 cm) was packed with 12 g of soil to achieve a soil height of approximately 6 cm. Column ends were covered with polyethylene frits. The soil column was saturated and equilibrated with 10 mM CaCl2 solution at an approach velocity of 2.5 mm/min in the upward direction with the aid of a peristaltic pump. Prewetting of soil was done from the column bottom to prevent formation of bubbles and to reduce the risk of channeling. When EDTA was used for leaching, 10 mM NaNO3 was used in the equilibration. Columns were equilibrated for 24 h, which was adequate to attain steadystate conditions (constant pH). Soil Leaching Procedure. After the equilibration step, 1 mL of QD dispersion and 1 mL of 1 mM KBr tracer were applied on the top surface of the soil column. The soil column was covered then inverted upside-down. Depending on the experiment, the selected leaching solution was continuously introduced into the column at approximately 1 mm/min in the upward direction. The bromide ion tracer leached out of the soil after addition of approximately 6 mL of CaCl2 or EDTA, which was then estimated to be the soil column pore-volume. Ten pore-volumes of leaching solution were loaded onto the columns, collecting 6-mL fractions of eluate in separate polypropylene tubes. Used soil columns were frozen at 40 °C and then cut into four equal sections (ca. 1.5 cm) for analysis. Leachates and soil sections were analyzed for the concentration of Cd, Se, and Br tracer. Due to the significantly high levels of endogenous Zn in soil, Zn from the CdSe/ZnS QDs was not analyzed. Soil columns spiked with 15 μg/mL of aqueous Cd2+ and SeO32- solution were prepared as positive control; SeO32- was chosen as spiking material because weathering of CdSe QDs has been reported to release this form of Se.12,23 Unspiked soil (tracer only) was used as the negative control. Leaching experiments were performed in duplicate. Approximately 50 250 nM QD dispersions (containing 20 100 μg/mL Cd2+) were used. Three leaching studies were conducted to assess the overall mobility and stability of the QDs in soil. The first experiment focused on the transport of the QDs in 10 mM CaCl2, which is used by the EPA to represent artificial rain.24 The second set of experiments aimed to investigate the influence of chelating acids on the transport of QDs. A solution of 10 mM EDTA in 10 mM NaNO3 was used as electrolyte; EDTA was introduced at the same concentration level as CaCl2 to provide a more direct comparison. In the first and second tests, spiked soils were leached on the same day. The third experiment studied the effect of aging the QDs in soil, using 10 mM EDTA in 10 mM NaNO3 as the leaching electrolyte. QD-spiked soils were allowed to age for 15 days before soil leaching. Each column was wrapped with Al-foil for protection from light. Analysis. Total Cd and Se (soil and leachates) were analyzed by inductively-coupled plasma mass spectrometry (ICP-MS) using a Thermo Scientific (Germany) X-Series 2 instrument with collision cell technology. Using 115In as an internal standard, the concentrations of total 111Cd and 78Se were determined using an external calibration curve. The detection limits for 111Cd and 78Se were 0.005 and 0.025 ng mL 1, respectively, based on three times the signal-to-noise ratio. Samples for ICP-MS were either acidified with 2% HNO3 (leachates) or digested in concentrated HNO3 and H2O2 (soil segments, EPA method 3050B). Recoveries for the extraction of Cd and Se in soil using the digestion 6344

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology

Figure 1. TEM images of pristine (a) CdSe and (b) CdSe/Zn QDs. Intact (c) CdSe and (d) CdSe/ZnS QDs that were found in the EDTA soil leachates. Emission spectra for the EDTA soil leachates from the (e) CdSe and (f) CdSe/ZnS QD set-ups in comparison with pristine QDs. The spectra for pristine CdSe/ZnS QD is scaled down by 1/15.

method were determined to be 104 ( 1% and 77 ( 2%, respectively. A combination bromide ion-selective electrode from Mettler Toledo was used to measure the Br concentration in the leachates. All absorption and emission spectra were collected using a Hewlett-Packard 8452A diode array spectrophotometer and Varian (USA) Cary Eclipse fluorimeter, respectively. Transmission electron microscopy (TEM) images and selected area electron diffraction (SAED) patterns were collected using a JEOL (Japan) JEM-2010 instrument operated at an accelerating voltage of 200 kV. Leachates analyzed by TEM were first freeze-dried and reconstituted with 1 mL of water to preconcentrate the sample. Samples were then prepared by slow evaporation of a drop of aqueous solution placed onto 400 mesh carbon/Formvar grids.

’ RESULTS AND DISCUSSION Characteristics of the QDs. The CdSe and CdSe/ZnS QDs were characterized by TEM and SAED (Figures 1a-b and S1a-b). The MPA-coated CdSe QDs, synthesized in-house, were rod-like with average aspect ratio, length, and diameter of 3 ( 1 nm, 14 ( 3 nm, and 5 ( 1 nm, respectively. The commercially procured polymer-coated CdSe/ZnS QDs, reported to have surface carboxylate groups, were also rod-like with average aspect ratio, length, and diameter of 2.0 ( 0.3, 12 ( 1 nm, and 6.3 ( 0.7 nm, respectively. Both QD samples exhibited well-defined and pronounced first excitonic absorption bands.

ARTICLE

The emission maxima of CdSe and CdSe/ZnS QDs were 652 and 654 nm, respectively. Compared to the CdSe QDs, the CdSe/ZnS QDs exhibited higher emission intensities even in very dilute concentrations (1.3 nM), which is expected since the higher bandgap ZnS shell substantially eliminates surface trap states that participate in nonradiative decay processes. Both QDs in water had negative ζ-potentials owing to the carboxylate functionalities on the surfaces of QDs; the ζ-potential distributions for CdSe and CdSe/ZnS QDs were centered at 17.7 and 20.5 mV, respectively. The absorption and emission spectra of the CdSe and CdSe/ ZnS QDs in water, CaCl2, and EDTA were stable over a 24-h period (Figures S3 and S4). No significant changes were observed in the absorption spectra for the CdSe and CdSe/ZnS QDs in water, CaCl2, and EDTA. Consistent with the emission spectra, shifts in peak positions were not observed indicating negligible changes of particle size during this time period. The decreased emission from both types of QDs in aqueous dispersions containing EDTA is likely not a result of disintegration of QDs because the absorption spectra were essentially unchanged with time. Since the leaching experiment was completed in 6 h, the QDs in the leaching solutions are expected to be stable. We assume that any changes observed with the QDs during the leaching experiment can be related to their behavior and stability in the soil matrix. (Further discussion is provided in the SI.) General Leaching Behavior of QDs. Soil column leaching profiles were plotted for the different experiments to show the distribution of Cd and Se in the soil and leachates. For the QDspiked set-ups, assuming that the dialyzed QDs did not contain free Cd2+ and SeO32- ions, the concentrations of Cd and Se determined by ICP-MS were attributed only to the QDs. Dispersions of QDs were dialyzed prior to use to avoid conflating transport of QDs with transport of free Cd2+ and SeO32- that may be present in the spiking solution. Leaching of Cd- and QD-spiked soils with CaCl2 revealed limited movement of Cd2+, CdSe, and CdSe/ZnS QDs (Figures 2 and S5a), with no Cd-containing species detected in any of the leachates. The ζ-potential distributions of the QDs in CaCl2 solution (centered at 6.5 mV for CdSe, and 4.8 mV for CdSe/ZnS QDs) suggest a higher tendency for the QDs to aggregate thereby reducing their mobilities upon additions of CaCl2. On the other hand, Se-containing species were detected in the leachates of Se-spiked soils. The distribution of Cd and Se in the QD-spiked soils is also well correlated, which suggest that the migrating species were intact QDs; this inference is based on the knowledge that free Cd2+ and SeO32- ions have different mobilities in soil (Figure S5). For the CdSe QD-spiked soil columns, >90% of the total Cd and Se was retained in the top 1.5-cm of soil. In contrast, for the CdSe/ZnS QD-spiked soil columns, >70% of the total Cd and Se was retained in the top 1.5-cm of soil. Thus, some movement of Cd- and Se-species was observed. Likewise, Cd2+ from the control solution moved slightly within the soil column after leaching with 10 porevolumes of CaCl2. The retentive behavior of Cd2+ is consistent with the fast sorption kinetics of Cd2+ to soil.25,26 Although no Cd2+ was detected in the leachates of the Cd2+-spiked soils, about 30% of the total Cd2+ moved beyond the top 1.5-cm soil. The CdSe QDs exhibited stronger soil sorption than CdSe/ ZnS QDs, which is somewhat surprising given that both QDs had similar surface functionalization (pendant surface COO groups), core composition, core (or core/shell) size, and shape. One possible origin of the observed difference is that the MPA6345

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology

ARTICLE

Figure 2. Soil column profiles for (a) CdSe and CdSe/ZnS QD-spiked soils leached using CaCl2 solution and (b) CdSe and (c) CdSe/ZnS QD-spiked soils leached using EDTA solution. Asterisks denote significant differences in concentrations between 0-d and 15-d soils (*) and leachates (*'); Student's t-test (two tailed) at 95% confidence level. Each bar graph section represents the concentration of Cd and Se in the different soil segments - from top to bottom of the 6-cm column - and in the leachates. Error bars correspond to standard deviation of concentrations determined from two soil columns. Recovery of Cd and Se species is within acceptable limits of error, and control experiments (Figure S5) provide validation of the amounts determined in the leaching experiments.

coated CdSe QDs were synthesized by ligand exchange of TDPA-coated CdSe QDs. While a sufficient number of TDPA ligands were probably displaced by MPA ligands to induce water dispersibility, it is unlikely that this reaction proceeded to completion. Incomplete exchange and removal of TDPA may have given rise to a different binding motif for these QDs in soil due to the hydrophobicity of the pendant alkyl chains from remnant TDPA compared to the hydrophilic polyelectrolyte overcoating on the CdSe/ZnS QDs. For the commercially procured QDs, the significantly larger polymeric passivating groups have a bulkier steric footprint but nevertheless exhibit appreciable aqueous dispersibility due to the high density of polar functional groups. No information is available with regard to the actual polymer material used to cover the QD surface. While the precise mechanism remains unclear, the addition of CaCl2 clearly influenced the mobility of the CdSe/ZnS QDs; CaCl2 also enabled desorption of solvated Cd2+ and SeO32-, which gave rise to some movement within the soil.26,27 Effect of EDTA on the Transport of CdSe and CdSe/ZnS QDs. Different types of organic acids, including long and short chain fatty acids, aromatic acids, and tricarboxylic acids derived from soil microorganisms and roots of plants, are commonly found in soil.28 While these compounds are important for mobilizing mineral nutrients, they can also influence the rate of release of heavy metal contaminants from soils.29 These acids, as well as their degradation products, may serve as coordinating ligands for metals and modify their solubility. In this study, EDTA, a synthetic organic acid that is used worldwide in household and industrial applications, was used to illustrate the effect of chelating acids on the mobility of QDs in soil. EDTA was used to represent the worst case scenario for the effect of organic acids.30 It is known that EDTA can mobilize Cd2+ in soil; therefore, we used Cd2+-spiked soil as a positive control to evaluate the

Figure 3. (a) Cd and (b) Se breakthrough curves for the different soil columns leached with 10 mM EDTA solution. C = detected concentration in the leachate, Co = initial spiked concentration. V = volume of leachate collected, Vo = column pore volume. Error bars correspond to standard deviation of concentrations determined from two soil columns.

leaching of QDs in the soil columns. The use of EDTA as an additive to facilitate remediation of heavy metal-polluted soil has also been proposed in the past.31,32 Addition of EDTA into the solution used for leaching experiments resulted in the leaching of soluble Cd2+ species, possibly Cd2+-EDTA complexes, which were detected in the first fraction of leachates. The Cd peak eluted close to the tracer peak (Figure 3a) indicating that the Cd2+-EDTA species were minimally retained in the soil; 98% of the total Cd2+ leached out from the Cd2+-spiked soil. A vast preponderance of the Cd2+ was mobilized, even some Cd2+ already present in the control soil. Unlike Cd2+, SeO32- species (Figure 3b) leached only after addition of 4 pore-volumes of EDTA, again illustrating the substantial differences in the mobilities of the cationic and anionic components (and verifying the retention of QD integrity when the concentrations of the two components can be correlated). Interestingly, Cd- and Se-containing species of comparable proportions eluted in the CdSe and CdSe/ZnS QD-spiked soils 6346

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology after addition of 1 pore-volume of EDTA. The ζ-potential distributions of the QDs in EDTA (centered at 40.2 mV for CdSe and 27.3 mV for CdSe/ZnS QDs) also suggest formation of stable aqueous dispersions. Given the distinctive relatively slower mobility of SeO32-, it can be inferred that the clusters of Cd and Se migrating are intact QDs rather than solubilized SeO32- and Cd2+ ions. Hence, we can assume that the first leachates contained 10% and 48% of the total CdSe and CdSe/ZnS QDs loaded onto the column, respectively. EDTA may have formed complexes with the QDs through exposed Cd2+ and/or Zn2+ surface sites, altering the dispersibility of QDs and allowing them to leach out of the column intact (Logarithms of formation constants for 1:1 Cd:EDTA and Zn: EDTA complexes are 17.4 and 17.5, respectively33). After addition of 10 pore-volumes of EDTA, some Cd- and Secontaining species were still retained in the soil spiked with CdSe and CdSe/ZnS QDs (Figure 2b-c). The surface metal centers of QDs are likely to be more shielded from the soil/ water interface than is solvated Cd2+. Comparing different QDs, the amphiphile-passivated CdSe/ZnS QDs moved more rapidly than the CdSe QDs. In the CdSe/ZnS QD-spiked soil, only 25% of the total Cd (37% total Se) loaded onto the column was retained, while in the CdSe QD-spiked soil, 56% of the total Cd (77% total Se) loaded onto the column was retained. Thus, EDTA may have coordinated preferentially to CdSe/ZnS QDs. Zwitterionic interactions between positively charged aminic nitrogen atoms on EDTA and the pendant carboxylate groups may also give rise to highly mobile and polar species that eluted rapidly through the soil column. (The polymer-coated CdSe/ ZnS QDs bear greater density of carboxylic acid groups than the MPA-coated CdSe QDs.) An additional secondary interaction invoked above may also contribute to the observed variation between the two types of QDs. Remnant TDPA groups present on the CdSe QD surface may contribute some hydrophobic character, thereby increasing the retention of CdSe QDs compared to CdSe/ZnS QDs. These results are consistent with our CaCl2 experiments, in which the CdSe QDs were less mobile than the CdSe/ZnS QDs. The levels of Cd and Se in the soil and leachates also correlated with each other providing further evidence that the migrating species may have been QDs. Aside from coordinative and/or zwitterionic interactions between QDs and EDTA, EDTA can also mobilize humic substances (HS) (via disruption of Ca2+-HS interactions) that could promote leaching of QDs in soil by disrupting QD-soil interactions and/or forming water-soluble QD-HS agglomerates.17,18 Leachates from the QD-spiked soils were further characterized to detect the presence of intact QDs. Absorption spectra of the CdSe and CdSe/ZnS QD EDTA leachates (Figure S6) were inconclusive with respect to indicating the presence of intact QDs since these were not significantly different from the spectra of the control. Compared to the CdSe QDs, the first CdSe/ZnS QD EDTA leachate exhibited measurable fluorescence (Figure 1e-f), with a peak at 652 nm, which is 2 nm blueshifted from the emission maximum of pristine CdSe/ZnS QDs. The persistence of measurable emission from CdSe/ ZnS QDs provides clear evidence that some intact QDs were found in the leachate. No emission peaks characteristic of CdSe QDs were found in the leachates, though the CdSe QDs exhibited weak emission when dispersed in aqueous solutions of EDTA (Figures S3 and S4) and in the leachate matrix (Figure S7). HS extracted at high concentrations with the addition of EDTA can also contribute to emission quenching.34

ARTICLE

TEM images (Figures 1c-d, S2a-b) and SAED patterns (Figure S1c-d) confirmed the presence of intact QDs in the EDTA leachates collected from the CdSe and CdSe/ZnS QD-spiked soil columns. The particles were still identifiable through their latticeresolved crystal planes despite being embedded within a matrix of amorphous material. Consistent with the blue-shifted emission, TEM images also indicated a decrease in the size of CdSe/ZnS QDs (∼5 nm in length and ∼2 nm in diameter), though freezedrying may also have contributed to some size changes. The retention of a crystalline lattice observed in the SAED patterns further corroborates that intact nanoparticles leached in the CdSe and CdSe/ZnS QD-spiked soil columns. Indeed, the ratios of the measured concentrations of Cd and Se in these leachates (0-day experiments) were also fairly close to the expected ratio of the QDs used (Table S2). Hence, the crystalline and fluorescent particles (containing Cd- and Se-species) that leached out after addition of 1 2 pore-volumes of EDTA (Figure 3) were intact QDs. Although some Cd2+ and Se2-/SeO32- ions may also have been released during the soil leaching process, based on our results, “free Cd2+ and Se2-/SeO32-” cannot be easily separated from intact QDs by simple dialysis as more complex Cd2+/Se2-/ SeO32--species may arise from the soil leaching process. (Dialysis experiments are discussed in more detail in the SI.) Effect of Soil Aging on Mobility of QDs. The effect of aging QDs was investigated to determine the distribution and chemical fate of QDs in soil over time. Using EDTA as the leaching solution, the Cd and Se distribution profiles for the CdSe and CdSe/ZnS QDs changed significantly 15 days after application in soil (Figure 2b-c). In the CdSe QD-spiked soil, the amount of Cd in the leachates increased from 30% to 70% (p < 0.05) after 15 days. However, increase in the amount of Se in the leachates was not significant (p > 0.05). In the case of CdSe/ZnS QDs, aging did not necessarily change the Cd concentration in the leachates (p > 0.05). Rather, aging appeared to have increased the retention of both Cd- and Se-containing species in the soil. The amount of Cd in the soil increased from 25% to 34% with aging. Furthermore, the amount of Se in the soil increased significantly from 37% to 75% (p < 0.05). Hence, unlike the 0-day experiments, Se did not move with Cd in the aged QD-spiked soil columns. These results were also contrary to the same-day leaching experiments, in which CdSe QDs were less mobile than CdSe/ZnS QDs. The widely different Cd and Se distribution profiles reveal that the aging of QDs in soil resulted in some degradation and/ or transformation of QDs because majority of the Cd- and Secontaining species did not move as a cluster. The calculated Cd: Se mole ratios in the leachates were higher than the expected ratios for the pristine QD (Table S2), indicating more Cdspecies in these samples. With the release of core elements upon degradation of QDs, it can be expected that addition of EDTA will leach out more Cd2+ and less SeO32- in the soil, given that Cd2+ is more mobile than SeO32- (Figure 3). Notably, although our data suggest that both types of QDs degraded during aging, intact CdSe and CdSe/ZnS QDs in the leachates were still discernible in the TEM images, albeit with reduced particle sizes (Figure S2c-d). The degradation and/or transformation of QDs likely involved removal and/or alteration of their surface capping groups through one or more mechanisms. For thiol-coated CdSe QDs, aging has been reported to result in increased mobility, which increases the formation of both Cd- and Se-species that were easily mobilized by EDTA. The gradual removal of thiol groups 6347

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology from surfaces of QDs has been reported to occur over time.11 Ligand-stripping processes would result in faster leaching of more Cd- and Se-containing species, including intact QDs and free Cd2+. Removal of thioalkyl groups from the CdSe QD surface could expose Cd2+ surface sites in a manner that could facilitate coordinative interactions between the QD core and EDTA giving rise to increased levels of Cd and Se in the leachates. Also, free Cd2+ could also be released, further increasing the level of Cd in the leachates. For the polymercoated CdSe/ZnS QDs, aging resulted in reduced mobility, with more Cd- and Se-containing species retained in the soil. Studies have shown that reactive oxygen species (ROS), like HOCl and H2O2, can cause degradation of polymer-coated core/shell QDs.35 While displacement of the polymer seems less likely compared to thioalkyl ligands, ROS could react with the surface functional groups, diffuse across the polymer coating, and promote surface oxidation of QDs. Reaction between the surface coating and ROS could result in the loss of polar pendant groups that impart the aqueous dispersibility of QDs, thereby decreasing their mobility. Similar to the CdSe QDs, free Cd2+ could also leach out of the soil column. In soil, microbially generated ROS, such as those produced by brown rot fungi,36 have been reported to accelerate formation of disulfides and oxidation of QDs. Metz et al.37 recently reported that poly (ethylene glycol)-thiol-coated QDs were degraded/transformed under oxidative environmental conditions mediated by microorganisms. Our emission spectra (Figure S8) reveal that both types of QDs remained intact in water during the aging period; therefore, the observed degradation and/or transformation of QDs can be attributed to the decreased stability of QDs in soil. Mechanistic characterization of apparent QD transformations represents a formidable challenge in the absence of validated separation methods for analysis of nanomaterials in complex environmental matrices. Though many researchers have suggested and demonstrated the use of field flow fractionation (coupled to a light scattering or an ICP-MS detector) for analysis of nanomaterials and QDs,38,39 to date, no reported method has shown the actual separation of different QD species in an environmentally relevant matrix. Potential Mobilization of QDs in the Environment. Surface chemistry plays a very important role in dictating the partitioning, mobility, and stability of QDs in soil. Results from this study suggest that, similar to Cd2+ ions, both CdSe and CdSe/ZnS QDs exhibit limited mobility in soil. It can be expected that QDs will remain on the top soil even with heavy rain, unless chelating agents such as EDTA are introduced on purpose or naturally occurring organic acids (such as plant exudates) are present. QDs in soil appear to be mobilized as intact and/or as etched QD species. The integrity of the QDs in soil could also be compromised over time, and to varying degrees, depending on the QD formulation. With aging, surface coatings can be displaced from QDs, degraded, and/or modified by species already present in soil, resulting in the transformation and/or degradation of QDs. Despite reports that core/shell QDs are stable,14,16 our results indicate that the longterm stability of these nanoparticles under environmental conditions is questionable. It appears that the large polymer coating protecting the QD core is still vulnerable to degradation and/or modification. While the mechanism by which QDs are transformed in soil has yet to be established, it is apparent that QDs will be susceptible to changing environmental conditions. The rate at which QDs degrade and their mobility in soil will

ARTICLE

dictate the relative distributions of intact and transformed QD species in soil leachates. Unless stabilized by natural organic matter or other species in the environment, QDs may ultimately be degraded in soil or in water and serve as a source for toxic mobile Cd and Se species.

’ ASSOCIATED CONTENT

bS

Supporting Information. Solutions and soil used. Details on QD dispersion controls and dialysis. SAED and TEM images of QDs and EDTA leachates. Absorbance and emission spectra of control QD dispersions (H2O, CaCl2, EDTA, and leachate matrix). Soil column profiles for Cd2+ and SeO32-. Absorbance spectra of EDTA leachates. Emission spectra of pristine CdSe QDs collected over 15 days. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 716 645 4220. Fax: 716 645 6963. E-mail: dianaaga@ buffalo.edu.

’ ACKNOWLEDGMENT This work was primarily supported by the Environmental Protection Agency (Grant#R833861). S.B. acknowledges partial support of this work from the National Science Foundation (NSF) under DMR 0847169. We acknowledge the NSF MRI Program CHE 0959565 for the ICP-MS. We acknowledge Dr. Yueling Qin for his assistance with HRTEM measurements. Although the research described in this work has been funded by the USEPA, 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. ’ REFERENCES (1) Lewis, N. S.; Crabtree, G. Basic research needs for solar energy utilization: Report of the basic energy sciences workshop on solar energy utilization, April 18 21, 2005; U.S. Department of Energy: 2005. (2) Kamat, P. V. Quantum dot solar cells. Semiconductor nanocrystals as light harvesters. J. Phys. Chem. C 2008, 112, 18737–18753. (3) Anikeeva, P. O.; Halpert, J. E.; Bawendi, M. G.; Bulovic, V. Quantum dot light-emitting devices with electroluminescence tunable over the entire visible spectrum. Nano Lett. 2009, 9 (7), 2532–6. (4) Gao, X.; Yang, L.; Petros, J. A.; Marshall, F. F.; Simons, J. W.; Nie, S. In vivo molecular and cellular imaging with quantum dots. Curr. Opin. Biotechnol. 2005, 16, 63–72. (5) Dunphy Guzman, K. A.; Taylor, M. R.; Banfield, J. F. Environmental Risks of Nanotechnology: National Nanotechnology Initiative Funding, 2000 2004. Environ. Sci. Technol. 2006, 40 (5), 1401–1407. (6) Owen, R.; Handy, R. Formulating the problems for environmental risk assessment of nanomaterials. Environ. Sci. Technol. 2007, 41 (16), 5582–5588. (7) Graham-Rowe, D. From dots to devices. Nat. Photonics 2009, 3 (6), 307–309. (8) U.S. EPA. EPA Nanotechnology White Paper. http://www.epa. gov/osainter/pdfs/nanotech/epa-nanotechnology-whitepaper-0207.pdf (accessed January 30, 2011). (9) HM Government, Characterizing the Potential Risks Posed by Engineered Nanoparticles: A Second U.K. Government Research Report. http://www.defra.gov.uk/environment/quality/nanotech/documents/ nanoparticles-riskreport07.pdf (accessed January 30, 2011). 6348

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349

Environmental Science & Technology (10) EC European Commission, Communication from the Commission: Towards a European Strategy for Nanoterchnology. http:// ec.europa.eu/nanotechnology/pdf/nano_com_en_new.pdf (accessed January 30, 2011). (11) Aldana, J.; Wang, Y. A.; Peng, X. Photochemical instability of CdSe nanocrystals coated by hydrophilic thiols. J. Am. Chem. Soc. 2001, 123, 8844–8850. (12) Mahendra, S.; Zhu, H.; Colvin, V. L.; Alvarez, P. J. J. Quantum dot weathering results in microbial toxicity. Environ. Sci. Technol. 2008, 42, 9424–9430. (13) Kirchner, C.; Liedl, T.; Kudera, S.; Pellegrino, T.; Mu~ noz Javier, A.; Gaub, H. E.; St€olzle, S.; Fertig, N.; Parak, W. J. Cytotoxicity of colloidal CdSe and CdSe/ZnS nanoparticles. Nano Lett. 2005, 5, 331–338. (14) Grabolle, M.; Ziegler, J.; Merkulov, A.; Nann, T.; ReschGenger, U. Stability and fluorescence quantum yield of CdSe-ZnS quantum dots-influence of the thickness of the ZnS shell. Ann. N.Y. Acad. Sci. 2008, 1130, 235–241. (15) Su, Y.; He, Y.; Lu, H.; Sai, L.; Li, Q.; Li, W.; Wang, L.; Shen, P.; Huang, Q.; Fan, C. The cytotoxicity of cadmium based, aqueous phase Synthesized, quantum dots and its modulation by surface coating. Biomaterials 2008, 30 (1), 19–25. (16) Slaveykova, V. I.; Startchev, K. Effect of natural organic matter and green microalga on carboxyl-polyethylene glycol coated CdSe/ZnS quantum dots stability and transformations under freshwater conditions. Environ. Pollut. 2009, 157 (12), 3445–3450. (17) Navarro, D. A.; Watson, D. F.; Aga, D. S.; Banerjee, S. Natural organic matter-mediated phase transfer of quantum dots in the aquatic environment. Environ. Sci. Technol. 2009, 43 (3), 677–682. (18) Navarro, D. A.; Banerjee, S.; Aga, D. S.; Watson, D. F. Partitioning of hydrophobic CdSe quantum dots into aqueous dispersions of humic substances: Influence of capping-group functionality on the phase-transfer mechanism. J. Colloid Interface Sci. 2010, 348 (1), 119–128. (19) Zhang, Y.; Chen, Y.; Westerhoff, P.; Crittenden, J. Stability and removal of water soluble CdTe quantum dots in water. Environ. Sci. Technol. 2008, 42 (1), 321–5. (20) Lecoanet, H. F.; Bottero, J.-Y.; Wiesner Mark, R. Laboratory assessment of the mobility of nanomaterials in porous media. Environ. Sci. Technol. 2004, 38 (19), 5164–9. (21) Jaisi, D. P.; Elimelech, M. Single-walled carbon nanotubes exhibit limited transport in soil columns. Environ. Sci. Technol. 2009, 43 (24), 9161–6. (22) Pong, B.-K.; Trout, B. L.; Lee, J.-Y. Modified Ligand-Exchange for Efficient Solubilization of CdSe/ZnS Quantum Dots in Water: A Procedure Guided by Computational Studies. Langmuir 2008, 24 (10), 5270–5276. (23) Katari, J. E. B.; Colvin, V. L.; Alivisatos, A. P. X-ray Photoelectron Spectroscopy of CdSe Nanocrystals with Applications to Studies of the Nanocrystal Surface. J. Phys. Chem. B 1994, 98 (15), 4109–4117. (24) U.S.EPA Fate, Transport and Transformation Test Guidelines: Leaching Studies; Office of Prevention, Pesticides, and Toxic Substances: October 2008, 2008. (25) Christensen, T. H. Cadmium soil sorption at low concentrations: I. Effect of time, cadmium load, pH, and calcium. Water, Air, Soil Pollut. 1984, 21 (1 4), 105–14. (26) Liu, C.-L.; Chang, T.-W.; Wang, M.-K.; Huang, C.-H. Transport of cadmium, nickel, and zinc in Taoyuan red soil using onedimensional convective-dispersive model. Geoderma 2006, 131 (1 2), 181–189. (27) Wang, W.-Z.; Brusseau, M. L.; Artiola, J. F. The use of calcium to facilitate desorption and removal of cadmium and nickel in subsurface soils. J. Contam. Hydrol. 1997, 25 (3 4), 325–336. (28) Jones, D. L.; Dennis, P. G.; Owen, A. G.; van Hees, P. A. W. Organic acid behavior in soils - misconceptions and knowledge gaps. Plant Soil 2003, 248 (1 2), 31–41. (29) Krishnamurti, G. S. R.; Cieslinski, G.; Huang, P. M.; Van Rees, K. C. J. Kinetics of cadmium release from soils as influenced by organic

ARTICLE

acids: implication in cadmium availability. J. Environ. Qual. 1997, 26 (1), 271–277. (30) Oviedo, C.; Rodriguez, J. EDTA: The chelating agent under environmental scrutiny. Quim. Nova 2003, 26 (6), 901–905. (31) Meers, E.; Lesage, E.; Lamsal, S.; Hopgood, M.; Vervaeke, P.; Tack, F.; Verloo, M. Enhanced phytoextraction: I. Effect of EDTA and citric acid on heavy metal mobility in a calcareous soil. Int. J. Phytorem. 2005, 7 (2), 129–142. (32) Sun, B.; Zhao, F. J.; Lombi, E.; McGrath, S. P. Leaching of heavy metals from contaminated soils using EDTA. Environ. Pollut. 2001, 113 (2), 111–120. (33) Morel, F. M. M. Principles of Aquatic Chemistry; Wiley: New York, 1983. (34) Kumke, M. U.; Loehmannsroeben, H. G.; Roch, T. Fluorescence quenching of polycyclic aromatic compounds by humic acid. Analyst 1994, 119 (5), 997–1001. (35) Mancini, M. C.; Kairdolf, B. A.; Smith, A. M.; Nie, S. Oxidative Quenching and Degradation of Polymer-Encapsulated Quantum Dots: New Insights into the Long-Term Fate and Toxicity of Nanocrystals in Vivo. J. Am. Chem. Soc. 2008, 130 (33), 10836–10837. (36) Cohen, R.; Jensen, K. A.; Houtman, C. J.; Hammel, K. E. Significant levels of extracellular reactive oxygen species produced by brown rot basidiomycetes on cellulose. FEBS Lett. 2002, 531 (3), 483–488. (37) Metz, K. M.; Mangham, A. N.; Bierman, M. J.; Jin, S.; Hamers, R. J.; Pedersen, J. A. Engineered nanomaterial transformation under oxidative environmental conditions: development of an in vitro biomimetic assay. Environ. Sci. Technol. 2009, 43 (5), 1598–1604. (38) Bouby, M.; Geckeis, H.; Geyer, F. W. Application of asymmetric flow field-flow fractionation (AsFlFFF) coupled to inductively coupled plasma mass spectrometry (ICPMS) to the quantitative characterization of natural colloids and synthetic nanoparticles. Anal. Bioanal. Chem. 2008, 392 (7 8), 1447–1457. (39) Williams, S. K. R.; Runyon, J. R.; Ashames, A. A. Field-Flow Fractionation: Addressing the Nano Challenge. Anal. Chem. 2011, 83 (3), 634–642.

6349

dx.doi.org/10.1021/es201010f |Environ. Sci. Technol. 2011, 45, 6343–6349