Environ. Sci. Technol. 2009 43, 8979–8984
Influence of Multiwalled Carbon Nanotubes Dispersed in Natural Organic Matter on Speciation and Bioavailability of Copper KI-TAE KIM,† AARON J. EDGINGTON,‡ STEPHEN J. KLAINE,‡ JAE-WEON CHO,† A N D S A N G D . K I M * ,† Department of Environmental Science and Engineering, Gwangju Institute of Science and Technology (GIST), 261 Cheom-dan Gwagi-ro, Buk-gu, Gwangju 500-712, Korea, and Department of Biological Sciences, Clemson University, Clemson, South Carolina 29634
Received March 3, 2009. Revised manuscript received October 5, 2009. Accepted October 8, 2009.
The dispersion of multiwalled carbon nanotubes (MWNTs) by natural organic matter (NOM) may influence the bioavailability of MWNTs and other contaminants. The speciation and bioavailability of copper (Cu) in MWNTs-associated NOM was studied using Daphnia magna. Cu titration data indicated that the binding affinity of Cu for MWNTs-associated NOM was lower than that for NOM alone. The free Cu2+ ion activity was increased even by the addition of a low nontoxic concentration of 1.0 mg/L MWNTs. The 96 h LC50 of MWNTs was determined to be 2.48 mg/L. The Fourier transform infrared (FTIR) spectra results showed clearly different features in Cu spiked between NOM and MWNTs-associated NOM, indicating that the interruption of Cu binding was probably due to steric stabilization of the MWNTs dispersed in NOM, which inhibited the complexation by rendering the functional groups in NOM less favorable to Cu. The mortality and biochemical reactive oxygen species (ROS) production in the D. magna bioassay were enhanced in MWNTs-associated NOM compared to NOM alone because of increased free Cu2+ ion activity as expected from the titration and FTIR results. This study suggests the bioavailability of Cu is enhanced by the presence of MWNTs interacting with NOM.
Introduction Multiwalled carbon nanotubes (MWNTs) are pure carbon macromolecules consisting of two or more concentric layers of rolled graphene sheets with various lengths and diameters. MWNTs have novel electronic, mechanical, thermal, and chemical properties, which allow their use as a major carbonbased nanomaterial within several technologies (1). However, the rapid increase in the investigation and commercialization of MWNTs has presented intentional and accidental exposure to human and environment receptors. The human health effects and ecological risks of MWNTs have received great attention (1, 2). Natural organic matter (NOM), a complex and heterogeneous mixture of a diverse group of molecules, plays an * Corresponding author e-mail:
[email protected]; phone: 82-62970-2445; fax: 82-62-970-2434. † Gwangju Institute of Science and Technology (GIST). ‡ Clemson University. 10.1021/es900647f CCC: $40.75
Published on Web 10/23/2009
2009 American Chemical Society
important role in the fate and transport of xenobiotic compounds in aquatic ecosystems (3). NOM is ubiquitously present at concentrations ranging from 1 to 100 mg L-1 in freshwater (4). Thus, considering the interaction of MWNTs with NOM is essential for assessing their behavior and ecological effects. A previous study reported that agglomeration of MWNTs originating from their strong hydrophobicity can be overcome by NOM; thereby, it is predicted that MWNTs would be stabilized in an aqueous environment (5). The π-π interaction between the graphene surface and aromatic rings is the major mechanism of NOM adsorption onto MWNTs, which depends on the hydrophobicity of the NOM source (i.e., aromatic contents) as well as water quality parameters such as pH and ionic strength (5-7). The stabilization of MWNTs implies increased bioavailability and mobility over that of agglomerates that can be easily deposited due to strong van der Waals forces. However, there has been no study regarding the bioavailability of MWNTs dispersed in NOM (8). NOM coating alters the physical and chemical properties of MWNTs. MWNTs with a modified surface due to coating by several types of dissolved organic matter (DOM) (i.e., humic acid, peptone, and R-phenylalanine) have been reported to decrease the sorption capacity for selected organic contaminants (9). The negatively charged polar functionalities also make the hydrophobic sites in MWNTs less active to hydrophobic organic chemicals. A large number of studies have demonstrated that NOM can attenuate the toxicity and bioavailability of trace metals in aquatic systems because of complexation (10-13). Therefore, MWNTs combined with NOM could provide a certain capacity to bind with heavy metals because of structural modification (i.e., surface functionality). MWNTs coated with NOM act as oxidized MWNTs, on which surface functional groups (e.g., hydroxyl, carboxyl, and carbonyl) can be introduced through acid treatment (14, 15). However, the role of MWNTs dispersed in NOM should be considered from the viewpoint of the bioavailability of the metals because MWNTs may interact with heavy metals. Little mechanistic work exists regarding the behavior of MWNTs dispersed in NOM. For example, could MWNTs modulate the binding of heavy metals to NOM? MWNTs released and stabilized into an aqueous environment would interact with coexisting contaminants, influencing their bioavailability. Meanwhile, copper (Cu) has been widely investigated as an important heavy metal in the aquatic environment because of its prevalence and high toxicity (10-13). It is also well-known that NOM binds to Cu, reducing the activity of “free” Cu, which is responsible for decreased toxicity or bioavailability (10-13). In this study, it was expected that the association of MWNTs with NOM would influence the bioavailability of Cu. The comparative Cu speciation and subsequent bioavailability in the presence of NOM alone and MWNTsassociated NOM were investigated using a Daphnia magna bioassay standardized for an aquatic toxicity test. The bioavailability of MWNTs dispersed in NOM was also evaluated.
Experimental Section Preparation of MWNTs. MWNTs were purchased from Hanhwa Nanotech (Seoul, Korea). According to commercial information, they were synthesized using a chemical vapor deposition (CVD) method, with purity of 95%. The reported diameter and length were approximately 10-15 nm and 10-20 µm, respectively. Suwannee River NOM (SR-NOM) was purchased from the International Humic Substances VOL. 43, NO. 23, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
8979
Society (IHSS) (St. Paul, MN). SR-NOM was dissolved in synthetic moderately hard water (MHW) to facilitate the bioassay and filtered using 0.22 µm cellulose membrane filter (Millipore, Bellerica, MA). The total organic carbon (TOC) composition of the 50 mg/L SR-NOM in MHW was measured to be 42% (i.e., 21 ( 1 mg C/L) using TOC analyzer (Sievers, Boulder, CO). Its hydrophobic, transphilic, and hydrophilic fractions were determined to be 66, 21, and 13%, respectively, via XAD 4/8 resins (Amberlite, U.S.A.) (16). Nominally weighed MWNTs (0.0030-0.0035 g) were transferred to 25 mL of 50 mg/L NOM solution in a 100 mL glass centrifuge tube (Kimble Kontes, Vineland, NJ). The solution was then sonicated for 5-7 min, with intensity of 30 W, using a Probe Sonicator (Cole Parmer, Vernon Hills, IL) to disperse the MWNTs. This sonication procedure was repeated after the addition of each 25 mL of NOM; finally, four rounds of sonication were conducted. After that, the solution was placed for 24 h to settle any undispersed MWNTs. The settled MWNTs were subtracted from the initial weight of the MWNTs; the concentration of actually dispersed MWNTs in 50 mg/L NOM was determined to be approximately 20 mg/L. The stock solutions of MWNTs dispersed in NOM were kept at room temperature prior to use. All MWNTs concentrations refer to mass of MWNTs dispersed in SR-NOM. Electron microscopic images were obtained using a JEOL transmission electron microscope (TEM), JEM-2100 (Jeol, Tokyo, Japan), operated at 200 kV electron volts. MWNTs dispersed in NOM were dried in a 100 °C oven for a day. Dried samples were kept in a desiccator and then mounted on a 200 mesh Cu carbon grid (Electron Microscopy Science, Hatfield, PA). The metal contents of the MWNTs were simultaneously analyzed using energy dispersive spectroscopy (EDS) (INCAx-sight, Oxfordshire, UK), equipped with TEM. The samples for EDS were mounted on a 200 mesh nickel (Ni) and Cu carbon grid to confirm the Cu and other metal contents in the MWNTs. Titrations. The free Cu2+ was determined using an Orion model 9272BN Cu-ISE, with an Ag/AgCl internal reference system, connected to an Orion pH meter (model 720A Plus, Boston, MA). Ion selective electrode (ISE) potentiometry is one of the analytical techniques able to measure Cu2+ speciation in NOM (10, 13, 17). Detailed procedures for Cu titration are provided in the accompanying Supporting Information. Cu titrations were performed on 0.22 µm filtered 50 mg/L NOM solution prepared in MHW (I ) 0.01 M and pH 8.0). The 100 mL of NOM solution was titrated with 5 × 10-4 and 10-2 M Cu(NO3)2 to cover a total Cu concentration range from 10-7 to 2 × 10-4 M. The titrant was added at 5-25 min intervals, until it reached a stable potential (
