Article pubs.acs.org/est
Increased Adsorption of Sulfamethoxazole on Suspended Carbon Nanotubes by Dissolved Humic Acid Bo Pan,*,† Di Zhang,† Hao Li,† Min Wu,† Zhenyu Wang,*,‡ and Baoshan Xing§ †
Faculty of Environmental Science & Engineering, Kunming University of Science & Technology, Kunming, China 650500 College of Environmental Science and Engineering, Ocean University of China, Qingdao, China 266100 § Stockbridge School of Agriculture, University of Massachusetts, Amherst, Massachusetts 01003, United States ‡
S Supporting Information *
ABSTRACT: Although dissolved organic matter (DOM) could effectively disperse carbon nanotubes (CNTs), sorption characteristics of DOMsuspended CNTs are unclear. In this study, we evaluated the relative contribution to the overall sorption from dissolved humic acid (DHA) coating (decreased sorption) and CNT dispersion (increased sorption). We observed that the sorption coefficients of sulfamethoxazole (SMX) on DHA-suspended CNTs were up to 2 orders of magnitude higher than that on aggregated CNTs. Although the mass percent of suspended CNTs were low (generally less than 1%), their contributions to SMX adsorption were up to 20%. Because DHA and SMX did not interact with each other due to their negative charges, the suspended CNTs may not be completely coated by DHA and they had considerable hydrophobic surface exposed. Importantly, this study provided the first evidence in aqueous phase of the significantly increased surface area of DHA-suspended CNTs relative to the aggregated ones based on 1H NMR relaxometry measurements. This study emphasizes that in comparison to aggregated CNTs, the suspended ones have amply exposed surface area and thus have greater environmental impacts, such as enhancing the mobility, transport, and possibly exposure of organic contaminants.
■
INTRODUCTION Carbon nanotubes (CNTs) are increasingly used as sensors, conductive composites, catalyst supports, and gas storage due to their unique properties of mechanical strength, electrical conductivity, optical activity, hydrogen storage capability, and large surface area.1−4 Therefore, they will be inevitably released into the environment via domestic wastewater effluent and/or during their manufacturing, application, and/or disposal, and may become a very important component of the environment.5−7 The coexistence of CNTs and dissolved organic matter (DOM) could widely occur in the environment because of the ubiquity of DOM and the increasing application of CNTs. The introduction of hydrophilic functional groups through DOM adsorption on CNTs decreased the hydrophobicity of CNTs 8 and increased CNT electrostatic repulsion9,10 and steric hindrance.9,11 Thus, DOM is an effective substance to disperse CNTs.8,12,13 The mobility and transport of dispersed CNTs and CNT-adsorbed organic chemicals could subsequently be promoted in natural aqueous environments, potentially enhancing the spread of various organic chemicals and their environmental risk. Most studies reported decreased adsorption in the presence of DOM. For example, DOM coating decreased adsorption of phenanthrene, naphthalene, 1-naphthol, 1,3-dinitrobenzene, 1,3,5-trinitrobenzene,14,15 sulfapyridine and sulfamethoxazole,16 and pyrene17 on CNTs. The investigators suggested that DOM coating occupied some sorption sites for these tested compounds and © XXXX American Chemical Society
thus decreased the availability of CNTs surface in addition to making CNTs surface less hydrophobic. A few studies, however, suggested that the dispersed nanoparticles may have higher adsorption than the aggregated ones. For example, Zhang et al. observed that sonication broke down aggregated CNTs and increased pyrene adsorption by up to 1.4 orders of magnitude. 17 Increased atrazine adsorption was observed in DOM-nC60 system at low DOM concentrations (2.5 to 5 mg/L), probably because of the exposed surface sites after dispersion by DOM.18 Wang et al. also proposed that the enhanced dispersion by DOM may have increased the sorption of phenanthrene, naphthalene, and 1-naphthol on CNTs, but it was offset by DOM coating to reduce the availability of surface sites.19 On the basis of the above discussion, we could notice that the key information that is missing is the adsorption of organic contaminants on the suspended CNTs. But no study directly reported the adsorption characteristics of the suspended CNTs and their potential importance for the environmental behavior and risk of organic contaminants and CNTs themselves. We thus made special efforts to identify the contribution of suspended CNTs to the overall adsorption. The primary Received: February 26, 2013 Revised: May 30, 2013 Accepted: June 6, 2013
A
dx.doi.org/10.1021/es4008933 | Environ. Sci. Technol. XXXX, XXX, XXX−XXX
Environmental Science & Technology
Article
dissolved humic acid (DHA) representing DOM. The pH values of these DHA solutions were adjusted to 7 using HCl, and the solution was diluted with background solution (0.02 M NaCl and 200 mg/L NaN3) to 250 mgC/L as a stock solution. The two DHA solutions were noted as DHA1 and DHA2 which were prepared using HA1 and HA4, respectively. Distinguishing the Adsorption Contributed by Suspended and Aggregated CNTs. The adsorption behavior of SMX in CNTs−DHA system was examined using a modified experimental design based on traditional batch sorption experiment, with a goal to distinguish the sorption contributed by suspended and aggregated CNTs. Briefly, 40 mL vials containing 20 mg of CNTs and 38 mL of background solution with different concentrations of DHA (0−110 mgC/L) were sonicated at maximum intensity for 90 min (AS 20500BT Ultrasonic Cleaner, Tianjin Automatic Science Instrument Co., Ltd.), and then shaken in an air-bath shaker at 25 °C for 5 d which was sufficient to reach DHA adsorption equilibrium. All the vials were centrifuged at 2000g for 20 min. The dark supernatants contained a significant fraction of suspended CNTs (up to 2.4%), and this DHA-mediated CNT suspension was stable for at least 7 d according to our preliminary study. The upper 50% of the supernatants (19 mL) were then carefully siphoned out and put into other vials. The CNTs− DHA systems were thus divided into suspended CNTs (S system) and suspended CNTs + aggregated CNTs (SA system). The adsorption difference between these two systems was the adsorption contributed by aggregated CNTs. The adsorption behaviors of SMX in both systems were studied using two batches of experiments: (a) SMX adsorption isotherms with various SMX and fixed DHA concentrations (the same suspended CNT concentration), and (b) SMX adsorption as affected by DHA concentrations (fixed SMX concentration but varied DOM and suspended CNT concentrations). According to our preliminary experiment, SMX adsorption on CNTs with sonication did not significantly suspend CNTs, and thus enhanced CNT suspending was solely attributed to DHA. Aliquots of 1 mL SMX solutions with different concentrations were added into all the glass vials (for both experiments a and b). All the vials were kept in the dark and were shaken in an air-bath shaker at 25 °C for 7 d. After equilibrated for 7 d, all the vials were centrifuged at 2000g for 20 min, and the supernatants were subjected to the quantification of suspended CNTs. The supernatants were also filtrated by 0.2 μm filters and subjected to quantification of SMX and DHA. Preliminary experiments indicated that the 0.2 μm filtration could efficiently retain the suspended CNTs but had no effect on SMX and DHA measurements when the filtrate volume was higher than 5 mL. Our previous studies also reported that no interaction between SMX and DHA was observed by dialysis equilibrium experiments and that DHA did not interfere with the quantification of SMX by HPLC. 26 Solution pH at equilibrium (around 7.5) was measured by a pH electrode (Leici Instruments, Shanghai, China). Detection of SMX, DHA, and CNTs. The concentrations of SMX in the supernatants were quantified at 265 nm by high performance liquid chromatography (HPLC, Agilent Technologies 1200) equipped with a reversed-phase C8 column (5 μm, 4.6 × 150 mm) and a UV detector. The mobile phase was 40:60 (v:v) of acetonitrile and deionized water with 0.1% acetic acid, and the flow rate was 1 mL/min. The detection limits were 0.05 mg/L. The concentrations of suspended CNTs were quantified using a UV−vis spectrometer (Shimadzu UV-2450)
hypothesis of this study is that the dispersed CNTs have more exposed surface area, and their adsorption for other contaminants will increase greatly compared to nondispersed CNTs. The latter part of the hypothesis could be examined by comparing the sorption with or without dispersed CNTs. But the first part of the hypothesis could not be tested using traditional surface area measurement (e.g., N2 or CO2 sorption experiment), because the sample-drying process will greatly alter the aggregation status of CNTs. The direct measurement of surface area in aqueous phase would provide the data of accessible surface area in water. 1H NMR relaxometry provided valuable information on pore size distribution in soils,20 wetting,21 swelling,22 and polymer adsorption on solid particles23,24 through water−solid interactions. A new instrument based on this technique was used to indirectly measure and compare the surface areas of suspended and aggregated CNTs in aqueous phase. An antibiotic, sulfamethoxazole (SMX), was used as the adsorbate in this study because of its negative charge at neutral pH and no interaction with DOM.25,26 SMX sorption behavior was investigated on three CNTs dispersed by two types of DOM. This study stresses that the sorption properties of suspended CNTs should be examined extensively because of their exposed surface area. This type of study will benefit the both the application of CNTs and their risk assessment.
■
EXPERIMENTAL SECTION Materials. CNTs used in the study were three multiwalled CNTs (purity >95%) which were hydroxylized (MH), carboxylized (MC), and graphitized (MG), respectively. They were purchased from Chengdu Organic Chemistry Co., Chinese Academy of Sciences. These CNTs were synthesized in the CH4/H2 mixture at 700 °C (MG at 2800 °C) by the chemical vapor deposition method. SMX was obtained from Bio Basic Inc. All the other chemicals were higher than analytical grade. All CNTs were characterized for their surface area (Autosorb-1C, Quantachrome) and elemental composition (MicroCube, Elementar, Germany). Preparation of DOM. A sediment sample was collected in Dianchi Lake (N 24°48′19.22″, E 102°39′52.16″), Yunnan province, China. The collected sample was freeze-dried, ground, and sieved through a 2 mm sieve. Plant residues were picked out manually. Humic acid (HA) was sequentially fractionated using traditional alkaline extraction methods.27 The sequentially fractionated HA from a same soil/sediment could be compared more easily than that from different soils/ sediments because the extraction from the same source excludes the interference due to different soil/sediment properties or structures. Briefly, 0.1 M NaOH and 0.1 M Na4P2O7 were mixed with the soil aggregates (50:1, v:w) to extract HA. After 12 h of equilibration, the mixture was centrifuged at 1000g for 10 min and the supernatant was collected. The extraction procedure was repeated successively, and the sequentially extracted solutions were collected separately. All the extractions were filtered, and HAs were precipitated with HCl. The precipitated HAs were washed using HF three times followed by distilled water until a negative test of chloride using AgNO3, freeze-dried, and ground into
