Article pubs.acs.org/est

Enhanced Transport of Phenanthrene and 1‑Naphthol by Colloidal Graphene Oxide Nanoparticles in Saturated Soil Zhichong Qi,† Lei Hou,† Dongqiang Zhu,‡ Rong Ji,‡ and Wei Chen*,† †

College of Environmental Science and Engineering/Ministry of Education Key Laboratory of Pollution Processes and Environmental Criteria/Tianjin Key Laboratory of Environmental Remediation and Pollution Control, Nankai University, Tianjin 300071, China ‡ State Key Laboratory of Pollution Control and Resource Reuse/School of the Environment, Nanjing University, Nanjing, Jiangsu 210093, China S Supporting Information *

ABSTRACT: With the increasing production and use of graphene oxide, the environmental implications of this new carbonaceous nanomaterial have received much attention. In this study, we found that the presence of low concentrations of graphene oxide nanoparticles (GONPs) significantly enhanced the transport of 1-naphthol in a saturated soil, but affected the transport of phenanthrene to a much smaller extent. The much stronger transport-enhancement effect on 1-naphthol was due to the significant desorption hysteresis (both thermodynamically irreversible adsorption and slow desorption kinetics) of GONP-adsorbed 1-naphthol, likely stemmed from the specific polar interactions (e.g., H-bonding) between 1-naphthol and GONPs. Increasing ionic strength or the presence of Cu(II) ion (a complexing cation) generally increased the transportenhancement capability of GONPs, mainly by increasing the aggregation of GONPs and thus, sequestering adsorbed contaminant molecules. Interestingly, modifying GONPs with Suwannee River humic acid or sodium dodecyl sulfate had little or essentially no effect on the transport-enhancement capability of GONPs, in contrast with the previously reported profound effects of humic acids and surfactants on the transport-enhancement capability of C60 nanoparticles. Overall, the findings indicate that GONPs in the aquatic environment may serve as an effective carrier for certain organic compounds that can interact with GONPs through strong polar interactions.

■

INTRODUCTION Graphene oxide (GO) is a new carbonaceous nanomaterial that has shown great promise in a number of applications, such as polymer composites, sensors, and biological applications, and is commonly used in mass production and manipulation of graphene and graphene-based nanomaterials.1 The rapidly increasing production and use of GO will increase the possibility of its environmental release, and understanding the fate and effects of GO is critical for its benign use and risk management.2−4 Because GO contains a range of O-functional groups such as carboxyl, carbonyl, hydroxyl, and phenol,5−7 it is hydrophilic and can easily disperse in aqueous solution,8−10 forming colloidal GO nanoparticles (GONPs) that can be quite mobile in aquatic environments.11,12 Additionally, it has been shown that GO has strong adsorption affinities for a range of environmental contaminants.13−17 Thus, when released into the environment GONPs may become an important medium with which toxic contaminants are associated and therefore, may significantly enhance the transport of the adsorbed contaminants.16,18,19 Thus, far, the effects of GONPs on the transport of environmental contaminants have not been studied. However, insights can be drawn from previous studies on the transport© 2014 American Chemical Society

enhancement effects of fullerene C60 nanoparticles (nC60) and carbon nanotubes (CNT).18,20−23 In our previous study, we found that the capability of nC60 to enhance the transport of organic contaminants (e.g., polychlorinated biphenyls and polycyclic aromatic hydrocarbons) is far greater than that of natural colloids such as dissolved organic matter.23 The strong transport-enhancement capability of nC60 stems primarily from the microporous structure of C60 aggregates that leads to desorption hysteresis of contaminant molecules entrapped within.23−26 In addition, solution chemistry conditions (e.g., humic acids and surfactants) can significantly affect the aggregation of C60 and consequently, the extent of nC60enhanced transport.21,22 The previous studies indicate that the capability of carbon nanoparticles to enhance contaminant transport is dictated largely by their aggregation properties. The aggregation properties of GONPs, in response to different solution chemistry conditions, differ distinctly from those of other Received: Revised: Accepted: Published: 10136

February 18, 2014 July 7, 2014 August 6, 2014 August 6, 2014 dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

Table 1. Experimental Setups and Breakthrough Results of Column Experiments effluent propertiesc

influent properties column

contaminant

contaminant concn. (μg/L)

GONPs concn. (mg/L)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

phenanthrene phenanthrene phenanthrene phenanthrene 1-naphthol 1-naphthol 1-naphthol 1-naphthol phenanthrene phenanthrene phenanthrene 1-naphthol 1-naphthol phenanthrene phenanthrene 1-naphthol 1-naphthol 1-naphthol

10.9 10.9 10.8 10.9 10.4 10.3 10.4 10.3 10.8 10.8 10.8 10.2 10.3 10.8 10.9 10.4 10.4 10.4

5.3 10.5 20.2 9.8 10.1 10.4 10.3 10.3 10.4 10.4 10.4 10.4 5.0

19

1-naphthol

10.3

5.0

20 21

1-naphthol 1-naphthol

10.3 10.4

5.0 5.0

5.2 10.3 20.1

adsorbed mass of contaminantb (%)

background solutiona 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 0.5 mM NaCl 2 mM NaCl 10 mM NaCl 10 mM NaCl (after) 10 mM NaCl 10 mM NaCl (after) 0.1 mM CuCl2 0.1 mM CuCl2 (after) 0.1 mM CuCl2 0.1 mM CuCl2 (after) 0.5 mM NaCl + 5 mg/L SRHA 0.5 mM NaCl + 5 mg/L SRHA (after) 0.5 mM NaCl + 5 mg/L SDS 0.5 mM NaCl + 5 mg/L SDS (after)

C/C0_GONPs (%)

C/C0_cont. (%)

97.2 96.2 97.6 95.2 90.0 89.7 91.8 91.2 80.8 82.1 81.8 81.8 100

58

99.2 ± 0.3

37.4 ± 0.5

53 54

100 ± 0.4 100 ± 1.3

35.2 ± 0.9 34.3 ± 0.4

96.8 ± 1.0 96.7 ± 0.5 98.2 ± 0.3 ± ± ± ± ± ± ± ± ± ± ± ± ±

0.8 0.8 0.6 0.1 1.6 1.5 0.9 0.6 0.6 1.0 1.1 0.9 0.3

0.75 1.8 2.5 11.9 9.3 34.3 59.7 81.2 3.5 3.9 4.9 81.4 87.0 4.3 4.9 50.1 80.7 33.9

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0 51 65 85 0 52 69 88 67 66 67 81 91 64 67 73 75 48

0.04 0.1 0.3 0.3 0.4 0.4 1.8 0.6 0.4 0.3 0.4 0.8 0.1 0.5 0.6 1.3 1.3 0.3

“After” in the parentheses indicates ionic strength was adjusted or Cu(II)/SRHA/SDS was added after adsorption of phenanthrene/1-naphthol to GONPs. bPercentage of contaminant adsorbed to GONPs in the influent. cAverage value of last three data points of respective breakthrough curve; cont. = contaminant.

a

carbon nanoparticles.11,12,27 For example, under relatively low ionic strength GO often exists as individually dispersed nanoflakes, and under high ionic strength, individual GO flakes may stack up to form GO aggregates.11,12,27 It is not known how the characteristic aggregation properties of GONPs are linked to their capabilities to enhance contaminant transport. Furthermore, in the previous studies on carbon nanoparticlesmediated contaminant transport, only nonpolar organic contaminants have been tested. GO can interact with organic contaminants via strong polar interactions.15,16 It is important to understand whether such specific polar interactions can affect the extent of desorption hysteresis of certain organic contaminants,28 and consequently, the extent of enhanced transport. The objectives of this study were to understand the potential strength of GONPs to enhance contaminant transport in porous media, and to understand the governing mechanisms controlling the transport-enhancement capability of GONPs. The effects of GONPsprepared under different solution chemistry conditions (including ionic strength, cations, dissolved organic matter, and surfactant)on the transport of a model nonpolar, hydrophobic aromatic (phenanthrene) and a model polar aromatic (1-naphthol) in a saturated soil were examined using column tests. The transport-enhancement effects of GONPs are linked to two key aspects: the polar interactions between GO and contaminants, and the aggregation properties of GONPs.

information provided by the supplier, the product was synthesized using a modified Hummers method;29 the sizes of the GO nanoplates were mainly 100−300 nm, and the thickness was 0.7−1.3 nm. The product contained 62.1% C (wt/wt), determined using an automatic elemental analyzer (Vario EL CUBE, Elementar Analysensysteme, Hanau, Germany); the surface C/O atomic ratio, determined with Xray photoelectron spectroscopy (MultiLab 2000, Thermo Electron Corp., England), was 2.2; and the Brunauer− Emmer−Teller (BET) surface area was 207.1 m2/g. The representative transmission electron microscope (TEM) images (JEM-2100, JEOL, Tokyo, Japan) of the GO product are shown in Supporting Information (SI) Figure S1; the Raman spectra (Renishew inVia Raman spectrometer, RM2000, UK) are shown in SI Figure S2 (the 2D/G intensity ratio further indicates that the product consists of more than one layer of nanosheet30,31); and the Fourier transform infrared (FTIR) transmission spectra (110 Bruker TENSOR 27 apparatus, Bruker Optics Inc., Germany) are shown in SI Figure S3. 14 C radioactively labeled phenanthrene and 1-naphthol with specific activities of 55 μCi/μmol were purchased from American Radiolabeled Chemicals (St. Louis, MO). Nonlabeled phenanthrene (99%) and 1-naphthol (99%) were purchased from Sigma−Aldrich (St. Louis, MO). The radioactively labeled compounds were diluted in methanol to obtain a phenanthrene stock solution of 46.6 mg/L and a 1-naphthol stock solution of 26.7 mg/L by mixing the labeled and nonlabeled compounds. Suwannee River humic acid (SRHA) was purchased from the International Humic Substance Society (St. Paul, MN), and was reported to be composed of 52.6% C (wt/wt), 4.3% H, 42.0% O, and 1.2% N.32 In this paper the concentrations of SRHA are

■

MATERIALS AND METHODS Materials. Graphene oxide (> 99%) was purchased from Nano Materials Tech Co. (Tianjin, China). Based on the 10137

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

Figure 1. Effects of different concentrations of GONPs (in 0.5 mM NaCl) on the transport of phenanthrene (Columns 1−4) and 1-naphthol (Columns 5−8). The left plots show the breakthrough curves of GONPs in each of the experiments, and the right plots show the breakthrough curves of phenanthrene or 1-naphthol in the respective experiments. The insert of the top-right plot was plotted on smaller scales to show the differences in phenanthrene breakthrough. The error bars represent the standard deviations of triplicates.

expressed as mg of SRHA per liter of solution. Sodium dodecyl sulfate (SDS) was purchased from Sigma−Aldrich (St. Louis, MO). Lula soil, containing 45% sand, 36% silt, and 19% clay,23 was collected from a ranch near Lula, OK. The fractional organic carbon (f OC) value of the soil was 0.37%. Preparation and Characterization of GONP Suspensions. Stock suspensions of GONPs were prepared using the following procedures. First, approximately 30 mg of GO powder was added to 300 mL of deionized (DI) water. Next, the mixture was ultrasonicated at 100 W (Vibra-Cell VCX800, Sonics & Material, Newtown, CT) for 30 min. Finally, the obtained GONP suspension was filtered with 0.45-μm membrane filters (Millipore Co., Billerica, MA) to remove large GONP aggregates. The concentrations of GONPs in the stock suspensions were verified by measuring the total organic carbon content (Shimadzu Scientific Instruments, Columbia, MD).33 The working suspensions, i.e., the influents of the column experiments, were obtained by diluting a stock suspension in solutions of varied solution chemistry. The particle size distribution (SI Figure S4) was measured by dynamic light scattering (DLS), using a ZetaPALS (Brookhaven Instruments, Holtsville, NY). Aggregation properties of GONPs were examined with TEM, and the samples were prepared by air-drying a drop of a suspension onto a copper TEM grid (Electron Microscopy Sciences, Hatfield, PA). Atomic force microscopy (AFM) analysis was carried out

with a J scanner of a Veeco Multimode Nanoscope VIII (Santa Barbara, CA). The detailed sample preparation methods are given in the SI. Column Experiments. Soil was dry-packed into Omnifit borosilicate glass columns (10 cm × 0.66 cm, Bio-Chem Valve Inc., Boonton, NJ) with 10-μm stainless-steel screens (Valco Instruments Inc., Houston, TX) on both ends. The columns were operated in an upward direction using syringe pumps (KD Scientific, Holliston, MA). The soil-packed columns were equilibrated by sequentially flushing with 100 mL (∼ 85 pore volumes (PV)) of DI water at a flow rate of 3 mL/h followed by 180 mL (∼ 150 PV) of background electrolyte solution.12 The porosity and dead volume were determined by inverse fitting the breakthrough curves (BTCs) of KBr (used as a conservative tracer). The detailed experimental setups of the column tests are summarized in Table 1; the column properties are summarized in SI Table S1. To prepare the influents, aliquots of a GONP stock suspension were added in different electrolytes in amber glass vials to give a GONP concentration of 5.0−20.2 mg/L in each vial. Immediately after adding the GONP stock suspension, a phenanthrene or 1-naphthol stock solution in methanol was added to each vial with a microsyringe to give a total phenanthrene or 1-naphthol concentration of approximately 10 μg/L (the volume percentage of methanol was kept below 0.05% (v/v) to minimize cosolvent effects). The vials 10138

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

and 5), and the decrease in aqueous phase concentration would have driven the desorption of the contaminants from GONPs. If the desorption from GONPs is instantaneous and reversible, then the percentage breakthrough of contaminant (i.e., the ratio of contaminant concentration in the effluent, C, to the contaminant concentration in the influent, C0) can be estimated using the following equation (see SI for the detailed derivation):

were sealed with Teflon-lined screw caps and tumbled endover-end at 3 rpm for 7 d. In selected experiments (Columns 11, 13, 15, 17, 19, and 21, Table 1), the influents were prepared by first diluting a GONP stock suspension in DI water or 0.5 mM NaCl solution, and next adding the phenanthrene or 1naphtol stock solution and equilibrating for 3 d (tumbling at 3 rpm), and then adding stock solutions of NaCl, CuCl2, or SRHA/SDS, and equilibrating again for 4 d (tumbling at 3 rpm). A negligible depletion solid-phase micro extraction approach was used to determine the concentrations of dissolved and GONP-adsorbed phenanthrene or 1-naphthol in the influents16 (see SI for detailed procedures). The adsorption/desorption isotherms of phenanthrene and 1naphthol to GONPs prepared under different solution chemistry conditions and to Lula soil were obtained to characterize the adsorptive interactions between the contaminants and GONPs/porous medium (see SI for detailed procedures). In a typical column experiment, the influent was loaded to the soil column with a syringe pump. The effluent was collected at every 3−4 PV. The collected sample was split into three aliquots, and the concentrations of both GONPs and phenanthrene/1-naphthol were measured. The concentrations of GONPs were determined by measuring the UV absorbance at 230 nm (UV-2401 UV/vis spectrophotometer, Shimadzu Scientific Instruments, Japan);34 the detection limit was 0.03 mg/L (aggregation of GONPs had negligible effects on UV absorbance; see SI Figure S5). In the presence of SRHA, the concentrations of GONPs were determined using the method of Chen et al.,35 by obtaining the calibration curve of UV absorbance of GO (at 230 nm) as a function of GO concentration in the presence of 5 mg/L SRHA (SI Figure S6).35 Phenanthrene and 1-naphthol were quantified with a liquid scintillation counter (LS6500, Beckman Coulter, Fullerton, CA). The detection limits were ∼0.04 μg/L for phenanthrene and ∼0.03 μg/L for 1-naphthol, respectively.

C /C 0=

V + CGONPs·V ·Kd_GONPs V + m _soil ·Kd_soil + CGONPs·V ·Kd_GONPs

(1)

where CGONPs (kg/L) is the concentration of GONPs in the effluent; V (ml) is the volume of the suspension flowed through the column; m_soil (g) is the mass of soil in the column; and Kd_GONPs (L/kg) and Kd_soil (L/kg) are the distribution coefficients of 1-naphthol to GONPs and soil, respectively (Kd_GONPs and Kd_soil can be obtained from the sorption isotherms in SI Figure S7). However, eq 1 would underestimate the breakthrough if desorption from GONPs is hysteretic (i.e., thermodynamically irreversible and/or kinetically slow). In SI Figures S8 and S9 the calculated C/C0 values are compared with the experimental data. It can be seen that the breakthrough of phenanthrene can more or less be predicted using eq 1 (at 20.1 mg/L GONPs breakthrough is underestimated). However, at all three GONP concentrations eq 1 significantly underestimates the breakthrough of 1-naphthol, indicating that desorption of 1-naphthol from GONPs deviated profoundly from the assumptions of instantaneous and reversible adsorption. Irreversible adsorption of 1-naphthol to GONPs was indicated by the desorption isotherm data (SI Figure S10); however, slow desorption kinetics cannot be completely ruled out. The markedly different desorption potentials between GONP-bound phenanthrene and GONPbound 1-naphthol indicate that the significant desorption hysteresis of 1-naphthol was likely attributable to the specific adsorptive interactions (e.g., H-bonding) between 1-naphthol and the surface O-functionalities of GONPs16 (the FTIR spectra of GO in SI Figure S3 verify that the GO product contains considerable amounts of strong H-bonding moieties such as phenolic hydroxyl group). It was proposed that polar interactions can result in irreversible adsorption of phenolic compounds to soil organic matter,36,37 which is similar to GO in the types and abundance of O-functionalities. Additionally, Pan et al. observed strong desorption hysteresis of 17α-ethinyl estradiol and bisphenol A from multiwalled CNT,38 and Ji et al. observed irreversible adsorption of phenol to single-walled CNT.39 A very interesting observation in Figure 1 was that the capability of GONPs to enhance the transport of phenanthrene appeared to be markedly weaker than that of nC60. Based on our previous study,23 nC60 with a concentration as low as 1.55 mg/L could achieve the similar transport-enhancement effect (in terms of C/C0 of phenanthrene) of 13.1 mg/L of GONPs (estimated using SI Figure S11). The superior transportenhancement effect of nC60 is even more striking when considering its much weaker adsorption affinity for phenanthrene than that of GONPsthe observed Kd values are 105.56−105.83 L/kg for GONPs (calculated from SI Figure S9) but only ∼10 4.38 L/kg for nC 60 . 23 Thus, at an nC 60 concentration of 1.55 mg/L, only 3.7% phenanthrene was adsorbed to nC60,23 whereas at a GONP concentration of 13.1 mg/L, 77% phenanthrene would be in the adsorbed phase. The

■

RESULTS AND DISCUSSION Enhanced Transport of Phenanthrene and 1-Naphthol by GONPs. The effects of GONPsat different concentrationson the transport of phenanthrene and 1naphthol are shown in Figure 1. A striking overall observation was that the presence of GONPs significantly enhanced the transport of 1-naphthol, but only affected the transport of phenanthrene to a much smaller extent. Because in all six GONP-mediated experiments (Columns 2−4 for phenanthrene; Columns 6−8 for 1-naphthol; Table 1) GONPs exhibited similar mobility: essentially 100% breakthrough was reached within a few PV (Figure 1a and b, left-hand-side plots), the vastly different effects of GONPs on the transport of phenanthrene and 1-naphthol appear to be related to the different capabilities of GONPs to bind the two compounds. For the GONP-mediated experiments, the mass of phenanthrene or 1-naphthol in the influent was distributed in equilibrium between GONPs and the aqueous solution. GONPs appeared to have similar adsorption affinities for the two compounds (see the adsorption isotherms in SI Figure S7). Accordingly, at a given GONP concentration, the mass fraction of GONP-adsorbed contaminant in the influent was similar between the two compounds (Table 1). When flowing through the soil columns, dissolved contaminants could be easily sorbed by soil grains (over 99% phenanthrene and over 90% 1naphthol could be retained, as demonstrated using Columns 1 10139

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

Figure 2. Effects of ionic strength on the capability of GONPs to enhance the transport of phenanthrene (Columns 3, 9−11) and 1-naphthol (Columns 7, 12, 13). The left plots show the breakthrough curves of GONPs in each of the experiments, and the right plots show the breakthrough curves of phenanthrene or 1-naphthol in the respective experiments. The term “after” in the parentheses indicates that ionic strength was adjusted after the adsorption of phenanthrene/1-naphthol to GONPs. The error bars represent the standard deviations of triplicates.

much stronger capability of nC60 than GONPs to enhance the transport of phenanthrene strongly indicates that the capabilities of carbon nanoparticles to enhance contaminant transport are largely controlled by their aggregation properties. Compared with the microporous nC60, well dispersed GO nanoflakes (see the AFM height profiles of GONPs in SI Figure S12) are much less effective in withholding adsorbed phenanthrene molecules. Effects of Ionic Strength on Transport-Enhancement Capability of GONPs. To further understand the important role of aggregation properties of GONPs in their transportenhancement capabilities, we conducted GONP-mediated transport experiments at higher ionic strength. Increasing ionic strength can result in aggregation of GONPs,11,12,27 which might lead to enhanced desorption hysteresis of contaminant molecules from GONPs, and consequently, enhanced transport of GONP-adsorbed compounds. Figure 2 compares the breakthrough data of phenanthrene and 1-naphthol in the presence of ∼10 mg/L GONPs at different ionic strengths of 0.5, 2, and 10 mM NaCl (Columns 3, 7, and 9−13). Note that for both phenanthrene and 1-naphthol, one specially designed experiment was carried out at 10 mM NaCl (Columns 11 and 13), in which adsorption equilibrium of phenanthrene/1naphthol was reached before allowing GONPs to aggregate by increasing the ionic strength of the influent to 10 mM NaCl. Under such experimental condition, it was anticipated that a

fraction of adsorbed contaminant molecules would have been entrapped during the aggregation of GONPs. Figure 2 shows that increasing ionic strength of the influent resulted in noticeable changes in the transport of GONPs, an effect consistent with the literature;11,12,27 however, the C/C0 values of GONPs at different ionic strength became similar after approximately 35 PV. Much greater effects were observed on the transport of phenanthrene and 1-naphthol, in that increasing ionic strength in the influent consistently enhanced the transport of both contaminants. Furthermore, a greater enhancement was observed when aggregation of GONPs occurred after the adsorption of contaminants. These observations are in line with the above-mentioned argument that the extent of desorption hysteresis depends largely on the aggregation properties of nanoparticles. Based on the AFM height profiles (SI Figure S12), aggregation of GONPs at elevated ionic strength (e.g., 10 mM NaCl) was evident, in that the thickness of GONP aggregates increased from 1.4 nm at 0.5 mM NaCl to 3.2 nm at 10 mM, indicating stacking of GO flakes27 (also see the particle size distribution data in SI Figure S4). Interestingly, the desorption isotherm data in SI Figure S13 seem to indicate that the extents of irreversible adsorption did not increase significantly with the aggregation of GONPs. Thus, it is possible that the aggregation of GONPs mainly affected the desorption kinetics of adsorbed molecules. 10140

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

Figure 3. Effects of Cu(II) ion on the capability of GONPs to enhance the transport of phenanthrene (Columns 3, 14, 15) and 1-naphthol (Columns 7, 16, 17). The left plots show the breakthrough curves of GONPs in each of the experiments, and the right plots show the breakthrough curves of phenanthrene or 1-naphthol in the respective experiments. The term “after” in the parentheses indicates that Cu(II) was added after the adsorption of phenanthrene/1-naphthol to GONPs. The error bars represent the standard deviations of triplicates.

Effects of Complexing Cation on Transport-Enhancement Capability of GONPs. The effects of Cu(II) ion (at 0.1 mM), a common divalent cation in aquatic environments and a strong complexing agent, on the transport-enhancement capability of GONPs were tested and the results are shown in Figure 3. Again, for both phenanthrene and 1-naphthol one specially designed experiment was conducted, by adding Cu(II) after the adsorption of phenanthrene/1-naphthol to GONPs (Columns 15 and 17). Figure 3 shows that Cu(II) had different effects on the transport-enhancement capabilities of GONPs for phenanthrene and 1-naphthol. For phenanthrene, the presence of Cu(II) consistently enhanced the transport-enhancement capability of GONPs, regardless whether Cu(II) was present during the adsorption of phenanthrene to GONPs or added after the adsorption (Figure 3a). For 1-naphthol, however, only when Cu(II) was added after the adsorption of 1-naphthol to GONPs, was the breakthrough of 1-naphthol significantly enhanced (compared with the case when the influent contained 0.5 mM NaCl). As a divalent cation, Cu(II) can cause significant aggregation of GONPs. The AFM height profiles (SI Figure S12) confirm that significant aggregation of GONPs occurred at 0.1 mM Cu(II): the thickness of GONPs was 25.1 nm at 0.1 mM Cu(II), 18 times higher than the thickness of 1.4 nm at 0.5 mM NaCl. The significantly enhanced aggregation of GONPs in the presence of Cu(II) should have increased the likelihood of

desorption hysteresis of the contaminants, via the abovementioned mechanism; this is also consistent with the results of desorption experiments (see SI Figure S14). Note that as a complexing cation Cu(II) can enhance the adsorption of 1-naphthol through a bridging mechanism, i.e., by forming complexes with both 1-naphthol and surface Ofunctional groups of GONPs. This is consistent with the stronger adsorption of 1-naphthol to GONPs in the presence of Cu(II), compared with the case when the solution contained only NaCl (SI Figure S14). In comparison, the adsorption affinities of GONPs for phenanthrene were similar at 0.1 mM Cu(II) and at 0.5 mM NaCl (SI Figure S15). One intriguing observation in Figure 3b was that whether Cu(II) was present during or after the adsorption of 1-naphthol to GONPs made a large difference in the breakthrough of 1-naphthol. In comparison, the timing of Cu(II) addition played a much smaller role in the breakthrough of phenanthrene (Figure 3a), nor did the timing of NaCl addition (at 10 mM) affect the breakthrough of 1-naphthol to such a large extent (Figure 2b). This interesting effect of Cu(II) on the transport of 1-naphthol might be linked to the specific mechanisms via which 1naphthol molecules are bound to GONPs under the two different scenarios. For example, the relative mass fractions of 1-naphthol bound to GONPs directly and 1-naphthol bound to GONPs through the bridging of Cu(II) likely would differ for the two cases, and the specific binding mechanisms might affect 10141

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

Figure 4. Effects of SRHA and SDS on the capability of GONPs to enhance the transport of 1-naphthol (Columns 6, 18−21). The left plots show the breakthrough curves of GONPs in each of the experiments, and the right plots show the breakthrough curves of 1-naphthol in the respective experiments. The term “after” in the parentheses indicates that SRHA/SDS was added after the adsorption of 1-naphthol to GONPs. The error bars represent the standard deviations of triplicates.

S16 and S18 show that SDS had negligible effects on the adsorption/desorption). The small or negligible effects of SRHA/SDS modification on the transport-enhancement capability of GONPs were surprising at first sight. In our previous studies,22 we found that modifying nC60 with trace amount of SRHA or SDS could vastly increase the transport-enhancement capability of nC60, because the intercalation of SRHA or SDS likely have enhanced the pore volume and altered the pore structures of nC60. The minimal effects of SRHA and SDS observed for GONPs were likely because in the presence of SRHA or SDS aggregation of GONPs was significantly inhibited and GONPs existed mainly as individually dispersed flakes. For example, the AFM height profiles (SI Figure S12) show that the thickness of the SRHA/ SDS-modified GONPs was only 1.3−1.4 nm, essentially the same as GONPs in DI water or 0.5 mM NaCl. Thus, the results in Figure 4 further highlight the important role of aggregation properties of carbon nanoparticles in their transport-enhancement capabilities. Environmental Implications. The findings of this study underline the potentially significant effects of GONPs on the transport of organic contaminants in aquatic environments. The extent of adsorption enhancement, however, is highly contaminant-specific. In particular, GONPs may serve as an effective carrier for certain polar organic compounds that can interact with GONPs through strong polar interactions.

the extent of desorption hysteresis. More studies are needed to fully understand the complex roles of Cu(II). Effects of Humic Acid and Surfactant Modification on Transport-Enhancement Capability of GONPs. In Figure 4 the effects of SRHA (as a model dissolved organic matter) and SDS (as a model surfactant) on the transport-enhancement capability of GONPs are shown. Figure 4a shows that when SRHA was present during the adsorption of 1-naphthol to GONPs (Column 18), transport of 1-naphthol was slightly inhibited, but only in the beginning of the experiment (up to approximately 25 PV); when SRHA was added after the adsorption of 1-naphthol to GONPs (Column 19), transport of 1-naphthol was enhanced slightly. SRHA can interact with GONPs through van der Waals interaction, π−π interactions, and H-bonding.40 When SRHA was present during the adsorption of 1-naphthol to GONPs, it could compete with 1-naphthol for the adsorption sites on GONPs. This is evident when comparing the adsorption isotherms of 1-naphthol in the presence and absence of SRHA (SI Figure S16). However, when SRHA was added after the adsorption of 1-naphthol to GONPs, the coating of SRHA on the surfaces of GONPs (see the TEM image in SI Figure S17) might have hindered the desorption of 1-naphthol from GONPs (possibly by inhibiting desorption kinetics, but not the extent of irreversible adsorption; SI Figure 18). Figure 4b shows that SDS had essentially no effects on the transport of 1-naphthol (SI Figures 10142

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

(11) Feriancikova, L.; Xu, S. Deposition and remobilization of graphene oxide within saturated sand packs. J. Hazard. Mater. 2012, 235−236, 194−200. (12) Lanphere, J. D.; Luth, C. J.; Walker, S. L. Effects of solution chemistry on the transport of graphene oxide in saturated porous media. Environ. Sci. Technol. 2013, 47, 4255−4261. (13) Gao, Y.; Li, Y.; Zhang, L.; Huang, H.; Hu, J.; Shah, S. M.; Su, X. Adsorption and removal of tetracycline antibiotics from aqueous solution by graphene oxide. J. Colloid Interface Sci. 2012, 368, 540− 546. (14) Pavagadhi, S.; Tang, A. L. L.; Sathishkumar, M.; Loh, K. P.; Balasubramanian, R. Removal of microcystin-LR and microcystin-RR by graphene oxide: Adsorption and kinetic experiments. Water Res. 2013, 47, 4621−4629. (15) Pei, Z.; Li, L.; Sun, L.; Zhang, S.; Shan, X.; Yang, S.; Wen, B. Adsorption characteristics of 1,2,4-trichlorobenzene, 2,4,6-trichlorophenol, 2-naphthol and naphthalene on graphene and graphene oxide. Carbon 2013, 51, 156−163. (16) Wang, F.; Haftka, J. J. H.; Sinnige, T. L.; Hermens, J. L. M.; Chen, W. Adsorption of polar, nonpolar, and substituted aromatics to colloidal graphene oxide nanoparticles. Environ. Pollut. 2014, 186, 226−233. (17) Zhao, G.; Li, J.; Ren, X.; Chen, C.; Wang, X. Few-layered graphene oxide nanosheets as superior sorbents for heavy metal ion pollution management. Environ. Sci. Technol. 2011, 45, 10454−10462. (18) Hofmann, T.; von der Kammer, F. Estimating the relevance of engineered carbonaceous nanoparticle facilitated transport of hydrophobic organic contaminants in porous media. Environ. Pollut. 2009, 157, 1117−1126. (19) Sen, T. K.; Khilar, K. C. Review on subsurface colloids and colloid-associated contaminant transport in saturated porous media. Adv. Colloid Interface Sci. 2006, 119, 71−96. (20) Fang, J.; Shan, X.; Wen, B.; Huang, R. Mobility of TX 100 suspended multiwalled carbon nanotubes (MWCNTs) and the facilitated transport of phenanthrene in real soil columns. Geoderma 2013, 207−208, 1−7. (21) Wang, L.; Fortner, J. D.; Hou, L.; Zhang, C.; Kan, A. T.; Tomson, M. B.; Chen, W. Contaminant-mobilizing capability of fullerene nanoparticles (nC60): Effect of solvent-exchange process in nC60 formation. Environ. Toxicol. Chem. 2013, 32, 329−336. (22) Wang, L.; Huang, Y.; Kan, A. T.; Tomson, M. B.; Chen, W. Enhanced transport of 2,2′,5,5′-polychlorinated biphenyl by natural organic matter (NOM) and surfactant-modified fullerene nanoparticles (nC60). Environ. Sci. Technol. 2012, 46, 5422−5429. (23) Zhang, L.; Wang, L.; Zhang, P.; Kan, A. T.; Chen, W.; Tomson, M. B. Facilitated transport of 2,2′,5,5′-polychlorinated biphenyl and phenanthrene by fullerene nanoparticles through sandy soil columns. Environ. Sci. Technol. 2011, 45, 1341−1348. (24) Cheng, X. K.; Kan, A. T.; Tomson, M. B. Naphthalene adsorption and desorption from aqueous C60 fullerene. J. Chem. Eng. Data 2004, 49, 675−683. (25) Cheng, X. K.; Kan, A. T.; Tomson, M. B. Uptake and sequestration of naphthalene and 1,2-dichlorobenzene by C60. J. Nanopart. Res. 2005, 7, 555−567. (26) Yang, K.; Xing, B. S. Desorption of polycyclic aromatic hydrocarbons from carbon nanomaterials in water. Environ. Pollut. 2007, 145, 529−537. (27) Qi, Z.; Zhang, L.; Wang, F.; Hou, L.; Chen, W. Factors controlling transport of graphene oxide nanoparticles in saturated sand columns. Environ. Toxicol. Chem. 2014, 33, 998−1004. (28) Wu, W.; Jiang, W.; Zhang, W.; Lin, D.; Yang, K. Influence of functional groups on desorption of organic compounds from carbon nanotubes into water: Insight into desorption hysteresis. Environ. Sci. Technol. 2013, 47, 8373−8382. (29) Becerril, H. A.; Mao, J.; Liu, Z.; Stoltenberg, R. M.; Bao, Z.; Chen, Y. Evaluation of solution-processed reduced graphene oxide films as transparent conductors. ACS Nano 2008, 2, 463−470.

Additionally, the transport-enhancement effects of GONPs might be increased under solution chemistry conditions that favor the aggregation of GONPs (but without resulting in significant settlement of GONPs). The potential effects of GONPs as a contaminant carrier should be given a full consideration in assessing the potential environmental risks of GONPs.

■

ASSOCIATED CONTENT

S Supporting Information *

Procedures of AFM analysis, procedures of adsorption/ desorption experiments and isotherms to GONPs, soil and fiber, determination of GONP concentrations in the presence of SRHA, estimation of contaminant breakthrough in the presence of GONPs, column properties, TEM images, Raman and FTIR spectra of GO product, particle size distribution of GONPs, effects of aggregation on GO analysis, phenanthrene breakthrough as a function of GONP concentration, AFM images and height profiles of GONPs, and TEM images of SRHA-coated GONPs. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone/fax: 86-22-6622-9516; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This project was supported by the Ministry of Science and Technology of China (Grant 2014CB932001), and the National Natural Science Foundation of China (Grants 21237002 and 21177063).

■

REFERENCES

(1) Chen, D.; Feng, H.; Li, J. Graphene oxide: Preparation, functionalization, and electrochemical applications. Chem. Rev. 2012, 112, 6027−6053. (2) Schmidt, C. W. Nanotechnology-related environment, health, and safety research. Environ. Health Perspect. 2009, 117, A158−A161. (3) Akhavan, O.; Ghaderi, E.; Emamy, H.; Akhavan, F. Genotoxicity of graphene nanoribbons in human mesenchymal stem cells. Carbon 2013, 54, 419−431. (4) Akhavan, O.; Ghaderi, E.; Akhavan, A. Size-dependent genotoxicity of graphene nanoplatelets in human stem cells. Biomaterials 2012, 33, 8017−8025. (5) Li, X. L.; Wang, H. L.; Robinson, J. T.; Sanchez, H.; Diankov, G.; Dai, H. J. Simultaneous nitrogen doping and reduction of graphene oxide. J. Am. Chem. Soc. 2009, 31, 15939−15944. (6) Bai, H.; Li, C.; Wang, X.; Shi, G. On the gelation of graphene oxide. J. Phys. Chem. C 2011, 115, 5545−5551. (7) Lu, N.; Yin, D.; Li, Z.; Yang, J. Structure of graphene oxide: Thermodynamics versus kinetics. J. Phys. Chem. C 2011, 115, 11991− 11995. (8) Kim, J.; Cote, L. J.; Kim, F.; Yuan, W.; Shull, K. R.; Huang, J. Graphene oxide sheets at interfaces. J. Am. Chem. Soc. 2010, 132, 8180−8186. (9) Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S. Preparation and characterization of graphene oxide paper. Nature 2007, 448, 457−460. (10) Li, D.; Müller, M. B.; Gilje, S.; Kaner, R. B.; Wallace, G. G. Processable aqueous dispersions of graphene nanosheets. Nat. Nanotechnol. 2008, 3 (6), 101−105. 10143

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144

Environmental Science & Technology

Article

(30) Calizo, I.; Balandin, A. A.; Bao, W.; Miao, F.; Lau, C. N. Temperature dependence of the Raman spectra of graphene and graphene multilayers. Nano Lett. 2007, 7, 2645−2649. (31) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Raman spectrum of graphene and graphene layers. Phys. Rev. Lett. 2006, 97, 18401. (32) International Humic Substances Society; http://www. humicsubstances.org/elements.html. (33) Wang, X.; Tian, H.; Yang, Y.; Wang, H.; Wang, S.; Zheng, W.; Liu, Y. Reduced graphene oxide/CdS for efficiently photocatalystic degradation of methylene blue. J. Alloys Compd. 2012, 524, 5−12. (34) Wang, G.; Wang, B.; Park, J.; Yang, J.; Shen, X.; Yao, J. Synthesis of enhanced hydrophilic and hydrophobic graphene oxide nanosheets by a solvothermal method. Carbon 2009, 47, 68−72. (35) Chen, G.; Liu, X.; Su, C. Distinct effects of humic acid on transport and retention of TiO2 rutile nanoparticles in saturated sand columns. Environ. Sci. Technol. 2012, 46 (13), 7142−7150. (36) Isaacon, P. J.; Frinj, C. R. Nonreversible sorption of phenolic compounds by sediment fractions: The role of sediment organic matter. Environ. Sci. Technol. 1984, 18 (1), 43−48. (37) Brusseau, M. L.; Rao, P. S. C.; Gillham, R. W. Sorption nonideality during organic contaminant transport in porous media. Crit. Rev. Environ. Control 1989, 19, 33−99. (38) Pan, B.; Lin, D.; Mashayekhi, H.; Xing, B. Adsorption and hysteresis of bisphenol A and 17α-ethinyl estradiol on carbon nanomaterials. Environ. Sci. Technol. 2008, 42, 5480−5485. (39) Ji, L.; Liu, F.; Xu, Z.; Zheng, S.; Zhu, D. Zeolite-templated microporous carbon as a superior adsorbent for removal of monoaromatic compounds from aqueous solution. Environ. Sci. Technol. 2009, 43, 7870−7876. (40) Hartono, T.; Wang, S.; Ma, Q.; Zhu, Z. Layer structured graphite oxide as a novel adsorbent for humic acid removal from aqueous solution. J. Colloid Interface Sci. 2009, 333, 114−119.

10144

dx.doi.org/10.1021/es500833z | Environ. Sci. Technol. 2014, 48, 10136−10144