Interactions of Graphene Oxide Nanomaterials with ... - ACS Publications

Article pubs.acs.org/est

Interactions of Graphene Oxide Nanomaterials with Natural Organic Matter and Metal Oxide Surfaces Indranil Chowdhury,† Matthew C. Duch,‡ Nikhita D. Mansukhani,‡ Mark C. Hersam,‡ and Dermont Bouchard*,§ †

National Research Council Research Associate, Athens, Georgia 30605, United States Departments of Material Science and Engineering, Chemistry, and Medicine, Northwestern University, Evanston, Illinois 60208, United States § National Exposure Research Laboratory, Ecosystem Research Division, United States Environmental Protection Agency, Athens, Georgia 30605, United States ‡

S Supporting Information *

ABSTRACT: Interactions of graphene oxide (GO) nanomaterials with natural organic matter (NOM) and metal oxide surfaces were investigated using a quartz crystal microbalance with dissipation monitoring (QCM-D). Three different types of NOM were studied: Suwannee River humic and fulvic acids (SRHA and SRFA) and alginate. Aluminum oxide surface was used as a model metal oxide surface. Deposition trends show that GO has the highest attachment on alginate, followed by SRFA, SRHA, and aluminum oxide surfaces, and that GO displayed higher interactions with all investigated surfaces than with silica. Deposition and release behavior of GO on aluminum oxide surface is very similar to positively charged poly-L-lysine-coated surface. Higher interactions of GO with NOM-coated surfaces are attributed to the hydroxyl, epoxy, and carboxyl functional groups of GO; higher deposition on alginate-coated surfaces is attributed to the rougher surface created by the extended conformation of the larger alginate macromolecules. Both ionic strength (IS) and ion valence (Na+ vs Ca2+) had notable impact on interactions of GO with different environmental surfaces. Due to charge screening, increased IS resulted in greater deposition for NOM-coated surfaces. Release behavior of deposited GO varied significantly between different environmental surfaces. All surfaces showed significant release of deposited GO upon introduction of low IS water, indicating that deposition of GO on these surfaces is reversible. Release of GO from NOM-coated surfaces decreased with IS due to charge screening. Release rates of deposited GO from alginate-coated surface were significantly lower than from SRHA and SRFA-coated surfaces due to trapping of GO within the rough surface of the alginate layer. toxic toward organisms, with GO being the most toxic.10−12 Increased applications and production will likely lead to release of graphene-based nanomaterials in the environment; hence, fate and transport of these emerging materials needs to be investigated for environmentally sustainable implementation of this industry.13,14

1. INTRODUCTION Graphene, a one carbon-atom-thick planar nanosheet, is known for its remarkable electrical, mechanical, and thermal properties.1,2 Due to their extraordinary physicochemical properties, graphene-based nanomaterials are being considered for numerous current and future applications in the electronic, medical, energy, and environmental sectors.3−7 Bulk quantities of graphene-based materials are usually synthesized from the oxidized form of graphene known as graphene oxide (GO).3 Due to oxygen contaning functional groups, GO has been considered as a precursor for graphene composites and hybrids.8,9 Graphene-based materials have been found to be © 2014 American Chemical Society

Received: Revised: Accepted: Published: 9382

April 28, 2014 July 14, 2014 July 15, 2014 July 15, 2014 dx.doi.org/10.1021/es5020828 | Environ. Sci. Technol. 2014, 48, 9382−9390

Environmental Science & Technology

Article

Several studies15,16 have investigated the interactions of graphene-based nanomaterials with organisms. For GO nanowalls, cellular membrane damage of bacteria caused by sharp edges of nanowalls is the primary mechanism, whereas higher charge transfer between reduced GO and bacteria can result in better antibacterial activity of reduced nanowalls.16 Another study15 showed that size affects the genotoxicity of graphene nanoplatelets in human stem cells. Reduced GO with lower lateral dimensions has been observed to be more toxic than GO with larger lateral dimensions.15 It has been shown that metal oxide semiconductors such as TiO2,17 ZnO,18 and WO319 can reduce GO under UV irradiation and that metal oxide semiconductors such as ZnO20 and TiO221 can photodegrade GO following photoreduction. Despite significant research on graphene nanomaterials’ applications, research on their environmental exposure is in initial stages, and there are few published reports22−25 on their fate. While aggregation of GO23 was found to follow colloidal theory, including Derjaguin−Landau−Verwey−Overbeek (DLVO) theory26 and the Schulze-Hardy rule,27 another study on GO aggregation kinetics24 found pH-dependent deprotonation of carboxylic groups on the edge of GO can influence the aggregation of GO, as well as GO’s orientation during aggregation. It has also been suggested that face-to-face aggregation is dominant at low pH, while edge-to-edge aggregation is significant in the presence of CaCl2.24 Ca2+ ions can form a bridge with GO functional groups and destabilize GO suspension.23 Stability of GO can be notably improved in the presence of natural organic matter (NOM) due to steric repulsion.23 Long-term stability studies23 showed that GO is highly stable in natural surface water, indicating interactions with other environmental surfaces may play key roles in the fate of these emerging materials in aquatic environments. Studies on GO transport through porous media25,28,29 found that GO is highly mobile in saturated porous media, that retention is reversible, that increased ionic strength (IS) resulted in greater GO retention primarily due to straining, and that observed transport trends can be explained by extended DLVO theory. A study on GO deposition and release using a quartz crystal microbalance with dissipation monitoring (QCM-D)22 showed deposition and release of GO can be influenced by ion valence, with CaCl2 systems exhibiting the lowest release due to bridging effects. NOM is mainly composed of humic substances and polysaccharides and is ubiquitous in the aquatic environment.30−32 Most natural surfaces in the aquatic environment are coated with NOM, and interactions of nanomaterials with NOM surfaces will be a major determinant of their fate in the environment. Besides NOM surfaces, metal oxide surfaces are some of the most common environmental surfaces. They are often positively charged at environmental pH (pH 5−9)30 and thus provide favorable attachment surfaces for negatively charged nanomaterials like GO. As a result, metal oxide surfaces can be one of the important factors controlling GO’s fate in the environment, but to date, there has been no study on interactions of GO with NOM and metal oxide surfaces. The objective of this work is to characterize GO interactions with environmentally relevant surfaces. In this study three different model NOM surfaces are investigated: Suwannee River humic acid (SRHA), Suwannee River fulvic acid (SRFA), and alginate. The model metal oxide surface investigated is aluminum oxide.

2. MATERIALS AND METHODS 2.1. GO Preparation and Characterization. A modified Hummers method was used to synthesize graphene oxide.10 Natural graphite flakes (3061 grade material from Asbury Graphite Mills) were treated with concentrated sulfuric acid and other oxidizing agents, followed by filtration, washing, and centrifugation to remove residual contaminants. Average thickness of synthesized GO was 0.85 ± 0.21 nm and average square root of the area was 179.2 ± 111.5 nm. Detailed physical dimensions, colloidal properties, and aggregation kinetics of synthesized GO are described in our previously published paper.23 2.2. Aquatic Chemistry. Laboratory grade NaCl and CaCl2 (Fisher Inc., PA) in a range of ionic strengths (IS) were used. pH was unadjusted at pH 5.5 ± 0.3. Prior to experiments, all electrolyte solutions were filtered through 0.1 μm filters (Anotop 25, Whatman, Middlesex, UK), degassed through sonication for 10 min, and stored in a water bath (Isotemp 3013 HD) at 27 °C. SRHA standard II and SRFA standard I were purchased from the International Humic Substances Society (St Paul, MN). Alginate (A112) was purchased from Sigma-Aldrich, MO. SRHA, SRFA, and alginate solutions were prepared following a procedure described elsewhere.33 2.3. Deposition and Release Study Using QCM-D. Deposition and release of GO onto different surfaces was investigated using QCM-D (E4, Q-Sense, Västra Frölunda, Sweden), following published methods.22,34,35 For all deposition and release experiments, deionized water was flowed across the crystal surface for initial stabilization while monitoring frequency and dissipation signals at the third overtone (f(3) and D(3), respectively). The crystal surfaces were then rinsed with background electrolyte until stabilized. For all experiments, a GO concentration of 1 mg/L was introduced into the chamber for at least 40 min at 0.1 mL/min in parallel flow configuration using a peristaltic pump (Ismatec SA, Switzerland), and then DI water was introduced into the chamber to monitor GO release. Solutions inside the chamber were maintained at 25 ± 0.2 °C. With deposition, the crystal sensors’ overtone frequencies decrease (i.e., become more negative), following the Sauerbrey relationship22,36 Δm = −

C Δfn n

(1)

where Δm is the deposited mass, Δf n is the shift in overtone frequency, n is the overtone number (1, 3, 5, 7, and···), and C is the crystal constant (17.7 ng/Hz·cm2 for the 5 MHz crystal). Deposition can also increase the crystal’s ability to dissipate energy which can be measured simultaneously with the dissipation unit (D) D=

Edissipation 2πEstored

(2)

where Edissipation is the energy dissipated in one oscillation, and Estored is the total energy stored in the oscillator. The Sauerbrey equation is not directly applicable in calculating deposited mass from frequency shift since deposited nanomaterials do not form a homogeneous rigid layer on QCM-D crystals.36,37 Relative deposition behaviors of nanomaterials can be determined from the frequency shifts monitored by QCM-D, however, by calculating deposition rate. Initial deposition rates rf are defined as rates of frequency shift in a time period t, respectively: 9383

dx.doi.org/10.1021/es5020828 | Environ. Sci. Technol. 2014, 48, 9382−9390


Article

Figure 1. A) Frequency (Δf 3) and B) dissipation shifts (ΔD3) at the third overtone during GO deposits on NOM and aluminum oxide surfaces in the presence of 30 mM NaCl. GO concentration was 1 mg/L. pH was unadjusted at 5.5 ± 0.3 and temperature was 25 °C. Different steps during experiment: i: injection of GO; ii: injection of 30 mM NaCl; iii: injection of DI water.

Figure 2. Initial deposition rates (rf) of GO on aluminum oxide surface as a function of A) NaCl and B) CaCl2. rf values were determined from initial slopes of frequency shifts at the third overtone (Δf 3). Error bars indicate standard deviation of three runs.

⎛ dΔf(3) ⎞ ⎟⎟ r f = ⎜⎜ ⎝ dt ⎠ t → 0

mentioned earlier; GO was then deposited on NOM-coated Si wafers. Prior to imaging, the samples were gently dried with nitrogen and mounted on stainless steel discs. Images were taken under ScanAsyst-Air mode with a Silicon Nitride cantilever (ScanAsyst-Air, Bruker AXC Inc., Madison, WI). AFM images were further analyzed for size distribution using Nanoscope Analysis software (Bruker AXC Inc., Madison, WI).

(3)

2.4. NOM and Metal Oxide Surface Preparation. Aluminum oxide (Al2O3, QSX306) surface was cleaned and then used without further modification. A silica-coated crystal sensor (QSX 303, Q-Sense, Västra Frölunda, Sweden) was utilized for deposition and release studies for NOM-coated surfaces. Prior to deposition experiments, the aluminum oxide and silica crystal sensors were cleaned following the protocol described in the QCM-D manual and mentioned elsewhere.22,34 NOM-coated surfaces were prepared following a published protocol.33 First, silica surfaces were coated with cationic poly-L-lysine hydrobromide (PLL, molecular weight 70,000−150,000 by viscosity, P-1274, Sigma-Aldrich, St. Louis, MO).22,33,35,38 Next, 2 mL of HEPES solution was flowed over the PLL layer followed by 2 mL of 1 mM NaCl. Then, 30 mg/L of SRHA, SRFA or alginate prepared in 1 mM NaCl was injected until the frequency shift stabilized. Adsorption of SRHA and SRFA led to a sharp frequency shift of about 10 Hz, while the alginate coating resulted in a 4-Hz frequency shift (Figure S1). After the NOM layers formed over the PLL-coated silica surface, 1 mM NaCl was flowed through QCM-D crystal for at least 20 min to remove any unadsorbed NOM. 2.5. Atomic Force Microscopy Imaging. Images of GOdeposited surfaces were taken using a Veeco Multimode Atomic Force Microscopy (AFM) with a Nanoscope V controller and an E Scanner (Bruker AXC Inc., Madison, WI). GO-deposited samples were prepared by drop casting on Si wafers to have sufficient deposition for AFM imaging. NOM layers were formed on Si wafers following QCM-D protocol

3. RESULTS AND DISCUSSION 3.1. Deposition of GO on Model Environmental Surfaces. Figure 1 shows the frequency (Δf 3) and dissipation shifts (ΔD3) at the third overtone during deposition of GO on different model environmental surfaces in the presence of 30 mM NaCl. Both Δf 3 and ΔD3 changed as GO suspension was injected in the QCM-D module, indicating that GO was depositing on the different surfaces being investigated (Figure 1). The highest changes in Δf 3 and ΔD3 were observed for the NOM surfaces and the aluminum oxide surface exhibited the lowest values. Deposition of GO on NOM and metal oxide surfaces is notably higher than that observed on silica surface in our previous study22 which indicates that NOM and metal oxide surfaces determine the fate of GO in the aquatic environment. Overall changes in Δf 3 and ΔD3 values during GO deposition for different surfaces are alginate > SRFA > SRHA > aluminum oxide > silica. The deposition processes on each surface, as a function of IS and ion valence, are discussed in the following sections. 3.2. Deposition of GO on Metal Oxide Surfaces. The deposition rates of GO on aluminum oxide surface, as a function of NaCl concentration, are shown in Figure 2A. The rf value of GO on aluminum oxide surface in 10 mM NaCl was 0.53 ± 0.11 Hz/min which is similar to deposition rates 9384



Article

Figure 3. Initial deposition rates of GO on NOM surfaces as a function of A) NaCl and B) CaCl2. rf values were determined from initial slopes of frequency shifts at the third overtone (Δf 3). Error bars indicate standard deviation of three runs.

(30 and 50 mM NaCl). This is unexpected, given that the SRHA surface is negatively charged and unfavorable for deposition of GO, while aluminum oxide surface is positively charged and favorable for deposition. These findings are quite different from deposition results for other carbon-based nanomaterials on SRHA surfaces.33,34 In the presence of NaCl, deposition of fullerene nanoparticles on SRHA surfaces resulted in significantly lower deposition than on silica, due to steric repulsion.33 Deposition of multiwalled carbon nanotubes has been observed to be slightly lower on SRHA surface than on silica;34 however, we found GO deposition on SRHA surfaces to be higher. Higher GO deposition on the SRHA surface may be due to the high amount of hydroxyl and carboxyl functional groups on GO which readily bind with functional groups of NOM. Recent studies showed that GO has high interactions with metal and organic based compounds.43−45 Gao et al.44 showed that engineered GO-coated sand can retain several-fold higher concentrations of heavy metal and organic dye than uncoated sand. Another study reported that spongy graphene can absorb petroleum products and organic solvents 10 times higher than conventional absorbents.45 Apul et al.43 found that GO can adsorb synthetic organic compounds better than carbon nanotubes and granular activated carbon. Since most naturally available surfaces are covered with NOM, higher interactions of GO with NOMcoated surfaces will govern the environmental fate of these emerging materials. Deposition rates of GO on SRHA surfaces, as a function of CaCl2 concentration, are presented in Figure 2B. The rf values of GO on SRHA surface in 0.5 and 1.0 mM CaCl2 were 0.23 ± 0.01 and 0.64 ± 0.03 Hz/min, respectively -- more than 15 times greater than rf values observed in deposition of GO on bare silica surface.22 Deposition rates of GO on SRHA surface in the presence of CaCl2 is notably higher than those on aluminum oxide surface, particularly at high IS (1 mM CaCl2). Increased CaCl2 concentration resulted in higher rf values on SRHA surface due to charge-screening, which follows classical colloidal theory.46 Although this is very similar to the trend observed for the monovalent ion (Na+) with respect to bare silica surface, rf values in the presence of Ca2+ were significantly higher (15 times) than in the presence of monovalent ions (3− 4 times). Chen and Elimelech33 observed a slightly higher deposition of fullerenes on SRHA surface than on bare silica in the presence of CaCl2 due to SRHA macromolecules undergoing complex formation with Ca2+ which reduces electrostatic and steric effects. 3.3.2. Deposition of GO on SRFA Surface. Figure 3A presents the deposition rates of GO on SRFA surfaces as a

observed on the positively charged PLL surface in our previous study.22 Prior studies34,39 showed that deposition of multiwalled carbon nanotubes and humic acid on aluminum oxide surface is significantly high and irreversible and quite similar to deposition on positively charged surfaces and that the zero point of charge of aluminum oxide is between pH 6 and 10.40,41 Both aluminum oxide and the PLL provide positively charged surfaces favorable for GO deposition. Deposition rates of GO on aluminum oxide surfaces remained quite constant with increased IS from 10 to 50 mM NaCl because deposition of GO on the aluminum oxide surface is favorable and further charge-screening due to increased IS cannot increase deposition. Deposition rates of GO on aluminum oxide surface started to decrease, however, with further increase of IS above 50 mM NaCl due to aggregation of GO which reduces its diffusion rate to the deposition surface.22,23 Similar GO deposition rates, as a function of NaCl concentration on PLL surface, were also observed in our previous study.22 Deposition rates of GO on the aluminum oxide surface, as a function of CaCl2 concentration, are shown in Figure 2B. The rf value of GO on the aluminum oxide surface in the presence of 0.5 mM CaCl2 is 0.40 ± 0.09 Hz/min, with a slight decrease in deposition rate at the 1.0 mM CaCl2 concentration due to aggregation. Overall, deposition behavior of GO on the aluminum oxide surface is very similar to that for positively charged PLL surfaces which indicates that interactions of GO with metal oxide surfaces in the environment will be governed primarily by favorable electrostatic conditions. 3.3. Deposition of GO on NOM Surfaces. 3.3.1. Deposition of GO on SRHA Surface. Initial deposition rates (rf) of GO on SRHA surfaces, as a function of NaCl concentration, appear in Figure 3A. The rf of GO in 10 mM NaCl is 0.29 ± 0.06 Hz/min, although our previous study22 found negligible GO deposition on silica surface at this IS. It has been shown that NOM is negatively charged under environmentally relevant conditions.42 Hence, in our study, interactions between GO and NOM-coated surfaces are electrostatically unfavorable. Increased IS results in higher rf values until 30 mM NaCl due to the charge-screening at high IS, following DLVO theory.26 Above 30 mM NaCl, however, rf values of GO decrease with IS due to aggregate formation23 at high IS which reduces diffusion rates of GO to the deposition surface.22 The rf values of GO in 30 and 50 mM NaCl are 1.42 ± 0.19 and 1.30 ± 0.05 Hz/min, which are four and three times higher, respectively, than rf for the silica surface.22 Overall, deposition of GO on SRHA surface is significantly higher than deposition on silica surface.22 The rf values of GO on SRHA surface are notably higher than on aluminum oxide surface, particularly in intermediate IS 9385



Article

Table 1. Summary of Release of Deposited GO from Various Environmental Surfaces upon Injection of DI Water SRHA-coated surface condition NaCl

CaCl2 a

ionic strength (mM) 10 30 50 80 0.5 1.0

release rate (min−1) 0.40 0.27 0.29 0.12 0.14 0.06

± ± ± ± ± ±

0.02 0.04 0.01 0.02 0.02 0.01

SRFA-coated surface

fraction release (%) 89.2 35.0 28.3 26.6 34.1 18.6

± ± ± ± ± ±

9.9 4.5 1.3 3.5 4.8 0.7

release rate (min−1) 0.37 0.24 0.14 0.10 0.09 0.04

± ± ± ± ± ±

0.04 0.03 0.01 0.02 0.01 0.01

alginate-coated surface

fraction release (%) 58.1 16.8 23.3 25.9 18.3 10.8

± ± ± ± ± ±

11.7 2.1 4.5 2.9 0.7 0.7

aluminum oxide surface

release rate (min−1)

fraction release (%)

release rate (min−1)

fraction release (%)

0.18 ± 0.08 ± 0.02 ± 0.02 ± NRa NRa

57.1 ± 33.2 ± 19.4 ± 13.5 ± NRa NRa

0.20 ± 0.15 ± 0.08 ± NRa 0.04 ± NRa

47.8 ± 48.9 ± 39.6 ± NRa 25.3 ± NRa

0.02 0.01 0.01 0.01

10.6 3.1 1.5 2.3

0.01 0.01 0.01 0.01

1.9 0.1 4.4 2.9

NR: negligible release.

3.4. Release of GO from Model Environmental Surfaces. 3.4.1. Release of GO from Metal Oxide Surface. Release rates and fractional release of deposited GO from environmental surfaces after introduction of DI water are summarized in Table 1. For the aluminum oxide surface, GO deposited in the presence of NaCl was released after introduction of DI water, indicating that its deposition on aluminum oxide surfaces is partially reversible. In our previous study,22 deposition of GO on positively charged PLL surfaces was found to be reversible, whereas multiwalled carbon nanotubes deposited on aluminum oxide surface were found to be irreversibly attached.34 Both the release rates and fractional release of deposited GO from the aluminum oxide surface decreased with NaCl concentration, indicating that the release process of GO from aluminum oxide surface is electrostatically controlled. Increase of NaCl concentration from 10 mM to 80 mM reduced release of GO from aluminum oxide surfaces, and a negligible amount of GO deposited in the presence of 80 mM NaCl was released. Release of GO deposited on the aluminum oxide surface in the presence of CaCl2 was notably lower than that of GO deposited in the presence of NaCl. About 25% of deposited GO was released in the presence of 0.5 mM CaCl2, while almost no GO was released in the presence of 1.0 mM CaCl2. A similar trend was observed on PLL surface in the presence of CaCl222 which is attributed to the binding capacity of GO functional groups with Ca2+ ions.22,24 3.4.2. Release of GO from NOM-Coated Surfaces. Significant release of deposited GO from SRHA surface was observed upon introduction of DI water under all conditions (Table 1); however, fractional release of GO decreased from about 90% to 30% as IS during deposition of GO increased from 10 mM to 50 mM NaCl. This is due to greater chargescreening at high IS which reduces the repulsive forces of GO with SRHA surfaces.46 A similar trend for the deposition of GO on silica surfaces was observed in our previous study.22 Release rates of GO from SRHA surface decreased from around 0.4 min−1 to 0.1 min−1 with IS, due to charge-screening. GO deposited on SRHA surface in the presence CaCl2 showed much lower (∼10−30%) release upon injection of DI water which is similar to the trend on metal oxide surface. Increased CaCl2 concentration reduced the release of GO from SRHA surface. On SRFA surface, release of GO was observed with introduction of DI water under all conditions. Release of GO from SRFA surface was notably lower than that for SRHA surface, however, which may be due to macromolecular differences between SRHA and SRFA. Since SRFA has more carboxyl functional groups than SRHA, higher deposition of GO on SRFA was observed, as mentioned in section 3.3.2. This

function of NaCl concentration. Like SRHA, the rf values of GO on SRFA surface were significantly higher than on silica. rf values of GO on SRFA surface also increased with IS, as expected from colloidal theory;46 however, rf values of GO on SRFA surfaces are generally higher than on SRHA surfaces particularly at intermediate IS (30 mM to 50 mM NaCl). A similar trend was observed in the presence of CaCl2, particularly at 0.5 mM (Figure 3B). This may be due to different macromolecular structures of SRHA and SRFA. According to the manufacturer (International Humic Substances Society, MN), SRHA used in this study contains 9.13 mequiv/g C carboxyl and 3.72 mequiv/g C phenolic functional groups. SRFA, on the other hand, has a higher concentration of carboxyl functional groups than phenolic (11.44 mequiv/g C carboxyl and 2.91 mequiv/g C phenolic). Moreover, GO has carboxyl functional groups on the edge and phenolic functional groups on the basal plane. The higher carboxyl functional group concentration on SRFA may be responsible for increased interaction with functional groups on GO. 3.3.3. Deposition of GO on Alginate Surface. Deposition rates of GO on alginate surface in the presence of NaCl are presented in Figure 3A. rf values of GO on alginate surface in 10 mM NaCl are significantly higher than on SRFA and SRHA surfaces. Moreover, deposition rates of GO on alginate surfaces decreased with IS, which is opposite to the trend observed for SRHA and SRFA surfaces. Deposition rates of GO on alginate surface as a function of CaCl2 concentration are shown in Figure 3B. rf values of GO on alginate surface in the presence of CaCl2 are higher than on both SRHA and SRFA surfaces at 0.5 mM CaCl2 but lower at 1 mM CaCl2. Deposition rates of GO on alginate surfaces decreased with CaCl2 concentration, however, which is similar to the trend observed for NaCl treatments. This may be due to macromolecular differences between alginate and SRHA or SRFA: the alginate surface is likely rougher than SRHA or SRFA surfaces because alginate macromolecules are polysaccharides and larger (12−80 kDa) than humic or fulvic acids (1−5 kDa).33,47,48 Specifically at low IS, alginate can take a more extended conformation from the PLL surface which can make the surface rougher and allow GO nanomaterials to be trapped within the alginate layer. Chen and Elimelech33 observed similar high deposition rates of fullerene nanoparticles on alginate surfaces, and Tong et al.49 found greater deposition of extracellular polymeric substances on an alginate layer. AFM images of NOM-coated surfaces (Figure S2) revealed that Rq (root mean squared of vertical deviations of roughness profile) and Ra (arithmetic average of vertical deviations of roughness profile) values are the highest for alginate-coated surface followed by SRHA and SRFA-coated surfaces which confirms the rougher surface of the alginatecoated layer. 9386



Article

Figure 4. Initial |ΔD(3)/Δf(3)| values of deposited GO layer on different environmental surfaces as a function of A) NaCl and B) CaCl2 concentrations.

significantly higher deposition of GO was observed on all NOM surfaces investigated here with respect to bare silica surface. Moreover, deposition rates of GO on NOM surfaces are notably higher (∼1.5−3 times) than on aluminum oxide surface even though aluminum oxide surface is positively charged due to Al3+ and presents a favorable surface for GO deposition. Thus, theoretical deposition rates of GO should be greatest on Al2O3 surface. During deposition, however, GO can deposit in multiple orientations such as on the basal plane or by edge attachment24,53 and mass deposition rates vary for different orientations. When GO deposits on a crystal surface in the flat orientation, it can cover more crystal surface area, resulting in lower mass deposition rates per unit surface area. The edge orientation, on the other hand, results in higher mass deposition rates per unit surface area. Thermodynamically, GO flat-oriented deposition is favored over edge deposition unless there is specific interaction between edges of GO and deposition surfaces. Various functional groups exist on NOM surfaces, however, which can interact with functional groups of GO -- in particular, the edge of GO contains carboxyl functional groups which can exhibit specific interactions with functional groups on NOM surfaces. GO can therefore deposit in multiple orientations on NOM surfaces which can increase its deposition rates. Also, NOM surfaces are rough due to extended conformations, which can trap GO in the NOM layer. As discussed ealier, macromolecular differences between SRHA, SRFA, and alginate significantly influence interactions between GO and NOM surfaces. Finally, QCM-D is an acoustic method, which can sense acoustically coupled water in addition to deposited materials, so it is possible that water trapped between GO flakes can cause frequency and dissipation shifts.53 A previous study53 showed that GO deposition on lipid membranes resulted ∼30% of mass deposited due to trapped water. NOM contains −COOH, −OH, and −NH2 functional groups,54−58 and the primary functional groups of GO are epoxy, hydroxyl, and carboxyl.8,59 Epoxy and hydroxyl groups are on the basal plane of GO, whereas carboxyl groups are in the edges. NOM molecules have been reported to adsorb on GO via hydrogen bonds, Lewis acid−base, and π-π interactions.55 Hydroxyl functional groups of NOM can form hydrogen bonds with oxygen-containing functional groups of GO.9,55 Lewis acid−base interactions can occur between NOM and GO due to proton complexation of the π electron system of graphene.55 Withdrawal of electron density by the higher electronegative oxygen-containing functional groups can create Lewis acid centers, which can interact with Lewis bases, particularly hydroxyl groups due to the lone electron pair of the oxygen atoms.60 π-π interactions between the aromatic rings of

interaction of GO with carboxyl functional groups might reduce the release of GO from SRFA surface. Increased IS from 10 mM to 50 mM reduced the release from about 60% to 20% of GO from SRFA surfaces. These release rates are very similar to those from SRHA surfaces. GO deposited on SRFA surface in the presence of CaCl2 showed a low release similar to SRHA surface; however, it was significantly lower from SRFA than SRHA. This is likely due to the higher amount of carboxyl functional groups on SRFA which results in stronger binding of GO with SRFA surface to reduce the release. Under all conditions investigated in this study, significant amounts of GO were released from alginate surfaces. The fractional release of deposited GO in the presence of NaCl from alginate surface is slightly lower than from SRFA and SRHA surfaces; however, release rates from alginate surfaces are much lower than from SRHA or SRFA surfaces. Moreover, there was no measurable release of GO observed in the presence of CaCl2. Again, this may be due to macromolecular differences between alginate and SRHA or SRFA. As mentioned in Section 3.3.3, alginate surface is rougher than SRHA or SRFA surfaces due to its extended linear conformation which can allow entrapment of GO within alginate layers. The additional physical trapping may reduce the release rates of GO from alginate surface. 3.5. Properties of Deposited GO Layer on Environmental Surfaces. Characteristics of layers deposited on QCM-D crystal sensors can be determined from the ratio of dissipation to frequency shifts.22,34,50−52 High |ΔD/Δf | values indicate a dissipative and soft deposited layer, while low |ΔD/ Δf | values are observed for rigid layers. |ΔD(3)/Δf(3)| at the third overtone, as a function of salt concentration on four different environmental surfaces, is presented in Figure 4. |ΔD(3)/Δf(3)| values on SRHA and SRFA surfaces are significantly higher (∼1.5−3 times) than for the Al2O3 surface, indicating that deposited GO layers on SRHA and SRFA surfaces are more dissipative and softer than on Al2O3 surface. This may be due to multilayer formation and deposition of GO in multiple orientations on SRHA and SRFA surfaces. We also observed significantly higher deposition rates on SRHA and SRFA surfaces than Al2O3 surface. Multilayer formation and deposition of GO in different orientations on SRHA and SRFA surfaces can allow more GO deposition per unit surface area of the QCM-D crystal, which can result higher deposition rates. |ΔD(3)/Δf(3)| values on alginate surface are significantly lower than SRHA and SRFA surfaces, however, indicating the deposited alginate surface is different from the other two NOM surfaces. 3.6. Mechanisms Involved in Deposition and Release of GO on Model Environmental Surfaces. Overall, 9387



Article

Figure 5. AFM images of NOM-coated and aluminum oxide surfaces after deposition of GO in the presence of 30 mM NaCl: deposited GO on A) aluminum oxide surface, B) SRHA surface, C) SRFA surface, and D) alginate surface.

NOM and GO can play a notable role in deposition.9,55,61 Hartono et al.55 proposed a detailed adsorption mechanism of GO where −COOH, −OH, and −NH2 functional groups of NOM can bind with the epoxy, hydroxyl, and carboxyl groups of GO via hydrogen bonding. AFM images of different environmental surfaces investigated in this study after deposition of GO in 30 mM NaCl, appear in Figure 5. AFM images showed that GO deposited on aluminum oxide surface primarily in the flat orientation which agrees with our hypothesis. Frost et al.53 also report a similar deposition process of GO on positively charged lipid membranes using QCM-D. AFM imaging of deposited GO on NOM surfaces showed a clear distinction compared to GO deposited on the aluminum oxide surface in that most deposited in multiple orientations and formed multilayer flakes, which agrees with our hypothesis. Distribution of height of the deposited GO layer on different surfaces is presented in Figure S3. On the Al2O3 surface, more than 70% of GO layers are less than 2.5 nm thick, while all GO layers on SRHA and SRFA-coated surfaces have thicknesses greater than 2.5 nm, and more than 30% of GO layers exceeded 5 nm thickness for SRHA and SRFA surfaces. This indicates deposited GO layers on SRHA and SRFA surfaces could be thicker due to possible multilayer formation and deposition in different orientations. Multiple orientation deposition and multilayer formation resulted in significantly higher mass deposition of GO on NOM surfaces than aluminum oxide surfaces.

that metal oxide patches in the soil or sediment will have a notable role in the fate of these emerging materials in the environment. The concentrations of monovalent and divalent ions in the natural environment are usually within 10 mM and 1 mM, respectively.30 Under these conditions, deposition rates of GO were generally higher on NOM surfaces than on the metal oxide surface investigated here, indicating that most of GO will be deposited on NOM-coated surfaces in the aquatic environment. However, deposition of GO on both NOM and metal oxide surfaces is found to be highly reversible which indicates release and remobilization of deposited GO from these surfaces in the natural aquatic environment is quite possible.

■

ASSOCIATED CONTENT

* Supporting Information S

Additional figure regarding height distributions determined from AFM images of deposited GO, representative frequency shifts during formation of the NOM-coated layer in QCM-D, and AFM images of NOM-coated layers. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 706-355-8333. Fax: 706-355-8026. E-mail: Bouchard. [email protected]. Notes

The authors declare no competing financial interest.

4. ENVIRONMENTAL IMPLICATIONS The data presented here indicates that GO nanomaterials have significantly higher interactions with NOM surfaces than silica. Since most naturally available surfaces in the aquatic environment are partially- or fully coated with NOM, our work indicates that NOM-coated surfaces will be a major factor controlling GO’s fate in the environment. Significant GO attachment was also observed with aluminum oxide indicating

■

ACKNOWLEDGMENTS Funding was provided for Indranil Chowdhury by a National Research Council (NRC) Research Associateship Award at EPA. Additional funding was provided by the University of California Center for the Environmental Implications of Nanotechnology (NSF-EPA under Cooperative Agreement # 9388



Article

(21) Akhavan, O.; Abdolahad, M.; Esfandiar, A.; Mohatashamifar, M. Photodegradation of Graphene Oxide Sheets by TiO2 Nanoparticles after a Photocatalytic Reduction. J. Phys. Chem. C 2010, 114 (30), 12955−12959. (22) Chowdhury, I.; Duch, M. C.; Mansukhani, N. D.; Hersam, M. C.; Bouchard, D. Deposition and Release of Graphene Oxide Nanomaterials Using a Quartz Crystal Microbalance. Environ. Sci. Technol. 2013, 48 (2), 961−969. (23) Chowdhury, I.; Duch, M. C.; Mansukhani, N. D.; Hersam, M. C.; Bouchard, D. Colloidal Properties and Stability of Graphene Oxide Nanomaterials in the Aquatic Environment. Environ. Sci. Technol. 2013, 47 (12), 6288−6296. (24) Wu, L.; Liu, L.; Gao, B.; Muñoz-Carpena, R.; Zhang, M.; Chen, H.; Zhou, Z.; Wang, H. Aggregation Kinetics of Graphene Oxides in Aqueous Solutions: Experiments, Mechanisms, and Modeling. Langmuir 2013, 29 (49), 15174−15181. (25) Feriancikova, L.; Xu, S. Deposition and remobilization of graphene oxide within saturated sand packs. J. Hazard. Mater. 2012, 235−236 (0), 194−200. (26) Overbeek, T. DLVO theory - Milestone of 20th century colloid science - Preface. Adv. Colloid Interface Sci. 1999, 83 (1−3), IX−XI. (27) Gregory, J. Particles in water: properties and processes; IWA Pub.; Taylor & Francis: London; Boca Raton, FL, 2006. (28) 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 (9), 4255−4261. (29) Liu, L.; Gao, B.; Wu, L.; Morales, V. L.; Yang, L.; Zhou, Z.; Wang, H. Deposition and transport of graphene oxide in saturated and unsaturated porous media. Chem. Eng. J. 2013, 229 (0), 444−449. (30) Crittenden, J. C.; Montgomery Watson, H. Water treatment principles and design; J. Wiley: Hoboken, NJ, 2005. (31) Wilkinson, K. J.; Balnois, E.; Leppard, G. G.; Buffle, J. Characteristic features of the major components of freshwater colloidal organic matter revealed by transmission electron and atomic force microscopy. Colloids Surf., A 1999, 155 (2−3), 287−310. (32) Walker, H. W.; Bob, M. M. Stability of particle flocs upon addition of natural organic matter under quiescent conditions. Water Res. 2001, 35 (4), 875−882. (33) Chen, K. L.; Elimelech, M. Interaction of Fullerene (C60) Nanoparticles with Humic Acid and Alginate Coated Silica Surfaces: Measurements, Mechanisms, and Environmental Implications. Environ. Sci. Technol. 2008, 42 (20), 7607−7614. (34) Chang, X.; Bouchard, D. C. Multiwalled Carbon Nanotube Deposition on Model Environmental Surfaces. Environ. Sci. Technol. 2013, 47 (18), 10372−10380. (35) Yi, P.; Chen, K. L. Influence of Surface Oxidation on the Aggregation and Deposition Kinetics of Multiwalled Carbon Nanotubes in Monovalent and Divalent Electrolytes. Langmuir 2011, 27 (7), 3588−3599. (36) Sauerbrey, G. Verwendung von Schwingquarzen zur Wagung dunner schichten und zur mikrowagung. Z. Phys. 1959, 155 (2), 206− 222. (37) Reviakine, I.; Johannsmann, D.; Richter, R. P. Hearing What You Cannot See and Visualizing What You Hear: Interpreting Quartz Crystal Microbalance Data from Solvated Interfaces. Anal. Chem. 2011, 83 (23), 8838−8848. (38) de Kerchove, A. J.; Elimelech, M. Formation of Polysaccharide Gel Layers in the Presence of Ca2+ and K+ Ions: Measurements and Mechanisms. Biomacromolecules 2006, 8 (1), 113−121. (39) Eita, M. In situ study of the adsorption of humic acid on the surface of aluminium oxide by QCM-D reveals novel features. Soft Matter 2011, 7 (2), 709−715. (40) Kosmulski, M. pH-dependent surface charging and points of zero charge. IV. Update and new approach. J. Colloid Sci. 2009, 337 (2), 439−448. (41) Yopps, J. A.; Fuerstenau, D. W. The zero point of charge of alpha-alumina. J. Colloid Sci. 1964, 19 (1), 61−71. (42) Buffle, J.; Wilkinson, K. J.; Stoll, S.; Filella, M.; Zhang, J. A Generalized Description of Aquatic Colloidal Interactions: The Three-

DBI-1266377). This paper has been reviewed in accordance with the USEPA’s peer and administrative review policies and approved for publication. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.

■

REFERENCES

(1) Geim, A. K.; Novoselov, K. S. The rise of graphene. Nat. Mater. 2007, 6 (3), 183−191. (2) Geim, A. K. Graphene: Status and Prospects. Science 2009, 324 (5934), 1530−1534. (3) Compton, O. C.; Nguyen, S. T. Graphene Oxide, Highly Reduced Graphene Oxide, and Graphene: Versatile Building Blocks for Carbon-Based Materials. Small 2010, 6 (6), 711−723. (4) Joshi, R. K.; Carbone, P.; Wang, F. C.; Kravets, V. G.; Su, Y.; Grigorieva, I. V.; Wu, H. A.; Geim, A. K.; Nair, R. R. Precise and Ultrafast Molecular Sieving Through Graphene Oxide Membranes. Science 2014, 343 (6172), 752−754. (5) Stankovich, S.; Dikin, D. A.; Dommett, G. H. B.; Kohlhaas, K. M.; Zimney, E. J.; Stach, E. A.; Piner, R. D.; Nguyen, S. T.; Ruoff, R. S. Graphene-based composite materials. Nature 2006, 442 (7100), 282− 286. (6) Segal, M. Selling graphene by the ton. Nat. Nanotechnol. 2009, 4 (10), 612−614. (7) Brownson, D. An overview of graphene in energy production and storage applications. J. Power Sources 2011, 196 (11), 4873−4885. (8) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. The chemistry of graphene oxide. Chem. Soc. Rev. 2010, 39 (1), 228−240. (9) Chen, D.; Feng, H.; Li, J. Graphene Oxide: Preparation, Functionalization, and Electrochemical Applications. Chem. Rev. 2012, 112 (11), 6027−6053. (10) Duch, M. C.; Budinger, G. R. S.; Liang, Y. T.; Soberanes, S.; Urich, D.; Chiarella, S. E.; Campochiaro, L. A.; Gonzalez, A.; Chandel, N. S.; Hersam, M. C.; Mutlu, G. M. Minimizing Oxidation and Stable Nanoscale Dispersion Improves the Biocompatibility of Graphene in the Lung. Nano Lett. 2011, 11 (12), 5201−5207. (11) Liu, S.; Zeng, T. H.; Hofmann, M.; Burcombe, E.; Wei, J.; Jiang, R.; Kong, J.; Chen, Y. Antibacterial Activity of Graphite, Graphite Oxide, Graphene Oxide, and Reduced Graphene Oxide: Membrane and Oxidative Stress. ACS Nano 2011, 5 (9), 6971−6980. (12) Akhavan, O.; Ghaderi, E.; Emamy, H.; Akhavan, F. Genotoxicity of graphene nanoribbons in human mesenchymal stem cells. Carbon 2013, 54 (0), 419−431. (13) Hu, X.; Zhou, Q. Health and Ecosystem Risks of Graphene. Chem. Rev. 2013, 113 (5), 3815−3835. (14) Arvidsson, R.; Kushnir, D.; Sandén, B. A.; Molander, S. Prospective Life Cycle Assessment of Graphene Production by Ultrasonication and Chemical Reduction. Environ. Sci. Technol. 2014, 48 (8), 4529−4536. (15) Akhavan, O.; Ghaderi, E.; Akhavan, A. Size-dependent genotoxicity of graphene nanoplatelets in human stem cells. Biomaterials 2012, 33 (32), 8017−8025. (16) Akhavan, O.; Ghaderi, E. Toxicity of Graphene and Graphene Oxide Nanowalls Against Bacteria. ACS Nano 2010, 4 (10), 5731− 5736. (17) Williams, G.; Seger, B.; Kamat, P. V. TiO2-Graphene Nanocomposites. UV-Assisted Photocatalytic Reduction of Graphene Oxide. ACS Nano 2008, 2 (7), 1487−1491. (18) Williams, G.; Kamat, P. V. Graphene−Semiconductor Nanocomposites: Excited-State Interactions between ZnO Nanoparticles and Graphene Oxide. Langmuir 2009, 25 (24), 13869−13873. (19) Akhavan, O.; Choobtashani, M.; Ghaderi, E. Protein Degradation and RNA Efflux of Viruses Photocatalyzed by Graphene−Tungsten Oxide Composite Under Visible Light Irradiation. J. Phys. Chem. C 2012, 116 (17), 9653−9659. (20) Akhavan, O. Graphene Nanomesh by ZnO Nanorod Photocatalysts. ACS Nano 2010, 4 (7), 4174−4180. 9389



Article

colloidal Component Approach. Environ. Sci. Technol. 1998, 32 (19), 2887−2899. (43) Apul, O. G.; Wang, Q.; Zhou, Y.; Karanfil, T. Adsorption of aromatic organic contaminants by graphene nanosheets: Comparison with carbon nanotubes and activated carbon. Water Res. 2013, 47 (4), 1648−1654. (44) Gao, W.; Majumder, M.; Alemany, L. B.; Narayanan, T. N.; Ibarra, M. A.; Pradhan, B. K.; Ajayan, P. M. Engineered Graphite Oxide Materials for Application in Water Purification. ACS Appl. Mater. Interfaces 2011, 3 (6), 1821−1826. (45) Bi, H.; Xie, X.; Yin, K.; Zhou, Y.; Wan, S.; He, L.; Xu, F.; Banhart, F.; Sun, L.; Ruoff, R. S. Spongy Graphene as a Highly Efficient and Recyclable Sorbent for Oils and Organic Solvents. Adv. Funct. Mater. 2012, 22 (21), 4421−4425. (46) Elimelech, M.; Gregory, J.; Jia, X.; Williams, R. A. Particle Deposition and Aggregation: Measurement, Modeling and Simulation; Butterworth-Heinemann: 1995; p 441. (47) Chen, K. L.; Mylon, S. E.; Elimelech, M. Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environ. Sci. Technol. 2006, 40 (5), 1516−1523. (48) Chen, K. L.; Elimelech, M. Influence of humic acid on the aggregation kinetics of fullerene (C60) nanoparticles in monovalent and divalent electrolyte solutions. J. Colloid Interface Sci. 2007, 309 (1), 126−134. (49) Tong, M.; Zhu, P.; Jiang, X.; Kim, H. Influence of natural organic matter on the deposition kinetics of extracellular polymeric substances (EPS) on silica. Colloids Surf., B 2011, 87 (1), 151−158. (50) Rodahl, M.; Hook, F.; Fredriksson, C.; Keller, C. A.; Krozer, A.; Brzezinski, P.; Voinova, M.; Kasemo, B. Simultaneous frequency and dissipation factor QCM measurements of biomolecular adsorption and cell adhesion. Faraday Discuss. 1997, 107, 229−246. (51) Feiler, A. A.; Sahlholm, A.; Sandberg, T.; Caldwell, K. D. Adsorption and viscoelastic properties of fractionated mucin (BSM) and bovine serum albumin (BSA) studied with quartz crystal microbalance (QCM-D). J. Colloid Interface Sci. 2007, 315 (2), 475−481. (52) Steinmetz, N. F.; Bock, E.; Richter, R. P.; Spatz, J. P.; Lomonossoff, G. P.; Evans, D. J. Assembly of Multilayer Arrays of Viral Nanoparticles via Biospecific Recognition: A Quartz Crystal Microbalance with Dissipation Monitoring Study. Biomacromolecules 2008, 9 (2), 456−462. (53) Frost, R.; Jönsson, G. E.; Chakarov, D.; Svedhem, S.; Kasemo, B. Graphene Oxide and Lipid Membranes: Interactions and Nanocomposite Structures. Nano Lett. 2012, 12 (7), 3356−3362. (54) Sutton, R.; Sposito, G. Molecular Structure in Soil Humic Substances: The New View. Environ. Sci. Technol. 2005, 39 (23), 9009−9015. (55) 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 (1), 114−119. (56) Schulten, H. R.; Schnitzer, M. A state of the art structural concept for humic substances. Naturwissenschaften 1993, 80 (1), 29− 30. (57) Leenheer, J. A. Progression from Model Structures to Molecular Structures of Natural Organic Matter Components. Ann. Environ. Sci. 2007, Vol. 1, Article 15. (58) Baalousha, M.; Motelica-Heino, M.; Coustumer, P. L. Conformation and size of humic substances: Effects of major cation concentration and type, pH, salinity, and residence time. Colloids Surf., A 2006, 272 (1−2), 48−55. (59) Gao, W.; Alemany, L.; Ci, L.; Ajayan, P. New insights into the structure and reduction of graphite oxide. Nat. Chem. 2009, 1 (5), 403−408. (60) Gonçalves, M.; Sánchez-García, L.; Oliveira Jardim, E. d.; Silvestre-Albero, J.; Rodríguez-Reinoso, F. Ammonia Removal Using Activated Carbons: Effect of the Surface Chemistry in Dry and Moist Conditions. Environ. Sci. Technol. 2011, 45 (24), 10605−10610. (61) 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 (24), 10454−10462.

9390


Interactions of Graphene Oxide Nanomaterials with ... - ACS Publications

Recommend Documents