Environmental Toxicology and Chemistry, Vol. 33, No. 5, pp. 1023–1029, 2014 # 2014 SETAC Printed in the USA
STABILITY AND AGGREGATION OF SILVER AND TITANIUM DIOXIDE NANOPARTICLES IN SEAWATER: ROLE OF SALINITY AND DISSOLVED ORGANIC CARBON HUANHUA WANG,yz ROBERT M. BURGESS,*x MARK G. CANTWELL,x LISA M. PORTIS,x MONIQUE M. PERRON,k FENGCHANG WU,z and KAY T. HOx yNational Research Council, US Environmental Protection Agency, Office of Research and Development, Narragansett, Rhode Island zState Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing, China xUS Environmental Protection Agency, Office of Research and Development, Narragansett, Rhode Island kUS Environmental Protection Agency, Office of Pesticide Programs, Washington, DC (Submitted 6 December 2013; Returned for Revision 5 January 2014; Accepted 18 January 2014) Abstract: The behavior and fate of nanoparticles (NPs) in the marine environment are largely unknown and potentially have important environmental and human health implications. The aggregation and fate of NPs in the marine environment are greatly influenced by their interactions with seawater and dissolved organic carbon (DOC). In the present study, the stability and aggregation of 30-nm–diameter silver nanoparticles (AgNPs) capped with citrate and polyvinylpyrrolidone (PVP; AgNP-citrate and AgNP-PVP) and 21-nm-diameter titanium dioxide (TiO2) NPs as affected by seawater salinity and DOC were investigated by measuring hydrodynamic diameters and zeta potentials. The added DOC (in humic acid form) stabilized the 3 types of NPs when the seawater salinities were 5 parts per thousand (ppt), but the stabilizing effect of DOC was reduced by a higher salinity (e.g., 30 ppt). In addition, AgNP-PVP was more stable than AgNPcitrate in seawater, indicating that surface capping agents and stabilization mechanisms govern the stability and aggregation of NPs. Statistical analysis showed that salinity is the most dominant influence on the stability and aggregation of AgNPs and TiO2NPs, followed by DOC. These findings expand our knowledge on the behavior of AgNPs and TiO2NPs in seawater and indicate that the fate of these NPs will be primarily to aggregate in the water column, precipitate, and accumulate in sediments following release into the marine environment. Environ Toxicol Chem 2014;33:1023–1029. # 2014 SETAC Keywords: Stability
Silver
Titanium dioxide
Nanoparticle
Seawater
to maintain colloidal stability in solution [7]. Given these uncertainties regarding metal nanomaterials, the objective of the present study was to investigate the effects of seawater salinity (i.e., ionic strength) and dissolved organic carbon (DOC) on the aqueous stability and aggregation of AgNPs and TiO2NPs. It has been shown that increasing the ionic strength of a solution will result in the destabilization and aggregation of NPs. For example, Domingos et al. [8] added mono- and divalent electrolytes (e.g., NaCl and CaCl2) to aqueous solutions and found that the presence of calcium resulted in aggregation of TiO2NPs despite the presence of dissolved organic matter (DOM; i.e., Suwannee River fulvic acid). However, these relatively simple electrolyte solutions were not representative of an environmentally complex medium such as seawater. Also, there is a paucity of published studies concerning the effects of salinity on the stability and aggregation of NPs over time (e.g., 10 d). The presence of DOM can influence the aggregation of NPs, and thereby their interactions with natural interfaces and organisms [9,10], and can alter the bioavailability and toxicity of NPs in solution [11]. In addition, metals associated with DOM have shown limited bioavailability to aquatic organisms, compared with the freely dissolved form of the metals [12]. For example, the aggregation rate of single-walled carbon nanotubes was slowed in the presence of Suwannee River humic acid, a form of DOM, possibly because of steric repulsion originating from the adsorbed macromolecular layer [9]. This type of enhanced colloidal stability is likely to lead to increased residence times for NPs in the overlying seawater and, consequently, to potentially elevated exposures to aquatic organisms. To date, no studies have focused on the role of
INTRODUCTION
Nanotechnology has taken the leap from the realm of research to integration into a growing number of applications in industrial and consumer products [1]. There are numerous nanoparticles (NPs) currently in production and use, including 2 important metal nanomaterials: silver nanoparticles (AgNPs) and titanium dioxide nanoparticles (TiO2NPs). Silver nanoparticles are used extensively in consumer goods, including textiles, cosmetics, and health-care products to utilize their strong antimicrobial activity [2] and in electronic components [3]. Titanium dioxide nanoparticles have a range of applications in consumer products, including sunscreens, cosmetics, paints, and photocatalysts, because of their unique optical properties [4]. Marine, estuarine, and coastal environments are the expected destination of many NPs, through industrial discharges and municipal effluents. However, there are very few studies on the stability and aggregation of NPs in seawater. Angel et al. [5] investigated the behavior of AgNPs over 72 h in seawater, and Ates et al. [6] examined the fate of TiO2NPs over 24 h. Despite these very recent additions to the literature, the behavior and fate of NPs in the marine environment are still not well understood. The high surface area to volume ratio of NPs results in high surface reactivity, which leads to particle aggregation and settling when in aqueous solution. For some nanomaterials, such as AgNPs, capping agents have been applied to the outer surface in an effort All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected]. Published online 27 January 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2529 1023
http://ehp.niehs.nih.gov Anatase (86%), rutile (14%) 3.5–4.5 21
a Based on transmission electron microscopy. AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle; na ¼ not available.
50 15 Spheroid
209 8
34.4 1.6
nanocomposix.com PVP surface 6.4 26.7 na 36.2 Spherical 28.7 2.1a
a
nanocomposix.com Citrate surface 7.7 29.0 3.1
a
Spherical
39.6
na
41.8
Composition or surface pH Zeta potential (mV) Surface area (m2 g1) Hydrodynamic diameter (nm)
nanoComposix (San Diego, CA, USA) nanoComposix (San Diego, CA, USA) Degussa/Evonik (Parsippany, NJ, USA)
Reconstituted seawater of varying salinities was prepared by diluting a filtered brine solution with ultrapure deionized water. Brine was produced by evaporating Narragansett Bay (RI, USA) seawater to a salinity of 100 parts per thousand (ppt; i.e., grams per liter). The natural seawater used to prepare the brine was filtered through a flow-through string filter with an effective pore size of 10 mm (Purolator Advanced Filtration). The brine was prepared and then stored at 20 8C in the dark to limit biological growth until use. Any brine showing evidence of growth (i.e., discoloration) was not
AgNP-citrate (BioPure) AgNP-PVP (BioPure) TiO2NP
Preparation of saline and DOC solutions
Shape
Nanoparticle suspensions of 30-nm–diameter Biopure citrate- and PVP-coated AgNPs were purchased from Nanocomposix at a concentration of 1000 mg L1. The specific characteristics of the materials can be found at www.nanocomposix.com (Table 1), and transmission electron microscopy (TEM) images are in Supplemental Data, Figure S1. The TiO2NPs were obtained from Degussa/Evonik. The batch of material used in the present study was characterized by the National Institute of Environmental Health Sciences and had an average particle size of 21 nm, a Brunauer–Emmett–Teller surface area of 50 15 m2 g1, and a crystalline structure of 86% anatase and 14% rutile (unpublished data; Table 1). These values are very close to those provided by the manufacturer. Dry TiO2NP powder was added to ultrapure water (Millipore) to make a working suspension of 1000 mg L1, followed by stirring for 15 min.
Diameter (nm)
Nanomaterials
Manufacturer
MATERIALS AND METHODS
Nanoparticle
DOM in the aggregation and stability of AgNPs and TiO2NPs in seawater. Our working hypothesis regarding the importance of NP stability and aggregation is that the more stable a NP is in aqueous solution, the greater potential for exposure, and potential bioavailability, to pelagic and epibenthic aquatic organisms. In turn, these conditions could enhance NP bioaccumulation and toxicity. Citrate and polyvinylpyrrolidone (PVP) are 2 popular surface capping agents for AgNPs and have different stabilization mechanisms [3], resulting in different behaviors in solution. For example, PVP-capped AgNPs have been found to be more stable than citrate-capped AgNPs in both simple NaCl and CaCl2 solutions [13]. Another unique attribute of AgNPs is their dissolution when in aqueous solution, resulting in the potential release of bioavailable Agþ free ions [14,15]. Relative to capping agents, PVP-capped AgNPs are more prone to Agþ release than are citrate-capped AgNPs in river and lake water [16]. To date, there has been a lack of information on the effect that surface capping agents have on the behavior of NPs in seawater. In the present study, seawater salinity and DOC were investigated relative to their influence on the stabilization and aggregation of AgNP-citrate, AgNP-PVP and TiO2NPs over 10 d. Measurements of hydrodynamic diameter and zeta potential were used to determine the degree of colloidal stability of the NPs. The dissolution or release of dissolved Ag from AgNPs was also evaluated in solutions with different salinities. Statistical analysis was performed to identify if salinity and DOC were dominant in influencing the stability and aggregation of Ag and TiO2NPs in seawater. The results will improve our understanding of the behavior and potential for exposure of NPs in the marine environment.
H. Wang et al.
Source
Environ Toxicol Chem 33, 2014
Table 1. Characteristics of Ag and TiO2 nanoparticles used in the present study
1024
Stability and aggregation of silver and titanium dioxide
used in the experiments. The seawater used to prepare the brine was natural seawater and had the conservative ionic composition of all seawater [17]. The pH of the 30 ppt seawater was approximately 7.5, and the DOC concentration was below 1 mg L1. Salinity was measured by refractometry (Reichert Technologies). A DOM solution was prepared with the sodium form of Aldrich humic acid in ultrapure deionized water. Characterization of this type of commercially available humic material by Kim et al. [18] showed a composition of 55%, 4.5%, 0.3%, 38%, and 2.3% of carbon, hydrogen, nitrogen, oxygen, and sulfur, respectively. Dissolved organic matter concentrations, expressed as DOC, were measured on a Shimadzu DOC-V instrument. Hydrodynamic particle size and zeta potential of the AgNPs and TiO2NPs
Three hydrodynamic diameter and zeta-potential studies were performed. Square polystyrene BI-SCP cells from Brookhaven Instrument were used. In the first study, 1000 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NP suspensions were sonicated at 40 kHz and 130 W (Branson 3510) for 1 h and then transferred to 0 ppt (ultrapure deionized water), 1 ppt, 5 ppt, and 30 ppt reconstituted seawater treatments with and without 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC to achieve a final concentration of 1.5 mg L1. In the second study, DOC concentration was held constant at 3.25 mg L1, while both salinity and NP concentrations were varied at 0 ppt, 1 ppt, 5 ppt, and 30 ppt reconstituted seawater and at 1 mg L1, 5 mg L1, and 10 mg L1, respectively. The NP suspensions were sampled at 0 h, 4 h, 24 h, and 240 h and the hydrodynamic diameters measured. Prior to analysis, the NP suspensions were resonicated and mixed. The pH values of the solutions varied from 7.5 to 8.5 in the first study and from 6.5 to 7.8 in the second study. In the third study, the zeta potentials of the test suspensions (i.e., 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs with 0 mg L1, 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC added at 0 ppt, 1 ppt, 5 ppt, and 30 ppt salinities) were measured 24 h after preparation. The hydrodynamic diameters and zeta potentials of the NPs were measured using a ZetaPALS instrument from Brookhaven Instrument. The upper instrumental detection limit for hydrodynamic diameter was 3000 nm; values greater than this were reported as >3000 nm. Dissolved Ag from AgNP-citrate and AgNP-PVP
Dissolution of the AgNP-citrate and AgNP-PVP suspended in 5 ppt and 30 ppt seawater solutions were studied over a period of 5 d and 10 d at room temperature (22 8C). The suspensions were filtered through 0.2-mm Millipore Millex poly(tetrafluoroethylene) (PTFE) filter membranes, and the dissolved Ag was measured by atomic absorption spectroscopy (SIMAA 6000; PerkinElmer). Compared with AgNPs, TiO2NPs do not readily release Ti ions in seawater [10]; therefore, the dissolved Ti was not determined in the present study. Statistical analyses
Unless otherwise noted, the mean and 1 standard deviation are reported. One-way analysis of variance using SPSS 16.0 for Windows (SPSS) followed by Dunnett’s multiple comparison test were performed to identify significant differences between treatments. Furthermore, SPSS multivariate analysis was performed to examine interactions between variables (i.e., salinity, DOC). Statistical significance was accepted at p < 0.05 or p < 0.01.
Environ Toxicol Chem 33, 2014
1025
RESULTS AND DISCUSSION
Effects of salinity and DOC on the hydrodynamic diameters of AgNPs and TiO2NPs over time
Increasing the salinity of the aqueous solutions resulted in the accelerated aggregation of AgNPs and TiO2NPs as evidenced by increasing NP hydrodynamic diameters. This effect is demonstrated by examining the 4 salinities (0 ppt, 1 ppt, 5 ppt, and 30 ppt) over 3 time periods (0 h, 24 h, and 240 h) under 4 DOC concentrations (0 mg L1, 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC) (Figure 1; and Supplemental Data, Table S1). For example, Figure 1A and D show the hydrodynamic diameters of 1.5 mg L1 AgNP-citrate with and without 10.50 mg L1 DOC added at various salinities over 240 h. Without DOC, the NPs start to aggregate at 1 ppt salinity to approximately 400 nm; at 5 ppt and 30 ppt, the AgNP-citrate aggregated to approximately 3000 nm over 24 h and likely continued to aggregate to 240 h. The DOC stabilized the AgNPcitrate, and as the DOC concentration increased, the stabilization effect became more pronounced. For example, with 3.25 mg L1 DOC added (Figure 1C), the hydrodynamic diameters were 185 nm at 1 ppt over 24 h (significantly lower than 385 nm without DOC added; Figure 1A). With 10.50 mg L1 DOC added, the hydrodynamic diameters were only 424 nm at 5 ppt after 240 h (Figure 1D). Similar to AgNP-citrate, AgNP-PVP stability was not enhanced by 0.25 mg L1 DOC at any of the salinities (Figure 1E–H; Supplemental Data, Table S1). However, in the 3.25 mg L1 and 10.50 mg L1 DOC treatments (Figure 1G and H), hydrodynamic diameters were much smaller at 5 ppt over 24 h compared to the treatment without DOC present (Figure 1E). In addition, 10.50 mg L1 DOC stabilized the AgNP-citrate at 424 nm and the AgNP-PVP at 817 nm at 5 ppt over 240 h (Figure 1H). Compared with effects on AgNP-citrate and AgNP-PVP, DOC had a greater stabilization effect on TiO2NPs (Figure 1I–L; Supplemental Data, Table S1). At 0 h, the hydrodynamic diameters of TiO2NPs were approximately 5 times larger than those of the AgNPs, indicating a nearly instantaneous aggregation of the TiO2NPs in water. The 0.25 mg L1 DOC treatment stabilized the TiO2NPs at approximately 300 nm to 400 nm at 1 ppt but had little effect at 5 ppt and 30 ppt (Figure 1J). But when the DOC was increased to 3.25 mg L1 and 10.50 mg L1 the TiO2NPs were stabilized at 1 ppt, 5 ppt, and 30 ppt (Figure 1K and L). In particular, the 10.50 mg L1 DOC treatment stabilized the TiO2NPs at 30 ppt over 24 h, compared with the treatment without DOC present (Figure 1I), which suggests a greater stabilization effect by DOC on TiO2NPs as compared to AgNP-citrate and AgNP-PVP, even at high salinity where stabilization was not achieved. We further investigated the impact of adding DOC (3.25 mg L1) on the stability and aggregation of AgNP-citrate, AgNPPVP and TiO2NPs with NP concentrations at 1 mg L1, 5 mg L1, and 10 mg L1 over 4 salinities (0 ppt, 1 ppt, 5 ppt, 30 ppt) and 4 measurement times (0 h, 4 h, 24 h, 240 h) (Figure 2 and Supplemental Data, Table S2). The DOC stabilized the 1 mg L1, 5 mg L1, and 10 mg L1 AgNP-citrate, AgNP-PVP and TiO2NPs. For example, for 1 mg L1 and 5 mg L1 AgNPcitrate with DOC added, the hydrodynamic diameters were 316 nm and 333 nm, respectively, at 5 ppt over 240 h, which is much lower than the 939 nm and 884 nm for 1 mg L1 and 5 mg L1 AgNP-citrate without DOC added (Figure 2A and B). For 10 mg L1 AgNP-citrate with DOC added, the hydrodynamic diameters were 690 nm and 666 nm at 1 ppt and 5 ppt over
1026
Environ Toxicol Chem 33, 2014
B 10000
1000
1000
100
100
160
200
240
1000
100
100
Time (h)
40
80
120
160
200
240
Time (h)
D
1000
1000
100
100
0
40
80
120
160
200
Time (h)
E
0
240
40
80
120
F
1000
1000
100
100
160
200
120
160
200
0
240
40
80
120
200
10000
1000
1000
100
100
0
40
80
120
160
200
0
240
40
80
120
160
200
240
I
10000
1000
1000
100
100
10
10 0
40
80
120
160
200
160
200
240
80
120
160
200
0
240
40
80
F
Time (h)
120
160
200
240
Time (h)
10000
10000
1000
1000
100
100
10
10 40
80
120
160
200
240
0
10000
1000
1000
100
100
10 80
120
80
160
200
120
160
200
240
200
240
Time (h)
10000
40
40
H
Time (h)
10
240
0
40
80
120
160
Time (h)
I
10000
0
240
40
80
Time (h)
K
40
Time (h)
J
120
10
10
0
Time (h)
Time (h)
80
D
100
0
10
10
40
Time (h)
100
G
10000
0
240
1000
240
H
200
1000
0
160
160
10000
Time (h)
Time (h)
G
120
10000
240
10
10 80
80
E
10000
40
40
Time (h)
Time (h)
10000
0
0
C
10
10
10
10
0
10000
Hydrodynamic diameter (nm)
1000
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
120
10000
Hydrodynamic diameter (nm)
10000
120
160
200
240
Hydrodynamic diameter (nm)
80
Hydrodynamic diameter (nm)
40
C
Hydrodynamic diameter (nm)
10000
10 0
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
10000
10
B
A
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
A
H. Wang et al.
10000
1000
100
10
Time (h)
L
0
40
80
120
160
200
240
Time (h) 10000
10000
1000
1000
100
100
10
10 0
40
80
120
160
Time (h)
200
240
0
40
80
120
160
200
240
Time (h)
Figure 1. Time-dependent hydrodynamic diameter (nanometers) of 1.5 mg L1 Ag nanoparticle-citrate (A–D), Ag nanoparticle-polyvinylpyrrolidone (E–H), and TiO2 nanoparticles (I–L) in deionized water and 1 ppt, 5 ppt, and 30 ppt seawater in the presence of 0 mg L1 (A,E,I), 0.25 mg L1 (B,F,J), 3.25 mg L1 (C,G,K), and 10.50 mg L1 dissolved organic carbon (D,H,L). Hydrodynamic diameter values greater than 3000 nm are shown as 3000 nm. & ¼ 0 ppt; ¼ 1 ppt; ¼ 5 ppt; ¼ 30 ppt. [Color figure can be viewed in the online issue which is available at wileyonlinelibrary.com]
Figure 2. Time-dependent hydrodynamic diameter of 1 mg L1, 5 mg L1, and 10 mg L1 Ag nanoparticle-citrate (A–C); 1 mg L1, 5 mg L1, and 10 mg L1 Ag nanoparticle-polyvinylpyrrolidone (D–F); and 1 mg L1, 5 mg L1, and 10 mg L1 TiO2 nanoparticles (G–I) without and with 3.25 mg L1 dissolved organic carbon (DOC) added over 240 h. & ¼ 0 ppt; ¼ 5 ppt; ¼ 30 ppt; ¼ 0 ppt þ 3.25 mg L1 DOC; ¼ 5 ppt þ 3.25 mg L1 DOC; ¼ 30 ppt þ 3.25 mg L1 DOC. Hydrodynamic diameter values greater than 3000 nm are shown as 3000 nm. [Color figure can be viewed in the online issue which is available at wileyonlinelibrary.com]
24 h, respectively (Figure 2C). These values were lower than those without DOC added (978 nm and >3000 nm at 1 ppt and 5 ppt, respectively). However, in the 30 ppt treatment when the AgNP-citrate concentrations increased to 5 mg L1 and 10 mg L1, the addition of DOC did not affect the hydrodynamic diameters. Another significant result was that the DOC stabilized the 5 mg L1 and 10 mg L1 AgNP-citrate at 0 ppt over 240 h (Figure 2B and C; Supplemental Data, Table S2).
Stability and aggregation of silver and titanium dioxide
1027
Table 2. Effect of salinity, time, and dissolved organic carbon (DOC) and their interactions on the hydrodynamic diameter of 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs over 240 ha p value Factor Salinity Time DOC Salinity time Salinity DOC Time DOC
AgNP-citrate
AgNP-PVP
TiO2NP
0.000 0.000 0.184 0.001 0.157 0.178
0.000 0.000 0.220 0.000
0.000 0.000 0.000 0.001 0.004 0.066
a Interaction results between salinity, time, and DOC are presented with a multiplication symbol and significant difference is defined when p < 0.05. AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle.
significantly influenced by salinity and time with a significant interaction between salinity and time. For TiO2NPs, the hydrodynamic diameters were also statistically significantly influenced by salinity, time, as well as DOC and interactions between salinity and time and between salinity and DOC (Table 2). These results confirm that salinity, time, and DOC significantly affected the stability and aggregation of NPs in aqueous solutions. Dissolution of Ag from AgNPs
There was evidence of dissolution of ionic silver from 1 mg L1 AgNP-citrate and AgNP-PVP into 5 ppt and 30 ppt seawater (Figure 3). However, it is noteworthy that, because the AgNP diameter is 30 nm and could penetrate through the 0.2-mm PTFE membrane filter, the operationally defined dissolved Ag from AgNPs determined in the present study may contain some AgNPs. Therefore, our methods for examining Ag dissolution into solution may have been inferior compared with other methods, such as ultrafiltration [5]. In the present study, 1 mg L1 AgNP-citrate dissolved less readily at 30 ppt (0.032 mg L1 Ag) than at 5 ppt (0.14 mg L1 Ag) after 5 d (Figure 3A), and AgNP-PVP dissolved more effectively at 30 ppt (0.58 mg L1 Ag) than at 5 ppt (0.086 mg L1 Ag) after 10 d (Figure 3B). The amount of dissolved Ag was not significantly different between AgNP-citrate and AgNP-PVP at either 5 ppt or 30 ppt over 5 d or 10 d. Angel et al. [5] reported that 40 mg L1 AgNP-citrate and AgNP-PVP dissolved 1.47 mg L1 and
A
B -1 Dissolved Ag (mg L )
-1 Dissolved Ag (mg L )
The 3.25 mg L1 DOC level limited the increase of the hydrodynamic diameters of 5 mg L1 and 10 mg L1 AgNPPVP at 5 ppt (Figure 2E and F). Without DOC added, the hydrodynamic diameter of 5 mg L1 AgNP-PVP was 1258 nm over 24 h and continuously aggregated to >3000 nm over 240 h, while in the treatments with DOC added the hydrodynamic diameter was 346 nm over 24 h and 556 nm over 240 h. The DOC also stabilized the 10 mg L1 AgNP-PVP at approximately 2400 nm at 5 ppt over 240 h. In agreement with earlier experiments summarized in Figure 1, the presence of DOC did not stabilize the AgNP-citrate and AgNP-PVP at 30 ppt (Figure 2). Figure 2 also clearly shows that the hydrodynamic diameters of 1 mg L1, 5 mg L1, and 10 mg L1 TiO2NPs at various seawater salinities over 240 h were less when 3.25 mg L1 DOC was added. Generally, the hydrodynamic diameters of 1 mg L1 and 5 mg L1 TiO2NPs at 1 ppt and 5 ppt over 240 h were greater than 1000 nm without DOC added but below 1000 nm when DOC was present (Figure 2G and H). Unlike the AgNPs, when the TiO2NPs concentration increased to 5 mg L1 and 10 mg L1, DOC still had a stabilizing effect at 30 ppt over 24 h (Figure 2H and I). Another finding is that in the 0 ppt treatment, the hydrodynamic diameters were larger in the presence of DOC compared to the absence of DOC for AgNP-citrate and AgNPPVP (Figure 2A–F) and TiO2NPs (Figure 2G–I). The above results show that DOC can stabilize AgNP-citrate, AgNP-PVP, and TiO2NP suspensions for periods of time (up to 10 d) and that the stabilization effect was increased with increased DOC concentrations. The stabilization effect of DOC was observed at lower salinities (0–5 ppt) but, except for TiO2NP, was not observed at a higher salinity (30 ppt), suggesting that the DOC concentrations used could not overcome the destabilization of AgNPs in higher–ionic strength environments. Likely mechanisms of DOC stabilization have already been suggested in the literature. These include alteration of a particle’s surface charge, which influences electrostatic repulsive forces, and the DOC coating on the particle’s surface introducing repulsive steric particle–particle interactions that pose significant barriers to aggregation [13]. In addition, the dissolution of AgNP may influence the mechanism of AgNP aggregation. The precipitation of AgCl resulting from the increase in dissolved Ag concentrations as a result of the dissolution of AgNPs may lead to inter-particle bridging between AgNPs, which will in turn enhance the nanoparticle aggregation kinetics [19]. According to El Badawy et al. [3], AgNP-citrate is stabilized by electrostatic repulsion and AgNP-PVP is stabilized by steric repulsion, with ionic strength and electrolyte type (e.g., NaNO3, NaCl, or Ca[NO3]2) having no effect on the aggregation of the AgNP-PVP. The TiO2NPs were found to be sterically stabilized in the presence of fulvic acids, resulting in some disaggregation of small nanoparticle aggregates [4]. Sterically stabilized quantum dots did not aggregate at increased ionic strengths (e.g., 0.15 M NaCl) [20], while sterically stabilized alginatecoated hematite NPs also aggregated after adding calcium (9.0 mM) [21]. In the present study, AgNP-citrate, AgNP-PVP, and TiO2NPs, when in the presence of DOC, aggregated as salinity and exposure time increased, suggesting that both particle electrostatic repulsive forces and steric repulsive interactions can be overwhelmed by an increase in ionic strength (i.e., a salting-out effect). Under the same conditions, AgNP-PVP was more stable than AgNP-citrate in seawater. In addition, multivariate analysis showed that the hydrodynamic diameters of the 2 types of AgNPs were statistically
Environ Toxicol Chem 33, 2014
0.8 5
0.6 0.4
*
0.2 0.0
10
5 Time (d)
0.8
*
0.6 0.4 0.2 0.0 5
10 Time (d)
Figure 3. Dissolution of 1 mg L1 Ag nanoparticle (NP)-citrate (A) and AgNP-polyvinylpyrrolidone (B) suspensions in 5 ppt seawater (white bars) and 30 ppt seawater (gray bars) over 5 d and 10 d. Nanoparticle suspensions were filtered through 0.2-mm Millipore poly(tetrafluoroethylene) membrane filters, and the dissolved Ag was measured by atomic absorption spectroscopy. Significant difference at p < 0.05 between 5 ppt and 30 ppt seawater treatment at a single time period for AgNP-citrate or AgNPpolyvinylpyrrolidone.
1028
Environ Toxicol Chem 33, 2014
1.32 mg L1 Ag, respectively, into seawater after 72 h. In the present study, at 5 ppt and 30 ppt, approximately 5 mg to 6 mg Ag was dissolved from 10 mg AgNP (i.e., 50% dissolution), much higher than the 0.7% to 3.7% reported by Angel et al. [5]. A potential explantion for the difference in magnitudes of Ag dissolution between the present study and that of Angel et al. [5] (i.e., 1–3.7% dissolution vs 50%) is the initial AgNP concentration. In the current study, the initial concentration was 1 mg L1 Ag, whereas Angel et al. [5] used an initial concentration 40 times higher. In seawater solution, concentrations of Ag may have been closer to silver’s solubility product in Angel et al. [5], resulting in less Ag dissolving, compared with the greater magnitude of Ag dissolution observed in the current study. Liu and Hurt [22] suggested that the presence of chloride ions in electrolyte solutions will lead to the formation of soluble silver chloride complexes, such as AgCl2– and AgCl32–, so even simple colloids contain 3 forms of silver: Ag0 solids, free Agþ, or its complexes. Generally, the free Agþ is considered the bioavailable form of silver [23]; while the formation of AgCl2– and AgCl32– as a result of the dissolution of nanosilver represents a possible form for silver to be transported throughout the marine environment, the ecological risk associated with these forms of silver is uncertain. Zeta potentials of Ag and TiO2NPs
The zeta potential is a measure of the electrokinetic potential in a colloidal system and, therefore, of its stability. As the zeta potential approaches 0 (from either the negative or the positive side) the solution becomes less stable. Generally, a suspension that exhibits a zeta potential greater than –30 mV or less than 30 mV is considered unstable and will result in particles aggregating and settling out of solution in the absence of other factors [24]. In these studies, all the zeta potentials were negative but were more negative, and therefore more stable, at 1 ppt and 5 ppt and less negative, and thus more unstable, at 30 ppt compared to 0 ppt (Table 3). As the DOC concentrations increased, the zeta potentials became even more negative at 1 ppt and 5 ppt, probably as a result of of the NP surface becoming coated with DOC and the charge on the particle becoming dominated more extensively by the presence of the organic carbon [10]. Also in the 30 ppt treatment, the zeta potentials became less negative, compared with the 0 ppt, 1 ppt, and 5 ppt treatments, suggesting that the stability of the suspension decreased. Complexation of seawater electrolyte ions, such as Ca2þ, with the carboxyl groups on the surface of the citrateAgNPs could be a reason for the reduction in the zeta potential and subsequent particle aggregation [3]. At 0 ppt salinity, the zeta potentials of the TiO2NP suspensions were more negative than those of AgNP-citrate and AgNP-PVP suspensions, indicating that TiO2NP suspensions were more stable. Overall, these results were consistent with the hydrodynamic diameter measurements. Statistical analysis showed that for all 3 types of NPs, zeta potentials were significantly influenced by salinity; for TiO2NP, the zeta potentials were also significantly influenced by DOC at p < 0.05. CONCLUSIONS
In the present study, the behavior and fate of 2 types of AgNPs and TiO2NPs in seawater solutions were evaluated. The stability of the NPs was influenced by salinity and DOC, with salinity being more dominant than DOC. As measured by hydrodynamic diameter and zeta potentials, the DOC stabilized the NPs at lower salinities (e.g., 0–5 ppt) up to 240 h but not at a
H. Wang et al. Table 3. Effect of salinity and dissolved organic carbon (DOC) on the zeta potentials (mV) of 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs Nominal DOC (mg L1) 0 0.25 3.25 10.5 0 0.25 3.25 10.5 0 0.25 3.25 10.5 0 0.25 3.25 10.5
Salinity (ppt)
AgNP-citrate
AgNP-PVP
TiO2NP
0 0 0 0 1 1 1 1 5 5 5 5 30 30 30 30
10.5 4.2 7.4 1.2 10.3 3.6 16.1 2.7 12.0 4.7 14.3 1.4 15.2 1.9 19.8 0.3 11.5 0.3 17.7 0.5 15.6 1.7 17 3.1 9.4 4.0 6.5 0.4 4.9 1.5 10.3 1.9
8.6 0.8 10.8 2.0 14.9 1.7 16.2 3.6 17.5 3.5 17.9 0.6 22.1 0.9 20.7 0.6 13.8 2.0 15.3 2.3 17.1 2.7 14.4 3.4 13.6 4.5 6.4 1.9 9.4 1.9 10.1 0.1
12.5 3.4 12.4 0.9 17.4 2.0 22.1 3.7 14.1 3.5 14.1 0.7 19.8 3.6 25.3 1.5 7.2 2.9 18.9 2.6 16.8 1.8 14.7 1.5 6.1 5.2 7.1 3.4 8.4 4.8 8.9 2.6
AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle; ppt ¼ parts per thousand. Significant differences compared with 0 ppt treatment for a given nanoparticle and DOC concentration at p < 0.05 by one-way analysis of variance. Significant differences compared with 0 ppt treatment for a given nanoparticle and DOC concentration at p < 0.01 by one-way analysis of variance.
higher salinity (e.g., 30 ppt). It appeared that DOC has a greater stabilization effect on TiO2NPs than on AgNP-citrate and AgNP-PVP. The present study also indicates that the stability and aggregation of NPs in seawater can be regulated by changing the salinity and DOC concentrations. Because of the many processes affecting the stability of AgNP-citrate, AgNP-PVP and TiO2NPs in the marine environment, it is very likely that these NPs, if released into the environment, will have relatively brief residence times in the water column; instead, they are likely to aggregate and accumulate in the sediments. These processes will influence the exposure to and effects on benthic communities and thereby influence the ecological risk of NPs to these communities. The present study is critical to understanding and predicting the potential health and environmental risks of NPs and for developing informed regulations. SUPPLEMENTAL DATA
Tables S1–S2. Figure S1. (482 KB DOCX). Acknowledgment—H. Wang was supported by the US Environmental Protection Agency (USEPA) as a National Research Council postdoctoral research associate. The authors thank W. Boothman and B. Taplin at the USEPA (Atlantic Ecology Division, Narragansett, RI) for their assistance in the laboratory. The authors also thank A. Parks, R. McKinney, and K. Rocha for their insightful reviews of the draft manuscript. Mention of trade names of commercial products and companies does not constitute endorsement or recommendation for use. This report has been reviewed by the USEPA’s Office of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, Narragansett, RI, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the agency. This is USEPA ORD005072.
REFERENCES 1. Beaudrie CEH, Kandlikar M, Satterfield T. 2013. From cradle-to-grave at the nanoscale: Gaps in U.S. regulatory oversight along the nanomaterial life cycle. Environ Sci Technol 47:5524–5534.
Stability and aggregation of silver and titanium dioxide 2. Wijnhoven SWP, Peijnenburg WJGM, Herberts CA, Hagens WI, Oomen AG, Heugens EHW, Roszek B, Bisschops J, Gosens I, van de Meent D, Dekkers S, De Jong WH, van Zijverden M, Sips AJAM, Geertsma RE. 2009. Nano-silver—A review of available data and knowledge gaps in human and environmental risk assessment. Nanotoxicology 3(2):109–138. 3. El Badawy AM, Luxton TP, Silva RG, Scheckel CK, Suidan MT, Tolaymat TM. 2010. Impact of environmental conditions (pH, ionic strength, and electrolyte type) on the surface charge and aggregation of silver nanoparticles suspensions. Environ Sci Technol 44:1260–1266. 4. Domingos RF, Tufenkji N, Wilkinson KJ. 2009. Aggregation of titanium dioxide nanoparticles: Role of a fulvic acid. Environ Sci Technol 43:1282–1286. 5. Angel BM, Batley GE, Jarolimek CV, Rogers NJ. 2013. The impact of size on the fate and toxicity of nanoparticulate silver in aquatic systems. Chemosphere 93:359–365. 6. Ates M, Daniels J, Arslan Z, Farah IO. 2013. Effects of aqueous suspensions of titanium dioxide nanoparticles on Artemia salina: Assessment of nanoparticle aggregation, accumulation, and toxicity. Environ Monit Assess 185:3339–3348. 7. Ju-Nam Y, Lead JR. 2008. Manufactured nanoparticles: An overview of their chemistry, interactions and potential environmental implications. Sci Total Environ 400:396–414. 8. Domingos RF, Peyrot C, Wilkinson KJ. 2010. Aggregation of titanium dioxide nanoparticles: Role of calcium and phosphate. Environ Chem 7:61–66. 9. Saleh NB, Pfefferle LD, Elimelech M. 2010. Influence of biomacromolecules and humic acid on the aggregation kinetics of single-walled carbon nanotubes. Environ Sci Technol 44:2412–2418. 10. Keller AA, Wang H, Zhou D, Lenihan HS, Cherr G, Cardinale BJ, Miller R, Ji Z. 2010. Stability and aggregation of metal oxide nanoparticles in natural aqueous matrices. Environ Sci Technol 44:1962–1967. 11. Yang SP, Bar-Ilan O, Peterson RE, Heideman W, Hamers RJ, Pedersen JA. 2013. Influence of humic acid on titanium dioxide nanoparticle toxicity to developing zebrafish. Environ Sci Technol 47:4718–4725.
Environ Toxicol Chem 33, 2014
1029
12. McCarthy JF. 1989. Bioavailability and toxicity of metals and hydrophobic organic contaminants. In Suffet IH, McCarthy P, eds, Aquatic Humic Substances: Influence on Fate and Treatment of Pollutants. American Chemical Society, Washington, DC, pp 263–277. 13. Huynh KA, Chen KL. 2011. Aggregation kinetics of citrate and polyvinylpyrrolidone coated silver nanoparticles in monovalent and divalent electrolyte solutions. Environ Sci Technol 45:5564–5571. 14. El Badawy AM, Silva RG, Morris B, Scheckel KG, Suidan MT, Tolaymat TM. 2011. Surface charge-dependent toxicity of silver nanoparticles. Environ Sci Technol 45:283–287. 15. Hoheisel SM, Diamond S, Mount D. 2012. Comparison of nanosilver and ionic silver toxicity in Daphnia magna and Pimephales promelas. Environ Toxicol Chem 31:2557–2563. 16. Dobias J, Bernier-Latmani R. 2013. Silver release from silver nanoparticles in natural waters. Environ Sci Technol 47:4140–4146. 17. Millero FJ, Sohn ML. 1992. Chemical Oceanography. CRC Press, Boca Raton, FL, USA. 18. Kim JI, Buckau G, Li GH, Duschner H, Psarros N. 1990. Characterization of humic and fulvic acids from Gorleben groundwater. Fresenius J Anal Chem 338:245–252. 19. Li X, Lenhart JJ, Walker HW. 2010. Dissolution-accompanied aggregation kinetics of silver nanoparticles. Langmuir 26:16690–16698. 20. Jiang J, Oberdorster G, Biswas P. 2009. Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies. J Nanopart Res 11:77–89. 21. Chen KL, Mylon SE, Elimelech M. 2006. Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environ Sci Technol 40:1516–1523. 22. Liu JY, Hurt RH. 2010. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ Sci Technol 44:2169–2175. 23. Berry WJ, Cantwell MG, Edwards PA, Serbst JR, Hansen DJ. 1999. Predicting toxicity of sediments spiked with silver. Environ Toxicol Chem 18:40–48. 24. Cosgrove T. 2005. Colloid Science: Principles, Methods and Applications. Blackwell, Oxford, UK.
STABILITY AND AGGREGATION OF SILVER AND TITANIUM DIOXIDE NANOPARTICLES IN SEAWATER: ROLE OF SALINITY AND DISSOLVED ORGANIC CARBON HUANHUA WANG,yz ROBERT M. BURGESS,*x MARK G. CANTWELL,x LISA M. PORTIS,x MONIQUE M. PERRON,k FENGCHANG WU,z and KAY T. HOx yNational Research Council, US Environmental Protection Agency, Office of Research and Development, Narragansett, Rhode Island zState Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing, China xUS Environmental Protection Agency, Office of Research and Development, Narragansett, Rhode Island kUS Environmental Protection Agency, Office of Pesticide Programs, Washington, DC (Submitted 6 December 2013; Returned for Revision 5 January 2014; Accepted 18 January 2014) Abstract: The behavior and fate of nanoparticles (NPs) in the marine environment are largely unknown and potentially have important environmental and human health implications. The aggregation and fate of NPs in the marine environment are greatly influenced by their interactions with seawater and dissolved organic carbon (DOC). In the present study, the stability and aggregation of 30-nm–diameter silver nanoparticles (AgNPs) capped with citrate and polyvinylpyrrolidone (PVP; AgNP-citrate and AgNP-PVP) and 21-nm-diameter titanium dioxide (TiO2) NPs as affected by seawater salinity and DOC were investigated by measuring hydrodynamic diameters and zeta potentials. The added DOC (in humic acid form) stabilized the 3 types of NPs when the seawater salinities were 5 parts per thousand (ppt), but the stabilizing effect of DOC was reduced by a higher salinity (e.g., 30 ppt). In addition, AgNP-PVP was more stable than AgNPcitrate in seawater, indicating that surface capping agents and stabilization mechanisms govern the stability and aggregation of NPs. Statistical analysis showed that salinity is the most dominant influence on the stability and aggregation of AgNPs and TiO2NPs, followed by DOC. These findings expand our knowledge on the behavior of AgNPs and TiO2NPs in seawater and indicate that the fate of these NPs will be primarily to aggregate in the water column, precipitate, and accumulate in sediments following release into the marine environment. Environ Toxicol Chem 2014;33:1023–1029. # 2014 SETAC Keywords: Stability
Silver
Titanium dioxide
Nanoparticle
Seawater
to maintain colloidal stability in solution [7]. Given these uncertainties regarding metal nanomaterials, the objective of the present study was to investigate the effects of seawater salinity (i.e., ionic strength) and dissolved organic carbon (DOC) on the aqueous stability and aggregation of AgNPs and TiO2NPs. It has been shown that increasing the ionic strength of a solution will result in the destabilization and aggregation of NPs. For example, Domingos et al. [8] added mono- and divalent electrolytes (e.g., NaCl and CaCl2) to aqueous solutions and found that the presence of calcium resulted in aggregation of TiO2NPs despite the presence of dissolved organic matter (DOM; i.e., Suwannee River fulvic acid). However, these relatively simple electrolyte solutions were not representative of an environmentally complex medium such as seawater. Also, there is a paucity of published studies concerning the effects of salinity on the stability and aggregation of NPs over time (e.g., 10 d). The presence of DOM can influence the aggregation of NPs, and thereby their interactions with natural interfaces and organisms [9,10], and can alter the bioavailability and toxicity of NPs in solution [11]. In addition, metals associated with DOM have shown limited bioavailability to aquatic organisms, compared with the freely dissolved form of the metals [12]. For example, the aggregation rate of single-walled carbon nanotubes was slowed in the presence of Suwannee River humic acid, a form of DOM, possibly because of steric repulsion originating from the adsorbed macromolecular layer [9]. This type of enhanced colloidal stability is likely to lead to increased residence times for NPs in the overlying seawater and, consequently, to potentially elevated exposures to aquatic organisms. To date, no studies have focused on the role of
INTRODUCTION
Nanotechnology has taken the leap from the realm of research to integration into a growing number of applications in industrial and consumer products [1]. There are numerous nanoparticles (NPs) currently in production and use, including 2 important metal nanomaterials: silver nanoparticles (AgNPs) and titanium dioxide nanoparticles (TiO2NPs). Silver nanoparticles are used extensively in consumer goods, including textiles, cosmetics, and health-care products to utilize their strong antimicrobial activity [2] and in electronic components [3]. Titanium dioxide nanoparticles have a range of applications in consumer products, including sunscreens, cosmetics, paints, and photocatalysts, because of their unique optical properties [4]. Marine, estuarine, and coastal environments are the expected destination of many NPs, through industrial discharges and municipal effluents. However, there are very few studies on the stability and aggregation of NPs in seawater. Angel et al. [5] investigated the behavior of AgNPs over 72 h in seawater, and Ates et al. [6] examined the fate of TiO2NPs over 24 h. Despite these very recent additions to the literature, the behavior and fate of NPs in the marine environment are still not well understood. The high surface area to volume ratio of NPs results in high surface reactivity, which leads to particle aggregation and settling when in aqueous solution. For some nanomaterials, such as AgNPs, capping agents have been applied to the outer surface in an effort All Supplemental Data may be found in the online version of this article. * Address correspondence to [email protected]. Published online 27 January 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/etc.2529 1023
http://ehp.niehs.nih.gov Anatase (86%), rutile (14%) 3.5–4.5 21
a Based on transmission electron microscopy. AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle; na ¼ not available.
50 15 Spheroid
209 8
34.4 1.6
nanocomposix.com PVP surface 6.4 26.7 na 36.2 Spherical 28.7 2.1a
a
nanocomposix.com Citrate surface 7.7 29.0 3.1
a
Spherical
39.6
na
41.8
Composition or surface pH Zeta potential (mV) Surface area (m2 g1) Hydrodynamic diameter (nm)
nanoComposix (San Diego, CA, USA) nanoComposix (San Diego, CA, USA) Degussa/Evonik (Parsippany, NJ, USA)
Reconstituted seawater of varying salinities was prepared by diluting a filtered brine solution with ultrapure deionized water. Brine was produced by evaporating Narragansett Bay (RI, USA) seawater to a salinity of 100 parts per thousand (ppt; i.e., grams per liter). The natural seawater used to prepare the brine was filtered through a flow-through string filter with an effective pore size of 10 mm (Purolator Advanced Filtration). The brine was prepared and then stored at 20 8C in the dark to limit biological growth until use. Any brine showing evidence of growth (i.e., discoloration) was not
AgNP-citrate (BioPure) AgNP-PVP (BioPure) TiO2NP
Preparation of saline and DOC solutions
Shape
Nanoparticle suspensions of 30-nm–diameter Biopure citrate- and PVP-coated AgNPs were purchased from Nanocomposix at a concentration of 1000 mg L1. The specific characteristics of the materials can be found at www.nanocomposix.com (Table 1), and transmission electron microscopy (TEM) images are in Supplemental Data, Figure S1. The TiO2NPs were obtained from Degussa/Evonik. The batch of material used in the present study was characterized by the National Institute of Environmental Health Sciences and had an average particle size of 21 nm, a Brunauer–Emmett–Teller surface area of 50 15 m2 g1, and a crystalline structure of 86% anatase and 14% rutile (unpublished data; Table 1). These values are very close to those provided by the manufacturer. Dry TiO2NP powder was added to ultrapure water (Millipore) to make a working suspension of 1000 mg L1, followed by stirring for 15 min.
Diameter (nm)
Nanomaterials
Manufacturer
MATERIALS AND METHODS
Nanoparticle
DOM in the aggregation and stability of AgNPs and TiO2NPs in seawater. Our working hypothesis regarding the importance of NP stability and aggregation is that the more stable a NP is in aqueous solution, the greater potential for exposure, and potential bioavailability, to pelagic and epibenthic aquatic organisms. In turn, these conditions could enhance NP bioaccumulation and toxicity. Citrate and polyvinylpyrrolidone (PVP) are 2 popular surface capping agents for AgNPs and have different stabilization mechanisms [3], resulting in different behaviors in solution. For example, PVP-capped AgNPs have been found to be more stable than citrate-capped AgNPs in both simple NaCl and CaCl2 solutions [13]. Another unique attribute of AgNPs is their dissolution when in aqueous solution, resulting in the potential release of bioavailable Agþ free ions [14,15]. Relative to capping agents, PVP-capped AgNPs are more prone to Agþ release than are citrate-capped AgNPs in river and lake water [16]. To date, there has been a lack of information on the effect that surface capping agents have on the behavior of NPs in seawater. In the present study, seawater salinity and DOC were investigated relative to their influence on the stabilization and aggregation of AgNP-citrate, AgNP-PVP and TiO2NPs over 10 d. Measurements of hydrodynamic diameter and zeta potential were used to determine the degree of colloidal stability of the NPs. The dissolution or release of dissolved Ag from AgNPs was also evaluated in solutions with different salinities. Statistical analysis was performed to identify if salinity and DOC were dominant in influencing the stability and aggregation of Ag and TiO2NPs in seawater. The results will improve our understanding of the behavior and potential for exposure of NPs in the marine environment.
H. Wang et al.
Source
Environ Toxicol Chem 33, 2014
Table 1. Characteristics of Ag and TiO2 nanoparticles used in the present study
1024
Stability and aggregation of silver and titanium dioxide
used in the experiments. The seawater used to prepare the brine was natural seawater and had the conservative ionic composition of all seawater [17]. The pH of the 30 ppt seawater was approximately 7.5, and the DOC concentration was below 1 mg L1. Salinity was measured by refractometry (Reichert Technologies). A DOM solution was prepared with the sodium form of Aldrich humic acid in ultrapure deionized water. Characterization of this type of commercially available humic material by Kim et al. [18] showed a composition of 55%, 4.5%, 0.3%, 38%, and 2.3% of carbon, hydrogen, nitrogen, oxygen, and sulfur, respectively. Dissolved organic matter concentrations, expressed as DOC, were measured on a Shimadzu DOC-V instrument. Hydrodynamic particle size and zeta potential of the AgNPs and TiO2NPs
Three hydrodynamic diameter and zeta-potential studies were performed. Square polystyrene BI-SCP cells from Brookhaven Instrument were used. In the first study, 1000 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NP suspensions were sonicated at 40 kHz and 130 W (Branson 3510) for 1 h and then transferred to 0 ppt (ultrapure deionized water), 1 ppt, 5 ppt, and 30 ppt reconstituted seawater treatments with and without 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC to achieve a final concentration of 1.5 mg L1. In the second study, DOC concentration was held constant at 3.25 mg L1, while both salinity and NP concentrations were varied at 0 ppt, 1 ppt, 5 ppt, and 30 ppt reconstituted seawater and at 1 mg L1, 5 mg L1, and 10 mg L1, respectively. The NP suspensions were sampled at 0 h, 4 h, 24 h, and 240 h and the hydrodynamic diameters measured. Prior to analysis, the NP suspensions were resonicated and mixed. The pH values of the solutions varied from 7.5 to 8.5 in the first study and from 6.5 to 7.8 in the second study. In the third study, the zeta potentials of the test suspensions (i.e., 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs with 0 mg L1, 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC added at 0 ppt, 1 ppt, 5 ppt, and 30 ppt salinities) were measured 24 h after preparation. The hydrodynamic diameters and zeta potentials of the NPs were measured using a ZetaPALS instrument from Brookhaven Instrument. The upper instrumental detection limit for hydrodynamic diameter was 3000 nm; values greater than this were reported as >3000 nm. Dissolved Ag from AgNP-citrate and AgNP-PVP
Dissolution of the AgNP-citrate and AgNP-PVP suspended in 5 ppt and 30 ppt seawater solutions were studied over a period of 5 d and 10 d at room temperature (22 8C). The suspensions were filtered through 0.2-mm Millipore Millex poly(tetrafluoroethylene) (PTFE) filter membranes, and the dissolved Ag was measured by atomic absorption spectroscopy (SIMAA 6000; PerkinElmer). Compared with AgNPs, TiO2NPs do not readily release Ti ions in seawater [10]; therefore, the dissolved Ti was not determined in the present study. Statistical analyses
Unless otherwise noted, the mean and 1 standard deviation are reported. One-way analysis of variance using SPSS 16.0 for Windows (SPSS) followed by Dunnett’s multiple comparison test were performed to identify significant differences between treatments. Furthermore, SPSS multivariate analysis was performed to examine interactions between variables (i.e., salinity, DOC). Statistical significance was accepted at p < 0.05 or p < 0.01.
Environ Toxicol Chem 33, 2014
1025
RESULTS AND DISCUSSION
Effects of salinity and DOC on the hydrodynamic diameters of AgNPs and TiO2NPs over time
Increasing the salinity of the aqueous solutions resulted in the accelerated aggregation of AgNPs and TiO2NPs as evidenced by increasing NP hydrodynamic diameters. This effect is demonstrated by examining the 4 salinities (0 ppt, 1 ppt, 5 ppt, and 30 ppt) over 3 time periods (0 h, 24 h, and 240 h) under 4 DOC concentrations (0 mg L1, 0.25 mg L1, 3.25 mg L1, and 10.50 mg L1 DOC) (Figure 1; and Supplemental Data, Table S1). For example, Figure 1A and D show the hydrodynamic diameters of 1.5 mg L1 AgNP-citrate with and without 10.50 mg L1 DOC added at various salinities over 240 h. Without DOC, the NPs start to aggregate at 1 ppt salinity to approximately 400 nm; at 5 ppt and 30 ppt, the AgNP-citrate aggregated to approximately 3000 nm over 24 h and likely continued to aggregate to 240 h. The DOC stabilized the AgNPcitrate, and as the DOC concentration increased, the stabilization effect became more pronounced. For example, with 3.25 mg L1 DOC added (Figure 1C), the hydrodynamic diameters were 185 nm at 1 ppt over 24 h (significantly lower than 385 nm without DOC added; Figure 1A). With 10.50 mg L1 DOC added, the hydrodynamic diameters were only 424 nm at 5 ppt after 240 h (Figure 1D). Similar to AgNP-citrate, AgNP-PVP stability was not enhanced by 0.25 mg L1 DOC at any of the salinities (Figure 1E–H; Supplemental Data, Table S1). However, in the 3.25 mg L1 and 10.50 mg L1 DOC treatments (Figure 1G and H), hydrodynamic diameters were much smaller at 5 ppt over 24 h compared to the treatment without DOC present (Figure 1E). In addition, 10.50 mg L1 DOC stabilized the AgNP-citrate at 424 nm and the AgNP-PVP at 817 nm at 5 ppt over 240 h (Figure 1H). Compared with effects on AgNP-citrate and AgNP-PVP, DOC had a greater stabilization effect on TiO2NPs (Figure 1I–L; Supplemental Data, Table S1). At 0 h, the hydrodynamic diameters of TiO2NPs were approximately 5 times larger than those of the AgNPs, indicating a nearly instantaneous aggregation of the TiO2NPs in water. The 0.25 mg L1 DOC treatment stabilized the TiO2NPs at approximately 300 nm to 400 nm at 1 ppt but had little effect at 5 ppt and 30 ppt (Figure 1J). But when the DOC was increased to 3.25 mg L1 and 10.50 mg L1 the TiO2NPs were stabilized at 1 ppt, 5 ppt, and 30 ppt (Figure 1K and L). In particular, the 10.50 mg L1 DOC treatment stabilized the TiO2NPs at 30 ppt over 24 h, compared with the treatment without DOC present (Figure 1I), which suggests a greater stabilization effect by DOC on TiO2NPs as compared to AgNP-citrate and AgNP-PVP, even at high salinity where stabilization was not achieved. We further investigated the impact of adding DOC (3.25 mg L1) on the stability and aggregation of AgNP-citrate, AgNPPVP and TiO2NPs with NP concentrations at 1 mg L1, 5 mg L1, and 10 mg L1 over 4 salinities (0 ppt, 1 ppt, 5 ppt, 30 ppt) and 4 measurement times (0 h, 4 h, 24 h, 240 h) (Figure 2 and Supplemental Data, Table S2). The DOC stabilized the 1 mg L1, 5 mg L1, and 10 mg L1 AgNP-citrate, AgNP-PVP and TiO2NPs. For example, for 1 mg L1 and 5 mg L1 AgNPcitrate with DOC added, the hydrodynamic diameters were 316 nm and 333 nm, respectively, at 5 ppt over 240 h, which is much lower than the 939 nm and 884 nm for 1 mg L1 and 5 mg L1 AgNP-citrate without DOC added (Figure 2A and B). For 10 mg L1 AgNP-citrate with DOC added, the hydrodynamic diameters were 690 nm and 666 nm at 1 ppt and 5 ppt over
1026
Environ Toxicol Chem 33, 2014
B 10000
1000
1000
100
100
160
200
240
1000
100
100
Time (h)
40
80
120
160
200
240
Time (h)
D
1000
1000
100
100
0
40
80
120
160
200
Time (h)
E
0
240
40
80
120
F
1000
1000
100
100
160
200
120
160
200
0
240
40
80
120
200
10000
1000
1000
100
100
0
40
80
120
160
200
0
240
40
80
120
160
200
240
I
10000
1000
1000
100
100
10
10 0
40
80
120
160
200
160
200
240
80
120
160
200
0
240
40
80
F
Time (h)
120
160
200
240
Time (h)
10000
10000
1000
1000
100
100
10
10 40
80
120
160
200
240
0
10000
1000
1000
100
100
10 80
120
80
160
200
120
160
200
240
200
240
Time (h)
10000
40
40
H
Time (h)
10
240
0
40
80
120
160
Time (h)
I
10000
0
240
40
80
Time (h)
K
40
Time (h)
J
120
10
10
0
Time (h)
Time (h)
80
D
100
0
10
10
40
Time (h)
100
G
10000
0
240
1000
240
H
200
1000
0
160
160
10000
Time (h)
Time (h)
G
120
10000
240
10
10 80
80
E
10000
40
40
Time (h)
Time (h)
10000
0
0
C
10
10
10
10
0
10000
Hydrodynamic diameter (nm)
1000
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
120
10000
Hydrodynamic diameter (nm)
10000
120
160
200
240
Hydrodynamic diameter (nm)
80
Hydrodynamic diameter (nm)
40
C
Hydrodynamic diameter (nm)
10000
10 0
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
10000
10
B
A
Hydrodynamic diameter (nm)
Hydrodynamic diameter (nm)
A
H. Wang et al.
10000
1000
100
10
Time (h)
L
0
40
80
120
160
200
240
Time (h) 10000
10000
1000
1000
100
100
10
10 0
40
80
120
160
Time (h)
200
240
0
40
80
120
160
200
240
Time (h)
Figure 1. Time-dependent hydrodynamic diameter (nanometers) of 1.5 mg L1 Ag nanoparticle-citrate (A–D), Ag nanoparticle-polyvinylpyrrolidone (E–H), and TiO2 nanoparticles (I–L) in deionized water and 1 ppt, 5 ppt, and 30 ppt seawater in the presence of 0 mg L1 (A,E,I), 0.25 mg L1 (B,F,J), 3.25 mg L1 (C,G,K), and 10.50 mg L1 dissolved organic carbon (D,H,L). Hydrodynamic diameter values greater than 3000 nm are shown as 3000 nm. & ¼ 0 ppt; ¼ 1 ppt; ¼ 5 ppt; ¼ 30 ppt. [Color figure can be viewed in the online issue which is available at wileyonlinelibrary.com]
Figure 2. Time-dependent hydrodynamic diameter of 1 mg L1, 5 mg L1, and 10 mg L1 Ag nanoparticle-citrate (A–C); 1 mg L1, 5 mg L1, and 10 mg L1 Ag nanoparticle-polyvinylpyrrolidone (D–F); and 1 mg L1, 5 mg L1, and 10 mg L1 TiO2 nanoparticles (G–I) without and with 3.25 mg L1 dissolved organic carbon (DOC) added over 240 h. & ¼ 0 ppt; ¼ 5 ppt; ¼ 30 ppt; ¼ 0 ppt þ 3.25 mg L1 DOC; ¼ 5 ppt þ 3.25 mg L1 DOC; ¼ 30 ppt þ 3.25 mg L1 DOC. Hydrodynamic diameter values greater than 3000 nm are shown as 3000 nm. [Color figure can be viewed in the online issue which is available at wileyonlinelibrary.com]
24 h, respectively (Figure 2C). These values were lower than those without DOC added (978 nm and >3000 nm at 1 ppt and 5 ppt, respectively). However, in the 30 ppt treatment when the AgNP-citrate concentrations increased to 5 mg L1 and 10 mg L1, the addition of DOC did not affect the hydrodynamic diameters. Another significant result was that the DOC stabilized the 5 mg L1 and 10 mg L1 AgNP-citrate at 0 ppt over 240 h (Figure 2B and C; Supplemental Data, Table S2).
Stability and aggregation of silver and titanium dioxide
1027
Table 2. Effect of salinity, time, and dissolved organic carbon (DOC) and their interactions on the hydrodynamic diameter of 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs over 240 ha p value Factor Salinity Time DOC Salinity time Salinity DOC Time DOC
AgNP-citrate
AgNP-PVP
TiO2NP
0.000 0.000 0.184 0.001 0.157 0.178
0.000 0.000 0.220 0.000
0.000 0.000 0.000 0.001 0.004 0.066
a Interaction results between salinity, time, and DOC are presented with a multiplication symbol and significant difference is defined when p < 0.05. AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle.
significantly influenced by salinity and time with a significant interaction between salinity and time. For TiO2NPs, the hydrodynamic diameters were also statistically significantly influenced by salinity, time, as well as DOC and interactions between salinity and time and between salinity and DOC (Table 2). These results confirm that salinity, time, and DOC significantly affected the stability and aggregation of NPs in aqueous solutions. Dissolution of Ag from AgNPs
There was evidence of dissolution of ionic silver from 1 mg L1 AgNP-citrate and AgNP-PVP into 5 ppt and 30 ppt seawater (Figure 3). However, it is noteworthy that, because the AgNP diameter is 30 nm and could penetrate through the 0.2-mm PTFE membrane filter, the operationally defined dissolved Ag from AgNPs determined in the present study may contain some AgNPs. Therefore, our methods for examining Ag dissolution into solution may have been inferior compared with other methods, such as ultrafiltration [5]. In the present study, 1 mg L1 AgNP-citrate dissolved less readily at 30 ppt (0.032 mg L1 Ag) than at 5 ppt (0.14 mg L1 Ag) after 5 d (Figure 3A), and AgNP-PVP dissolved more effectively at 30 ppt (0.58 mg L1 Ag) than at 5 ppt (0.086 mg L1 Ag) after 10 d (Figure 3B). The amount of dissolved Ag was not significantly different between AgNP-citrate and AgNP-PVP at either 5 ppt or 30 ppt over 5 d or 10 d. Angel et al. [5] reported that 40 mg L1 AgNP-citrate and AgNP-PVP dissolved 1.47 mg L1 and
A
B -1 Dissolved Ag (mg L )
-1 Dissolved Ag (mg L )
The 3.25 mg L1 DOC level limited the increase of the hydrodynamic diameters of 5 mg L1 and 10 mg L1 AgNPPVP at 5 ppt (Figure 2E and F). Without DOC added, the hydrodynamic diameter of 5 mg L1 AgNP-PVP was 1258 nm over 24 h and continuously aggregated to >3000 nm over 240 h, while in the treatments with DOC added the hydrodynamic diameter was 346 nm over 24 h and 556 nm over 240 h. The DOC also stabilized the 10 mg L1 AgNP-PVP at approximately 2400 nm at 5 ppt over 240 h. In agreement with earlier experiments summarized in Figure 1, the presence of DOC did not stabilize the AgNP-citrate and AgNP-PVP at 30 ppt (Figure 2). Figure 2 also clearly shows that the hydrodynamic diameters of 1 mg L1, 5 mg L1, and 10 mg L1 TiO2NPs at various seawater salinities over 240 h were less when 3.25 mg L1 DOC was added. Generally, the hydrodynamic diameters of 1 mg L1 and 5 mg L1 TiO2NPs at 1 ppt and 5 ppt over 240 h were greater than 1000 nm without DOC added but below 1000 nm when DOC was present (Figure 2G and H). Unlike the AgNPs, when the TiO2NPs concentration increased to 5 mg L1 and 10 mg L1, DOC still had a stabilizing effect at 30 ppt over 24 h (Figure 2H and I). Another finding is that in the 0 ppt treatment, the hydrodynamic diameters were larger in the presence of DOC compared to the absence of DOC for AgNP-citrate and AgNPPVP (Figure 2A–F) and TiO2NPs (Figure 2G–I). The above results show that DOC can stabilize AgNP-citrate, AgNP-PVP, and TiO2NP suspensions for periods of time (up to 10 d) and that the stabilization effect was increased with increased DOC concentrations. The stabilization effect of DOC was observed at lower salinities (0–5 ppt) but, except for TiO2NP, was not observed at a higher salinity (30 ppt), suggesting that the DOC concentrations used could not overcome the destabilization of AgNPs in higher–ionic strength environments. Likely mechanisms of DOC stabilization have already been suggested in the literature. These include alteration of a particle’s surface charge, which influences electrostatic repulsive forces, and the DOC coating on the particle’s surface introducing repulsive steric particle–particle interactions that pose significant barriers to aggregation [13]. In addition, the dissolution of AgNP may influence the mechanism of AgNP aggregation. The precipitation of AgCl resulting from the increase in dissolved Ag concentrations as a result of the dissolution of AgNPs may lead to inter-particle bridging between AgNPs, which will in turn enhance the nanoparticle aggregation kinetics [19]. According to El Badawy et al. [3], AgNP-citrate is stabilized by electrostatic repulsion and AgNP-PVP is stabilized by steric repulsion, with ionic strength and electrolyte type (e.g., NaNO3, NaCl, or Ca[NO3]2) having no effect on the aggregation of the AgNP-PVP. The TiO2NPs were found to be sterically stabilized in the presence of fulvic acids, resulting in some disaggregation of small nanoparticle aggregates [4]. Sterically stabilized quantum dots did not aggregate at increased ionic strengths (e.g., 0.15 M NaCl) [20], while sterically stabilized alginatecoated hematite NPs also aggregated after adding calcium (9.0 mM) [21]. In the present study, AgNP-citrate, AgNP-PVP, and TiO2NPs, when in the presence of DOC, aggregated as salinity and exposure time increased, suggesting that both particle electrostatic repulsive forces and steric repulsive interactions can be overwhelmed by an increase in ionic strength (i.e., a salting-out effect). Under the same conditions, AgNP-PVP was more stable than AgNP-citrate in seawater. In addition, multivariate analysis showed that the hydrodynamic diameters of the 2 types of AgNPs were statistically
Environ Toxicol Chem 33, 2014
0.8 5
0.6 0.4
*
0.2 0.0
10
5 Time (d)
0.8
*
0.6 0.4 0.2 0.0 5
10 Time (d)
Figure 3. Dissolution of 1 mg L1 Ag nanoparticle (NP)-citrate (A) and AgNP-polyvinylpyrrolidone (B) suspensions in 5 ppt seawater (white bars) and 30 ppt seawater (gray bars) over 5 d and 10 d. Nanoparticle suspensions were filtered through 0.2-mm Millipore poly(tetrafluoroethylene) membrane filters, and the dissolved Ag was measured by atomic absorption spectroscopy. Significant difference at p < 0.05 between 5 ppt and 30 ppt seawater treatment at a single time period for AgNP-citrate or AgNPpolyvinylpyrrolidone.
1028
Environ Toxicol Chem 33, 2014
1.32 mg L1 Ag, respectively, into seawater after 72 h. In the present study, at 5 ppt and 30 ppt, approximately 5 mg to 6 mg Ag was dissolved from 10 mg AgNP (i.e., 50% dissolution), much higher than the 0.7% to 3.7% reported by Angel et al. [5]. A potential explantion for the difference in magnitudes of Ag dissolution between the present study and that of Angel et al. [5] (i.e., 1–3.7% dissolution vs 50%) is the initial AgNP concentration. In the current study, the initial concentration was 1 mg L1 Ag, whereas Angel et al. [5] used an initial concentration 40 times higher. In seawater solution, concentrations of Ag may have been closer to silver’s solubility product in Angel et al. [5], resulting in less Ag dissolving, compared with the greater magnitude of Ag dissolution observed in the current study. Liu and Hurt [22] suggested that the presence of chloride ions in electrolyte solutions will lead to the formation of soluble silver chloride complexes, such as AgCl2– and AgCl32–, so even simple colloids contain 3 forms of silver: Ag0 solids, free Agþ, or its complexes. Generally, the free Agþ is considered the bioavailable form of silver [23]; while the formation of AgCl2– and AgCl32– as a result of the dissolution of nanosilver represents a possible form for silver to be transported throughout the marine environment, the ecological risk associated with these forms of silver is uncertain. Zeta potentials of Ag and TiO2NPs
The zeta potential is a measure of the electrokinetic potential in a colloidal system and, therefore, of its stability. As the zeta potential approaches 0 (from either the negative or the positive side) the solution becomes less stable. Generally, a suspension that exhibits a zeta potential greater than –30 mV or less than 30 mV is considered unstable and will result in particles aggregating and settling out of solution in the absence of other factors [24]. In these studies, all the zeta potentials were negative but were more negative, and therefore more stable, at 1 ppt and 5 ppt and less negative, and thus more unstable, at 30 ppt compared to 0 ppt (Table 3). As the DOC concentrations increased, the zeta potentials became even more negative at 1 ppt and 5 ppt, probably as a result of of the NP surface becoming coated with DOC and the charge on the particle becoming dominated more extensively by the presence of the organic carbon [10]. Also in the 30 ppt treatment, the zeta potentials became less negative, compared with the 0 ppt, 1 ppt, and 5 ppt treatments, suggesting that the stability of the suspension decreased. Complexation of seawater electrolyte ions, such as Ca2þ, with the carboxyl groups on the surface of the citrateAgNPs could be a reason for the reduction in the zeta potential and subsequent particle aggregation [3]. At 0 ppt salinity, the zeta potentials of the TiO2NP suspensions were more negative than those of AgNP-citrate and AgNP-PVP suspensions, indicating that TiO2NP suspensions were more stable. Overall, these results were consistent with the hydrodynamic diameter measurements. Statistical analysis showed that for all 3 types of NPs, zeta potentials were significantly influenced by salinity; for TiO2NP, the zeta potentials were also significantly influenced by DOC at p < 0.05. CONCLUSIONS
In the present study, the behavior and fate of 2 types of AgNPs and TiO2NPs in seawater solutions were evaluated. The stability of the NPs was influenced by salinity and DOC, with salinity being more dominant than DOC. As measured by hydrodynamic diameter and zeta potentials, the DOC stabilized the NPs at lower salinities (e.g., 0–5 ppt) up to 240 h but not at a
H. Wang et al. Table 3. Effect of salinity and dissolved organic carbon (DOC) on the zeta potentials (mV) of 1.5 mg L1 AgNP-citrate, AgNP-PVP, and TiO2NPs Nominal DOC (mg L1) 0 0.25 3.25 10.5 0 0.25 3.25 10.5 0 0.25 3.25 10.5 0 0.25 3.25 10.5
Salinity (ppt)
AgNP-citrate
AgNP-PVP
TiO2NP
0 0 0 0 1 1 1 1 5 5 5 5 30 30 30 30
10.5 4.2 7.4 1.2 10.3 3.6 16.1 2.7 12.0 4.7 14.3 1.4 15.2 1.9 19.8 0.3 11.5 0.3 17.7 0.5 15.6 1.7 17 3.1 9.4 4.0 6.5 0.4 4.9 1.5 10.3 1.9
8.6 0.8 10.8 2.0 14.9 1.7 16.2 3.6 17.5 3.5 17.9 0.6 22.1 0.9 20.7 0.6 13.8 2.0 15.3 2.3 17.1 2.7 14.4 3.4 13.6 4.5 6.4 1.9 9.4 1.9 10.1 0.1
12.5 3.4 12.4 0.9 17.4 2.0 22.1 3.7 14.1 3.5 14.1 0.7 19.8 3.6 25.3 1.5 7.2 2.9 18.9 2.6 16.8 1.8 14.7 1.5 6.1 5.2 7.1 3.4 8.4 4.8 8.9 2.6
AgNP ¼ silver nanoparticle; PVP ¼ polyvinylpyrrolidone; TiO2NP ¼ titanium dioxide nanoparticle; ppt ¼ parts per thousand. Significant differences compared with 0 ppt treatment for a given nanoparticle and DOC concentration at p < 0.05 by one-way analysis of variance. Significant differences compared with 0 ppt treatment for a given nanoparticle and DOC concentration at p < 0.01 by one-way analysis of variance.
higher salinity (e.g., 30 ppt). It appeared that DOC has a greater stabilization effect on TiO2NPs than on AgNP-citrate and AgNP-PVP. The present study also indicates that the stability and aggregation of NPs in seawater can be regulated by changing the salinity and DOC concentrations. Because of the many processes affecting the stability of AgNP-citrate, AgNP-PVP and TiO2NPs in the marine environment, it is very likely that these NPs, if released into the environment, will have relatively brief residence times in the water column; instead, they are likely to aggregate and accumulate in the sediments. These processes will influence the exposure to and effects on benthic communities and thereby influence the ecological risk of NPs to these communities. The present study is critical to understanding and predicting the potential health and environmental risks of NPs and for developing informed regulations. SUPPLEMENTAL DATA
Tables S1–S2. Figure S1. (482 KB DOCX). Acknowledgment—H. Wang was supported by the US Environmental Protection Agency (USEPA) as a National Research Council postdoctoral research associate. The authors thank W. Boothman and B. Taplin at the USEPA (Atlantic Ecology Division, Narragansett, RI) for their assistance in the laboratory. The authors also thank A. Parks, R. McKinney, and K. Rocha for their insightful reviews of the draft manuscript. Mention of trade names of commercial products and companies does not constitute endorsement or recommendation for use. This report has been reviewed by the USEPA’s Office of Research and Development, National Health and Environmental Effects Research Laboratory, Atlantic Ecology Division, Narragansett, RI, and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the agency. This is USEPA ORD005072.
REFERENCES 1. Beaudrie CEH, Kandlikar M, Satterfield T. 2013. From cradle-to-grave at the nanoscale: Gaps in U.S. regulatory oversight along the nanomaterial life cycle. Environ Sci Technol 47:5524–5534.
Stability and aggregation of silver and titanium dioxide 2. Wijnhoven SWP, Peijnenburg WJGM, Herberts CA, Hagens WI, Oomen AG, Heugens EHW, Roszek B, Bisschops J, Gosens I, van de Meent D, Dekkers S, De Jong WH, van Zijverden M, Sips AJAM, Geertsma RE. 2009. Nano-silver—A review of available data and knowledge gaps in human and environmental risk assessment. Nanotoxicology 3(2):109–138. 3. El Badawy AM, Luxton TP, Silva RG, Scheckel CK, Suidan MT, Tolaymat TM. 2010. Impact of environmental conditions (pH, ionic strength, and electrolyte type) on the surface charge and aggregation of silver nanoparticles suspensions. Environ Sci Technol 44:1260–1266. 4. Domingos RF, Tufenkji N, Wilkinson KJ. 2009. Aggregation of titanium dioxide nanoparticles: Role of a fulvic acid. Environ Sci Technol 43:1282–1286. 5. Angel BM, Batley GE, Jarolimek CV, Rogers NJ. 2013. The impact of size on the fate and toxicity of nanoparticulate silver in aquatic systems. Chemosphere 93:359–365. 6. Ates M, Daniels J, Arslan Z, Farah IO. 2013. Effects of aqueous suspensions of titanium dioxide nanoparticles on Artemia salina: Assessment of nanoparticle aggregation, accumulation, and toxicity. Environ Monit Assess 185:3339–3348. 7. Ju-Nam Y, Lead JR. 2008. Manufactured nanoparticles: An overview of their chemistry, interactions and potential environmental implications. Sci Total Environ 400:396–414. 8. Domingos RF, Peyrot C, Wilkinson KJ. 2010. Aggregation of titanium dioxide nanoparticles: Role of calcium and phosphate. Environ Chem 7:61–66. 9. Saleh NB, Pfefferle LD, Elimelech M. 2010. Influence of biomacromolecules and humic acid on the aggregation kinetics of single-walled carbon nanotubes. Environ Sci Technol 44:2412–2418. 10. Keller AA, Wang H, Zhou D, Lenihan HS, Cherr G, Cardinale BJ, Miller R, Ji Z. 2010. Stability and aggregation of metal oxide nanoparticles in natural aqueous matrices. Environ Sci Technol 44:1962–1967. 11. Yang SP, Bar-Ilan O, Peterson RE, Heideman W, Hamers RJ, Pedersen JA. 2013. Influence of humic acid on titanium dioxide nanoparticle toxicity to developing zebrafish. Environ Sci Technol 47:4718–4725.
Environ Toxicol Chem 33, 2014
1029
12. McCarthy JF. 1989. Bioavailability and toxicity of metals and hydrophobic organic contaminants. In Suffet IH, McCarthy P, eds, Aquatic Humic Substances: Influence on Fate and Treatment of Pollutants. American Chemical Society, Washington, DC, pp 263–277. 13. Huynh KA, Chen KL. 2011. Aggregation kinetics of citrate and polyvinylpyrrolidone coated silver nanoparticles in monovalent and divalent electrolyte solutions. Environ Sci Technol 45:5564–5571. 14. El Badawy AM, Silva RG, Morris B, Scheckel KG, Suidan MT, Tolaymat TM. 2011. Surface charge-dependent toxicity of silver nanoparticles. Environ Sci Technol 45:283–287. 15. Hoheisel SM, Diamond S, Mount D. 2012. Comparison of nanosilver and ionic silver toxicity in Daphnia magna and Pimephales promelas. Environ Toxicol Chem 31:2557–2563. 16. Dobias J, Bernier-Latmani R. 2013. Silver release from silver nanoparticles in natural waters. Environ Sci Technol 47:4140–4146. 17. Millero FJ, Sohn ML. 1992. Chemical Oceanography. CRC Press, Boca Raton, FL, USA. 18. Kim JI, Buckau G, Li GH, Duschner H, Psarros N. 1990. Characterization of humic and fulvic acids from Gorleben groundwater. Fresenius J Anal Chem 338:245–252. 19. Li X, Lenhart JJ, Walker HW. 2010. Dissolution-accompanied aggregation kinetics of silver nanoparticles. Langmuir 26:16690–16698. 20. Jiang J, Oberdorster G, Biswas P. 2009. Characterization of size, surface charge, and agglomeration state of nanoparticle dispersions for toxicological studies. J Nanopart Res 11:77–89. 21. Chen KL, Mylon SE, Elimelech M. 2006. Aggregation kinetics of alginate-coated hematite nanoparticles in monovalent and divalent electrolytes. Environ Sci Technol 40:1516–1523. 22. Liu JY, Hurt RH. 2010. Ion release kinetics and particle persistence in aqueous nano-silver colloids. Environ Sci Technol 44:2169–2175. 23. Berry WJ, Cantwell MG, Edwards PA, Serbst JR, Hansen DJ. 1999. Predicting toxicity of sediments spiked with silver. Environ Toxicol Chem 18:40–48. 24. Cosgrove T. 2005. Colloid Science: Principles, Methods and Applications. Blackwell, Oxford, UK.
