Article pubs.acs.org/est
Effects of pH, Electrolyte, Humic Acid, and Light Exposure on the Long-Term Fate of Silver Nanoparticles Wei Zhou,†,‡ Yen-Ling Liu,‡ Audrey M. Stallworth,‡ Chunsong Ye,† and John J. Lenhart*,‡ †
School of Power and Mechanical Engineering, Wuhan University, Wuhan 430072, People’s Republic of China Department of Civil, Environmental and Geodetic Engineering, The Ohio State University, Columbus, Ohio 43210, United States
‡
Downloaded via UNIV OF WINNIPEG on June 22, 2018 at 15:04:57 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: We investigated the evolution in silver nanoparticle (AgNP) properties during a series of 10−50 day experiments on suspensions with different pH (5−9), electrolyte type (NaNO3 and NaCl) and concentration (2 and 6 mM), Suwannee River humic acid (SRHA) concentration (0−13.2 mg C/L), and light exposure (artificial sun light exposure for 8 h per day or dark). Of these factors, pH most influenced the AgNPs’ properties as it modifies surface charge as well as AgNP dissolution and oxidation and Ag+ reduction reactions. As a result, particle behavior differed in basic and acidic conditions. Trends with pH varied, however, based on the electrolyte and SRHA concentration. In the presence of chloride which forms AgCl(s), for example, we observed the particle size decreased with increasing pH. The opposite was observed in identical systems in NaNO3. This behavior was modified by SRHA, with increasing SRHA reducing dissolution and enhancing stability. Light exposure enhanced processes resulting in AgNP dissolution, resulting in higher dissolved Ag concentrations than under similar conditions in the dark. Overall, our results highlight how AgNP properties evolve over time and provide insight needed to confidently extend model system behavior to predict the environmental fate of AgNPs.
■
that particle properties will change.9,23 Extending the results of these ideal studies to environmental conditions with confidence will thus require an understanding of how the physicochemical properties of the particles change over time in environmental systems and how the resulting changes reflect exposure conditions. Concern regarding the release and transformation of AgNPs in the environment reflects the potential risk they pose to humans and other organisms.24 For example, in vitro studies report that AgNPs are toxic to a variety of organs including the lung, liver, brain, digestive system, vascular system, immune system, and reproductive organs.25 It is also reported that AgNP suspensions represent a potential source of toxicity for other organisms, such as zebra fish,26 lemma gibba plant cells,27 and Chlamydomonas reinhardtii.28 This research indicates that nanoparticle size, chemical composition, crystal structure, surface area as well as the release rate of ionic silver are expected to be important variables in determining toxicity. The fact that the toxicity varies so strongly with the form and properties of the particles29 further highlights the need to
INTRODUCTION Nanomaterials (NM) are increasingly being used to enhance the performance of medical, commercial, and industrial products such as pharmaceuticals, cosmetics, textiles, surface coatings, electronic components, and food packaging.1,2 Due to this increasing and widespread use, the release of NM into the environment during use is nearly certain.1,3,4 Owing to its unique optical, catalytic, sensing, and antimicrobial properties, nanosilver is currently one of the most commonly used engineered nanomaterials.5 According to Nowack et al.,6 more than 50% of the nanomaterials used for industrial purposes contain nanosilver, with an estimated global production of silver nanoparticles (AgNPs) of approximately 320 ton/year. AgNP release into the environment can occur during the production, transport, erosion, washing, and disposal of AgNP products.7 Upon release into the environment, AgNP fate in aquatic systems reflects processes such as aggregation, flocculation, redox reactions, dissolution, sulfidation, reactions with natural organic matter (NOM), and light exposure.8−15 These processes subsequently dictate AgNP fate, transport, and bioavailability as they control the AgNP’s physical and chemical properties such as size, surface charge, and nature of surface coating.16−22 There is extensive research regarding the influence of many of these factors on AgNP fate using freshly made or commercially available AgNPs; however, upon extended exposure to environmental factors it seems likely © 2016 American Chemical Society
Received: Revised: Accepted: Published: 12214
June 29, 2016 September 20, 2016 October 14, 2016 October 14, 2016 DOI: 10.1021/acs.est.6b03237 Environ. Sci. Technol. 2016, 50, 12214−12224
Article
Environmental Science & Technology
prepared to systematically investigate variations in pH, electrolyte, and NOM concentration. We also explicitly took the duration of light exposure into consideration by conducting experiments that alternated 8 h light exposure with 16 h dark exposure. The particle and suspension properties were investigated using a combination of dynamic light scattering (DLS), UV−vis absorption spectroscopy, electrophoretic mobility, and transmission electron microscopy (TEM). The release of silver was also monitored by measuring the dissolved silver concentration. As expected, pH dominated the properties of the particles over time due to its influence on particle dissolution and NOM adsorption. However, its influence was a function of the electrolyte type and NOM concentration, with particle behavior in systems with chloride reflecting the influence of AgCl(s) and those with NOM reflecting its ability to enhance particle stability at low pH. Interestingly, light irradiation had the smallest influence on changing particle properties. Overall, our results provide insight into how AgNP properties evolve as they “age” in aquatic systems and suggest under slightly basic pH conditions (e.g., pH 9) that properties evolve to the extent that studies done with fresh AgNPs may not accurately depict their true fate.
improve our understanding of the transformation of AgNPs in the environment. The behavior of AgNPs in the environment is sensitive to a variety of physical and chemical processes, with pH, electrolyte, NOM, and light exposure being among the most important.8,14,30−33 Changing pH influences the surface charge of AgNPs, which is a dominant factor in determining nanosilver fate. For example, the pH-dependent surface charge of carbonate-coated AgNPs becomes positive below pH 4, and as a result, increased agglomeration was observed.34 Similarly, the rate of silver release from AgNPs and extent of dissolution is pH-dependent and it is observed to decrease with increasing pH.34−36 This reflects the dependence in the oxidation of AgNPs on both protons and dissolved O2, with the oxidative dissolution of AgNPs correspondingly increasing as each of these increase.37 Chloride is an important anion in natural aqueous systems, and it exhibits a strong influence on AgNP fate as it factors into particle stability and dissolution. For example, in their study on AgNP stability Li et al.38 report that chloride reacts with dissolved silver resulting from AgNP dissolution to form a AgCl layer on the AgNPs. A low Cl/Ag ratio appears to enhance the stability of AgNPs because it inhibits their dissolution; however, at high Cl/Ag ratios dissolution is enhanced due to the formation of soluble Ag−Cl complexes.33,39 Because Ag toxicity is often governed by the amount of dissolved Ag in solution, the amount of chloride will also have a strong effect on the toxicity of AgNPs.40 NOM is a heterogeneous mixture of natural macromolecules that is known to be redox-reactive.41 This property of NOM is critical for the cycling of many elements in the environment, such as Fe3+, Hg2+, and I2.42 In the case of AgNPs, NOM is observed to reduce Ag+ to form AgNPs,43,44 with higher NOM concentrations enhancing this process.14 In addition, NOM adsorption to the surface of AgNPs can enhance their stability as the adsorbed organic molecules can induce repulsive steric (or electrosteric) interactions.8 Exposure of AgNPs to light produces a variety of outcomes.14,45−49 For example, AgNPs are broken into fragments by light exposure as the photoejection of electrons induces a positive charge on the AgNPs that results in their disintegration into smaller-sized particles.50 In addition, light irradiation enhances silver release as it oxidizes the surface to form an oxide layer that exhibits greater solubility than metallic silver15,45 The reduction in size50 or transformation of capping layers51 upon exposure to light can also alter surface properties leading to enhanced aggregation. In the presence of NOM, photoreduction of Ag+ by NOM inhibits the dissolution of AgNPs and is observed to enhance particle stability.51 Thus, AgNP fate appears to be strongly influenced by light irradiation. Although studies on AgNP fate in the aqueous environment are extensive, the emphasis in most cases is on short-term studies, such as particle aggregation.31,32,52−54 Studies with light exposure also exist, but they tend to utilize continuous exposure over extended periods of time (e.g., 48 h) as opposed to the alternating light and dark exposure conditions that occur in the environment.50,55,56 In most cases, the particles employed are also freshly prepared8,54,57,58 or sourced commercially.59−61 Thus, in order to extend the results of these studies and many others to evaluate AgNP fate, an understanding of how the particle properties evolve in the environment is required, as this evolution in the properties may alter particle fate. In this study, we conducted a series of long-term experiments using samples
■
MATERIALS AND METHODS Materials. Silver nitrate (99.8%) and D-maltose (99%) were purchased from Sigma-Aldrich in the form of a powder and were used without further purification. Trace metal grade ammonium hydroxide was purchased from Fisher Scientific. Suwannee River humic acid standard II (SRHA) was purchased from the International Humic Substances Society (IHSS, www. humicsubstances.org). All other reagents were analytical grade or better. The deionized water used in the experiments was supplied from a deionized water system (Milli-Q, Millipore) with a resistivity of 18.2 MΩ·cm. All solutions used to synthesize AgNPs were filtered through 0.1 μm cellulose ester membranes (Millipore) prior to use. Solutions containing NaCl, NaNO3, NaHCO3, NaOH, HNO3, and SRHA used to adjust chemical conditions in the experimental samples were filtered through 0.45 μm PTFE membranes (Sartorius) before use. The labware and glassware used in the experiments were washed with 10% nitric acid, rinsed thoroughly with deionized water, oven-dried, and stored prior to use under dust-free conditions. Silver Nanoparticles. This study utilized silver nanoparticles synthesized by the reduction of the Ag(NH3)2+ complex with D-maltose following Li et al.38 Details of particle synthesis are summarized in the Supporting Information. Particle hydrodynamic diameter was measured by DLS (90Plus, Brookhaven Instruments Corp., Holtsville, NY). The UV−vis spectra were collected over the wavelength range of 200−700 nm using a Shimadzu UV-4201PC UV−vis spectrophotometer. Particle morphology and size were observed with a TEM (Tecnai G2 Spirit, FEI) adjusted to 80 kV. Samples for TEM were prepared by placing one drop (ca. 12.5 μL) of the working suspension on a 200 mesh copper grid. The sample was dried under flowing nitrogen after extra water was removed by wicking with filter paper. The total silver concentration of the AgNP stock suspensions was determined by digesting the suspensions at a ratio of 9 mL of concentrated HNO3 and 1 mL of stock AgNP solution in a microwave digestor (Milestone Inc.) at 180 ± 5 °C for 9.5 min.62,63 The digested solutions were then diluted with DI water to a known volume and HNO3 concentration of ca. 2% before being 12215
DOI: 10.1021/acs.est.6b03237 Environ. Sci. Technol. 2016, 50, 12214−12224
Article
Environmental Science & Technology Table 1. Summary of Experimental Conditions expt
pH
SRHA concn (mg C/L)
A B C D E
5 and 8.5 7 5 and 9 7 7
4.4 and 13.2 0, 6.6, and 13.2 0, 2.2, and 6.6 0, 2.2, and 6.6 0−13.2
electrolyte 2 2 6 2 2
mM mM mM mM mM
NaCl NaCl vs 2 mM NaCl + 4 mM NaNO3 NaNO3 NaCl vs 2 mM NaNO3 NaCl
light condition dark dark dark dark dark
vs vs vs vs
light light light light
duration (days) 50 20 50 50 10
Figure 1. Change in hydrodynamic radius over time for the AgNPs under dark conditions in (a) 2 mM NaCl as a function of pH (5 and 8.5) and SRHA (4.4 and 13.2 mg C/L), (b) 6 mM NaNO3 as a function of pH (5 and 9) and SRHA concentration (0, 2.2, 6.6 mg C/L), and (c) 2 mM NaCl or 2 mM NaNO3 at pH 7 as a function of SRHA concentration (0, 2.2, 6.6 mg C/L). The conditions in panel d are the same as those in panel a except that the samples were exposed to light. The lines are presented to guide the eye.
measured by an inductively coupled plasma optical emission spectrometer (ICP-OES, Vista AX, Varian, Inc.). The total dissolved silver concentration was measured by ICP-OES following centrifugal ultrafiltration of the samples using Amicon Ultra-4 centrifugal filtration cartridges with the 10 kDa membrane (Millipore). Samples were centrifuged for 10 min at 7500g using a LegendRT centrifuge (SORVALL, Inc.). The samples were preserved following centrifugal ultrafiltration through the addition of 2% HNO3. The electrophoretic mobility of the AgNPs was measured at a temperature of 22 ± 0.5 °C with a Brookhaven Instruments ZetaPALS. Error estimates were based on 10 measurements conducted on two separate samples for each set of experimental conditions. Preparation of Experimental Samples. A series of samples was prepared in polycarbonate bottles in order to systematically evaluate changes in the properties of the AgNPs under long-term exposure scenarios as a function of pH,
electrolyte type, electrolyte concentration, and SRHA concentration (see Table 1). Aliquots of 0.05 M NaHCO3, 0.025 M NaOH, and 0.025 M HNO3 were added as needed to adjust the pH value to the desired target and facilitate equilibration with atmospheric CO2. The humic acid stock solutions (220 mg C/ L) were prepared by dissolving 250 mg of SRHA in 500 mL of Milli-Q water. These solutions were stirred for 12 h in the dark, filtered with 0.45 μm PTFE membranes (Sartorius) to remove any particulate material, and adjusted to a pH value of 7.0 ± 0.1 using 0.025 M trace metal grade NaOH (Fisher Scientific). The dissolved organic carbon (DOC) concentration was determined with a Shimadzu TOC-VCSN analyzer. Aliquots of the stock SRHA were added to make the final experimental SRHA concentration. The total Ag concentration was constant across all samples at 2.78 ± 0.21 mg/L, which included an initial dissolved silver concentration of approximately 0.030 mg/L based upon the addition of appropriate volumes of the stock AgNP suspensions that had a total Ag concentration of 69.5 ± 12216
DOI: 10.1021/acs.est.6b03237 Environ. Sci. Technol. 2016, 50, 12214−12224
Article
Environmental Science & Technology
depicted in Figure 1a, a shorter duration experiment was conducted using systems at a constant chloride concentration and pH with varying SRHA and electrolyte concentration (see experiment B in Table 1). Over the 20 day time period of this experiment, differences in the particle diameter for the 2 and 6 mM electrolyte systems were little, on the order of 1 nm (Figure SI 2). This small change in size paled in comparison to the size differences observed in systems with and without SRHA, and thus, the size differences in Figure 1a appear to primarily reflect the influence of pH and SRHA concentration. Systems prepared solely in NaNO3 (experiment C in Table 1) exhibited trends that were opposite with those observed for systems with NaCl. Prior results by Li et al.68 evaluating the aggregation of AgNPs reported differing behavior in systems with different electrolyte anions, suggesting our observations reflect innate differences in the chemistry of chloride and nitrate in these systems. At elevated pH, the particle size in our systems with NaNO3 remained constant; however, at low pH the size decreased (Figure 1b). The decrease in size at pH 5 was suppressed as the concentration of SRHA increased as the particle diameter decreased 15, 8, and 5 nm for SRHA concentrations of 0, 2.2, and 6.6 mg C/L, respectively. At pH 9, changes in the particle diameter exhibited similar, albeit smaller, variations with SRHA concentrations increasing 2, 1, and 0.3 nm at SRHA concentration of 0, 2.2, and 6.6 mg C/L. Results collected at pH 7 (see experiment D in Table 1) were similar to those observed at pH 5 in that the change in particle size for AgNPs in NaNO3 was greater than in NaCl (Figure 1c). The results also showed the same dependence on SRHA concentration, with the change in particle size decreasing as the SRHA concentration increased. That the changes in particle size in systems with increasing SRHA were muted at neutral and acidic pH values likely reflects that SRHA adsorbed to the AgNP surfaces more readily at those pH values than it did at basic pH.38 Changes in particle size in systems equilibrated with NaCl exposed to light were generally similar in trend to those in systems that were in the dark in that particles tended to decrease in size and that increasing SRHA concentration muted this size decrease (Figure 1d). When compared to changes observed in the dark (Figure 1a), however, the magnitude of change in systems exposed to light was larger. For example, at pH 8.5 the decrease in size for AgNPs in 2 mM NaCl and 13.2 mg C/L SRHA was 20 nm when exposed to light (Figure 1d) compared to 9 nm in the dark (Figure 1a). For systems in NaNO3, changes in AgNP size when exposed to light exhibited marked dependence on SRHA concentration. In the presence of SRHA, the trends were roughly similar to those in the dark in that the particle size decreased over time (e.g., compare Figure SI 3a and Figure 1b). However, the overall change in particle size for the systems exposed to light was lower than those in the dark, and except for at pH 9 (Figure SI 3a), there was little influence on particle size with increasing SRHA (e.g., Figure SI 3b). These trends in changing particle size were also quite lower than those observed for the NaCl systems as well. In the absence of SRHA, the change in particle size was quite variable and exhibited a strong dependence on pH. At pH 5, the AgNP size increased roughly 600 times to ca. 40 μm in the presence of light by the second day of the experiment. As time continued, the size decreased, and by day 25 the particle size decreased to roughly the starting value (Figure SI 3c). Similar, albeit smaller, results were observed at pH 7 (Figure SI 3d). For comparison, in dark
5.2 mg/L. This concentration was similar to that used in other studies (e.g., 1.3 mg/L total silver used by Li and Lenhart,64 2 mg/L total silver used by Liu and Hurt,37 and 8 mg/L total silver used by Peretyazhko et al.36), but higher than that anticipated to occur in environmental systems.6 The electrolyte type and concentration was reached using stock 1 M NaCl and 1 M NaNO3 solutions. The samples were equilibrated on an orbital shaker at 30 rpm and exposed to artificial sun light from a 150 W Xe ozone-free lamp (model 6255, Newport Inc.) 8 h each day. The light spectrum from the lamp was corrected by a filter (Global filter Air Mass 1.5/model 81094, Newport Inc.) to provide a spectrum of light that approximated that of real sun light. See Li et al. for additional details of the light exposure setup.64 Sample locations were systematically changed during the course of the experiment to ensure each sample got similar levels of light exposure. Light condition in this paper refers to 8 h of light exposure per 24 h which approximated natural conditions.65 The dark controls comprised an identical set of samples that were covered with aluminum foil that excluded light. The aging experiments were carried out in room temperature (22 ± 2 °C) over a time frame of 10−50 days. Samples for each experiment were prepared in duplicate, and thus, the results represent the average of these two samples.
■
RESULTS Characterization of AgNPs. The properties of the freshly prepared AgNPs were consistent with those previously reported for particles synthesized in this manner.38 The average hydrodynamic diameter was measured using DLS as 76.7 ± 1.4 nm. TEM imaging indicated the particles were roughly spherical and monodisperse with a size similar to that determined using DLS (Figure SI 1a). The UV−vis spectrum of the particles was characterized by the presence of the surface plasmon band characteristic of these particles at a wavelength of 440.7 ± 1.8 nm (Figure SI 1b), which is indicative of an oxide surface coating.66 Particles prepared in this manner are stable over extended periods of time without the addition of additional capping agents38,67 as the oxide layer and potentially residual amounts of maltose or its oxidation products are sufficient to impart stability. Change in Particle Size. The size of the particles changed over the course of the experiments as a consequence of AgNP dissolution, aggregation, and secondary phase precipitation. This change in size was a function of system pH, SRHA concentration, electrolyte type, and light exposure. For example, in systems equilibrated under dark conditions with a constant chloride concentration of 2 mM (experiment A in Table 1), we observed over a period of 50 days at pH 5 that the hydrodynamic diameter of particles increased from 71 to 75 nm (Figure 1a). Under similar conditions at pH 8.5, however, the particle size decreased. The decrease in size at pH 8.5 was a function of SRHA concentration, with the observed decrease in size increasing in magnitude from 3 nm in 4.4 mg C/L to 9 nm in 13.2 mg C/L SRHA. The same cannot be said for these systems at pH 5, where the results did not exhibit any dependence on SRHA concentration. Although the chloride concentration in experiment A was constant, the total electrolyte concentration varied as a result of the pH adjustment, and at pH 8.5 it was approximately 6 mM compared to 2 mM at pH 5. Li et al.38 suggest altering electrolyte concentration can result in AgNP dissolution. To establish whether the slight variation in electrolyte concentration in experiment A contributed to the particle size changes 12217
DOI: 10.1021/acs.est.6b03237 Environ. Sci. Technol. 2016, 50, 12214−12224
Article
Environmental Science & Technology
Figure 2. Change in dissolved silver over time for the AgNP systems as a function of SRHA concentration (0, 2.2, and 6.6 mg/L) at (a) pH 5 and 6 mM NaNO3 in the absence of light, (b) pH 5 and 6 mM NaNO3 in the presence of light, (c) pH 7 and 2 mM NaNO3 in the absence of light, and (d) pH 7 and 2 mM NaNO3 in the presence of light. The lines represent the fits to the data with the pseudo-first-order kinetic model (eq 6).
conditions the change in size exhibited a relatively consistent decreasing trend with increasing time (see Figure 1, parts b and c). Samples exposed to light at pH 9 did not exhibit any increase in size, as the AgNP size decreased from 75 to 53 nm (Figure SI 3e). The decrease in size was not uniform with time as the size was relatively stable for the first 20 days of the study, which roughly correlated to the time period where we observed large size changes for this experimental condition at pH 5 and 7. Change in Dissolved Silver Concentration. The dissolved silver concentration exhibited time-dependent trends that reflected the influence of system pH, electrolyte type, SRHA concentration, and light exposure. The initial (t = 0) dissolved silver in all experiments was low (
