at SciVerse ScienceDirect

Environmental Pollution 182 (2013) 141e149

Contents lists available

Environmental Pollution

journal homepage: www.elsevier .com/locate/envpol

Behavior of Ag nanoparticles in soil: Effects of particle surface coating,aging and sewage sludge amendment

Annie R. Whitley a,b, Clément Levard b,c, Emily Oostveen a,b, Paul M. Bertsch a,b,Chris J. Matocha a, Frank von der Kammer d, Jason M. Unrine a,b,*

aUniversity of Kentucky, Department of Plant and Soil Sciences, Lexington, KY 40546, United StatesbCenter for the Environmental Implications of Nanotechnology, Duke University, Durham, NC 27708, United StatescCEREGE, UMR 7330, CNRS-Aix Marseille University, BP 80, 13545 Aix en Provence, FrancedDepartment of Environmental Geosciences, Vienna University, Althanstrasse 14, A-1090 Vienna, Austria

a r t i c l e i n f o

Article history:Received 18 March 2013Received in revised form29 May 2013Accepted 5 June 2013

Keywords:AgricultureBiosolidsDetectionNanotechnologyWeathering

* Corresponding author.E-mail address: jason.unrine@uky.edu (J.M. Unrine

0269-7491/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.envpol.2013.06.027

a b s t r a c t

This study addressed the relative importance of particle coating, sewage sludge amendment, and agingon aggregation and dissolution of manufactured Ag nanoparticles (Ag MNPs) in soil pore water. Ag MNPswith citrate (CIT) or polyvinylpyrrolidone (PVP) coatings were incubated with soil or municipal sewagesludge which was then amended to soil (1% or 3% sludge (w/w)). Pore waters were extracted after 1 weekand 2 and 6 months and analyzed for chemical speciation, aggregation state and dissolution. Ag MNPcoating had profound effects on aggregation state and partitioning to pore water in the absence ofsewage sludge, but pre-incubation with sewage sludge negated these effects. This suggests that Ag MNPcoating does not need to be taken into account to understand fate of AgMNPs applied to soil throughbiosolids amendment. Aging of soil also had profound effects that depended on Ag MNP coating andsludge amendment.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction

The growth of nanotechnology has raised public concern aboutpotential environmental and human health effects of manufacturednanoparticles (MNPs) released to the environment (Helland et al.,2006; Wiesner et al., 2006). Production of consumer productscontaining MNPs continues to increase despite the lack of sufficientknowledge concerning how they may affect the environment(Luoma, 2008; Project on Emerging Nanotechnologies, 2008). Thelack of detection and in situ characterization capabilities fornanomaterials in complex biological and environmental matricesalso hinders the development of regulations for MNPs in theenvironment (von der Kammer et al., 2012; Weinberg et al., 2011).One major class of MNPs currently being used is Ag MNPs, due totheir antimicrobial properties (Levard et al., 2012). Ag MNP con-taining products, such as paint and textiles have already beenshown to release Ag MNPs through normal use (Benn andWesterhoff, 2008; Kaegi et al., 2010).

).

All rights reserved.

In some instances particles may directly enter terrestrial envi-ronments as they are shed from Ag MNP containing products suchas paints (Kaegi et al., 2010). However, a large fraction of Ag MNPsare predicted to enter wastewater treatment plants (WWTP) viasewage streams where they are likely to efficiently partition to thesewage sludge and be sulfidized (Gottschalk et al., 2009; Kaegi et al.,2011; Kim et al., 2012; Lombi et al., 2013). In the United States andelsewhere, the majority (about 60% in the US) of sewage sludge isapplied to agricultural lands as biosolids (EPA, 1995). Because ofthis, we expect agricultural soils to be a major repository for MNPsand a source of Ag MNPs to aquatic environments through erosionand runoff. It has been shown that aggregation and dissolutionbehavior of Ag MNPs can have important implications for envi-ronmental fate and toxicity (Bone et al., 2012; Unrine et al., 2012).

Ag MNP behavior in soil has not been widely investigated(Coutris et al., 2012; Sagee et al., 2012), in part due to difficultiesassociated with tracking MNPs in the complex soil matrix (Blaseret al., 2007; Kaegi et al., 2011). Conclusions drawn from aquaticstudies may have limited relevance to terrestrial systems. Behaviorof MNPs in soil may differ from aquatic systems due to the uniquebiological, physical and chemical characteristics of soil that differfrom aqueous systems. These characteristics vary widely amongsoil types and through time and space. For example, aggregation of

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149142

the particles onto the solid phase of the soil is a complicating factorthat cannot be taken into account in simple aqueous studies. To ourknowledge, few previous studies have been published that attemptto characterize AgMNP aggregation and dissolution behavior in soilpore water. A recent study investigated Ag MNP dissolution in soil(Cornelis et al., 2012); however, this study did not investigate therole of particle coating, aging or sludge amendment.

Changes in Ag MNP behavior due to modification by themanufacturer (e.g., surface coating) or transformations in theenvironment via contact with naturally occurring minerals ororganic matter (NOM), as well as other ligands such as sulfide orchloride further increases difficulties associated with assessing therisk of MNPs to human health and the environment. Differences insurface coating alone can affect Ag MNP aggregation and dissolu-tion behavior under differing environmental conditions (Bone et al.,2012; El Badawy et al., 2010; Kittler et al., 2010; MacCuspie, 2011;Unrine et al., 2012). Likewise, environmental constituents such asNOM have been shown to promote particle stability for both MNPsand naturally occurring particles (King and Jarvie, 2012), in somecases by coating the particle surface (Bertsch and Seaman, 1999;Fabrega et al., 2009; Hotze et al., 2010; Zhang et al., 2009). Inwastewater treatment plants, Ag MNPs are sulfidized whichsignificantly reduces Ag solubility and results in decreased toxicity(Hirsch, 1998; Kaegi et al., 2011; Kim et al., 2010; Ratte, 1999;Reinsch et al., 2012). However, little is known of Ag MNPbehavior following application of sewage sludge to agriculturalsoils including the influence of sulfidation and surface coating.

The objective of this study was to determine the aggregationand dissolution behavior of AgMNPs in soil porewater as a functionof surface chemistry, sewage sludge biosolids pre-incubation andamendment rate as well as soil aging time. To observe effects of AgMNP surface coating and biosolids pre-incubation, we aged soilscontaining Ag MNPs having different surface coatings and undercontrolled laboratory conditions with various incubation timesbetween 1 week and 6 months. The Ag MNPs were introduced tothe soil either directly or through amendment of sewage sludgecontaining the Ag MNPs which were pre-incubated for 1 week.Previous studies have shown rapid transformation (within mi-nutes) of AgNPs both when incubated with processed biosolids(Colman et al., 2013) or in pilot wastewater treatment plants (Lombiet al., 2013). We expected differences in surface coating to result indissimilar Ag MNP behavior, with sterically stabilized poly-vinylpyrrolidone (PVP) Ag MNPs being more stable against aggre-gation than the low molecular weight organic acid (citrate, CIT)coated particles which would be subject to removal of coatingthrough desorption as well as screening of surface charge bydivalent cations in soil solution. Further, we expected sulfidation tonegate the effects of manufactured coating (Levard et al., 2011;Lombi et al., 2013) within pore waters and aging to yield adecline in pore water Ag because of the very low solubility of Ag2S.We adapted techniques developed and used in our previousresearch to track the chemical and physical changes of Ag MNPs inthe soils including asymmetrical flow field-flow fractionation (AF4)(Unrine et al., 2012) and extended X-ray absorption fine structurespectroscopy (EXAFS) (Shoults-Wilson et al., 2011a,b; Unrine et al.,2012). We extracted pore water from soils and analyzed them usingAF4 coupled to static/dynamic multi-angle laser light scattering(MALLS/DLS) and an inductively coupled plasma mass spectrom-eter (ICP-MS) to assess Ag particle size distribution. Ultracentrifu-gation followed by ICP-MS analysis was also used to assess theproportion of dissolved Ag in pore waters. EXAFS of soil solids andtransmission electron microscopy-energy dispersive spectroscopy(TEM-EDS) of AF4 fractions were used to examine chemical speci-ation and physical form of Ag in soils and soil pore water,respectively.

2. Materials and methods

2.1. Silver nanoparticles synthesis and characterization

Two types of Ag MNPs were used having differing surface coatings. First, 60 nmnominal diameter polyvinylpyrrolidone (PVP) Ag MNPs were synthesized as pre-viously described (Cheng et al., 2011). We also used 60 nm nominal diameter citrate(CIT) coated Ag MNPs made via reduction of AgNO3 by boiling in sodium citrate(Turkevich et al., 1951). Primary particle size and shape was examined usingtransmission electron microscopy (TEM) using a Jeol 2010 F field emission gunelectron microscope (Tokyo, Japan). AgNO3 was used to compare the MNP treat-ments with ionic Ag behavior. Size distributions were verified in triplicate for bothtypes of Ag MNPs and soil nanoparticles via batch dynamic light scattering (DLS)with a Malvern Zeta-Sizer Nano-ZS (Malvern Instruments, Malvern, UK). Intensityweighted (Z-average, estimated using cumulants analysis) hydrodynamic diameters(dh) are reported (Koppel, 1972). Electrophoretic mobility was also determined at pH6 in 18 MU deionized water using phase analysis light scattering (PALS).

2.2. Soil preparation and aging

Yeager sandy loam (YSL) soil from Estill County, KY was air-dried and sievedto < 1 mm. This soil has been thoroughly characterized with respect to pH,composition, and cation exchange capacity (Table S-1; Shoults-Wilson et al., 2011a).Methods for determination of soil field capacity and results are outlined in theSupplemental information (SI). Aerobically digested sewage sludge was obtainedfrom a municipal wastewater treatment facility (Winchester, KY) and was nottreated any further (such as with liming or composting). Chemical composition ofthe sludge is reported in Table S-2.

Sludge was spiked with Ag MNP suspensions or AgNO3 solution to obtain a finalconcentration of 200 mg Ag kg�1 solid (soil þ sludge) when combined with soil.Therefore, the sludge was spiked to concentrations of 20,000 or 6666 mg Ag/kg drysludge. These are unrealistically high Ag concentrations; however, based on pre-liminary estimates, a concentration of 200 mg/kg would assure us that at the end ofsix months it would be possible to detect and characterize the Ag NPs in pore water,while also keeping the actual sludge amendment rate and soil organic carboncontent at realistic levels. These soil Ag MNP concentrations are also similar to whatis needed to elicit adverse effects in soil invertebrates in this soil (Shoults-Wilsonet al., 2011a, 2011b, 2011c). It is important to note that this was a one time-amendment, and after repeated amendments in the field, Ag could accumulateover time. The Ag MNPs were synthesized in the colloidal phase and never dried, sono dispersion step was necessary. Spiked sludge was incubated for 1 week in anenvironmental chamber at 20 �C and rewetted daily to maintain 80% moisture.Sludge was then combined and mixed with a wooden stick with 10 g YSL to achieveeither 1 or 3% sludge dry mass. These amendment rates fall within the range ofannual biosolids loading rates for a variety of agricultural applications in the U.S. andwould be equivalent to 22 or 67 dry t/ha assuming a tillage depth of 17 cm (EPA,2000). Soil samples without sludge were spiked directly with 200 mg Ag kg�1 soilfor each treatment. Soil mixtures were prepared in 20 mL glass scintillation vials.Blank samples consisted of YSL at 0, 1, and 3% sludge with no Ag addition. Sampleswere maintained at 19% (v/w) moisture content in an environmental chamber at20 �C in the dark for 1 week, 2months or 6months. Soils were rewetted as necessaryevery three days to maintain constant moisture content. Three replicates wereincluded for each Ag treatment at each time point.

2.3. Soil pore water extraction and dissolved Ag measurements

To extract sufficient soil pore water for analysis, we added a volume of 18 MU

deionized (DI) water equivalent to 2.5� the moisture content of the soils to obtain asaturated paste (the point at which all pore space is occupied with water but there isno free water (Rhoades, 1982)). The saturated paste was added to a 20 mL syringeplugged with borosilicate glass wool which had been pre-moistened with DI water.The syringe was suspended from the top of a 50 mL centrifuge tube and centrifuged8 min (25 �C) at 1000 rpm to allow the pore water to elute into the centrifuge tube.The obtained fraction is procedurally defined pore water and will subsequently bereferred to simply as “pore water”. Pore water extracts were filtered with 30 mm,1.0 mmborosilicate glass fiber syringe filters (GE Osmonics, Fairfield, CT, USA) prior toanalysis. This step was required to remove particles larger than 1 mm prior to AF4analysis to avoid steric inversion where particles larger than 1 mm elute in thereverse order of particles smaller than 1 mm (Fedotov et al., 2011). Recovery of AgMNPs and AgNO3 through the pore water extraction apparatus and 1 mm filter weredetermined by passing a solution/suspension (either in DI water or soil pore watersfrom each of the treatments) of known Ag concentration through. Total Ag in porewaterwas determined by digesting samples in 7.5M concentrated trace-metal gradeHNO3 followed by dilution and ICP-MS (Agilent Technologies 7500cx; Santa Clara,CA, USA) analysis. Total organic carbon (TOC) and nitrogen (TN) in pore waters weredetermined using a FlashEA 1112 elemental analyzer (ThermoFisher Scientific Inc.,Waltham, MA, USA). Anion (Fl�, Cl�, NO2

�, Br�, NO32�, PO4

�, and SO4�) concentrations

in pore waters extracted from blank soil samples (0, 1, and 3% sludge) and sludgewere measured at experiment start and after aging using a Metrohm 792 Basic ion

Table 1Linear combination fits of model compounds to EXAFS spectra from soil sludgemixtures containing either AgNO3, polyvinylpyrrolidone coated Ag MNPs (PVP) orcitrate coated Ag MNPs (citrate).

AgNO3 0% AgNO3 3% PVP 0% PVP 3% Citrate 0% Citrate 3%

Ag (0) 0.13 0.12 1.00 0.30 1.00 0.22Ag2S 0.33 0.52 0.70 0.78AgCl 0.30Ag-acetate 0.25Ag-GSH 0.36R-value 0.094 0.095 0.044 0.057 0.061 0.054

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149 143

chromatograph (Herisau, Switzerland) having a MetroSep RP guard disc and aMetroSep A column. The mobile phase was 3.2 mmol L�1 Na2CO3 and 1 mmol L�1

NaHCO3. Cations (Naþ, Mg2þ, Al3þ, Kþ, Ca2þ, Mn2þ, Fe) were also determined via ICP-MS analysis. Pore water pH was determined immediately after 1 mm filtration.

After pore water extraction, 200 mg dried sample soil was digested in 9 mLtrace-metal grade HNO3 and 3 mL HCl using a MARS Express microwave digestionsystem (CEM, Matthews, NC) according to EPA method 3052 (EPA, 1996). Total Agin the digestates was determined using ICP-MS following EPA method 6020 (EPA,1998) including blanks, duplicate digestions, and standard reference materials(SRM 2711a Montana II soil and 2781 Domestic sludge, National Institute ofStandards and Technology). For dissolved Ag measurements, samples werecentrifuged at 239,311 � g for 90 min in order to remove Ag particles larger than1 nm. These parameters would also have removed Ag ions bound to DOM with ahydrodynamic radius larger than 7e10 nm and clay or other mineral particleslarger than 4e5 nm. Ultrasupernatants were acidified 0.15 M HNO3 to preserve foranalysis.

2.4. Statistical analyses

For statistical comparison among Ag concentrations we tested homoscedasticityusing the Bartlett test and Normality using the ShapiroeWilk test. Since the datawere non-normally distributed and had nonhomogeneous variance we used theKruskaleWallace test to determine differences in mean Ag concentrations. Indi-vidual differences among means were determined with the Wilcoxon RankeSumtest (p < 0.05).

2.5. AF4-ICP-MS analysis

An asymmetrical flow field-flow fractionation (AF4) system was used toseparate particles based on hydrodynamic radius (Wyatt Eclipse 3, Santa Barbara,CA, USA). All samples were analyzed within 12 h of extraction because extendedstorage at 4 �C beyond this time led to decreased intensity of the Ag signal in thefractograms, likely resulting from the aggregation of particles. Parameters used forAF4 are shown in Table S-3. The eluent from the AF4 channel entered an on-lineMALLS/DLS detector (Wyatt DAWN HELEOS-II) which measured light scatteringintensity at 18 angles with DLS measured at 100.3� and an ICP-MS system used forelement specific detection. Masses monitored on the ICP-MS included Ag (m/z ¼ 107, 109), Al (m/z ¼ 27), Fe (m/z ¼ 56), Mn (m/z ¼ 55), and Si (m/z ¼ 28). WyattASTRA version 5.3.4.11 was used to process light scattering data and Agilent ICP-MS chromatographic software version C.01.00 was used to process elementaldistribution fractograms collected via the ICP-MS. A flow splitter diverted aportion of the sample flow to waste to reduce the eluent flow rate to the optimalflow rate for the ICP-MS nebulizer (0.25 mL min�1). The portion diverted to thewaste was also diverted to a fraction collector (Agilent 1200 series), with fractionscollected for additional analyses including TEM. For TEM analysis, particles fromAF4 fractions were deposited onto lacey carbon coated Cu TEM grids placed on3 kDa ultrafiltration membranes within centrifugal filtration devices. The deviceswere centrifuged allowing the solutes to pass through the ultrafiltration mem-brane, while particles were deposited onto the grid. Dried grids were analyzed byTEM-EDS.

2.6. Validation of AF4-separations

To calibrate AF4 separation we analyzed bovine serum albumin (MW 66,463;7 nm dh, Sigma), National Institute of Standards and Technology (NIST) traceablepolystyrene latex spheres (20, 46 and 200 nm diameters, Thermo Scientific), alcoholdehydrogenase (MW 150,000; 9.2 nm dh, Sigma), reference Au nanoparticles (22 nmnominal diameter, Nanocomposix, San Diego, CA; 30 and 60 nm nominal diameters,NIST Standard Reference Materials (SRM) 8012 and 8013) and Au particles (80 and98 nm diameter, British Biocell International, Cardiff, United Kingdom). Calibrationcurves of retention time versus diameter of reference particles were used todetermine particle size. Pore water particle sizes were also validated using DLS/MALLS. Geometric diameters calculated using DLS/MALLS were compared to cali-brated values to provide independent cross validation that the separation wasworking properly. Sewage sludge particles strongly absorbed at 658 nm (thewavelength of the MALLS laser) so data from DLS/MALLS could not be fit to particlesize models for soils amended with 1 and 3% sludge and TEM was instead used toconfirm calibrated particle sizes for samples containing sludge.

2.7. EXAFS

Following extraction of pore waters, 1 week soils treated with AgNO3, PVP-AgMNP and CIT-Ag MNP and soils amended with 3% sludge and treated likewisewere analyzed by synchrotron-based extended X-ray absorption fine structure(EXAFS) spectroscopy using linear combination fits of model compounds to deter-mine speciation of Ag retained in the soils. Details of the EXAFS data collection andanalysis can be found in the SI.

3. Results

3.1. Initial particle characterization

The intensity weighted (Z-average) hydrodynamic diameters ofthe PVP-Ag MNPs and CIT-Ag MNPs in 18 MU DI water were84.4 � 0.5 and 51.4 � 0.1, respectively. Zeta potentials of PVP-AgMNPs and CIT-Ag MNPs were �24.7 � 14.4 mV and�101.0 � 17 mV (Hückel approximation) at pH 6.0 in DI water,respectively. The primary particle size of PVP and CIT-Ag MNPsdetermined by TEM were 53 � 14 and 84 � 24 (errors representstandard deviation of the particle size distribution) for PVP-AgMNPs and CIT-Ag MNPs, respectively (Fig. S-1,2).

3.2. EXAFs and pore water chemistry

Pore water chemistry and Ag speciation in soils were largelyaltered following the addition of sludge to soil. Pore waters fromsludge amended soils had increased amounts of Naþ, Ca2þ, Cl�, Br�,SO4

2�, and most notably Kþ and PO42� (Table S-4,5). Pore water

organic carbon concentration also increased (Fig. S-3). Agingresulted in increased concentrations of SO4

2�, Mg2þ, Ca2þ, and Fe2þ,decreased Cl- and decreased organic carbon content. On average,pore waters from 0%, 1%, and 3% sludge amended soils had 2.0, 5.3and 5.6 g L�1 total suspended solids, respectively. Due to limitationson X-ray beam time available for data acquisition, we only analyzed1 week aged samples from the 0 and 3% sludge amendments, to getan idea of the degree of initial chemical transformation of theparticles after interaction with sewage sludge. The best linearcombination fits (LCF) of EXAFS data (Table 1, Fig. S-5) suggests thatAg speciation in AgNO3 treated soil without sludge contains Ag2S(33%), AgCl (30%), Ag-acetate (25%), and Ag metal (13%). The best fitfor soils amended with 3% sludge treated with AgNO3 has anincreased proportion of Ag2S (52%) as well as Ag-glutathione (36%)that was used as a model compound for Ag bound to a thiol group,while the proportion of Ag metal (12%) changed little. In soilswithout sludge, 100% of Ag in PVP-Ag MNP and CIT-Ag MNPremained as Ag metal, while incubation in sludge amended soil ledto significant transformations to Ag2S (70% and 78%, respectively).

3.3. Total Ag in pore water

Total porewater Ag concentrations (T-Ag) are expressed as the %Ag in soil present within the pore water to normalize for slightdifferences between total Ag concentrations in individual soilsamples. Recovered acid leachable Ag for whole soil microwavedigestions of the SRMs was 93 � 3.2% and 93 � 1.5% for SRM 2711aand 2781, respectively (n¼ 8).Wemeasured 73�13%, 81�10% and68 � 12% (mean � standard deviation) recovery of PVP-Ag MNPs,CIT-Ag MNPs and Ag ions for the pore water extraction apparatuswhen passing solutions of known Ag concentration in filtered porewater through the apparatus. In soils aged without sludge for 1

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149144

week T-Ag concentrations in pore water were as high as 41%(190 mg Ag L�1) and as low as 1% (4.4 mg Ag L�1) of the total addedAg for CIT-Ag MNP and PVP-Ag MNP treatments, respectively. T-Agconcentrations in pore water for the non-amended AgNO3 treatedsoils were not significantly different from the PVP-Ag MNP treatedsoils. In comparison to AgNO3 and PVP-AgMNP treated soils, CIT-AgMNP treated soil had as much as 40 times more pore water T-Agafter 1 week in non-amended soils. Soils treated with AgNO3 andPVP-Ag MNPs had significantly higher T-Ag concentrations fromsludge amended soils compared to non-amended soils, but with nosignificant difference between sludge amendment rates. Additionof sludge had no significant impact on T-Ag concentrations for CIT-Ag MNP soils at 1 week (Fig. 1).

Fig. 1. Total Ag expressed as a percentage of the soil concentration (w/w%) containedwithin filtered (1.0 mm) pore water samples from soils treated with silver nitrate(AgNO3), polyvinylpyrrolidone coated Ag nanoparticles (PVPeAg MNP), or citratecoated Ag nanoparticles (CIT-Ag MNP) amended with 0, 1, or 3% sludge (w/w; drymass) and aged for 1 week, 2 or 6 months. Bars with the same letter are not signifi-cantly different among treatments (AgNO3, PVP-Ag MNP, CIT-Ag MNP) for a givenincubation time (1 week, 2 or 6 months). All treatments were significantly differentthan control samples amended with 0, 1, or 3% sludge and aged for 1 week, 2 or 6months.

In most cases, less T-Ag was observed in soil pore watersfollowing 2 months of aging. Proportions of T-Ag in pore water innon-amended soils were similar (w 5%) for both Ag MNP types;however, proportion of T-Ag for the AgNO3 treatment was signifi-cantly lower. Sludge amendment caused a slight increase in con-centration of T-Ag in all treatments, although this effect was onlystatistically significant for AgNO3 at the 3% amendment rate. Therewas a significant decrease in T-Ag in all CIT-Ag MNP treatments at 2months relative to 1 week.

Ag was still present in all soil pore waters following 6 months ofaging. However, less than 2% of the added Ag was recovered in porewaters from non-amended AgNO3 treated soils and all Ag treat-ments amended with sludge. Significantly more T-Ag was presentin pore water at six months in non-amended Ag MNP treatments;approximately 9.9% and 7.1% of added Ag were measured for PVP-Ag MNP and CIT-Ag MNP treatments. Unlike AgNO3 and CIT-AgMNP treated soils, T- Ag concentrations increased with aging fornon-amended PVP-Ag MNP soil.

3.4. Dissolved Ag in pore water

Dissolved Ag is operationally defined here as Ag remainingwithin the supernatant after ultracentrifugation of pore water (D-Ag) and expressed as a percentage of the soil concentration tonormalize for small differences in soil Ag concentrations (Fig. 2).We obtained 95 � 2%, 107 � 5% and 49 � 2% recovery of PVP-AgMNPs, CIT-Ag MNPs and Ag ions after spiking filtered pore waterand thenfiltering through the 1 mm glass microfiber filter. Recoveryof Agþ was similar regardless of whether DI water (49 � 2%), orpore water from unamended soil (49 � 3%) or amended soil(47 � 13%) was used. Recovery of MNPs through filters could not bereliably measured in the presence of soil pore water due to thepossibility of heteroaggregate formation. Recovery of Ag for ultra-centrifugation was 79 � 0.4%. Because recovery was unexpectedlylow for Ag ions, concentrations were corrected for recovery fordissolved Ag measurements. In the absence of sewage sludge, therewas more D-Ag in the AgNO3treated soils than in the Ag MNPtreatments; however, the concentrations decreased from a high of13% of the soil content after 1 week of aging to 0.5% after 6 monthsof aging. Concentrations of D-Ag remained very low (<2% of soil Agcontent) for all nanoparticle treatments regardless of aging time orsewage sludge amendment rate. Sewage sludge amendment and/oraging for six months reduced D-Ag concentrations in AgNO3treated soils to similar low levels as the Ag MNP treatments (<2%)at all time points.

3.5. AF4 multidetection analysis

The size distribution and quantity of Ag containing particlesfrom AgNO3 treated soils was altered with addition of sewagesludge. In porewater from the non-amended AgNO3 treated soil, Agis only observed in fractograms at the first time point (1 week;Fig. 3). The average dh is approximately 250 nm. Following aging, Agis absent from fractograms and is likely immobilized due to bindingto the soil solids. It is important to note that any Ag complexes <

5 kDa (w 1 nm) would permeate the AF4 membrane andwould notappear in the fractograms (Moon et al., 2006). Measurements of T-Ag and D-Ag account for such losses. Addition of sludge to AgNO3

treated soils decreased the average particle size to <90 nm. Inseveral fractograms, what appears to be tailing of the void peak ispossibly Ag bound to NOM (<5 nm), which can be resolved fromthe void peak at a higher crossflow (Fig. S-6). With the addition ofsludge, Ag appears in this portion of the fractogram even after agingsoils for 2 months. Increasing the sludge amendment rate increasesthe intensity of this small size fraction.

Fig. 2. Dissolved Ag present in soil pore water expressed as a percentage of soil Ag concentration as a function of Ag form (PVP Ag MNPs ¼ polyvinylpyrrolidone coated Agmanufactured nanoparticles; CIT-Ag MNPs ¼ citrate coated Ag manufactured nanoparticles), sewage sludge amendment and aging time. Error bars indicate standard deviations.

Fig. 3. Asymmetrical flow field-flow fractograms using ICP-MS detection of Ag (m/z ¼ 107) for pore waters extracted from polyvinlylpyrrolidone coated Ag nanoparticles (PVPeAgMNP; AeC), citrate coated Ag nanoparticles (CIT-Ag MNP; DeF) or silver nitrate (AgNO3; GeI) treated soils amended with 0% (A, D, G), 1% (B, E, H), or 3% (C, F, I) sludge (w/w; drymass) and aged for 1 week, 2 or 6 months. The y-axis displays the normalized Ag intensity. The x-axis displays calibrated hydrodynamic diameters.

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149 145

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149146

Soils treated with PVP-Ag MNP typically had more Ag contain-ing particles in soil pore water than AgNO3 treatments. After 1week, pore water Ag from non-amended PVP-Ag MNP treated soilwas present in only small quantities, yet with aging a larger peak (2months) or peaks (6 months) were observed (Fig. 3). In the sludgetreated PVP-Ag MNP samples, as observed for AgNO3, the sizedistribution of Al and Si containing particles which may be allu-minosilicate clay particles, was different than the trace for Ag(Fig. 4). Data from Fe and Mn are not shown due to low intensity,although size distributions generally followed those seen for Al andSi. Ag eluted with a peak at 40 nm. Al and Si displayed a broadparticle size distribution from around 50 nm to around 250 nm.Particles containing Ag recovered in sludge amended PVP-Ag MNPsoils exhibit size distributions that were slightly larger than theparticles observed in the AgNO3 sludge treatments at 1 week. Peakintensities decrease with aging, but Ag containing particles are stillpresent in sludge amended PVP-Ag MNP soils after 6 months.Average Ag containing particle sizes are approximately 75 nm and55 nm for 1% and 3% sludge PVP-Ag MNP soils at 6 months.

Fractograms for non-amended CIT-Ag MNP treated soils displayAg containing peaks comparable in size to the pristine particles(w80 nm) at all time points. Such particles were extractable evenafter aging soils for 6 months. Addition of sludge decreased theaverage Ag containing particle size distribution to <80 nm, com-parable to the AgNO3 and PVP-Ag MNP treatments with sludge atthe 1 week and two month time points. At 6 months of aging, Agappeared to be absent in pore waters extracted from CIT-Ag MNPsoils with sludge. Almost all Ag MNP treatments had greaterquantities of Ag present within the fractograms than observed forAgNO3.

3.6. Validity of AF4 separation

We obtained linear calibration curves for a variety of particlesize standards (protein, latex and gold) with r2 values exceeding0.99 (Fig. S-7). Average Ag recovery was 100 � 8% for samples at alltime points (recovery was measured for every injection; n ¼ 108).Given that the batch DLS diameter (which was intensity weighted)for CIT-Ag MNPs was smaller than the mass weighted diameterfrom AF4-ICP-MS as well as the TEM diameter, it is likely that thebatch DLS result was inaccurate. The larger hydrodynamic radius ofthe PVP-Ag MNPs than the geometric (TEM) diameter, may stemfrom comparing number weighted (TEM) distributions to mass(AF4-ICP-MS) and intensity (DLS) weighted distributions. Fracto-grams from individual soil replicates slightly varied in intensity and

Fig. 4. Representative asymmetrical flow field-flow fractogram using ICP-MS detectionof Al (m/z ¼ 27), Si (m/z ¼ 28) and Ag (m/z ¼ 107) for pore water extracted from soilamended with 3% sludge containing PVP-Ag MNP (w/w; dry mass) and incubated for 1week. The y-axis displays the relative peak intensities because concentrations of Al andSi in soil pore waters were much higher than Ag concentrations. The x-axis displayscalibrated hydrodynamic diameter determined from reference particles.

size distribution; representative samples are shown in Fig. 3 andindividual replicates are available in the SI (Fig. S-8). In addition,light scattering data from MALLS indicated that particle sizes weresimilar to calibrated particle sizes (Fig. S-9), although some minordifferences were observed. Due to excessive absorption of the laserby particles from sludge amended soils, fitting of the MALLS data toavailable models was not possible. Further validation of AF4 sepa-ration in sludge amended samples was provided through TEManalysis of collected fractions as described below.

3.7. TEM-EDS of AF-4 fractions

TEM was used to validate AF4 particle size distributions andcharacterize Ag containing particles in 1 week treatments for selectsoil-sludge samples. Intact Ag MNPs similar in size to the pristineparticles are verified in pore water extracted from non-amendedCIT-Ag MNP treated soil in AF4 fractions that corresponded to thepristine primary particle size (Fig. 5, S10). EDS analysis confirmedthe composition of these particles as primarily containing Ag withsome traces of S (Fig. S-11). No Ag nanoparticles are observed viaTEM in pore water from non-amended PVP-Ag MNP soil from thesize fraction that corresponded to the pristine particle size; how-ever, the concentrations were very low. In both PVP-Ag MNP andCIT-Ag MNP soils amended with 3% sludge, a collection of smallerAg nanoparticles were observed with sizes corresponding to cali-brated sizes from AF4 at which the particles were collected (Fig. 5,S10). These were confirmed with EDS to contain Cl and S (Fig. S-11),although some particles appeared to contain Ag metal coresjudging from the electron density and morphology of the particles.All of the particles within these clusters appear to be primarilycomposed of Ag, S and Cl and do not appear to be heteroaggregates.

4. Discussion

Surface coatingwas demonstrated to influence AgMNP colloidalstability in pore water extracted from soil for up to six months inthe absence of sewage sludge. Initially, relatively high concentra-tions of intact CIT-AgMNPswere observed in porewaterwhile PVP-Ag MNPs were present at very low concentrations and likely boundto solid phases in the soil. This finding is in agreement with anotherstudy which observed PVP-Ag MNPs to have a high affinity for soilsolids in a sandy loam soil (VandeVoort and Arai, 2012). Other workhas predicted that PVP-Ag MNPs would be among the most mobileAg MNP under environmental conditions compared to severalother capped Ag MNPs, including CIT-coated (El Badawy et al.,2010; Tejamaya et al., 2012). However, our results suggest that af-ter 1 week electrostatically stabilized CIT-Ag MNPs were morestable in soil porewater than sterically stabilized PVP-AgMNPs. It islikely that the uncharged PVP coating has a relatively high affinityfor the soil solid phases, while the CIT coating, possessing a netnegative charge would repel negatively charged soil surfaces.Although the PVP- AgMNPs are sterically stabilized, this apparentlydoesn’t stabilize them against heteroaggregation with soil solids.We previously observed that sterically stabilized PVP-Ag MNPs aresusceptible to heteroaggregation in aquatic systems (Unrine et al.,2012). It is also possible that CIT, having a lower molecularweight, more readily exchanges with dissolved organic matter(DOM) than the high molecular weight PVP, since high molecularweight compounds would have more points of attachment to theparticle surface causing them to desorb more slowly. Previousstudies have shown DOM to have a stabilizing effect on Ag MNPs(Bone et al., 2012; King and Jarvie, 2012). Between 1 week and sixmonths, there was a clear increase in the amount of particles pre-sent within pore water for PVP-Ag MNPs, but a decrease for theamount of particles present for CIT-AgMNPs.We postulate that this

Fig. 5. Transmission electron microscopy (TEM) images confirming the presence of Ag nanoparticles in pore waters extracted after 1 week from non-amended CIT-Ag MNP treatedsoil (aec), 3% sludge (w/w; dry mass) CIT-Ag MNP treated soil (def) and 3% sludge amended PVP-Ag MNP treated soil (gei). Accompanying energy dispersive spectra (EDS) can befound in the Appendix (Fig. S-11).

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149 147

may be the result of slow exchange between coating and DOM orother ligands, which may have increased the net negative chargefor the PVP coated particles, which had a low starting zeta poten-tial, while decreasing the net negative charge for the CIT coatedparticles, which had a high starting zeta potential. Alternatively, itcould be a result of over coating of the PVP or CIT with DOM.

Pre-incubation of Ag with sludge and subsequent amendmentto soils had a large effect on the ensuing behavior of Ag and seemedto negate the effect of initial AgMNP coating. Surface coating (CITorPVP) had little effect on Ag MNP particle size in pore waters in soilsamended with Ag containing sludge, as evidenced by similarity ofthe fractograms for CIT-AgMNPs and PVP-AgMNPs. Comparatively,non-amended soils had higher pore water Ag concentrations at sixmonths for Ag MNP treatments compared to AgNO3, likely as aresult of higher affinity of Ag ions to soil solids than Ag MNPs.Application of Ag to soil through sludge resulted in similar T-Agpore water concentrations in all treatments, likely due to similar-ities in chemical speciation of Ag in sludge. While Ag speciation inYSL soil was found to be 100% Ag (0) for both Ag MNP treatments,introduction of Ag MNP to sludge transformed the particlesresulting in extensive sulfidation. Lack of transformation of AgMNPs in sludge-free soil is in agreement with our previous resultswhere little oxidation of AgMNPs was shown up to 28 days of agingin soils (Shoults-Wilson et al., 2011a,b). Results from EXAFS LCF arein agreement with previous studies showing sulfidation of Ag MNPas a major environmental transformation (Kim et al., 2010; Levardet al., 2011, 2012; Lowry et al., 2012; Lombi et al., 2013). It is

likely that sulfidation occurred during the incubation with sludgeand not as a result of contact with the soil since our previousstudies (Shoults-Wilson et al., 2011a,b) showed that the YeagerSandy Loam does not cause sulfidation of AgNPs. It is likely thatwith more prolonged digestion during the wastewater treatmentprocess at environmentally realistic concentrations that completesulfidation may occur (Lombi et al., 2013). Levard et al. (2011)observed that the PVP surface coating did not inhibit sulfidationor subsequent aggregation. Comparable results among PVP and CIT-Ag MNPs suggest that the CIT coating also does not inhibit sulfi-dation. Lombi et al. (2013) also recently confirmed that regardlessof core composition (Ag or AgCl) or coating composition, Ag2S is thedominant form of Ag found in sewage sludge. A small fraction of Ag(0) was observed in sludge treatments for both AgNO3 and AgMNPs. It is possible that this is the result of formation of a coreeshell Ag(0)-Ag2S structure as previously described (Levard et al.,2011).

In soils amended with sludge containing Ag MNPs at 1 week, asharp, tailing, Ag peak appeared in the fractograms with a peakparticle size around 35 nm. This peak had a different size distri-bution than the peaks for Al and Si, suggesting that it was not Agbound to clay particles. Maurer et al. (2013) suggested that sulf-hydryl groups of soil organic matter are themajor site of adsorptionof Agþ, a soft metal, rather than oxygen bearing functional groupson clay particles. The peak eluting near the void volume in thefractograms is similar to the expected size of organic matter(Unrine et al., 2012). Analysis of fractions collected from this peak

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149148

by TEM confirmed the presence of Ag electron rich particles in theAg MNP treatments, but clay particles enriched with Ag were notobserved. Particles recovered from the fractograms of sludgetreated soils appeared to consist of Ag (0) cores with weatheredrims consisting of Ag2S and or AgCl or in some cases Ag (0) particleswithout rims, confirming the EXAFS results which showed thepresence of Ag(0) and Ag2S in the samples. We propose two pos-sibilities to explain why these particles were smaller than theparticles in the distributions in of the pristine particles: (1) some ofthe observed particles have been re-precipitated as Ag (0), AgCl orAg2S from the release of dissolved Ag from oxidative dissolution ofthe added MNPs, and/or (2) some of the particles have actuallybeen weathered to a smaller size through dissolution processes,until the dissolution process was slowed due to the formation of ashell of Ag2S or AgCl. At week one, we also observed Ag within asimilar size range in the AgNO3 treatment when sludge was added.These could have also been precipitated Ag (0) particles. Maureret al. (2012) observed ionic Ag reduced to Ag(0) as a majorpathway for the removal of free Agþ ions from solution whenexposed to an anaerobic soil with organic matter. After six monthsof aging, few particles were observed in pore waters from anysludge amended soil, regardless of Ag form.

The present study suggests that surface coating dictates AgMNPpartitioning to pore water when directly introduced to soil, butinitial Ag MNP coating has less relevance when added via sewagesludge amendment. Non-amended soil treated with CIT-Ag MNPshad ten-fold more Ag in soil pore water after 1 week than soiltreated directly with PVP-Ag MNP or AgNO3. In sludge amendedsoils, similar pore water Ag concentrations and size distributions ofAg containing particles are observed for both Ag MNP and AgNO3treatments; although fractograms demonstrated that more Ag waspresent in colloidal form in pore water for Ag MNP treatments thanthe AgNO3 treatment.

The amount of dissolved Ag found in pore water was very lowfor all treatments except for AgNO3 treated soils at 1 week and 2months. After six months very little (<2%) of the total Ag was foundas dissolved Ag in the pore water regardless of Ag form or sludgeamendment. This suggests that exposure of organisms to free Agions in pore water would be similar in soils that have aged for morethan 2 months regardless of how the Ag was introduced to the soil.Cornelis et al. (2012) recently investigated dissolution of Ag MNPsin a variety of soils and found that partitioning of Agþ to soil solidsand/or DOM was so strong that little could be detected in soil porewater after ultrafiltration. Regardless of the strong partitioning ofAg to soil solids and DOM, Ag MNPs and AgNO3 do cause toxicity insoil organisms, including behavioral avoidance at very low con-centrations (<10 mg Ag kg�1 soil) (Shoults-Wilson et al., 2011c).

5. Conclusion

Silver form (Ag ion versusMNP), MNP coating, aging and contactwith sewage sludge all have profound effects on Ag behavior in soilpore water. In unamended soils Ag MNP coating has clear effects onpartitioning to pore water, but not after contact with sewagesludge. During the first few months after amendment with sewagesludge, the chemical speciation was similar regardless of Ag source(nanoparticles versus ions), but the physical form of the Ag com-pounds (particles suspended in pore water versus bound to soilsolid phase) differed. Based on this finding it may not be sufficientto assume that Ag introduced into biosolids as MNPs would behavethe same in soils as Ag introduced as Ag salts, even though thechemical speciation is ultimately similar. On the other hand, it maybe sufficient to assume that taking into account Ag NP coating is notnecessary for risk assessment of biosolids amendment since theeffect of coating on particle behavior will be virtually eliminated by

sulfidation during the wastewater treatment process. Studies areneeded to see if this also carries through to the toxicity of particleswith different coatings after sulfidation. The differences observedbetween behavior of Ag added as Ag ions versus MNPs suggeststhat studies of the chemical speciation alone (e.g. EXAFS) are notsufficient and that information on the physical form of Ag con-taining particles in pore water is needed. Because of this differencein behavior, more studies on the fate and effects of Ag MNPsintroduced into soils through biosolids amendment are warranted.Using AF4-ICP-MS combined with EXAFS, TEM and EDS analyses isan approach that may have considerable utility in future studies.Total Ag concentrations used in this work exceed recent projectedconcentrations of Ag MNPs in sewage sludge amended soils(Gottschalk et al., 2009). However, the present study lays importantgroundwork for detection and characterization nanomaterials insoil. Future studies could focus on refining these methods to makethem capable of characterizing these transformations at lowerconcentrations in order to enable characterization and detection ofnanoparticles in the environment.

Acknowledgment

This researchwas supported by the United States EnvironmentalProtection Agency (U.S. EPA) and National Science Foundation(NSF) through cooperative agreements CR-83515701 (Office ofResearch and Development) and EF-0830093 (Center for Environ-mental Implications of Nanotechnology) and through the EPA Sci-ence to Achieve Results (STAR) program (RD-83485701 and RD-83457401 ). It has not been formally reviewed by EPA or NSF. Theviews expressed in this document are solely those of the authors.EPA and NSF do not endorse any products or commercial servicesmentioned in this publication. The authors also wish to acknowl-edge the assistance of J. Ye, S. Hunyadi, S. Marinakos, M. Vandiviere,A. Gondikas and J. Nelson. Portions of this researchwere carried outat the Stanford Synchrotron Radiation Lightsource, a directorate ofSLAC National Accelerator Laboratory and an Office of Science UserFacility operated for the U.S. Department of Energy Office of Scienceby Stanford University.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.envpol.2013.06.027.

References

Benn, T.M., Westerhoff, P., 2008. Nanoparticle silver released into water fromcommercially available Sock Fabrics. Environ. Sci. Technol. 42, 4133e4139.

Bertsch, P.M., Seaman, J.C., 1999. Characterization of complex mineral assemblages:implications for contaminant transport and environmental remediation. Proc.Nat. Acad. Sci. U. S. A. 96, 3350e3357.

Blaser, S., Scheringer, M., MacLeod, M., Hungerbuhler, K., 2007. Estimation of cu-mulative aquatic exposure and risk due to silver: contribution of nano-functionalized plastics and textiles. Sci. Total Environ. 390, 396e409.

Bone, A.J., Colman, B.P., Gondikas, A.P., Newton, K.M., Harrold, K.H., Cory, R.M.,Unrine, J.M., Klaine, S.J., Matson, C.W., Di Giulio, R.T., 2012. Biotic and abioticinteractions in aquatic microcosms determine fate and toxicity of Ag nano-particles: part 2-Toxicity and Ag speciation. Environ. Sci. Technol. 46, 6925e6933.

Cheng, Y., Yin, L., Lin, S., Wiesner, M., Bernhardt, E., Liu, J., 2011. Toxicity reduction ofpolymer-stabilized silver nanoparticles by sunlight. J. Phys. Chem. C 115, 4425e4432.

Colman, B.P., Arnaout, C.L., Anciaux, S., Gunsch, C.K., Hochella Jr., M.F., Kim, B.,Lowry, G.V., McGill, B.M., Reinsch, B.C., Richardson, C.J., Unrine, J.M., Wright, J.P.,Yin, L., Bernhardt, E.S., 2013. Low concentrations of silver nanoparticles inbiosolids cause adverse ecosystem responses under realistic field scenario. PLoSOne 8, e57189.

Cornelis, G., DooletteMadeleine Thomas, C., McLaughlin, M.J., Kirby, J.K., Beak, D.G.,Chittleborough, D., 2012. Retention and dissolution of engineered silver nano-particles in natural soils. Soil Sci. Soc. Am. J. 76, 891e902.

A.R. Whitley et al. / Environmental Pollution 182 (2013) 141e149 149

Coutris, C., Joner, E.J., Oughton, D.H., 2012. Aging and soil organic matter contentaffect the fate of silver nanoparticles in soil. Sci. Total Environ. 420, 327e333.

El Badawy, A., Feldhake, D., Venkatapathy, R., 2010. State of the Science LiteratureReview: Everything Nanosilver and More. Environmental Protection Agency,Washington, DC, p. 221.

EPA, 1995. A Guide to the Biosolids Risk Assessments for the EPA Part 503 Rule.United States Environmental Protection Agency, Washington, DC, USA.

EPA, 1996. Method 3052: Microwave Assisted Acid Digestion of Siliceous andOrganically Based Matrices. United States Environmental Protection Agency,Washington, DC, USA.

EPA, 1998. Method 6020: Inductively Coupled Plasma Mass Spectrometry. UnitedStates Environmental Protection Agency, Washington, DC, USA.

EPA, 2000. Biosolids Technology Fact Sheet: Land Application of Biosolids. UnitedStates Environmental Protection Agency, Washington, DC, USA. EPA 832-F-00e064.

Fabrega, J., Fawcett, S.R., Renshaw, J.C., Lead, J.R., 2009. Silver nanoparticle impact onbacterial growth: effect of pH, concentration, and organic matter. Environ. Sci.Technol. 43, 7285e7290.

Fedotov, P., Vanifatova, N., Shkinev, V., Spivakov, B., 2011. Fractionation and char-acterization of nano- and microparticles in liquid media. Anal Bioanal. Chem.400, 1787e1804.

Gottschalk, F., Sonderer, T., Scholz, R.W., Nowack, B., 2009. Modeled environmentalconcentrations of engineered nanomaterials (TiO2, ZnO, Ag, CNT, fullerenes) fordifferent regions. Environ. Sci. Technol. 43, 9216e9222.

Helland, A., Kastenholz, H., Thidell, A., Arnfalk, P., Deppert, K., 2006. Nanoparticulatematerials and regulatory policy in Europe: an analysis of stakeholder per-spectives. J. Nanopart Res. 8, 709e719.

Hirsch, M.P., 1998. Availability of sludge-borne silver to agricultural crops. Environ.Toxicol. Chem. 17, 610e616.

Hotze, E.M., Phenrat, T., Lowry, G.V., 2010. Nanoparticle aggregation: challenges tounderstanding transport and reactivity in the environment. J. Environ. Qual. 39,1909e1924.

Kaegi, R., Sinnet, B., Zuleeg, S., Hagendorfer, H., Mueller, E., Vonbank, R., Boller, M.,Burkhardt, M., 2010. Release of silver nanoparticles from outdoor facades. En-viron. Pollut. 158, 2900e2905.

Kaegi, R., Voegelin, A., Sinnet, B., Zuleeg, S., Hagendorfer, H., Burkhardt, M.,Siegrist, H., 2011. Behavior of metallic silver nanoparticles in a pilot wastewatertreatment plant. Environ. Sci. Technol. 45, 3902e3908.

Kim, B., Murayama, M., Colman, B.P., Hochella, M.F., 2012. Characterization andenvironmental implications of nano- and larger TiO2 particles in sewage sludge,and soils amended with sewage sludge. J. Environ. Monit. 14, 1128e1136.

Kim, B., Park, C.S., Murayama, M., Hochella, M.F., 2010. Discovery and character-ization of silver sulfide nanoparticles in final sewage sludge products. Environ.Sci. Technol. 44, 7509e7514.

King, S.M., Jarvie, H.P., 2012. Exploring how organic matter controls structuraltransformations in natural aquatic nanocolloidal dispersions. Environ. Sci.Technol. 46, 6559e6967.

Kittler, S., Greulich, C., Diendorf, J., Köller, M., Epple, M., 2010. Toxicity of silvernanoparticles increases during storage because of slow dissolution underrelease of silver ions. Chem. Mat 22, 4548e4554.

Koppel, D.E., 1972. Analysis of macromolecular polydispersity in intensity correla-tion spectroscopy: the method of cumulants. J. Chem. Phys. 57, 4814e4820.

Levard, C., Hotze, E.M., Lowry, G.V., Brown, G.E., 2012. Environmental trans-formations of silver Nanoparticles: impact on stability and toxicity. Environ. Sci.Technol. 46, 6900e6914.

Levard, C., Reinsch, B.C., Michel, F.M., Oumahi, C., Lowry, G.V., Brown, G.E., 2011.Sulfidation processes of PVP-coated silver nanoparticles in aqueous Solution:impact on dissolution rate. Environ. Sci. Technol. 45, 5260e5266.

Lombi, E., Donner, E., Taheri, S., Tavakkoli, E., Jamting, A., McClure, S., Naidu, R.,Miller, B., Scheckel, K., Vasilev, K., 2013. Transformation of four silver/silverchloride nanoparticles during anaerobic treatment of wastewater and post-processing of sewage sludge. Environ. Pollut. 176, 193e197.

Lowry, G.V., Espinasse, B.P., Badireddy, A.R., Richardson, C.J., Reinsch, B.C.,Bryant, L.D., Bone, A.J., Deonarine, A., Chae, S., Therezien, M., Colman, B.P., Hsu-Kim, H., Bernhardt, E.S., Matson, C.W., Wiesner, M.R., 2012. Long-term trans-formation and fate of manufactured Ag nanoparticles in a simulated large scalefreshwater emergent wetland. Environ. Sci. Technol. 46, 7027e7036.

Luoma, S., 2008. Silver Nanotechnologies and the Environment: Old Problems orNew Challenges. WoodrowWilson International Center for Schollars, Project onEmerging Nanotechnologies, The Pew Charitable Trusts. PEN-15.

MacCuspie, R., 2011. Colloidal stability of silver nanoparticles in biologically relevantconditions. J. Nanopart Res. 13, 2893e2908.

Maurer, F., Christl, I., Hoffmann, M., Kretzschmar, R., 2012. Reduction and reox-idation of humic acid: influence on speciation of cadmium and silver. Environ.Sci. Technol. 46, 8808e8816.

Moon, J., Kim, S.H., Cho, J., 2006. Characterizations of natural organic matter as nanoparticle using flow field-flow fractionation. Coll. Surf A-Physicochem Eng. Asp287, 232e236.

Project on Emerging Nanotechnologies, 2008. Inventory of Nanotechnology-basedConsumer Products Currently on the Market. Woodrow Wilson InternationalCenter for Scholars and Pew Charitable Trusts. www.nanoproject.org/inventories/consumer.

Ratte, H., 1999. Bioaccumulation and toxicity of silver compounds: a review. Envi-ron. Toxicol. Chem. 18, 89e108.

Reinsch, B.C., Levard, C., Li, Z., Ma, R., Wise, A., Gregory, K.B., Brown, G.E., Lowry, G.V.,2012. Sulfidation of silver nanoparticles decreases Escherichia coli growth in-hibition. Environ. Sci. Technol. 46, 6992e7000.

Rhoades, J.D., 1982. Soluble salts. In: Page, A.L., et al. (Eds.), Methods of Soil Analysis:Part 2: Chemical and Microbiological Properties. Monograph Number 9 seconded. ASA, Madison, WI, pp. 167e179.

Sagee, O., Dror, I., Berkowitz, B., 2012. Transport of silver nanoparticles (AgNPs) insoil. Chemosphere 88, 670e675.

Shoults-Wilson, W.A., Reinsch, B.C., Tsyusko, O.V., Bertsch, P.M., Lowry, G.V.,Unrine, J.M., 2011a. Effect of silver nanoparticle surface coating on bio-accumulation and reproductive toxicity in earthworms (Eisenia fetida). Nano-toxicology 5, 432e444.

Shoults-Wilson, W.A., Reinsch, B.C., Tsyusko, O.V., Bertsch, P.M., Lowry, G.V.,Unrine, J.M., 2011b. Role of particle size and soil type in toxicity of silvernanoparticles to earthworms. Soil Sci. Soc. Am. J. 75, 365e377.

Shoults-Wilson, W.A., Zhurbich, O.I., McNear, D.H., Tsyusko, O.V., Bertsch, P.M.,Unrine, J.M., 2011c. Evidence for avoidance of Ag nanoparticles by earthworms(Eisenia fetida). Ecotoxicology 20, 385e396.

Tejamaya, M., Römer, I., Merrifield, R.C., Lead, J.R., 2012. Stability of citrate, PVP, andPEG coated silver nanoparticles in Ecotoxicology media. Environ. Sci. Technol.46, 7011e7017.

Turkevich, J., Stevenson, P.C., Hillier, J., 1951. A study of the nucleation and growthprocesses in the synthesis of colloidal gold. Discuss. Faraday Soc. 11, 55e75.

Unrine, J.M., Colman, B.P., Bone, A.J., Gondikas, A.P., Matson, C.W., 2012. Biotic andabiotic interactions in aquatic microcosms determine fate and toxicity of Agnanoparticles. Part 1. Aggregation and dissolution. Environ. Sci. Technol. 46,6915e6924.

VandeVoort, A.R., Arai, Y., 2012. Environmental chemistry of silver in soils: currentand historic perspective. Adv. Agron. 114, 59e90.

von der Kammer, F., Ferguson, P.L., Holden, P.A., Masion, A., Rogers, K.R., Klaine, S.J.,Koelmans, A.A., Horne, N., Unrine, J.M., 2012. Analysis of engineered nano-materials in complex matrices (environment and biota): general considerationsand conceptual case studies. Environ. Toxicol. Chem. 31, 32e49.

Weinberg, H., Galyean, A., Leopold, M., 2011. Evaluating engineered nanoparticles innatural waters. Trends Anal Chem. 30, 72e83.

Wiesner, M.R., Lowry, G.V., Alvarez, P., Dionysiou, D., Biswas, P., 2006. Assessing therisks of manufactured nanomaterials. Environ. Sci. Technol. 40, 4336e4345.

Zhang, Y., Chen, Y., Westerhoff, P., Crittenden, J., 2009. Impact of natural organicmatter and divalent cations on the stability of aqueous nanoparticles. WaterRes. 43, 4249e4257.