ORIGINAL PAPER

Comparative toxicity study of Ag, Au, and Ag–Au bimetallicnanoparticles on Daphnia magna

Ting Li & Brian Albee & Matti Alemayehu & Rocio Diaz &

Leigha Ingham & Shawn Kamal & Maritza Rodriguez &

Sandra Whaley Bishnoi

Received: 26 April 2010 /Revised: 7 June 2010 /Accepted: 8 June 2010 /Published online: 26 June 2010# Springer-Verlag 2010

Abstract A comparative assessment of the 48-h acutetoxicity of aqueous nanoparticles synthesized using thesame methodology, including Au, Ag, and Ag–Au bimet-allic nanoparticles, was conducted to determine theirecological effect in freshwater environments through theuse of Daphnia magna, using their mortality as atoxicological endpoint. D. magna are one of the standardorganisms used for ecotoxicity studies due to theirsensitivity to chemical toxicants. Particle suspensions usedin toxicity testing were well-characterized through acombination of absorbance measurements, atomic force orelectron microscopy, flame atomic absorption spectrometry,and dynamic light scattering to determine composition,aggregation state, and particle size. The toxicity of allnanoparticles tested was found to be dose and compositiondependent. The concentration of Au nanoparticles thatkilled 50% of the test organisms (LC50) ranged from 65–75 mg/L. In addition, three different sized Ag nanoparticles(diameters=36, 52, and 66 nm) were studied to analyze thetoxicological effects of particle size on D. magna; however,it was found that toxicity was not a function of size andranged from 3–4 μg/L for all three sets of Ag nanoparticlestested. This was possibly due to the large degree ofaggregation when these nanoparticles were suspended instandard synthetic freshwater. Moreover, the LC50 valuesfor Ag–Au bimetallic nanoparticles were found to bebetween that of Ag and Au but much closer to that of Ag.

The bimetallic particles containing 80% Ag and 20% Auwere found to have a significantly lower toxicity toDaphnia (LC50 of 15 μg/L) compared to Ag nanoparticles,while the toxicity of the nanoparticles containing 20% Agand 80% Au was greater than expected at 12 μg/L. Thecomparison results confirm that Ag nanoparticles weremuch more toxic than Au nanoparticles, and that theintroduction of gold into silver nanoparticles may lowertheir environmental impact by lowering the amount of Agwhich is bioavailable.

Keywords Forensics/toxicology . AAS . AFM .X-rayspectroscopy . Nanoparticles/nanotechnology .

Metals/heavy metals

Introduction

The field of nanomaterials is a fast-growing area and hasgained great attention by scientists and industry manufac-turers because of its multi-functionality, along with pro-cessing properties that can be tailored. In the past few years,there has been an explosion of interest in the use ofnanomaterials in assays regarding every field, such asdetection of gases [1], metal ions, pH [2, 3], and DNA [4,5] or protein [6, 7], and especially popular as deliverysystems and molecular diagnostics. In certain cases, assaysbased on nanomaterials have offered significant advantagesover conventional assessments in sensitivity [8] andselectivity [9]. Nano-sized materials serve as an idealcandidate for various applications due to their extremelysmall size with correspondingly large surface-to-volumeratio. Furthermore, their properties may modified byvarying the size, shape and composition through syntheticways [10–14].

T. Li :B. Albee :M. Alemayehu :R. Diaz : L. Ingham :S. Kamal :M. Rodriguez : S. Whaley Bishnoi (*)BCPS Department, Illinois Institute of Technology,3101 South Dearborn Street,Chicago, IL 60616, USAe-mail: bishnoi@iit.edu

Anal Bioanal Chem (2010) 398:689–700DOI 10.1007/s00216-010-3915-1

Since nanomaterials have shown an exponential appli-cation in domains such as chemistry and biology, thegrowing use of these nanoparticles inevitably leads to theirrelease into the environment. The impact of this intentionaland unintentional disposal is largely unknown, especially infreshwater ecosystems [15, 16]. In addition, silver nano-particles and various silver-based materials containing ionicsilver or metallic silver are already being commercializedfor their antimicrobial activity [17–20]. This antibacterialproperty was applied to numerous interesting applications,including as coatings on medical devices [18], adjuvant intextiles [21], and in bone cements [22]. Such toxicity isbeneficial for these applications; however, it also suggestseco-toxicological issues upon release [23]. Based on thisadoption rate, it is crucial to understand the degree ofhazard regarding silver-based nanoparticles to the environ-ment. However, there have only been a limited number ofstudies analyzing the interactions of aquatic organisms withmanufactured silver nanoparticles released into the envi-ronment [17, 24–28].

Several mechanisms of the toxic effect of silver colloidparticles have been hypothesized, with the most commonbeing the release of Ag+ from nanoparticles, leading toother issues such as the increased production of reactiveoxygen species (ROS) [29–31]. This suggests that nano-particle toxicity effects depend on a particle’s composition,size, and shape. Some antimicrobial studies conductedusing the scanning tunneling electron microscopy (STEM)and X-ray energy dispersive spectrometry (XEDS), showedsilver nanoparticles not only at the surface of cellmembrane, but also inside the bacteria [32]. Asharani etal. observed a concentration-dependent increase in mortal-ity and hatching delay when zebrafish embryos were treatedwith Ag nanoparticles (NPs) [26]. They found that Ag NPsaccumulated inside of nuclei and could be responsible forDNA damage that led to further embryonic developmentdamage. Furthermore, Oberdorster et al. proposed that theendocytosis and biokinetics were largely dependent onnanoparticles surface conditions, in vivo surface modifica-tions [33], as well as free-radical generations [34].

Crustaceans such as D. magna, Daphnia pulex, andCeriodaphnia dubia are approved test models for effluentrelease into the environment by the Environmental Protec-tion Agency and other regulatory agencies [35–37] becauseof their sensitivity towards potential pollutants, such asmetal ion species. Since they are filter-feeders, they havealso been included as test organisms by many groupsconducting nanotoxicity studies. For example, Lovern et al.studied the uptake and release of gold nanoparticles in thegut of D. magna [38] as well as the mortality of D. magnawhen exposed to either titanium dioxide (TiO2) or fullerenenanoparticles [39]. They found that TiO2 and carbonnanomaterials had toxicities that were dependent on sample

preparation and ranged in the 0.46–7.9 mg/L for fullerenesand was ∼5.5 mg/L for TiO2. They also found that gold wasrelatively benign with lowest observable effect concentra-tions of 500 μg/L [38]. According to their reports,behavioral and physiological changes were observed whenthese organisms were exposed to sub-lethal concentrationof carbon-based nanoparticles but not TiO2. Zhu et al.studied the acute toxicity and uptake of six manufacturednanomaterial suspensions on D. magna [40]. Largeamounts of dark material were found in the gut tract ofDaphnia after exposure but not in the control group. Thiscase was consistent with the results from Lovern et al.indicating that the accumulation of nanoparticles occurs inthe gut in a dose-dependent manner [38].

Our objective in this comparative study was to assessthe potential toxicity that Ag–containing nanoparticlesmay have upon release into aquatic environments, byassessing the effects of particle size, aggregation state,and composition on toxicity via testing Ag NPs ofdifferent sizes, Au NPs, and Ag–Au bimetallic NPs withvarious compositions. Though several studies have beendone to assess the toxicity of silver ion, including thosein daphnids [20, 25, 27, 28], only a few have comparedsilver ion toxicity to that of silver in a nanoparticle form.In addition, many of these studies do not provideanalytical data to quantify the amount of silver present intheir testing assays, which we have found to be critical forunderstanding the relationship between particle exposureand toxicity. In this study, we have analyzed the toxicity ofion versus nanoparticle with respect to silver and gold,including nanoparticles that are bimetallic in nature. Wehave analyzed acute toxicity of Ag NPs of varying size bychanging the ratio between Ag+ and the reducing/cappingagent, sodium citrate. Several studies have demonstratedthat surface chemistry, including functionalization, couldaffect the toxicity of nanomaterials [41–43]. Though bettercontrol over particle dimensions and polydispersity hasbeen achieved by others through the use of surfactants orcapping agents [26, 44], we specifically focused on thesynthesis of nanoparticles with a common surface chem-istry to remove potential matrix effects from organiccapping agents or complications in data interpretationdue to changes in surface chemistry.

We exposed D. magna to varying concentrations ofeach nanoparticle in a synthetic freshwater solution toassess the particle concentration that resulted in mortalityof 50% of the test population in 48 h (LC50). The currentmechanism of Ag NP toxicity to aquatic organisms inliterature is the localized release of Ag+, which results inan interference in sodium transport [45]. Based on thismechanism, it was expected that smaller particles wouldshow a more detrimental effect on Daphnia due to theirgreater surface area. In order to compare the degree of

690 T. Li et al.

toxicity as a function of composition, Au nanoparticles,silver ion, and Ag–Au bimetallic nanoparticles wereassessed during the research. By analyzing the acutetoxicity range of nanoparticles on Daphnia, a comprehen-sion of their potential impact on the aquatic environmentcan be gained. In order to provide a good understanding ofthe actual entities that are responsible for toxicity intesting waters, particle suspensions employed in toxicitytesting were well-characterized using a combination ofvisible light absorbance measurements, atomic force orelectron microscopy, flame atomic absorption spectrome-try, and dynamic light scattering to determine composi-tion, aggregation state, and particle sizes.

Experimental

Materials

Silver nitrate (AgNO3), tetra chloroauric acid (HAuCl4) andsodium citrate were purchased from Sigma-Aldrich. All thesolutions were prepared by distilled deionized water.

Preparation of gold nanoparticles

Gold nanoparticles were prepared by the reduction ofchloroauric acid by using sodium citrate. Typically, 1 mlof 1% HAuCl4 and 1.5 ml 1% sodium citrate were addedto the 100 ml of Milli-Q water to form a solution, whichwas heated until boiling. The formation of nanoparticleswas carried out at boiling temperature for approximatelyten minutes. During this period, a color change from lightyellow to ruby red was observed. Particles were charac-terized using visible absorbance spectrophotometry(Jasco, V-530), dynamic light scattering (DLS, MalvernNanoS), and atomic force microscopy (AFM, Agilent5500 AFM, tapping mode). Ten microliters of stocksolution of gold nanoparticles suspension was placed ona silicon substrate and dried before placed in the AFM forimaging.

Preparation of silver nanoparticles

Stable silver nanoparticles were formed when an aqueoussolution of AgNO3 (∼1 mM) was boiled in the presence ofsodium citrate using a procedure similar to that demon-strated by Pillai et al. [46]. A 1-mM AgNO3 solution wasprepared by dissolving 21.2 mg of AgNO3 powder in125 mL Milli-Q water. Then the solution was heated untilboiling and then 1% sodium citrate was added dropwise toobtain silver nanoparticles with different molar ratios ofsilver ion to citrate. For example, through the addition of 5,10, or 15 mL of 1% citrate to the silver nitrate solution, the

Ag: citrate ratios of 1:1.6, 1:3.1, or 1:4.2 were obtained.After heating for 1 h, the color of the solution slowly turnedgrayish yellow, indicating the reduction of the silver ion.Particles were characterized with visible spectrophotometry,dynamic light scattering, and transmission electron micros-copy (TEM, JEOL JEM2010 electron microscopy, UIC)analysis. A small amount of the final dispersion wasdropped onto a carbon-coated copper grid placed on filterpaper. After evaporation of water, TEM micrographs wererecorded at 200 kV. The element analysis was carried out atthe same time by a Thermo Noran Vantage X-ray energydispersive spectroscopy (XEDS) system with a 40-mm lightelement detector.

Preparation of Ag–Au bimetallic nanoparticles

Synthesis of Ag-Au bimetallic nanoparticles was carriedout through the co-reduction of metal precursor salts using2% sodium citrate as a reducing agent as well as a cappingagent, using a similar method to that used by Pal et al. [47].To a 100-ml round-bottom flask, 49 ml of DI water wastaken. Then, 0.2 ml of AgNO3 (10

−2 M) as well as 0.8 mlof HAuCl4 (10

−2 M) were added. The mixture solution washeated until boiling. A 0.5-ml volume of 2% citrateaqueous solution was added at the boiling point. Thereaction mixture was heated for an hour. Bimetallic nano-particles with a different mole ratio of silver to gold saltswere synthesized, such as 0.8 ml Au salts combined with0.2 ml of AgNO3. Total metal ion concentration wasmaintained at 10−4 M. Regarding the concern of depositionof AgCl, the reaction was kept at 100°C during thesynthesis to prevent its deposition, which made the productof [Ag+] and [Cl−] to be lower than the solubility product(1.2×10−6 at 100°C; 1.8×10−10 at 20°C). The formation ofnanoparticles was noticed by the change in color andconfirmed by optical spectroscopy. Particles were charac-terized with visible spectrophotometry, DLS, and TEM.

Particle stability studies

In order to determine whether the Daphnia were beingexposed to individual nanoparticles or aggregated species,all nanoparticles were examined by dynamic light scatter-ing at three stages: immediately after synthesis, after beingsuspended in standard synthetic freshwater (SSF) for0.5 h, and suspension in SSF for 24 h. The nanoparticleswere monitored at a concentration of 5 mg/L in SSF toallow a high-enough concentration for detection by DLSbut also a high-enough ionic strength to determine the realeffect of SSF on particle dispersion. Polydispersity index(PDI) represents how uniform the particle sizes are insuspension with 0 being monodisperse and 1 beingpolydisperse [41].

Study of Ag, Au, and Ag–Au bimetallic nanoparticles on Daphnia magna 691

Atomic absorption

An atomic absorption spectrophotometer (Varian SpectraAA 55B) was used to quantitatively measure the ionicsilver in nanoparticle samples using an air/acetylene(oxidant/fuel) mixture. In order to separate the Ag NP, theNP solutions were centrifuged (Eppendorf, Minispin). Theseparation was conducted by centrifuging 2.2 ml of silvercolloid solution, running at 12K rpm for 15 min. Then2 mL of supernatant was collected and further diluted 10×with deionized water. The solution appeared transparent incolor after separation. The supernatant was then analyzedby AA and compared to the original silver colloid samples.For calibration, fresh silver standard solutions were gener-ated by dissolving 20 mg of silver nitrate in 1,000 mL ofdeionized water. The stock solution was serially diluted toproduce 10, 5.0, 1.0, 0.5, 0.1 mg/L Ag ion solutions.

Daphnia toxicity analysis

The standard synthetic freshwater (SSF) was preparedaccording to the EPA protocol [36]. The media wereprepared by adding different amount of magnesium sulfate(MgSO4), sodium bicarbonate (NaHCO3), potassium chlo-ride (KCl), and calcium sulfate dihydrate (CaSO4·2H2O) todeionized water. All chemicals were acquired from Sigma-Aldrich and used without further purification. The mediawere aerated overnight before use to ensure sufficientoxygenation. Daphnia cultures (ARO strain, AquaticResearch Organisms, Hampton, NH) were maintained at20–22 °C at ambient light (16:8 light:dark cycle) and fed acombination of Spirulina (Salt Creek, Inc, Salt Lake City,UT) and Roti-Rich (Florida Aqua Farms, Dade City, FL).The feeding for the cultures in every 1,600 ml of SSF waterconsisted of three to five drops of algae concentrate andthree drops of Roti-Rich on alternate days. The toxicity ofdifferent nanoparticles was investigated on 24-h old D.magna neonates with a SSF control. This SSF controlgroup was required to have survival rate of at least 100% tomake the toxicity test valid. Acute sample toxicity testswere conducted using a multi-concentration test, consistingof a control and different concentrations of nanoparticles.Each test was performed by placing four Daphnia neonatesinto 30 ml of SSF with nanoparticles into a 50-ml conicalcentrifuge tube. The nanoparticle solutions were dilutedusing SSF as a solvent according to the desired concen-trations. Each test included a set of three to five concen-trations and the SSF control group with four replicate tubesfor each individual concentration. Mortality was assessedafter 24 and 48 h to determine the acute toxicities onDaphnia. These tests were conducted in agreement with theOECD Guideline for Testing of Chemicals, “Daphnia sp.,Acute Immobilisation Test and Reproduction Test” [37].

Results and discussion

Gold NP Characterization

UV–vis absorbance spectrometry is one of the primarymethods for the analysis of gold and silver colloids sincethey are plasmonic in nature [48, 49]. The oscillation of theconduction electrons, i.e., the plasmon, is very sensitive tochanges in particle size, dielectric constant of the medium,and aggregation state of the particles. Therefore, freshlysynthesized Au NPs were analyzed between the wave-lengths of 400–800 nm (Fig. 1). Our results for pure AuNPs were consistent with literature having an absorptionmaximum at 523 nm (Fig. 1b) which is indicative of NPs inthe 20 nm range [50]. Particle size and morphology of AuNPs were characterized using AFM by drying the particlesonto a silicon substrate. Figure 2 shows that the particleswere monodispersed with a diameter of ∼15 nm based on z-axis cross sectional analysis of the particles. DLS was usedto monitor the particle size and aggregation state of the NPsin solution, which gave an indication of the chemicalspecies to which the Daphnia were exposed. DLS measure-ments have been previously shown to be helpful forunderstanding the results of nanoparticle toxicity studiesin biological media [41]. As synthesized, Au NPs had afairly uniform particle distribution with an average hydro-dynamic diameter of 21 nm and a relatively smallpolydispersity index (PDI) of 0.20 (Table 1). When theseparticles were placed into standard synthetic freshwater(SSF) at a concentration of 5 mg/L for 0.5- and 24-h timeperiods, the gold NPs underwent a small amount ofaggregation as determined by the shift in diameter to 29.4and 35.7 nm, respectively. Since there was very littlechange in PDI (∼0.2), we assumed that the Daphnia wereexposed was mostly individual gold NPs and not largeaggregates of gold.

Ag NP Characterization

Ag NP solutions were synthesized using different ratios ofsilver nitrate to sodium citrate and the results monitoredusing absorption spectroscopy and DLS. According toabsorption spectroscopy, increasing the amount of reducingagent (citrate) shifted the surface plasmon peak to longerwavelengths, 408 to 433 nm, (Fig. 1a) which was indicativeof an increase in particle size [46]. According to DLS, asthe ratio of Ag:citrate was increased from 1:1.6 to 1:4.2, theaverage particle size (hydrodynamic diameter) increasedfrom 36 to 66 nm. The absorption peaks were relativelybroad suggesting a large distribution in particle sizes, whichwas in agreement with the large polydispersity index (PDI)found by DLS (Table 1). We attempted to use AFM tocorrelate particle size and shape information to the data

692 T. Li et al.

Fig. 2 Scanning AFM image of Au NPs (N≅500), including topography and amplitude images

Fig. 1 Absorption spectra of asilver colloid synthesized withdifferent molar ratios of Ag:citrate and b pure gold colloidand bimetallic particles withdifferent silver to gold ratios. cPhotograph comparing the ap-pearance of the nanoparticlessynthesized

Study of Ag, Au, and Ag–Au bimetallic nanoparticles on Daphnia magna 693

found with absorption spectroscopy and DLS, but the AFMof the Ag NPs showed a high level of polydispersity(results not shown). Further characterization was done withSTEM to determine the state of the originally synthesizedparticles. STEM results of the Ag NP samples (Fig. 3)showed a wide range of NP sizes. In addition, the Ag NPsamples were composed of a mixture of spherical,prismatic, and rod-shaped colloids. Rods had lengths upto 20 times the diameter of the smallest spheres. Overall,the rod particles were present at a concentration of <1% ofthe total colloidal particles. Elemental mapping with XEDSshowed that the Ag NPs were composed primarily of pureAg with only small amounts of sulfur and oxygen present.When the particles were originally dispersed into SSF for0.5 h, some solutions underwent small amounts ofaggregation; however, significant aggregation was detectedin all Ag NP solutions after prolonged exposure to SSF(24 h). Since this was consistent for all three formulationsof Ag NPs, it suggests that the Daphnia were subjected to adynamic suspension of particles, originally containingrelatively small NPs but undergoing significant aggregationover the assay time period. Flame atomic absorptionspectrometry (AAS) was used to quantify the amount ofsilver present in stock solutions after synthesis anddetermine whether excess Ag+ remained after reduction ofsilver nitrate. It was determined that for two of the Ag NPformulations (1:1.6 and 1:3.1), very little Ag was lostduring synthesis; however, for the Ag:citrate ratio of 1:4.2,the [Ag] was significantly lower than expected due to silverplating onto the reaction vessel. This emphasized theimportance of quantifying [Ag] prior to toxicity profiling.When NPs were separated out of the stock solutions bycentrifugation and the supernatants measured using AAS,no [Ag] was detectable in the supernatant, suggesting thatall of the bioavailable Ag was in a NP form.

Bimetallic Ag–Au NP Characterization

Ag–Au (80:20) NPs: while only relatively small shifts(<50 nm) in the absorption spectra are generally seen in

monometallic Ag and Au nanoparticles as a function ofsynthetic method, the spectral properties of Ag–Au bimet-allic nanoparticles have been previously shown to rangebetween 410 and 525 nm, shifting linearly to the red withincreasing Au content [47, 51]. Ag and Au have almostidentical lattice constants (0.408 for Au and 0.409 for Ag),which leads to alloy formation under some syntheticconditions [47, 52]. However, it has also been shown thatbimetallic core-shell nanoparticles can be formed [52, 53]by controlling the synthetic methods used. Some research-ers state that the presence of a single plasmon absorptionpeak is indicative of true alloy formation [51, 53], whileothers have shown that core-shell bimetallic particles canhave multiple plasmon resonance bands [53]. We synthe-sized two different bimetallic Ag–Au NPs, one that wassilver-rich (Ag:Au 80:20) and the other which was gold-rich (Ag:Au 20:80). These particles each displayed a singleabsorption band with a steady red-shift in the particles withincreasing percentage of gold (Fig. 1b).

As expected, the Ag–Au (80:20) NPs exhibited aplasmon resonance at 455 nm which corresponds toparticles that are mostly Ag in character. When the Ag–Au (80:20) NPs were analyzed by STEM, a very unusualresult was observed—only pure Au NPs were detected byXEDS with diameters ranging from 20–27 nm (results notshown). This was in contradiction of the absorptionspectrum which suggested that the particles were mostlyAg in character, since their plasmon resonance (λmax=455 nm) was much closer to that seen in pure Ag NPs(λmax=408 nm) versus pure Au NPs (λmax=523 nm) [54].When Ag–Au (80:20) NP suspensions were analyzed byDLS, the average hydrodynamic diameter of particlespresent in the Ag–Au (80:20) NP suspension was 77 nmwith an extremely narrow particle distribution (PDI=0.09).These particles were extremely stable even when dispersedin SSF at 5 mg/L with little change in average diameter andonly a slight increase in PDI after 24 hrs. There are threepossible scenarios to explain the discrepancy between theabsorption spectroscopy and XEDS measurements, thesesolutions contain (1) a mixture of pure Au NPs and pure Ag

Table 1 DLS results of the nanoparticles used in the toxicity tests

Average diameter in nm (PDI)

NP sample Original solutions NPs in SSF—0.5 h NPs in SSF—24 h

Au 21 (0.21) 29 (0.20) 36 (0.27)

Ag Ag: citrate = 1:1.6 36 (0.32) 58 (0.37) 438 (0.80)

Ag: citrate = 1:3.1 52 (0.40) 71 (0.40) 378 (0.66)

Ag: citrate = 1:4.2 66 (0.24) 59 (0.44) 553 (0.90)

Ag - Au Ag: Au = 20:80 41 (0.30) 14 (0.58) 17 (0.61)

Ag: Au = 80:20 72 (0.08) 73 (0.13) 77 (0.15)

The molar ratio between Ag and citrate is given for Ag NP and molar ratio between Ag and Au is given for bimetallic particles

694 T. Li et al.

NPs, (2) a mixture of Ag ion and Au NPs, or (3) core-shellparticles with a Au-rich shell. Scenario 1 was consideredunlikely because of the monodispersity of the NPsproduced; Ag NPs are generally polydispersed and not inthe same size regime as Au NPs produced using a similarmethodology (Table 1). In addition, previous studies haveshown that co-mixed particle solutions contain two absorp-tion maximums and not the presence of a single peak aswas observed in this case. Further analysis of the solutionsby AAS testing was conducted in attempt to explain thesource of this discrepancy (Table 2). AAS confirmed thepresence of silver in the Ag–Au (80:20) NPs in a solidform, but since no dissolved Ag was detected aftercentrifugal separation of the particulate matter, it eliminatedthe possibility of scenario 2. Since it has been previouslyshown that Au shells can form around Ag NPs, scenario 3still a possibility [55] but one would expect to have seensome signal from Ag in the XEDS analysis. Finally, therelies the possibility that the disagreement in the XEDS andSTEM results are due to a sampling error in the TEManalysis. Further analysis is underway to gain furtherunderstanding of the particles formed in the Ag–Au(80:20) synthesis; however, both absorption spectroscopy

and AAS agree that the particles contain Ag in some formwhich was of the greatest importance to understanding thetoxicity results.

Ag–Au (20:80) NPs: the plasmon resonance for this setof particles showed a maximum absorbance at 508 nm,consistent with particles that are mostly Au in character,either alloyed or core-shell in nature [54–56]. STEManalysis was used to investigate NP size and morphologyof Ag–Au (20:80) NPs (Fig. 4). The STEM images of Ag–Au (20:80) NPs (Fig. 4) suggested that the bimetallic NPshad a spherical shape with the average diameter of ∼30 nm.Elemental analysis with XEDS was carried out at the sametime to study the structural compositions of the particlesand determine whether the particles were alloyed or core-shell structures. XEDS confirmed that the particles werecomposed of both Ag and Au. However, the center of theparticles had a larger than anticipated amount of Au (ave.=88.5%) compared to the molar ratios (80%) used insynthesis. In addition, when the edges of the particles wereanalyzed, it was discovered that they had significantlyhigher Ag concentrations (ave.=33.5%). This suggestedthat instead of forming a uniform alloy, the Ag–Au (20:80)NPs possessed a core-shell structure with a core composed

Fig. 3 STEM images, a brightfield and b dark field, of Ag NPssynthesized via reduction ofAgNO3 by sodium citrate(Ag:citrate=1: 3.1)

Sample Measured [Ag] (mg/L) Calculateda [Ag] (mg/L)

Ag 1:1.6 (supernatant) ND N/A

Ag:Citrate 1:1.6 13.8 17.9

1:3.1 (supernatant) ND N/A

1:3.1 11.9 18.4

1:4 (supernatant) ND N/A

1:4 7.8 22.3

Ag: Au 20:80 (supernatant) ND N/A

20:80 13 27

80:20 (supernatant) ND N/A

80:20 32 31

Table 2 AA results of [Ag]from stock solutions of Ag andAg–Au NP samples

ND not detectable, N/A notapplicablea Calculated based on [Ag] used insynthesis

Study of Ag, Au, and Ag–Au bimetallic nanoparticles on Daphnia magna 695

primarily of Au and a Ag-rich shell. The segregation of Agon the surface of bimetallic NPs has been observedpreviously using different synthetic conditions [55].According to STEM, the size dispersity of these NPs wasrelatively narrow, compared to that found with Ag-alone(Fig. 3), but larger than that seen in pure Au NPs. The DLSresults were in agreement with STEM, in that the particlesin suspension had a polydispersity between that of Au andAg NPs (PDI=0.3) and an average diameter of 41 nm.When these particles were suspended in SSF at 5 mg/L,DLS reported a shift in average hydrodynamic diameter tosmaller particles but an increase in polydispersity. This maybe due to the loss of some of the largest particles due toaggregation and dissolution, while the smaller, more stableparticles remained in solution. This result was consistenteven after 24 h in SSF, suggesting that the Daphnia wereexposed to a variety of particle sizes during the course ofthe assay but much smaller NPs than that used in Ag-onlyNP testing. AAS results from this suspension also con-firmed the presence of Ag in NP form but not in thesupernatant after NP separation (Table 2); however, it wasdiscovered that the [Ag] was lower than expected due toloss during synthesis to the reaction vessel.

Toxicity profiling with D. magna

The possible toxic effects of exposure to Au, Ag, and Ag–Au NPs were investigated using D. magna. Daphnia werechosen because of their status as a common indicator ofenvironmental toxicity by a variety of agencies, includingthe U.S. Environmental Protection Agency, the Organiza-tion for Economic Co-operation and Development, and theAmerican Society for Testing and Materials. They have alsobeen shown to be a strong indicator of aquatic toxicity andto be a valid test model species prior to mammalian testing[57].

The LC50 of Au NPs in D. magna was found to be∼70 mg/L. The daphnids that died during this assessmentshowed changes in motility, prior to becoming static at thebottom of the centrifuge tube and then dying. Previousstudies showed that when exposed to high concentrations ofcolloidal solutions, Daphnia demonstrated changes inbehavior including increased swimming rates and tendencyto go to the surface of the solution [38]. Au ion solutions inthe form of aurochloric acid were also tested and found tohave toxicity around 2 mg/L after 48 h, which wassignificantly more toxic than the NP form of Au. Since it

Fig. 4 STEM images of Ag–Au (20:80) bimetallic NPs. Dark fieldTEM images (a, d) and bright field TEM images (b, e) are shown forrepresentative particles. The results from XEDS elemental analyses

are shown in tabular form (c, f) for the respective images. XEDSdemonstrates that most of the silver is located on the edges of theparticles, suggesting a core-shell structure

696 T. Li et al.

required significantly large concentrations of Au NPs tohave an acute effect on Daphnia, non-fatal effects such as adecrease in birth rate or changes in embryo developmentwere also monitored. No statistically significant differenceswere found in either birth rates or embryo developmentwhen daphnids were exposed to Au NPs for 8 days at sub-lethal dosages of 10 mg/L. However, daphnids exposed tothis concentration of Au NPs did have a statistically higherdeath rate than control daphnids not exposed to NPs overthe same time period. The NPs appeared to adhere to theexternal appendages of the daphnids and caused changes inswimming patterns. In addition, daphnids exposed to AuNPs were observed to molt more frequently than controldaphnids, which may have contributed to the shortening oftheir lifespans. The toxicity of gold NPs on Daphnia at highconcentrations may be due to inhibition of nutrient uptakewithin the gut, since it has been previously shown that AuNPs concentrate within the gut prior to being eliminated[38].

The difference in acute toxicity between Au and Agcontaining solutions was a key focus of this study. SinceAg+ is known to inhibit sodium influx in aquatic species[45] and Ag NPs previously caused embryonic develop-ment problems in zebrafish [26], we tested both the ionicand NP forms of Ag for their toxicity towards D. magna.Food was eliminated from all ion experimental testingvessels during the 48-h test, because the presence of algaeled to metal reduction and NP formation. In addition, sincethe presence of organic matter has been previously shownto inhibit silver ion uptake by Daphnia [27], we wanted tobe able to differentiate effects of unmodified NPs ofdifferent compositions. The LC50 concentrations (Fig. 5)are shown for the calculated [Ag] (Fig. 5a) and measured[Ag] values (Fig. 5b). The 48-h LC50 for silver ion wasdetermined to be 2 μg/L which was in general agreementwith literature [58], making it the most toxic species withinour test series. In addition, daphnids were found to die

within a 2-h timeframe when exposed to Ag+ concentra-tions greater than 3 mg/L.

For the Ag NPs produced using different ratios of silverion to citrate, the toxicity was found to decrease withincreasing amounts of added citrate (corresponding to anincrease in particle size) when the calculated values of [Ag]were used (Fig. 5a), but to be similar to one another once[Ag] loss was taken into account (Fig. 5b). The medianlethal concentration was found to be ∼3 μg/L, which islarger than that found for silver oxide coated Ag NPs in D.pulex [25]. It was initially thought that the apparent increasein toxicity of lower Ag:citrate ratios might be due to thepresence of excess Ag+ in solution; however, AAS resultsafter separation of NPs from the stock solutions showedthat there was not a significant amount of Ag+ present insolution (Table 2). In addition, toxicity did not correlatewith particle size once final [Ag] was taken into account(Fig. 5b) though the results using initial [Ag] suggested thattoxicity increased as particle size decreased (Fig. 5a). DLStests of the Ag NPs in SSF did not show significantaggregation of the particles within the first half-hour ofbeing placed into SSF; however the unmodified Ag NPs didaggregate significantly after 24 h in SSF water (Table 1).This increase in particle aggregation may be one reasonwhy toxicity did not correlate to particle size in this study;though previous studies of Ag+ uptake in Daphnia showedthat this is an extremely fast process, occurring in less than1 h [27]. The mechanism of Ag NP toxicity is stillunknown [25]; however, it is known that the soluble iondisrupt sodium ion influx and can lead to mortality [28].

The toxicity assessments of Ag–Au bimetallic NPs werealso conducted at 48 h without food supply. Our originalhypothesis was that bimetallic particles would havesignificantly reduced toxicities, based on the many reportsof Ag–Au bimetallic particles forming alloys. If theparticles were to form a uniform alloy, then fewer Agatoms would present themselves on the particle’s surface,

Fig. 5 Toxic effects of aqueousnanoparticle solutions and silverion on Daphnia at 48 h, repre-sented as LC50 values in termsof a calculated [Ag] and bcorrected [Ag] values accordingto AA

Study of Ag, Au, and Ag–Au bimetallic nanoparticles on Daphnia magna 697

which should reduce NP toxicity. Since Au was shown tobe approximately 1,000-fold less toxic than Ag (70 mg/Lversus 3 μg/L), it was expected that the difference intoxicity between Ag NPs and bimetallic Ag–Au NPs wouldbe significant. If the ratio used in synthesis was represen-tative of the atoms present on the surface, then it would beexpected that the toxicity of the Ag–Au (80:20) NPs wouldbe close to that found in Ag NPs. The LC50 value for theAg–Au (80:20) NPs at 15 μg/L was actually significantlyhigher than expected, and therefore less toxic (Fig. 5). Thiswas very surprising and could be explained by theformation of a gold shell on the surface of these particles,as suggested previously, which would decrease the bio-availability of Ag. Another possibility was that some of theAg present was in the form of AgCl, which has been shownto be significantly less toxic to daphnids than Ag+ [28]. Ifthe LC50 was stated in terms of overall metal concentration[Ag+Au], the difference was actually greater with an LC50

of 19 μg/L for Ag–Au (80:20) NPs.The LC50 of the Ag–Au (20:80) NPs was determined to

be 12 μg/L in terms of [Ag], which means that the particlesare slightly more toxic than what was expected from the Agmole ratio of the particles (15 μg/L). The reason for thisdifference may be due to the fact that these particles consistof a Ag-enriched shell. When the LC50 was stated in termsof overall [Ag+Au] the result would be 59 μg/L, whichshowed that the introduction of Au into the NP did inhibitoverall NP toxicity. It is possible that since the bimetallicNPs were shown not to aggregate significantly, comparedto Ag NPs, that the aggregation state of the NPs also had animpact of the observed toxicity level. Overall, the ingestionof Ag-containing NPs allowed a significant amount of theAg atoms present to interact with the Daphnia uponingestion. This may lead to a highly localized release ofsilver ion to the organism, resulting in a greater impact thanexpected; however, the active component is currentlyunknown.

Conclusion

Comparative toxicity assessments of a series of well-characterized Au, Ag, and bimetallic Ag–Au NPs wereinvestigated with D. magna, a freshwater filter-feedingcrustacean commonly used in environmental studies. Basedon the experimental results, it can be concluded that all theparticles tested have shown dose-dependent eco-toxicological effects on D. magna. The LC50 value of AuNPs was found to be ∼1,000× greater than Ag-based NPs,thereby exhibiting a much lower toxic effect on daphnids.AAS was used to measure the concentration of silver, [Ag],in Ag NP and Ag–Au bimetallic NP solutions. Themeasured [Ag] included all forms of silver, including ionic

and NP forms based on calibration data. For all suspensionsused in toxicity testing, NPs were analyzed in theirsynthesized form, as well as after a separation step wasconducted. In all cases, the dissolved Ag was below thelimit of detection for the technique, which means that over95% of the [Ag] present was in a solid form and notthe ionic form, suggesting that the active component in thetoxicity testing of Ag NPs and bimetallic NPs was thereduced form of silver and not the presence of excessdissolved Ag+. Changing the ratio between silver nitrateand sodium citrate during Ag NP synthesis led to anincrease in the average particle size produced; however, nodifference in toxicity was observed once the overall [Ag]was corrected using AAS results. Ag NPs were shown toaggregate significantly in standard synthetic freshwater,which may contribute to the NP toxicity levels.

The toxicity effects of Ag–Au bimetallic NPs wereobserved between that of Ag NPs and Au NPs; howevermuch closer to that of Ag NPs. Ag–Au (80:20) NPs weresignificantly less toxic than expected and Ag–Au (20:80)NPs were slightly more toxic than expected. We have takengreat care to ensure that the particles tested were synthe-sized in the same manner, using citrate as both the reducingand capping agent for the nanoparticles which keeps theirsurface chemistry consistent. Since citrate-reduced Ag NPsdemonstrate a significant amount of aggregation, it may beof interest to investigate NPs with protective surfacechemistries and assess their role in toxicity. In this regard,our future research will be focused on the toxicity study ofdifferent surface coating particles using D. magna.

Acknowledgments Financial funding for this work was providedthrough start-up funds provided to S.W.B. by the College of Scienceand Letters at the Illinois Institute of Technology. T.L. was fundedthrough the Educational and Research Initiative Fund by the IITGraduate College. M.R. and S.K. were funded through a ProjectSEED grant given by the American Chemical Society. L.I. was fundedby an ACS Fellows grant given by the American Chemical Society.Additional support for materials and supplies were funded through thePhysical Science Initiative Cohort Project by the Illinois State Boardof Education. The authors thank Dr. Alan Nicholls, Interim AssociateDirector of the University of Illinois at Chicago’s Research ResourceCenter, for assistance with TEM collection. The authors also thankKangmin Xu and Professor Xiaoping Qian for assistance with AFMmeasurements. The authors also thank Y. Huang, Y.J. Lin, D. Jezek,Professor Mitch Dushay, and Y. Cai for helpful discussions.

References

1. Baraton MI, Merhari L (2004) Nanoparticles-based chemicalgas sensors for outdoor air quality monitoring microstations.Mater Sci Eng B Solid State Mater Adv Technol 112:206–213

2. Lim JK, Joo SW (2006) Gold nanoparticle-based pH sensor inhighly alkaline region at pH > 11: surface-enhanced Ramanscattering study. Appl Spectrosc 60:847–852

698 T. Li et al.

3. Bishnoi SW, Rozell CJ, Levin CS, Gheith MK, Johnson BR,Johnson DH, Halas NJ (2006) All-optical nanoscale pH meter.Nano Lett 6:1687–1692

4. Jayakumar R, Chennazhi KP, Muzzarelli RAA, Tamura H, NairSV, Selvamurugan N (2010) Chitosan conjugated DNA nano-particles in gene therapy. Carbohydrate Polymers 79:1–8

5. Taton TA, Mirkin CA, Letsinger RL (2000) Scanometric DNAarray detection with nanoparticle probes. Science 289:1757–1760

6. Cai C, Bakowsky U, Rytting E, Schaper AK, Kissel T (2008)Charged nanoparticles as protein delivery systems: a feasibilitystudy using lysozyme as model protein. Eur J Pharm Biopharm69:31–42

7. Grubisha DS, Lipert RJ, Park H-Y, Driskell J, Porter MD (2003)Femtomolar detection of prostate-specific antigen: an immunoas-say based on surface-enhanced Raman scattering and immunogoldlabels. Anal Chem 75:5936–5943

8. Thaxton CS, Georganopoulou DG, Mirkin CA (2006) Goldnanoparticle probes for the detection of nucleic acid targets. ClinChim Acta 363:120–126

9. Yanez-Sedeno P, Pingarron JM (2005) Gold nanoparticle-basedelectrochemical biosensors. Anal Bioanal Chem 382:884–886

10. Van Hyning DL, Zukoski CF (1998) Formation mechanisms andaggregation behavior of borohydride reduced silver particles.Langmuir 14:7034–7046

11. Adair JH, Suvaci E (2000) Morphological control of particles.Curr opin colloid interface sci 5:160–167

12. Jana NR, Gearheart L, Murphy CJ (2001) Evidence for seed-mediated nucleation in the chemical reduction of gold salts to goldnanoparticles. Chem Mater 13:2313–2322

13. Velikov KP, Zegers GE, van Blaaderen A (2003) Synthesis andcharacterization of large colloidal silver particles. Langmuir19:1384–1389

14. D'Souza L, Suchopar A, Richards RM (2004) In situ approachesto establish colloidal growth kinetics. J Colloid Interface Sci279:458–463

15. Tiede K, Hassellov M, Breitbarth E, Chaudhry Q, Boxall ABA(2009) Considerations for environmental fate and ecotoxicitytesting to support environmental risk assessments for engineerednanoparticles. J Chromatogr A 1216:503–509

16. Ostrowski A, Martin T, Conti J, Hurt I, Harthorn B (2009)Nanotoxicology: characterizing the scientific literature, 2000–2007. J Nanopart Res 11:251–257

17. Wijnhoven SWP, Peijnenburg W, 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, SipsA, Geertsma RE (2009) Nano-silver—a review of available dataand knowledge gaps in human and environmental risk assessment.Nanotoxicology 3:109–U178

18. Stevens KNJ, Crespo-Biel O, van den Bosch EEM, Dias AA,Knetsch MLW, Aldenhoff YBJ, van der Veen FH, Maessen JG,Stobberingh EE, Koole LH (2009) The relationship between theantimicrobial effect of catheter coatings containing silver nano-particles and the coagulation of contacting blood. Biomaterials30:3682–3690

19. Prema P, Raju R (2009) Fabrication and characterization of silvernanoparticle and its potential antibacterial activity. BiotechnolBioprocess Eng 14:842–847

20. Gao J, Youn S, Hovsepyan A, Llaneza VL, Wang Y, Bitton G,Bonzongo JCJ (2009) Dispersion and toxicity of selectedmanufactured nanomaterials in natural river water samples: effectsof water chemical composition. Environ Sci Technol 43:3322–3328

21. Benn TM, Westerhoff P (2008) Nanoparticle silver released intowater from commercially available sock fabrics. Environ SciTechnol 42:4133–4139

22. Shahverdi AR, Fakhimi A, Shahverdi HR, Minaian S (2007)Synthesis and effect of silver nanoparticles on the antibacterialactivity of different antibiotics against Staphylococcus aureusand Escherichia coli. Nanomed Nanotechnol Biol Med 3:168–171

23. Simon-Deckers A, Brun E, Gouget B, Carriere M, Sicard-Roselli C(2008) Impact of gold nanoparticles combined to X-ray irradiation onbacteria. Gold Bulletin 41:187–194

24. Mueller NC, Nowack B (2008) Exposure modeling of engineerednanoparticles in the environment. Environ Sci Technol 42:4447–4453

25. Griffitt RJ, Luo J, Gao J, Bonzongo JC, Barber DS (2008) Effectsof particle composition and species on toxicity of metallicnanomaterials in aquatic organisms. Environ Toxicol Chem27:1972–1978

26. Asharani PV, Wu YL, Gong ZY, Valiyaveettil S (2008)Toxicity of silver nanoparticles in zebrafish models. Nanotech-nology 19:8

27. Glover CN, Wood CM (2005) Accumulation and elimination ofsilver in Daphnia magna and the effect of natural organic matter.Aquat Toxicol 73:406–417

28. Glover CN, Wood CM (2004) Physiological interactions of silverand humic substances in Daphnia magna: effects on reproductionand silver accumulation following an acute silver challenge.Comp Biochem Physiol C Toxicol Pharmacol 139:273–280

29. AshaRani PV, Low Kah Mun G, Hande MP, Valiyaveettil S(2008) Cytotoxicity and genotoxicity of silver nanoparticles inhuman cells. ACS Nano 3:279–290

30. Choi O, Hu ZQ (2008) Size dependent and reactive oxygenspecies related nanosilver toxicity to nitrifying bacteria. EnvironSci Technol 42:4583–4588

31. Carlson C, Hussain SM, Schrand AM, K. Braydich-Stolle L, HessKL, Jones RL, Schlager JJ (2008) Unique cellular interaction ofsilver nanoparticles: size-dependent generation of reactive oxygenspecies. J Phys Chem B 112:13608–13619

32. Morones JR, Elechiguerra JL, Camacho A, Holt K, Kouri JB,Ramirez JT, Yacaman MJ (2005) The bactericidal effect of silvernanoparticles. Nanotechnology 16:2346–2353

33. Oberdorster G, Oberdorster E, Oberdorster J (2005) Nanotoxicol-ogy: an emerging discipline evolving from studies of ultrafineparticles. Environ Health Perspect 113:823–839

34. AshaRani PV, Mun GLK, Hande MP, Valiyaveettil S (2009)Cytotoxicity and genotoxicity of silver nanoparticles in humancells. ACS Nano 3:279–290

35. ASTM Standard E1193 - 97, 2004 "Standard Guide for Conduct-ing Daphnia magna Life-Cycle Toxicity Tests". (West Consho-hocken, PA: ASTM International)

36. EPA 2002 Methods for Measuring the Acute Toxicity of Effluentsand Receiving Waters to Freshwater and Marine Organisms. edAgency U S E P

37. OECD 1984 Daphnia sp., acute immobilisation test and repro-duction test. In: Part I - the 24 h EC 50 acute immobilisation test,pp 2 - 5

38. Lovern SB, Owen HA, Klaper R (2008) Electron microscopy ofgold nanoparticle intake in the gut of Daphnia magna. Nano-toxicology 2:43–48

39. Lovern SB, Klaper R (2006) Daphnia magna mortality whenexposed to titanium dioxide and fullerene (C-60) nanoparticles.Environ Toxicol Chem 25:1132–1137

40. Zhu XS, Zhu L, Chen YS, Tian SY (2009) Acute toxicities of sixmanufactured nanomaterial suspensions to Daphnia magna. JNanopart Res 11:67–75

41. Murdock RC, Braydich-Stolle L, Schrand AM, Schlager JJ,Hussain SM (2008) Characterization of Nanomaterial Dispersionin Solution Prior to In Vitro Exposure Using Dynamic LightScattering Technique. Toxicol Sci 101:239–253

Study of Ag, Au, and Ag–Au bimetallic nanoparticles on Daphnia magna 699

42. Ahamed M, Karns M, Goodson M, Rowe J, Hussain SM,Schlager JJ, Hong YL (2008) DNA damage response to differentsurface chemistry of silver nanoparticles in mammalian cells.Toxicol Appl Pharmacol 233:404–410

43. Thevenot P, Cho J, Wavhal D, Timmons RB, Tang LP (2008)Surface chemistry influences cancer killing effect of TiO2 nano-particles. Nanomed Nanotechnol Biol Med 4:226–236

44. Kora AJ, Manjusha R, Arunachalam J (2009) Superiorbactericidal activity of SDS capped silver nanoparticles:synthesis and characterization. Mater Sci Eng C Mater BiolAppl 29:2104–2109

45. Evanoff DD, ChumanovG (2005) Synthesis and optical properties ofsilver nanoparticles and arrays. Chemphyschem 6:1221–1231

46. Pillai ZS, Kamat PV (2004) What factors control the size andshape of silver nanoparticles in the citrate ion reduction method? JPhys Chem B 108:945–951

47. Pal A, Shah S, Kulkarni V, Murthy RSR, Devi S (2009) Templatefree synthesis of silver-gold alloy nanoparticles and cellularuptake of gold nanoparticles in Chinese Hamster Ovary cell.Mater Chem Phys 113:276–282

48. Hayat MA (1989) Colloidal gold: principles, methods, andapplications. Academic, San Diego

49. Chumanov G, Sokolov K, Cotton TM (1996) Unusual extinctionspectra of nanometer-sized silver particles arranged in two-dimensional arrays. J Phys Chem 100:5166–5168

50. Tullman JA, Finney WF, Lin YJ, Bishnoi SW (2007) Tunableassembly of peptide-coated gold nanoparticles. Plasmonics 2:119–127

51. Mallin MP, Murphy CJ (2002) Solution-phase synthesis of sub-10 nm Au–Ag alloy nanoparticles. Nano Lett 2:1235–1237

52. Yang Y, Gong X, Zeng H, Zhang L, Zhang X, Zou C, Huang S(2010) Combination of digestive ripening and seeding growth as ageneralized route for precisely controlling size of monodispersednoble monometallic, shell thickness of core - shell and composi-tion of alloy nanoparticles. The Journal of Physical Chemistry C114:256–264

53. Gonzalez CM, Liu Y, Scaiano JC (2009) Photochemical strategiesfor the facile synthesis of gold–silver alloy and core-shellbimetallic nanoparticles. The Journal of Physical Chemistry C113:11861–11867

54. Link S, Wang ZL, El-Sayed MA (1999) Alloy formation of gold-silver nanoparticles and the dependence of the plasmon absorptionon their composition. J Phys Chem B 103:3529–3533

55. Chen HM, Liu RS, Jang LY, Lee JF, Hu SF (2006) Characteriza-tion of core-shell type and alloy Ag/Au bimetallic clusters byusing extended X-ray absorption fine structure spectroscopy.Chem Phys Lett 421:118–123

56. Srnova-Sloufova I, Lednicky F, Gemperle A, Gemperlova J(2000) Core-shell (Ag)Au bimetallic nanoparticles: analysis oftransmission electron microscopy images. Langmuir 16:9928–9935

57. Martins J, Oliva Teles L, Vasconcelos V (2007) Assays withDaphnia magna and Danio rerio as alert systems in aquatictoxicology. Environ Int 33:414–425

58. Wood C M, La Point TW, Armstrong D E, Birge WJ, Brauner C J,Brix KV, Call D J, Crecelius E A, Davies P H, Gorsuch JW,Hogstrand C, Mahony J D, McGeer J C and O’Connor T P 2002Silver in the environment: transport, fate, and effects, ed Andren AW and Bober TW (Pensacola, Florida: SETAC) pp 25–63

700 T. Li et al.