Science of the Total Environment 409 (2011) 1351–1356

Contents lists available at ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r.com/ locate /sc i totenv

Synergistic toxic effect of nano-TiO2 and As(V) on Ceriodaphnia dubia

Demin Wang a, Ji Hu a,c, David R. Irons a, Jianmin Wang a,b,⁎a Department of Civil, Architectural and Environmental Engineering, Missouri University of Science and Technology, Rolla, MO 65409, USAb State Key Laboratory of Pollution Control and Resource Reuse, College of Environmental Science and Engineering, Tongji University, Shanghai 200092, Chinac Key Laboratory of Marine Chemistry Theory and Technology, Ministry of Education, Ocean University of China, Qingdao 266100, China

⁎ Corresponding author. Department of Civil, ArchEngineering, Missouri University of Science and TechnTel.: +1 573 341 7503; fax: +1 573 341 4729.

E-mail address: wangjia@mst.edu (J. Wang).

0048-9697/$ – see front matter © 2010 Elsevier B.V. Aldoi:10.1016/j.scitotenv.2010.12.024

a b s t r a c t

a r t i c l e i n f o

Article history:Received 17 October 2010Received in revised form 6 December 2010Accepted 17 December 2010Available online 15 January 2011

Keywords:Nanotoxicologyn-TiO2

ArsenicSynergistic toxic effect

Due to the active development and application of nanotechnology, engineered nanomaterials (ENMs) arebecoming a new class of environmental pollutants that may significantly impact the environment and humanhealth. While many toxicity investigations have been conducted, there is little information about thesynergistic effect of ENMs and other toxic compounds in the environment. In order to extend this knowledge,the combined effect of TiO2 nanoparticles (n-TiO2) and As(V) were investigated. High concentrations of As(V)can accumulate on the n-TiO2 surface. Cultured Ceriodaphnia dubia (C. dubia) species were used to examinethe synergistic toxic effect through exposure to 1) n-TiO2 suspensions, 2) As(V) solutions, and 3) mixtures ofn-TiO2 and As(V) suspensions. Results showed that n-TiO2 alone was not toxic when the concentration wasless than 400 mg/L and that the 24-hour median lethal concentration (LC50) of As(V) alone was 3.68±0.22 mg/L. However, in the presence of low concentrations of n-TiO2, the toxicity of As(V) increasedsignificantly. At the same initial As(V) concentration, the toxicity of n-TiO2 first increased, reached amaximum, and then decreased with an increase in n-TiO2 concentration. Hydrodynamic size and sorptioncapacity were most important parameters for toxicity.

itectural and Environmentalology, Rolla, MO 65409, USA.

l rights reserved.

© 2010 Elsevier B.V. All rights reserved.

1. Introduction

According to current predictions, the sales of nanotechnology-integrated products could total more than a trillion dollars by 2015(Nanowerk, 2007). This rapid growth could result in the release of largequantities of engineered nanomaterials (ENMs) into the environment ifappropriate regulations and controls are not implemented quickly. Therelease of ENMs from commercial products into the aquatic environmenthas already been reported (Benn and Westerhoff, 2008; Geranio et al.,2009); however, regulations pertaining to ENMs are still in the earlystages due to lack of sufficient information (Anthony et al., 2010;Chatterjee, 2008; Choi et al., 2009; Wiesner et al., 2009). Beforeregulations can move forward, a more complete understanding of thetoxicity of ENMs is necessary.

Previous research has examined potential human health impacts(Nel et al., 2006; Oberdörster et al., 2005) and ecological effects ofENMs (Baun et al., 2008; Klaine et al., 2008; Moore, 2006). Whilethese findings provide valuable information on the toxicity of ENMswhen they act alone, it is rare in the environment that there are noother toxicants (such as heavy metals) present. In the environment,ENMs can interact with these toxicants to give different responses ontarget organisms. Significant accumulations of toxic substances at

nanoparticle surfaces have been reported (Guan et al., 2009; Guptaand Ghosh, 2009; Hristovski et al., 2007, 2008). In principle, theseaccumulated toxic species play important roles in the dose–responserelationship in a toxicity test; in this case, organisms are affected notonly by the ENMs but also by the accumulated toxicants (Kim et al.,2010; Sun et al., 2009). Some researchers have observed an enhancedtoxic effect caused by ENM impurities (Gao et al., 2009; Griffitt et al.,2009, 2008) or mixtures of toxicants and ENMs (Kim et al., 2010; Sunet al., 2009). On the contrary, other research has shown decreasingtoxicity of environmental contaminants when particulate matter ororganic ligands were present (Clements and Kiffney, 1994; Cooperet al., 2009; Ma et al., 2002, 1999; Mahar andWatzin, 2005). As of yet,toxic effects of ENMs in the presence of other environmentalcontaminants have seldom been reported. It is believed that thesetoxic effects are more complex due to tremendous uncertaintiesremaining, even for ENMs alone (Wiesner et al., 2009). Theinteractions in these complex systems are not only limited to theinteractions of the toxicants with the organism, but also include theinteractions of ENMs with the organism and the combined effects ofENMs and toxicants on the organism. To probe the actual effect inthese realistic conditions, it is necessary to examine the response ofaquatic species to ENMs in the presence of other environmentaltoxins.

The TiO2 nanoparticle (n-TiO2) is a commercially important ENM.The toxic effect of n-TiO2 alone on daphnia species has been evaluated.However, LC50 values reported by different researchers are inconsistentdue to the different sources and surface properties of the ENMs used in

1352 D. Wang et al. / Science of the Total Environment 409 (2011) 1351–1356

the experiments (Hall et al., 2009;Hund-Rinke and Simon, 2006; Lovernand Klaper, 2006; Lovern et al., 2007; Warheit et al., 2007). Arsenic is ahighly toxic contaminant found in groundwater in many regions in theworld. The toxic effect of arsenic alone can also be found in the literature(He et al., 2009; L. Spehar and T. Fiandt, 1986; Shaw et al., 2007). Theobjective of this study was to determine the synergistic toxic effects ofn-TiO2 and As(V) on Ceriodaphnia dubia (C. dubia).

2. Materials and methods

2.1. Chemicals and nanomaterials

CaSO4·2H2O (98%), KCl (99%), Na2SeO4 (99%), CuCl2·2H2O, NaHCO3

(100.2%, Asb5.0 mg/kg),MgSO4 (Asb0.001%), andHNO3 (67% inpurity)werepurchased fromFisher Scientific (Pittsburgh, PA).NaH2AsO4·7H2Owas purchased from Alfa Aesar (Ward Hill, MA). TiO2 (anatase, 99%)nanoparticle (see Appendix A for characterization)was purchased fromSkyspring Nanomaterials Inc. (Houston, TX, USA). Millipore water wasproduced using a Synergy® ultrapure water system (Billerica, MA).

2.2. Culturing of C. dubia

Starter C. dubia were purchased from MBL Aquaculture (Sarasota,FL, USA) and the food (YTC and Algae) was purchased from ABS Inc.(Fort Collins, Co, USA). Mass culture, individual culture, and toxicitytests were performed in a SVC-6AX laminar flow hood (Streamline®laboratory products, Fort Myers, FL, USA) to prevent contaminationfrom any other source of particles in the air. The culture medium wasprepared by dissolving appropriate amounts of NaHCO3, CaSO4·2H2O,MgSO4, KCl, and Na2SeO4 into Millipore water according to the EPAstandardmethod (USEPA, 2002), which has a pH of 7.8 and a hardnessof 85 mg/L as CaCO3. Healthy neonates, with an age of less than 24 hafter 7-day individual culturing (three-brood) were used for thisstudy.

2.3. Toxicity test of As(V) or n-TiO2

Stock As(V) solution (150 mg/L) was prepared by dissolving thenecessary amount of NaH2AsO4 into 1 L of Millipore water. OperatingAs(V) solutions were prepared by diluting the As(V) stock solutionusing culture medium in six plastic bottles (125 mL, Nalgene). TiO2

suspensions were prepared by adding appropriate amounts of n-TiO2

into a 100 mL culture medium in six other plastic bottles. Thesuspensions were mixed with food before performing a toxicity test.Five medicine cups (30 mL polyethylene; Fisher) were used asreactors for each concentration and 15 mL of an As(V) or nano-TiO2

suspension were added to each cup. The EPA standard method(USEPA, 2002) was followed in other parts of the experiment toexamine toxic effects. The toxicity experiments were repeated at leastthree times. The residual suspensions were used to analyze theresidual concentration of As(V) and to characterize the n-TiO2 inaqueous suspensions. Due to the settling of n-TiO2 as a function oftime, the suspensions were renewed every 48 h. The residualconcentration of TiO2 after 48 h was found above 10 mg/L with aninitial concentration of 200 mg/L.

2.4. Synergistic effect of As(V) and n-TiO2

To investigate the synergistic effect, two types of experimentswere conducted. The first set of experiments examined the toxic effectof As(V) with fixed n-TiO2 concentrations. The other set of experi-ments studied the toxicity of TiO2 with fixed As(V) concentrations. Allexperiments were repeated at least three times for statistics analysis.

For the first set of experiments, a series of As(V) solutions atconcentrations of 0, 0.45 mg/L, 0.75 mg/L, 1.5 mg/L, 2.25 mg/L, and4.5 mg/L were prepared by diluting the stock solution using the

culture medium. The volume of these solutions was 100 mL. Sameamount of n-TiO2 was added while mixing. The concentrations ofn-TiO2 used for different experiments were 1.00±0.05 mg/L, 10±2 mg/L, 50±5 mg/L, 100±5 mg/L, 200±5 mg/L, and 300±5 mg/L,respectively. Seventy-five milliliters of the final suspensions weredivided equally into fivemedicine cups to initiate the toxicity test. Theresidual 25 mL suspensions were used to analyze residual As(V)concentrations.

For the second experiment, As(V) solutions at constant concentrationswere prepared by diluting the stock solution using the culturemedium insix bottles (125 mL) to a volume of 100mL. The TiO2 nanoparticle wasweighed and added to these bottles at different concentrations. After theaddition of food, followed by mixing, toxicity tests were performed andresidual solutions were analyzed to determine As(V) concentrations. Theinitial concentrations of As(V) were 2.5 mg/L and 3.0 mg/L, respectively.

2.5. Sorption of As(V) on n-TiO2

Fifty milliliter solutions with different concentrations of As(V)solutions (up to 15 mg/L)were prepared by diluting the stock solutionwith the culturemedium in polypropylene tubes (50 mL). These tubeswere covered and mixed in a shaker for 10 min. The pH values of theAs(V) solutions were in the range of 7.6–7.8. TiO2 particles wereadded to each tube at a concentration of 200±5 mg/L. The mixedsuspensions were shaken for 24 h to achieve sorption equilibrium.

2.6. As(V) analysis

After toxicity or sorption experiments, residual suspensions werecollected and transferred to 50 mL centrifugal tubes. Theweight of eachsamplewas 30.00±0.05 g (including the tube itself). The actual volumeof the As(V) aliquots was approximately 12 mL. The aliquots werecentrifuged at 15,000 rpm for 20 min at 20 °C. After centrifuging,supernatants were collected and acidified using a 0.2% HNO3 solution.Depending on the initial concentration of the solution, the supernatantswere diluted with a dilution factor in the range of 2 to 10. Theconcentration of As(V) in the dilute solution was determined usingGFAA (PerkinElmer AAnalyst™ 600), and the residual As(V) concen-tration in the original suspensions was calculated based on dilutionfactors.

3. Results

3.1. Toxic effect of As(V) or n-TiO2 on C. dubia

The toxic effect of As(V) or n-TiO2 alone on C. dubia has beeninvestigated by other researchers. In order to obtain experimentalresults under the same experimental condition for this research, thetoxicity experiments for As(V) and for n-TiO2 alone were repeated toserve as controls on assessing synergistic effects. The experimentswere carried out by following the procedure described in the standardmethod of USEPA (2002). The results were considered valid when thesurvival rate in the control group (without contaminants) was greaterthan 90%. Figs. 1 and 2 show the results of the toxic effects of As(V)and n-TiO2, respectively.

Fig. 1 shows the acute toxicity of As(V) on C. dubia at differentconcentrations and durations. Mortality was observed when theconcentration of As(V) was greater than 2.0 mg/L. In a concentrationrange of 2.0 mg/L to 6.0 mg/L, partial mortality was observed. Basedon these experiments, the LC50 was estimated to be 3.68±0.22 mg/Lin 24 h, with the TSK program suggested by EPA (USEPA) andTOXCALC Software (v 5.0.32) from Tidepool Scientific SoftwareCompany (McKinley Ville, CA, USA). This value is close to the resultsreported by Shaw et al. (2007) with a value of 2.5 mg/L to 3.9 mg/L.

Fig. 2 shows the acute toxicity test results of n-TiO2 on C. dubia atdifferent durations. There was nomortality when the concentration of

Total As Concentration (mg/L)

0 2 4 6 8

Mo

rtal

ity

(%)

0

20

40

60

80

100

24hr48hr72hr96hr

Fig. 1. Toxic effect of As(V) on C. dubia at different durations. Experiment conditions:solution=culture medium, volume=15 mL, pH=7.8, hardness=85 mg/L as CaCO3,T=20 °C.

As(V) Concentration (mg/L)

0 1 2 3 4 5

Mo

rtal

ity

(%)

0

20

40

60

80

100As(V) onlyTiO2 +As (initial)TiO2 +As (residual)

Fig. 3. Toxic effect of As(V) in the presence of n-TiO2 particles. Experiment conditions:suspension=culture medium and TiO2, volume=15 mL, pH=7.8, hardness=85mg/Las CaCO3, T=20 °C, TiO2 concentration=50mg/L.

1353D. Wang et al. / Science of the Total Environment 409 (2011) 1351–1356

n-TiO2 was less than 400 mg/L. When the n-TiO2 concentration washigher, mortality of C. dubia was found. However, we suspected thatthe death resulted mainly from the rising of body weight rather thanthe toxicity of n-TiO2. The accumulation of nanoparticles increasedthe body weight of C. dubia significantly and limited their mobility(see Appendix A). Different toxic effects of n-TiO2 on daphnia specieshave been reported in the literature (Hall et al., 2009; Warheit et al.,2007; Zhu et al., 2009). Hall et al. and Zhu et al. reported a LC50 valueof 7.6 mg/L and 143.4 mg/L, respectively. However, Warheit et al.stated that n-TiO2 was a low concern in aquatic environments. In thepresent research, we considered n-TiO2 as non-lethal to C. dubia if theconcentrationwas less than 400 mg/L. Therefore, the synergistic effectwas only examined when the concentrations of n-TiO2 were less than300 mg/L.

3.2. Synergistic toxic effect of As(V) and n-TiO2

While there was no significant toxic effect of n-TiO2 on C. dubia atconcentrations of less than 300 mg/L, a synergistic effect may occur ifAs(V) is also present in the system. To validate this hypothesis, thetoxic effects of As(V) were examined in the presence of n-TiO2

particles. Figs. 3 and 4 show the experimental results.Fig. 3 shows the toxicity of As(V) on C. dubiawith and without the

presence of 50 mg/L n-TiO2. The mortality was also plotted using thetotal (or initial) As(V) concentration and the dissolved (or residual)As(V) concentration, respectively. Results indicated that the As(V)

TiO2 Concentration (mg/L)

0 200 400 600 800

Mo

rtal

ity

(%)

0

20

40

60

80

10024 hr48hr72hr96hr

Fig. 2. Toxic effect of n-TiO2 at different durations. Experiment conditions: suspension=culture medium and TiO2, volume=15mL, pH=7.8, hardness=85mg/L as CaCO3,T=20 °C.

toxicity was significantly enhanced by 50 mg/L n-TiO2 in the culturemedium. The calculated LC50 for As(V) based on the total As(V)concentration was only 1.43 mg/L, a smaller value than the LC50 for As(V) alone (3.68 mg/L). In addition, the toxicity based on the residualAs(V) concentration in the medium was much more severe. Thecalculated As(V) LC50 for As(V) was only 0.89 mg/L.

Fig. 4 shows the effect of various concentrations of n-TiO2 for thesame initial As(V) concentration range. All experiments wereconducted with a 24 hour duration and n-TiO2 concentrations of 1.0,10.0, 50.0, 100, 200, and 300 mg/L, respectively. A peakmortality valuewas observed at a concentration of 50 mg/L n-TiO2. Under lower (1 and10 mg/L) and higher (200, 300 mg/L) concentrations of n-TiO2, theenhancing effect was less pronounced. It is suspected that not only theresidual As(V) but also the adsorbed As(V) contributed to the overalltoxic effect on the model organisms, i.e. C. dubia. When the n-TiO2

concentration is greater than 50 mg/L, more surface is available for As(V) adsorption, therefore the adsorbed As(V) concentration and thesoluble As(V) concentration decrease, resulting in less toxicity.

3.3. Impact of As(V) on the toxic effect of n-TiO2

As shown in Fig. 2, there was no mortality when the concentrationof n-TiO2 was less than 400 mg/L. However, when As(V) wasintroduced to the a culture medium, a synergistic effect occurred,even at low concentrations of nanoparticles. Figs. 5 and 6 show theresults at As(V) concentrations of 2.5 and 3.0 mg/L, respectively.

As Concentration (mg/L)0 1 2 3 4 5

Mo

rtal

ity

(%)

0

20

40

60

80

1001 ppm10ppm50ppm100ppm300ppmAs only

Fig. 4. Comparisons of toxic effect of As(V) at different n-TiO2 concentrations.Experiment conditions: suspension=culture medium and TiO2, volume=15 mL,pH=7.8, hardness=85 mg/L as CaCO3, T=20 °C.

0 50 100 150 200 250 300

Mo

rtal

ity

(%)

0

20

40

60

80

10024hr48hr72hr96hr

Total TiO2 Concentration (mg/L)

Fig. 5. Synergistic effect of TiO2 at the As(V) concentration of 2.5 mg/L. Experimentconditions: suspension=culture medium and TiO2, volume=15 mL, pH=7.8, hard-ness=85 mg/L as CaCO3, T=20 °C, [As(V)]=2.5 mg/L.

Residual Concentration (µg/L)0 1000 2000 3000 4000

Sp

ecif

ic A

dso

rpti

on

Am

ou

nt

(mg

/g)

0

2

4

6

8

10

12

Fig. 7. Sorption of As(V) onto n-TiO2 surface. Experiment conditions: suspension=culture medium and TiO2, volume=50 mL, pH=7.8, hardness=85 mg/L as CaCO3,T=20 °C, n-TiO2 concentration=1000 mg/L.

1354 D. Wang et al. / Science of the Total Environment 409 (2011) 1351–1356

Without n-TiO2, there was no mortality at the initial As(V)concentration of 2.5 mg/L (Fig. 1). This observation was alsoconfirmed for the control group in this experiment. However, withthe addition of different concentrations of n-TiO2, the mortalityincreased significantly until the concentration reached 50 mg/Ln-TiO2, thendecreased slowlywith the increase of n-TiO2 concentration.This toxic effectwasdifferent from theeffect observed in the presence ofparticulate matter of a larger size (Ma et al., 2002, 1999). Addition ofn-TiO2 reduced soluble As(V), but increased the total uptake of As(V)through the gastrointestinal tract. The accumulation of n-TiO2 in thegastrointestinal tract was observed in microscopy examinations (seeAppendix A.). However, at higher concentrations of n-TiO2, the residualAs(V) concentration was reduced significantly; therefore, the specificadsorption of As(V) on particle surface decreased, resulting in reducedAs(V) toxicity.

Fig. 6 shows that at an initial As(V) concentration of 3.0 mg/L,partial mortality was found even without the presence of n-TiO2. Thiswas consistent with the results shown in Fig. 1. The addition of n-TiO2

increased the mortality in a manner similar to that in Fig. 5. Due to ahigher initial As(V) concentration, a wider concentration range ofn-TiO2 was found for 100% mortality.

3.4. Sorption of As(V) on n-TiO2

The surface concentration of As(V) was suspected to play a veryimportant role in the synergistic toxic effect. Sorption experiments

0 50 100 150 200 250 300

Mo

rtal

ity

(%)

0

20

40

60

80

10024hr48hr72hr96hr

Total TiO2 Concentration (mg/L)

Fig. 6. Synergistic effect of n-TiO2 at the As(V) concentration of 3.0 mg/L. Experimentconditions: suspension=culture medium and TiO2, volume=15 mL, pH=7.8, hard-ness=85 mg/L as CaCO3, T=20 °C, [As(V)]=3.0 mg/L.

were carried out by mixing known amounts of nanoparticles with As(V) solutions at different concentrations. Fig. 7 shows the results. Itshould be noted that the sorption of As(V) onto the food was alsoexamined. Therewas no significant adsorption of As(V) by food due tothe low concentration of algae.

TiO2 nanoparticles have high affinity for As(V) species (Sun et al.,2009). At pH=7.8, the surface of n-TiO2 was positively charged, asevidencedbyameasuredpHzpc in the range 9.1–9.5 (see Supplementarydata). Therefore, protonated surface sites were available for As(V)adsorption. The Freundlich isotherm, q=KC1/n was used to fit thesorption data using SigmaPlot non-linear regression (version 11.0,2008); a correlation coefficient R2=0.99was obtained. The parametersfor the isothermwere K=1.05 and 1/n=0.27. Based on this result, thesorbed As(V) also contributed to the toxic response once n-TiO2

particles entered the gastrointestinal tract of C. dubia.

4. Discussion

The impact of engineered ENMs on human health and theenvironment is not well understood. Existing research concentrateson pure ENMs under ideal conditions. It is unlikely in a realisticenvironment that no other toxic species co-exist with ENMs. In thepresent research, when both As(V) and n-TiO2 were present in theculture medium, their interactions also impacted the mortality ofC. dubia. As shown in Fig. 7, the residual As(V) concentrationsignificantly decreased in the presence of n-TiO2; however, theoverall toxicity significantly increased. Therefore, the interactionbetween As(V) and n-TiO2 impacts the toxicity to C. dubia in twoways. First, the decrease in residual As(V) concentration reduces thetoxic effect. This is similar to the scenario in which both heavy metalsand other carriers (such as metal-complexing ligands, particles, andsediments) are present in a toxicity experiment (Stumm and O'Melia,1968). Second, the sorption of As(V) onto the n-TiO2 surfacecontributes to the toxicity once nanoparticles enter the body.

Accordingly, the mortality of C. dubia results from two uptakepathways. The first one is the uptake of As(V) species. The calculatedLC50 should be comparable to that of a system with As(V) only. Thesecond one is the uptake of n-TiO2, which also carries the sorbed As(V) to C. dubia through the gastrointestinal tract or body surface.However, as shown in Fig. 2, no mortality was found when the n-TiO2

concentration was less than 400 mg/L. For the following discussion,the contribution of bare n-TiO2 to mortality can be neglected.

Themortality of C. dubia largely resulted from the uptake of As(V) inboth soluble and sorbed forms. Therefore, the mortality was related notonly to the total concentration of As(V) but also to the concentration ofn-TiO2 in the suspension. At lower concentrations of As(V) (b0.4 mg/L)

1355D. Wang et al. / Science of the Total Environment 409 (2011) 1351–1356

(Fig. 4), the presence of n-TiO2 did not affectmortality. The uptake of As(V) from both residual and sorbed forms was less than a lethal dose. Athigher As(V) concentrations (N0.4 mg/L), mortality was dependent onthe concentrations of n-TiO2. When the concentration of n-TiO2 waslow, higher mortality was observed due to the higher residual andspecific surface concentrations of As(V). However, when the concen-tration of n-TiO2 was high (N100 mg/L), lower mortality was observeddue to the lower residual and specific surface concentrations of As(V).Therefore, for the same initial concentration of As(V), the mortalityincreased first, and then decreased with increasing n-TiO2 concentra-tion. The maximum mortality was observed at 50 mg/L of n-TiO2.

It should be noted that the concentration of suspended n-TiO2 isnot linearly related to the added amount of n-TiO2 due to aggregationand settling. Increased particle addition could enhance aggregation,resulting in a lower suspended particle concentration (Zhu et al.,2009). As a result, the overall toxicity could also be decreased byreduced uptake of TiO2 by C. dubia when n-TiO2 concentrations arehigh. Table 1 shows the calculated LC50 of As(V) with different n-TiO2

concentrations. The reduced LC50 values in the presence of n-TiO2

suggest a synergistic effect between As(V) and n-TiO2.There are two pathways for the uptake of nanoparticles by C. dubia.

The first pathway is the sorption of nanoparticles on the body surface.The degree of uptake is governed by the surface potentials of the bodyand the nanoparticles. However, this pathwaymaynot contribute to thetoxic effect because sorbednanoparticles canbe easily removed.C. dubiausually change their shell at regular intervals (see Appendix A.). Adiscarded shell with nanoparticles was observed, indicating that sorbedparticles can be removed during shell change. In addition, some n-TiO2

particles may penetrate through the shell and enter the body. In thiscase, the contribution of sorbed As(V) to mortality can be consideredinsignificant compared to other uptake pathways. The other pathway isthe oral uptake of nanoparticles as “faking food”. ENMs in thegastrointestinal tract of daphnia species have been observed andreported in the literature (Geller and Müller, 1981). In this pathway,the hydrodynamic size of the ENMs is critical to the uptake rate. To beeffectively transported to the gastrointestinal tract by daphnia species,foodmust fall in the appropriate size range. The acceptable particle sizeranges from hundreds of nanometers to several micrometers (Gellerand Müller, 1981) for filter-feeder such as C. dubia. As a result, onlyparticles of n-TiO2 with hydrodynamic radii within this range can beingested efficiently by C. dubia and contribute to the toxic effect.

5. Conclusion

The mortality of C. dubia in this research is mostly a result of theuptake of As(V) in both dissolved and sorbed forms. At a fixedconcentration of As(V), addition of n-TiO2 has a dual effect on thetotal uptake of As(V) by C. dubia. At lower concentrations of n-TiO2, thedissolved As(V) concentration is reduced slightly, while the sorbed As(V) concentration on the n-TiO2 surface is high. The mortality increaseswith increasing n-TiO2 concentrations in this concentration range.When the concentration of n-TiO2 reaches a certain value, both thedissolved As(V) concentration and the sorbed As(V) are significantlyreduced. In addition, a higher n-TiO2 concentration induces better

Table 1LC50 of As(V) in the presence of n-TiO2 at different concentrations.

CTiO2

(mg/L)LC50 total(mg/L)

LC50 residual(mg/L)

EFt⁎

0 3.68 3.68 11 3.01 2.92 1.2210 3.18 2.78 1.1650 1.43 0.89 2.57100 2.05 1.30 1.79300 2.77 – 1.32

⁎ Enhancement factor (EF) calculated from the ratio of LC50 without and with n-TiO2.

aggregation, which reduces the availability of ENMs for uptake byC. dubia. As a result, mortality decreases with further increasing n-TiO2

concentrations. However, when the concentration of n-TiO2 is fixed, themortality is only dependent on the As(V) concentration. The higher theAs(V) concentration, the greater the mortality.

Acknowledgements

This researchwas partially supported through grants from LeonardWood Institute, Fort LeonardWood,MO, USA (contract # 281173) andState Key Laboratory of Pollution Control and Resource Reuse, TongjiUniversity, China. Ji Hu was supported by the China ScholarshipCouncil through a state-sponsored scholarship program.

Appendix A. Supplementary data

Supplementary data to this article can be found online atdoi:10.1016/j.scitotenv.2010.12.024.

References

Anthony S, Tran L, Aitken R, Donaldson K. Nanoparticles, human health hazard andregulation. J R Soc 2010:s119–29.

Baun A, Hartmann NB, Grieger K, Kusk KO. Ecotoxicity of engineered nanoparticles toaquatic invertebrates: a brief review and recommendations for future toxicitytesting. Ecotoxicology 2008;17:387–95.

Benn TM, Westerhoff P. Nanoparticle silver released into water from commerciallyavailable sock fabrics. Environ Sci Technol 2008;42:4133–9.

Chatterjee R. The challenge of regulating nanomaterials. Environ Sci Technol 2008;42:339–43.

Choi JY, Ramachandran G, Kandlikar M. The impact of toxicity testing costs onnanomaterial regulation. Environ Sci Technol 2009;43:3030–4.

Clements WH, Kiffney PM. Integrated laboratory and field approach for assessingimpacts of heavy metals at the Arkansas River, Colorado. Environ Toxicol Chem1994;13:397–404.

Cooper NL, Bidwell JR, Kumar A. Toxicity of copper, lead, and zinc mixtures toCeriodaphnia dubia and Daphnia carinata. Ecotoxicol Environ Saf 2009;72:1523–8.

Gao J, Youn S, Hovsepyan A, Llaneza VL, Wang Y, Bitton G, et al. Dispersion and toxicityof selected manufactured nanomaterials in Natural River water samples: effects ofwater chemical composition. Environ Sci Technol 2009;43:3322–8.

Geller W, Müller H. The filtration apparatus of Cladocera: filter mesh-sizes and theirimplications on food selectivity. Oecologia 1981;49:316–21.

Geranio L, Heuberger M, Nowack B. The behavior of silver nanotextiles during washing.Environ Sci Technol 2009;43:8113–8.

Griffitt RJ, Luo J, Gao J, Bonzongo JC, Barber DS. Effects of particle composition andspecies on toxicity of metallic nanomaterials in aquatic organisms. Environ ToxicolChem 2008;27:1972–8.

Griffitt RJ, Hyndman K, Denslow ND, Barber DS. Comparison of molecular andhistological changes in zebrafish gills exposed to metallic nanoparticles. ToxicolSci 2009;107:404–15.

Guan XH, Su T, Wang J. Quantifying effects of pH and surface loading on arsenicadsorption on NanoActive alumina using a speciation-based model. J Hazard Mater2009;166:39–45.

Gupta K, Ghosh UC. Arsenic removal using hydrous nanostructure iron(III)–titanium(IV) binary mixed oxide from aqueous solution. J Hazard Mater 2009;161:884–92.

Hall S, Bradley T, Moore JT, Kuykindall T, Minella L. Acute and chronic toxicity of nano-scale TiO2 particles to freshwater fish, cladocerans, and green algae, and effects oforganic and inorganic substrate on TiO2 toxicity. Nanotoxicology 2009;3:91–7.

He W, Megharaj M, Naidu R. Toxicity of tri- and penta-valent arsenic, alone and incombination, to the cladoceran Daphnia carinata: the influence of microbialtransformation in natural waters. Environ Geochem Health 2009;31:133–41.

Hristovski K, Baumgardner A, Westerhoff P. Selecting metal oxide nanomaterials forarsenic removal in fixed bed columns: from nanopowders to aggregatednanoparticle media. J Hazard Mater 2007;147:265–74.

Hristovski KD, Westerhoff PK, Crittenden JC, Olson LW. Arsenate removal bynanostructured ZrO2 spheres. Environ Sci Technol 2008;42:3786–90.

Hund-Rinke K, SimonM. Ecotoxic effect of photocatalytic active nanoparticles (TiO2) onalgae and daphnids. Environ Sci Pollut Res 2006;13:225–32.

Kim KT, Klaine SJ, Lin S, Ke PC, Kim SD. Acute toxicity of a mixture of copper and single-walled carbon nanotubes to Daphnia magna. Environ Toxicol Chem 2010;29:122–6.

Klaine SJ, Alvarez PJJ, Batley GE, Fernandes TF, Handy RD, Lyon DY, et al. Nanomaterialsin the environment: behavior, fate, bioavailability, and effects. Environ ToxicolChem 2008;27:1825–51.

L. Spehar R, T. Fiandt J. Acute and chronic effects of water quality criteria-based metalmixtures on three aquatic species. Environ Toxicol Chem 1986;5:917–31.

Lovern SB, Klaper R. Daphnia magna mortality when exposed to titanium dioxide andfullerene (C60) nanoparticles. Environ Toxicol Chem 2006;25:1132–7.

Lovern SB, Strickler JR, Klaper R. Behavioral and physiological changes in Daphniamagnawhen exposed to nanoparticle suspensions (titanium dioxide, nano-C60, andC60HxC70Hx). Environ Sci Technol 2007;41:4465–70.

1356 D. Wang et al. / Science of the Total Environment 409 (2011) 1351–1356

Ma H, Kim SD, Cha DK, Allen HE. Effect of kinetics of complexation by humic acid ontoxicity of copper to Ceriodaphnia dubia. Environ Toxicol Chem 1999;18:828–37.

Ma H, Kim SD, Allen HE, Cha DK. Effect of copper binding by suspended particulatematter on toxicity. Environ Toxicol Chem 2002;21:710–4.

Mahar AM,WatzinMC. Effects of metal and organophosphate mixtures on Ceriodaphniadubia survival and reproduction. Environ Toxicol Chem 2005;24:1579–86.

Moore MN. Do nanoparticles present ecotoxicological risks for the health of the aquaticenvironment? Environ Int 2006;32:967–76.

Nanowerk. Debunking the trillion dollar nanotechnology market size hype. http://www.nanowerk.com/spotlight/spotid=1792.php2007.

Nel A, Xia T, Mädler L, Li N. Toxic potential of materials at the nanolevel. Science2006;311:622–7.

Oberdörster G, Oberdörster E, Oberdörster J. Nanotoxicology: an emerging disciplineevolving from studies of ultrafine particles. Environ Health Perspect 2005;113:823–39.

Shaw JR, Glaholt SP, Greenberg NS, Sierra-Alvarez R, Folt CL. Acute toxicity of arsenic toDaphnia pulex: influence of organic functional groups and oxidation state. EnvironToxicol Chem 2007;26:1532–7.

Stumm W, O'Melia CR. Stoichiometry of coagulation. J Am Water Works Assn 1968;60:514–39.

Sun H, Zhang X, Zhang Z, Chen Y, Crittenden JC. Influence of titanium dioxidenanoparticles on speciation and bioavailability of arsenite. Environ Pollut2009;157:1165–70.

USEPA. Methods for measuring the acute toxicity of effluents and receiving waters tofreshwater and marine organisms. http://www.epa.gov/waterscience/methods/wet/disk2/2002.

USEPA. Trimmed Spearman–Karber method. http://www.epa.gov/EERD/stat2.htm.Warheit DB, Hoke RA, Finlay C, Donner EM, Reed KL, Sayes CM. Development of a base

set of toxicity tests using ultrafine TiO2 particles as a component of nanoparticlerisk management. Toxicol Lett 2007;171:99-110.

Wiesner MR, Lowry GV, Jones KL, Hochella Jr MF, Digiulio RT, Casman E, et al.Decreasing uncertainties in assessing environmental exposure, risk, and ecologicalimplications of nanomaterials. Environ Sci Technol 2009;43:6458–62.

Zhu X, Zhu L, Chen Y, Tian S. Acute toxicities of six manufactured nanomaterialsuspensions to Daphnia magna. J Nanopart Res 2009;11:67–75.