Desalination 248 (2009) 771–782 Acute toxicity of boron, titanium dioxide, and aluminum nanoparticles to Daphnia magna and Vibrio fischeri Nikolay Strigula*, Liana Vaccaria, Catherine Galduna, Mahmoud Waznea, Xuyang Liua, Christos Christodoulatosa, Kristin Jasinkiewiczb a Stevens Institute of Technology, Center for Environmental Systems, Castle Point on Hudson, Hoboken, NJ 07030, USA b US Army, RDECOM-ARDEC, Environmental Technology Division, Picatinny, NJ 07806, USA Tel: 201-9524260; Fax: 201-2168321; email: nstrigul@stevens.edu Received 19 November 2008; accepted 15 January 2009 Abstract The acute toxicity of four different nanosized particulate materials (titanium dioxide, boron nanoparticles, and two types of aluminum nanoparticles (ALEX and L-ALEX)) were evaluated using two tests: the Microtox toxicity test and the acute toxicity test with Daphnia magna. The results were analyzed in order to calculate LD50 at 24 and 48 h. It was found that titanium dioxide nanoparticles show a low level of toxicity, and LD50 values cannot be calculated. Conversely, boron nanoparticles with EC50 ranging from 56 to 66 mg L 1, depending upon the age of the solution, can be classified as ‘‘harmful’’ to aquatic microorganisms (EC50 in the range 10–100 mg L 1). We have also discussed possible mechanisms of nanoparticle toxicity and potential problems in ecotoxicological testing of nanomaterials. The studied nanomaterials can be ranked in the following order according to their Daphnia acute toxicity: boron nanoparticles>ALEX>L-ALEX> TiO2. Keywords: nanomaterials; acute toxicity; boron nanoparticles; aluminum nanoparticles; titanium dioxide nanoparticles; Daphnia magna test; Vibrio fischeri test 1. Introduction Numerous nanomaterials have recently been developed, and numerous practical applications have been found in water treatment, medicine, cosmetics, and engineering [1,2]. Ecotoxicolog*Corresponding author. Presented at the Conference on Protection and Restoration of the Environment IX, Kefalonia Greece, June 30–July 3, 2008 ical properties of nanomaterials are largely unknown [2–4]; however, this problem has recently received substantial attention [5]. Several reports published over the last couple of years, including the first review papers [6–10], address ecotoxicological properties of different nanomaterials. This paper presents results concerning ecotoxicity of boron, aluminum, and titanium dioxide nanoparticles. These materials have already 0011-9164/09/$– See front matter © 2009 Published by Elsevier B.V. doi:10.1016/j.desal.2009.01.013 772 N. Strigul et al. / Desalination 248 (2009) 771–782 given rise to numerous applications and are being manufactured at the industrial level. For instance, boron nanoparticles are being considered as a potential fuel source that releases energy after metal oxidation, and have been employed in medical research [11]. Aluminum nanoparticles are already widely employed in explosive combinations [12]. Titanium dioxide nanoparticles belong to the currently most used group of nanomaterials known as photocatalysts and adsorbents [13]. TiO2 particles are being used in sunscreen products, and as a catalyst in sterilization and chemical engineering. Also, it has an application as an adsorbent in water cleaning [14,15]. Some data concerning toxicity of the nanoparticles similar to those tested in this study recently became available. Several reports available on the ecotoxicity of TiO2 nanoparticles show that this material has relatively low toxicity [10,16–19]. In particular, Heinlaan et al. [20] compared toxicity of ZnO, CuO, and TiO2 nanoparticles to Vibrio fischeri, Daphnia magna, and Thamnocephalus platyurus and reported that TiO2 particles did not show any toxicity in concentrations up to 20 mg L 1. Similarly Griffitt et al. [19] did not find any toxic effects from TiO2 nanoparticles in standard ecotoxicological tests with zebrafish, Daphnia, and algae. Zhu et al. [21] showed that TiO2 and Al2O3 nanoparticles did not affect zebrafish at early developmental stages. Wagner et al. [22] reported that aluminum nanoparticles have greater toxicity to macrophages than aluminum oxide nanoparticles. In particular, aluminum particles administered at 25 mg mL 1 level for 24 h inhibited cell phagocytotic ability. Doshi et al. [23] have shown that ALEX and L-ALEX do not have significant toxicity in the soil environment in a realistic range of concentrations. At this point, we are not aware of any published ecotoxicological results concerning the boron nanoparticles. However, Petersen et al. [11] reported that boron nanoparticles inhibited tumor growth in an animal model. The first objective of this work is to evaluate acute toxicity of four different nanoparticles (boron, titanium dioxide, and two different aluminum [ALEX and L-ALEX] nanoparticles) using two standard ecotoxicological tests, the D. magna acute toxicity test and the Microtox toxicity test. The tests with D. magna have been performed according to the standard protocol Daphnia sp., acute immobilization test (OECD test No. 201). The Microtox test employs the bioluminescent bacterium (V. fischeri) as the test organism. The preliminary toxicity estimates obtained with these tests often correlate with other tests on fish, cell cultures, and rodents and these tests are considered to be convenient tools for evaluation of nanomaterials toxicity [6,10,24]. The second objective of this study is to evaluate possible methodological issues caused by the ‘‘nano’’ nature of the tested materials. Nanomaterials often have unique chemical and physical properties that generate a substantial interest in nanotechnology; however, from another side, nanomaterials potentially may have biological and environmental effects that have not been observed in other materials [5,7,8,10,16]. Also, nanomaterials may cause new types of biases in applications of the standardized ecotoxicological tests. These uncertainties present a challenge for ecotoxicological research, and some modifications of the standard ecotoxicological tests may be needed in respect to nanomaterials. Some particular issues that this study revealed concern the stability of nanoparticles in water suspensions and light-adsorbing properties of nanoparticles. 2. Materials and methods 2.1. Nanoparticles: characterization and preparation The nanoparticle suspensions were prepared by a procedure consisting of the preparation of suspension of the stock powder in D.I. water (for the Microtox tests) or the standard N. Strigul et al. / Desalination 248 (2009) 771–782 freshwater (for the Daphnia tests) followed by ultrasonication for 30 min to break up aggregates, which were commonly formed during storage. In the long-term experiment, the nanoparticle suspensions were kept at room temperature on a convenient laboratory rotator, which mixed the samples at a constant rate of about 100 rpm. The analysis of the particle size distribution showed that this procedure is sufficient to obtain homogeneous nanoparticle suspension (see [25] for the details). The 100-nm nanoaluminum particles, ALEX and L-ALEX (as a powder of 1–10 mm agglomerates), were obtained from Argonide Corp. ALEX and L-ALEX were produced by the electroexploded wire technique. The obtained particles are immediately passivated with air to produce a coating of aluminum oxide (ALEX) or with a carboxylic acid to produce surface carboxylate ligand chains covering the metal surface. Nanocrystalline TiO2 was provided by Dr. Meng and prepared by hydrolysis of titanium sulfate solution according to Meng et al. [14]. The particles are about 6 nm in diameter and were delivered in agglomerates of 0.5–2.0 mm [15,26]. Boron nanoparticles of average nominal size 10–20 nm were obtained from Alfa Aesar (Ward Hill, MA) and kept in a pure argon atmosphere. Based on scanning electron microscopy–energy dispersive spectrometer (SEM-EDS) analysis, boron content was measured at 70%, which is less than that reported by the manufacturer (95%). The same value was obtained in acid digestion analysis. This is probably due to the oxidation of boron during production or transportation, based on the measured value of oxygen at 28% using SEM-EDS analysis [25]. 2.2. Acute Daphnia sp. immobilization test (OECD 202) The Daphnia sp., acute immobilization test (OECD test No. 202) was used to give an estimate of the toxic effects of the studied materials 773 on freshwater invertebrates. This test was performed using Daphtoxkit bioassays produced by MicroBioTests Inc. according to the OECD 202 protocol. The Daphtoxkit bioassay included the standard freshwater similar to the one prescribed by the OECD test, Daphnia eggs, and plates consisting of 24 experimental wells of 10 mL. Six distinct concentrations including control (250, 80, 25, 8, 2.5, and 0 mg L 1, respectively) of the focal nanomaterial were used in each test. Experimental design included 6 groups with 20 animals (4  5) exposed to each of the tested concentrations. Animals from each group were separated into 4 smaller groups of 5 animals to provide 4 independent replications located in different experimental wells. The animals tested were young daphnids, each less than 24 h old at the beginning of the test, in different concentrations of the tested nanomaterials. Young D. magna were hatched from eggs, which were incubated for 72 h in well-aerated standard freshwater. Two hours prior to exposing the Daphnia to the various nanoparticles, they were fed using powdered Spirulina and not fed again for the duration of the test. The experimental wells were filled with wellaerated suspension of the focal nanomaterial. Then the groups of 5 daphnids each were added to these wells of effluent and remained there for a period of 48 h in darkness. After each 24 and 48 h had elapsed, the experimental cells were examined to determine the number of dead and immobilized neonates. These numbers were recorded, and immobilization of the daphnids in the test substances was compared with control values from the same tray. 2.3. Microtox acute toxicity test (ASTM Standard D-5660) The Microtox test is a metabolic inhibition test with luminescent marine bacteria (V. fischeri 774 N. Strigul et al. / Desalination 248 (2009) 771–782 NRRL B-11177). This test was performed according to the ASTM Standard (D-5660). Lyophilized bacteria were rehydrated with the included reconstitution solution (ultrapure water) just prior to performing the test. Luminescent bacteria were incubated under constant temperature in glass vials with 1 mL of 2% NaCl solution and the toxicant presented in various concentrations. Nine concentrations of each nanomaterial obtained by a sequence of 1:1 dilutions ranging from 0.4395 to 112.5 mg L 1 were tested. At 5- and 10-min time intervals, the luminescence of each individual vial was measured. The sample toxicity was evaluated based on the decrease of the intensity of light produced by the luminescent bacteria; the statistical analysis including the light absorbance correction was done according to the test description. using dynamic light scattering (DLS). A monochromatic coherent He–Ne laser with a fixed wavelength of 633 nm was used as a light source. The intensity of scattered light was measured by a detector at 1738 and an autocorrelation function was accumulated for 10 s. For each experiment, 1 mL of the suspension was introduced into a polystyrene disposable cuvette (Sarstedt, Germany). The hydrodynamic radius (intensity based) was calculated using the Stokes–Einstein equation. 2.6. Statistical analysis LD50 values and their confidence intervals were calculated by the probit analysis using the EPA Probit Analysis Program, version 1.5 (U.S. Environmental Protection Agency). Other statistical calculations have been performed using Statistica 7.0 software (StatSoft Inc.). 2.4. Water characteristics Water pH was measured using Accumet AR25 pH meter with Accumet 13-620-285 pH electrode from Fisher Scientific Inc. Light absorbance of nanomaterials in water was measured with the Hewlett-Packard 8452A spectrophotometer. 2.5. Dynamic light scattering The particle size distribution was obtained by Nano Zetasizer (Malvern, Worcestershire, UK) 3. Results and discussion 3.1. Acute Daphnia sp. immobilization test (OECD 202) Titanium dioxide nanoparticles showed a low level of toxicity, and LD50 values at 24 and 48 h could not be calculated (Table 1). However, animals exposed to 80 and 250 mg L 1 of TiO2 nanoparticles were significantly slower after 24 h than the animals exposed to the lower concentrations, and it was evident that TiO2 particles Table 1 The results of the Daphnia magna test with TiO2 nanoparticles Concentration Control Exposure time 24 h 48 h 24 h A B C D Mortality % 5/0 5/0 5/0 5/0 0 5/0 5/0 3/2 4/1 15 5/0 5/0 5/0 5/0 0 8 mg L 1 25 mg L 48 h 24 h 48 h 24 h 48 h 24 h 48 h 24 h 48 h 4/1 3/2 3/2 3/2 35 4/1 4/1 4/1 5/0 15 1/4 3/2 2/3 2/3 50 5/0 5/0 4/1 4/1 10 5/0 5/0 4/1 3/2 15 5/0 5/0 5/0 5/0 0 4/1 3/2 5/0 4/1 20 5/0 3/2 4/1 4/1 20 3/2 2/3 4/1 4/1 35 2.5 mg L 1 1 80 mg L 1 250 mg L 1 A,B,C,D are four different experimental groups; k/m indicates number of alive (k) and dead (m) animals in each group. 775 N. Strigul et al. / Desalination 248 (2009) 771–782 Table 2 The results of the Daphnia magna test with boron nanoparticles Concentration Control Exposure time 24 h 48 h 24 h A B C D Mortality % 5/0 5/0 5/0 5/0 0 3/2 4/1 3/2 5/0 25 5/0 5/0 5/0 5/0 0 8 mg L 1 25 mg L 48 h 24 h 48 h 24 h 48 h 24 h 48 h 24 h 48 h 5/0 4/1 5/0 4/1 10 4/1 4/1 5/0 4/1 15 0/5 2/3 3/2 1/4 70 2/3 2/3 3/2 2/3 55 0/5 0/5 0/5 1/4 95 0/5 0/5 0/5 0/5 100 0/5 0/5 0/5 0/5 100 0/5 0/5 0/5 0/5 100 0/5 0/5 0/5 0/5 100 2.5 mg L 1 1 80 mg L 1 250 mg L 1 A,B,C,D are four different experimental groups; k/m indicates number of alive (k) and dead (m) animals in each group. accumulated in significant amounts in the digestive tract. Boron nanoparticles (Table 2), in concentrations higher than 80 mg L 1, killed all daphnids in 24 h. The LD50 value for 24 h is 19.5 mg L 1 with 14.2–27.0 mg L 1 95% confidence interval, and for 48 h it is 6.7 mg L 1 with 3.661–9.696 mg L 1 95% confidence interval. The probit analysis algorithm converged for the 24 h data. However to provide the algorithm convergence for 48 h data, we had to ignore the mortality in the control group (as the probit transformation can be done only if the mortality changes monotonically as the concentration of the pollutant increases). After 24 h, animals exposed to 2.5 mg L 1 of boron nanoparticles were actively swimming similarly to the daphnids in the control group, while animals in 8 mg L 1 were less active and had some amounts of boron particles in the digestive system. Daphnids in 25 mg L 1 were very slow and had relatively large amounts of particles inside. For the nanoaluminum particles that differ in surface coatings (ALEX particles coated with a thin layer of aluminum oxide are not hydrophobic while L-ALEX particles coated with carboxylate ligand are hydrophobic), the following preliminary LD50 values were found: ALEX (see Table 3) – the LD50 for 24 h is 219.6 mg L 1 with 118.0–774.4 mg L 1 95% confidence interval, and for 48 h it is 7.483 (the confidence interval cannot be calculated as the mortality does not change monotonically); L-ALEX – the LD50 value for 48 h is 107.588 mg L 1 with a very large 95% confidence interval 15.549–2468.308 mg L 1 (Table 4) and the Table 3 The results of the Daphnia magna test with ALEX nanoparticles Concentration Control Exposure time 24 h 48 h 24 h A B C D Mortality % 5/0 5/0 5/0 5/0 0 5/0 5/0 4/1 5/0 5 5/0 5/0 5/0 5/0 0 8 mg L 1 25 mg L 48 h 24 h 48 h 24 h 48 h 24 h 48 h 24 h 48 h 4/1 5/0 3/2 3/2 25 5/0 4/1 5/0 5/0 5 1/4 2/3 1/4 1/4 75 5/0 5/0 4/1 5/0 5 2/3 2/3 1/4 1/4 70 5/0 3/2 4/1 1/4 35 1/4 2/3 3/2 1/4 65 3/2 2/3 2/3 3/2 50 0/5 0/5 0/5 1/4 95 2.5 mg L 1 1 80 mg L 1 250 mg L 1 A,B,C,D are four different experimental groups; k/m indicates number of alive (k) and dead (m) animals in each group. 776 N. Strigul et al. / Desalination 248 (2009) 771–782 Table 4 The results of the Daphnia magna test with L-ALEX nanoparticles Concentration Control Exposure time 24 h 48 h 24 h A B C D Mortality % 5/0 5/0 5/0 5/0 0 5/0 5/0 3/2 4/1 15 4/1 4/1 4/1 3/2 25 8 mg L 1 25 mg L 48 h 24 h 48 h 24 h 48 h 24 h 48 h 24 h 48 h 4/1 4/1 4/1 3/2 25 5/0 4/1 4/1 5/0 10 5/0 4/1 3/2 3/2 25 3/2 4/1 4/1 3/2 20 2/3 2/3 3/2 2/3 55 4/1 5/0 2/3 3/2 30 4/1 3/2 1/4 2/3 50 3/2 3/2 3/2 4/1 45 1/4 1/4 2/3 3/2 65 2.5 mg L 1 1 80 mg L 1 250 mg L 1 A,B,C,D are four different experimental groups; k/m indicates number of alive (k) and dead (m) animals in each group. effects were not sufficient to compute LD50 value for the 24-h exposure (mortality did not exceed 50%). While L-ALEX particles demonstrate moderate toxicity in suspension, these particles are hydrophobic and the real concentration in suspension is much lower than declared since numerous particles stayed on the water surface. Also the actual concentration of particles in water may not be proportional to input concentration of nanoparticles. The homogeneous suspensions of L-ALEX could not be prepared even after long-term ultrasound treatment and mixing. The lack of homogeneous character of the suspensions explains the observed pattern in which mortality levels stabilized at values around 30% and 55% after 24 and 48 h, respectively and a dose–response relationship was not observed at high L-ALEX nominal particle concentrations (Table 4). Boron nanoparticles were most toxic of all the tested nanomaterials (Figs 1 and 2). Also, in contrast with the other nanomaterials, boron nanoparticles were probably more chemically active as boron nanoparticles slightly increased water pH up to 8 with saturation effect (Fig. 3) while TiO2, ALEX, and L-ALEX nanoparticles in all concentrations did not significantly change the pH of the standard freshwater (pH was about 92 92 72 Survival at 48h, % Survival at 24h, % 72 52 32 Boron L-Alex TiO2 Alex 12 52 32 Boron L-Alex TiO2 Alex 12 −8 −8 1 10 Log of concentration, mg L–1 100 Fig. 1. Survival of young Daphnia after exposure to nanomaterials in different concentrations at 24 h. 1 10 100 Log of concentration, mg L–1 Fig. 2. Survival of young Daphnia after exposure to nanomaterials in different concentrations at 48 h. N. Strigul et al. / Desalination 248 (2009) 771–782 7.5–7.7) during the experiment. However, this change of pH was not so large to explain mortality of animals exposed to the boron nanoparticles. Hund-Rinke and Simon [16] modified the Daphnia test to address the role of photocatalytic reactions in nanoparticle toxicity. In contrast, the reported experiments were conducted in darkness that excluded possible contribution of nanoparticle photoreactions to their toxicity. No water mixing was performed during the 48-h acute tests, and TiO2, ALEX, and boron nanoparticles precipitated in significant amount from suspension during the first 24 h, while hydrophobic L-ALEX floated to the surface. Apparently, the studied nanomaterials not only precipitated or floated to the surface, but also agglomerated to some extent [25]. Therefore the physical conditions in water during the first 24 h of the experiment were different compared to the conditions in the second 24 h as a substantial amount of nanoparticles initially being in suspension precipitated or floated. Nanoparticles were observed in the digestive track of young Daphnia. Animals may actively eat nanoparticles or their agglomerates, mistaking them for food. It is possible that the particles agglomerate further in the digestive system, eventually blocking it and causing death. However, our results show that this possible indirect toxicity mechanism does not contribute significantly to the 48 h acute toxicity. For instance, after 48 h, some actively moving animals exposed to high concentrations of TiO2 had a significant amount of TiO2 in their bodies. However, these observations were made using shortterm acute toxicity tests, and they probably do not indicate adaptation of Daphnia to nanoparticles. We anticipate that the possible toxicity mechanism described above might have some importance in case of a long-term exposure of Daphnia to nanoparticles. We can conclude that the acute toxicity of nanoparticles on Daphnia is determined by some specific chemical properties of each material, rather than 777 a physical–chemical mechanism common to different nanomaterials. 3.2. Microtox acute toxicity test (ASTM Standard D-5660) The results of the acute toxicity test are consistent with our findings using the Daphnia test. Only the boron nanoparticles exhibited significant toxicity (V. fischeri NRRL B-11177) in the range of concentrations 0–120 mg L 1 among the tested nanomaterials. ALEX, LALEX, and TiO2 nanoparticles did not inhibit luminescent activity of V. fischeri (less than 5% light reduction after 15 min), but these materials adsorbed light proportionally to the concentration (in the range 20–30% at the highest concentration, 120 mg L 1). Our results concerning the toxicity of TiO2 nanoparticles to Daphnia and V. fischeri confirm recently published data [19,20] that TiO2 does not have acute toxicity to these organisms in a realistic range of concentrations. With respect to the Microtox test, several issues were identified for the boron nanoparticles, as well as other nanomaterials, which need to be addressed to eliminate the possible biases: (1) Nanoparticles may adsorb and scatter light. A light absorbance correction is then a necessary experimental step. (2) Nanoparticles of different sizes can have different effects on microorganisms and different mobility in aquatic solution. The dynamics of size particle distribution is an important parameter, which was monitored with a zetaanalyzer. (3) Toxicity observed after nanoparticles amendment can be a result of the interference of the toxicity of nanoparticles of certain size fractions and the toxicity of soluble compounds, which are products of the nanoparticles’ dissolution. Therefore, it is important to be able to evaluate the dissolution kinetics and separate toxicity of the nanoparticles and their dissolution products. While the first two problems, i.e., light absorbance correction and dynamics of nanoparticle 778 N. Strigul et al. / Desalination 248 (2009) 771–782 distribution in solution, can be evaluated using standard methods, the third problem (i.e., interference of possible toxic effects of the nanoparticles themselves and the products of their dissolution) is nontrivial. To address this problem, filtration and filtration applied together with centrifugation were explored; however, both these methods were not able to completely remove particles from aqueous samples. After filtration and centrifugation, an aqueous sample with a homogenous distribution of small-sized nanoparticle was obtained. Therefore, to this point, we have not been able to measure the dissolution kinetics and decouple the toxicity effects of boron particles and soluble boron compounds. 3.2.1. Boron nanoparticles in solution Boron nanoparticle suspensions were characterized immediately after preparation and after mixing for set intervals of time to study their aggregation behavior. When a particle suspension with a concentration of 250 mg L 1 was freshly prepared, the hydrodynamic diameter was measured at 320 nm and the measured zeta- potential was 32.5 mV. After mixing in an end-over-end rotator for 1 day, the hydrodynamic diameter increased to 435 nm most likely due to aggregation (Fig. 4). The dispersion was then filtered through 0.2-mm size membrane, and the measured hydrodynamic diameter decreased to 101 nm. The hydrodynamic diameter further decreased to 60 nm after the filtered sample was centrifuged at 12,500 rpm for 30 min. A duplicate sample of the original suspension was analyzed for hydrodynamic diameter measurement after the sample was mixed for 1 month. The hydrodynamic diameter was measured at 447 nm, slightly greater than that measured after a mixing period of one day, and the zeta-potential was measured at 33.4 mV. Similarly, the hydrodynamic diameter of the suspension decreased to 78 nm after filtration and it further decreased to 64 nm after centrifugation. The experimental data indicated that the hydrodynamic diameter of the boron suspension increased significantly after mixing for 1 day; however, only, a slight change was observed between a mixing period of one day and one month (Fig. 4). Furthermore, the hydrodynamic diameter of the suspension decreased significantly upon filtration and it further decreased upon centrifugation. Boron nanoparticles in water absorb light for parameters of the Microtox detector (490 nm, 10 mm path width). The detector design indicates 8.1 8 pH 7.9 7.8 7.7 7.6 7.5 −10 40 90 140 190 240 Concentration of boron nanoparticles, mg L–1 Fig. 3. Change of pH in the suspension of boron nanoparticles in the standard freshwater after 48 h. 779 N. Strigul et al. / Desalination 248 (2009) 771–782 600 Hydrodynamic size (nm) 500 400 300 200 100 0 0 day 1 day 1 month suspension suspension suspension 1 month filtered 1 day fltered 1 month 1 day filtered & filtered & centrifuged centrifuged Fig. 4. Hydrodynamic size (nm) of the boron nanoparticles in suspension depends on the time and filtration method. for the fast estimation of the light adsorbance when the tested substances are not very toxic. 3.2.1. Toxicity of boron nanoparticles to Vibrio fischeri The aqueous mixture of the particles shows similar toxicity properties (Fig. 5) immediately after sample preparation (EC50–55.85 mg L 1) on the next day (EC50–56.62 mg L 1) and after 60 days (EC50–65.98 mg L 1). % effect after 15 minutes of exposure that it is light absorbance rather than scattering, which may also contribute to the light losses. The standard method (according to the test manual) is to estimate light absorbance of the samples by using the spectrophotometer. It included the preparation of the similar series of dilutions of the same nanomaterials and their measurements. For the boron nanoparticles, this method demonstrated linear dependence of the light adsorbance on the particle concentration (y = 0.0036x + 0.0036, r2 = 0.9998) in the concentration range 0–125 mg of particles per liter. We have found that this method may be not very convenient in some cases, where expensive nanomaterials are tested. We have tested an alternative method to check the light adsorbance of the samples. In the Microtox test we have added one more measurement at 1 min, which was practically immediate after the mixing of the samples in the test vials. Considering the reduction of light at 1 min as results of light absorbance of nanoparticles we have obtained similar results for boron nanoparticles (y = 0.0037x + 0.0041, r2 = 0.9909) as well as with other tested nanomaterials. We suggest that this method may be suitable 120 100 80 60 40 20 0 0 day, 1 day, 60 days. −20 0 20 40 60 80 100 120 Nanoparticles concentration mg L–1 Fig. 5. Dose–response curves for boron nanoparticles in the 60-day experiment. % effect after 15 minutes of exposure 780 N. Strigul et al. / Desalination 248 (2009) 771–782 100 80 60 40 20 0 nonfiltered, filtered, filtered and centrifuged, −20 0 20 40 60 80 100 120 Nanoparticles concentration in the nonfiltered sample, mg L–1 Fig. 6. Dose–response curves of boron nanoparticles in water suspension for nonfiltered, filtered, and filtered and centrifuged samples after 24 h. The original mixture of particles inhibits fluorescent bacteria more significantly than filtered samples, or filtered and centrifuged samples (Figs 6 and 7). The initial toxicity of the nanoparticle suspension (1 day after suspension preparation) is associated mostly with the nanoparticles’ toxicity, and the contribution of the dissolution product is not significant (Fig. 6). However, after 60 days, the observed toxicity could be associated mostly with the toxicity of the dissolution products (Fig. 7). The boron nanoparticles with EC50 ranging from 56 to 66 mg L 1, depending upon aging % effect after 15 minutes of exposure 90 80 70 60 50 40 30 20 10 0 −10 nonfiltered, filtered and centrifuged, −20 0 20 40 60 80 100 120 Nanoparticles concentration in the nonfiltered sample, mg L–1 Fig. 7. Dose–response curves of boron nanoparticles suspension and filtered and centrifuged samples after 60 days. time/age of solution, can be classified as ‘‘harmful’’ to aquatic microorganisms (EC50 in the range 10–100 mg L 1) according to the Commission Directive 93/67/EEC from the European Union for the assessment of risk to man and the environment of substances. The toxicity of a water suspension of boron nanoparticles could be explained by: (1) toxicity of the nanoparticles themselves, and (2) toxicity of the products of the particles’ dissolution. The first factor dominates at the earliest time of boron nanoparticles release in water, and the second factor might dominate in aged suspensions. Further research is necessary to investigate (1) the ecotoxicity of boron particles using standard OECD tests (i.e. fish, invertebrates, plants, and soil tests), (2) mechanisms of boron particles toxicity, (3) the solubility of boron nanoparticles, and (4) the toxicity of the dissolution products of boron nanoparticles. 4. Conclusions The results suggest that toxic responses related more to chemical characteristics and the dissolved phase concentration of the different nanoparticles. Several issues specific to nanoparticles, which need to be addressed to eliminate the possible biases in toxicity tests, were identified: (1) Nanoparticles adsorb light. Light absorbance correction is then a necessary experimental step for certain toxicity tests. (2) Toxicity of nanoparticles may be a result of the effects from nanoparticles itself, dissolution products, and nanoparticle agglomerates that develop during the experiment. The studied nanomaterials can be ranked in the following order according to their Daphnia acute toxicity: boron nanoparticles >ALEX > L-ALEX> TiO2. Acknowledgments This research was supported by TACOM/ ARDEC, Picatinny Arsenal. The authors are N. Strigul et al. / Desalination 248 (2009) 771–782 grateful to Washington Braida and David Vaccari for their valuable comments. References [1] Nanoscience and Nanotechnologies: Opportunities and Uncertainties, The Royal Society & The Royal Academy of Engineering, Royal Society Publications, London, UK (2004). [2] U.S. Environmental Protection Agency Nanotechnology White Paper (2005) (http://www.epa. gov/osa/pdfs/EPA_nanotechnology_white_paper_ external_review_draft_12-02-2005.pdf). [3] V.L. Colvin, The potential environmental impact of engineered nanoparticles. Nat. Biotechnol., 21 (2003) 1166–1170. [4] R. Owen and R. Handy, Formulating the problems for environmental risk assessment of nanomaterials. Environ. Sci. Technol., 41 (2007) 5582–5588. [5] R. Behra and H. Krug, Nanoecotoxicology: Nanoparticles at large. Nat. Nanotechnol., 3 (2008) 253–254. [6] M. Crane, R.D. Handy, J. Garrod and R. Owen, Ecotoxicity test methods and environmental hazard assessment for engineered nanoparticles. Ecotoxicology, 17 (2008) 421–437. [7] R.D. Handy, F. Von Der Kammer, J.R. Lead, M. Hassellov, R. Owen and M. Crane, The ecotoxicology and chemistry of manufactured nanoparticles. Ecotoxicology, 17 (2008) 287–314. [8] R.D. Handy, R. Owen and E. Valsami-Jones, The ecotoxicology of nanoparticles and nanomaterials: Current status, knowledge gaps, challenges, and future needs. Ecotoxicology, 17 (2008) 315–325. [9] M. Mortimer, K. Kasemets, M. Heinlaan, I. Kurvet and A. Kahru, High throughput kinetic Vibrio fischeri bioluminescence inhibition assay for study of toxic effects of nanoparticles. Toxicol. In Vitro, 22 (2008) 1412–1417. [10] A. Kahru, H.-C. Dubourguier, I. Blinova, A. Ivask and K. Kasemets, Biotests and biosensors for ecotoxicology of metal oxide nanoparticles: A minireview. Sensors, 8 (2008) 5153–5170. [11] M.S. Petersen, C.C. Petersen, R. Agger, M. Sutmuller, M.R. Jensen, P.G. Sørensen, M.W. Mortensen, T. Hansen, T. Bjørnholm, H.J. Gundersen, R. Huiskamp and M. Hokland, Boron nanoparticles inhibit tumour growth by boron neutron capture therapy in the murine B16-OVA model. Anticancer Res., 28 (2008) 571–576. 781 [12] P. J. Kaste and B.M. Rice, Novel energetic materials for the future force: the army pursues the next generation of propellant and explosives. The AMPTIAC Quarterly, 8 (2004) 84–90. [13] A. Fujishima, T.N. Rao and D.A. Tryk, Titanium dioxide photocatalysis. J. Photochem. Photobiol. C, 1 (2000) 1–21. [14] X. Meng, M. Dadachov, G.P. Korfiatis and C. Christodoulatos, Methods of preparing a surface-activated titanium oxide product and of using same in water treatment processes. U.S. Patent Application Number 6,919,029. (2005). [15] M.E. Pena, G.P. Korfiatis, M. Patel, L. Lippincott and X. Meng, Adsorption of As(V) and As(III) by nanocrystalline titanium dioxide. Water Res., 39 (2005) 2327–2337. [16] K. Hund-Rinke and M. Simon, Ecotoxic effect of photocatalytic active nanoparticles (TiO2) on algae and daphnids. Environ. Sci. Pollut. Res., 13 (2006) 225–232. [17] S.B. Lovern and R. Klaper, Daphnia magna mortality when exposed to titanium dioxide and fullerene (C) nanoparticles. Environ. Toxicol. Chem., 25 (2006) 1132–1137. [18] S.B. Lovern, J.R. Strickler and R. Klaper, Behavioral and physiological changes in Daphnia magna when exposed to nanoparticle suspensions (titanium dioxide, nano-C, and CHxCHx). Environ. Sci. Technol., 41 (2007) 4465–4470. [19] R.J. Griffitt, J. Luo, J. Gao, J.C. Bonzongo and D.S. Barber, Effects of particle composition and species on toxicity of metallic nanomaterials in aquatic organisms. Environ. Toxicol. Chem., 27 (2008) 1972–1978. [20] M. Heinlaan, A. Ivask, I. Blinova, H.-C. Dubourguier and A. Kahru, Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere, 71 (2008) 1308–1316. [21] X. Zhu, L. Zhu, Z. Duan, R. Qi, Y. Li and Y. Lang, Comparative toxicity of several metal oxide nanoparticle aqueous suspensions to Zebrafish (Danio rerio) early developmental stage. J. Environ. Sci. Health. A, 43 (2008) 278–284. [22] A.J. Wagner, C.A. Bleckmann, R.C. Murdock, A.M. Schrand, J.J. Schlager and S.M. Hussain, Cellular interaction of different forms of aluminum nanoparticles in rat alveolar macrophages. J. Phys. Chem. B, 111, 25, (2007) 7353–7359. 782 N. Strigul et al. / Desalination 248 (2009) 771–782 [23] R. Doshi, W. Braida, C. Christodoulatos, M. Wazne and G. O’Connor, Nano-aluminum: Transport through sand columns and environmental effects on plants and soil communities. Environ. Res., 106 (2008) 296–303. [24] K.L.E. Kaiser, Correlations of Vibrio fischeri bacteria test data with bioassay data for other organisms. Environ. Health Perspect., 106 (Sup. 2) (1998) 583–591. [25] X. Liu, M. Wazne, C. Christodoulatos, K.L. Jasinkiewicz. Aggregation and deposition behavior of boron nanoparticles in porous media. J. Colloid Interface Sci., 330 (2009) 90–96. [26] C.C. Choy, M. Wazne and X. Meng, Application of an empirical transport model to simulate retention of nanocrystalline titanium dioxide in sand columns. Chemosphere, 71, (2008) 1794–1801.