Photochemical Transformation and Photoinduced Toxicity Reduction

Mar 27, 2014 - ...
2 downloads 5 Views 3MB Size
Article pubs.acs.org/est

Photochemical Transformation and Photoinduced Toxicity Reduction of Silver Nanoparticles in the Presence of Perfluorocarboxylic Acids under UV Irradiation Yang Li,† Junfeng Niu,*,† Enxiang Shang,† and John Crittenden*,†,‡ †

State Key Laboratory of Water Environment Simulation, School of Environment, Beijing Normal University, Beijing 100875, People’s Republic of China ‡ School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States S Supporting Information *

ABSTRACT: The impact of perfluorocarboxylic acids (PFCAs) with carbon chain length C2 to C8 on the dissolution, aggregation, reactive oxygen species (ROS) generation, and toxicity of citrate-coated AgNPs was investigated under UV irradiation. The presence of PFCAs decreased dissolution, aggregation, ROS generation, and toxicity of AgNPs because the negatively charged PFCAs sorbed on AgNP surface enhanced their stability. Both dissolution and aggregation rate of AgNPs decreased with chain length of PFCAs under UV irradiation, primarily because PFCAs with longer chain length sorbed on AgNP surface could form thicker coatings. The dissolution of AgNPs followed pseudo-first-order kinetics, and the rate constant decreased from 0.58 h−1 with C2 to 0.30 h−1 with C8. The hydrodynamic diameters of AgNPs linearly increased under UV irradiation with aggregation rates ranged from 72.1 to 143.5 nm/h. O2•− generation was observed in AgNP suspension with quantum yield of 0.12%, but was completely suppressed by PFCAs because they inhibited the interaction between photoelectrons and O2. A linear correlation was established between dissolved Ag+ and bacterial survival rates of AgNPs with and without PFCAs under UV irradiation. This study highlights the necessity of considering coexisting organic contaminants when investigating the environmental behaviors of AgNPs.



INTRODUCTION Silver nanoparticles (AgNPs) have many attributes that include unique optical, antimicrobial, catalysis, and electrical conductivity properties.1,2 Such striking traits have made them attractive for incorporation into a variety of commercial products, including clothes, pharmaceuticals, catalysts, and textiles.1,2 Widespread usage may increase their potential release into the natural aquatic or built environment, which could harm the environment or human health.3,4 When AgNPs are released into the natural aquatic environment, they will inevitably be exposed to light from the sun or artificial lighting. As AgNPs are photosensitive materials,5,6 their kinetics of dissolution, aggregation, and reactive oxygen species (ROS) generation are impacted by their exposure to light.7,8 For instance, UV lamp, solar, or simulated sunlight can significantly enhance the Ag+ release rates by AgNPs compared to the dark conditions.2,9 The dipole−dipole attraction between particles driven by solar or simulated sunlight irradiation leads to aggregation and sedimentation of citrate-coated AgNPs.2,6,10 Under UV irradiation, the local surface Plasmon resonance of AgNPs induces a strong adsorption of the incident photon energy and thus facilitates their generation of superoxide radicals (O2•−) and hydroxyl radicals (•OH).2 Therefore, it is imperative to perform a comparative study on the light-induced © 2014 American Chemical Society

generation of different species (ions, aggregates, or isolated NPs) of AgNPs and ROS, which might significantly change their transport and biological effects in ecosystems. In recent decades, the release of some emerging persistent organic pollutants (POPs) into natural water is a big issue due to their wide application in industry.11,12 For example, the concentrations of perfluorocarboxylic acids (PFCAs) in the natural aquatic environment increased exponentially.11 Different from other POPs, these recalcitrant organic pollutants have high solubility,13,14 resulting in their wide distribution in water with high concentration. As coexisting pollutants, they can interact with the suspended NPs and then form complex compounds because the high surface-to-volume ratio inherent to NPs facilitates the adsorption of PFCAs onto NP surfaces.11 As an anionic surfactant, the attached PFCAs could alter the electrostatic or electrosteric force between NPs, which subsequently changed the stability, photochemical activity, and toxicity of NPs. Recent studies have demonstrated the necessity to pay more attention to the combined effect of NPs Received: Revised: Accepted: Published: 4946

February 4, 2014 March 25, 2014 March 27, 2014 March 27, 2014 dx.doi.org/10.1021/es500596a | Environ. Sci. Technol. 2014, 48, 4946−4953

Environmental Science & Technology

Article

PFCAs in the filtrate were analyzed by high performance liquid chromatography coupled with a triple-stage quadrupole mass spectrometer (API3200; Applied Biosystems, U.S.A.).23 The control experiments without AgNPs demonstrated that the evaporation and sorption of PFCAs by bottle surface or ultrafilter was negligible. Measurement of Dissolution under UV Exposure. Periodically throughout the experiments (0, 0.5, 1, 1.5, 2, 3, and 4 h), the AgNP suspension with or without PFCAs were collected and filtered by the same ultrafilter. 24 After centrifugation, the filtrate was mixed with trace-metal grade HNO3 (67−70%, w/w). The mixture was analyzed by inductively coupled plasma-mass spectrometry (ICP-MS, Elan DRC II, PerkinElmer, U.S.A.) to determine the dissolved amount of Ag+.25 Control experiments were performed in the dark to measure the chemical dissolution of AgNPs. Assessment of Aggregation under UV Exposure. TEM images of AgNPs before and after 4-h UV exposure were obtained to image silver aggregates and derive their fractal dimensions (Df), which was derived by eq 1:26

and recalcitrant organic pollutants on NP toxicity.15−17 Substantial efforts have focused on the attachment of organic contaminants on nC60 and their combined toxicity effect.15−17 As a typical metal NP, AgNPs have many special characteristics different from nC60, such as photochemical dissolution.2,10 Is the potential contaminant-carrying effect of AgNPs on their toxicity similar to nC60 or more complicated due to the ion release of AgNPs? Investigation on the interaction between AgNPs and PFCAs and their combined effect mechanism on physicochemical processes and toxicity of AgNPs could contribute to understanding the transport and bioavailability of AgNPs in the natural aqueous environment. In this study, we comprehensively evaluated the effects of straight-chain PFCAs with chain lengths from C2 to C8 on the kinetics of three major aqueous processes of AgNPs, including dissolution, aggregation, ROS generation, and subsequently their toxicity toward E. coli cells under UV-365 irradiation. Citrate-coated AgNPs are selected because they are recommended by the Economic Co-operation and Development (OECD) organization for standard use in toxicity testing.18,19 UV-365 was chosen to enhance dissolved Ag+ and ROS generation concentrations, which facilitate detecting Ag+ and ROS in a timely fashion and investigating their toxic effects on bacteria.20,21 In addition, UV-365 is a primary component of UV irradiation that can reach the Earth’s surface.8,22 To the best of our knowledge, this is the first study focused on the carrier effect of AgNPs in the aquatic environment and investigating the influence of environmentally relevant organic contaminants on the physicochemical properties and toxicity of AgNPs.

n = kδ −Df

(1)

where n is the number of the nonoverlapping equal boxes that fills the projected surface area of the agglomerate, k is a constant, and δ is a box size (cm). The values of Df were measured using a box counting algorithm in ImageJ software with the plugin FracLac_2003 K (NIH, Bethesda, MD, http:// www.rsb.info.nih.gov/ij).26 Detection of Photogenerated ROS. Electron spin resonance spectrometry (ESR; Bruker ESP-300E, Germany) equipped with a quanta-Ray Nd:YAG laser system as the irradiation source at λ 355 nm was employed to determine the photogenerated ROS. The settings of ESR spectrometer were as follows: (1) center field of 3480.00 G; (2) microwave power of 10 mW; (3) receiver gain of 1.00 × 105; (4) modulation frequency of 100.00 kHz; and (5) modulation amplitude of 2.071 G. 2,2,6,6-tetramethyl-4-hydroxy-piperidinyloxy (TEMP) was used as spin trap for 1O2 and 2,5,5-dimethyl-1-pyrroline-Noxide (DMPO) was used as spin trap for •OH and O2•−. Indicator assays were conducted to confirm the photogenerated ROS by AgNPs with or without PFCAs. Twenty μM of para-chlorobenzoic acid (pCBA), 100 μM of 3-bis(2methoxy-4-nitro-5-sulfophehyl)-2H-tetrazolium-5-carboxanilide (XTT), and 850 μM of furfuryl alcohol (FFA) were used as indicators for •OH, O2•−, and 1O2, respectively.20,21,24 The quantum yield (QY) for ROS photogeneration was calculated by eq 2:27,28



MATERIALS AND METHODS Photochemical Experiments. The chemicals used in this study and characterization methods of AgNPs are shown in Sections S1 and S2 of the Supporting Information (SI), respectively. All photochemical experiments, including measurement of dissolution, assessment of aggregation, detection of ROS, and assessment of toxicity were conducted under irradiation of a ultraviolet lamp (UVP model UVGL-21) with a wavelength of 365 nm (UV-365). The light intensity in the center of the reaction solution was 1.8 × 10−9 Einstein·cm−2· s−1. For the dissolution, aggregation, and toxicity experiments, 100 mL of aqueous suspension containing 100 μg/L PFCAs and AgNPs with total silver concentration of 100 μg/L were put into polypropylene beakers. However, photogenerated ROS concentrations were not detectable at such low AgNP concentration. Thus, the ROS detection experiments were performed at a total silver concentration of 5 mg/L. The initial pH of AgNP suspension was 5.6. No electrolytes were added to the AgNP suspensions to eliminate colloidal instability of AgNPs. The reaction temperature was maintained at 22 ± 2 °C by a constant-temperature water bath. The light intensity at different depths of the reaction solution is shown in SI Figure S1. Prior to irradiation, the suspensions of AgNPs and PFCAs were mixed in polypropylene beakers and then stirred for 4 h in the dark to achieve sorption−desorption equilibrium. Preliminary experiments suggested that 1 h was sufficient to achieve sorption equilibrium for all PFCAs. Periodically throughout the experiments (0, 0.5, 1, 1.5, 2, 3, and 4 h), the mixture was collected and filtered by an Amicon Ultra-4 centrifugal ultrafilter containing porous cellulose membranes with a nominal particle size limit of 1−2 nm (3K, Millipore, U.S.A.) to remove NPs from the suspension. The concentrations of

QY = (R /P) × 100%

(2)

where R is the number of photogenerated ROS molecules per unit time (s−1), and P is the number of adsorbed photons per unit time (Einstein s−1 ). R was determined by ROS concentration: R = (C·v ·A v )/t

(3)

where C is the ROS concentration (μM), v is the volume of reaction solution (100 mL), Av is Avogadro’s number (6.02 × 1023), and t is the irradiation time (4 × 3600 s). The number of P can be calculated as follows:29

P = S·

∫x

x2

1

4947

Ia·dx

(4)

dx.doi.org/10.1021/es500596a | Environ. Sci. Technol. 2014, 48, 4946−4953

Environmental Science & Technology

Article

Table 1. Correlation Coefficients (R2) and Fitting Parameters for UV-Induced Dissolution Kinetics of AgNPs in the Absence or Presence of PFCAs with Different Chain Length PFCAs

no PFCAs

C2

C3

C4

C5

C6

C7

C8

R2 k (h‑1) [Ag+]max (μg/L)

0.99 0.65 35.5

0.98 0.58 31.4

0.99 0.61 29.9

0.97 0.51 21.7

0.98 0.43 22.1

0.97 0.42 19.8

0.96 0.34 18.8

0.94 0.30 15.5

where S is the cross-sectional area of the beaker exposed to UV irradiation (58.0 cm2), Ia is the number of photons absorbed per volume of solution at particular point (Einstein·cm−3·s−1, SI eq S3), x is the depth of reaction solution (cm), and x1 and x2 are the lower and upper limits for the depth of the reaction solutions (0 and 2.5 cm), respectively. Assessment of Toxicity under UV Exposure. E. coli cells were cultured and harvested according to the published method.30 For the toxicity assays, the density of the bacteria cells was approximately 108 CFU/mL. The mixed suspension was exposed to UV light for 4 h, which was chosen because the bacteria were not inactivated with exposure less than 4 h. Moreover, the 4-h UV dose ranged from 2160 to 8640 mW·s/ cm2 at different depths of the reaction solution, which was greater than the EPA-proposed minimum UV dosage (12−186 mW·s/cm2) in the drinking water disinfection.31 The samples were collected to determine the bacterial viability by the traditional plate count method.15,21,23 Statistical Analysis. All photochemical experiments were conducted at least in triplicate to confirm their reproducibility. The data points were expressed as mean values with standard deviations (SD) of the three parallel samples. Statistical significance was evaluated by Student’s t-test with p = 0.05. The t-Test was performed by Origin 7.5.

significantly affect the light exposure and photochemical dissolution of AgNPs. There is apparent dependence of k and [Ag+]max values on chain length of PFCAs, which followed the order of kAgNPs (no PFCAs) ≈ 1.1kC2 ≈ 1.1kC3 ≈ 1.3kC4 ≈ 1.5kC5 ≈ 1.5kC6 ≈ 1.9kC7 ≈ 2.2kC8 and [Ag+]max, AgNPs (no PFCAs) ≈ 1.1[Ag+]max, C2 ≈ 1.2[Ag+]max, C3 ≈ 1.6[Ag+]max, C4 ≈ 1.6[Ag+]max, C5 ≈ 1.8[Ag+]max, C6 ≈ 1.9[Ag+]max, C7 ≈ 2.3[Ag+]max, C8, respectively. The addition of PFCAs apparently decreased the values of k and [Ag+]max, which were inversely proportional to the chain length of PFCAs. The amount of PFCAs sorbed by AgNPs at equilibrium time (qe, mg/g) is shown in the inset of Figure 1. According to the



RESULTS AND DISCUSSION Modeling Analysis of Photochemical Dissolution Kinetics. The characterization of citrate-coated AgNPs are shown in SI Section S4. SI Figure S7(a),(b) shows the photochemical dissolution kinetics of AgNPs with or without PFCAs under UV irradiation. In comparison with the photochemical dissolution, the 4-h released Ag+ concentrations of AgNPs with or without PFCAs in the dark were all less than 10.0 μg/L (