Using Rhodamine 6G-Modified Gold Nanoparticles To Detect Organic

Jan 26, 2011 - We developed a gold nanoparticle (Au NP)-based fluorescence sensor for the detection of mercury ions in aqueous solutions. After introd...
0 downloads 0 Views 2MB Size
Download PDF
ARTICLE pubs.acs.org/est

Using Rhodamine 6G-Modified Gold Nanoparticles To Detect Organic Mercury Species in Highly Saline Solutions Hsin-Yun Chang,† Tung-Ming Hsiung,† Yu-Fen Huang,*,‡ and Chih-Ching Huang*,†,§ †

Institute of Bioscience and Biotechnology and §Center for Marine Bioenvironment and Biotechnology (CMBB), National Taiwan Ocean University, Keelung, Taiwan ‡ Department of Biomedical Engineering and Environmental Sciences, National Tsing Hua University, Hsinchu, Taiwan

bS Supporting Information ABSTRACT: We developed a gold nanoparticle (Au NP)-based fluorescence sensor for the detection of mercury ions in aqueous solutions. After introducing bovine serum albumin (BSA) to a solution of rhodamine 6G (R6G) and 3-mercaptopropionic acid (MPA)-modified Au NPs, the as-prepared BSA@R6G/MPA-Au NP probe could sense mercury ions under high salt conditions. This probe operated through a mechanism involving mercury species depositing onto the surfaces of the Au NPs and releasing R6G molecules into the solution, causing the fluorescence intensity of the BSA@R6G/MPA-AuNP solution to increase. We improved the selectivity of the nanosensor by adding masking agents (ethylenediamine tetraacetic acid (EDTA) and Na2S) or tellurium nanowires (Te NWs) to the sample solutions. In the presence of 1.0 mM EDTA and 10 μM Na2S, the selectivities of this system toward phenylmercury (PhHg(I)) over other metal ions and mercury species were greater than 200- and 10-fold, respectively. The limit of detection (LOD), at a signal-to-noise ratio of 3, for PhHg(I) was 20 nM. Selective detection of the total organic mercury (methylmercury (MeHg(I)), ethylmercury (EtHg(I)), and PhHg(I)) was possible when using the BSA@R6G/MPA-Au NPs in conjunction with Te NWs (3.0 nM). The selectivity of this nanosensor system for the total organic mercury over Hg(II) was remarkably high (100-fold) with an LOD for organic mercury of 10 nM. We also demonstrated the feasibility of using the developed nanosensor for rapid determination of mercury species in river, sea, and tap water as well as in fish samples.

1. INTRODUCTION Mercury and its organometallic compounds (mainly methylmercury) are the most toxic pollutants in aquatic ecosystems.1 The most important chemical forms of mercury in aquatic ecosystems are inorganic mercurous (Hg2(II)) and mercuric (Hg(II)) cations and organic mono- and dialkylated and/or -arylated mercury species.2 Inorganic Hg(II) ions, which are water soluble, have strong affinities for many inorganic and organic ligands, especially those containing sulfur atoms; they are caustic and carcinogenic materials that exhibit high cellular toxicities.3 Monomethylmercury (MeHg), the most prevalent organomercury species, is a potent neurotoxin that impairs the central nervous system and, in severe cases, causes irreversible damage to the brain.4 Furthermore, fetuses and infants are more sensitive to methylmercury exposure than are adults. In aquatic environments, MeHg(I) is formed through the biomethylation of inorganic Hg(II), thereby accumulating in food webs. It is also biomagnified by factors of up to 107 in predatory fish, causing adverse effects in humans and other wildlife that consume fish.4 Consequently, the U.S. Environmental Protection Agency (EPA) has set the maximum allowable level of Hg(II) in drinking water at 2 ppb5 and a maximum level of 0.3 μg/g in tissue (wet weight) for fresh water fish, estuarine fish, and shellfish tissue.6 Therefore, development of reliable and rapid routine methodologies for determination of Hg(II) in rivers and larger bodies of water, as well as methylmercury in foodstuffs, is an attractive challenge. Several analytical techniques based on atomic spectrometry, including cold vapor atomic absorption spectroscopy (CV-AAS), cold vapor atomic fluorescence spectrometry (CV-AFS), and r 2011 American Chemical Society

inductively coupled plasma mass spectrometry (ICP-MS), have been developed for analysis of the total mercury in aquatic ecosystems.7 Although these analytical strategies provide low detection limits, for point-of-use applications, they are rather complicated, time consuming, and costly. In addition, only those methods incorporating preseparation procedures, such as sequential leaching and extraction procedures that allow for separation of inorganic forms from total mercury, are suitable for speciation of mercury.7 Some of these techniques are readily subject to interference when analyzing highly saline solutions. Thus, the search continues for alternative approaches that are simple, rapid, and sufficiently sensitive for determination of mercury species in real samples. Recently, researchers have developed several colorimetric and fluorescent Au NP-based and oligonucleotide-based assays for detection of mercury.8-19 For example, in the presence of a suitable masking reagent, such as pyridinedicarboxylic acid, thiolfunctionalized gold nanoparticles (Au NPs) exhibit high selectivity for Hg(II) ions.9-11 Hg(II) ions can induce aggregation of NPs, leading to a significant colorimetric change, a quenching of the fluorescence, or an enhancement of the hyper-Rayleigh scattering intensity.9-19 Although these methods provide greater affordability and portability to Hg(II) ion detection, few of them Received: October 6, 2010 Accepted: January 6, 2011 Revised: December 31, 2010 Published: January 26, 2011 1534

dx.doi.org/10.1021/es103369d | Environ. Sci. Technol. 2011, 45, 1534–1539

Environmental Science & Technology Scheme 1. Schematic Representation of the Preparation of the BSA@R6G/MPA-Au NP Fluorescent Sensor for Detection of Mercury Ions Based on Displacement of R6G Units on Au NPs

can be applied to environmental matrices that feature high salinities. In addition, they have poor abilities to accurately identify and quantify different mercury species. Herein, we describe a simple, rapid, and inexpensive fluorescence assay for detecting Hg(II) and organomercury species, such as MeHg(I), ethylmercury (EtHg(I)), and phenylmercury (PhHg(I)) in salt solutions and fish samples. Through introducing bovine serum albumin (BSA) to a solution of rhodamine 6G (R6G) and 3-mercaptopropionic acid (MPA)-modified Au NPs (14.2 nm), we prepared the BSA@R6G/MPA-Au NP probe for sensing mercury ions. We employed BSA to protect the R6G/MPA-Au NPs from serious aggregation in high salt solutions. This probe operated based on the fact that mercury ions deposited on the surfaces of the Au NPs induce the release of R6G molecules into solution and thus restore the florescence of R6G (see Scheme 1). In the presence of masking reagents (ethylenediamine tetraacetic acid (EDTA) and Na2S) and tellurium nanowires (Te NWs), our BSA@R6G/MPAAu NP sensor is capable of detecting the specific level of PhHg(I) and the total content of the organic mercury species, respectively. This approach allows for accurate analysis of the levels of methylmercury in fish reference materials without the need for complicated pretreatment processes to extract the inorganic mercury ions from the total mercury content.

2. EXPERIMENTAL METHODS 2.1. Materials. MPA, R6G, trisodium citrate, EDTA, and all of the metal salts used in this study were purchased from Aldrich (Milwaukee, WI). Hydrogen tetrachloroaurate(III) trihydrate, sodium thiosulfate pentahydrate, and sodium dodecyl sulfate (SDS) were obtained from Acros (Geel, Belgium). We purchased tellurium dioxide powder (particle size not determined or provided; 99.9%) and hydrazine monohydrate (N2H4 3 H2O, 80%) from Showa (Tokyo, Japan). Milli-Q ultrapure water was used in each experiment. The sodium phosphate buffers (pH 3-10) were prepared with phosphoric acid and trisodium phosphate. 2.2. BSA@R6G/MPA-Au NPs. We prepared the Au NPs (14.2 ( 0.6 nm) according to the previous report.11 An aliquot of MPA solution (1.0 mM, 187.5 μL) was stirred into a solution of the 14.2 nm Au NPs (15 nM, 10 mL) and incubated for 1 h. Modification of MPA on Au NPs was via the strong S-Au bond (∼40 kcal/mol). The MPA-Au NP (4 mL) was diluted with 5 mM sodium phosphate (pH 5) to 9.4 mL, and then R6G solution (10 μM, 0.6 mL) was added, and the resulting solution was equilibrated at ambient temperatures for another 1 h. Prior to use, we added BSA (0.1 mM,

ARTICLE

300 μL) to stabilize the as-prepared R6G/MPA-Au NPs against the high salt conditions. The fluorescence spectrum of the BSA@R6G/ MPA-Au NP solution (excitation at 520 nm; Synergy 4 Multi-Mode Microplate Reader; Biotek Instruments, Winooski, VT) revealed very weak fluorescence, suggesting that very little R6G was free in solution. 2.3. Te NWs. We added slowly TeO2 powder (16 mg) to N2H4 3 H2O (10 mL) in a beaker at room temperature under constant magnetic stirring. The powder dissolved completely within 10 min. The solution changed color sequentially from colorless, to amber, to purple, and eventually to blue within 1 h. The blue color indicated formation of Te NWs.20 The solution was diluted 5-fold with 10 mM SDS and then stirred for another 30 min. The average diameter and length of the Te NWs were 12.1 ((1.9) nm and 456 ((110) nm, respectively. The corresponding concentration of the as-prepared Te NWs was 3.0 nM.20 2.4. Detection of Heavy Metal Ions. A stock solution of metal ions (0.1 M) was prepared in 0.1 M HNO3 and diluted to 0100 μM in ultrapure water. We separately added aliquots of metal ion solutions (50 μL) to 5 mM sodium phosphate buffer (pH 5) solutions containing 0.6 nM BSA@R6G/MPA-Au NPs to give final volumes of 500 μL. After equilibration at ambient temperatures for 1 h, the mixtures were transferred separately into 96-well microtiter plates, and their fluorescence spectra were recorded using a Synergy 4 Multi-Mode Microplate Reader. For selective determination of MeHg species, we equilibrated masking agents with the metal ion solutions for 10 min prior to addition of BSA@R6G/MPA-Au NPs. Herein, we provide only the final concentrations of the species. 2.5. Analysis of Water Samples. Water samples collected from the East China Sea, a river on the campus of the National Taiwan Ocean University (NTOU), and a tap were filtered through a 0.2 μm membrane. Aliquots of the water samples (250 μL) were spiked with standard solutions of mercury species (10 μL) at the desired concentrations. We then diluted the spiked samples to 500 μL using a solution containing the BSA@R6G/MPA-Au NPs (0.6 nM) and 5 mM sodium phosphate (pH 5). The spiked samples were analyzed separately using ICP-MS and the developed sensing technique. 2.6. Analysis of Methylmercury in Fish Samples. To isolate the mercury species from the selected fish (certified reference material (CRM) dogfish muscle DORM-2), 1.0 g of the fish sample was add to an extraction reagent (10 mL), comprising 6.0 M HCl and 0.1 M NaCl and sonicated for 45 min at 55 °C. After filtration (filter paper no. 389; disk diameter = 12.5 cm), the supernatant was diluted 10-fold with 5 mM sodium phosphate buffer (pH 5) and analyzed using the BSA@R6G/MPA-Au NP sensor.

3. RESULTS AND DISCUSSION 3.1. Preparation of BSA@R6G/MPA-Au NPs for Sensing Hg(II). In a previous study, we used rhodamine B (RB)/MPA-

capped Au NPs as nanosensors for selective detection of Hg(II) species.11 Rhodamine dye molecules are highly fluorescent dyes that adsorb noncovalently onto the surfaces of Au NPs. Once they are in close proximity to the Au NPs, the fluorescence of the dye molecules is quenched efficiently through FRET and electron transfer processes. Addition of Hg(II) species, however, can restore the fluorescence of dye by removing these molecules from the surfaces of the Au NPs. A combination of the electrostatic and π-π interactions between the negative MPA and citrate moieties and the positive groups of the dye molecules controlled the interaction of Au NPs and rhodamine dye molecules (R6G or RB).21 The positively charged R6G molecules (containing imino groups) showed stronger interactions with Au 1535

dx.doi.org/10.1021/es103369d |Environ. Sci. Technol. 2011, 45, 1534–1539

Environmental Science & Technology

ARTICLE

Figure 1. Fluorescence spectra of solutions of (a) R6G and (b, c) BSA@R6G/MPA-Au NPs in the (b) absence and (c) presence of Hg(II) (1.0 μM). Concentrations of the Au NPs, MPA, R6G, and BSA were 0.6 nM, 750 nM, 60 nM, and 0.6 μM, respectively. (Inset) Fluorescence photographs of (a) R6G and (b, c) BSA@R6G/MPA-Au NPs in the (b) absence and (c) presence of Hg(II) upon excitation under a handheld UV lamp (365 nm). Buffer: 5 mM sodium phosphate, pH 5. Excitation wavelength: 520 nm. Fluorescence intensities (IF) are plotted in arbitrary units (au).

NPs. We should note that the carboxyl and imino groups in rhodamine B made it uncharged overall, whereas the carboxyl group in R6G was esterified and therefore carried a positive charge. From the fluorescence intensities of the supernatants of MPA-Au NPs (6 nM) in the presence of RB (600 nM) and R6G (600 nM), we estimated the amount of dye absorbed on each Au NP was 62 RB molecules and 95 R6G molecules, respectively. The stronger electrostatic interaction likely provided lower fluorescence background of the R6G/MPA-Au NP probe and a wider dynamic range for detection of mercuric ions. To expand their use to detection of mercury in highly saline solutions, we further protected the R6G/MPA-Au NPs with bulky BSA units. Au NPs exhibit high tolerance toward the salinity of aqueous solutions after their capping with bovine serum albumin (BSA), which appears to bind spontaneously to the surfaces of negatively charged Au NPs through a predominantly electrostatic mechanism, although a contribution from the hydrophobic interactions with the surface adlayer is also probably involved.22 The BSA-capped R6G/MPA-Au NPs were mainly due to the steric interactions (bulky proteins on the surface preventing neighboring Au NPs from getting in close enough proximity to interact and aggregate) and high hydrophilicity of BSA, thereby stabilizing the R6G/MPA-Au NPs. The BSA@R6G/MPAAu NPs were stable (no aggregation)23 in 5 mM sodium phosphate solutions (pH 5) containing NaCl at concentrations of up to 500 mM when the concentration ratio of BSA to Au NPs ([BSA]/[Au NPs]) was as high as 1000 (see Figure S1 of the Supporting Information). We performed proof-of-concept experiments for detection of Hg(II) ions using the BSA@R6G/MPA-Au NPs (0.6 nM). Our BSA@ R6G/MPA-Au NP probe exhibited a low fluorescence background intensity and showed a 4.1-fold increase in fluorescence upon addition of 1.0 μM Hg(II) in a solution of 5 mM sodium phosphate at pH 5 (see Figure 1). 3.2. Detection of Mercury(II) and Organomercury Ions. To determine if BSA@R6G/MPA-Au NPs are also able to detect other mercury species, we investigated the fluorescence intensities of the BSA@R6G/MPA-Au NPs that followed addition of specific mercury species (1.0 μM): Hg(II), MeHg(I), EtHg(I), and PhHg(I). All of these species (Hg(II) and organomercury

Figure 2. (A) Values of ((IF - IF0)/IF0) for responses of the BSA@ R6G/MPA-Au NPs (0.6 nM) for Hg(II), PhHg(I), EtHg(I), and MeHg(I) in 5 mM sodium phosphate (pH 5). (B) ICP-MS-based quantitation of the mercury atoms on the Au NPs. The concentration of each mercury species was 1.0 μM. IF and IF0 in A represent the fluorescence intensities of the BSA@R6G/MPA-Au NPs at 550 nm in the presence and absence of mercury species, respectively. Error bars represent the standard deviations from four repeated experiments. Other conditions were the same as those described in Figure 1.

ions) had strong affinities for the BSA@R6G/MPA-Au NPs through strong Hg-Au(I) metallophilic and Hg-carboxylate group of MPA interactions,18,24-26 leading to significant increases in the fluorescence. In a control experiment, we observed that all mercury species (Hg(II), MeHg(I), EtHg(I), and PhHg(I)) over the concentration range from 100 nM to 10 μM did not cause changes in the fluorescence intensity of R6G solutions (100 nM), revealing no fluorescence quenching due to the heavy atom effect. The fluorescence enhancements ((IF - IF0)/IF0) in which IF and IF0 represent the fluorescence intensities of BSA@ R6G/MPA-Au NPs at 550 nm in the presence and absence of the mercury species, respectively (see Figure 2A), followed the order Hg(II) (4.1) > PhHg(I) (3.7) > EtHg(I) (2.7) > MeHg(I) (1.8). This trend was probably due to the fact that Hg(II) species form the most stable complexes with the carboxylic acid units of the surface MPA units (log Kf = ca. 10.1) as opposed to other organomercury species (log Kf = ca. 3-5);25,26 as a result, more R6G molecules departed from the Au NP surfaces and, thereby, restored the higher fluorescence signal of R6G in the presence of Hg(II). To confirm this hypothesis, we used ICP-MS to quantify the mercury atoms obtained from each precipitate after a series of centrifugation/washing steps to remove any unbound species. Deposition of mercury atoms on the Au NPs followed the order Hg(II) > PhHg(I) > EtHg(I) > MeHg(I), a trend that was consistent with that observed for restoration of the fluorescence (see Figure 2B). When considering the relatively large effect of PhHg(I), we could not exclude the steric effects induced during 1536

dx.doi.org/10.1021/es103369d |Environ. Sci. Technol. 2011, 45, 1534–1539

Environmental Science & Technology the approach of these species toward the Au NPs, thereby causing release of more R6G molecules. According to the literature, MeHg(I) exhibits a stronger potency toward alkanethiol units than does Hg(II).25,26 We suspected that detachment of MPA molecules from the Au NP surfaces in the presence of MeHg(I), which would decrease the coverage area of the thiol ligands, would allow R6G molecules to readsorb onto the surfaces of the Au NPs, leading to the lower level of fluorescence restoration. 3.3. Effect of pH. The as-prepared BSA@R6G/MPA-Au NPs were stable in sodium phosphate buffer at pH values greater than 4.0 as determined by the lack of changes in their surface plasmon resonance band at 520 nm (see Figure S2 of the Supporting Information). The values of ((IF - IF0)/IF0) of the probe solutions in response to the Hg(II), MeHg(I), EtHg(I), and PhHg(I) species decreased gradually upon increasing the pH from 5 to 10 (see Figure S3A of the Supporting Information). Because the fluorescence of R6G was insensitive to the pH, the lower values of ((IF - IF0)/IF0) for the BSA@R6G/MPA-Au NP solutions in the presence of mercury ions under alkaline conditions indicated slight desorptions of the R6G molecules from the Au NP surfaces. At low pH values, R6G molecules tended to be protonated to a greater degree; consequently, their interactions with the MPA-Au NP surfaces were enhanced. Our BSA@R6G/ MPA-Au NP solutions exhibited relatively low background intensities (IF0; data not shown). Furthermore, formation of metal oxides and metal hydroxides became predominant at higher pH values.25,26 Therefore, under alkaline conditions, the mercury species had lower complexion abilities for the MPA units on the Au NP surfaces, and as a result, fewer R6G molecules were displaced. ICP-MS analysis confirmed that the deposition percentage of mercury atoms decreased upon increasing the pH (see Figure S3B of the Supporting Information). Because larger values of ((IF - IF0)/IF0) resulted in more sensitive detection of the mercury species, we chose solutions with pH values of 5 for our subsequent experiments to ensure lower detection limits. 3.4. Selectivity and Sensitivity. To ensure that the nanosensor was highly selective for mercury species, we determined the relative fluorescence intensity of the BSA@R6G/MPA-Au NPs (0.6 nM) in the presence of 22 different metal species (1.0 μM). Figure 3A reveals that addition of Cd(II), Ag(I), Pb(II), and the four mercury species to solutions of BSA@R6G/MPA-Au NPs resulted in an apparent fluorescence restoration whereas the other remaining ions had insignificant effects under the same experimental conditions. These results are similar to those of our previous report,11 suggesting that Cd(II), Ag(I), and Pb(II) are capable of depositing on Au NPs. Notably, chelation, mediated by heavy metal ions (e.g., Pb(II) or Cd (II)), of the carboxylic group of MPA could cause R6G to depart from the surfaces of the Au NPs.11,27 Next, we investigated the effects of masking reagents—EDTA and Na2S as well as mixtures of EDTA and Na2S—on the BSA@R6G/MPA-Au NP sensing system. Gratifyingly, our nanosensor was specific for PhHg(I) with respect to the other metal ion species in the presence of 1.0 mM EDTA and 10 μM Na2S (see Figure 3D). EDTA forms much more stable complexes with divalent metal ions—such as Hg(II) (log β2 = 20.28), Pb(II) (19.8), and Cd(II) (18.2)—than with other metal ions25,26 and greatly assists in suppressing the interference caused by heavy metals. To improve the selectivities of the BSA@R6G/ MPA-Au NPs for organomercury species, we used Na2S as a second masking agent to take advantage of the stronger formation constant of Ag2S (pKf = 50.1). The strong affinity of S2(Na2S) for Hg(II) ions (log Kf ≈ 53) and for MeHg(I) and

ARTICLE

Figure 3. (A) Selectivities of the BSA@R6G/MPA-Au NP (0.6 nM) sensor for different metal ions (1.0 μM) and four mercury species (1.0 μM) in the (A) absence of a masking reagent and (B-E) presence of (B) EDTA (1.0 mM), (C) Na2S (10 μM), (D) EDTA (1.0 mM) and Na2S (10 μM), and (E) Te NWs (3.0 nM). Error bars represent the standard deviations from five repeated experiments. Other conditions were the same as those described in Figure 2.

EtHg(I) (log Kf ≈ 21) suggested that most of the mercury species, except for the bulky PhHg(I) ions, preferred to form stable complexes with S2- ions, leading to fewer R6G molecules detaching from the Au NP surfaces (see Figure 3C). As indicated in Figure 3D, BSA@R6G/MPA-Au NPs (0.6 nM) in 5 mM sodium phosphate (pH 5) containing masking reagents (1.0 mM EDTA and 10 μM Na2S) responded with high selectivity for PhHg(I) ions (1.0 μM) with respect to the other metal ions (200-fold) and with respect to the other mercury species (10fold) at the same concentration. The fluorescence intensity of the BSA@R6G/MPA-Au NPs (0.6 nM) in 5 mM sodium phosphate (pH 5) containing masking reagents (1.0 mM EDTA and 10 μM Na2S) gradually increased upon increasing concentration of PhHg(I) ions (see Figure S4 of the Supporting Information). A linear correlation (R2 = 0.96) existed between the value of ((IF - IF0)/IF0) and the concentration over the range from 100 nM to 2.5 μM. The limit of detection (LOD), with a signal-to-noise (S/N) ratio of 3, of PhHg(I) was 20 nM. To the best of our knowledge, this system is the first nanosensor for selective detection of PhHg(I) ions. Tellurium possesses a strong affinity for inorganic metal ions, such as Hg(II); in contrast, its interactions with organomercury species are relatively weak (see Figure 4). For that reason, the use of as-prepared Te NWs as a masking reagent greatly suppressed the interference from the Cd(II), Ag(I), Pb(II), Pt(II), and Hg(II) ions, allowing the BSA@R6G/MPA-Au NP sensor to exhibit excellent selectivity for organomercury ions (see Figure 3E). The BSA@ R6G/MPA-Au NPs (0.6 nM) in 5 mM sodium phosphate buffer (pH 5) containing Te NWs (3.0 nM) exhibited 100-fold selectivity for MeHg(I), EtHg(I), and PhHg(I) over the other metal ions. Under optimal conditions, we quantified the levels of MeHg(I), 1537

dx.doi.org/10.1021/es103369d |Environ. Sci. Technol. 2011, 45, 1534–1539

Environmental Science & Technology

Figure 4. (A) UV-vis absorption spectra of Te NWs (3.0 nM) in the (a) absence and (b-e) presence of (b) Hg(II) (10 μM), (c) MeHg(I) (10 μM), (d) EtHg(I) (10 μM), and (e) PhHg(I) (10 μM). (B) Deposition of mercury atoms per Te NW in the presence of Hg(II), PhHg(I), EtHg(I), and MeHg(I) in 5 mM sodium phosphate buffer (pH 5); error bars represent the standard deviations from three repeated experiments. (C) TEM images of Te NWs (3.0 nM) in the (a) absence and (b) presence of Hg(II) (10 μM); the arrows highlight the HgTe composites. Other conditions were the same as those described in Figure 3.

EtHg(I), and PhHg(I) by monitoring the fluorescence enhancement ((IF - IF0)/IF0) of the solution of BSA@R6G/MPA-Au NPs (0.6 nM) containing Te NWs (3.0 nM). When the concentration of the respective organomercury ion increased, we observed a gradual increase in the value of ((IF - IF0)/IF0) (see Figure S5 of the Supporting Information). The plots of ((IF - IF0)/IF0) against the concentrations of MeHg(I), EtHg(I), and PhHg(I) over the range from 50 nM to 1.0 μM were all linear with correlation coefficients (R2) of 0.98, 0.99, and 0.98, respectively. This method enabled the analyses of MeHg(I), EtHg(I), and PhHg(I), each with an LOD of 10 nM. Although the sensitivity of this BSA@R6G/MPA-Au NP sensor is 1-2 orders of magnitude less than those of other analytical methods (e.g., ICP-MS), its low cost, ease of use, and lack of sample pretreatment suggest that it has potential application for selective detection of organomercury ions. 3.5. Detection of Organomercury Species in Real Water Samples. To validate that our proposed sensing strategy could have practical application for organomercury analysis in environmental samples, we applied the BSA@R6G/MPA-Au NP sensor to determine the levels of mercury species in river, sea, and tap water samples. Prior to analysis, each of the three samples was filtered through a 0.45 μm membrane and diluted 2-fold in 5 mM

ARTICLE

Figure 5. Relative fluorescence intensities of the BSA@R6G/MPA-Au NP (0.6 nM) sensor for Hg(II), MeHg(I), EtHg(I), and PhHg(I) in ultrapure water, river, sea, and tap water samples. Prior to use, the four water samples were diluted 2-fold in 5 mM sodium phosphate buffer (pH 5) and spiked with 1.0 μM of a mercury species. (A) EDTA (1.0 mM) and Na2S (10 μM) and (B) 3.0 nM Te NWs were preformed as masking reagents, respectively. Error bars represent the standard deviations from three repeated experiments. Other conditions were the same as those described in Figure 2.

sodium phosphate buffer (pH 5). Figure 5A reveals that the fluorescence restoration of the BSA@R6G/MPA-Au NPs (0.6 nM) in the presence of the masking agents (1.0 mM EDTA and 10 μM Na2S) increased significantly for the samples spiked with PhHg(I) ions (1.0 μM); in contrast, only slight or no changes in fluorescence occurred upon addition of Hg(II), MeHg(I), or EtHg(I). ICP-MS analyses confirmed that there were no detectable mercury species in any of these three original water samples. By applying a standard addition method, we estimated the recoveries of PhHg(I) ions in the river, sea, and tap water samples to be 92.8%, 95.7%, and 96.9%, respectively (n = 3). Thus, our proposed method is applicable to the practical analyses of specific PhHg(I) species in environmental samples. Similarly, we observed apparent increases in the fluorescence intensities of the BSA@R6G/MPA-Au NP solutions (0.6 nM) when we spiked a mixture of MeHg(I), EtHg(I), and PhHg(I) (1.0 μM; [MeHg(I)]:[EtHg(I)]:[PhHg(I)] = 1:1:1) in sample solutions containing 3.0 nM Te NWs (see Figure 5B); in contrast, no significant changes occurred before or after addition of inorganic Hg(II) under otherwise identical experimental conditions. The recoveries of total organomercury species in the river, sea, and tap water samples were 90.5%, 100.5%, and 109.6%, respectively. The fact that such high recoveries were possible from complex, highly saline seawater samples suggested that our BSA@R6G/ MPA-Au NP sensor is a practical tool for determination of the total organomercury in environmental samples. 3.6. Detection of MeHg(I) in Fish Sample DORM-2. To evaluate whether our developed nanosensor could provide rapid 1538

dx.doi.org/10.1021/es103369d |Environ. Sci. Technol. 2011, 45, 1534–1539

Environmental Science & Technology screening of the total organomercury content in a fish sample, we tested its effectiveness in the analysis of the certified reference material (CRM) dogfish muscle DORM-2, which had a MeHg(I) content of 4.47 ( 0.32 mg/kg and a total mercury content of 4.64 ( 0.25 mg/kg. Through a standard addition of 0-100 nM MeHg(I) in the presence of Te NWs (3.0 nM) as the masking reagent, we determined that the concentration of MeHg(I) in the fish sample was 4.23 ( 0.42 mg/kg (n = 4) (see Figure S6, Supporting Information). On the basis of a t test (95% confidence level, four degrees of freedom) and an F test (95% confidence level), the results obtained from our method were in good accordance with those determined using ICP-MS (4.35 ( 0.39 mg/kg). We demonstrated the feasibility of using this developed nanosensor for rapid determination of mercury species in river, sea, and tap water samples as well as in fish samples. Relative to our previous RBAu NP sensor,11 the present probe provided the advantages of a high salt tolerance, better selectivity, and greater practicality. To the best of our knowledge, our probe is the first example of Au NP-based sensor for selective detection of inorganic and organic mercury ions. The simple, rapid, and cost-effective sensing system holds great potential for use in detection of mercury species in real samples.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information as noted in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Phone: 011-886-2-24622192 Ext. 5517 (C.-C.H); 011-886-35715131 (Y.-F.H.). E-mail: [email protected] (C.-C.H); [email protected] (Y.-F.H.).

’ ACKNOWLEDGMENT This study was supported by the National Science Council of Taiwan under contract 99-2113-M-019-001-MY2. ’ REFERENCES (1) Boening, D. W. Ecological effects, transport, and fate of mercury: a general review. Chemosphere 2000, 40 (12), 1335–1351. (2) Valko, M.; Morris, H.; Cronin, M. T. D. Metals, toxicity and oxidative stress. Curr. Med. Chem. 2005, 12 (10), 1161–1208. (3) Onyido, I.; Norris, A. R.; Buncel, E. Biomolecule-mercury interactions: Modalities of DNA base-mercury binding mechanisms. Remediation strategies. Chem. Rev. 2004, 104 (12), 5911–5929. (4) Murata, K.; Grandjean, P.; Dakeishi, M. Neurophysiological evidence of methylmercury neurotoxicity. Am. J. Ind. Med. 2007, 50 (10), 765–771. (5) Mercury Update: Impact on Fish Advisories; EPA Fact Sheet EPA823-F-01-011; EPA, Office of Water: Washington, DC, 2001. (6) Guidance for implementing the January 2001 methylmercury water quality criterion; United States Environmental Protection Agency: EPA-823-R-10-001, 2010. (7) Leermakers, M.; Baeyens, W.; Quevauviller, P.; Horvat, M. Mercury in environmental samples: Speciation, artifacts and validation. Trend Anal. Chem. 2005, 24 (5), 383–393. (8) Nolan, E. M.; Lippard, S. J. Tools and tactics for the optical detection of mercuric ion. Chem. Rev. 2008, 108 (9), 3443–3480. (9) Huang, C.-C.; Chang, H.-T. Parameters for selective colorimetric sensing of mercury(II) in aqueous solutions using mercaptopropionic acidmodified gold nanoparticles. Chem. Commun. 2007, 12, 1215–1217.

ARTICLE

(10) Huang, C.-C.; Yang, Z.; Lee, K.-H.; Chang, H.-T. Synthesis of Highly Fluorescent Gold Nanoparticles for Sensing Mercury(II). Angew. Chem., Int. Ed. 2007, 46 (36), 6824–6828. (11) Huang, C.-C.; Chang, H.-T. Selective gold-nanoparticle-based “Turn-On” fluorescent sensors for detection of mercury(II) in aqueous solution. Anal. Chem. 2006, 78 (24), 8332–8338. (12) He, S.; Li, D.; Zhu, C.; Song, S.; Wang, L.; Long, Y.; Fan, C. Design of a gold nanoprobe for rapid and portable mercury detection with the naked eye. Chem. Commun. 2008, 40, 4885–4887. (13) Li, T.; Dong, S.; Wang, E. Label-free colorimetric detection of aqueous mercury ion (Hg2þ) using Hg2þ-modulated G-quadruplexbased DNAzymes. Anal. Chem. 2009, 81 (6), 2144–2149. (14) Wang, H.; Wang, Y.; Jin, J.; Yang, R. Gold nanoparticle based colorimetric and “Turn-On” fluorescent probe for mercury(II) ions in aqueous solution. Anal. Chem. 2008, 80 (23), 9021–9028. (15) Wang, Y.; Yang, F.; Yang, X. Colorimetric biosensing of mercury(II) ion using unmodified gold nanoparticle probes and thrombin-binding aptamer. Biosens. Bioelectron. 2010, 25 (8), 1994–1998. (16) Wang, L.; Li, T.; Du, Y.; Chen, C.; Li, B.; Zhou, M.; Dong, S. Au NPs-enhanced surface plasmon resonance for sensitive detection of mercury(II) ions. Biosens. Bioelectron. 2010, 25 (12), 2622–2626. (17) Darbha, G. K.; Singh, A. K.; Rai, U. S.; Yu, E.; Yu, H.; Ray, P. C. Selective detection of mercury (II) ion using nonlinear optical properties of gold nanoparticles. J. Am. Chem. Soc. 2008, 130 (25), 8038–8043. (18) Xie, J.; Zheng, Y.; Ying, J. Y. Highly selective and ultrasensitive detection of Hg2þ based on fluorescence quenching of Au nanoclusters by Hg2þ-Auþ interactions. Chem. Commun. 2010, 46 (6), 961–963. (19) Lin, C.-Y.; Yu, C.-J.; Lin, Y.-H.; Tseng, W.-L. Colorimetric sensing of silver(I) and mercury(II) ions based on an assembly of tween 20 stabilized gold nanoparticles. Anal. Chem. 2010, 82 (16), 6830–6837. (20) Lin, Z.-H.; Yang, Z.; Chang, H.-T. Preparation of fluorescent tellurium nanowires at room temperature. Cryst. Growth Des. 2008, 8 (1), 351–357. (21) Zheng, A.; Chen, J.; Wu, G.; Wei, H.; He, C.; Kai, X.; Wu, G.; Chen, Y. Optimization of a sensitive method for the “switch-on” determination of mercury(II) in waters using Rhodamine B capped gold nanoparticles as a fluorescence sensor. Microchim Acta 2009, 164 (1-2), 17–27. (22) Brewer, S. H.; Glomm, W. R.; Johnson, M. C.; Knag, M. K.; Franzen, S. Probing BSA binding to citrate-coated gold nanoparticles and surfaces. Langmuir 2005, 21 (20), 9303–9307. (23) Ghosh, S. K.; Pal, T. Interparticle coupling effect on the surface plasmon resonance of gold nanoparticles: From theory to applications. Chem. Rev. 2007, 107 (11), 4797–4862. (24) Katz, M. J.; Sakai, K.; Leznoff, D. B. The use of aurophilic and other metal-metal interactions as crystal engineering design elements to increase structural dimensionality. Chem. Soc. Rev. 2008, 37 (9), 1884–1895. (25) Sillen, L. G.; Martell, A. E. Stability Constants of Metal-Ion Complexes, Supplement No. 1; The Chemical Society: London, 1971. (26) Morel, F. M. M.; Hering, J. G. Principles and applications of Aquatic Chemistry; Wiley: New York, 1993. (27) Kim, Y.; Johnson, R. C.; Hupp, J. T. Gold nanoparticle-based sensing of “Spectroscopically Silent” heavy metal ions. Nano Lett. 2001, 1 (4), 165–167.

1539

dx.doi.org/10.1021/es103369d |Environ. Sci. Technol. 2011, 45, 1534–1539