Asymmetric Flow-Field Flow Fractionation Hyphenated ICP-MS as an

Jun 20, 2015 - Asymmetric Flow-Field Flow Fractionation Hyphenated ICP-MS as an Alternative to Cloud Point Extraction for Quantification of Silver Nan...
4 downloads 3 Views 5MB Size
Download PDF
Article pubs.acs.org/ac

Asymmetric Flow-Field Flow Fractionation Hyphenated ICP-MS as an Alternative to Cloud Point Extraction for Quantification of Silver Nanoparticles and Silver Speciation: Application for Nanoparticles with a Protein Corona Thilak K. Mudalige,* Haiou Qu, and Sean W. Linder* Office of Regulatory Affairs, Arkansas Regional Laboratory, U.S. Food and Drug Administration, 3900 NCTR Road, Jefferson, Arkansas 72079, United States ABSTRACT: Production and application of nanoparticles in consumer products is at an all-time high due to the emerging field of nanotechnology. Direct detection and quantification of trace levels of nanoparticles within consumer products is very challenging and problematic. Although multiple methodologies are available for this purpose, each method has its own set of limitations. Herein, we developed an analytical platform consisting of asymmetric flow-field flow fractionation (AF4) coupled with inductively coupled plasma mass spectroscopy (ICP-MS) for the speciation and quantification of silver ions and silver nanoparticles at the ng/kg level (ppt). AF4 is utilized to concentrate the nanoparticles, and ICP-MS acts as the detector. The protein corona that forms upon exposure of nanoparticles to bovine serum albumin was utilized as a nanoparticle stabilization and AF4 recovery enhancement mechanism. Speciation of silver ions and nanoparticles was achieved with the assistance of penicillamine as a complexation ligand. The effect of nanoparticle size, surface coating, and ionization state toward the detection and quantification of the developed methodology was evaluated. The detection limit was found to be 4 ng/kg with the application of a 5 mL sample loop. Further application of this developed methodology on environmentally relevant samples was demonstrated by the analysis of Arkansas River water spiked with silver nanoparticles and nanoparticle spiked into humic acid solution (50 mg/L) at an environmentally relevant level. including cloud point extraction, asymmetric flow-field flow fractionation, hydrodynamic chromatography, single-particle ICP-MS, and capillary electrophoresis;11−19 however, low detection limits and subordinate extraction efficiencies hinder broad spectrum application of these techniques for samples having metallic nanoparticles available at single-digit ng/kg levels. Silver speciation is another challenge, and methodologies are available for samples having detectable silver concentrations.20,21 The quantification of nanoparticles at single-digit ng/kg levels is generally carried out with the help of an enrichment step prior to the analysis.18 Cloud point extraction is the prominent methodology for selective enrichment of nanoparticles. In this method, Triton X-114 is used as a surfactant, forming a hydrophobic cloud phase at around 40 °C that the nanoparticles are partitioned into.11,18 For quantification, the Triton X-114-rich cloud phase containing the nanoparticles is separated by mild centrifugation and undergoes an acid digestion followed by elemental analysis using ICP-MS or atomic emission/absorbance spectroscopy.11 Even though

ue to advancements within the emerging field of nanotechnology, the application of metallic nanoparticles in consumer products is rapidly increasing.1,2 Out of all metallic nanoparticles, silver nanoparticles (AgNPs) are the most utilized, as a result of their antimicrobial properties.3,4 There are over 438 products available on the open market that contain AgNPs including dietary supplements, food packing materials, medical products (catheters and wound dressing), and surface coatings of washing machines and refrigerators.5,6 The presence of AgNPs in these consumer products increases the risk of human exposure. The toxic effects attributed to silver nanoparticles have been reported in the literature for multiple cell lines, including human and mouse.7 Furthermore, the Environmental Protection Agency has conditionally approved the use of AgNPs as a pesticide.8 As the use of AgNPs in consumer products becomes more widespread, the potential for environmental contamination will reach an all-time high.9 Consequently, the development of methods to detect those nanoparticles is a necessity for regulatory agencies like the U.S. Food and Drug Administration (FDA).10 Direct detection and quantification of AgNPs in ultralow concentrations is a challenge even for modern techniques, such as inductively coupled plasma mass spectrometry (ICP-MS). Multiple methodologies are available for the isolation and detection of nanoparticles in environmental and consumer product samples,

D

This article not subject to U.S. Copyright. Published XXXX by the American Chemical Society

Received: April 27, 2015 Accepted: June 20, 2015

A

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

to the ICP-MS detector for quantification (Figure 1b). Herein we report the development of a methodology to enrich, quantify, and speciate ultratrace levels of silver nanoparticles across multiple particle sizes and coatings, including BSAcoated silver nanoparticles, which were unable to be enriched using cloud point extraction. Furthermore, we report the application of this method to detect and quantify AgNPs in environmentally relevant samples such as Arkansas River water. The feasibility of silver speciation, which is the ability to separate and quantitate silver ions from metallic nanoparticles, was also investigated and confirmed.

cloud point extraction is an excellent and well-documented methodology for ultratrace quantification of nanoparticles, its extraction efficiency is significantly reduced when encountering hydrophilic protein-coated nanoparticles.11 Upon exposure to protein solutions, nanoparticles become coated with a layer of protein, called the protein corona. It is well-documented that the formation of a protein corona makes the nanoparticle more hydrophilic, preventing its extraction into the relatively hydrophobic cloud phase.11 The reported extraction efficiency for bovine serum albumin (BSA)-coated silver nanoparticles is in the single-digit percentage range.11 However, percent recovery depends on the composition of protein in the corona, and Yu et al. have reported over 67% recovery of silver nanoparticles exposed to HepG2 cells.22 In the case of silver nanoparticles, the protein corona stabilizes the nanoparticle under harsh conditions and protects it from oxidation, ligand exchange, and even acid dissolution at certain acidity levels.23,24 In this study, we utilize the remarkable stability of nanoparticles coated with a BSA corona to develop an enrichment and quantitation methodology. We coupled asymmetric flow-field flow fractionation (AF4) with ICP-MS, applying AF4 as preconcentration step instead of a size-based separation technique. The mechanism and instrumentation of AF4 has been thoroughly discussed in the literature.17,25−28 In short, AF4 uses a narrow ribbon-shaped channel, in which one wall of the channel is lined with a porous membrane. A typical AF4 analysis consists of two steps, focusing and separation. In the focusing step, the sample is injected into the channel using an injection port, while solvent is simultaneously being pumped into the channel from both ends. During this process, the solvent and any ionic species pass through the membrane to the waste stream and all particulate matter is retained on the membrane (Figure 1a). In this study, we injected a large volume of sample over a longer period of time, extending the focusing step to remove any dissolved electrolytes. Particulate matter from the sample is concentrated in an area of no net liquid flow called the focusing zone. Upon completion of the focusing step, collected particles are passed through the channel

■

EXPERIMENTAL SECTION Materials and Reagents. Type I ultrapure water (18 MΩ· cm), obtained from a Thermo Scientific Barnstead Nanopure System (Waltham, MA), was utilized for all solution preparations. Nitric (67−69%) and hydrochloric acid (34− 37%) (optima grade), BSA lyophilized powder, and ethylenediaminetetraacetic acid (EDTA) solution (500 mM, pH 8.0) were purchased from Fisher Scientific (Houston, TX). Silver (1000 mg kg−1) single-element ICP-MS standard solutions were acquired from Spex CertiPrep (Metuchen, NJ) and Ultra Scientific (Kingstown, RI). An indium (100 mg kg−1) single-element ICP-MS standard solution was purchased from Inorganic Ventures (Christiansburg, VA). Isopropyl alcohol (99.999%, electronic grade) was purchased from Sigma-Aldrich (St. Louis, MO). Humic acid was purchased from Alfa Aesar (Wardhill, MA). Sodium dodecyl sulfate (SDS) and D(−)-penicillamine (99%) were purchased from Acros Organics (New Jersey). AgNPs with nominal diameters of 10, 30, and 60 nm with citrate coating and polyvinylpyrrolidone (PVP, 40 kDa) coating were purchased from Nanocomposix Inc. (San Diego, CA). The nanoparticle solutions were received in a chilled condition between 2 and 4 °C and stored at 4 °C under dark conditions to prevent oxidation. Millipore precut regenerated cellulose membranes (30 kDa) were purchased from Wyatt Technology (Santa Barbara, CA). Carbon-coated copper grids (300 mesh) were purchased from Electron Microscopy Sciences (Hatfield, PA). Disposable folded capillary cells were purchased from Malvern Instruments (Worcestershire, UK) and used for measuring the ζ-potential of particles. Functionalization and Characterization Methods. Silver Nanoparticle Functionalization. BSA-coated AgNP suspensions were prepared by mixing citrate- or PVP-stabilized AgNPs with the required amount of BSA to keep the BSA concentration constant at 0.2 mg/mL. After the BSA was added, the solutions were mixed at room temperature for 1 h to assist in protein corona formation. BSA-coated AgNPs were used directly for AF4 analysis. Determination of Particle Size and ζ-Potential. AgNPs were studied on a JEOL 1400 transmission electron microscope (Peabody, MA) operated at 80 kV. Ten microliters of the nanoparticle suspension was placed directly on a 300 mesh copper grid and allowed to rest for 5 min. The excess liquid was removed with filter paper followed by air drying overnight. Protein-coated samples were separated from excess protein by centrifugation, followed by suspension in water. Ten microliters of the protein-coated nanoparticle suspension was placed directly on a 300 mesh copper grid and allowed to rest for 5 min. The excess liquid was removed with filter paper, and the sample was stained with a freshly prepared aqueous solution of 1% uranyl acetate. The micrographs were acquired using a TVIPS TemCam F416 camera (Tiete Video and Image

Figure 1. Schematic illustration of preconcentration (a) and elution (b) of nanoparticles using AF4 channel assembly. B

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Processing Systems GmbH, Gauting, Germany). Image-J (National Institutes of Health) was used in the processing of the images to obtain size statistics of the AgNPs. The ζ-potential measurements were performed using a Malvern Instruments (Worcestershire, UK) Zetasizer Nano ZS DLS system. For ζ-potential measurements, the functionalized particles were isolated by centrifugation at 10 000g for 10 min and subsequently suspended in 0.5 mM sodium citrate at a pH of 7.5. Measurements were performed in triplicate using a disposable folded capillary cell from Malvern. Determination of Particle Concentration by ICP-MS. The quantification of AgNPs was carried out by ICP-MS. Briefly, a calculated volume of the nanoparticle suspension was added to microwave digestion vessels and digested with 2 mL of nitric acid and 1 mL of hydrochloric acid using a CEM (Matthews, NC) MARS 5 microwave-accelerated reaction system (CEM MARSXpress Teflon vessels; CEM MARSXpress vessel capping station). The samples were subsequently analyzed by an Agilent Technologies (Santa Clara, CA) 7700x ICP-MS (micromist nebulizer; autosampler ASX-500; MassHunter workstation software version A.01.02) under no collision cell gas mode. A 6-point calibration curve with indium as an internal standard was used for analysis. A second silver standard from a different source was used for the independent calibration verification. Nanoparticle Enrichment by AF4. A Wyatt Technology (Santa Barbara, CA) Eclipse 4 AF4 system composed of a flow control unit and channel compartment (short channel with 145 mm length and 350 μm spacer) was coupled to an Agilent Technologies (Santa Clara, CA) 1200 high-performance liquid chromatography (HPLC) system, which contains a quaternary pump (G1311B) with built-in vacuum degasser, autosampler (G1329B), 2 position/6 port valve (G1158A), and a 5.0 mL, 1.6 mm i.d. PEEK sample loop. A PVDF (polyvinylidene fluoride) hydrophilic 0.1 μm membrane filter (EMD Millipore, Billerica, MA) was used immediately after the pump. The pump was used to control liquid flow into the channel compartment; a 2 position/6 port valve was used for sample introduction, and an autosampler was only used to generate an injection signal for the ICP-MS. Precut 30 kDa MC membranes were soaked overnight in 20% isopropyl alcohol, washed with type I ultrapure water, and equilibrated with the carrier fluid inside the AF4 channel for 1 h. Five milliliters of sample was injected during each analysis using the conditions listed in Table 1. The ICP-MS signal was

Table 2. ICP-MS Operating Parameters RF power sample depth plasma gas carrier gas dilution gas reaction gas monitored isotope (m/z) dwell time peristaltic pump

contaminants are being leached from the HPLC pump piston heads. As further verification, we introduced the same solvent using a peristaltic pump and did not observe any zirconium oxide interference. The internal standard flow rate was kept constant at 0.2 mL/min. A SeaSpray nebulizer (Glass Expansion, West Melbourne, Australia) having a flow capacity up to 2 mL/min was used for sample introduction.

■

RESULTS AND DISCUSSION Particle Characterization. In this study, we used fresh AgNPs stored under controlled conditions to minimize ionization of the silver nanoparticles in solution. The absence of measurable ionic silver in the nanoparticle solution was verified by centrifugation and subsequent analysis of the supernatant for the silver. To verify BSA functionalization, ζpotential of the nanoparticle solution was measured with a Malvern Nano ZS, as shown in Table 3. A nanoparticle’s ζpotential depends on multiple factors including the type and concentration of the electrolyte in solution as well as the solution’s pH. We used 0.5 mM sodium citrate at a pH of 7.5 for the ζ-potential measurements. To minimize any interference due to unbound BSA, excess BSA was removed from the nanoparticle solution by centrifugation at 10000g for 10 min. This purification step was carried out twice, each time suspending the pellet in 0.5 mM citrate buffer. The ζ-potential of citrate-coated nanoparticles decreased with BSA coating, indicating a change in surface functionality.30 In the case of PVP-coated nanoparticles, the ζ-potential increased toward a more negative value upon BSA coating. In both cases, the resulting ζ-potential values for BSA-coated nanoparticles were found to be similar regardless of the original particle stabilizer, indicating the ability of BSA to replace the original coating. These results are comparable with recent literature values using identical measurement conditions.30 The ζ-potential of the 10 nm particles was not measured due to technical limitations of the instrumentation. Furthermore, the nanoparticle size distribution (metallic core) was measured by TEM, and the results from these analyses are included in Table 3. There was no change in the size of the silver core upon BSA coating. A representative TEM micrograph of 30 nm AgNPs after BSA coating is shown in Figure 2. The protein corona is clearly visible due to negative staining with uranyl acetate. Method Optimization. Here we developed a robust methodology for both the detection of silver nanoparticles in dilute solutions and the speciation of soluble ionic silver in solution using AF4 as an enrichment mechanism and ICP-MS as a detector. There were multiple technical challenges during the method development, including poor nanoparticle recoveries due to membrane fouling during AF4 nanoparticle enrichment. To enhance the recovery of silver nanoparticles, we applied a carrier fluid containing 0.01% SDS and 2 mM

Table 1. Liquid Flow Program for Nanoparticle Analysis step

designation

1 2 3

elution (prefocus) focus focus + injection

4 5

focus elution

1550 W 8.4 mm 15.0 L min−1 1.21 L min−1 0.15 L min−1 H2, 2.6 mL/min 107 (Ag+), 115 (In+) 1.0 s (107Ag), 0.5 s (115In) 0.3 rps

time (min), flow (mL/min) 0.0−2.0, channel flow 1.0, cross flow 0.0 2.0−3.0, channel flow 1.0, focus flow 0.6 3.0−20, channel flow 1.0, focus flow 0.6, injection flow 0.4 20−30, channel flow 1.0, focus flow 0.6 30−50, channel flow 1.0, cross flow 0.0

collected with a 1 second dwell time for 107Ag and 0.5 second for 115In (Table 2). A 3% isopropyl alcohol solution containing 20 μg/kg indium as an internal standard and 2 mM EDTA as a stabilizer was introduced into the sample before detection by ICP-MS using an Agilent isocratic pump (G1310A) and a Tconnector. Polyatomic interference of zirconium oxide on silver was observed and eliminated with the application of hydrogen as a reaction cell gas.29 We speculate that the zirconium C

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

Table 3. Nanoparticle Characterization nanoparticle size (TEM) nm ζ-potential (mV), before BSA coatinga ζ-potential (mV), after BSA coatinga a

PVP 60 58.8 (±7.1) −6.50 (±0.3) −24.0 (±0.3)

citrate 60 60.0 (±7.5) −45.4 (±0.9) −20.2 (±0.7)

PVP 30 27.1 (±5.3) −9.47 (±0.3) −24.5 (±2.0)

citrate 30 31.7 (±4.0) −39.2 (±1.0) −19.3 (±0.5)

PVP 10 9.8 (±1.9)

citrate 10 9.5 (±2.3)

n = 3.

EDTA at a pH of 8.5 with an extremely low focusing flow (0.6 mL/min). The carrier fluid improved the recovery for PVPcoated particles but not for citrate-coated particles. Citratecoated particles showed a run-to-run increase in recovery, indicating membrane fouling and saturation. ICP-MS results indicate that all of the ionic silver passed through the membrane as an EDTA complex. As an alternative methodology, we selected BSA as a particle coating due to the remarkable stability of BSA-coated AgNPs toward salt-induced aggregation and dissolution even at elevated salt concentrations and acidic conditions.23,24 It has been reported that BSA-coated AgNPs are stable for months under ambient conditions.23 In the case of BSA-coated AgNPs, particle recovery was consistent, but EDTA was unable to eliminate ionic silver from the sample. It has also been reported that silver ions can bind to BSA, likely to thiol-bearing cysteine residues on the surface of the protein.31,32 The binding of free silver ions to the BSA provides a mechanism for enrichment and for detection of ionic silver using the same asymmetric flow-field flow fractionation methodology. For the speciation of silver, penicillamine, a stronger chelating agent was applied at a concentration of 0.1 mM directly to the sample just before injection. The penicillamine was able to extract the silver ions from the BSA

Figure 2. Transmission electron micrograph of BSA-coated 30 nm silver nanoparticles.

Figure 3. Schematic illustration of detection and speciation: (1) silver nanoparticles in the presence of BSA; (2) silver ions in the presence of BSA; (3) silver ions in the presence of BSA and penicillamine ligand (4); silver ions and nanoparticles in the presence of BSA; (5) silver ions and nanoparticles in the presence of BSA and penicillamine. D

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

coating, forming a penicillamine−silver complex, which was able to pass through the AF4 membrane. The channel volume was calculated to be 382 μL, assuming a 145 μm reduction of channel height due to membrane swelling and compression. During the 17 min injection time, 5.0 mL of sample and 5.2 mL of carrier fluid flowed into the channel, passing through the membrane to the waste stream, resulting in a replacement of the channel fluid a total of 26.7 times. During the additional 10 min focusing step, another 6.0 mL of carrier fluid passed through the membrane for an additional 15.7 times replacement of the fluid in the channel. Combined, the injection and focusing steps result in the replacement of the fluid in the AF4 channel more than 42 times, providing adequate liquid replacement for silver speciation. Upon optimization of particle functionalization and selection of chelating agents, we focused on ICP-MS as a detection technique due to signal stability and reproducibility. The coelution of excess BSA with the sample can cause an increase in viscosity, decrease in surface tension, and carbon enhancement of the ICP-MS signal. To eliminate those shortcomings, an internal standard consisting of 20 μg/kg indium containing 2 mM EDTA and 3% isopropyl alcohol was introduced using a Tconnecter just before sample introduction to the ICP-MS. The flow rate from the HPLC pump used for the internal standard introduction was kept at a constant flow of 0.2 mL/min. The addition of isopropyl alcohol was used to normalize carbon enhancement effects and negate the effects on nebulization due to the variation of surface tension and viscosity. The detection, quantitation, and speciation of nanoparticles and silver ions are schematically represented in Figure 3. This figure clearly indicates the pass-through of silver ions to waste in the presence of penicillamine and the detection of a protein-bound silver ion by ICP-MS in the absence of a chelating agent. Effect of AgNP Size and Ionization State on ICP-MS Detection. To evaluate the effect of particle size and ionization state on ICP-MS sensitivity, we analyzed all nanoparticles and silver ions using identical BSA concentrations using direct introduction of the nanoparticles to the ICP-MS. All samples were introduced using the peristaltic pump attached to ICPMS, and an indium internal standard was introduced using a HPLC pump. The silver concentration of each sample was kept at 80 μg/kg. Each analysis was performed in triplicate without any chelating agent to keep ionic silver bound to the BSA. Results from this study are shown in Figure 4 as a percentage of citrate-coated 60 nm silver particles. These results indicate that the effect of particle size and ionization state on the sensitivity of ICP-MS detection is negligible. They also show that the signal is at an acceptable level, having a recovery range of 92− 104%. Effect of Nanoparticle Size, Original Coating, and Matrix on Recovery of Nanoparticles. To determine the concentration of an unknown sample, 60 nm citrate-coated AgNPs were functionalized with BSA and used to generate a calibration curve. Fractograms collected with ICP-MS for each concentration are shown in Figure 5a, and the resulting calibration curve is plotted in Figure 5b. The linearity of the calibration curve was explored up to 1000.0 ng/kg and shown to have a correlation of 0.998. The detection limit was measured using 3σ criteria by making seven injections of 30 nm citrate-coated AgNPs with a concentration of 50 ng/kg and found to be 4 ng/kg for this method with a run-to-run variation of 4.5%. Particle size and coating effects on nanoparticle recovery were evaluated in triplicate. As indicated in Figure 6,

Figure 4. ICP-MS signal of 80 μg/kg suspension of citrate-coated 60, 30, and 10 nm particles and PVP-coated 60, 30, and 10 nm particles, where the original coating was replaced with BSA.

Figure 5. (a) Fractograms of calibration standards of BSA-coated 60 nm particles and (b) calibration curve using BSA-coated 60 nm particles.

the particle recovery for each size range was found to be at a satisfactory level (recovery range of 85−105%), proving the applicability of the developed methodology for the analysis of nanoparticles of varying sizes from 10 to 60 nm. Relatively lower recoveries were observed with both 10 nm citrate- and PVP-stabilized particles. We can speculate this observation is the effect of nanoparticle size on protein corona formation. The pore size of 30 kDa molecular weight cutoff filters is about 2 E

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

ionic silver. Ionic silver spikes were analyzed in triplicate without the application of penicillamine, and recoveries were found to be 100.2%, proving the applicability of enrichment and detection of ionic silver. For the speciation of silver, 30 nm AgNPs (300 ng/kg) were spiked with 3000 ng/kg ionic silver and analyzed in triplicate. This analysis indicated particle recoveries of 89% after elimination of ionic silver by penicillamine. To confirm the total elimination of ionic silver in the presence of penicillamine, only the ionic silver spike (3000 ng/kg) was analyzed in triplicate, and we did not detect any silver ions, confirming complete pass through to the waste stream. In this case, the silver recovery was undetectable, which indicated the method’s ability to perform speciation using penicillamine.

■

CONCLUSION Here we have developed a robust enrichment and quantification methodology for ultratrace level analysis of AgNPs using a hyphenated AF4/ICP-MS technique with detection limits of 4.0 ng/kg with respect to metal concentration. Detection limit was calculated by analysis of seven replicates at 50 ng/kg level and the 3σ method. In previous publications, nanoparticles having a protein corona were unable to be extracted using cloud point extraction; however, our methodology can enrich and detect such particles without further processing. Protein corona formation was utilized for the stabilization of nanoparticles and confirmed with electron microscopy and ζ-potential determination. Applicability for the detection of ionic silver and silver speciation was confirmed. Environmentally relevant samples, AgNPs spiked to Arkansas River water and 50 mg/L humic acid solution, were analyzed and shown to provide satisfactory particle recovery. The developed methodology provides a new analytical route for the quantification and speciation of silver in consumer products and environmentally relevant samples. Experiments to extend this method to other types of nanoparticles are underway.

Figure 6. Percentile recovery of silver nanoparticle and ions with respect to particle coating and size.

nm; thus a small metal culture might pass through the membrane, and application of the membrane with lower molecular weight cutoff may be necessary in the case of ultrasmall nanoparticles. The application of this method for the analysis of environmental samples was tested with a triplicate analysis of 30 nm citrate-stabilized AgNPs spiked into Arkansas River water at a final silver concentration of 300 ng/kg. The particle recovery was found to be 95% of the original concentration. Silver nanoparticles were also spiked into a 50 mg/L humic acid solution at final silver concentration of 300 ng/kg and analyzed to evaluate the effects of organic matter on particle recovery. In both cases, recoveries were found to be 99%, proving the applicability of the developed methodology for environmental samples. Focusing and concentration of humic acid was clearly visible in the channel at the injection and focus (Figure 7), but the nanoparticles were not concentrated

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. Phone +1-870-5434665. Fax: +1-870-543-4041. *E-mail: [email protected]. Phone: +1-870-543-4667. Fax: +1-870-543-4041. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS These studies were conducted using the Nanotechnology Core Facility (NanoCore) located on the U.S. Food and Drug Administration’s Jefferson Laboratories campus (Jefferson, AR), which houses the FDA National Center for Toxicological Research and the FDA Office of Regulatory Affairs Arkansas Regional Laboratory. We thank Dr. Patrick Sisco, Dr. Jin-Hee Lim, Dr. Yasith Nanayakkara, and Dr. Marilyn Khanna for their support and valuable comments on the draft manuscript. The views expressed in this document are those of the researchers and should not be interpreted as the official opinion or policy of the U.S. Food and Drug Administration, Department of Health and Human Services, or any other agency or component of the U.S. government. The mention of trade names, commercial products, or organizations is for clarification of

Figure 7. Image of AF4 channel assembly showing focusing zone with enrichment of humic acid and nanoparticles at the end of the focusing step (top) and total elution of concentrated materials at the end of the run (bottom).

enough to visualize with the naked eye. Membrane fouling was not observed for 30 consecutive runs of a solution containing 50 mg/L humic acid and 0.2 mg/mL BSA. Application of very low cross-flow rate may help to prevent the membrane fouling. Arkansas River water was analyzed with and without penicillamine to determine the natural silver concentration of the river water, which was found to be 24.3 (±1.8) ng/kg of F

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX

Analytical Chemistry

Article

the methods used and should not be interpreted as an endorsement of a product or a manufacturer.

■

REFERENCES

(1) Aitken, R. J.; Chaudhry, M. Q.; Boxall, A. B. A.; Hull, M. Occup. Med. 2006, 56, 300−306. (2) Kessler, R. Environ. Health Perspect. 2011, 119, A120−A125. (3) Sondi, I.; Salopek-Sondi, B. J. Colloid Interface Sci. 2004, 275, 177−182. (4) Rai, M.; Yadav, A.; Gade, A. Biotechnol. Adv. 2009, 27, 76−83. (5) Project on Emerging Nanotechnologies, C.P.I.; http://www. nanotechproject.org/cpi (accessed April 27, 2015). (6) Chen, X.; Schluesener, H. J. Toxicol. Lett. 2008, 176, 1−12. (7) Ahamed, M.; AlSalhi, M. S.; Siddiqui, M. K. J. Clin. Chim. Acta 2010, 411, 1841−1848. (8) Environmental Protection Agency. EPA-HQ-OPP-2009-1012, 2009. (9) Nowack, B.; Bucheli, T. D. Environ. Pollut. 2007, 150, 5−22. (10) Hamburg, M. A. Science 2012, 336, 299−300. (11) Hartmann, G.; Baumgartner, T.; Schuster, M. Anal. Chem. 2014, 86, 790−796. (12) Poda, A. R.; Bednar, A. J.; Kennedy, A. J.; Harmon, A.; Hull, M.; Mitrano, D. M.; Ranville, J. F.; Steevens, J. J. Chromatogr. A 2011, 1218, 4219−4225. (13) Liu, J.-f.; Chao, J.-b.; Liu, R.; Tan, Z.-q.; Yin, Y.-g.; Wu, Y.; Jiang, G.-b. Anal. Chem. 2009, 81, 6496−6502. (14) Tiede, K.; Boxall, A. B. A.; Wang, X.; Gore, D.; Tiede, D.; Baxter, M.; David, H.; Tear, S. P.; Lewis, J. J. Anal. At. Spectrom. 2010, 25, 1149−1154. (15) Qu, H.; Mudalige, T. K.; Linder, S. W. Anal. Chem. 2014, 86, 11620−11627. (16) Mitrano, D. M.; Barber, A.; Bednar, A.; Westerhoff, P.; Higgins, C. P.; Ranville, J. F. J. Anal. At. Spectrom. 2012, 27, 1131−1142. (17) Gray, E. P.; Bruton, T. A.; Higgins, C. P.; Halden, R. U.; Westerhoff, P.; Ranville, J. F. J. Anal. At. Spectrom. 2012, 27, 1532− 1539. (18) Chao, J.-b.; Liu, J.-f.; Yu, S.-j.; Feng, Y.-d.; Tan, Z.-q.; Liu, R.; Yin, Y.-g. Anal. Chem. 2011, 83, 6875−6882. (19) Pettibone, J. M.; Gigault, J.; Hackley, V. A. ACS Nano 2013, 7, 2491−2499. (20) Soto-Alvaredo, J.; Montes-Bayón, M.; Bettmer, J. Anal. Chem. 2013, 85, 1316−1321. (21) Zhou, X.-X.; Liu, R.; Liu, J.-F. Environ. Sci. Technol. 2014, 48, 14516−14524. (22) Yu, S.-j.; Chao, J.-b.; Sun, J.; Yin, Y.-g.; Liu, J.-f.; Jiang, G.-b. Environ. Sci. Technol. 2013, 47, 3268−3274. (23) Martin, M. N.; Allen, A. J.; MacCuspie, R. I.; Hackley, V. A. Langmuir 2014, 30, 11442−11452. (24) Dominguez-Medina, S.; Blankenburg, J.; Olson, J.; Landes, C. F.; Link, S. ACS Sustainable Chem. Eng. 2013, 1, 833−842. (25) Tasci, T. O.; Johnson, W. P.; Fernandez, D. P.; Manangon, E.; Gale, B. K. Anal. Chem. 2013, 85, 11225−11232. (26) M-M, P.; Siripinyanond, A. J. Anal. At. Spectrom. 2014, 29, 1739−1752. (27) Giddings, J. C.; Caldwell, K. D. Anal. Chem. 1984, 56, 2093− 2099. (28) Giddings, J. C.; Yang, F. J.; Myers, M. N. Anal. Chem. 1976, 48, 1126−1132. (29) Guo, W.; Hu, S.; Zhang, J.; Zhang, H. Sci. Total Environ. 2011, 409, 2981−2986. (30) Brewer, S. H.; Glomm, W. R.; Johnson, M. C.; Knag, M. K.; Franzen, S. Langmuir 2005, 21, 9303−9307. (31) Lenz, G. R.; Martell, A. E. Biochemistry 1964, 3, 745−750. (32) Grade, S.; Eberhard, J.; Neumeister, A.; Wagener, P.; Winkel, A.; Stiesch, M.; Barcikowski, S. RSC Adv. 2012, 2, 7190−7196.

G

DOI: 10.1021/acs.analchem.5b01592 Anal. Chem. XXXX, XXX, XXX−XXX