Article pubs.acs.org/ac
Flow Injection Single Particle Inductively Coupled Plasma Mass Spectrometry: An Original Simple Approach for the Characterization of Metal-Based Nanoparticles Ram P. Lamsal, Gregory Jerkiewicz, and Diane Beauchemin* Queen’s University, Department of Chemistry, 90 Bader Lane, Kingston, Ontario K7L 3N6, Canada
ABSTRACT: In recent years, single-particle inductively coupled plasma mass spectrometry (spICPMS) has emerged as a reliable tool that can both count metal-containing nanoparticles and measure their mass, thereby allowing sizing if their shape, density, and composition are known. However, the methodology associated with the current spICPMS approach for mass determination requires determination of both the sample uptake rate and the sample introduction efficiency of the nebulization system. In this paper, the proof of concept of a novel approach based on flow injection (FI) analysis coupled to ICPMS, i.e., FIspICPMS, is presented. Unlike the established technique, this method does not require a determination of the transport efficiency and of the sample uptake rate for the accurate measurement of particle mass. It also only requires a measurement of the transport efficiency for determination of the particle number. Unlike the traditional spICPMS approach, the measurement of transport efficiency by FI-spICPMS is not affected by changes in sample uptake rate. The efficiency of FI-spICPMS is demonstrated through accurate determination of the particle number and size of 60 nm citrate-coated gold nanoparticles suspended in high-purity water. Despite being simpler, the method provides similar results to those obtained by the established spICPMS method. With a 5 ms dwell time and 200 μs settling time, the size detection limit is 20 nm, i.e., the same as with spICPMS.
W
chemical composition, crystal geometry, particle size and shape, surface chemistry, and agglomeration/aggregation state.1,5−7 Some studies have focused on assessment of the toxicological effects of ENMs in various biological media and environmental conditions.8,9 However, relatively few studies have attempted the quantitative analysis of ENMs at exposure concentrations of biological or environmental relevance.10 Indeed, characterizing and quantifying engineered NPs in environmental conditions remains a substantial challenge.5,11,12 This arises from the extremely low concentrations of engineered NPs13,14 in
ith the startling development of nanotechnology, there is rapidly growing interest in the production and use of engineered nanomaterials (ENMs) in diverse sectors, ranging from medicine, cosmetics, and food processing to energy production and storage as well as electronics and textiles.1 Despite its promises, the rapidly growing application of ENMs in commercial products raises serious concern about the potential environmental and toxicological impacts during the manufacture, use, and disposal of ENMs and related products.2 Appropriate risk assessment of ENMs in the natural environment and biological media necessitates reliable techniques for detecting, quantifying, and characterizing nanoparticles (NPs).3,4 In contrast to bulk materials, the fate, transport, and stability of ENMs are linked to their unusual physicochemical properties including specific surface area, © XXXX American Chemical Society
Received: July 12, 2016 Accepted: September 29, 2016
A
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry environmental matrices (typically on the order of ng L−1), the compromised method sensitivity in complex environmental matrix, and spectroscopic interferences originating from colloids.15 Consequently, an adequate assessment of the risks associated with nanotechnology requires the development and/ or improvement of NP metrology in environmental and biological systems.3 Inductively coupled plasma mass spectrometry (ICPMS) operating in time-resolved analysis mode is considered an emerging analytical approach for the detection and characterization of low concentrations of metal-based NPs.16−19 Because it can detect individual NPs, it is commonly called single particle ICPMS (spICPMS).12 In addition to allowing NP quantification, spICPMS can provide particle size distribution if the shape, density, and composition of particles are known and particle number concentration simultaneously with the proportion of dissolved metal and particulate species.20,21 Generally, introduction of a very dilute suspension of NPs into ICPMS is carried out so that there is at most one particle per droplet in the aerosol generated by the nebulizer and that a signal intensity spike is generated for each individual particle following vaporization, atomization, and finally ionization in the plasma with the packet of ions then generated being detected.22 If the particle number concentration in the suspension is sufficiently low, and a short measurement dwell time is used, each intensity spike represents at most a single particle event. The frequency and intensity of the pulses can then be used to derive the particle number concentration and particle mass (which can be converted into its size if its shape, density, and composition are known). The theoretical aspects for sizing and counting particles by spICPMS were first outlined by Degueldre et al.16−19 in a series of studies on the characterization of colloids in aqueous suspensions, which also demonstrated the applicability of spICPMS for screening inorganic colloids. Laborda et al.23 reviewed other prospective applications of spICPMS, further demonstrating its usefulness as a characterization tool for NPs. Calibration for NP size characterization can be carried out using aqueous standards of mostly spherical NPs having welldefined diameters. However, this approach is limited by the availability of monodisperse standard NPs of similar composition, shape, and size as those of the sample NPs.24 Pace et al.25 introduced an alternative calibration strategy for spICPMS analysis using continuous nebulization of standard solutions of dissolved metals to establish an external calibration curve. Unlike calibration with monodisperse droplets of standard solution, it eliminates the need for a dedicated droplet generator.26 However, this technique requires accurate knowledge of the transport efficiency and sample uptake rate for determination of the mass flux of analyte so that the NP mass can then be determined by relating the signal intensity of the spikes to the external calibration curve.25 To date, most spICPMS methods use a dissolved standard instead of monodisperse NPs for calibration despite the fact that this approach critically depends on an accurate measurement of transport efficiency and of the sample uptake rate.25 Transport efficiency is instrument-specific and can also depend on operating conditions and sample types. Regular monitoring is thus required for accurate particle number and mass determinations by spICPMS.25 The objective of this work is to present the proof-of-concept of a new simple approach for measuring the mass of NPs and counting them based on flow injection (FI) coupled to ICPMS,
i.e., FI-spICPMS. This method does not require determination of the transport efficiency or of the sample uptake rate for measurement of the mass, and thus size, of NPs. Only measurement of the transport efficiency is required for determination of the particle number. These reduced requirements inherently increase sample throughput. Toncelli et al.7 have used FI, but for a completely different purpose: to perform online dilution of seawater for monitoring Ag NPs in seawater and for the measurement of Ag NPs in marine microorganisms by spICPMS. For the first time, this work shows that, because FI readily allows calculation of the mass of injected analyte, an external calibration of intensity relative to injected mass can simply be constructed, which can then be used to find the mass of analyte in each spike (particle) of the sample without any need to measure the transport efficiency or sample uptake rate. The suitability of FI-spICPMS is demonstrated through a comparison to conventional spICPMS, where similar results were obtained for a standard material of Au NPs.
■
EXPERIMENTAL SECTION Chemicals. A monodisperse suspension of Au NPs in water (99.99% Au purity), with a particle number concentration of 2.3 × 1010 particles/mL and a particle diameter of 60.6 ± 5.9 nm, was purchased from nanoComposix (San Diego, CA, USA). These NPs, which are coated and stabilized with 2 mM citrate, have a nearly spherical shape. Their mass concentration in the suspension was 0.052 mg mL−1. Aqueous Au standard solutions in the 0.1−10 ng mL−1 range were prepared in 2% HCl from 1000 mg L−1 monoelement Au standard solution (SCP Science, Baie d’Urfé, Québec, Canada), doubly deionized water (DDW) (18.2 MΩ cm−1), and sub-boiled HCl (ACS grade; Fisher Scientific, Ottawa, Canada). All DDW was purified using an Arium Pro UV|DI water purification system (Sartorius Stedim Biotech, Göttingen, Germany). HCl was purified with a DST-1000 sub-boiling distillation system (Savillex, Minnetonka, USA). Instrumentation. The research was conducted on a Varian 820MS (Varian Inc., Australia) quadrupole-based ICPMS instrument equipped with Ni sampler and skimmer cones with 0.9 and 0.4 mm diameter orifices, respectively. The sample introduction system consisted of a MicroMist concentric nebulizer (Glass Expansion, Victoria, Australia) fitted into a Peltier-cooled Scott double-pass spray chamber (SCP Science, Quebec, Canada) maintained at 0 °C. Data acquisition was carried out in time-resolved analysis mode using the instrument’s Bruker Quantum software. The instrument was tuned daily (including torch alignment) using a solution containing 5 μg L−1 of Be, Mg, Co, In, Ce, Pb, and Ba in 2% (v/v) HNO3 for optimal sensitivity and minimum oxide and double-charged ion levels. The resulting ICPMS operating conditions are summarized in Table 1. spICPMS Analysis Procedure. The commercial Au NP suspension was diluted with DDW to the concentrations indicated in Table 2 in polypropylene vials on the day of the analysis to avoid NP dissolution. Immediately prior to analysis, the diluted suspensions were sonicated (Branson 5800, MI, USA) for 10 min to ensure full dispersion of the NPs. A few small blocks of ice were added to the sonication bath to prevent temperature increase during sonication. The sample uptake rate was checked every day in triplicate by weighing a vial containing DDW before and after 2 min of aspiration and was found relatively constant at 0.24−0.25 mL min−1. B
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
In this work, a flow injection variation was used where 50 μL of standard suspension was injected (in triplicate). As the injected volume is known, there is no requirement to measure the sample flow rate. Indeed, the number of NPs nebulized is then simply the known NP concentration multiplied by the injection volume. The ratio of the number of NPs detected over that injected then gives the transport efficiency. Data Processing for spICPMS. The raw intensity data were exported to Microsoft Excel 2013 for data processing. The signal intensity (in counts per s) of each measurement point was first multiplied by the dwell time. To discriminate particle events (i.e., the pulses) from the background signal,25 an iterative approach was selected18,30,31 using a five times the standard deviation (5σ) iterative algorithm criterion.10 The average intensity (μ) and standard deviation (σ) were calculated for the whole data set with all data points greater than (μ + 5σ) collected as NP events and removed from the data set. This process was repeated with a new (μ + 5σ) of the remaining data. NP events were collected and removed from the set until no more points could be identified as particle events; the remaining data set then corresponded to the background signal, unresolved particles, and the dissolved analyte fraction. The particle number concentration, Np (number of particles mL−1), was obtained from the pulse frequency of NP events f(IP) (number of pulses min−1) using16−19
Table 1. ICPMS Operating Conditions parameter
value
Ar plasma gas flow rate (L min−1) Ar auxiliary gas flow rate (L min−1) Ar sheath gas flow rate (L min−1) nebulizer gas flow rate (mL min−1) sample uptake ratea (mL min−1) sampling depth (mm) RF power (kW) dwell time (ms) settling time (μs) monitored signal total measurement time (s)
18 1.80 0.04 0.98 0.25 5.5 1.40 5 200 197 Au+ 100 (spICPMS); 250 (FI-spICPMS)
a
The sample uptake rate in spICPMS is the DDW carrier uptake rate in FI-spICPMS.
Table 2. Average Diameter ± Standard Deviation (n = 3) of Au NPs Suspended in Water at Different Mass Concentrations As Determined by spICPMS and FIspICPMS measured average diameter (nm)
a
mass concentration (ng L−1)
spICPMS
FI-spICPMS
50 200 500
61.82 ± 0.87 64.4 ± 4.8 67.2 ± 7.3
60.23 ± 0.60 64.0 ± 4.2 66.4 ± 5.6
reference valuea (nm) 60.6 ± 5.9
Np =
On the basis of transmission electron microscopy measurements.
f (IP) qliq × ηn
(1) −1
where qliq is the sample flow rate (in mL min ) and ηn is the transport efficiency. The particle size was determined by following a protocol developed by Pace et al.25 using external calibration with dissolved standards. For a given dwell time, tdt, and analyte concentration, C, the mass flux W was determined using
FI-spICPMS Analysis Procedure. An FI system with a 100 μL injection loop connected to a universal automatic actuator (Anachem Ltd., Luton, England) was coupled to the nebulizer of ICPMS. For quantification, a four-point calibration curve was constructed of blank-subtracted peak area relative to the absolute analyte content (calculated as 100 μL × concentration of standard injected in μg L−1) by injecting standard solutions into a DDW carrier. Linear regression analysis was performed using the data analysis function in Microsoft Office Excel 2013 to obtain the equation of the line of best fit, which was then used to convert the total peak area for the injected suspension into the total absolute mass of injected Au NPs. The mass of an individual NP was then obtained based on the fraction of the area due to that NP. The injection valve, sample loop, and capillary tubing connecting the valve to the nebulizer were cleaned by pumping 2% HNO3 through them between sample and standard injections to prevent carry-over of dissolved and particulate Au. Determination of Transport Efficiency. Two approaches were compared. A mass-based method was used to determine the transport efficiency in triplicate directly through capture of the aerosol escaping the spray chamber.27 A known mass of dry silica gel was loaded into a 1 mL micropipette tip (cut so as to match the inner diameter of the torch injector) and then attached to the outlet of the spray chamber.28 The ratio of the mass of 1 μg L−1 Au solution trapped to that aspirated gave the transport efficiency. The other approach is based on spICPMS theory that each NP entering the plasma produces a signal pulse. If a standard suspension of known NP number concentration is nebulized, the ratio of the number of NPs detected to the number of NPs nebulized for a given period of time at a known sample uptake rate is the transport efficiency.29
W = ηn × qliq × tdt × C
(2)
and used to construct a calibration line relating the average intensity (counts/event) for each Au standard solution to mass per event. This calibration curve was then used to determine the corresponding mass of each particle event, mp, using the equation mp =
[(IP − Ibdg) − b] m
(3)
where m and b are the slope and y-intercept of the calibration curve, Ibdg is the average background intensity, and IP is the pulse intensity. The mass was finally converted to particle size assuming a spherical geometry and full ionization of all NPs in the plasma using d=
3
⎡ 6 × mp ⎤ ⎥ ⎢ ⎣ ρ×π ⎦
(4)
where ρ is the density of the bulk metal. Finally, the particle sizes were binned to generate a size distribution histogram. Data Processing for FI-spICPMS. After converting the pulse intensity (in counts per s) to counts through multiplication by the dwell time, the particle events were then discriminated from the background as described in the previous section. The identified particle events were summed to get the total intensity produced by a known amount of NP suspension. C
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. Steps involved in (a) spICPMS and (b) FI-spICPMS.
solution). The equation of the line of best fit was then applied to the background-corrected peak area of the injected sample to find the total mass of injected Au. The mass of each particle, mp, was then calculated by multiplying the total mass (mT) by ⎡I ⎤ the signal fraction ⎣⎢ IP ⎦⎥ (i.e., signal from the particle divided by T
The total background intensity over the same length of time was also calculated and subtracted from the total intensity to remove any contribution from dissolved Au. The particle number concentration was simply obtained by dividing the number of NP events counted over the entire FI peak, n(IP), divided by the Vinjected sample injection volume (0.1 mL) by the transport efficiency Np =
n(IP) Vinjected × ηn
the total FI peak signal) ⎡I ⎤ mp = mT × ⎢ P ⎥ ⎣ IT ⎦
(5)
A calibration curve was established by plotting the background-corrected total intensity (peak area) as a function of the mass of Au injected (which is equal to the injection volume multiplied by the analyte concentration in the standard
(6)
Finally, the mass was converted into a size using the density of the bulk metal and assuming a spherical geometry. A particle size distribution was generated by counting the number of D
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 2. Time-resolved signal generated by a suspension of Au NPs (60 nm nominal diameter) using (a) established spICPMS and (b) FIspICPMS and the corresponding size distributions in (c) and (d), respectively.
Although there is no significant difference within individual pairs of results, scrutiny of Table 2 reveals that the particle size obtained by FI-spICPMS is systematically slightly smaller than that obtained by spICPMS. This likely arises from the additional small dilution that is inherent with FI, as dispersion of the sample plug occurs in the carrier on the way to the nebulizer, which may then further reduce the coincidence of particles. Indeed, Figure 3 shows that dispersion occurred, as
particles of each size. The steps involved in spICPMS and FIspICPMS are summarized in Figure 1.
■
RESULTS AND DISCUSSION spFI-ICPMS Relative to spICPMS for Particle Size Determination of Au NPs. Figure 2 shows the time-resolved profiles obtained by spICPMS and FI-spICPMS using the same ICPMS operating conditions for a 52 ng L−1 suspension of Au NPs along with the corresponding histograms of the particle size distribution. The mean particle size calculated using these size distributions is 61.82 ± 0.87 nm (n = 3) and 60.23 ± 0.60 nm (n = 3) without and with FI, respectively (Table 2). Both of these results are in good agreement with the reference value. However, no measurement of the sample uptake rate was required for the FI approach. Although Au NPs are certified to be 60 nm in diameter, in reality, they are almost spherical with (111) and (100) facets predominating on their faces, edges having (110) facets, and corners having all three types of facets. For further verifying that the FI approach did not affect the results, the suspension of standard Au NPs was analyzed at two other dilution levels. Again, very similar mean particle sizes were obtained by spICPMS and FI-spICPMS (Table 2). A similar precision was also achieved by the two methods. Irrespective of the method, precision is significantly higher at higher dilution level. As the mass concentration increases at constant dwell time, precision not only deteriorates, but the particle size is overestimated. This is a result of a greater frequency of particle coincidence, where two or more particles are erroneously counted as one particle, as the particle concentration increases without concurrently decreasing the dwell time. The recovery in number of particles indeed decreased to 70 and 50% at 200 and 500 ng L−1, respectively. For spICPMS and FI-spICPMS to provide accurate results, only one NP should be detected during a dwell time.23 In any case, whether the concentration and/or measurement conditions were selected appropriately or not, Table 2 shows that FI-spICPMS systematically gave similar results to those of spICPMS.
Figure 3. Time-resolved signal from a 100 μL injection of 1 μg L−1 Au standard solution using the operating conditions in Table 1.
the 100 μL aliquot of 1 μg L−1 Au standard solution took 24 s to reach the nebulizer and then 140 s to be measured, whereas in the absence of dispersion 24 s would have been sufficient to measure the whole 100 μL volume at 0.25 mL min−1. This significant dispersion likely also explains the apparent loss of NPs for several seconds at around 130 s in Figure 2b. Indeed, unless the suspension is very homogeneous, heterogeneity will be amplified by the dispersion process. This has no effect on the results however, as the whole flow injection peak is integrated. The smallest detectable NP size, which corresponds to the smallest pulse height that can be distinguished from the E
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry Table 3. Determination of Au NP Number Concentration (n = 3) method
transport efficiency (%)
nominal (particles L−1)
measured (particles L−1)
recovery (%)
sp-ICPMS FI-spICPMS
5.20 ± 0.02 5.55 ± 0.06
2.30 × 10 2.30 × 1010
(1.91 ± 0.02) × 1010 (2.09 ± 0.01) × 1010
83.5 ± 1.5 90.7 ± 5.0
10
■
background,20,32 was determined as three times the standard deviation of the blank count rate (3σ, n = 10) divided by sensitivity, i.e., the slope of the calibration curve established with standard solutions of 0.1−10 ng mL−1.20 The size detection limit for Au NPs in DDW was calculated to be 20 nm without or with FI. Again, similar results were obtained without the additional requirement of measuring the sample uptake rate. spFI-ICPMS Relative to spICPMS for Particle Number Determination. The particle number concentrations obtained using eq 1 for spICPMS and eq 5 for FI-spICPMS are compared in Table 3, which also includes the transport efficiency measured by the mass-based and new FI-spICPMS particle number-based methods. A good agreement with the nominal particle number concentration and good recoveries were obtained by both approaches. The roughly 7% higher number of particles obtained with FI-spICPMS than with spICPMS is commensurate with the 7% higher transport efficiency that was then measured with the particle number method. Any change in sample flow rate during transport efficiency measurements by the mass-based and the traditional spICPMS particle number-based methods would affect the resulting transport efficiency, hence affecting the particle number concentration. With FI-spICPMS, this source of discrepancy is eliminated, as a fixed sample loop volume is used, which does not require measurement of the sample flow rate as long as the whole flow injection peak is measured. Hence, the FI-spICPMS particle number approach for measuring transport efficiency is unaffected by variation in sample uptake rate.
■
AUTHOR INFORMATION
Corresponding Author
*Phone: +1 613 533 2619. Fax: +1 613 533 6669. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The authors gratefully acknowledge funding from the Natural Sciences and Engineering Research Council of Canada (Grants 39487-2013 and 477963-2015). R.P.L. thanks Queen’s School of Graduate Studies for graduate awards.
■
REFERENCES
(1) Nel, A.; Xia, T.; Mädler, L.; Li, N. Science 2006, 311, 622−627. (2) Wiesner, M. R.; Lowry, G. V.; Alvarez, P.; Dionysiou, D.; Biswas, P. Environ. Sci. Technol. 2006, 40 (14), 4336−4345. (3) Klaine, S. J.; Alvarez, P. J. J.; Batley, G. E.; Fernandes, T. F.; Handy, R. D.; Lyon, D. Y.; Mahendra, S.; McLaughlin, M. J.; Lead, J. R. Environ. Toxicol. Chem. 2008, 27 (9), 1825−1851. (4) Howard, A. G. J. Environ. Monit. 2010, 12 (1), 135−142. (5) MacCuspie, R. I.; Rogers, K.; Patra, M.; Suo, Z.; Allen, A. J.; Martin, M. N.; Hackley, V. J. Environ. Monit. 2011, 13 (5), 1212−1226. (6) Chinnapongse, S. L.; MacCuspie, R. I.; Hackley, V. A. Sci. Total Environ. 2011, 409 (12), 2443−2450. (7) Toncelli, C.; Mylona, K.; Tsapakis, M.; Pergantis, S. A. J. Anal. At. Spectrom. 2016, 31 (7), 3036−3049. (8) Maurer-Jones, M. A.; Gunsolus, I. L.; Murphy, C. J.; Haynes, C. L. Anal. Chem. 2013, 85 (6), 3036−3049. (9) Quigg, A.; Chin, W.; Chen, C.; Zhang, S.-J.; Jiang, Y.; Miao, A.; Schwehr, K. a; Xu, C.; Santschi, P. H. ACS Sustainable Chem. Eng. 2013, 1 (7), 686−702. (10) Tuoriniemi, J.; Cornelis, G.; Hassellöv, M. Anal. Chem. 2012, 84 (9), 3965−3972. (11) Pettibone, J. M.; Gigault, J.; Hackley, V. A. ACS Nano 2013, 7 (3), 2491−2499. (12) Liu, J.; Murphy, K. E.; MacCuspie, R. I.; Winchester, M. R. Anal. Chem. 2014, 86 (7), 3405−3414. (13) Gottschalk, F.; Sonderer, T.; Scholz, R. W.; Nowack, B. Environ. Sci. Technol. 2009, 43 (24), 9216−9222. (14) Mueller, N. C.; Nowack, B. Environ. Sci. Technol. 2008, 42 (12), 4447−4453. (15) Hassellöv, M.; Readman, J. W.; Ranville, J. F.; Tiede, K. Ecotoxicology 2008, 17 (5), 344−361. (16) Degueldre, C.; Favarger, P. Y. Colloids Surf., A 2003, 217 (1−3), 137−142. (17) Degueldre, C.; Favarger, P. Y.; Wold, S. Anal. Chim. Acta 2006, 555 (2), 263−268. (18) Degueldre, C.; Favarger, P. Y.; Bitea, C. Anal. Chim. Acta 2004, 518 (1−2), 137−142. (19) Degueldre, C.; Favarger, P.-Y. Talanta 2004, 62 (5), 1051− 1054. (20) Laborda, F.; Jimenez-Lamana, J.; Bolea, E.; Castillo, J. R. J. Anal. At. Spectrom. 2011, 26 (7), 1362−1371. (21) Mitrano, D. M.; Barber, A.; Bednar, A.; Westerhoff, P.; Higgins, C. P.; Ranville, J. F. J. Anal. At. Spectrom. 2012, 27 (7), 1131. (22) Laborda, F.; Bolea, E.; Jiménez-Lamana, J. Trends Environ. Anal. Chem. 2016, 9, 15−23. (23) Laborda, F.; Bolea, E.; Jiménez -Lamana, J. Anal. Chem. 2014, 86 (5), 2270−2278.
CONCLUSIONS
A novel analytical technique for determining the mass (and size, for a known geometry, density, and composition) of NPs was developed based on FI-spICPMS. It was demonstrated to provide a similar mean particle size for a standard material of Au NPs diluted in DDW without the need to measure the transport efficiency and the sample uptake rate, thereby eliminating two sources of error. Furthermore, as the analysis is carried out in a closed system, contamination is minimized. Moreover, the small volume injected (100 μL in this proof-ofconcept work, which could be further reduced) minimizes sample consumption and enables a higher throughput than possible with spICPMS, even more so as measurement of the sample uptake rate is eliminated. This technique is expected to be very beneficial for the analysis of NPs in complex environmental samples, as the discrete sample introduction followed by immediate rinsing by the carrier effectively minimizes memory effects. The only additional requirement compared to spICPMS is for investment in an inexpensive lowpressure sample injection valve. Although this work focused on Au NPs, the approach is applicable to any type of NPs containing metallic or metalloid elements. Future work will investigate in more detail the influence of the dwell time as well as that of the sample injection volume, manifold configuration, and so forth on the sizing of NPs. Its applicability will be tested for NPs of various sizes made of different elements. F
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry (24) Telgmann, L.; Metcalfe, C. D.; Hintelmann, H. J. Anal. At. Spectrom. 2014, 29, 1265−1272. (25) Pace, H. E.; Rogers, N. J.; Jarolimek, C.; Coleman, V. A.; Higgins, C. P.; Ranville, J. F. Anal. Chem. 2011, 83 (24), 9361−9369. (26) Lee, W.-W.; Chan, W.-T. J. Anal. At. Spectrom. 2015, 30, 1245− 1254. (27) Smith, D. D.; Browner, R. F. Anal. Chem. 1982, 54 (3), 533− 537. (28) Agatemor, C.; Beauchemin, D. Anal. Chim. Acta 2011, 706 (1), 66−83. (29) Pace, H. E.; Rogers, N. J.; Jarolimek, C.; Coleman, V. A.; Higgins, C. P.; Ranville, J. F. Anal. Chem. 2012, 84 (10), 4633. (30) Navratilova, J.; Praetorius, A.; Gondikas, A.; Fabienke, W.; von der Kammer, F.; Hofmann, T. Int. J. Environ. Res. Public Health 2015, 12 (12), 15756−15768. (31) Mitrano, D. M.; Lesher, E. K.; Bednar, A.; Monserud, J.; Higgins, C. P.; Ranville, J. F. Environ. Toxicol. Chem. 2012, 31 (1), 115−121. (32) Laborda, F.; Jiménez-Lamana, J.; Bolea, E.; Castillo, J. R. J. Anal. At. Spectrom. 2013, 28, 1220−1232.
G
DOI: 10.1021/acs.analchem.6b02656 Anal. Chem. XXXX, XXX, XXX−XXX
