Letter pubs.acs.org/ac
Determination of Size- and Number-Based Concentration of Silica Nanoparticles in a Complex Biological Matrix by Online Techniques Dorota Bartczak,† Phil Vincent,‡ and Heidi Goenaga-Infante*,† †
LGC Limited, Queens Road, Teddington, Middlesex TW11 0LY, United Kingdom NanoSight, a Malvern Instruments Limited, London Road, Amesbury, Wiltshire SP4 7RT, United Kingdom
‡
S Supporting Information *
ABSTRACT: We propose for the first time methodology for the determination of a number-based concentration of silica (SiO2) nanoparticles (NP) in biological serum using nanoparticle tracking analysis (NTA) as the online detector for asymmetric flow field-flow fractionation (AF4). The degree of selectivity offered by AF4 was found necessary to determine reliably number-based concentration of the measured NP in the complex matrix with a relative measurement error of 5.1% (as relative standard deviation, n = 3) and without chemical sample pretreatment. The simultaneous online coupling to other size and concentration detectors, such as multiangle light scattering (MALS) and ICPMS, for the measurement of the same NP suspension, was used to confirm the particle size determined with NTA and the equivalent particle number determined by AF4/NTA, respectively. The size- and number-based concentration data obtained by independent techniques were in a good agreement. The developed methodology can easily be extended to other types of particles or particle suspensions and other complex matrices provided that the particle size is above the limit of detection for NTA.
T
There are a few techniques, including light scattering-based systems such as nanoparticle tracking analysis (NTA)5 or electrical current-based platforms such as tunable resistive pulse sensing (TRPS),6 capable of counting nanoparticles directly in suspension. However, these techniques may suffer from matrixinduced interferences. An attractive way to overcome issues caused by the matrix, while achieving improved selectivity, is the use of online sample fractionation systems such as asymmetric flow field-flow fractionation (AF4), enabling gentle separation of nanoparticles from the matrix, without chemical sample pretreatment. To date, AF4 has been successfully coupled online to particle size detectors, such as multiangle light scattering (MALS),7 as well as to mass concentration detectors, such as ICPMS,8 and applied to the characterization of SiO2 NP, in terms of size and mass concentration.9,10 However, to the authors’ knowledge, online coupling of AF4 to scattering-based particle counting detectors has not been reported so far. Alternatively, single particle ICPMS (sp ICPMS) has emerged as a promising tool for the determination of the number-based concentration of nanomaterials by direct infusion of nanoparticles suspended in largely diluted samples. However, for difficult elements such as silicon, with poor ionization and high signal contribution of procedural blanks,
here are over 1300 nanomaterial-containing consumer products currently available on the market, ranging from cosmetics and food additives to paints or clothing, with global revenue expected to be worth €30 billion by 2015.1 Being one of the most highly produced and commonly used nanomaterials, silica (SiO2) nanoparticles (NP) can be found in a variety of industrial applications, e.g., in paints and coatings, microelectronic devices, and food production as an anticaking agent. Despite their broad commercial use, relevant regulations concerning nanomaterials have only recently started to emerge. For example, the presence of a nanomaterial and, more specifically, of a material with 50% (1−50% in specific cases) or more particles with one or more external dimensions in the size range of 1−100 nm2 should now be clearly indicated on the label of food3 and cosmetic4 products marketed in the European Union. More specifically, the EU recommendation stated that, for regulatory purposes, the size distribution of a material should be presented as size distribution based on the number concentration (i.e., the number of objects within a given size range divided by the number of objects in total) rather than the mass fraction of nanoscale particles, since a small mass fraction may contain the largest number of particles. Despite these requirements, reliable methodology for numberbased characterization of nanomaterials in complex samples is yet to be developed. Such methods will support regulatory and scientific assessment of potential hazards and the toxicological impact of nanomaterials on humans and the environment. © XXXX American Chemical Society
Received: March 19, 2015 Accepted: May 13, 2015
A
DOI: 10.1021/acs.analchem.5b01052 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
Figure 1. Schematic representation of the online characterization setup.
the lower size detection limit is still relatively poor (around 200 nm using a standard instrumental setup).11 This work investigates, for the first time, the potential of simultaneous online coupling of AF4 to NTA, ICPMS, and MALS for the characterization of SiO2 NP, in terms of size, size distribution, size-based elemental composition, and numberbased concentration, in a bovine serum matrix. The degree of selectivity offered by AF4 and its impact on the measurement error of NTA for the number-based concentration of the silica particles in the complex matrix was studied by online coupling of AF4 to NTA, without chemical sample preparation. ICPMS analysis of AF4 size fractions provided information on particle mass concentration, as well as on size-based elemental composition. From the Si mass fraction data, the equivalent number of particles was estimated and compared with that obtained by AF4 coupled to NTA. The online use of MALS detection was useful to obtain information on the particle size, as well as the characteristics of the particle capping layer. Since the developed methodology comprises two complementary size detectors and two concentration detectors, a comparison of the SiO2 NP number-based concentrations obtained with AF4/ NTA and AF4/ICPMS, as well as the size data generated with AF4/NTA and AF4/MALS, was achieved. This is in line with the recently published requirement by Linsinger et al.,12 expressing the importance of a multimethod approach to characterization of nanomaterials in complex samples. The online system used in this work was assembled as shown in Figure 1. An additional connector and software-controlled switching valve were introduced into the standard AF4/MALS/ ICPMS,13 allowing coupling of the NTA platform and a stopflow operation (dashed line) in an online mode. Equal portions of the sample outer-flow from the AF4/MALS were delivered to NTA and ICPMS (via T connector), while the valve was further redirecting the “NTA flow” either to the instrument or to waste, switching automatically every 15 or 30 s. First, the switching valve was opened to NTA, allowing acquisition of sample into the instrument. The acquired solution was then temporarily immobilized, prior to recoding NTA movies by flicking the valve and directing the sample flow to waste. Once the movie was recorded, the valve was switched back and the whole cycle was repeated several times, ensuring consistent sample influx in a time-resolved manner, as for all the other detectors. NTA movies were analyzed following a procedure described in the Supporting Information S-1, and the obtained
NP counts were plotted in Excel 2010 against given time points, resulting in NP concentration over time fractogram. The total number of particles present was calculated from the area of an integrated peak(s) representing separated analyte and the known amount of the sample injected into the AF4. NTA concentration values were further compared with ICPMS readouts. However, since ICPMS typically provides elemental concentrations,14 the ICPMS signal (Supporting Information S2) was converted to equivalent particle number (Supporting Information S-3). We show how these calculated values agree with values measured by AF4/NTA. Coupled online NTA and MALS enabled direct complementary size measurements on the separated population(s) of NP and additional molecular weight (MW, MALS detector) determination. The use of ultraviolet light (UV) detection offered insights into the characteristics of the matrix itself. The results suggest that the empirically obtained MW values agree well with the theoretically calculated (Supporting Information S-3) values, as well as the literature reports. The developed method was applied to a sample of aminated SiO2 NP suspended in pure fetal bovine serum, containing a cocktail of proteins and smaller organic and inorganic moieties (see Supporting Information S-1 for experimental details). It is important to obtain reliable data on NP size and number-based concentration in such a complex matrix in order to support development of standardized protocols for toxicological screens in NP safety assessment,15 hence our choice of matrix. Total sample separation and characterization time was 35 min, under the system operating conditions described in the Supporting Information S-1. Representative MALS, UV, NTA, and ICPMS fractograms collected over a 35 min time frame are shown in Figure 2. Displayed graphs indicate that the SiO2 NP fraction (marked in blue, with an elution time of approximately 20 min) can be separated from the excess matrix (serum proteins, marked in green) with the methodology developed here. As mentioned earlier, NP concentration, expressed as particle number, was derived from the total peak area on the NTA fractogram and the amount of injected material, while Si concentration was measured by ICPMS, using elemental Si standards as calibrants and a post-AF4 calibration approach14 (Supporting Information S-2). This approach was followed due to the absence of like-for-like particles suitable for external calibration. Since the ICP ionization efficiency of SiO2 NP and dissolved Si may not necessarily be the same, the difference in B
DOI: 10.1021/acs.analchem.5b01052 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
number of particles measured in the same sample with NTA alone, using a dilution factor that matches the outer-flow of the AF4, was significantly lower [(0.47 ± 0.10) × 1012 NP/g; average ± stdev, n = 3]. This could be attributed to the high background scattering of the protein matrix, requiring an increase in the detection threshold up to a level at which most of the silica particles are no longer counted. NP size values (Table 1) of 90.1 ± 5.1 and 87.9 ± 3.4 nm (average ± stdev, n = 3), obtained with coupled online AF4/ NTA and AF4/MALS, respectively, are in good agreement and suggest the presence of a capping layer with a thickness of around 4.0 nm (NTA) or 2.9 nm (MALS), if 82 nm diameter of the original material is assumed (Supporting Information S3). These values are in line with bovine serum albumin (main component of serum16) dimensions along the shorter axis,17 indicating formation of a single capping layer. The NP MW of 4.1 × 108 ± 8.7 × 106 g/mol measured by MALS additionally confirms the presence of a protein capping layer, since theoretical MW of “bare” 82 nm silica NP is around 3.47 × 108 g/mol (derived from weight of a single NP; Supporting Information S-3). It is only when the presence of protein monolayer with an estimated number of 781 molecules per single NP (assuming 3.7 × 1012/cm2 surface coverage18) or equivalent of (5.3 ± 0.3) × 107 g/mol MW is accounted for, that the calculated MW of capped NP (around 4.0 × 108) agrees with the measured value. The UV-measured MW of albumin [(6.8 ± 0.4) × 104 g/mol] is in good agreement with the literature data,19 offering additional and valuable insights into the characteristics of the matrix itself. The proteins present on the surface of silica NP were not detected by UV, most likely because their concentration was lower than the detection limit of this technique. In summary, the potential of the methodology developed here, which is based on the online coupling of NTA with AF4 and supported by theoretical calculations, has been demonstrated for the determination of number-based concentration of SiO2 NP in a complex bovine serum matrix. The degree of selectivity offered by AF4 was found necessary to determine reliably the number-based concentration of the SiO2 NP in the complex matrix by online coupling of AF4 to NTA with relative standard deviation of 5.1% (n = 3) and without chemical sample pretreatment. The online coupling of AF4 to ICPMS is only used to provide confirmatory measurements to the AF4/ NTA-based method, since unlike for NTA, the use of the AF4/ ICPMS technique for number-based concentration measurements is limited to particles with very well-known character-
Figure 2. Representative time-resolved fractograms collected online from a sample of silica nanoparticles suspended in pure bovine serum.
quantified Si concentration in mineralized (elemental) samples versus nanoparticulate form, both calibrated against elemental standards (Supporting Information S-4), was investigated and the average recovery was found to be 80.6 ± 4.3% (n = 4). Mass balance calculations, based on the obtained ICPMS fractograms (Supporting Information S-5), were also performed, and it was found that approximately 13.2 ± 3.9% Si (in the NP fraction) cannot be recovered from the AF4, probably due to interactions with the AF4 membrane or NP dissolution (due to the dissolved Si transported to the waste and, hence, not detected by ICPMS). The equivalent number of particles in the sample was calculated by taking into account the absolute NP recovery (67.6 ± 3.5%), as explained in Supporting Information S-4 and calculated by correcting for the difference between Si ions and NP nebulization efficiency and the silicon content in the separated fraction (801.5 ± 38.7, Table 1). The calculated value of (4.40 ± 0.21) × 1012 NP/g (based on Si measurements by AF4/ICPMS) is in agreement with measured value of (4.10 ± 0.21) × 1012 NP/g (based on measurements by AF4/NTA). This is a very important result, since it demonstrates the feasibility and potential of the online AF4/ NTA approach developed here for number-based concentration measurement of NP in a complex matrix with a relative standard deviation of 5.1% (n = 3). For comparison, the
Table 1. Concentration, Size, and Molecular Weight of Nanoparticles and Matrix Proteins in the Sample Determined Using the Online Techniques (Average ± Stdev, n = 3)a
a
Marked in blue are the number-based particle concentration values, in orange the particle size, in violet the MW of the particles, and in green the MW of proteins. C
DOI: 10.1021/acs.analchem.5b01052 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
(16) Zheng, X.; Baker, H.; Hancock, W. S.; Fawaz, F.; McCaman, M.; Pungor, E., Jr. Biotechnol. Prog. 2006, 22, 1294−1300. (17) Squire, P. G.; Moser, P.; O’Konski, C. T. Biochemistry-US 1968, 7, 4261−4271. (18) Brewer, S. H.; Glomm, W. R.; Johnson, M. C.; Knag, M. K.; Franzen, S. Langmuir 2005, 21, 9303−9307. (19) Hirayama, K.; Akashi, S.; Furuya, M.; Fukuhara, K. Biochem. Biophys. Res. Commun. 1990, 173, 639−646.
istics (i.e., density and the size of inorganic core). The results presented here also suggest that no significant SiO2 NP losses in the AF4/NTA system were observed.
■
ASSOCIATED CONTENT
S Supporting Information *
Experimental details, ICPMS totals, comparison of particulate and elemental Si, post-AF4 quantification, mass balance, and NP number calculations. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b01052.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Fax: +44(0) 2089432767. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the European Metrology Research Programme. We thank Z. Varga and M. Palmai from the Hungarian Academy of Sciences for preparation of aminated silica nanoparticles, as well as G. Roebben and V. Kestens from IRMM for providing base silica material and ampoules of the aminated silica. Biomedical Imaging Unit, University of Southampton, is gratefully acknowledged for support with TEM imaging.
■
REFERENCES
(1) McWilliams, A. BCC Market Research Reports; BCC Research: Wellesley, Massachusetts, USA, 2010; NAN031D, pp 1−276. (2) EC Commission Recommendation, 2011/696/EU; European Union: Luxembourg, 2011. (3) EC Regulation 1169/2011 of the European Parliament and of Council Official Journal of the European Union; European Union: Luxembourg, 2011; L 304/18. (4) EC Regulation 1223/2009 of the European Parliament and of Council Official Journal of the European Union; European Union: Luxembourg, 2009; L 342/59. (5) Saveyn, H.; De Baets, B.; Thas, O.; Hole, P.; Smith, J.; Van der Meeren, P. J. Colloid Interface Sci. 2010, 352, 593−600. (6) Kozak, D.; Anderson, W.; Vogel, R.; Trau, M. Nano Today 2011, 6, 531−545. (7) Wyatt, P. J. J. Colloid Interface Sci. 1998, 197, 9−20. (8) Hassellöv, M.; Lyven, B.; Haraldsson, C.; Sirinawin, W. Anal. Chem. 1999, 71, 3497−3502. (9) Aureli, F.; D’Amato, M.; Raggi, A.; Cubadda, F. J. Anal. Atom. Spectrom. 2015, DOI: 10.1039/C4JA00478G. (10) Barahona, F.; Geiss, O.; Urbán, P.; Ojea-Jimenez, I.; Gilliland, D.; Barrero-Moreno, J. Anal. Chem. 2015, 87, 3039−3047. (11) Peters, R.; Herrera-Rivera, Z.; Undas, A.; van der Lee, M.; Marvin, H.; Bouwmeester, H.; Weigel, S. J. Anal. Atom. Spectrom. 2015, DOI: 10.1039/C4JA00357H. (12) Linsinger, T.; Roebben, G.; Gilliland, D.; Calzolai, L.; Rossi, F.; Gibson, N.; Klein, C. JRC Reference report EUR 25404 EN; EUR − Scientific and Technical research series; European Union: Luxembourg, 2012; ISSN 1831-9424, DOI:10.2787/63490. (13) Schmidt, B.; Loeschner, K.; Hadrup, N.; Mortensen, A.; Sloth, J. J.; Bender Koch, C.; Larsen, E. H. Anal. Chem. 2011, 83, 2461−2468. (14) Heroult, J.; Nischwitz, V.; Bartczak, D.; Goenaga-Infante, H. Anal. Bioanal. Chem. 2014, 406, 3919−3927. (15) Teeguarden, J. G.; Hinderliter, P. M.; Orr, G.; Thrall, B. D.; Pounds, J. G. J. Toxicol. Sci. 2007, 95, 300−312. D
DOI: 10.1021/acs.analchem.5b01052 Anal. Chem. XXXX, XXX, XXX−XXX
