Analytica Chimica Acta 839 (2014) 8–13

Contents lists available at ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Elemental ratios for characterization of quantum-dots populations in complex mixtures by asymmetrical flow field-flow fractionation online coupled to fluorescence and inductively coupled plasma mass spectrometry Mario Menendez-Miranda, Maria T. Fernandez-Arguelles, Jose M. Costa-Fernandez *, Jorge Ruiz Encinar, Alfredo Sanz-Medel * Department of Physical and Analytical Chemistry, University of Oviedo, Avda, Julian Claveria 8, 33006 Oviedo, Spain

H I G H L I G H T S

G R A P H I C A L A B S T R A C T

 The hyphenated system allows unequivocal identification of nanoparticle populations.  AF4 separation permitted detection of unexpected nanosized species in a sample.  ICP-QQQ provides elemental ratios with adequate accuracy in every nanoparticle.  Purity and chemical composition of different quantum dot samples were assessed.

A R T I C L E I N F O

A B S T R A C T

Article history: Received 8 April 2014 Received in revised form 12 June 2014 Accepted 19 June 2014 Available online 23 June 2014

Separation and identification of nanoparticles of different composition, with similar particle diameter, coexisting in heterogeneous suspensions of polymer-coated CdSe/ZnS quantum dots (QDs) have been thoroughly assessed by asymmetric flow field-flow fractionation (AF4) coupled on-line to fluorescence and inductively coupled plasma mass spectrometry (ICPMS) detectors. Chemical characterization of any previously on-line separated nanosized species was achieved by the measurement of the elemental molar ratios of every element involved in the synthesis of the QDs, using inorganic standards and external calibration by flow injection analysis (FIA). Such elemental molar ratios, strongly limited so far to pure single nanoparticles suspensions, have been achieved with adequate accuracy by coupling for the first time an ICP-QQQ instrument to an AF4 system. This hyphenation turned out to be instrumental to assess the chemical composition of the different populations of nanoparticles coexisting in the relatively complex mixtures, due to its capabilities to detect the hardly detectable elements involved in the synthesis. Interestingly such information, complementary to that obtained by fluorescence, was very valuable to detect and identify unexpected nanosized species, present at significant level, produced during QDs synthesis and hardly detectable by standard approaches. ã 2014 Elsevier B.V. All rights reserved.

Keywords: Asymmetric flow field-flow fractionation Nanoparticles Quantum dots Triple quadrupole inductively coupled plasma mass spectrometry

1. Introduction * Corresponding authors. Tel.: +34 985102970. E-mail addresses: [email protected] (J.M. Costa-Fernandez), [email protected] (A. Sanz-Medel). http://dx.doi.org/10.1016/j.aca.2014.06.034 0003-2670/ ã 2014 Elsevier B.V. All rights reserved.

Luminescent semiconductor nanoparticles, or quantum dots, have become powerful fluorescent probes for the imaging of

M. Menendez-Miranda et al. / Analytica Chimica Acta 839 (2014) 8–13

biological samples [1]. Particularly, the core–shell nanoparticle type, (e.g. CdSe/ZnS QDs), are the most widely used QDs so far in bioanalysis [2]. It is considered that the success or failure in the use of the QDs for bioanalytical applications is largely determined by the quality and chemical stability of the nanocrystals [3]. However, despite the myriad of the established QDs synthesis routes [4], and the new ones continuously being developed [5], there is still an urgent need for effective procedures for the through characterization of the resulting synthesis solutions, especially at the chemical level. Of course, an appropriate nanoparticle characterization is also mandatory to evaluate their eventual toxicological impact [6– 10]. Conventional bulk characterization techniques such as UV–vis absorption and fluorescence spectroscopy provide highly valuable information about size, size distribution, mass and number concentration and extinction coefficients [11]. On one hand, other techniques including transmission electron microscopy (TEM) and dynamic light scattering (DLS) are used to assess the size of the core and the hydrodynamic radius of the nanoparticles, respectively. Furthermore, TEM combined with EDAX can provide additional elemental information about the composition of the sample, although the sensitivity of the technique is limited. However, it should be mentioned that these techniques present some important limitations (e.g. high matrix interferences, limited size resolution observed when analyzing nanoparticles with a core diameter of only few nanometers suspended in water media, eventual aggregation problems during aqueous sample handling before the measurement, etc.) when analyzing heterogeneous and complex mixtures such as raw synthesis solutions, environmental or biological samples. Nanoparticle tracking analysis (NTA) is also a powerful characterization method for nanoparticles. NTA enables the visualization of the sample, gives an approximate particle concentration and size information. However the operational size interval of the NTA technique (typically from 30 to 1000 nm) is not suited for analysis of NPs suspensions with sizes in the few nanometer range, as the QDs used in the present work. Interestingly, in this context, the measurement of elemental molar ratios using ICP-optical emission spectroscopy (ICPOES) [5], mass spectrometry (ICPMS) [12–14] or thermogravimetry in conjunction with elemental analysis [15], has been employed to characterize the chemical composition of single-type and pure metal nanoparticle suspensions. Such powerful approaches, allowed the authors to control ligand density of gold nanostructures [13–15], or the chemical composition of alloys core/ multishell quantum dots obtained under different synthesis conditions [5,12]. However, it should be considered that NPs tend to aggregate or to partially decompose, which poses a challenge to all those conventional bulk characterization techniques. Moreover, different synthesis can potentially produce unexpected by-products, which could not completely be removed using typical purification techniques (ultrafiltration, successive centrifugations) and are hardly detectable by standard approaches. Besides, the complete removal of free metal ions released from the nanoparticles in their suspensions must be assured to obtain accurate elemental ratios [16]. Thus, separation methods might play a critical role for a more complete characterization of the QDs suspensions. In this context, previous works showed the great promise of size exclusion chromatography and hydrodynamic chromatography for NPs characterization [17,18]. Unfortunately the chromatographic resolution of such techniques in the nanoscale size range turned out to be very limited. Alternatively, asymmetrical flow field-flow fractionation (AF4) has demonstrated its great potential to separate QDs and other NPs ranging from few nm up to several mm, with a low risk of sample degradation or aggregation, due to the lack of destructive forces during separation [19–22]. Further

9

interfacing of the AF4 to sensitive elemental specific detectors such as ICPMS could provide not only nanoparticle size separation but also elemental information at the mg L 1 level [23–25]. In a rather recent report, interfacing of a FFF separation method to ICP-MS has demonstrated to provide the ability to detect and size nanoparticles at the mg L 1 level [24]. The additional information gained from the simultaneous elemental-specific nature of the ICPMS allowed the authors to detect the metals present in a single type of quantum dots. In this article, the necessity to improve the instrumental capabilities to be able to measure sulphur properly needed to achieve a complete characterization of many engineered metal nanoparticles was remarked. Herein, on-line sensitive determination of elemental molar ratios by ICPMS has been performed, with adequate accuracy, on the individual nanosized species separated by AF4 and originally present in heterogeneous nanoparticle suspensions (containing more than one type of QDs). Such innovative information, in combination with that obtained by the use of an on-line molecular fluorescence detector, turned out to be critical to control and assess the quality and chemical composition of polymer-coated CdSe/ZnS quantum dots (QDs) mixtures (in which different nanoparticle species are present). Different QDs synthetic protocols were intentionally selected in order to have different quality and purity levels in the resulting QDs suspensions and thus, to evaluate the capability of the strategy for the characterization of relatively complex mixtures. A triple quad (ICP-QQQ) was coupled for the first time to the AF4 due to its capabilities to detect, free of interference and with high sensitivity, difficult elements such as selenium, zinc and especially sulphur [26], which has been proved to be critical for the characterization of the different species formed during QDs syntheses. 2. Materials and methods 2.1. Chemicals and solutions All chemicals used for QD synthesis and modification were of analytical grade and used as received without further purification. The precursors used for the synthesis of the nanoparticles were selenium powder (99.99%), cadmium oxide (99.99%), hexamethyldisilathiane, 1.0 M diethyl zinc solution in hexane, trioctylphosphine (TOP, 90%), trioctylphosphine oxide (TOPO, 99%) and anhydrous chloroform (99%), all of them purchased from Sigma– Aldrich (Schnelldorf, Germany). Hexylphosphonic acid (HPA) was obtained from Alfa Aesar (Karlsruhe, Germany). Ammonium acetate was used to prepare a 0.2 g L 1 buffered solution (pH = 7.3) employed as mobile phase in the AF4 instrument. Elemental ratios in samples were determined by external FIA calibration using Merck (Darmstadt, Germany) certified 1000 mg L 1 standards of Cd, S, Se and Zn. 2.2. Instrumentation UV–vis absorption spectra were recorded at room temperature on a Genesys 10S Thermo Scientific Spectrophotometer (Thermo Scientific, Germany). Fluorescence spectra were recorded on a Varian Cary Eclipse (Agilent, Germany) fluorescence spectrophotometer using a fixed excitation wavelength of 350 nm with both excitation and emission slit width of 10 nm. All measurements were carried out using quartz cuvettes. The core size and the concentration of the QDs were calculated according to Peng's equations based on measurements of absorbance and fluorescence [11]. Dynamic light scattering (DLS) measurements were carried out with a NanoZS90 (Malvern Instruments, USA). The AF4 system used was an AF2000 Focus model purchased from Postnova Analytics Inc. (Landsberg, Germany). Details of the

10

M. Menendez-Miranda et al. / Analytica Chimica Acta 839 (2014) 8–13

AF4 technique principles can be found elsewhere [27]. After fractionation, the effluent was first directed to a fluorescence detector (Agilent 1200 Series, Germany, operated at an excitation wavelength of 280 nm and at emission wavelengths of 450 and 600 or 620 nm depending on the sample) and then to an ICP-QQQ detector (Agilent 8800, Tokyo, Japan), both coupled on-line to the AF4 system. Separation conditions were adjusted to allow separation of the different nanoparticle species present in the suspensions obtained after the QDs synthesis. Experimental conditions optimized for the AF4 separation of the QDs species are summarized in Table S1 (see Supplementary data). A twoposition and six-port injection valve (Rheodyne, IDEX Health & Science, Germany) was placed between the fluorimeter detector and the ICPMS to allow an on-line flow injection of a calibration solution containing Cd, Se, S and Zn standards. Simultaneous detection of S, Zn, Se and Cd was carried out with an ICP-QQQ using a micromist nebulizer [26]. Oxygen was introduced in the collision/reaction cell at a flow rate of 0.35 mL min 1. Operation conditions were daily optimized using a tuning solution. S and Se were detected in mass shift mode (48SO+ and 96SeO+, respectively) after their reaction with oxygen in the cell. Zn and Cd elements, which do not react with oxygen, were measured in on-mass MS/MS mode (66Zn+ and 111Cd+ respectively). The integration time for each of the targeted isotopes 32S, 34S, 64Zn, 66Zn, 80Se, 82Se, 110Cd and 111Cd was 100 ms. Integration of the fractogram peaks was performed using the MassHunter software (Agilent). Computation of the different elemental molar ratios within the nanoparticles were carried out by resorting to external calibration, performed by FIA, of the different elements present in the core and the shell of the QDs. Peak area ratios observed could be translated into elemental molar ratios using the corresponding elemental calibration curves. Confidence interval associated to the elemental molar ratios provided all along the text corresponded to one standard deviation (the numbers of replicates used to obtain such SD values was always n = 3). 2.3. Synthesis of CdSe/ZnS “One-Pot” quantum dots CdSe/ZnS QDs were synthesized using CdO as precursor via the organometallic route described elsewhere [28,29]. Once the synthesis was finished, the mixture was centrifuged to remove excess of reagents and then the precipitated QDs were redispersed in anhydrous chloroform and stored at room temperature in the dark.

(based on poly-maleic anhydride functional groups), following a procedure described elsewhere [31]. Results and discussion Two different hydrophilic QDs samples, obtained following the different procedures described in the Experimental Section, were analyzed. In the “1-Pot QDs” and “2-Pot QDs” samples, the ZnS shell growing was performed in the presence and absence, respectively, of the excess of reagents (Cd and Se) employed for the core nucleation. The quality of the synthesized QDs was first evaluated from UV– vis TEM and DLS measurements (see Figs. S1, S2 and S3 of the appendix Supplementary data). Absorption UV–vis spectrophotometry can be used to estimate the size, concentration and extinction coefficient of the QDs, with good accuracy, following the approach proposed by Peng and co-workers [11]. Electron microscopy showed the presence nanoparticles with relatively low size dispersion (see Fig. S2). The same samples were also assessed using DLS (see Fig. S3). Results showed that synthetic routes assayed provided a single distribution with an adequate polydispersity index (PDI) ranging from 0.236 to 0.248. Additionally, the full width at half maximum (FWHM) of the fluorescence emission is a commonly employed parameter to estimate the size distribution of the nanoparticles (FWHM values lower than 35–40 nm is considered to indicate highly monodisperse NPs). Fig. 1 shows the observed emission fluorescence spectra of the two types of nanoparticle suspensions investigated here. It can be observed that “1-Pot QDs” showed an intense emission band with a maximum at 620 nm, and an FWHM of 75 nm, indicating an undesirable degree of size dispersion. Additionally, a less intense emission band, with a maximum at 455 nm, is apparent. This can be attributed to the presence of impurities or different size nanoparticles originated during the synthesis of the CdSe/ZnS QDs. Conversely, as also shown in Fig. 1, the fluorescence spectrum of the “2-Pot QDs”, exhibits a much narrower emission band (FWHM  40 nm) and there is not a clear evidence of the presence of another species population emitting at 450 nm. However, an undesirable fluorescence emission background in the range between 400 and 500 nm is still apparent. Both the emission registered at 450 nm and the high photoluminescent background observed in this spectral region could be originated from coexisting species or different less abundant nanoparticle populations. In principle, such emission would not represent a big constraint for further photoluminescent applications as the analytical signal (emission maximum of the QDs fluorescence, centered at 610 nm) is well separated from that unforeseen photoluminescent emission at 450 nm. However, such unknown

2.4. Synthesis of CdSe/ZnS “Two-Pot” quantum dots In this protocol, the CdSe core of the QDs synthesized following the approach previously described (“one-pot synthesis”) were subjected to a new purification step [30]. Hence, once the CdSe cores reached the desired size, the solution was cooled down to room temperature, and the nanoparticles were precipitated with methanol. The supernatant containing the excess of reagents was discarded, and the nanoparticles were redispersed in hexane. Afterwards, the CdSe cores dispersion was heated up to 190  C and the Zn/S/TOP solution were added to allow the growth of the ZnS shell around the CdSe cores following the same procedure already described in the “one-pot synthesis”. 2.5. Surface modification for QDs aqueous suspension The fluorescent QDs used in our experiments were synthesized from organometallic precursors. In order to make the nanoparticles hydrophilic to render them compatible with bioanalytical applications, the QDs were coated with an amphiphilic polymer

Fig. 1. Fluorescence spectra. Dashed line: “1-Pot QDs”; solid line: “2-Pot QDs”.

M. Menendez-Miranda et al. / Analytica Chimica Acta 839 (2014) 8–13

species could likely interfere significantly in all those bioanalytical applications in which the metal nanoparticles act as fluorescent/ metal tags since those by-products could be also bioconjugated to the recognition elements (i.e. antibodies), thus interfering the final determination. In addition, we observed that standard purification procedures (i.e. centrifugation) are not efficient enough to get rid of such unknown species as they could have similar size to the sought NPs. Therefore, it is an urgent need to further characterize the desired products and the eventual impurities present in the nanoparticle suspension obtained from the different syntheses. For that purpose, we resorted to a separation of particles populations by AF4, coupled on-line to fluorescence and also to ICPMS detectors. The fluorescence detector was used to monitor any fluorescent emission from the QDs core, as well as from any possible fluorescent species present in the final nanoparticle suspension. On the other hand, the use of the ICPMS detector was intended to monitor the different elements (Cd, Se, S and Zn), that constitute the core and shell of CdSe/ZnS QDs, as well as other possible metal nanoparticles eventually obtained. In this context, it is worth stressing that the new ICP-QQQ used here allows the sensitive [26] and interference-free detection of highly interfered elements such as Se and Zn and especially, S. In addition, ICPMS detection also allowed us to evaluate the metal nanoparticle recoveries after AF4 separations, which in all the cases were over 70%. The obtained fractogram of “1-Pot QDs”, measuring the fluorescent emission at the two different wavelengths observed previously in the bulk analysis, 450 and 620 nm (), is shown in Fig. 2a. It is apparent that there are three peaks, likely corresponding to three species with different sizes and/or composition in the nanometer range. It is clear that AF4 offers

Fig. 2. Fractogram of “1-Pot QDs” with (a) fluorescence detection: dashed and solid lines correspond to emission registered at 450 and 620 nm, respectively. (b) ICPMS detection, isotopes measured: solid line: 32S; dotted line: 66Zn; dash-dot line: 80Se; dashed line: 111Cd.

11

very valuable additional information to that of the conventional bulk spectrophotometric techniques typically used for characterizing nanoparticle populations. In fact, bulk analysis (see Fig. 1) could lead to the wrong conclusion that only 2 species were present at the most, the desired most abundant and its small impurity emitting at 620 and 450 nm, respectively. Such interpretation, however, turned out to be rather limited since Fig. 2a clearly shows that there are at least three significant species. Of course, it is expected that most intense peak at 620 nm (25 min) could correspond to the sought CdSe/ZnS QDs. The other 2 species with lower hydrodynamic volumes eluting at 13 and 17 min could most likely correspond to degraded (poor polymer coated) CdSe/ZnS QDs, free polymer micelles and/or undesired ZnS nanoparticles formed during the shell formation. In order to corroborate such assumptions, the chemical compositions of the different peaks observed were measured by ICPMS and the fractogram obtained is shown in Fig. 2b. Interestingly the elemental content of the different species/peaks clearly differed. As expected, the peak eluting at 25 min contained significant amounts of the four elements monitored. Moreover, the calibration plots obtained using FIA could be used to translate the different intensity area ratios into elemental molar ratios. In this sense, the Cd/Se molar ratio obtained 2.14  0.19, clearly indicated the presence of a multishell due to an epitaxial growth of CdS layer before or mixed with the ZnS. This was possible as the excess of the Cd reagent was still present when the Zn and S reagents were added during the shell formation [5]. However, taking into account that the mol of S detected in this peak should be combined to Zn and Cd in the shell, the remaining mol of Cd can be easily computed and therefore the Cd/Se molar ratio in the nanoparticle core estimated. The value obtained, 1.35  0.16, matches well with the value obtained previously for the single and unshelled CdSe QDs, 1.27  0.03 [12]. Elemental MS fractogram in Fig. 2b also revealed that the second major peak corresponded to CdS nanoparticles, which are known to have smaller size than the CdSe/ZnS QDs and an intense fluorescence emission at 450 nm [32]. The Cd/S molar ratio computed in this peak, 1.16  0.13, again showed a slight excess of the metallic element, in accordance to results already published for CdSe [5,12,33] and PbSe [34] nanoparticles. The presence of traces of Zn and Se in such nanoparticle or the coelution of unknown species cannot be neglected. However, the Zn/S molar ratio measured in such peak, 0.013  0.001 was very far away from the value expected for eventual ZnS nanostructures. Interestingly, in spite of the weak emission band observed at 450 nm in the bulk fluorescence spectrum (Fig. 1), up to 38% of the total Cd detected as nanoparticle in this sample is present in such unexpected CdS nanoparticle population. Finally, this approach was followed to clarify the composition of the band eluting at around 14 min, which showed some fluorescence emission at 600 and 450 nm as well (Fig. 2a). The presence of Zn and Se exclusively (Fig. 2b) and the intense emission at 450 nm seemed to support the idea of a ZnSe nanoparticle population, which could be formed from the excess of the Se reagent present at the time of the addition of the Zn compound used for the shell formation. However the molar Zn/Se ratio observed, 9.8  0.5, ruled out such possibility. Therefore, although bulk fluorescence and DLS characterization of “1-Pot QDs” indicated that the size distribution of the sample was acceptable, analysis by AF4 system with elemental and molecular detection shows that different populations coexist in the nanoparticle suspension having different hydrodynamic volumes and, what is more interesting, different chemical composition. With the aim of further assessing the potential of AF4 with elemental/molecular detection to carry out more complete characterization of QDs, a possibly different nanoparticle suspension (“2-pot synthesis”) was analyzed. Following a 2-pot

12

M. Menendez-Miranda et al. / Analytica Chimica Acta 839 (2014) 8–13

procedure, a purification step is accomplished right after the core nucleation and growth take place, thus minimizing the presence of Cd or Se precursors excess. In this case, the addition of the precursors of the shell (i.e. Zn2+ and S2 ) is done in the virtual absence of Cd2+ and Se2 species. As expected, a simpler fluorescence fractogram profile was obtained, as shown in Fig. 3a. Only two well differentiated species eluting at 16 min and 26 min respectively were observed with intense photoluminescence emission at 450 and about 600 nm, respectively. Complementarily, ICPMS fractogram (Fig. 3b) corroborated that the species eluting at 26 min could be again ascribed to CdSe/ZnS QDs. In this case, the global Cd/Se molar ratio, 1.51  0.15, was found to be significantly lower than the global ratio found in Fig. 1 for the “1-Pot QDs” at similar elution time (2.14  0.19). This fact can be explained by the significant removal of the Cd reagent excess at the time of the shell formation in the “2-Pot QDs” synthesis, which resulted in a shell layer formed mostly of ZnS, with lower amounts of CdS [5]. In fact, the Zn/S molar ratio computed, 0.65  0.08, clearly indicated that the main component of the mixed shell was ZnS. Conversely, the main component in the shell of the “1-Pot QDs” (Fig. 2b, peak at 26 min) was CdS, as clearly shown by the much lower Zn/S ratio observed, 0.22  0.02. These results are in good agreement with recent studies [5] where highly different CdS/ZnS compositions of the multishell were obtained after varying the Cd precursor content under strictly controlled synthesis conditions. Moreover, after computing again the CdSe molar ratio in the nanoparticle core of the “2-pot QDs” (Fig. 3b), the value obtained (1.35  0.22) was statistically undistinguishable from the value estimated before for the “1-Pot QDs” synthesis, which seems to internally validate the approach.

The independent CdS nanoparticle population previously observed for the “1-Pot QDs” synthesis at 17 min (Fig. 2b) disappeared completely. On the other hand, a second broad peak (from 13 to 20 min) is observed in the fractogram of the “2-Pot QDs”, which had a strong maximum fluorescent emission at 450 nm (Fig. 3a). Interestingly, the elemental composition of such species revealed that it was almost exclusively made of Zn (Fig. 3b). Analysis of this fraction by TEM confirmed the absence of metallic nanoparticles as well. Previous studies carried out in our laboratory showed that, depending on the environmental conditions used, certain loses of the polymer coating from the surface of the QDs can occur [17]. Such polymer residues (micelles) emit strong luminescence at about 450 nm and could likely correspond to this broad peak in Fig. 3a. In this context, the coelution of a significant Zn signal at this time (Fig. 3b) can be explained considering the strong affinity of the maleic acid-based polymers used in our synthesis for free metals, especially Zn2+ [35], which can be present as shell reagent excess or after releasing from the QDs outer shell. As a matter of fact, such Zn complexation by polymer-based species could also explain the unknown Znenriched peak (appearing at about 14 min) and its later shoulder (17 min) previously observed in the fractogram of Fig. 2b (“1-Pot QDs”). The synthetic procedures followed in these studies resulted in significantly heterogeneous samples and therefore were ideal to actually demonstrate the capabilities of the proposed methodology to chemically characterize metallic nanoparticles suspensions. In any case, an aliquot of “1-Pot QDs” was subjected to a further purification process aiming to improve the quality of the product (see Supplementary data). This sample was again successfully analysed by the developed methodology and confirmed the improvement in the homogeneity of the nanoparticle suspension. In any case, results obtained are shown in the supplementary data and are in agreement with those obtained for the synthesis solutions already described in the text. 3. Conclusions

Fig. 3. Fractogram of “2-Pot QDs” with (a) fluorescence detection: dashed and solid lines correspond to emission registered at 450 and 620 nm, respectively. (b) ICPMS detection, isotopes measured: solid line: 32S; dotted line: 66Zn; dash-dot line: 80Se; dashed line: 111Cd.

In brief, AF4 enabled a time-resolved nanosize separation of the eventual species present in the heterogeneous nanoparticle suspensions obtained after carrying out different QDs syntheses protocols. The ICP-QQQ, used for the first time as detector of AF4, provided on-line elemental and isotopic information that turned out to be critical to assess the chemical composition of the different populations of nanoparticles coexisting in a given sample. This performance derives from the ICP-QQQ capabilities to detect Se, Zn and especially S, with high sensitivity and minimal interferences. Accurate elemental molar ratios in each of the peaks detected within each synthesis route, even those corresponding to unexpected species, could be then computed. It was observed that, after simple application of the bulk molecular techniques, typically used for characterizing NPs, to the analysis of nanoparticle suspensions obtained following different QDs synthesis routes, it could be concluded that a single (or at least, a dominant) type of nanoparticles was present in the samples. Such information, however, was far away from the real situation. The AF4-ICP-QQQ/ fluorescence fractograms obtained proved the presence of a mixture of different metal nanoparticles, at significant level, coexisting in the different suspensions under study. The approach and results provided in this work could be highly valuable to control the quality of the products obtained in a metal nanoparticle synthesis, in terms of purity and chemical composition of the NPs obtained. Indeed, this methodology could be a powerful diagnostic tool to evaluate the stability of metal nanomaterials and to optimize or control the synthesis of novel engineered hybrid nanoparticles.

M. Menendez-Miranda et al. / Analytica Chimica Acta 839 (2014) 8–13

Acknowledgments Financial support provided by the Spanish Ministry of Education (CTQ-2010-02309) and Agilent Foundation is also acknowledged. M.M.M. acknowledges the Ph.D. grant (BP12-046) from Principado de Asturias (Spain). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.aca.2014.06.034. References [1] S.G. Penn, L. He, M.J. Natan, Nanoparticles for bioanalysis, Curr. Opin. Chem. Biol. 7 (2003) 609–615. [2] X. Ji, J. Zheng, J. Xu, V.K. Rastogi, T. Cheng, J.J. DeFrank, R.M. Leblanc, (CdSe)ZnS quantum dots and organophosphorus hydrolase bioconjugate as biosensors for detection of paraoxon, J. Phys. Chem. B 109 (2005) 3793–3799. [3] J.M. Montenegro, I. Nabiev, A. Sukhanova, S. Agarwal, J.M. de la Fuente, V. Grazu, A. Greiner, W.J. Parak, Controlled antibody/(bio-) conjugation of inorganic nanoparticles for targeted delivery, Adv. Drug Deliv. Rev. 65 (2013) 677–688. [4] R. Martinez-Mañez, The Supramolecular Chemistry of Organic–Inorganic Hybrid Materials, 1st ed., Wiley, New York, 2010, pp. 155–191. [5] S. Jun, E. Jang, Bright and stable alloy core/multishell quantum dots, Angew. Chem. Int. Ed. 52 (2013) 679–682. [6] T.S. Hauck, R.E. Anderson, H.C. Fischer, S. Newbigging, W.C.W. Chan, In vivo quantum-dot toxicity assessment, Small 6 (2010) 138–144. [7] H. Fischer, L. Liu, K. Pang, W.C.W. Chan, Pharmacokinetics of nanoscale quantum dots: in vivo distribution, sequestration, and clearance in the rat, Adv. Funct. Mater. 16 (2006) 1299–1305. [8] S.J. Soenen, P. Rivera-Gil, J.M. Montenegro, W.J. Parak, S.C. de Smedt, K. Braeckmans, Cellular toxicity of inorganic nanoparticles: common aspects and guidelines for improved nanotoxicity evaluation, Nano Today 6 (2011) 446– 465. [9] E. Bolea, J. Jimenez-Lamana, F. Laborda, I. Abad-Alvaro, C. Blade, L. Arolab, J.R. Castillo, Detection and characterization of silver nanoparticles and dissolved species of silver in culture medium and cells by AsFlFFF-UV–visICPMS: application to nanotoxicity tests, Analyst 139 (2014) 914–922. [10] F. Ribeiro, J.A. Gallego-Urrea, K. Jurkschat, A. Crossley, M. Hassellöv, C. Taylor, M.V.M. Amadeu, S.S. Loureiro, Silver nanoparticles and silver nitrate induce high toxicity to Pseudokirchneriella subcapitata, Daphnia magna and Danio rerio, Sci. Total Environ. 466 (2014) 232–241. [11] W.W. Yu, L. Qu, W. Guo, X. Peng, Experimental determination of the extinction coefficient of CdTe, CdSe, and CdS nanocrystals, Chem. Mater. 15 (2003) 2854– 2860. [12] A.R. Montoro, J.R. Encinar, M.T. Fernandez-Arguelles, J.M. Costa-Fernandez, A. Sanz-Medel, Elemental mass spectrometry: a powerful tool for an accurate characterisation at elemental level of quantum dots, Chem. Commun. (2009) 3107–3109. [13] W. Jiang, B.D. Hibbert, G. Moran, R. Akter, Measurement of gold and sulfur mass fractions in L-cysteine-modified gold nanoparticles by ICP-DRC-MS after acid digestion: validation and uncertainty of results, J. Anal. At. Spectrom. 27 (2012) 1465–1473. [14] H. Hinterwirth, S. Kappel, T. Waitz, T. Prohaska, W. Lindner, M. Lämmmerhofer, Quantifying thiol ligand density of self-assembled monolayers on gold nanoparticle by inductively coupled plasma-mass spectrometry, ACS Nano 7 (2013) 1129–1136. [15] Q. Huifen, J. Rongchao, Controlling nanoparticles with atomic precision: the case of Au144(SCH2CH2Ph)60, Nano Lett. 9 (2009) 4083–4087. [16] M. Bouby, H. Geckeis, F.W. Geyer, Application of asymmetric flow field-flow fractionation (AsFlFFF) coupled to inductively coupled plasma mass spectrometry (ICPMS) to the quantitative characterization of natural colloids and synthetic nanoparticles, Anal. Bioanal. Chem. 392 (2008) 1447–1457.

13

[17] L. Trapiella-Alfonso, A.R. Montoro, J.R. Encinar, J.M. Costa-Fernandez, R. Pereiro, A. Sanz-Medel, New integrated elemental and molecular strategies as a diagnostic tool for the quality of water soluble quantum dots and their bioconjugates, Nanoscale 3 (2011) 954–957. [18] E.P. Gray, T.P. Bruton, C.P. Higgins, R.U. Halden, P. Westerhoff, J.F. Ranville, Analysis of gold nanoparticle mixtures: a comparison of hydrodynamic chromatography (HDC) and asymmetrical flow field-flow fractionation (AF4) coupled to ICP-MS, J. Anal. At. Spectrom. 27 (2012) 1532–1539. [19] H. Hagendorfer, R. Kaegi, J. Traber, S.F.L. Mertens, R. Scherrers, C. Ludwig, A. Ulrich, Application of an asymmetric flow field flow fractionation multidetector approach for metallic engineered nanoparticle characterization – prospects and limitations demonstrated on Au nanoparticles, Anal. Chim. Acta 706 (2011) 367–378. [20] A. Zattoni, D.C. Rambaldi, P. Reschiglian, M. Melucci, S. Krol, A.M. Coto-Garcia, A. Sanz-Medel, D. Roessner, C. Johann, Asymmetrical flow field-flow fractionation with multi-angle light scattering detection for the analysis of structured nanoparticles, J. Chromatogr. A 1216 (2009) 9106–9112. [21] J. Gigault, T.J. Cho, R.I. MacCuspie, V.A. Hackley, Gold nanorod separation and characterization by asymmetric-flow field flow fractionation with UV–vis detection, Anal. Bioanal. Chem. 405 (2013) 1191–1202. [22] H.E. Pace, E.K. Lesher, J.F. Ranville, Influence of stability on the acute toxicity of CdSe/ZnS nanocrystals to Daphnia magna, Environ. Toxicol. Chem. 29 (2010) 1338–1344. [23] B. Schmidt, K. Loeschner, N. Hadrup, A. Mortensen, J.J. Sloth, C.B. Koch, E.H. Larsen, Quantitative characterization of gold nanoparticles by field-flow fractionation coupled online with light scattering detection and inductively coupled plasma mass spectrometry, Anal. Chem. 83 (2011) 2461–2468. [24] A.J. Bednar, A.R. Poda, D.M. Mitrano, A.J. Kennedy, E.P. Gray, J.F. Ranville, C.A. Hayes, F.H. Crocker, A.J. Steevens, Comparison of on-line detectors for field flow fractionation analysis of nanomaterials, Talanta 104 (2013) 140–148. [25] V. Nischwitz, H. Goenaga-Infante, Improved sample preparation and quality control for the characterisation of titanium dioxide nanoparticles in sunscreens using flow field flow fractionation on-line with inductively coupled plasma mass spectrometry, J. Anal. At. Spectrom. 27 (2012) 1084– 1092. [26] S.D. Fernández, N. Sugishama, J.R. Encinar, A. Sanz-Medel, Triple quad ICPMS (ICPQQQ) as a new tool for absolute quantitative proteomics and phosphoproteomics, Anal. Chem. 84 (2012) 5851–5857. [27] R.N. Qureshi, W.T. Kok, Optimization of asymmetrical flow field-flow fractionation (AF4), LC–GC Eur. 23 (2010) 18–25. [28] A. Peng, X. Peng, Nearly monodisperse and shape-controlled CdSe nanocrystals via alternative routes: nucleation and growth, J. Am. Chem. Soc. 123 (2001) 183–184. [29] L. Trapiella-Alfonso, J.M. Costa-Fernández, R. Pereiro, A. Sanz-Medel, Development of a quantum dot-based fluorescent immunoassay for progesterone determination in bovine milk, Biosens. Bioelectron. 26 (2011) 4753– 4759. [30] B.O. Dabbousi, J. Rodriguez-Viejo, F.V. Mikulec, J.R. Heine, H. Mattoussi, R. Ober, K.F. Jensen, M.G. Bawendi, (CdSe)ZnS core–shell quantum dots: synthesis and characterization of a size series of highly luminescent nanocrystallites, J. Chem. Phys. Chem. B 101 (1997) 9463–9475. [31] M.T. Fernandez-Arguelles, A. Yakovlev, R.A. Sperling, C. Luccardini, S. Gaillard, A. Sanz-Medel, J.M. Mallet, J.C. Brochon, A. Feltz, M. Oheim, W.J. Parak, Synthesis and characterization of polymer-coated quantum dots with integrated acceptor dyes as FRET-Based nanoprobes, Nano Lett. 7 (2007) 2613–2617. [32] J.S. Steckel, J.P. Zimmer, S. Coe-Sullivan, N.E. Stott, V. Bulovi c, M.G. Bawendi, Blue Luminescence from (CdS)ZnS Core–Shell Nanocrystals, Angew. Chem. Int. Ed. 43 (2004) 2154–2158. [33] J. Jasieniak, P. Mulvaney, From Cd-rich to Se-rich – the manipulation of CdSe nanocrystal surface stoichiometry, J. Am. Chem. Soc. 129 (2007) 2841–2848. [34] I. Moreels, K. Lambert, D.D. Muynck, F. Vanhaecke, D. Poelman, J.C. Martins, G. Allan, Z. Hens, Composition and size-dependent extinction coefficient of colloidal PbSe quantum dots, Chem. Mater. 19 (2007) 6101–6106. [35] R. Hasanzadeh, P.N. Moghadam, N. Samadi, Synthesis and application of modified poly (styrene-alt-maleic anhydride) networks as a nano chelating resin for uptake of heavy metal ions, Polym. Adv. Technol. 24 (2013) 34–41.