Strategies for ambient aerosols characterisation using synchrotron X-ray fluorescence: a review

Ursula E.A. Fittschen

Institute of Inorganic and Applied Chemistry, University of Hamburg, Martin-Luther-King-Platz 6, D-20146 Hamburg, Germany. E-mail: [email protected]

Introduction

Ambient aerosols are of interest in many different regards, e.g., human health,¹ cloud formation and their effect on global temperatures, as well as transport of nutrients to the oceans. The elemental load and the particle size are important information required to determine the impact of the aerosols in terms of their penetration into the lungs, and their toxic effects on human health, their activities as condensation nuclei and their transport efficiencies.^2,3 To be able to resolve atmospheric events like a storm front which lasts from 2 h to 6 h, monitoring intervals of 1 h are desirable. Accordingly, trace elemental detection limits for micro amounts of sample materials are required. Until now characterisation of aerosols representative for short time scales (1 h and less) remained a challenging analytical task. Size-resolved determination of major particulate matter components using conventional bulk analytical techniques typically fall short of  being capable of providing time resolutions of between 6 h and 24 h. Fast bulk on-line methods are available for non-metal compounds. Calibration with model aerosols allows, to some extent, for quantification on a mass per sampled air volume.⁴ Quantification of specific single-particles or particle ensemble components is still a challenge.⁵ A major drawback of these on-line methods is their destructive nature, which in general does not allow for further analyses of the sample. In this sense X-ray fluorescence related techniques have been established in many fields of aerosol research. In this review aerosol characterisation using synchrotron excitation X-ray fluorescence related techniques are highlighted with respect to studies of environmental relevance and atmospheric elemental transport. The review does not claim to give a full overview of all the work that has been done in these fields, but highlights procedures and results by selected examples.

Size fractionation

The general approach for size fractionated sampling of atmospheric aerosols in most off-line procedures is the separation of particles of different inertia by impaction. This is translated into their aerodynamic diameter, by assuming a standard density (1 kg L^–1) and shape (spherical). Impactors are often designed to separate particulate matter in air as dictated by regulation, monitoring PM10 (particulate matter Ø < 10 µm), PM2.5 (particulate matter Ø < 2.5 µm) and ultrafine dust (particulate matter Ø < 1 µm).

The general setup of an impactor is depicted in Figure 1.

The air is sucked through the impactor by a vacuum pump. The air flow is controlled by nozzles and directed to the impaction plates. The air has to go round the plates. Particles of high inertia cannot follow the airflow and impact on the plates. Usually the plates are covered with a filter material to collect the particles and allow for determination of mass of, e.g. PM10 or PM2.5. The amounts of airborne matter collected during 1 h is limited by the airflow through the impactor, which is in the range of 1 m³ h^–1 for the devices used in the work cited here.

To achieve the sensitivity allowing short sampling times with synchrotron radiation X-ray fluorescence (SR-XRF) analysis background noise stemming from the sample carrier has to be minimised. Two to three nm penetration of the primary X-rays into the reflector and therefore minimal background scattering is observed in total reflection XRF (TXRF) geometry.⁶ Total reflection of X-rays is realised by using polished reflectors and very small angles of the excitation beam, less than the critical angle of total reflection of X-rays. In TXRF, Si-wafer impaction plates have been used with the dual purpose of being, first, the impaction plate for aerosol collection and, subsequently, the sample carrier reflector for total reflection X-ray fluorescence.^7,8,9 It has been shown that SR-TXRF is a very potent method for elemental determination of size fractionated aerosols.^10,11 In addition to SR-TXRF, other synchrotron-assisted XRF techniques have been applied successfully for size-resolved elemental determination in aerosols. Bukowiecki et al.¹² used a conventional 45° geometry micro-XRF setup and a thin 1.8 µm thickness polymer film (Mylar^®) as impaction plates and sample carrier, which produce negligible background scatter comparable to TXRF.

Low limits of detection

For both; TXRF geometry and micro-XRF geometry low limits of detection (DL) are usually in the single digit pg m^–3 regime for, e.g., transition metals from Ti to Zn and Pb. Bukowiecki et al.¹² report DLs of 17 pg m^–3 Fe and 19 pg m^–3 Pb (hourly monitoring and 15 s XRF spectra collection time) studying ambient air in areas in Zürich, Switzerland. Fittschen et al.¹³ found DLs of 0.8 pg m^–3 Fe and 2 pg m^–3 Pb (hourly monitoring and 100 s XRF spectra collection time) monitoring air in Hamburg, Germany. For both studies the bending magnet beam line L at DORISIII at the DESY, Hamburg, was used as the SR source. SR-XRF analysis not only benefits from higher flux of the excitation beam and polarisation but often the SR-beam line allows for optimising excitation and detection parameters. This has been demonstrated for the analysis of P in atmospheric aerosols. XRF analysis of low Z elements like P usually is unfavourable because of low fluorescence yield and absorption effects. At the ANKA Fluo beam-line at Karlsruhe the excitation energy was set to 3.5 keV to have optimal excitation of the P K-edge and to prevent from Ca Ka escape peak interferences exciting below the Ca K-edge energy of 4.04 keV by Fittschen et al.¹⁴ In addition, the evacuated sample environment reduced absorption of the relatively low energy P K lines, which are otherwise easily absorbed in air atmosphere. These optimisations resulted in DLs for P of 200 pg m^–3 (hourly monitoring and 100 s XRF spectra collection time; nine size fractions from 15 nm to 16 µm).
P in aerosols is of interest with regard to nutrient transport to the far oceans. The relevance of intercontinental transport of P from Saharan dust to the South-American rain forests has been controversially discussed.^15,16 Because of the low absolute amounts of P (< 10 ng m^–3) in atmospheric aerosols the impact of atmospheric P transport on marine plankton and plant growth in nutrient depleted areas remains unclear.

With the high sensitivity for P established using SR-TXRF it was possible to show the high variability of P in aerosols with respect to their particle size fraction and location. Figure 2 shows TXRF spectra of aerosols of different particle sizes. It can be seen that the 250–500 nm fraction (blue) shows the highest P content, which corresponded to 180 ng m^–3, and the coarse fraction (1–2 µm; red) the lowest content of 7 ng m^–3.

Automated sampling, fractionation and small particle collection

The impaction devices used in the studies cited here provide different advantages. Best suited for continuous monitoring of atmospheric aerosols is the rotating drum impactor first designed by Lundgren in 1967.¹⁷ A modified design was developed at the “Eidgenössische Materialprüfungs- und Forschungsanstalt” (Empa)¹². The aerosols are collected on a polymer film through a line shape orifice. The film is installed on a wheel, which is rotated after a certain time, e.g. 1 h, to collect a new charge of aerosol. Figure 3 shows the rotating drum impactor.

This device collects the aerosols in three size fractions corresponding to PM10, PM2.5 and PM1 with a low cut off of 100 nm for the smallest fraction. So, recent measurements indicate a slightly higher cut-off.¹⁹ Much better size fractionation of 11 fractions starting from 16 µm and with a low particle cut off of 15 nm is achieved with the low pressure Berner impactor.¹⁰ The higher number of fraction-ation steps allows for better diagnostic of air mass origin and chemical species of the elements in the aerosol. The aerosols are collected in multiple spots, which allow for obtaining subsamples but also reduces the effective mass to be analysed. Here a single orifice impactor shows better performance like the May impactor used in a number of SR-TXRF studies by the group of the KFKI Atomic Energy Research Institute in Budapest, Hungary.¹¹ Nonetheless, this device has a low cut off at 250 nm, which means the fine dust fraction cannot be adequately represented.

Calibration

Reliable calibration of the SR-XRF analysis is ambitious. References need to be well defined in both size and shape and generally of a microscopic spatial dimension to represent the sample adequately. Different procedures have been introduced. The most flexible approach is to use “Drop on Demand” (DOD) printing technology to generate custom made deposits in size, shape and elemental composition introduced by Fittschen et al.^10,20 The deposit may be used as either an internal standard or external reference. The elemental mass is determined by the volume spotted by the printing device. Figure 4 shows Auger electron maps and a scanning electron microscopy (SEM) image of dried Ni deposits on a Si wafer slide.

A modified approach was applied using additives to generate homogeneous films with a DOD-printer.²¹ The procedure allowed for custom made mass per area reference. Another approach is to use reference-free quantification established  by Beckhoff at the “Physikalisch-Technische Bundesanstalt” (PTB)^22,23 at the BESSYII beamline, and lithographic techniques²⁴ or sputtering to generate homogeneous or structured elemental deposits.²⁵ The latter has not only been used in elemental determination of aerosols, but has become a common diagnostic tool in SR-XRF.

Subsequent analysis with SR-TXRF-XANES

As mentioned previously; the quasi non-destructive XRF analysis of aerosols allows for further/prior characterisation of the sample. The collected aerosols may be digested and analysed using other analytical tools like inductively coupled plasma mass spectrometry ICP-MS or ICP-OES (ICP-optical emission spectrometry), which enable one to obtain additional information.²⁶

With respect to the elucidation of aerosol origin, source apportionment and alteration in the atmosphere chemical, speciation is of utmost importance. X-ray based techniques are excellently suited to study matter in its solid state assuming the sensitivity is sufficient. Micro-X-ray diffraction (XRD) analysis on single particles several micrometres in diameter has been applied by Lind et al.²⁷ but falls short for smaller particles. Electron probe micro analysis (EPMA) analysis has served very well in this regard and has been widely applied to identify air borne contaminant with respect to historical site conservation.²⁸

A different approach is to use X-ray absorption near edge fine structure analysis (XANES). Meirer et al.²⁹ have shown that speciation in terms of determining valence of arsenic in xylem sap of cucumber plants with concentrations of 30 ppb is feasible using TXRF-XANES.

The result shows that As(V) is converted to As(III) by the plant. In general those oxidation state conversions help detoxification. The element may be converted to a less toxic form or to more volatile compounds, which are emitted from the leaf. Such processes result in a decrease of toxic metal concentration in an organism.

With respect to aerosols the species of a metal may also give important information on elemental transport by air masses and transport conditions. Osán et al.³⁰ determined species of Zn and Cu in aerosols collected at different urban locations in Hungary. The species and size of the particle was used to draw a conclusion on the origin of the particle. They identified Zn(NO₃)₂ to be the main compound in the sub-micrometre fraction rather than ZnSO₄. Both compounds indicate Zn being associated to secondary aerosol particles. They found detection limits for TXRF-XANES to be as low as 50 pg Zn. An  amount of 60 pg Zn was found sufficient for a reasonable determination of the species, which for the May impactor corresponds to a concentration of 140 pg m^–3. These analyses were performed at beam line L at DORISIII at the DESY. Speciation on low Z elements is also feasible at beam lines dedicated to soft X-rays, e.g. Osán et al.³¹ studied nitrate and ammonia content in aerosols from the Antarctic at the Physikalisch-Technische Bundesanstalt (PTB) plane grating monochromator (PGM) beam line at BESSYII.

Besides source apportionment and health aspects, speciation of the oxidation state is important with regard to nutrient fluxes. Fe(II) as a micro nutrient plays an important role in air borne transport of nutrients to the ocean. Whereas Fe is abundant in freshwater reservoirs, it is a growth limitation factor in the ocean. This is because in the alkaline milieu of the ocean Fe hydroxide/oxides easily sediment. In contrast to Fe(III), Fe(II) is easily consumed by plankton.^32,33 Fittschen et al.¹³ show that using SR-TXRF XANES the oxidation state of Fe from atmospheric aerosols (sampled for 1 h collected in nine size fractions from 15 nm to 16 µm) could be determined for 200 pg Fe absolute, corresponding to 3 ng m^–3. Figure 5 shows the XANES spectra of selected Fe references and Figure 6 shows the XANES spectra of the aerosol particles. Fe(III) oxidation state was found for all size fractions.

Conclusions

SR-XRF, in micro-XRF as well as TXRF geometry, is a sensitive analytical tool allowing for elemental determination and speciation of aerosol particulates with limits of detection in the pg m^–3 range for many elements and collection times as low as 1 h. Hourly monitoring of PM10, 2.5 and ultrafine dust is feasible as well as high size fractionation of particles and particle cut off as low as 15 nm allowing one to solve different scientific questions.

Acknowledgements

The author thanks Magnus Menzel for proofreading and Markus Furger, Nicolas Bukowiecki, Peter Lienemann, Gerald Falkenberg and Janos Osan for discussions on the subject of SR-XRF aerosol characterisation on several occasions.

References

1. K. Katsouyanni, K. Katsouyanni, G. Touloumi, E. Samoli, A. Gryparis, A. Le Tertre, Y. Monopolis, G. Rossi, D. Zmirou, F. Ballester, A. Boumghar, H.R. Anderson, B. Wojtyniak, A. Paldy, R. Braunstein, J. Pekkanen, C. Schindler and J. Schwartz, “Confounding and effect modification in the short-term effects of ambient particles on total mortality: results from 29 European cities within the APHEA2 project”, Epidemiology 12, 521 (2001). doi: http://dx.doi.org/10.1097/00001648-200109000-00011
2. R. Swap, M. Garstang, S. Greco, R. Talbot and P. Kollberg, Tellus 44B, 133 (1994).
3. P. Carbo, M.D. Krom, W.B. Homoky, L.G. Benning and B. Herut, “Impact of atmospheric deposition on N and P geochemistry in the southeastern Levantine basin”, Deep Sea Research II 52, 3041 (2005). doi: http://dx.doi.org/10.1016/j.dsr2.2005.08.014
4. J.L. Jimenez, J.T. Jayne, Q. Shi, C.E. Kolb, D.R. Worsnop, I. Yourshaw, J.H. Seinfeld, R.C. Flagan, X. Zhang, K.A. Smith, J.W. Morris and P. Davidovits, “Ambient aerosol sampling using the Aerodyne Aerosol Mass Spectrometer”, J. Geophys. Res. 108, 8425 (2003). doi: http://dx.doi.org/10.1029/2001JD001213
5. K.-P. Hinz and B. Spengler, “Instrumentation, data evaluation and quantification in on-line aerosol mass spectrometry”, J. Mass Spectrom. 42(7), 843 (2007). doi: http://dx.doi.org/10.1002/jms.1262
6. C. Streli, G. Pepponi, P. Wobrauschek, C. Jokubonis, G. Falkenberg, G. Záray, J. Broekaert, U. Fittschen and B. Peschel, “Recent results of synchrotron radiation induced total reflection X-ray fluorescence analysis at HASYLAB, beamline L”, Spectrochim. Acta B 61, 1129 (2006). doi: http://dx.doi.org/10.1016/j.sab.2006.08.010
7. B. Schneider, “The determination of atmospheric trace metal concentrations by collection of aerosol particles on sample holders for total-reflection X-ray fluorescence”, Spectrochim. Acta B 44, 519 (1989). doi: http://dx.doi.org/10.1016/0584-8547(89)80059-X
8. R. Klockenkämper, H. Bayer, A. v. Bohlen, M. Schmeling and D. Klockow, “Collection of airborne particulate matter for a subsequent analysis by total reflection x-ray fluorescence”, Anal. Sci. 11, 495 (1995). doi: http://dx.doi.org/10.2116/analsci.11.495
9. M. Schmeling and D. Klockow, “Sample collection and preparation for analysis of airborne particulate matter by total reflection X-ray fluorescence spectrometry”, Anal. Chim. Acta 346, 121 (1997). doi: http://dx.doi.org/10.1016/S0003-2670(97)00022-6
10. U.E.A. Fittschen, S. Hauschild, M.A. Amberger, G. Lammel, C. Streli, S. Förster, P. Wobrauschek, C. Jokubonis, G. Pepponi, G. Falkenberg and J.A.C. Broekaert, “A new technique for the deposition of standard solutions in total reflection X-ray fluorescence spectrometry (TXRF) using pico-droplets generated by inkjet printers and its applicability for aerosol analysis with SR-TXRF”, Spectrochim. Acta B 61, 1098 (2006). doi: http://dx.doi.org/10.1016/j.sab.2006.09.009
11. V. Groma, J. Osan, A. Alsecz, S. Török, F. Meirer, C. Streli, P. Wobrauschek and G. Falkenberg, IDÕJÁRÁS—J. Hungarian Meteorological Service 112(2), 83 (2008).
12. F.N. Bukowiecki, M. Hill, R. Gehrig, C.N. Zwicky, P. Lienemann, F. Hegedus, G. Falkenberg, E. Weingartner and U. Baltensperger, “Trace metals in ambient air: hourly size-segregated mass concentrations determined by synchrotron-XRF”, Environ. Sci. & Technol. 39, 5754 (2005). doi: http://dx.doi.org/10.1021/es048089m
13. U.E.A. Fittschen, F. Meirer, C. Streli, P. Wobrauschek, J. Thiele, G. Falkenberg and G. Pepponi, “Characterization of atmospheric aerosols using synchroton radiation total reflection X-ray fluorescence and Fe K-edge total reflection X-ray fluorescence-X-ray absorption near-edge structure”, Spectrochim. Acta B 63, 1489 (2008). doi: http://dx.doi.org/10.1016/j.sab.2008.10.016
14. U.E.A. Fittschen, C. Streli, F. Meirer and M. Alfeld, “Determination of phosphorus and other elements in atmospheric aerosols using synchrotron total-reflection X-ray fluorescence”, X-Ray Spectrom. 42, 368 (2013). doi: http://dx.doi.org/10.1002/xrs.2457
15. W. Werner. Z. Umweltchem. Ökotox. 11(6), 343 (1999).
16. F. Andrade, C. Orsini and W. Maenhaut. “Relation between aerosol sources and meteorological parameters for inhalable atmospheric particles in São Paulo city, Brazil”, Atmos. Environ. 28, 2307 (1994). doi: http://dx.doi.org/10.1016/1352-2310(94)90484-7
17. D. Lundgren, “An aerosol sampler for determination of particulate concentration as a function of size and time”, J. Air Pollut. Control Assoc. 17, 225 (1967). doi: http://dx.doi.org/10.1080/00022470.1967.10468972
18. N. Bukowiecki, P. Lienemann, C.N. Zwicky, M. Furger, A. Richard, G. Falkenberg, K. Rickers, D. Grolimund, C. Borca, M. Hill, R. Gehrig and U. Baltensperger, “X-ray fluorescence spectrometry for high throughput analysis of atmospheric aerosol samples: The benefits of synchrotron X-rays”, Spectrochim. Acta B 63, 929 (2008). doi: http://dx.doi.org/10.1016/j.sab.2008.05.006
19. M. Furger, personal communication.
20. U.E.A. Fittschen, N.H. Bings, S. Hauschild, S. Förster, A. Kiera, E. Karavani, G. Falkenberg and J. Thiele, “Characteristics of dried residues of pico droplets suitable as reference in direct analysing techniques”, Anal. Chem. 80, 1967 (2008). doi: http://dx.doi.org/10.1021/ac702005x
21. M. Furger, A. Richard, N. Bukowiecki, P. Lienemann, K. Appel, U. Flechsig, A.S.H. Prevot and U. Baltensperger, “Trace elements in ambient aerosols analyzed and quantified with synchrotron radiation induced XRF”, Hasylab annual report (2010).
22. B. Beckhoff, “Reference-free X-ray spectrometry based on metrology using synchrotron radiation”, J. Anal. At. Spectrom. 23, 845 (2008). doi: http://dx.doi.org/10.1039/b718355k
23. F. Reinhardt, J. Osan, S. Török, A.E. Pap, M. Kolbe and B. Beckhoff, “Reference-free quantification of particle-like surface contaminations by grazing incidence X-ray fluorescence analysis”, J. Anal. At. Spectrom. 27, 248 (2012). doi: http://dx.doi.org/10.1039/c2ja10286b
24. J. Osán, F. Reinhardt, B. Beckhoff, A.E. Pap and S. Török, “Probing patterned wafer structures by means of grazing incidence X-ray fluorescence analysis”, ECS Trans. 25, 441 (2009).
25. M. Krämer, R. Dietsch, Th. Holz, D. Weißbach, G. Falkenberg, R. Simon, U. Fittschen, T. Krugmann, M. Kolbe, M. Müller and B. Beckhoff, “Ultrathin multi-element layer stacks—a new type of reference samples for µ-XRF and TXRF”, Adv. X-ray Anal. 54, 299 (2010).
26. M. Menzel and U.E.A. Fittschen, “TXRF analysis of airborne silver nanoparticles from fabrics”, Anal. Chem. (2014). doi: http://dx.doi.org/10.1021/ac404017u
27. O.C. Lind, B. Salbu, L. Skipperud, K. Janssens, J. Jaroszewicz and W. De Nolf, “Solid state speciation and potential bioavailability of depleted uranium particles from Kosovo and Kuwait”, J. Environ. Radioactivity 100, 301 (2009). doi: http://dx.doi.org/10.1016/j.jenvrad.2008.12.018
28. V. Kontozova-Deutsch, C. Cardell, M. Urosevic, E. Ruiz-Agudo, F. Deutsch and R. Van Grieken, “Characterization of indoor and outdoor atmospheric pollutants impacting architectural monuments: the case of San Jeronimo Monastery (Granada, Spain)”, Environ Earth Sci 63, 1433 (2011). doi: http://dx.doi.org/10.1007/s12665-010-0657-5
29. F. Meirer, G. Pepponi, C. Streli, P. Wobrauschek, V.G. Mihucz, G. Zaray, V. Czech, J.A.C. Broekaert, U.E.A. Fittschen and G. Falkenberg, X-Ray Spectrom. 36, 408 (2007). doi: http://dx.doi.org/10.1002/xrs.993
30. J. Osán, F. Meirer, V. Groma, S. Török, D. Ingerle, C. Streli and G. Pepponi, Spectrochim. Acta B 65, 1008 (2010).
31. J. Osán, S. Török, B. Beckhoff, G. Ulm, H. Hwang, C.-U. Ro, C. Abete and R. Fuoco, “Nitrogen and sulfur compounds in coastal Antarctic fine aerosol particles—an insight using non-destructive X-ray microanalytical methods”, Atmos. Environ. 40, 4691 (2006). doi: http://dx.doi.org/10.1016/j.atmosenv.2006.04.033
32. M. Andrea and P. Crutzen, “Atmospheric aerosols: biogeochemical sources and role in atmospheric chemistry”, Science 276, 1052 (1997). doi: http://dx.doi.org/10.1126/science.276.5315.1052
33. L. Spokes, T. Jickells and B. Lim, “Solubilisation of aerosol trace metals by cloud processing: a laboratory study”, Geochim. Cosmochim. Acta 58, 3281 (1994). doi: http://dx.doi.org/10.1016/0016-7037(94)90056-6