Chemical Speciation of Individual Airborne Particles by the Combined

Jul 30, 2010 - Spectral data processing was performed using the Perkin-Elmer SpectrumIMAGE software, where images were corrected for atmospheric influ...
0 downloads 12 Views 7MB Size
Download PDF
Anal. Chem. 2010, 82, 7987–7998

Chemical Speciation of Individual Airborne Particles by the Combined Use of Quantitative Energy-Dispersive Electron Probe X-ray Microanalysis and Attenuated Total Reflection Fourier Transform-Infrared Imaging Techniques† Young-Chul Song, JiYeon Ryu, Md Abdul Malek, Hae-Jin Jung, and Chul-Un Ro* Department of Chemistry, Inha University, 253, Yonghyun-dong, Nam-gu, Incheon 402-751, Korea In our previous work, it was demonstrated that the combined use of attenuated total reflectance (ATR) FTIR imaging and quantitative energy-dispersive electron probe X-ray microanalysis (ED-EPMA), named low-Z particle EPMA, had the potential for characterization of individual aerosol particles. Additionally, the speciation of individual mineral particles was performed on a single particle level by the combined use of the two techniques, demonstrating that simultaneous use of the two single particle analytical techniques is powerful for the detailed characterization of externally heterogeneous mineral particle samples and has great potential for characterization of atmospheric mineral dust aerosols. These single particle analytical techniques provide complementary information on the physicochemical characteristics of the same individual particles, such as low-Z particle EPMA on morphology and elemental concentrations and the ATRFT-IR imaging on molecular species, crystal structures, functional groups, and physical states. In this work, this analytical methodology was applied to characterize an atmospheric aerosol sample collected in Incheon, Korea. Overall, 118 individual particles were observed to be primarily NaNO3-containing, Ca- and/or Mg-containing, silicate, and carbonaceous particles, although internal mixing states of the individual particles proved complicated. This work demonstrates that more detailed physiochemical properties of individual airborne particles can be obtained using this approach than when either the low-Z particle EPMA or ATR-FT-IR imaging technique is used alone. Energy-dispersive electron probe X-ray microanalysis (EDEPMA), based on a scanning electron microscope (SEM) equipped with an ultrathin window energy-dispersive X-ray (EDX) detector, can be used to simultaneously detect the morphology and constituent elements, including low-Z elements such as C, N, and O (within a microscopic size volume), of individual particles. The characteristic X-ray lines of the low-Z elements undergo extremely † Part of the special issue “Atmospheric Analysis as Related to Climate Change”. * To whom correspondence should be addressed. Phone: +82 32 860 7676. Fax: +82 32 867 5604. E-mail: [email protected].

10.1021/ac1014113  2010 American Chemical Society Published on Web 07/30/2010

strong attenuation while propagating through the particle volume, therefore, the estimation of the matrix effect is important. Quantitative determination of low-Z elements was a necessary development for the study of atmospheric individual particles because these elements (C, N, O) are abundantly present in atmospheric particles and quantitative information is necessary for speciation of individual microscopic particles. Indeed, many environmental particles contain low-Z elements in the form of nitrates, sulfates, oxides, or mixtures including a carbon matrix. A quantitative EPMA technique, low-Z particle EPMA, based on a Monte Carlo calculation combined with reverse successive approximations, allows for quantitative determination of chemical elements, including low-Z elements.1-3 For a decade, the quantitative single particle analysis has been successfully used for the characterization of various types of atmospheric aerosol samples.4-8 Although the low-Z particle EPMA is powerful for characterization of airborne individual particles, its limited capability for unambiguous molecular speciation is sometimes disadvantageous. Fourier transform infrared spectroscopy (FT-IR) is a powerful technique for functional group analysis and molecular speciation of organic and inorganic chemical compounds. Thus far, in research pertaining to atmospheric aerosols, many studies regarding the characterization and quantification of functional groups in size-segregated aerosol samples have been performed using either transmission, reflectance, or attenuated total reflectance (ATR) FT-IR spectroscopic techniques.9-20 FT-IR spectroscopic imaging techniques can provide further information on the spatial (1) Ro, C.-U.; Osan, J.; Van Grieken, R. Anal. Chem. 1999, 71, 1521–1528. (2) Ro, C.-U.; Osan, J.; Szaloki, I.; de Hoog, J.; Worobiec, A.; Van Grieken, R. Anal. Chem. 2003, 75, 851–859. (3) Ro, C.-U.; Kim, H.; Van Grieken, R. Anal. Chem. 2004, 76, 1322–1327. (4) Ro, C.-U.; Kim, H.; Oh, K.-Y.; Yea, S. K.; Lee, C. B.; Jang, M.; Van Grieken, R. Environ. Sci. Technol. 2002, 36, 4770–4776. (5) Ro, C.-U.; Hwang, H.; Kim, H.; Chun, Y.; Van Grieken, R. Environ. Sci. Technol. 2005, 39, 1409–1419. (6) Kang, S.; Hwang, H.; Park, Y.; Kim, H.; Ro, C.-U. Environ. Sci. Technol. 2008, 42, 9051–9057. (7) Geng, H.; Park, Y.; Hwang, H.; Kang, S.; Ro, C.-U. Atmos. Chem. Phys. 2009, 9, 6933–6947. (8) Geng, H.; Ryu, J.; Jung, H.; Chung, H.; Ahn, K.; Ro, C.-U. Environ. Sci. Technol. 2010, 44, 2348–2353. (9) Coury, C.; Dillner, A. M. Atmos. Environ. 2009, 43, 940–948. (10) Coury, C.; Dillner, A. M. Atmos. Environ. 2008, 42, 5923–5932. (11) Hopey, J. A.; Fuller, K. A.; Krishnaswamy, V.; Bowdle, D.; Newchurch, M. J. Appl. Opt. 2008, 471, 2266–2274. (12) Maria, S. F.; Russell, L. M. Environ. Sci. Technol. 2005, 39, 4793–4800.

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

7987

distribution of chemical constituents within a sample, as the obtained IR image contains full spectral information at each pixel of the image corresponding to a unique spatial location on the sample. This FT-IR imaging technique has been applied in a wide variety of research fields.21-24 In atmospheric aerosol research, however, few studies regarding ambient aerosol or smog chamber particles using FT-IR imaging have been reported.25-28 In these studies, aerosol collection spots on a cascade impactor were analyzed and information on the average chemical species for agglomerated aerosol particles obtained. Because atmospheric particles are chemically and morphologically heterogeneous and the average composition and average aerodynamic diameter do not fully describe the population of the particles, FT-IR imaging stands to be a much more powerful technique for the characterization of aerosol particles if individual particles are to be analyzed on a single particle basis. Recently, attenuated total reflection FTIR (ATR-FT-IR) imaging instruments with a spatial resolution of 3.1 µm at 1726 cm-1 became commercially available.21,22,29 However, the spatial resolution of these ATR-FT-IR imaging instruments still remains inadequate for single particle analysis as environmentally relevant aerosol particles are of (sub)micrometer size. Their morphologies are also important for elucidating their sources and atmospheric reaction histories in the air.30,31 ATR-FT-IR signals can vary extensively according to chemical composition as linear absorption coefficients for different chemicals can differ by several orders of magnitude, resulting in the limited quantification capability. As an extreme example, NaCl particles are IR-inactive, and thus, ATR-FT-IR is blind for the identification of NaCl particles. (13) Maria, S. F.; Russell, L. M.; Turpin, B. J.; Porcja, R. J.; Campos, T. L.; Weber, R. J.; Huebert, B. J. J. Geophys. Res. 2003, 108 (D23), 8637, DOI: 10.1029/ 2003JD003703. (14) Maria, S. F.; Russell, L. M.; Turpin, B. J.; Porcja, R. J. Atmos. Environ. 2002, 36, 5185–5196. (15) Reff1, A.; Turpin, B. J.; Porcja, R. J.; Giovennetti, R.; Cui1, W.; Weisel, C. P.; Zhang, J.; Kwon, J.; Alimokhtari, S.; Morandi, M.; Stock, T.; Maberti, S.; Colome, S.; Winer, A.; Shendell, D.; Jones, J.; Farrar, C. Indoor Air 2005, 15, 53–61. (16) Blando, J. R.; Porcja, R. J.; Li, T.-H.; Bowman, D.; Lioy, P. J.; Turpin, B. J. Environ. Sci. Technol. 1998, 32, 604–613. (17) Shaka, H.; Saliba, N. A. Atmos. Environ. 2004, 38, 523–531. (18) Laurent, J.-P.; Allen, D. T. Aerosol Sci. Technol. 2004, 38 (S1), 82–91. (19) Garnes, L. A.; Allen, D. T. Aerosol Sci. Technol. 2002, 36, 983–992. (20) Allen, D. T.; Palen, E. J.; Haimov, M. I.; Hering, S. V.; Young, J. R. Aerosol Sci. Technol. 1994, 21, 325–342. (21) Van Dalen, G.; Heussen, P. C. M.; Den Adel, R.; Hoeve, R. B. J. Appl. Spectrosc. 2007, 61, 593–602. (22) Crane, N. J.; Bartick, E. G.; Perlman, R. S.; Huffman, S. J. Forensic Sci. 2007, 52, 48–53. (23) Chalmers, J. M.; Everall, N. J.; Schaeberle, M. D.; Levin, I. W.; Lewis, E. N.; Kidder, L. H.; Wilson, J.; Crocombe, R. Vib. Spectrosc. 2002, 30, 43–52. (24) Tungol, M. W.; Bartick, E. G.; Montaser, A. Appl. Spectrosc. 1990, 44, 543–549. (25) Kellner, R.; Malissa, H. Aerosol Sci. Technol. 1989, 10, 397–407. (26) Palen, E. J.; Allen, D. T.; Pandis, S. N.; Paulson, S.; Seinfeld, J. H.; Flagan, R. C. Atmos. Environ. 1993, 27A, 1471–1477. (27) Palen, E. J.; Allen, D. T.; Pandis, S. N.; Paulson, S.; Seinfeld, J. H. Atmos. Environ. 1992, 26A, 1239–1251. (28) Allen, D. T.; Palen, E. J. Aerosol Sci. 1989, 20, 441–455. (29) Perkin Elmer. Spatial Resolution in ATR-FT-IR imaging: measurement and interpretation. Technical Note, 2006. (30) Giere, R.; Blackford, M.; Smith, K. Environ. Sci. Technol. 2006, 40, 6235– 6240. (31) Vester, B. P.; Ebert, M.; Barnert, E. B.; Schneider, J.; Kandler, K.; Schu ¨ tz, L.; Weinbruch, S. Atmos. Environ. 2007, 41, 6102–6115.

7988

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

Our previous work32 demonstrated that the combined technique using ATR-FT-IR imaging and low-Z particle EPMA had the potential for characterization of individual aerosol particles. Furthermore, the speciation of individual mineral particles was performed on a single particle level by the combined use of the two techniques, demonstrating that the combined use of the two single particle analytical techniques is powerful for the detailed characterization of externally heterogeneous mineral particle samples and has a great potential for the characterization of mineral dust aerosols within the atmosphere.33 The single particle analytical techniques provide complementary information on the physicochemical characteristics of the same individual particles, such as low-Z particle EPMA for the morphology and elemental concentrations and the ATR-FT-IR imaging for molecular species, crystal structures, functional groups, and physical states. In this work, this analytical methodology was applied to characterize an atmospheric aerosol sample collected in Incheon, Korea, in order to investigate its practical application for characterization of complicated external mixtures of particles. Overall, 118 individual particles were observed to be mainly NaNO3-containing, Ca- and/ or Mg-containing, silicate, and carbonaceous particles, although internal mixing states of individual particles were complicated. This work demonstrates that more detailed physiochemical properties of individual airborne particles can be obtained than when either low-Z particle EPMA or an ATR-FT-IR imaging technique is used alone. EXPERIMENTAL SECTION Sample. To investigate the practical application of the combined use of low-Z particle EPMA and ATR-FT-IR imaging techniques for the detailed characterization of chemical species in airborne particles on a single particle level, a total of 118 individual particles in a stage sample collected on Al foil using a three-stage Dekati PM-10 cascade impactor were analyzed. The Dekati PM-10 impactor has, at a sampling flow of 10 L/min, aerodynamic cutoffs of 10, 2.5, and 1.0 µm for stages 1-3, respectively. A stage 2 sample (2.5-10 µm size range) was used in this work. The sampling was done just for 20 min, to avoid the collection of agglomerated particles, starting from 11 a.m. on October 13, 2008. The sampling was performed at the roof of a campus building of Inha University (15 m above ground level), located in Incheon, a densely populated (population, 2.7 million; area, 427 km2) Korean city with many different local emission sources, including 7 industrial complexes, 2 seaports with 10 wharfs, and 1 international airport. A new town (area, ∼33 km2) has been under construction with major building activity and heavy transport. Incheon is on the coast of the Yellow Sea and adjacent to the megacity of Seoul (population, 10.3 million; area, 605 km2). The sampling site was regarded as being susceptible to various urban source processes, including high area traffic loads. Therefore, this sample is expected to be a complicated heterogeneous mixture of urban particles that can be used for investigation of the practical applicability of the combined use of the low-Z particle EPMA and ATR-FT-IR imaging techniques. The collected sample was put in plastic carriers, sealed, and (32) Ryu, J.; Ro, C.-U. Anal. Chem. 2009, 81, 6695–6707. (33) Jung, H.-J.; Malek, Md. A.; Ryu, J.; Kim, B.; Song, Y.-C.; Kim, H.; Ro, C.-U. Anal. Chem. 2010, 82, 6193–6202.

stored in desiccators before low-Z particle EPMA and ATRFT-IR imaging measurements. Low-Z Particle EPMA. To obtain the morphological and elemental compositional information, a quantitative ED-EPMA single particle analysis, low-Z particle EPMA, was applied prior to ATR-FT-IR imaging measurements. The measurements were carried out using a Jeol JSM-6390 SEM equipped with an Oxford Link SATW ultrathin window EDX detector. The resolution of the detector was 133 eV for Mn-KR X-rays. The X-ray spectra were recorded under the control of Oxford INCA Energy software. To achieve optimal experimental conditions, such as a low background level in the spectra and high sensitivity for low-Z element analysis, an accelerating voltage of 10 kV was selected. The beam current was 1.0 nA for all measurements. A more detailed discussion of the measurement conditions is given elsewhere.1 The net X-ray intensities of the elements were obtained by nonlinear least-squares fitting of the collected spectra using the AXIL program.34 The elemental concentrations of the individual particles were determined from their X-ray intensities by application of a Monte Carlo calculation combined with reverse successive approximations.2,3 The quantification procedure provided results accurate within 12% relative deviations between the calculated and nominal elemental concentrations when the method was applied to various types of standard particles, such as NaCl, Al2O3, CaSO4 · 2H2O, Fe2O3, CaCO3, and KNO3.35,36 The low-Z particle EPMA method can provide quantitative information on the chemical composition, while the particles can be classified based upon their chemical species. ATR-FT-IR Imaging Technique. All ATR-FT-IR imaging measurements were performed using a Perkin-Elmer Spectrum 100 FT-IR spectrometer interfaced to a Spectrum Spotlight 400 FT-IR microscope. For ATR imaging, an ATR accessory employing a germanium hemispherical internal reflection element (IRE) crystal with a diameter of 600 µm was used. The ATR accessory was mounted on the X-Y stage of the FT-IR microscope, with the IRE crystal making contact with the sample via a force lever. The ultimate spatial resolution of the IR imaging is approximately equal to the wavelength of the incident IR radiation. However, the hemispherical IRE crystal acts like a lens, condensing the IR beam when it strikes the IRE. The extent of the condensation is proportional to the refractive index of the IRE material (4.0 for germanium).37 Thus, a spatial resolution of 3.1 µm at 1726 cm-1 (5.79 µm) was achieved beyond the ultimate spatial resolution limit. A 16 × 1 pixel mercury cadmium telluride (MCT) array detector was used to obtain FT-IR images with a pixel size of 1.56 µm. For each pixel, an ATR-FT-IR spectrum, ranging from 720 to 4000 cm-1 with a spectral resolution of 8 cm-1, was obtained from four interferograms that were coadded and Fourier-transformed. The position of the crystal on the sample was determined using a visible optical microscope equipped with a light-emitting diode and a charge-coupled device camera. Additionally, the optical image was used to identify the same (34) Vekemans, B.; Janssens, K.; Vincze, L.; Adams, F.; Van Espen, P. X-Ray Spectrom. 1994, 23, 278–285. (35) Ro, C.-U.; Osan, J.; Szaloki, I.; Oh, K.-Y.; Kim, H.; Van Grieken, R. Environ. Sci. Technol. 2000, 34, 3023–3030. (36) Ro, C.-U.; Oh, K.-Y.; Kim, H.; Chun, Y.-S.; Osan, J.; de Hoog, J.; Van Grieken, R. Atmos. Environ. 2001, 35, 4995–5005. (37) Griffiths, P. R.; de Haseth, J. A. Fourier Transform Infrared Spectrometry, 2nd ed.; John Wiley & Sons, Inc.: Hoboken, NJ, 2007.

single particles that were analyzed using low-Z particle EPMA before the ATR-FT-IR imaging measurement, as the optical microscopy provides an image of sufficient spatial resolution to help locate the same image field observed by low-Z particle EPMA. Spectral data processing was performed using the Perkin-Elmer SpectrumIMAGE software, where images were corrected for atmospheric influences and the spectral information at each pixel was extracted by application of principal component analysis (PCA). RESULTS AND DISCUSSION In Figure 1 are shown the secondary electron image (SEI) obtained prior to ATR-FT-IR imaging measurement (Figure 1A), the visible light optical image (Figure 1B), the ATR-FT-IR image (Figure 1C), and the SEI after ATR-FT-IR imaging measurement (Figure 1D) of the same 118 individual airborne particles on Al foil. The image field containing 118 particles was chosen as different types of particles were present and well separated from each other. First, the morphologies and chemical compositions of all the particles on the image field were obtained by low-Z particle EPMA. In order to locate the same image field for the ATR-FT-IR imaging measurements, visible light optical microcopy was employed as it can provide an image of sufficient spatial resolution to help locate the same image field observed by low-Z particle EPMA (Figure 1B). After the location of the image field, the sample was made in contact with the IRE crystal for the ATRFT-IR imaging measurement. The ATR-FT-IR image (Figure 1C) was obtained by the application of principle component analysis (PCA) after the first differentiation of original ATR-FT-IR spectra at all of the pixels in the image. Although the qualities of the SEI and ATR-FT-IR image differ due to the inherently different spatial resolutions of the images (∼100 nm for the SEI and 3.1 µm for the ATR-FT-IR image), the same patterns of particle location among the images ensured that the same particles of micrometer size were seen. Their equivalent diameters ranged from 1.6 to 13.4 µm (determined from the SEI obtained before ATR-FT-IR imaging measurement), where the equivalent diameter was calculated by assuming that a particle with the same area for a particle on the SEI was circular. The area of each image is approximately 400 × 300 µm2. Considering that the pixel size of the ATR-FT-IR image is 1.56 × 1.56 µm2, the number of pixels for the shown ATR-FT-IR image is ∼4 900. All of the pixels in the image contain full IR spectra, ranging from 4000 to 720 cm-1. Although the experimental ATR-FT-IR imaging data were obtained on a 1.56 × 1.56 µm2 pixel size, the manufacture’s software interpolates ATR-FT-IR imaging pixel data onto a display image with many more pixels, such that the final display image looks better than the actual image. In the PCA analysis, the ATR-FT-IR spectral data for all of the pixels in the image were considered, and similar spectra with significant encountering frequencies were grouped as a principle component. The ATR-FT-IR image indicates that four types of particles (four principle components) are dominant: the NaNO3containing particles (displayed in pink; particle nos. 17, 51, 64, 69, 73, 81, 97, and 111); the Ca- and/or Mg-containing particles (displayed in brown; particle nos. 1, 10, 18, 33, 35, and 42); the silicate mineral particles (displayed in green; particle nos. 2, 3, 7, 12, 20, 22, and 41); the mixture of particles of silicates and Ca-containing species (yellowish colored; particle nos. 4, Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

7989

Figure 1. (A) Secondary electron image (SEI) before the ATR-FT-IR imaging measurement; (B) visible light optical image; (C) ATR-FT-IR image; and (D) SEI after the ATR-FT-IR imaging measurement of the same 118 individual airborne particles on Al foil.

28, and 78). Some particles present in the SEIs are not shown in the ATR-FT-IR image (particle nos. 8, 9, 14, and 40, to name a few). In that case, their FT-IR spectra of the missing particles in the ATR-FT-IR image were extracted from the raw ATR-FTIR image data at the locations known from the SEIs. The SEIs (Figure 1A,D) clearly show morphologies and locations of the 118 particles before and after the ATR-FT-IR imaging measurements. For the ATR-FT-IR imaging measurements, the sample had to be in contact with the IRE crystal so that some force was applied to the sample during the contact. When good contact was made and the sample was well pressed against the IRE crystal, particles were (partly or fully) embedded into the ductile Al collecting foil. Some nonconductive mineral particles sitting on the Al foil (particle nos. 4, 15, and 19) appear bright because of high secondary and backscattered electron yields of insulating particles (Figure 1A), whereas the particles on the SEI taken after FT-IR measurements appear dark (Figure 1D) as electrons can flow from the embedded particles into the metallic foil, resulting in low secondary and backscattered electron yields.38 With comparison of parts A and D of Figure 1, it is clear that most particles spread out and some particles were reoriented and/or broken off into small parts as the force by the IRE crystal was applied on the particles. In addition to the inherently different spatial resolutions for the SEI and ATR-FTIR image, modification of the particle shape owing to the contact with the IRE crystal aided broader ATR-FT-IR images of the individual particles. The SEIs before and after ATR-FT-IR imaging measurement (38) Goldstein, J. I.; Newbury, D. E.; Joy, D. C.; Lyman, C.; Echlin, P.; Lifshin, E.; Sawyer, L.; Michael, J. Scanning Electron Microscopy and X-ray Microanalysis, 3rd ed.; Kluwer-Plenum: New York, 2003.

7990

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

show the original morphology of individual particles and the modified morphology seen during the ATR-FT-IR imaging measurement, respectively. On the basis of morphological, X-ray spectral, and ATR-FT-IR spectral data, 118 individual particles were classified into different particle types, such as NaNO3-containing, Ca- and/or Mg-containing, silicate, and carbonaceous particles. In Table S1 of the Supporting Information are shown the analysis results for all 118 particles determined by the combined use of the two techniques. The elemental concentrations and chemical species of the particles obtained from the low-Z particle EPMA data are listed in the third and fourth columns of Table S1 in the Supporting Information, respectively. Details pertaining to the low-Z particle EPMA data treatment and interpretation can be found in previous papers by the authors.6–8 Herein is a brief summary of how the particles were classified. The low-Z particle EPMA can provide quantitative information on the chemical composition. First, particles were regarded to be composed of only one chemical species when the chemical species constituted at least 90% of the atomic fraction. Second, efforts were made to specify chemical species, even for particles internally mixed with two or more chemical species. The mixed particles were specified on the basis of all of the chemical species with >10% in the formula fraction. Third, it is known that ED-EPMA has high detection limits of 0.1-1.0% in weight mainly due to its high Bremsstrahlung background level. Thus, elements with less than 1.0% of atomic concentration were not included in the chemical speciation procedure. All ATR-FT-IR absorption peaks and chemical species obtained from the ATR-FT-IR spectral data are listed in the fifth and sixth

Figure 2. X-ray spectra of typical NaNO3-containing particles.

columns of Table S1 in the Supporting Information, respectively. As it is difficult to quantitatively assess molecular species of individual particles using ATR-FT-IR spectral data when they are internal mixtures, the ATR-FT-IR spectral data were used for qualitative molecular speciation and/or functional group analysis. To specify the molecular species of individual particles using ATR-FT-IR spectral data, a homemade ATR-FT-IR spectral library was initially consulted; inorganic compounds and minerals commonly encountered in airborne particle samples were measured using ATR-FT-IR techniques to generate a custom ATR-FT-IR library. When the ATR-FTIR spectrum of an individual particle was too complex to clearly specify all of the chemical species of the particle, the IR absorption peaks were examined for functional group analysis based on the IR peak data available in the literature.10,11,13–20 In the seventh column of Table S1 in the Supporting Information are listed the chemical species identified by the combined use of the two single particle analytical techniques, which are particle types used for the discussion hereafter. Figures S1-S9 of the Supporting Information are SEIs obtained in higher magnifications (×650 to ×1100) than an overall SEI (Figure 1A with ×300), where more detailed morphologies of all the 118 particles are shown together with their particle types. For the notation of a particle type, a unique notation system was devised; a particle notated as (Ca, Mg)(NO3, SO4) is an internal mixture of Ca(NO3)2, CaSO4, Mg(NO3)2, and MgSO4 species. A particle notated as montmorillonite/organic/NaNO3/H2O indicates that the particle is a mixture of montmorillonite, organic species, and NaNO3, with the most abundant chemical species listed first. However, water present in the particle is placed last just to indicate its presence.

NaNO3-Containing Particles. Among the 118 particles analyzed, the number of NaNO3-containing particles is 13, with no observation of primary genuine sea salt particles. The NaNO3-containing particles are reaction products of primary sea salt particles and HNO3.39 All of the NaNO3-containing particles (see particle nos. 51, 69, 81, and 82 in Figure S1 of the Supporting Information) are in crystalline form as they appear bright and crystalline. X-ray spectral and elemental concentration data of the NaNO3-containing particles are shown in Figure 2, indicating that they are pure NaNO3 particles, with exception to particle no. 69. ATR-FT-IR spectra of the NaNO3-containing particles are shown in Figure 3. The strong ATR-FT-IR absorption peaks observed at 1345 and 832 cm-1 indicate NO3-. The weak peak at ∼1787 cm-1 is characteristic for the crystalline NaNO3 species (see ATR-FT-IR spectra of the bulk NaNO3 powders (Aldrich) and aqueous NaNO3 solution, shown in a left-hand inset of Figure 3.) Also, the absence of absorption peaks of H2O at ∼3370 and 1635 cm-1 indicates the NaNO3-containing particles are crystalline. Particle nos. 51, 69, 81, and 82 have needle- or rodlike features around the central NaNO3-containing particles (Figure S1 in the Supporting Information). The CaSO4 species was reported to crystallize in a needlelike form when sea salt particles become crystallized.40 Indeed, the peak at ∼1130 cm-1 from SO42- is present in the ATR-FT-IR spectra of (39) Laskin, A.; Iedema, M. J.; Cowin, J. P. Environ. Sci. Technol. 2002, 36, 4948–4955. (40) Andreae, M. O.; Charlson, R. J.; Bruynseels, F.; Storms, H.; Van Grieken, R.; Maenhaut, W. Science 1986, 132, 1620–1622.

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

7991

Figure 3. ATR-FT-IR spectra of typical NaNO3-containing particles. Data of the 2200-2530 cm-1 region where atmospheric CO2 peaks are present and were deleted for clarity.

the NaNO3-containing particles (Figure 3). As the X-ray spectra were obtained just on the central NaNO3 particles, only the NaNO3 species was detected using low-Z particle EPMA, whereas ATR-FT-IR signals were obtained from large areas (areas of ∼5.8 µm diameter33), including the central NaNO3 and surrounding particles. The SEI of particle no. 51, obtained after ATR-FT-IR imaging measurement, is shown in Figure S10A of the Supporting Information. Particle no. 51 was shattered into many pieces of small crystalline particles when it came into contact with the IRE crystal for ATR-FT-IR imaging measurement, indicating that the NaNO3-containing particle was polycrystalline. The X-ray spectra obtained from several small particles also indicate that particle no. 51 is polycrystalline with different chemical forms (Figure S10B-F); the chemical species of region no. 1 in Figure S10A is (Ca, Na, Mg)(SO4, NO3)/C with major CaSO4, based on its elemental concentration data, whereas the chemical species of region nos. 2-5 are NaNO3/ C, (Na, Mg)(NO3, SO4, Cl), (Na, Mg)NO3, and C/aluminosilicate/(Na, Mg)(NO3, SO4, Cl), respectively. Some of the shattered particles also contain carbonaceous species according to their X-ray spectral data. A broad ATR-FT-IR absorption peak at ∼1600 cm-1 (Figure 3) is from the carbonaceous species, due likely to the organic functional groups of the carboxylate anion.41,42 All the NaNO3-containing particles show additional peaks at ∼1260 cm-1, arising from NO2-. In the right-hand inset of (41) Hay, M. B.; Myneni, S. C. B. Geochim. Cosmochim. Acta 2007, 71, 3518– 3532. (42) Cabaniss, S. E.; Leenheer, J. A.; McVey, I. F. Spectrochim. Acta 1998, A54, 449–458.

7992

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

Figure 3, the ATR-FT-IR spectra of the aqueous NaNO2 solution and a mixture solution of NaNO3 and NaNO2 confirm that the peak at ∼1260 cm-1 is from NaNO2. Further ATR-FT-IR measurements were performed for several mixture solutions of NaNO3 and NaNO2 in different proportions, with the ATRFT-IR spectrum of a 4 (NaNO3):1 (NaNO2) mixture resembling those obtained from the NaNO3-containing single particles, indicating the significant presence of the NaNO2 species. Some studies have reported the formation of NaNO3 from sea salt particles,7,8,39 and yet this is the first observation on the formation of NaNO2 from sea salt particles on a single particle basis. The HONO, ClNO2, and HNO3 might be involved in the formation of NaNO2,43,44 implying NaNO2 as a new sink for HONO. The photochemical nitrite formation from concentrated aqueous nitrate solution45 could also be a potential mechanism for NaNO2 formation. As the relative humidity (RH) during the sampling was ∼56% and the deliquescence and efflorescence RHs (DRH and ERH, respectively) of NaNO3 were reported to be 74.5 and 35%, respectively,46 the NaNO3-containing particles must be crystallized somewhere else prior to coming to the sampling site. (43) Thornton, J. A.; Kercher, J. P.; Riedel, T. P.; Wagner, N. L.; Cozic, J.; Holloway, J. S.; Dube, W. P.; Wolfe, G. M.; Quinn, P. K.; Middlebrook, A. M.; Alexander, B. Nature 2010, 464, 271–274. (44) Ziemba, L. D.; Dibb, J. E.; Griffin, R. J.; Anderson, C. H.; Whitlow, S. I.; Lefer, B. L.; Rappengluck, B.; Flynn, J. Atmos. Environ. 2009, DOI: 10.1016/j.atmosenv.2008.12.024. (45) Roca, M.; Zahardis, J.; Bone, J.; El-Maazawi, M.; Grassian, V. H. J. Phys. Chem. A 2008, 112, 13275–13281. (46) Tang, I. N.; Fung, K. H. J. Chem. Phys. 1997, 106, 1653–1660.

Figure 4. X-ray spectra of typical Ca- and/or Mg-containing particles.

However, as discussed below, it is worth noting that the Caand/or Mg-containing nitrate particles were collected as water droplets. Ca- and/or Mg-Containing Particles. Among the 118 particles, the number of Ca- and/or Mg-containing particles is 50. As shown in Table S1 in the Supporting Information, the numbers of Ca-, (Ca,Mg)-, and Mg-containing particles are 35, 9, and 6, respectively. Among the 35 Ca-containing particles, just two particles are CaCO3. The others are reacted (aged) Ca-containing particles, all of which contain Ca(NO3)2. Unlike NaNO3containing particles collected in the crystalline form, nearly all the Ca- and Mg-containing particles (see particle nos. 45, 46, 47, 68, and 87 in Figure S1 in the Supporting Information) were collected as water droplets because they appear circular and dark on their SEIs. Typical X-ray and ATR-FT-IR spectra of the Ca- and/ or Mg-containing particles (particle nos. 28, 46, 47, and 91) are shown in Figures 4 and 5, respectively. The presence of the ATRFT-IR absorption peaks of H2O between 3320 and 3440 cm-1 and 1620-1630 cm-1 also indicates that the Ca- and/or Mgcontaining particles contain water and that they were collected as water droplets, except particle no. 28, which contains H2O but looks bright and angular on its SEI (see particle no. 28 in Figure S2 in the Supporting Information). It was reported that DRHs of Ca(NO3)2 and Mg(NO3)2 were 18 and 53%, respectively; they had no ERHs.46,47 Therefore, the observation that Ca- and/ or Mg-containing nitrate particles were collected as water droplets can be explained by their hygroscopic properties. The capability of ATR-FT-IR imaging and low-Z particle EPMA techniques to (47) Mikhailov, E.; Vlasenko, S.; Martin, S. T.; Koop, T.; Poschl, U. Atmos. Chem. Phys. 2009, 9, 9491–9522.

identify the physical phase of individual aerosol particles is highly valuable as the phase of airborne particles is one of the most important physical properties of aerosols, given their different contributions to the radiative forcing of aerosols with the same chemical compositions but in different phases. Particle no. 28 is a (Ca, Mg)(CO3, NO3)/(montmorillonite, quartz)/organic/Cl/H2O particle based on its low-Z particle EPMA and ATR-FT-IR spectral data. As shown in Figure 4A, chemical elements of particle no. 28 are O (58.7% in atomic fraction), Ca (14.7%), C (10.6%), Si (5.1%), Al (3.5%), Mg (3.4%), N (3.3%), and Cl (0.5%). The ATR-FT-IR spectrum of particle no. 28 indicates that Al and Si are for montmorillonite ((Na, Ca)0.33(Al, Mg)2Si4O10(OH)2 · nH2O) and quartz (SiO2) minerals, according to their characteristic absorption peaks at 1006 and 912 cm-1 (montmorillonite) and 1042, 796, and 778 cm-1 (quartz). The observation of absorption peaks at 1356 and 817 cm-1 indicates the presence of the NO3- species, for which the counter cations are most probably Ca2+ and/or Mg2+ (see ATR-FT-IR spectra of aqueous Ca(NO3)2 and Mg(NO3)2 solutions shown in the insets of Figure 5). EPMA detects the Cl element, and the chlorine can exist as Cl- and/or as an impurity element in montmorillonite or quartz. However, it is quite rare to encounter the Cl element in airborne silicate particles but very common to encounter Cl- in particles containing Mg2+ and NO3-. Absorption peaks at 1400 and 871 cm-1 are from CO32(see the spectrum of the bulk CaCO3 powders shown in the left-hand inset of Figure 5). As the peak at 1597 cm-1 is for organic species, carbonaceous species exist as CO32- and organic species. In the ATR-FT-IR spectrum, the CO32- peak at 1400 cm-1 is much stronger than the organic peak at 1597 Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

7993

Figure 5. ATR-FT-IR spectra of typical Ca- and/or Mg-containing particles. Data of the 2200-2530 cm-1 region where atmospheric CO2 peaks are present and were deleted for clarity.

cm-1; however, it is difficult to estimate the relative content of CO32- and the organic species based on the ATR-FT-IR spectral data as linear absorption coefficients of different chemical species can possibly vary by several orders of magnitude. Our future work will include the quantitative aspects of the ATRFT-IR imaging technique for chemical species commonly observed in airborne particles. On the basis of the atomic concentration data from low-Z particle EPMA, the carbon content (10.6%) is much larger than the N content (3.3%), implying that CO32- is more abundant than NO3-. The relative intensities of the ATR-FT-IR CO32- and NO3- peaks at 1400 and 1356 cm-1, respectively, are difficult to be accurately estimated, but the peak of CO32- at 871 cm-1 is stronger than that of NO3- at 817 cm-1, indicating that CO32- is probably more abundant than NO3-. As the (Ca, Mg)CO3 and silicate minerals are major species of particle no. 28, the particle of soil origin looks bright and angular on the SEI. However, according to the ATR-FT-IR spectral data, the particle also contains minor nitrate, organics (probably humic substance), and water, clearly indicating partial nitrate formation from the reaction between carbonate and HNO3 in or on this particle. Particle no. 46 appears circular, somewhat bright at a central part, and dark in the surrounding area (Figure S1 in the Supporting Information), indicating that it was collected as a water droplet with the major chemical species at the central and surrounding areas being different. As shown in Figure 4B, the chemical elements of particle no. 46 are O (63.9% in atomic fraction), Mg (10.6%), C (8.9%), N (8.4%), S (3.9%), Ca (2.5%), Cl (1.6%), and Si (0.3%). The NO3-, CO32-, and SO42- are major 7994

Analytical Chemistry, Vol. 82, No. 19, October 1, 2010

species based on the ATR-FT-IR spectrum (Figure 5). Intensities of the NO3- peaks at 1346 and 820 cm-1 are comparable with those of the CO32- peaks at 1420 and 855 cm-1, implying similar contents of the NO3- and CO32- species. This is supported by the similar elemental C and N contents (8.9 and 8.4%, respectively). Carbon is mostly in the form of CO32-, as no organic peak is observed at ∼1600 cm-1. The counter cations for those anions are most probably Mg2+ and Ca2+, with the Mg species more abundant than the Ca species based on the elemental concentration data (Mg, 10.6% and Ca, 2.5%). The peak at 1050 cm-1 is from a silicate mineral; however, it is difficult to assign the specific silicate mineral type due to its very low signal intensity. The ATR-FT-IR peaks of the water at 3378 and 1630 cm-1 are very strong, confirming that particle no. 46 was collected as a water droplet. The bright central part is for a carbonate and a silicate mineral and the dark area for nitrate, sulfate, and chloride. Particle no. 46 is a (Mg, Ca)(CO3, NO3, SO4)/Cl/silicate/H2O particle, based on its low-Z EPMA and ATR-FT-IR spectral data, probably formed by coagulation of a particle of marine origin and that of soil origin. Particle no. 47 is a (Ca, Mg)(CO3, NO3, NO2, SO4)/organic/ Cl/H2O particle based on its low-Z EPMA and ATR-FT-IR spectral data (see Figures 4C and 5). Nitrate is more abundant than carbonate (see elemental concentration). The conversion of nitrate and sulfate from carbonate proceeded extensively, resulting in its circular and dark morphology on the SEI (Figure S1 in the Supporting Information). Particle no. 91 is a Mg(CO3, NO3)/Cl/ organic/H2O particle, looking circular and completely dark (see Figure S6 in the in the Supporting Information and Figures 4D

and 5). As discussed for the four exemplar Ca- and/or Mgcontaining particles, the Ca- and/or Mg-containing particles are reacted (aged, secondary) particles converted from carbonates, all of which contain nitrate and some also sulfate. The carbonates are mostly of soil origin, but the presence of Cl in the Ca- and/or Mg-containing particles cannot rule out that some portions of the particles are of marine origin. Among the overall 50 Ca- and/or Mg-containing particles, the chemical composition of any two particles are not identical, based on the low-Z particle EPMA and ATR-FT-IR spectral data. This kind of detailed characterization of individual airborne particles demonstrates the great potential of the combined application of the two single-particle analytical techniques. Silicate Particles. Airborne mineral dust particles, naturally borne of soils, are the most abundant particulate matter in coarse atmospheric aerosols; the recent recognition of multiple roles of mineral dust particles in atmospheric processes has made research into mineral dust a central topic in environmental studies.48,49 Minerals are classified into different types according to their chemical composition, such as silicates, carbonates, oxides, sulfates, and phosphates.50 Silicate minerals, which include quartz, feldspar, pyroxene, olivine, mica, and clay minerals, are the most abundant minerals, constituting ∼90% of the crust of the earth. Particles with