Natural Abundance 13C and 14C Analysis of Water-Soluble Organic

WSOC scatters sunlight(9, 10) and thus has a direct effect on climate by cooling the ... for a small set of samples from the Southeast and Northeast U...
0 downloads 0 Views 217KB Size
Download PDF
Anal. Chem. 2010, 82, 7973–7978

Natural Abundance 13C and 14C Analysis of Water-Soluble Organic Carbon in Atmospheric Aerosols† ¨ rjan Gustafsson*,‡ Elena N. Kirillova,‡ Rebecca J. Sheesley,‡,§ August Andersson,‡ and O Department of Applied Environmental Science (ITM) and Bert Bolin Climate Research Centre, Stockholm University, Sweden, and Department of Environmental Science, Baylor University, Waco, Texas 76798 Water-soluble organic carbon (WSOC) constitutes a large fraction of climate-forcing organic aerosols in the atmosphere, yet the sources of WSOC are poorly constrained. A method was developed to measure the stable carbon isotope (δ13C) and radiocarbon (∆14C) composition of WSOC for apportionment between fossil fuel and different biogenic sources. Synthetic WSOC test substances and ambient aerosols were employed to investigate the effect of both modern and fossil carbon contamination and any method-induced isotope fractionation. The method includes extraction of aerosols collected on quartz filters followed by purification and preparation for off-line δ13C and ∆14C determination. The preparative freeze-drying step for isotope analysis yielded recoveries of only ∼70% for ambient aerosols and WSOC probes. However, the δ13C of the WSOC isolates were in agreement with the δ13C of the unprocessed starting material, even for the volatile oxalic acid probe (6.59 ( 0.37‰ vs 6.33 ( 0.31‰; 2 sd). A 14C-fossil phthalic acid WSOC probe returned a fraction modern biomass of 0.999, indicating the ∆14C-WSOC method to be free of both fossil and contemporary carbon contamination. Application of the δ13C/∆14C-WSOC method to source apportion climate-affecting aerosols was illustrated be constraining that WSOC in ambient Stockholm aerosols were 88% of contemporary biogenic C3 plant origin. Carbonaceous aerosols constitute one of the largest uncertainties in the assessment of radiative climate forcing.1,2 These air † Part of the special issue “Atmospheric Analysis as Related to Climate Change”. * To whom correspondence should be addressed. Phone: +46 70-3247317. E-mail: [email protected]. ‡ Stockholm University. § Baylor University. (1) Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K. B., Tignor, M., Miller, H. L., Eds.; Cambridge University Press: Cambridge, U.K., 2007. (2) Kiehl, J. T. Geophys. Res. Lett. 2007, 34, L22710, DOI: 10.1029/ 2007GL031383.

10.1021/ac1014436  2010 American Chemical Society Published on Web 09/15/2010

particles also have serious impacts on human respiratory health.3,4 Unfortunately, the relative contribution of different sources to various components of carbonaceous aerosols in the atmosphere is poorly constrained. The water-soluble organic carbon (WSOC) component frequently makes up 20-70% of the total organic aerosol (TOA).5-8 The high abundance and hydrophilic character of WSOC result in both direct and indirect influences on the radiative forcing of the aerosol. WSOC scatters sunlight9,10 and thus has a direct effect on climate by cooling the earth surface. WSOC also influences the hygroscopicity and surface tension of aerosols and thereby have an indirect aerosol effect by affecting the ability of particles to act as cloud condensation nuclei (CCN) and thus of the cloud albedo.11-13 Hence, analysis of WSOC in the atmosphere is important for aerosol-climate studies. Analysis of WSOC properties diagnostic of its sources is warranted both to increase our understanding of the contribution from various natural and anthropogenic sources and to inform decisions on mitigation of anthropogenic aerosol sources. WSOC stem from oxidative processes of biogenic volatile organic compounds but also from poorly constrained contributions of primary organic aerosols emitted by incomplete combustion of biomass and fossil fuel.7,14,15 The carbon isotopic signals are (3) Downs, S. H.; Schindler, C.; Liu, L. J. S.; Keidel, D.; Bayer-Oglesby, L.; Brutsche, M. H.; Gerbase, M. W.; Keller, R.; Kunzli, N.; Leuenberger, P.; Probst-Hensch, N. M.; Tschopp, J. M.; Zellweger, J. P.; Rochat, T.; Schwartz, J.; Ackermann-Liebrich, U. N. Engl. J. Med. 2007, 357, 2338–2347. (4) Nel, A. Science 2005, 308, 804–806. (5) Saxena, P.; Hildemann, L. J. Atmos. Chem. 1996, 24, 57–109. (6) Decesari, S.; Facchini, M. C.; Matta, E.; Lettini, F.; Mircea, M.; Fuzzi, S.; Tagliavini, E.; Putaud, J. P. Atmos. Environ. 2001, 35, 3691–3699. (7) Mayol-Bracero, O. L.; Guyon, P.; Graham, B.; Roberts, G.; Andreae, M. O.; Decesari, S.; Facchini, M. C.; Fuzzi, S.; Artaxo, P. J. Geophys. Res. 2002, 107, 8091, DOI: 10.1029/2001JD000522. (8) Yu, J. Z.; Yang, H.; Zhang, H. Y.; Lau, A. K. H. Atmos. Environ. 2004, 38, 1061–1071. (9) Mladenov, N.; Reche, I.; Olmo, F. J.; Lyamani, H.; Alados-Arboledas, L. J. Geophys. Res. 2010, 115, G00F11, DOI: 10.1029/2009JG000991. (10) Kim, J.; Yoon, S. C.; Kim, S. W.; Brechtel, F.; Jefferson, A.; Dutton, E. G.; Bower, K. N.; Cliff, S.; Schauer, J. J. Atmos. Environ. 2006, 40, 6718–6729. (11) Miyazaki, Y.; Kondo, Y.; Han, S.; Koike, M.; Kodama, D.; Komazaki, Y.; Tanimoto, H.; Matsueda, H. J. Geophys. Res. 2007, 112, D22S30, DOI: 10.1029/2007JD009116. (12) Facchini, M. C.; Mircea, M.; Fuzzi, S.; Charlson, R. J. Nature 1999, 401, 257–259. (13) Padro´, L. T.; Tkacik, D.; Lathem, T.; Hennigan, C. J.; Sullivan, A. P.; Weber, R. J.; Huey, L. G.; Nenes, A. J. Geophys. Res. 2010, 115, D09204, DOI: 10.1029/2009JD013195. (14) Miyazaki, Y.; Aggarwal, S. G.; Singh, K.; Gupta, P. K.; Kawamura, K. J. Geophys. Res. 2009, 114, D19206, DOI: 10.1029/2009JD011790.

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

7973

powerful in source apportionment as they are intensive properties of the aerosol carbon fractions. Natural abundance radiocarbon (∆14C) analysis allows quantitative apportionment between fossil fuel vs biogenic and biomass combustion sources. For ∆14CWSOC there is minimal risk for interference from isotope fractionation. First, the dynamic range of ∆14C (+225 to -1000 ‰) is wide compared with the magnitude of observed C isotope effects (maximum a few per thousand, ‰). Furthermore, ∆14C values are per definition always normalized to a δ13C of -25‰.16,17 This means that any C isotope effect during formation/ transport is not affecting the ∆14C. For δ13C, some minor isotope fractionating processes cannot be ruled out a priori. However, reported δ13C isotope effects during aerosol formation18 and reactions relevant for δ13C-WSOC during transport19,20 is on the order of 0-2‰ and thus will not influence the ability of δ13CWSOC to distinguish, e.g., between C3 and C4 biosynthetic pathways. Methodologies other than δ13C and ∆14C can, in some situations, be applied for WSOC source analysis. For instance, correlative analysis of WSOC and certain ions (e.g., K+ and SO42-) or molecular markers (e.g., isoprene, levoglucosan, and dicarboxylic acids) have been explored.14,21,22 However, these indirect and extensive chemical marker approaches are challenged by varying emission factors and differing nonconservative behavior during long-range atmospheric transport. The isotope tracers and the molecular markers may favorably be used in concert. Analytical techniques for employing δ13C and ∆14C in aerosol source apportionment have been developed for several aerosol carbon fractions but not yet for WSOC. The most common C isotopic application is for the bulk TOA.23-30 Methods have also been developed and applied for 14C-based measurement of (15) Decesari, S.; Facchini, M. C.; Matta, E.; Mircea, M.; Fuzzi, S.; Chughtai, A. R.; Smith, D. M. Atmos. Environ. 2002, 36, 1827–1832. (16) Stuiver, M.; Polach, H. A. Radiocarbon 1977, 19, 355–363. (17) Zencak, Z.; Reddy, C. M.; Teuten, E. L.; Xu, L.; McNichol, A. P.; Gustafsson, O. Anal. Chem. 2007, 79, 2042–2049. (18) Widory, D. Combust. Theory Modell. 2006, 10, 831–841. (19) Rudolph, J.; Anderson, R. S.; Czapiewski, K. V.; Czuba, E.; Ernst, D.; Gillespie, T.; Huang, L.; Rigby, C.; Thompson, A. E. J. Atmos. Chem. 2003, 44, 39–55. (20) Aggarwal, S. G.; Kawamura, K. J. Geophys. Res. 2008, 113, D14301, DOI: 10.1029/2007JD009365. (21) Kumagai, K.; Iijima, A.; Tago, H.; Tomioka, A.; Kozawa, K.; Sakamoto, K. Atmos. Environ. 2009, 43, 3345–3351. (22) Snyder, D. C.; Rutter, A. P.; Collins, R.; Worley, C.; Schauer, J. J. Aerosol Sci. Technol. 2009, 43, 1099–1107. (23) Currie, L. A.; Benner, B. A., Jr.; Kessler, J. D.; Klinedinst, D. B.; Klouda, G. A.; Marolf, J. V.; Slater, J. F.; Wise, S. A.; Cachier, H.; Cary, R.; Chow, J. C.; Watson, J. E.; Druffel, R. M.; Masiello, C. A.; Eglinton, T. I.; Pearson, A.; Reddy, C. M.; Gustafsson, O.; Quinn, J. G.; Hartmann, P. C.; Hedges, J. I.; Prentice, K. M.; Kirchstetter, T. W.; Novakow, T.; Puxbaum, H.; Schmid, H. J. Res. Natl. Inst. Stand. Technol. 2002, 107, 279–298. (24) Szidat, S.; Jenk, T. M.; Gaggeler, H. W.; Synal, H. A.; Hajdas, I.; Bonani, G.; Saurer, M. Nucl. Instrum. Methods Phys. Res., Sect. B 2004, 223, 829– 836. (25) Tanner, R. L.; Parkhurst, W. J.; McNichol, A. P. Aerosol Sci. Technol. 2004, 38, 133–139. (26) May, B.; Wagenbach, D.; Hammer, H.; Steier, P.; Puxbaum, H.; Pio, C. Tellus B 2008, 61, 464–472. (27) Szidat, S.; Pre´voˆt, A. S. H.; Sandradewi, J.; Alfarra, M. R.; Synal, H.-A.; Wacker, L.; Baltensperger, U. Geophys. Res. Lett. 2007, 34, L05820, DOI: 10.1029/2006GL028325. ¨ . Atmos. Environ. 2007, 41, 7895– (28) Zencak, Z.; Elmquist, M.; Gustafsson, O 7906. (29) Schichtel, B. A.; Malm, W. C.; Bench, G.; Fallon, S.; McDade, C. E.; Chow, J. C.; Watson, J. G. J. Geophys. Res. 2008, 113, D02311, DOI: 10.1029/ 2007JD008605.

7974

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

combustion-derived aerosol components such as elemental and soot carbon24,27,28,30 as well as for compound-specific radiocarbon analysis of aerosol samples.28,31-34 The radiocarbon signal in aerosol WSOC has only been reported twice earlier.35,36 In those pioneering studies, an intriguingly large fossil contribution to the WSOC fraction (20-40%) was found for a small set of samples from the Southeast and Northeast United States, respectively. Unfortunately, the employed WSOC isolation methods for extracting, purifying, and preparing WSOC for small-sample14C-AMS were not tested for low-level carbon contamination and/or isotope fractionation, which may be important for small-sample 14 C AMS analysis of environmental isolates.23 Carbon isolates for 14C analysis, such as WSOC, are amenable to analytical process contamination by carbon of either higher or lower 14C/ 12 C isotopic composition.17,37 A particularly worrisome form of carbon contamination is from trace amounts of 14C-labeled compounds, which are used in many biologically and/or environmentally oriented research fields. Contamination by very small amounts of such compounds during laboratory handling is sufficient to cause an extreme error in the radiocarbon analysis, since the natural abundance of 14C relative to 12C is in the order of 1.2 × 10-12 (e.g., only 30 ag of 14C is sufficient to change the ∆14C value of a 25 µg C sample from +100 to +1100‰17,38,39). A second putative source of error is method-induced isotopic fractionation. That is a process wellknown to occur in various separation methods such as those based on chromatography and vaporization.40,41 The objective of this study is to develop a method to provide isolates of the WSOC fraction of atmospheric aerosols that are suitable for off-line δ13C (isotope ratio mass spectrometry) and ∆14C (accelerator mass spectrometry) quantification. The method builds on existing protocols for extraction and quantification of WSOC mass and adds further steps for preparing the water-solubilized organic carbon for isotope measurements. Both well-defined WSOC chemical probes and WSOC extracted from ambient aerosol are used to test for process carbon contamination and integrity of the method to preserve the carbon isotopic signal. (30) Gustafsson, O.; Krusa, M.; Zencak, Z.; Sheesley, R. J.; Granat, L.; Engstrom, E.; Praveen, P. S.; Rao, P. S. P.; Leck, C.; Rodhe, H. Science 2009, 323, 495–498. (31) Mandalakis, M.; Gustafsson, O. J. Chromatogr., A 2003, 996, 163–172. (32) Mandalakis, M.; Gustafsson, O.; Alsberg, T.; Egeback, A. L.; Reddy, C. M.; Xu, L.; Klanova, J.; Holoubek, I.; Stephanou, E. G. Environ. Sci. Technol. 2005, 39, 2976–2982. (33) Kumata, H.; Uchida, M.; Sakuma, E.; Uchida, T.; Fujiwara, K.; Tsuzuki, M.; Yoneda, M.; Shibata, Y. Environ. Sci. Technol. 2006, 40, 3474–3480. (34) Sheesley, R. J.; Krusa, M.; Krecl, P.; Johansson, C.; Gustafsson, O. Atmos. Chem. Phys. 2009, 9, 3347–3356. (35) Weber, R. J.; Sullivan, A. P.; Peltier, R. E.; Russell, A.; Yan, B.; Zheng, M.; de Gouw, J.; Warneke, C.; Brock, C.; Holloway, J. S.; Atlas, E. L.; Edgerton, E. J. Geophys. Res. 2007, 112, D13302, DOI: 10.1029/2007JD008408. (36) Wozniak, A. S.; Bauer, J. E.; Sleighter, R. L.; Dickhut, R. M.; Hatcher, P. G. Atmos. Chem. Phys. 2008, 8, 5099–5111. (37) Mollenhauer, G.; Montlucon, D.; Eglinton, T. I. Radiocarbon 2005, 47, 413– 424. (38) Brown, K.; Tompkins, E. M.; White, I. N. H. Mass Spectrom. Rev. 2006, 25, 127–145. (39) Buchhols, B. A.; Freeman, S. P. H. T.; Haack, K. W.; Vogel, J. S. Nucl. Instrum. Methods Phys. Res., Sect. B 2000, 172, 404–408. (40) Holmstrand, H.; Mandalakis, M.; Zencak, Z.; Gustafsson, O.; Andersson, P. J. Chromatogr., A 2006, 1103, 133–138. (41) Aeppli, C.; Holmstrand, H.; Andersson, P.; Gustafsson, O. Anal. Chem. 2010, 82, 420–426.

EXPERIMENTAL SECTION Chemicals and Preparation of Standard Solutions. Malic acid, cis-pinonic acid (98%) and phthalic acid were obtained from Sigma-Aldrich (Bellefonte, PA). Oxalic acid dihydrate (100%) was purchased from Alfa Aesar (Ward Hill, MA). These synthetic standards were selected as WSOC test substances as they have been reported to be present in the WSOC fraction of atmospheric aerosols.42,43 Sucrose, purchased from Scharlau Chemie (Barcelona, Spain), was selected as a WSOC test substance because of its known contemporary (modern 14C) origin and thus sensitively diagnostic for any fossil carbon contamination during the analytical process. Similarly, the WSOC organic acid probes are 14C extinct (synthesized from fossil petroleum feedstock) and may thus indicate any process contamination of biogenic/ modern carbon. Acetone and potassium hydrogen phthalate were obtained from Merck (Darmstadt, Germany). Solutions and extracts were prepared in Milli-Q water (18.2 MΩ quality) from a Milli-Q purification system (Millipore, Billerica, MA). Aerosol Sampling and Aerosol Bulk Carbon Analysis. To further test the δ13C/∆14C-WSOC method with a realistic matrix, high-volume total suspended particle (TSP) air samples were obtained as detailed earlier.30,34 The ambient aerosols were collected during 3-day intervals onto 142 mm quartz fiber filters (QFF; Millipore, Billerica, MA), precombusted at 450 °C for 5 h, from a 15 m platform on the top floor of the Geoscience building at Stockholm University campus (59°21′58′′N, 18°03′29′′E), Stockholm, Sweden, August 19-October 2, 2009. The site is located in a mixed grassland-forested area (a National park) about 120 m away from a highway. Field blanks were obtained by placing and removing the blank filters in the filter holder at site. The aerosol bulk organic carbon (OC) and elemental carbon (EC, a key POA component) contents were analyzed with a standard thermaloptical transmission analyzer (Sunset Laboratory, Tigard, OR).44 WSOC Extraction and Measurement of WSOC Mass Concentration. The full analytical scheme of the δ13C/∆14CWSOC method is graphically summarized in the Supporting Information (Figure S-1). The initial WSOC extraction steps largely follow established protocols.7,42,45,46 Briefly, our specific methodology was to subject filter punches/subsamples (1 cm × 1.5 cm size) to 15 min extraction in 20 mL of Milli-Q water immersed in an ultrasonic bath (Ultrasons, JP Selecta, Abrera, Spain). The number of filter pieces ranged between two and four, as determined by the OC load measured by the thermal-optical transmission instrument. Next, large filter residues were removed from the solution by centrifugation at 1500 rpm for 10 min (Sigma 4-15, Labex Instrument AB, Helsingborg, Sweden). The supernatant was filtered using 0.02 µm cutoff aluminum syringe filters (Anotop 10 Plus; Whatman, Maidstone, Kent, U.K.) prewashed with 2 mL of acetone and 10 mL of Milli-Q water. About 3 mL of the extract solution was used for syringe rinsing before the sample filtration. (42) Graham, B.; Mayol-Bracero, O. L.; Guyon, P.; Roberts, G. C.; Decesari, S.; Facchini, M. C.; Artaxo, P.; Maenhaut, W.; Koll, P.; Andreae, M. O. J. Geophys. Res. 2002, 107, 8047, DOI: 10.1029/2001JD000336. (43) Myhre, C. E. L.; Nielsen, C. J. Atmos. Chem. Phys. 2004, 4, 1759–1769. (44) Birch, M. E.; Cary, R. A. Analyst 1996, 121, 1183–1190. (45) Mader, B. T.; Yu, J. Z.; Xu, J. H.; Li, Q. F.; Wu, W. S.; Flagan, R. C.; Seinfeld, J. H. J. Geophys. Res. 2004, 109, D06206, DOI: 10.1029/2003JD004105. (46) Ram, K.; Sarin, M. M. J. Aerosol Sci. 2010, 41, 88–98.

WSOC was quantified as total dissolved organic carbon (DOC) in the filtered solutions using a high-temperature catalytic oxidation (HTCO) instrument (Shimadzu-TOC-VCPH analyzer; Shimadzu, Kyoto, Japan) following the nonpurgeable organic carbon (NPOC) protocol.47 The NPOC analysis is used to measure dissolved nonvolatile organic carbon and includes automatic acidification of samples to a pH less than 2.0, sparging with ultrapure (CO2 free) air and quantification of CO2 produced by sample combustion with a nondispersive infrared gas analyzer. Each sample was run in three replicate injections. The complete WSOC field blank was constrained for three different dates during the Stockholm campaign (e0.2, 0.52, and 0.44 µg/cm2). Considering that the WSOC load per filter area in the typical ambient aerosol samples ranged from ∼5-30 µg/ cm2, it is estimated that the WSOC field blanks correspond to an average around 2% (range 1-7%) of the field signal. Preparation of WSOC Isolates for Isotopic Analyses. A filter area corresponding to at least 300 µg of WSOC was targeted for 14C analysis. The corresponding subsampled filter was placed into precombusted (5 h at 450 °C) glass Petri dishes (6.0 cm i.d.) for acidification in order to remove inorganic carbon. Acidification was performed by fumigation of the filter samples in open Petri dishes held in a desiccator also containing a glass beaker with 50 mL of 12 M (37%) hydrochloric acid for 24 h. Finally, the samples were dried at 60 °C for 1 h. Decarbonated filters were then extracted for WSOC in 10 mL of Milli-Q water in 50 mL glass centrifuge vials using the same ultrasonication method as described for WSOC concentration measurement above. The WSOC solutions were subsequently filtered into 30 mL polycarbonate vials (NALGENE, Rochester, NY). The solutions were frozen and freeze-dried to dryness under vacuum (40-50 mbar) at -20 to -10 °C for about 16 h using a low-carbon background freeze drier (Alpha 2-4 LSC, Martin Christ, Osterode an Harz, Germany). The fluffy residue material was redissolved in 300 µL of Milli-Q water and transferred into 5 mm × 12 mm silver capsules (Sa¨ntis Analytical, Uppsala, Sweden), which had been precombusted at 450 °C for 5 h. Each sample was divided into two capsules: 100 µL was placed into the capsule for carbon concentration and stable carbon isotope measurements and the rest into the capsule for radiocarbon measurements. Finally, samples were evaporated in the oven at 60 °C, and dried WSOC samples were ready for isotope ratio measurements. Measurement of WSOC Composition for Carbon Isotopes. Stable carbon isotope (δ13C-WSOC) measurements were performed at the Department of Geological Sciences of Stockholm University (Stockholm, Sweden). The instrumental method employed sample combustion with a Carlo Erba NC2500 analyzer connected via a split interface to reduce the gas volume to a Finnigan MAT Delta V mass spectrometer. Longterm precision of such δ13C is 0.1-0.2‰ relative to the universal VPDB standard. The ∆14C determination of the freeze-dried WSOC isolates was performed at the U.S. National Ocean Sciences Accelerator Mass Spectrometry (NOSAMS) facility of the Woods Hole Oceanographic Institution (Woods Hole, MA). The final AMS preparation steps include combustion of the samples in a quartz (47) Sharp, J. H.; Benner, R.; Bennett, L.; Carlson, C. A.; Fitzwater, S. E.; Peltzer, E. T.; Tupas, L. M. Mar. Chem. 1995, 48, 91–108.

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

7975

tube with CuO and Ag to obtain carbon dioxide followed by reduction to graphite. The accelerator mass spectrometry analysis followed previously published procedures.48,49 The results of the AMS analysis were reported as fraction modern and ∆14C relative to NBS Oxalic Acid I standard.16,28 From the ∆14C values fractional contributions of contemporary biomass/biogenic sources vs radiocarbon-extinct fossil fuel, sources can be determined using the isotopic mass balance equation: ∆14Csample ) ∆14Cbiomassfbiomass + ∆14Cfossil(1 - fbiomass) (1) where ∆14C sample is the measured radiocarbon content of a WSOC sample and ∆14Cfossil is -1000‰. The ∆14Cbiomass endmember is between +70‰ and +225‰. The first value corresponds to the ∆14C of contemporary CO250 and thus freshly produced biomass. The second value is for the ∆14C of wood logged in the 1990s.28,51 Given the current understanding that WSOC in the boreal zone is overwhelmingly from biogenic precursors52 and the expectation of very limited wood burning in the Stockholm area in the summer, the ∆14C end-member value for freshly produced biomass (+70‰) is employed here. Method Tests. The δ13C/∆14C-WSOC method was developed and tested by assessing both the recovery of carbon mass and isotopic fractionation of test substances and ambient aerosols. The efficiency of the standard literaturebased WSOC extraction method was tested both with a kinetic experiment and by adding known amounts of probe WSOC molecules to the ambient filter-aerosol matrix. Extraction times of 5-15-25-35 min were assessed (Supporting Information, Table S-1). Known amounts of sucrose and phthalic acid were spiked to ambient Stockholm aerosol filters and their recoveries probed in a standard additions fashion (Supporting Information, Table S-2). The mass recovery during both the extraction steps and the isolation for isotopic analyses were also probed by well-defined synthetic WSOC analogues. The key test of the δ13C/∆14C-WSOC method was the assessment of how well it preserves the δ13C and ∆14C signals of WSOC during processing. For this, both synthetic WSOC probes and ambient aerosols were employed. Synthetic WSOC probes were selected to span a wide range in starting δ13C values from -30 (phthalic acid) to -6 ‰ (oxalic acid). To test for low-level process contamination of both contemporary and fossil carbon, both sucrose (no fossil carbon) and phthalic acid (no contemporary carbon) were analyzed for their “∆14C-WSOC” content. Finally, three ambient Stockholm summer aerosol samples were analyzed (48) McNichol, A. P.; Gagnon, A. R.; Jones, G. A.; Osborne, E. A. Radiocarbon 1992, 34, 321–329. (49) Pearson, A.; McNichol, A. P.; Schneider, R. J.; Von Reden, K. F.; Zheng, Y. Radiocarbon 1998, 40, 61–75. (50) Levin, I.; Kromer, B.; Schmidt, M.; Sartorius, H. Geophys. Res. Lett. 2003, 30, 2194. (51) Klinedinst, D. B.; Currie, L. A. Environ. Sci. Technol. 1999, 33, 4146– 4154. (52) Tunved, P.; Hansson, H. C.; Kerminen, V. M.; Strom, J.; Dal Maso, M.; Lihavainen, H.; Viisanen, Y.; Aalto, P. P.; Komppula, M.; Kulmala, M. Science 2006, 312, 261–263.

7976

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

for their inherent δ13C-WSOC and ∆14C-WSOC content and compared with system expectations from this aerosol regime. RESULTS AND DISCUSSION Characteristics of Ambient Aerosol Regime Used for Method Testing. The natural aerosol of the Stockholm suburban national park in late summer-early fall exhibited bulk aerosol carbon levels typical for the boreal zone (Supporting Information, Figure S-2). The mean Stockholm WSOC concentration was 1.0 ± 0.3 µg m-3 (n ) 18; Supporting Information, Figure S-2) with a mean standard deviation of 7% (based on five triplicate extractions). This compares well with a broad range of reported WSOC concentrations using similar methods for both rural (0.5-4.8 µg m-3 53,54) and urban (1.4-1.6 µg m-3 55-57) sites. The WSOC/ OC ratios for the Stockholm test samples were 14-40% (30 ± 7%), and the WSOC/TC ratios were 12-35% (26 ± 6%) with a positive correlation between WSOC and both OC and TC (R2 of 0.69 and 0.66; data not shown). WSOC/OC as well as OC/EC ratios may be used for interpreting aerosol composition and different source contribution. Generally higher WSOC/OC ratios are found for aged aerosols and SOA57,58 while traffic emissions are characterized by low WSOC/ OC.57 During the summer season, it is expected that WSOC/TC would be higher due to input of biogenic secondary organic aerosols (SOA).55 The Stockholm aerosol WSOC/TC is in the same range as previously reported for both urban (34-36% for PM10 and PM2.5, respectively) and rural (23-66%)54,55,59 sites. WSOC Mass Recoveries during Extraction, Purification, and Preparation for Isotopic Analyses. The yields of the synthetic WSOC test substances during filter extraction and purification were tested. An average WSOC test substance extraction yield of 91 ± 6% of filter-spiked carbon mass was recorded with the lowest value for malic acid (86 ± 3%) and highest for phthalic acid (98 ± 5%) (Figure 1A). These high recoveries, combined with WSOC/OC ratios of the tested Stockholm aerosol in same range as typical ambient aerosols, suggest that the employed 15 min ultrasonication-mediated water extraction procedure is successful in extracting WSOC from quartz filters. Furthermore, the test of the duration of the extraction demonstrated similar extraction efficiencies for the whole series 5-15-25-35 min (Supporting Information, Table S-1). The standard addition test of matrix effects on the extraction yield of the test substances gave complete recoveries (Supporting Information, Table S-2). The analytical steps of freeze-drying for preparing WSOC extracts for carbon isotope analysis were tested with both welldefined WSOC chemical standards and ambient aerosols. The carbon mass recovery of WSOC test substances after freeze-drying was on average 71 ± 20% with lowest recovery for oxalic acid (44 (53) Carvalho, A.; Pio, C.; Santos, C. Atmos. Environ. 2003, 37, 1775–1783. (54) Jaffrezo, J.-L.; Aymoz, G.; Delaval, C.; Cozic, J. Atmos. Chem. Phys. 2005, 5, 2809–2821. (55) Viana, M.; Chi, X.; Maenhaut, W.; Querol, X.; Alastuey, A.; Mikuska, P.; Vecera, Z. Atmos. Environ. 2006, 40, 2180–2193. (56) Yttri, K. E.; Dye, C.; Braathen, O. A.; Simpson, D.; Steinnes, E. Atmos. Chem. Phys. 2009, 9, 2007–2020. (57) Saarikoski, S.; Timonen, H.; Saarnio, K.; Aurela, M.; Jarvi, L.; Keronen, P.; Kerminen, V. M.; Hillamo, R. Atmos. Chem. Phys. 2008, 8, 6281–6295. (58) Aggarwal, S. G.; Kawamura, K. Atmos. Environ. 2009, 43, 2532–2540. (59) Temesi, D.; Molnar, A.; Meszaros, E.; Feczko, T.; Gelencser, A.; Kiss, G.; Krivacsy, Z. Atmos. Environ. 2001, 35, 4347–4355.

Figure 1. Carbon mass recovery tests for the two major steps of the δ13C/∆14C-WSOC method: (A) carbon mass yields after filter extraction; (B) carbon mass yield after freeze-drying, resolvation in water, and preparation of WSOC samples for isotopic analysis. The samples Stockholm 1 (090831-090902), Stockholm 2 (090907090909), and Stockholm 3 (090918-090921) are ambient aerosols collected in suburban Stockholm (see Experimental Section).

± 7%) and highest recovery for sucrose (96 ± 1%) (Figure 1B). Oxalic acid (a C2 dicarboxylic acid) was selected to represent a worst-case scenario as it is one of the smallest molecule, with the highest vapor pressure, found in ambient WSOC.60 Oxalic acids can thus be expected to be more sensitive to losses due to volatilization during freeze-drying than the average WSOC component, yet nearly half of this challenging test substance is recovered with the current protocol. The carbon mass recovery of the WSOC extract from the Stockholm ambient aerosol was 73 ± 11% (n ) 3; Figure 1B). The best test of suitability of the δ13C/∆14C-WSOC preparation steps is the extent of isotope shift in the resulting isolate relative to the isotope signal of the starting material. Evaluation of the Consistency of WSOC Carbon Isotopic Composition. Both the stable (δ13C) and radiocarbon (∆14C) results of the chemical standards and of ambient aerosol indicate absence of any significant carbon isotope contamination and/or method-induced isotope fractionation. In spite of just ∼70% carbon mass recoveries of the WSOC test chemicals (60) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, 2nd ed.; Wiley: New York, 2006.

Figure 2. δ13C-WSOC (A) and ∆14C -WSOC (B) results after freezedrying, indicating the absence of any significant process carbon contamination and/or isotopic fractionation. Black-filled circles represent δ13C of unprocessed starting matter of WSOC-analogue probe compounds, white filled circles represent δ13C of the WSOC isolate after freeze-drying, and gray-filled circles represent δ13C-TOC. The samples Stockholm 1 (090831-090902), Stockholm 2 (090907090909), and Stockholm 3 (090918-090921) are ambient Stockholm aerosols (see the Experimental Section) employed to demonstrate the application of the δ13C/∆14C-WSOC method.

(Figure 1B), the δ13C-WSOC values were indistinguishable from the δ13C signature in the unprocessed starting material (Figure 2A). The method yields consistent results over a wide range of δ13C, spanning from the 13C-heavy signal of oxalic acid (-6.59 ± 0.37‰ for unprocessed vs 6.33 ± 0.31‰ for its WSOC isolate; 2 SD) to the 13C-depleted signal of, e.g., cis-pinonic acid (-26.30 ± 0.33‰ for unprocessed vs -25.89 ± 0.25‰ for its WSOC isolate; 2 SD). The application of the method to ambient aerosols was demonstrated by the consistency of the δ13C for three of the Stockholm WSOC samples with values narrowly remaining between -25.1 ± 0.2‰ and -25.6 ± 0.2‰ for samples collected over a 3 week period (Figure 2A), consistent with a dominant input of C3 precursor molecules as would be expected in this temperate-boreal vegetation zone. From this late summer suburban Stockholm location, the δ13CWSOC obtained with the current method is enriched by ∼1.5‰ relative to bulk organic aerosol (i.e., δ13C-TOC) (Figure 2A), which is consistent with other field studies.61 (61) Fisseha, R.; Saurer, M.; Jaggi, M.; Siegwolf, R. T. W.; Dommen, J.; Szidat, S.; Samburova, V.; Baltensperger, U. Atmos. Environ. 2009, 43, 431–437.

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

7977

Table 1. 13C and 14C Isotope Composition of Both Well-Defined WSOC-Analogue Probe Molecules and of WSOC Fraction of Ambient Stockholm Aerosol sample

δ13C ‰

∆14C ‰

fbiomassa

phthalic acid, unprocessed phthalic acid, freeze-dried sucrose, unprocessed sucrose, freeze-dried Stockholm 1 Stockholm 2 Stockholm 3

-30.4 ± 0.2

-992.4 ± 72

0.0071 ± 0.0005

-29.8 ± 0.2

-992.0 ± 85

0.0075 ± 0.0006

-26.3 ± 0.2 -25.9 ± 0.2 -25.1 ± 0.2 -25.2 ± 0.2 -25.6 ± 0.2

+84.4 ± 0.3 +69.9 ± 0.3 -57.6 ± 0.1 +124.6 ± 0.7 -53.2 ± 0.3

1.0135 ± 0.0037 0.9999 ± 0.0039 0.8807 ± 0.0019 1.0511 ± 0.0058 0.8849 ± 0.0054

a Applying an end-member of ∆14C ) 70‰ for contemporary biomass.50

The accelerator mass spectrometry radiocarbon results of the WSOC isolates indicated good performance of the methodology for ∆14C-WSOC. The result for the fossil phthalic acid suggested absence of any hot/modern 14C contamination during the method processing (Figure 2B and Table 1). The phthalic acid by the WSOC freeze-drying method yielded a practically extinct ∆14C signal of -992 ± 86‰, which is indistinguishable from the ∆14C of the unprocessed phthalic acid (Table 1) and the AMS detection limit of -1000‰.48 Although contamination by trace residues of 14C-labeled compounds is the “worst-case” scenario in a laboratory dedicated to 14C analysis, contamination by fossil C is perhaps more likely since most chemicals used in a laboratory setting are produced from fossil petroleum feed stock. Contamination from general laboratory dust has likely an intermediate, yet inherently unknown, 14C signal. Fortunately, the ∆14C-WSOC signal for the 14C-modern test substance sucrose indicates absence in the method of any significant fossil C contamination (Figure 2B and Table 1). The sucrose WSOC standard recorded a ∆14C of +69.9 ± 0.3‰, which is indistinguishable from the unprocessed sucrose (Table 1) and furthermore consistent with modern photosynthesis of sugar beet from atmospheric CO2 with a ∆14C around +70‰.50 The application of the developed ∆14C-WSOC method to ambient aerosols was demonstrated for the three Stockholm aerosol samples and yielded overall quite contemporary biomass 14C signals (Figure 2B and Table 1). The two samples collected August 31 to September 2 and September 19-21 have very similar contribution of contemporary biomass to WSOC (88%). This is consistent with the expectation that the main component of the extracted WSOC in this summer aerosol regime is biogenic SOA, as in the European boreal region gas-to-particle formation from biogenic monoterpenes, substantially contribute to aerosol mass and number.52 The presence of a fossil component (up to 12%) in the samples indicates the contribution to WSOC from oxidation of fossil-fuel derived VOCs (to form water-soluble SOA) or aging of POA with fossil fuel combustion origins in the (62) Stone, E. A.; Snyder, D. C.; Sheesley, R. J.; Sullivan, A. P.; Weber, R. J.; Schauer, J. J. Atmos. Chem. Phys. 2008, 8, 1249–1259. (63) Mandalakis, M.; Gustafsson, O.; Reddy, C. M.; Li, X. Environ. Sci. Technol. 2004, 38, 5344–5349.

7978

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

intercepted air masses (potentially from the nearby highway). The sample collected September 7-9 has a ∆14C-WSOC signal (+124.6 ± 0.7‰; Table 1), which indicates additional contribution from wood combustion. Average 20th century wood fuel from the boreal zone has an end member ∆14C composition of +225‰.27,28,51 Biomass burning has been suggested to contribute to WSOC elsewhere.46,62 The result for this single sample likely reflects contribution from local outdoor bonfires on the university campus/ natural park and/or household open fire places, which are quite common in Sweden.34,63 Taken together, the combined isotopic results of the test suite of well-defined WSOC probes and of ambient aerosols suggest that the developed method can be employed to elucidate the δ13C and ∆14C signal of the WSOC fraction of carbonaceous aerosols. This isotopic fingerprinting is useful information for distinguishing between fossil and contemporary sources of climate- and health-affecting aerosols. CONCLUSIONS A method for analysis of the stable carbon isotope and radiocarbon content of the water-soluble organic carbon (WSOC) component of ambient atmospheric aerosols was developed. Critical method aspects involve freeze-drying of the WSOC extract in preparation for the off-line carbon isotope measurements. The preservation of probe molecule δ13C and ∆14C signals from the unprocessed starting material into the WSOC isolates render a high integrity to the isotope method. These results demonstrate the absence of any significant process contamination of either fossil or modern carbon as well as the absence of any measurable method-induced δ13C isotope fractionation. The Stockholm late summertime field demonstration of radiocarbon analysis of ambient particulate WSOC revealed a high contribution from biogenic sources, which is expected for summer in the boreal zone. Taken together, these results show that the method can be applied to isolate WSOC in ambient aerosols for carbon-isotope based apportionment between fossil and biomass/biogenic derived sources of these key climate-affecting atmospheric constituents. ACKNOWLEDGMENT We acknowledge financial support from the Swedish Research Council for Environment, Agricultural Sciences and Spatial Plan¨ .G. also acknowlning (FORMAS Contract No. 214-2009-970). O edges financial support as an Academy Researcher from the Swedish Royal Academy of Sciences, and A.A. acknowledges financial support from the Knut and Alice Wallenberg Foundation. This study also benefitted from the research environments provided by the Bert Bolin Climate Research Centre and the Delta Facility (a core facility for compound-specific isotope analysis), both at the Stockholm University School of Natural Sciences. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review June 1, 2010. Accepted September 8, 2010. AC1014436