Environ. Sci. Technol. 2006, 40, 5917-5922
Molecular Size Evolution of Oligomers in Organic Aerosols Collected in Urban Atmospheres and Generated in a Smog Chamber MARKUS KALBERER,* MIRJAM SAX, AND VERA SAMBUROVA Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland
Only a minor fraction of the total organic aerosol mass can be resolved on a molecular level. High molecular weight compounds in organic aerosols have recently gained much attention because this class of compound potentially explains a major fraction of the unexplained organic aerosol mass. These compounds have been identified with different mass spectrometric methods, and compounds with molecular masses up to 1000 Da are found in secondary organic aerosols (SOA) generated from aromatic and terpene precursors in smog chamber experiments. Here, we apply matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) to SOA particles from two biogenic precursors, R-pinene and isoprene. Similar oligomer patterns are found in these two SOA systems, but also in SOA from trimethylbenzene, an anthropogenic SOA precursor. However, different maxima molecular sizes were measured for these three SOA systems. While oligomers in R-pinene and isoprene have sizes mostly below 600-700 Da, they grow up to about 1000 Da in trimethylbenzene-SOA. The final molecular size of the oligomers is reached early during the particle aging process, whereas other particle properties related to aging, such as the overall acid concentration or the oligomer concentration, increase continuously over a much longer time scale. This kinetic behavior of the oligomer molecular size growth can be explained by a chain growth kinetic regime. Similar oligomer mass patterns were measured in aqueous extracts of ambient aerosol samples (measured with the same technique). Distinct differences between summer and winter were observed. In summer a few single mass peaks were measured with much higher intensity than in winter, pointing to a possible difference in the formation processes of these compounds in winter and summer.
Introduction Although a major fraction of the ambient aerosol is composed of organic material (often 20-50%, ref 1), only a small fraction is known on a molecular level. To assess the influence of ambient particles on the radiative balance of the atmosphere as well as on health related issues, the chemical properties of the organic mass fraction has to be known in greater detail. Despite the large effort in the past decades and the identification of hundreds of single compounds, the chemical * Corresponding author phone: +41 44 632 2929; fax: +41 44 632 1292; e-mail:
[email protected]. 10.1021/es0525760 CCC: $33.50 Published on Web 08/25/2006
2006 American Chemical Society
identity of 80-90% of the organic mass remains unknown (2, 3). In recent years evidence from field and laboratory studies showed that possibly a large fraction of the organic aerosol is composed of molecules with high molecular weights. In ambient aerosols, a class of compounds with chemical properties similar to humic substances was found, and was, therefore, named humic-like substances (HULIS) (4). Spectroscopic methods such as ultraviolet, fluorescence, or infrared spectroscopy were most often used to characterize this organic mass fraction but electrospray ionization mass spectroscopy and laser desorption/ionization mass spectrometry (LDI-MS) were used as well (5-8). Because of the large number of compounds detected with mass spectrometric methods (7, 8), and because the chemical properties of these compounds are mostly unknown, it is difficult to determine their concentration or their molecular weight distribution. Chromatographic methods (e.g., size exclusion chromatography) and mass spectrometry have been used to estimate molecular weight distributions. Samburova et al. (8) found a small fraction of water-soluble HULIS with a molecular weight up to about 3500 Da. The majority of the high molecular weight compounds, however, seem to be in the range of 400-700 Da (6-8). Concentration determinations are chiefly based on spectroscopic detection methods using humic or fulvic acid as quantification standards (e.g., ref 8). These studies showed that up to 30% of the total organic aerosol mass might be due to these high molecular weight compounds. The origin and formation processes of these compounds are largely speculative, but secondary formation from biomass combustion precursors or from other secondary organic aerosol (SOA) precursors are likely processes (8, 9). Complementing the findings from field studies, several laboratory experiments in recent years showed that high molecular weight compounds are also formed in SOA particles generated in smog chamber experiments. Compounds with a molecular weight >900 Da were measured in SOA particles generated from R-pinene in a smog chamber and were analyzed with electrospray ionization mass spectrometry (10-12). Possible SOA precursors and SOA components were oxidized in laboratory experiments (13, 14) and reaction products with physicochemical properties similar to high molecular weight compounds present in atmospheric aerosols were found. In an earlier study (15) we showed, using laser desorption/ionization mass spectrometry (LDI-MS), that compounds of increasing molecular weight up to about 1000 Da are observed during aging of SOA particles generated for 1,3,5-trimethylbenzene (TMB). In this study, this mass spectrometric method is extended to SOA particles generated from R-pinene and isoprene and compared with the high molecular weight compounds found in TMB-SOA and in ambient samples collected at an urban site in downtown Zurich, Switzerland. In addition, kinetic aspects of the oligomer evolution are discussed and compared with long-term time trends of the particle bulk oxidation.
Experimental Section SOA particles were generated in the PSI smog chamber, a 27 m3 Teflon bag suspended in a thermostated housing. The chamber and routine equipment for the analysis of gaseous compounds and aerosol particles is described in detail by Paulsen et al. (16). All experiments were performed at 20 (( 1)°C. The chamber is equipped with four xenon arc lamps (4kW each, XBO 4000 W/HS, OSRAM) to mimic the tropoVOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5917
spheric photochemistry as closely as possible. The organic precursor (i.e., TMB, R-pinene or isoprene) was introduced into the smog chamber by evaporation in a heated flask and NO, NO2, and propene (used as a radical starter) were flushed into the chamber from gas bottles. The starting concentrations of organic SOA precursors for the experiments discussed here were 650 ppb for TMB (160 ppb NO and NO2 each), 300 ppb for R-pinene (120 ppb NO and NO2 each), and 2085 ppb for isoprene (540 ppb NO and 805 ppb NO2). All experiments were performed at 40-50% relative humidity. Only for the TMB experiments 300 ppb propene was used as radical starter. Paulsen et al. (2006) showed that propene has most likely no significant influence on the oligomer formation (17). An experiment was started by turning on the lights after all reactants were flushed into the bag. All four lamps were turned on during the entire duration of the experiments. Samples were collected 1-2 h after particle nucleation started for the mass spectrometric measurements. Sampling times were usually 1-2 h. A series of samples were collected as a function of time up to 20 h. Particles were deposited on 10 mm stainless steel plates in an 11-stage impactor (18). After sampling of 0.7-1.5 m3 of air (depending on the particle concentration in the chamber) the plates were placed directly on the sample holder of the mass spectrometer and measured without further treatment. (Matrix assisted) laser desorption/ ionization (MALDI) measurements were performed with a time-of-flight mass spectrometer (Axima-CFR, KRATOS Analytical Shimadzu, Manchester, UK, laser wavelength 337 nm, pulse width 3 ns, linear mode, delayed extraction optimized for m/z ) 500). MALDI-MS allows for fast and detailed analysis of the molecular weight distribution of polymers and oligomers. However, limitations have to be kept in mind interpreting the MALDI data. A number of studies on synthetic polymers (e.g., refs 19-22) showed that instrumental parameters (e.g., laser power, detector voltage, delay time) and sample preparation (e.g., matrix/sample ratio and type of cationization) can influence the molecular size distribution measured with MALDI-MS, resulting mostly in size distributions lower than determined with reference methods. Narrow molecular size distributions with a polydispersity index (PI) < 1.5 are generally less susceptible to such artifacts. PI determined for the oligomers measured here are in the range of 1.03-1.16 (see below). The laser power had to be kept in a narrow range between 8 and 15 µJ. At higher laser power, significant fragmentation and plasma formation was observed, which resulted in noninterpretable mass spectra. At lower laser powers, decreasing signal intensities were observed, but no shift of the molecular size distribution. This indicates that, within these laser power limits, the molecular size distribution is not affected significantly by the laser energy. It is noteworthy that SOA components generated from TMB could be easily measured without the addition of any matrix. In contrast, SOA from R-pinene and isoprene yielded very poor signal intensity (about 100-1000 times lower) when no matrix was added. The higher signal intensities for TMBSOA might be due to compounds present in TMB-SOA that efficiently absorb the UV light of the laser at 337 nm, e.g., aromatic compounds, which are not expected to be present in R-pinene- and isoprene-SOA. To increase the signal intensity for R-pinene- and isoprene-SOA compounds, graphite (1-2 µm diameter, Aldrich) or TiN matrix particles were deposited on the impactor plates before aerosol sampling, which resulted in much higher signal intensities. The addition of matrix in laser mass spectrometry methods has two main advantages. First, the matrix compound efficiently absorbs the laser energy. Often small aromatic compounds, such as 3,5-dihydroxybenzoic acid or inorganic matrixes such as graphite, are used. Laser energy absorption 5918
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006
by the matrix ensures an efficient desorption of the analyte molecules. Second, organic matrix molecules are often involved in the complex ionization reactions of the sample molecules. However, this often results in matrix-sample molecule adducts, which complicate data interpretation in a complex sample mixture. With inorganic matrixes, cationization reactions (e.g., by Na+ or K+ present as impurities or added to the matrix) often become dominant ionization processes for the analytes. If inorganic matrixes are cleaned thoroughly, almost no signal is measured from the pure inorganic matrix above approximately m/z 200. Comparative tests with two different inorganic matrixes, graphite and TiN particles, showed very similar mass distributions for isoprene SOA (data not shown). The overall features of the oligomer distribution are the same for both matrixes, and the maxima of the repetitive units with ∆m 14-16 are measured at the same m/z for both matrixes. This largely excludes that the inorganic matrixes used here significantly interfere with the oligomer mass distribution of the SOA particles. Ambient samples were collected in August/September 2002 and February/March 2003 at an urban background site in downtown Zurich, Switzerland. Day and night samples were collected with a high-volume sampler at a flow rate of 0.5 m3 min-1 (Digitel, Switzerland) during 12 h for 1-3 consecutive days and nights, respectively, resulting in a total sampling time of 12-36 h. Ten 15 mm diameter cuts of each filter were extracted with milli-Q water. Filters were extracted by soaking them for two hours in milli-Q water. Samples were then centrifuged to separate insoluble particles and filter material from the supernatant, from which 50 µL were added to the sample plate of the mass spectrometer. After evaporation of the water in a desiccator, the samples were directly measured without adding any matrix. Extractions in organic solvents were much less efficient and resulted in much lower signal intensities (see ref 8 for more details). Blank filters were analyzed during both campaigns and only negligible signals were detected.
Results and Discussion Smog Chamber Samples. The repetitive regular mass pattern for R-pinene and isoprene above about m/z 200-250 and m/z 400 for TMB is typical for oligomers and polymers (Figure 1a-c). The oligomer size distribution of isoprene-SOA extends to about m/z 600 (Figure 1b), only slightly smaller than the distribution of R-pinene oligomers (Figure 1a). The intensity maxima for the oligomer distributions are different of the three systems. While the peaks with the highest intensities in isoprene-SOA are at about m/z 280, the maximum in R-pinene-SOA is measured at about m/z 380 and for TMB at about m/z 480. The oligomers in TMB-SOA extend up to masses of m/z 900 (Figure 1c). High-intensity mass peaks are separated in all three SOA systems by ∆m 14, 16, and 18. 3-4 masses often have high signal intensities followed by a group of peaks with lower signal intensities. In polymers, such regular repetitive mass patterns are usually observed with mass differences equal to the monomer building blocks of the polymer. The small mass differences observed (i.e., ∆m 14-18) are explained by the incorporation of monomers with different molecular masses into the oligomer. For example, the mass of a dimer formed from two C3 compounds is 14 Da smaller than a dimer formed from a C3 and a C4 compound (with a CH2 group more than the C3 monomer), although the smallest monomer is much larger than the observed mass difference. Oligomeric mass signatures are observed soon after the onset of particle nucleation generated from R-pinene and isoprene, as well as from TMB-SOA (Figure 1). For an R-pinene sample collected 1.5-2.5 h after the start of the experiment (Figure 2a), hardly any masses are measured >
FIGURE 1. MALDI mass spectra of (a) r-pinene-SOA, (b) isopreneSOA, and (c) TMB-SOA collected 4.5-6.5 h after starting the experiments showing the final molecular size distributions. After an initial growth, the measured oligomer size distribution does not grow to molecules larger than that shown here, even after 15-20 h aging time (see Figure 3). m/z 500. After about 4.5 h (Figure 2b) the oligomer size distribution clearly extends to about m/z 650. After this initial growth of the oligomer chain length, the peak intensity distribution remains quite constant. After about 19 h (Figure 2c) the molecular mass distribution of the oligomer looks almost identical to that observed after 4.5 h, and only a slight increase of the high mass oligomers around m/z 600 is observed. The same general growth behavior of the oligomer chain length is seen for isoprene- and TMB-SOA. This kinetic behavior of the molecular weight growth of oligomers can be explained with standard polymer growth models. Figure 3a shows two abundant kinetic regimes for molecular size growth of polymers, i.e., step growth and chain growth polymerization (23). In the kinetic regime of step growth, a large number of monomers can initiate the polymerization process leading to many rather small polymer molecules over most of the reaction time. High molecular weight molecules are only formed toward the end of the reaction, by which time most monomers have been incorporated into the polymer. In contrast, the polymer chain extension proceeds via an active site at the polymer molecule in chain growth kinetics, which is often a radical. This leads to high molecular weight polymers much earlier in the reaction scheme, and the chain length is largely independent of reaction time and percentage of monomer conversion. After an initial short increase, this results in a constant polymer chain length, although, with an increasing amount of polymer mass. Number average molecular weight (Mn), weight average molecular weight (Mw), and polydispersity index (PI ) Mw/Mn) are values often used to characterize the
FIGURE 2. MALDI mass spectra of r-pinene-SOA collected after different aging times in the smog chamber. (a) 2 h, (b) 4.5 h, and (c) 19.5 h after start of the experiment. molecular size distribution of polymers and oligomers (23). Mn and Mw are defined as
Mn )
∑ NM ∑N i
i
i
and
∑ NM , ∑N M 2
Mw )
i
i
i
i
respectively, where N is the number (abundance) and M is the molecular weight of the oligomer molecule i. For the three investigated SOA systems, Mn and Mw are either almost constant over the entire experiment time or are only slightly increasing (Figure 3b). PI values range from 1.028 to 1.040 for R-pinene and 1.049-1.076 for TMB to 1.114-1.157 for isoprene indicating rather narrow oligomer size distributions. PI values did not change significantly over time. Assuming that compounds incorporated into the oligomer have molecular masses of about 70-150 Da (which is a typical mass range of known oxidation products of the three SOA systems investigated here), the oligomers have a maximum chain length of only 10-15 monomers, which points to a very efficient termination reaction for the oligomerization reaction. Because the oligomer mass fraction in the SOA particle increases steadily over a much longer time scale than the growth of molecular size (see below), it is unlikely that a depletion of the respective monomers over time in the VOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5919
FIGURE 3. (a) Theoretical model of polymer kinetics showing chain growth and step growth polymerization. (b) Temporal change of the number average molecular weight (Mn, open symbols) and weight average molecular weight (Mw, solid symbols) for TMB (green), r-pinene (blue), and isoprene (red) SOA. (c) In contrast to the rapid cease of the oligomer chain growth, the particle bulk is continuously oxidized over a much longer time scale. Concentration changes of acids in r-pinene particles as measured with FT-IR (1) (24), IC-MS (b) (26), and AMS (- - -) (25) as well as carbonyls measured by FT-IR (9) (24). particle or that slower oligomerization rates are responsible for a cease of the molecular size growth of the oligomers. The relatively fast cease in growth of the molecular weight of the oligomers is in contrast with the oligomer concentration in R-pinene- and TMB-SOA particles as measured with a volatility tandem differential mobility analyzer (VTDMA) in earlier studies (15, 17). This technique measures the nonvolatile particle fraction at elevated temperatures (e.g., 100, 150, and 200 °C), which is assumed to represent mostly oligomers. The VTDMA measurements show a steadily increasing oligomer mass fraction over more than 20 h. After about 5-6 h, only 50-60% of the SOA particle volume is nonvolatile at 100 °C for TMB- and R-pinene-SOA and increases continuously to about 80 and 90% after about 25 h (17). The kinetics of the formation of SOA oligomers, where a final molecular weight of the oligomers is reached relatively fast but the oligomer mass fraction is continuously increasing, closely resembles the kinetic behavior of a chain growth polymerization (Figure 3a). The total particle bulk is increasingly oxidized over a time scale of at least 17 h as measured with Fourier transform infrared spectroscopy (FT-IR) of R-pinene-SOA particles collected with an impactor (24), similar to the slow and steady increase of the oligomer mass fraction but in contrast to the oligomer molecular size evolution. An especially strong 5920
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006
increase of acid functional groups by a factor of 2.5 is observed with FT-IR over this time range (24) (Figure 3c). A similar increase of acids, although smaller, is measured with aerosol mass spectrometry (AMS), where a mass fragment assigned to acid functional groups (m/z 44) is increasing continuously up to a factor of 1.5 over about 15 h (25). The same trend can be observed with ion chromatography-mass spectrometry (IC-MS) measuring the concentration of single acids in the particles (26) and in the concentration of carbonyls measured with FT-IR (24). The increase in oxidized functional groups, measured with three different techniques (FT-IR, AMS, IC-MS), can partially be explained by an increasing absorption of small oxidized compounds from the gas phase (e.g., acids). However, particle volume measurements showed the total particle mass growth ceases after 6-8 h (15). Thus, the continuous oxidation of the particle bulk can be explained only during the first 6-8 h (when the particle mass is still growing) with absorption of oxidized compounds from the gas phase. At a later stage of the particle aging, oxidation of particle components might be chiefly responsible for the observed continuously increasing oxidation of the particles. An oxidation of the oligomer compounds is difficult to determine with the current methods, because an oxidation will likely shift the mass of an oligomer by ∆m 16, which is equal to the repetitive unit of the oligomer series. Ambient Samples. Figure 4a and b show LDI-MS from ambient samples collected in downtown Zurich in summer 2002 and winter 2003, respectively (8). In both spectra a signal is measured at every m/z between m/z 100-550. This chemical background decreases in intensity toward higher masses up to approximately m/z 500-550. Above m/z 550 generally only a few isolated peaks are measured up to approximately m/z 700. In addition to this chemical background, regular ∆m 14 and 16 mass series are measured with slightly higher signal intensities. This structured background, especially dominant in most winter samples, is frequently visible throughout a large part of the mass spectrum starting at approximately m/z 200 and extending up to approximately m/z 400-450. These series are qualitatively similar to the oligomers measured in the smog chamber samples, although they are much less pronounced. It can, therefore, be speculated that the same oligomerization reactions that occur in the smog chamber are also important in the ambient atmosphere. However, due to a greater number of potentially oligomerzing reactants in the ambient atmosphere, there is also a higher number of oligomers which can be formed, resulting in a less pronounced mass series pattern than in the smog chamber. More intensive peaks, up to 10-20 times higher than the structured background, are measured in most samples. These high intensity peaks are much more pronounced in all 20 summer samples (Figure 4a) than in the 25 winter samples (Figure 4b), where only a few high-intensity peaks are visible. In summer, the structured background peaks between m/z 200-500 have on average maximum intensities of about 10% of the highest peaks, whereas in winter the structured background is increased and reaches on average 25% of the highest intensity peaks. All samples, 20 filters in summer and 25 filters in winter, show the seasonal differences described above. These higher intensity peaks are also often grouped in ∆m 14 and 16 series, although at different m/z than the ones of the structured background. It is interesting to note that the same high-intensity peaks are seen in all daytime and nighttime samples over the whole sampling period of one month (August/September 2002). Thus, it seems that a few selected high mass compounds (about 20-40) are present in consistently increased concentrations during summer. This could point to very specific formation processes of the high
FIGURE 4. Typical LDI mass spectra of an aqueous extract of 24hour samples collected at an urban background site in Zurich, Switzerland in (a) summer and (b) winter. A structured chemical oligomer background (- - -) with ∆ m/z 14-16 is present in winter with a relative higher intensity than in summer. molecular weight compounds during summer that are different from formation processes in winter, or to a source that is much more intense during summer. These oligomers seem to be highly stable in the ambient atmosphere since they are observed constantly over several weeks. Similarly, the oligomer distributions observed in the laboratory experiments were very stable over time scales up to 20 h (Figure 2). Thus, the apparent high stability of these oligomers in the atmosphere makes them ideal candidates for tracer compounds of SOA mass in the ambient atmosphere, potentially differentiable into anthropogenic and biogenic fractions. This would allow for a source apportionment of these important organic particle fractions. Up until now, source apportionment for SOA mass is very difficult to estimate and no reliable organic tracers exist for ambient SOA mass. A major challenge to overcome before such oligomers could be used as SOA tracers is an accurate quantification, which is, however, not yet possible with current analytical methods. Further mass spectrometric investigations will need to look in more detail into the chemical structure of these high mass compounds, e.g., with tandem mass spectrometric methods or with high-resolution mass determination.
Acknowledgments We thank Christoph Hueglin at EMPA, Du ¨ bendorf for providing the PM10 sampler for the field measurements, and the smog chamber crew, Josef Dommen, Axel Metzger, and Jonathan Duplissy at the Paul Scherrer Institute for their help during the smog chamber experiments. Financial support by the ETH grant TH-10./01-2 and by the Swiss National Science Foundation (grants 2169-061393.00/2 and 200020-103605/1) is greatly acknowledged.
Literature Cited (1) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics; John Wiley & Sons: New York, 1998.
(2) Rogge, W. F.; Mazurek, M. A.; Hildemann, L. M.; Cass G. R.; Simoneit, B. R. T. Quantification of urban organic aerosol at a molecular levelsidentification, abundance and seasonal variation. Atmos. Environ. 1993, 27, 1309-1330. (3) Puxbaum, H.; Rendl, J.; Allabashi, R.; Otter, L.; Scholes, M. C. Mass balance of the atmospheric aerosol in a South African subtropical savanna (Nylsvley, May 1997). J. Geophys. Res. 2000, 105, 20697-20706. (4) Havers, N.; Burba, P.; Lambert, J.; Klockow, D. Spectroscopic characterization of humic-like substances in airborne particulate matter. J. Atmos. Chem. 1998, 29, 45-54. (5) Zappoli, S.; Andracchio, A.; Fuzzi, S.; Facchini, M. C.; Gelencse´r, A.; Kiss, G.; Kriva´csy, Z.; Molna´r, A.; Me´sza´ros, E.; Hansson, H. C.; Rosman, K.; Zebuhr, Y. Inorganic, organic and macromolecular components of fine aerosol in different areas of Europe in relation to their water solubility. Atmos. Environ. 1999, 33, 2733-2743. (6) Kriva´csy, Z.; Kiss, G.; Varga, B.; Galambos, I.; Sa´rva´ri, Z.; Gelencse´r, A.; Molna´r, A.; Fuzzi, S.; Facchini, M. C.; Zappoli, S.; Andracchio, A.; Alsberg, T.; Hansson, H. C.; Persson, L. Study of humic-like substances in fog and interstitial aerosol by sizeexclusion chromatography and capillary electrophoresis. Atmos. Environ. 2000, 34, 4273-4281. (7) Kiss, G.; Tomba´cz, E.; Varga, B.; Alsberg, T.; Persson, L. Estimation of the average molecular weight of humic-like substances isolated from fine atmospheric aerosol. Atmos. Environ. 2003, 37, 3783-3794. (8) Samburova, V.; Szidat, S.; Hueglin, C.; Fisseha, R.; Baltensperger, U.; Zenobi, R.; Kalberer, M. Seasonal variation of high molecular weight compounds in the water-soluble organic fraction of urban aerosols. J. Geophys. Res. 2005, 110, D23210, doi: 10.1029/ 2005JD005910. (9) Facchini, M. C.; Fuzzi, S.; Zappoli, S.; Andracchio, A.; Gelencse´r, A.; Kiss, G.; Kriva´csy, Z.; Me´sza´ros, E.; Hansson, H. C.; Alsberg, T.; Zebuhr, Y. Partitioning of the organic aerosol component between fog droplets and interstitial air. J. Geophys. Res. 1999, 104, 26821-26832. (10) Tolocka, M.; Jang, M.; Ginter, J.; Cox, F.; Kamens, R.; Johnston, M. Formation of Oligomers in Secondary Organic Aerosol. Environ. Sci. Technol. 2004, 38, 1428-1434. (11) Iinuma, Y.; Boge, O.; Miao, Y.; Sierau, B.; Gnauk, T.; Herrmann, H. Laboratory studies on secondary organic aerosol formation from terpenes. Faraday Discuss. 2005, 130, 279-294. (12) Gao, S.; Ng, N. L.; Keywood, M.; Varutbangkul, V.; Bahreini, R.; Nenes, A.; He, J.; Yoo, K. Y.; Beauchamp, J. L.; Hodyss, R. P.; Flagan, R. C.; Seinfeld, J. H. Particle phase acidity and oligomer formation in secondary organic aerosol. Environ. Sci. Technol. 2004, 38, 6582-6589. (13) Limbeck, A.; Kulmala, M.; Puxbaum, H. Secondary organic aerosol formation in the atmosphere via heterogeneous reaction of gaseous isoprene on acidic particles. Geophys. Res. Lett. 2003, 30, doi: 10.1029/2003GL017738. (14) Hoffer, A.; Kiss, G.; Blazso´, M.; Gelencse´r, A. Chemical characterization of humic-like substances (HULIS) formed from a lignin-type precursor in model cloud water. Geophys. Res. Lett. 2004, 31, L06115, doi: 10.1029/2003GL018962. (15) Kalberer, M.; Paulsen, D.; Sax, M.; Steinbacher, M.; Dommen, J.; Fisseha, R.; Prevot, A. S. H.; Frankevich, V.; Zenobi, R.; Baltensperger, U. Identification of polymers as major components of atmospheric organic aerosols. Science, 2004, 303, 16591662. (16) Paulsen, D.; Dommen, J.; Kalberer, M.; Pre´voˆt, A.; Richter, R.; Sax, M.; Steinbacher, M.; Weingartner, E.; Baltensperger, U. Secondary organic aerosol formation and reaction chamber characterization using irradiations of 1,3, 5 trimethylbenzene-NOx-H2O. Environ. Sci. Technol. 2005, 39, 26682678. (17) Paulsen, D.; Weingartner, E.; Alfarra, M. R.; Baltensperger, U. Volatility measurements of photochemically- and nebulizer- generated organic particles. J. Aerosol Sci. 2006, in press. (18) Maenhaut, W.; Hillamo, R.; Makela, T.; Jaffrezo, J. L.; Bergin, M. H.; Davidson, C. I. A new cascade impactor for aerosol sampling with subsequent PIXE analysis. Nucl. Instrum. Methods Phys. Res. 1996, 109B, 482-487. (19) Schriemer, D. C.; Li, L. A. Mass discrimination in the analysis of polydisperse polymers by MALDI time-of-flight mass spectrometry 0.1. Sample preparation and desorption/ionization issues. Anal. Chem. 1997, 69, 4169-4175. (20) Schriemer, D. C.; Li, L. A. Mass discrimination its the analysis of polydisperse polymers by MALDI time-of-flight mass specVOL. 40, NO. 19, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5921
(21) (22)
(23) (24)
trometry 0.2. Instrumental issues. Anal. Chem. 1997, 69, 41764183. Nielen, M. W. F. Maldi time-of-flight mass spectrometry of synthetic polymers. Mass Spectrom. Rev. 1999, 18, 309-344. Zhu, H. H.; Yalcin, T.; Li, L. Analysis of the accuracy of determining average molecular weights of narrow polydispersity polymers by matrix-assisted laser desorption ionization timeof-flight mass spectrometry. J. Am. Soc. Mass Spectrom. 1998, 9, 275-281. Odian, G. Principles of Polymeriaztion; John Wiley & Sons: New York, 1991. Sax, M.; Zenobi, R.; Baltensperger, U.; Kalberer, M. Time resolved infrared spectroscopic analysis of aerosol formed by photooxidation of 1,3,5-trimethylbenzene and R-pinene. Aerosol Sci. Technol. 2005, 39, 822-830.
5922
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 19, 2006
(25) Baltensperger, U.; Kalberer, M.; Dommen, J.; Paulsen, D.; Alfarra, R.; Coe, H.; Fisseha, R.; Gascho, A.; Gysel, M.; Nyeki, S.; Sax, M.; Steinbacher, M.; Prevot, A.; Sjo¨gren, S.; Weingartner, E.; Zenobi, R. Secondary organic aerosols from anthropogenic and biogenic precursors. Faraday Discuss. 2005, 130, 265-278. (26) Fisseha, R.; Dommen, J.; Sax, M.; Paulsen, D.; Kalberer, M.; Maurer, R.; Ho¨fler, F.; Baltensperger, U. Identification of organic acids in secondary organic aerosol from chamber experiments. Anal. Chem. 2004, 76, 6535-6540.
Received for review December 23, 2005. Revised manuscript received May 5, 2006. Accepted June 6, 2006. ES0525760
