Abstract -cdn.com



Supplementary MaterialMolecular Marker Characterization and Source Appointment of Particulate Matter and Its Organic Aerosols Jong-Kyu Choia,d, Soo-Jin Banb, Yong-Pyo Kimc, Yong-Hee Kimd, Seung-Muk Yia, Kyung-Duk Zoha *a Department of Environmental Health, School of Public Health, Seoul National University, Seoul 151-742, KoreabNational Institute of Environmental Research, Ministry of Environment, Incheon, 404-708, KoreacDepartment of Environmental Science and Engineering, Ewha Womans University, Seoul, 120-750, Koread Research Institute of Public Health & Environment, Incheon Metropolitan city, Incheon, 400-036, KoreaSubmitted to ChemosphereCorresponding AuthorKyung-Duk ZohTel: +82-2-880-2737Fax: +82-2-762-2888Email: zohkd@snu.ac.kr1. Supplementary Tables The analysis methods for the particle-phase organic compounds and instrument conditions GC×GC-TOF/MS for analyzing organic species are reported in Table S1. More detailed descriptions about QA/QC (i.e. MDL, RSD (%), RPD (%), and recovery) were listed in Table S2. Factor loadings from principal component analysis of organic aerosol in PM were also listed Table S2. Table S3 shows factor loadings from principal component analysis of organic aerosol in PM after varimax rotation. Table S4. Pearson correlation coefficients for individual PAHs between TSP and PM2.5The analysis methods for the particle-phase organic compounds were well documented in the following references (Schauer et al. 2002; Sheesley et al. 2004; Choi et al. 2012). In brief, a quarter of the each quartz filter was extracted with 50 mL of dichloromethane, two times sonication, followed by 50 mL of hexane extraction. Before the extraction step, surrogate standard consisting of pyrene-d10, tetracosane-d50, and hexanoic acid-d6 were added to each sample. Insoluble particles from the extracts are removed by filtration over a syringe filter and the intermediate filtrates are concentrated by turbovap to the final volume of 1 mL. After final extraction, each sample was spiked with a series of deuterated internal standards containing tetracosane-d50 and 6-PAHs (naphthalene-d8, acenaphthene-d10, phenanthrene-d10, chrysene-d12, perylene-d12), respectively. Half of the volume of the final extract was methylated using diazomethane. The other half of the volume of the extract was reacted with silylation reagent containing the mixtures of BSTFA, and 1% chlorotrimethylsilane to derivatize COOH and OH groups to the corresponding trimethylsilyl (TMS) esters and ethers, respectively. Individual organic compounds in PM samples were analyzed by GCGC-TOF/MS (Hamilton et al., 2004). Table S1. GC and MS analysis conditionsParameterConfigurationInjectionSplitlessInjection volume2 μLTemperatureInjection Port : 250℃Temperature?program First column ovenRate (℃/min)Target temp (℃)Duration (min)Initial605530020Secondary column ovenRate (℃/min)Target temp (℃)Duration (min)Initial705531520He gas flow1.2 mL/min ColumnFirst column : DB-5MS (cross-linked 5% phenyl methyl silicone 30m,ID;0.25mm, film thickness; 0.25μm)Secondary column : DB-17MS (cross-linked 5% phenyl methyl silicone 1m,ID;0.18mm, film thickness; 0.18 μm)Ionization energy EM volt (1800)TempTransfer line : 300℃, Ion source chamber : 230℃Solvent Delay (min) 3MS Data Collection ModeScanMS Scan Range (amu)35-600Quality assurance and control (QA/QC) Quality assurance and control (QA/QC) procedures were carried out for data certification. The detailed QA/QC data was described in supplemental materials (Table S2). For QA in the analysis of the samples, blank filters were simultaneously examined using the same methods as described above. Background contamination was periodically monitored (every 20 samples) using field blanks that were simultaneously processed with the field samples. The background contamination was less than 5% of the associated samples for all analytes. The relative percent difference (RPD) between sampled concentrations was also used to evaluate the accuracy of measurement for each pollutant and was typically within ±10 % of the standard value. The relative standard deviation (RSD, %) expresses the standard deviation as a percentage of the mean. The RSDs of ionic species, metallic elements, and individual organic species averaged approximately 0.8, 1.4, and 1.4%, respectively. The method detection limit (MDL) was calculated as three times the value of the standard deviation, obtained from seven consecutive analyses of low level samples. The MDL values of ionic species, metallic elements, and individual organic species were estimated to be 0.01~0.05 g/m3, 0.0005~0.004 g/m3, and 0.003~0.079 ng/m3, respectively. Recoveries of ionic species and metallic elements were determined by spiking a standard solution into a blank filter once every 20 samples and the recovery (%) of organic species was calculated from the extraction recovery of the surrogate organic standards spiked. The recoveries were estimated to be 91, 98, 80, 81, and 83% for ionic species, metallic elements, alkanes, alkanoic acids, and polycyclic aromatic hydrocarbons (PAH), respectively. Table S2. Method detection limits, RSD (%), and RPD (%) of target analytes.AnalyteMDLRSDRPDg/m3%%OC 0.255 2.1 1.9WSOC0.027 0.3 0.6WIOC0.024 4.6 0.0Na+0.055 1.0 2.0NH4+0.021 0.2 0.3K+0.048 0.3 0.4Cl-0.021 0.3 0.3NO3-0.013 2.1 3.5SO42-0.047 0.4 0.5Mg0.002 2.1 3.6Al0.0005 1.0 1.7P0.003 2.7 3.6Ca0.004 2.4 4.2Ti0.001 1.0 1.8V0.001 1.2 2.0Cr0.001 1.0 1.8Mn0.004 0.6 1.0Fe0.002 1.7 3.0Ni0.001 0.9 1.6Cu0.002 1.6 2.8Zn0.001 1.2 2.1As0.002 1.4 2.5Se0.002 1.7 2.9Sr0.001 1.6 2.7Cd0.001 1.0 1.8Sn0.001 1.1 1.9Sb0.002 1.4 2.4Pb0.003 0.8 1.31. The method detection limit (MDL) was calculated as three times the value of the standard deviation, obtained from seven consecutive analyses of low level samples. 2. The relative standard deviation (RSD, %) expresses the standard deviation as a percentage of the mean. 3. The RPD (relative percent difference, %) was estimated from two time measurement of sample. Table S2. (continued)_AnalyteMDLRSDRPDng/m3%%Heptadecane 0.0030.90.2 Octadecane0.0480.61.3 Nonadecane0.0110.71.2 Eicosane0.0060.61.0 Docosane0.0091.32.5 Tetracosane0.0051.32.5 Hexacosane 0.0131.92.1 Heptacosane0.0131.92.1 Nonacosane 0.0111.92.1 Dotriacontane0.0090.71.0 Triacontane0.0032.55.0 tetratriacontane0.0042.55.0 Hexanoic acid0.0692.74.8 Heptanoic acid0.0493.05.2 Nonanoic acid0.0501.52.6 Decanoic acid0.0400.80.1 Undecanoic acid0.0701.11.9 Dodecanoic acid0.0581.30.8 Tridecanoic acid0.0331.83.3 Tetradecanoic acid0.0440.71.3 Pentadecanoic acid0.0371.10.3 Hexadecanoic acid0.0302.64.6 Heptadecanoic acid0.0461.63.2 Octadecanoic acid0.0370.90.3 Nonadecanoic acid0.0461.31.8 Eicosanoic acid0.0372.44.1 Heneicosanoic acid0.0522.04.0 Tricosanoic acid0.0581.22.4 Tetracosanoic acid0.0642.65.2 Butanedioic acid0.0441.83.3 Pentanedioic acid0.0371.10.3 Hexanedioic acid0.0302.64.6 Nonanedioic acid0.0461.31.8 Naphthalene0.0151.61.0 Acenaphthene0.0230.81.0 Acenaphthylene0.0210.50.8 Fluorene0.0161.12.2 Phenanthrene0.0161.12.2 Anthracene0.0250.30.4 Fluoranthene0.0190.30.6 Table S2. (continued)AnalyteMDLRSDRPDng/m3%%Pyrene0.0072.14.1 Benzo[a]fluoranthene0.0081.42.7 Benzo[b]fluoranthene0.0041.32.6 Benzo[k]fluoranthene0.0031.32.6 Benzo[a]pyrene0.0031.32.6 Benzo[e]pyrene0.0031.32.6 Benzo[b]triphenylene0.0082.95.7 Benzo[ghi]perylene0.0061.01.2 Chrysene0.0031.42.4 Indeno[1,2,3-cd]pyrene0.0051.01.0 17α(H),21β(H)-(22R)-Homohopane 0.0291.50.6 17α(H),21β(H)-(22S)-Homohopane)0.0350.30.1 17α(H),21β(H)-30-Norhopane0.0312.44.6 17α(H),21β(H)-Hopane0.0361.01.4 17α(H)-22,29,30-Trisnorhopane0.0700.81.5 ααα 20R Cholestane 0.0270.81.5 ααα(20R,24R)-24-Ethylcholestane0.0791.60.1 αββ 20R Cholestane 0.0721.31.1 αββ(20R,24R)-24-Ethylcholestane0.0541.30.1 αββ(20R,24S)-24-Ethylcholestane 0.0251.32.6 9,10-Anthracenedione0.0122.02.6 9H-Fluorenone0.0252.73.4 Benzofuran0.0060.3<0.1 11H-Benzo[a]fluorenone0.0051.32.5 7H-Benzo[c]fluorenone0.0051.32.5 naphtho[1,2-c]furan 0.0051.32.5 Cholestol 0.0141.21.9 Levoglucosan0.0051.32.5 Retene0.0051.32.5 Squalene0.0051.32.5 Dibutyl phthalate0.0051.32.5 Benzothiazole 0.0051.32.5 Dehydroabetic acid0.0461.31.8 Phenanthrene-2methyl0.0151.61.0 Phenanthrene-3methyl0.0151.61.0 Phenanthrene-1methyl0.0151.61.0 Phenanthrene-1,7dimethyl0.0161.61.0 Pyrene-1methyl0.0072.14.1 Pyrene-4methyl0.0072.14.1 Chrysene-1methyl0.0031.42.4 1,2-Benzenecaboxylic acid0.0461.31.8Table S3. Factor loadings from principal component analysis of organic aerosol in PM after varimax rotationF1F2F3F4F5F6F7F8F9F10Combustion 1(LMW-PAHs)0.880 Biomass burning -0.330 0.354 Vegetative detritus 0.891 SOA 1 0.780 SOA 2 0.322 0.819 Combustion 2(HMW-PAHs)0.859 0.355 Motor vehicle 0.808 OC 0.653 0.443 0.300 EC-0.364 0.319 0.528 SOC0.362 0.660 0.433 POC-0.362 0.322 0.536 WSOC0.425 0.493 0.559 WIOC 0.618 Na -0.487 0.324 0.415 NH4 0.495 0.767 K 0.359 0.711 Cl0.517 0.566 NO3 0.512 0.710 SO4 0.743 Mg 0.824 Al 0.392 0.333 0.498 -0.326 P0.377 -0.352 0.398 Ca 0.823 Ti 0.844 V 0.711 Cr0.330 0.703 Fe 0.856 Mn 0.317 0.752 Ni-0.369 0.665 Cu 0.494 Zn 0.497 0.311 0.357 As 0.817 Pb 0.325 0.433 0.369 N-HEXD 0.656 N-HEPD 0.501 0.344 N-OCTD0.772 0.457 N-NONAD0.871 N-EICO0.778 N-HENEI0.865 N-DOCO0.721 0.360 0.382 N-TRICO0.625 0.314 N-TETRACO 0.613 0.570 N-PENTACO 0.619 0.544 N-HEXACO 0.896 N-HEPTACO 0.632 0.535 N-OCTACO 0.858 N_NONACO 0.348 N_TRICO 0.865 N_DOTRICO 0.866 Table S3. (continued) F1F2F3F4F5F6F7F8F9F10FLU0.919 PYR0.917 B(A)F0.851 0.369 B(B)F0.842 0.351 B(K)F0.653 0.359 BGHIPE0.807 0.461 CHRYSN0.888 0.331 INCDPY0.835 0.403 BA30NH 0.557 0.381 AB_HOP 0.691 HOPANE 0.678 0.350 CHOLESTANE 0.574 N_HEXDA 0.428 0.688 N_HEPDA 0.350 0.569 N_OCTDA 0.301 N_NONDA 0.525 N_EICOA 0.381 0.803 N_HENEICOA 0.643 N_TRICOSA 0.791 N_TETRACOSA0.353 0.815 9H-FLUORENE0.878 CHOLESTEROL0.359 0.318 LOVOGUCOSAN -0.328 0.358 RETENE0.909 SQUALENE0.560 -0.312 0.323 DB PHTHA 0.777 BENZOTHIO 0.408 0.313 0.493 NAPHTHFUR 0.604 0.560 BUTANDIOA 0.819 PENTADIOA 0.803 NONANDIOA cis-PINOIC ACID 0.358 OlLEIC ACID 0.702 DEHYDROABIEA 0.774 0.420 CO0.440 0.771 SO20.442 0.553 O3-0.450 -0.592 NO0.396 0.822 NO2 0.792 WS -0.315 -0.505 -0.33 TEMP-0.746 0.408 HUM RAD-0.349 0.364 -0.338 Table S4. Pearson correlation coefficients for individual PAHs between TSP and PM2.5TSPPM2.5FLUORAPYRENEB(A)FB(B)FB(K)FBGHIPECHRYSNINCDPYFLUORAPYRENEB(A)FB(B)FB(K)FBGHIPECHRYSNINCDPYTSPFLUORA1.00 1.00**0.99** 0.99** 0.88** 0.97** 0.99**0.98**0.96** 0.96**0.55 0.84** 0.84** 0.91** 0.78**0.55 PYRENE　1.00 0.99** 0.99** 0.91** 0.98** 1.00**0.98** 0.96** 0.96**0.57 0.85** 0.85** 0.92** 0.77** 0.52 B(A)F　　1.00 0.99** 0.87** 0.98** 0.99**0.99** 0.95** 0.95** 0.55 0.86** 0.86** 0.90** 0.74** 0.57 B(B)F　　　1.00 0.91** 0.99** 0.99**0.99** 0.97** 0.98**0.63* 0.88** 0.88** 0.91** 0.80**0.60* B(K)F　　　　1.00 0.92** 0.92**0.89** 0.91** 0.92**0.72* 0.86** 0.86** 0.90** 0.72** 0.48 BGHIPE　　　　　1.00 0.99** 0.99** 0.94** 0.95** 0.69* 0.90** 0.90** 0.89** 0.71** 0.56 CHRYSN　　　　　　1.00 0.99** 0.96** 0.96** 0.61* 0.87** 0.87** 0.92** 0.74** 0.53 INCDPY　　　　　　　1.00 0.95** 0.95** 0.64* 0.90** 0.90** 0.90** 0.71** 0.55 PM2.5FLUORA　　　　　　　　1.00 1.00** 0.62* 0.88** 0.88** 0.95** 0.88** 0.67* PYRENE　　　　　　　　　1.00 0.67* 0.91** 0.91** 0.94** 0.86** 0.66* B(A)F　　　　　　　　　　1.00 0.79** 0.79** 0.59* 0.40 0.46 B(B)F　　　　　　　　　　　1.00 1.00 0.86** 0.67* 0.53 B(K)F　　　　　　　　　　　　1.00 0.86** 0.67* 0.53 BGHIPE　　　　　　　　　　　　1.00 0.79** 0.66* CHRYSN　　　　　　　　　　　　　　1.00 0.68* INCDPY　　　　　　　　　　　　　　　1.00 2. Supplemental figures Fig. S1. Location of the study sites in Incheon, KoreaFig. S2. The diagonistic factor of PMF model using 41 molecular markers only. (a) IM, IS, and rotational freedom as a function of the factors chosen in PMF, (b) Q-value for the different factor solutions and the change of “FPEAK” parameter.Fig. S3. The diagonistic factor of PMF model using traditional 21items couple with 41 molecular markers. (a) IM, IS, and rotational freedom as a function of the factors chosen in PMF, (b) Q-value for the different factor solutions and the change of “FPEAK” parameter.Fig. S4. Source profiles obtained from organic data (prediction ± standard deviation) using 41organic marker species in Incheon, Korea.Fig. S5. Timeseries plot for each source contribution to OC mass concentrationscalculated from PMF model using 41 organic marker species.Fig. S6. Source profiles obtained from TSP samples (prediction ± standard deviation) using 63species in Incheon, Korea.Fig. S7. Timeseries plot for each source contribution of TSP using 63species in Incheon, KoreaFig. S8. The source contributions (%) of identified sources to TSP mass concentrations calculated from PMF model using 63species.Fig.S1. Location of the study sites in Incheon, Korea- PMF analysis for Organic Carbon using 41 molecular markersPMF diagnostics (e.g., model error, Q and rotational ambiguity, rotmat) were based on those described by Lee et al. (1999). We investigated the Q-value for different numbers of factors and values of the rotational parameter (FPEAK), as well as variations in the maximum individual column mean (IM), the maximum individual column standard deviation (IS), and rotational freedom for the different factors used in PMF models (see Fig. S2(a) and S2(b)). As the number of factors approached a critical value, IM and IS clearly decreased. We also investigated the maximum rotmat, which exhibited a significant increase from seven to ten factors (Figs. S2). (a)(b)Fig.S2. The diagonistic factor of PMF model using 41 molecular markers only. (a) IM, IS, and rotational freedom as a function of the factors chosen in PMF, (b) Q-value for the different factor solutions and the change of “FPEAK” parameter.PMF analysis for TSP using 62 compounds (traditional 21 items + 41 molecular markers)(a)(b)Fig.S3.The diagonistic factor of PMF model using traditional 21items couple with 41 molecular markers. (a) IM, IS, and rotational freedom as a function of the factors chosen in PMF, (b) Q-value for the different factor solutions and the change of “FPEAK” parameter.Fig. S4. Source profiles obtained from organic data (prediction ± standard deviation) using 41organic marker species in Incheon, Korea.Fig. S5. Timeseries plot for each source contribution to OC mass concentrationscalculated from PMF model using 41 organic marker species.Fig.S6. Source profiles obtained from TSP samples (prediction ± standard deviation) using 63 species in Incheon, Korea.Fig.S7. Timeseries plot for each source contribution of TSP using 63 species in Incheon, Korea.Fig.S8. The source contributions (%) of identified sources to TSP mass concentrations calculated from PMF model using 63 species. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download