Environ. Sci. Technol. 2007, 41, 99-105
Chemical Compositions of Fine Particulate Organic Matter Emitted from Chinese Cooking YUNLIANG ZHAO,† MIN HU,* SJAAK SLANINA, AND YUANHANG ZHANG State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences, Peking University, Beijing, 100871, P.R. China
Food cooking can be a significant source of atmospheric particulate organic matter. In this study, the chemical composition of particulate organic matter (POM) in PM2.5 emitted from four different Chinese cooking styles were examined by gas chromotography-mass spectrometry (GC-MS). The identified species are consistent in the emissions from different Chinese cooking styles and the quantified compounds account for 5∼10% of total POM in PM2.5. The dominant homologue is fatty acids, constituting 73∼85% of the quantified compounds. The pattern of n-alkanes and the presence of β-sitosterol and levoglucosan indicate that vegetables are consumed during Chinese cooking operations. Furthermore, the emissions of different compounds are impacted significantly by the cooking ingredients. The candidates of organic tracers used to describe and distinguish emissions from Chinese cooking in Guangzhou are tetradecanoic acid, hexadecanoic acid, octadecanoic acid, oleic acid, levoglucosan, mannosan, galactosan, nonanal, and lactones. During the sampling period, the relative contribution of Chinese cooking to the mass concentration of atmospheric hexadecanoic acid should be less than 1.3% in Guangzhou.
Introduction Organic aerosols are composed of hundreds of organic compounds, some of them are carcinogenic and mutagenic (1-3), some can impact atmospheric chemical processes, and all particles with suitable characteristics have a large influence on solar radiation (4-7). Previous studies (8-13) have reported that food cooking is a significant source of atmospheric particulate organic matter. Emissions from meat cooking contribute about 20% of the concentration of particulate organic matter (POM) in PM2.5 in Los Angeles area (8-11). In addition, 7% of the total primary emissions of C16 and C18 n-alkanoic acids are attributed to commercial deep frying operations (12); and 20∼75% of the ambient concentrations of polycyclic aromatic hydrocarbons (PAHs) are from meat cooking (13). The composition of particles emitted from cooking operations is strongly dependent on cooking procedures, such as cooking temperature, ingredients, duration, and other factors (8, 14-17). The differences are obvious when the compounds released by Chinese cooking operations are * Corresponding author phone: +86-10-62759880; e-mail:
[email protected]. † Current address: Department of Earth and Environmental Sciences, RutgerssNewark, NJ, 07102.. 10.1021/es0614518 CCC: $37.00 Published on Web 11/22/2006
2007 American Chemical Society
compared with those released by meat cooking operations; for example, both monosaccharide anhydrides and β-sitosterol are identified in Chinese cooking, but not in U.S. meat cooking (15, 17). These differences mainly result from cooking ingredients. Beef is the main ingredient in meat cooking (15, 16); however, pork, poultry, beef, seafood, and vegetables are regularly consumed in Chinese cooking (18). Previous studies of Chinese cooking mainly focused on special compounds in fumes of cooking, like PAHs (19-26), except for the study conducted by He et al. (17), which extended the scope of the chemical compounds to n-alkanes, fatty acids, dicarboxylic acids sterols, and PAHs in the emissions from Hunan and Cantonese cooking styles. Chinese cooking can be characterized by eight representative cooking styles with different cooking ingredients and methods. The differences of compositions of POM in emissions caused by these differences among Chinese cooking styles must be taken into account when molecular markers are selected. The aims of this study are to investigate the organic chemical compositions of PM2.5 produced by different Chinese cooking styles in more detail and to look for candidates of organic tracers for emissions of Chinese cooking. To meet these demands, samples were collected from four different commercial Chinese restaurants with distinct characteristics of different cooking styles; meanwhile, atmospheric PM2.5 was also collected in the same city.
Experimental Section Sampling. Four representative Chinese restaurants were selected in Guangzhou city: Cantonese style, Hunan style, Sichuan style, and Dongbei style restaurant. All of the four tested restaurants are commercial, and the size of these restaurants is above average. The differences among Chinese cooking styles include methods, ingredients, and seasonings for cooking. Cantonese style is based on fresh local ingredients and lightly seasoned stir-frying, quick-frying, sauteing, and stewing. Hunan style is famous for spicy and sour dishes, mainly using steaming, simmering and stir-frying. Sichuan style, famous for its sharp and spicy taste, mainly employs sauteing and stewing. Dongbei style is like home cooking and famous for pot-roast, and mainly uses simmering, marinating, flavor-potting, and roasting. The cooking oil used by each restaurant varies: Cantonese style with peanut oil and blended oil; Sichuan style with blended oil; Dongbei style with Canola oil and peanut oil; Hunan style with peanut oil. The exhausts from the tested restaurants were treated by different methods: the Cantonese style restaurant by a train of static, water-washing, and activated carbon; the other three restaurants by water-washing. The smoke temperature was about 40 °C at the vent of exhausts. To get sufficient loading, two medium-volume samplers with 78L min-1of flow rate (Dizhi Instrument Factory, China) were deployed at the vent of exhausts. Six samples were collected in the single restaurant during rush hours of lunch (three samples) and supper (three samples) on quartz fiber filters (90 mm diameter, Whatman) and the single sampling lasted 2 h. Before sampling, the samplers were cleaned with Milli-Q water; and the filter holders and open ends were wrapped by clean foil immediately after cleaning. In addition, quartz fiber filters were baked at 450 °C for 4.5 h to remove any organic contaminants. After sampling, samples were stored in a refrigerator (-13 °C) within 2 h. The relative concentrations of PM2.5 and PM10 emitted from restaurants were monitored by two Dustraks (TSI) samplers. During the sampling period, atmospheric PM2.5 were also collected in the same city by means of a VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
99
FIGURE 1. Relative concentrations of PM2.5 and the ratios of PM2.5 to PM10 (Cantonese style). Hi-volume sampler (Andersen). Organic carbon and element carbon were measured by carbon analyzer (Sunset lab) and molecular compounds were measured by gas chromotography-mass spectrometry (GC-MS). Sample Extraction. Each group of filters was spiked with a known amount of perdeuterated tetracosane (n-C24D50), perdeuterated benzoic acid (C7HO2-D5) and perdeutrated anthracene (C14D10) (Aldrich) as recovery standards prior to extraction. The amount added was calculated based on the amount of organic matter in the extracted filters. Samples were extracted by three successive portions of dichloromethane and methanol (Tedia) (3:1; V:V) for three 15-min periods with ultrasonic extractors at room temperature, and the solvent was changed every 15 min. After filtration, the extracts were distilled under lightly negative pressure and reduced to 3∼5 mL; then the solution was concentrated to 1 mL under a high purity nitrogen gas stream. The extract of 1 mL was divided into three fractions. One fraction was analyzed directly by GC-MS to attain the concentrations of nonpolar organic compounds, like n-alkanes. Another fraction was derivatized by Bis(trimethylsilyl)trifluoroacetamide (BSTFA) plus 1% trimethylchlorosilane (TMCS) (Supelco) at 70 °C for 2 h, which was analyzed to attain the concentrations of polar organic compounds, like n-alkanoic acids. The third one was stored at 4 °C as backup. GC-MS Analysis. The samples were analyzed by GCMS (Agilent, GC model 6890plus, MSD model 5973N) using a 60m × 0.25 mm diameter DB-5MS capillary column (0.25 µm film thickness, J&W scientific). Hexamethylbenzene (Aldrich) was spiked as a co-injection standard for all samples. The temperature program for GC consisted of a 10-min isothermal hold at 60 °C followed by ramp of 6 °C min-1 to 300 °C, finally isothermal hold at 300 °C for 50 min. The MS analyses were performed at electron energy of 70 eV. The major organic compounds were initially identified by computer matching to standard reference mass fragmentograms (National Institute of Standards and Technology (NIST) library), then confirmed by authentic standards. The recovery, calculated by recovery standards, ranges from 46 to 120%, mainly from 69 to 114%, during extraction and analysis. The relative standard deviation of standard chemicals is within 10%. In this study, the detected compounds in the field and laboratory blanks are nonanoic acid and hexadecanoic acid. Compared with their concentrations in the samples, their concentrations in the blanks were too low to impact the concentrations of interested compounds; therefore, the blanks were not subtracted from the samples. Compounds Quantification. Quantification was performed using GC-MS by an authentic standard relative to co-injection standard (Hexamethylbenzene). For some organic compounds where no authentic standards were available, a secondary quantification was used. Compounds were selected based on similar retention time and molecular structure. 100
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 1, 2007
Results and Discussion Mass Size Distribution of Particles. In the case of emissions of Chinese cooking operations, two distinct emission peaks are observed, about 11:30-13:30 and 18:00-20:00. Though the mass concentrations of PM2.5 from different restaurants vary significantly (639∼4854 µg/m3), and the concentrations of PM2.5 from the same restaurant also vary significantly among each sampling period, the ratios of PM2.5 to PM10 are above 0.8 (Figure 1). In addition, POM (OC × 1.4) accounts for 84% of mass concentrations of PM2.5 from tested Chinese restaurants. Chemical Compositions of Fine Particles. The identified homologues include n-alkanes, fatty acids, dicarboxylic acids, n-alkanals, n-alkanones, PAHs, sterols, monosaccharide anhydrides, and other compounds (Table 1). The quantified compounds from different restaurants account for approximate fraction of POM in PM2.5: Cantonese style (5.1 ( 0.8%), Sichuan style (10.0 ( 3.0%), Dongbei cooking (6.9 ( 0.9%), Hunan cooking (6.1 ( 1.1%). They are in the same order of magnitude as that in meat cooking (about 5∼7%) (15, 16); however, they are less than those of Cantonese and Hunan styles in Shenzhen (above 20%). The difference can be attributed to the quantity of quantified fatty acids. He et al. (17) reported fatty acids accounted for 90% of quantified compounds in Shenzhen Chinese cooking; while just 73∼81% in this study. N-alkanes constitute a minor fraction, but their distributions indicate that vegetables are consumed. Both sterols and monosaccharide anhydrides are detected in the emissions of Chinese cooking, but only cholesterol is reported in the studies of meat cooking (15, 16). Nonanedioic acid is the dicarboxylic acid of highest concentration. N-alkanes. Normal alkanes from C14 to C33 are quantified. Their distributions as the function of carbon number are similar among four tested restaurants (Figure 2). If the concentration ratio of odd-to-even number homologues is defined as the carbon preference index(CPI), the distributions of n-alkanes (carbon number > 23) from tested restaurants show distinct odd-to-even carbon preference (CPI ) 6.2∼10.1) and Cmax at C29 or C31, which indicate the origin of n-alkanes (carbon number > 23) is the high plant wax (27, 28). The distributions of n-alkanes of Cantonese style and Hunan style in Shenzhen also show odd-to-even preference above C23 and Cmax at C29, but the distributions below C23 are dissimilar to those in Guangzhou Chinese cooking (17). In Guangzhou, Cantonese style has the highest emissions, followed by Sichuan, Hunan, and Dongbei styles. Cantonese and Hunan styles in Shenzhen cause lower emissions than those in Guangzhou, while Cantonese style releases more than Hunan style in both cities (17). N-alkanoic and -Alkenoic Acids. Normal fatty acids with even carbon numbers from C4 to C34 as triglycerides and phospholipids are found in high concentrations in animal and vegetable fats; C16 and C18 fatty acids, including
TABLE 1. Average Concentrations of Organic Compounds from Chinese Cooking (ng/mg of POM) Cantonese style n-tetradecane n-pentadecane n-hexadecane n-heptadecane n-octadecane n-nonadecane n-eicosane n-heneicosane n-docosane n-tricosane n-tetracosane n-pentacosane n-hexacosane n-heptacosane n-octacosane n-nonacosane n-triacontane n-hentriacontane n-tritriacontane
Sichuan style
N-Alkanes 19 ( 17 23 ( 9 70 ( 11 75 ( 11 128 ( 33 43 ( 10 139 ( 44 50 ( 12 71 ( 24 55 ( 19 72 ( 14 75 ( 35 97 ( 25 65 ( 19 167 ( 80 104 ( 61 192 ( 64 143 ( 76 214 ( 71 161 ( 71 135 ( 33 102 ( 54 292 ( 102 303 ( 105 81 ( 22 59 ( 43 181 ( 60 116 ( 41 62 ( 69 45 ( 69 552 ( 223 212 ( 79 16 ( 25 31 ( 23 560 ( 197 367 ( 149 26 ( 41 90 ( 55
phenanthrene fluoranthene pyrene retene benzo[ghi]fluoranthene chrysene benzo[b]fluoranthene benzo[k]fluoranthene benzo[a]fluoranthene benzo[b]pyrene benzo[a]pyrene perylene indeno[1,2,3-c,d]pyrene benzo[ghi]pyrene
Polycyclic Aromatic Hydrocarbons 7(3 6(2 121 ( 36 238 ( 97 243 ( 71 466 ( 191 24 ( 14 1(3 160 ( 30 119 ( 49 59 ( 24 23 ( 8 48 ( 30 10 ( 3 40 ( 35 8(3 11 ( 17 nd 52 ( 30 7(4 39 ( 42 1(2 1(4 nd 65 ( 58 4(7 196 ( 116 27 ( 11
nonanal decanal undecanal dodecanal tridecanal tetradecanal pentadecanal
1365 ( 393 49 ( 81 33 ( 82 8 ( 21 nd 62 ( 69 62 ( 69
N-Alkanals 2281 ( 443 272 ( 80 154 ( 92 147 ( 100 117 ( 72 235 ( 59 305 ( 123
2-nonanone 2-decanone 2-undecanone 2-dodecanone 2-tridecanone 2-tetradecanone 2-pentadecanone 2-hexadecanone 2-heptadecanone
N-Alkanones 116 ( 90 326 ( 90 116 ( 91 361 ( 108 58 ( 50 259 ( 72 nd 102 ( 42 nd 35 ( 86 nd 23 ( 56 234 ( 152 592 ( 262 nd 16 ( 26 667 ( 180 1003 ( 489
5-propyldihydro-2(3H)-furanone 5-butyldihydro-2(3H)-furanone 5-pentyldihydro-2(3H)-furanone 5-hexyldihydro-2(3H)-furanone 5-heptyldihydro-2(3H)-furanone 5-octyldihydro-2(3H)-furanone 5-nonyldihydro-2(3H)-furanone 5-decyldihydro-2(3H)-furanone 5-undecyldihydro-2(3H)-furanone 5-dodecyldihydro-2(3H)-furanone 5-tridecyldihydro-2(3H)-furanone 5-tetradecyldihydro-2(3H)-furanone
6 ( 16 21 ( 45 20 ( 42 8 ( 21 5 ( 11 12 ( 20 4(9 43 ( 40 50 ( 56 809 ( 125 19 ( 23 329 ( 78
tetradecanamide hexadecanamide 9-octadecenamide octadecanamide
nd 376 ( 99 185 ( 73 129 ( 33
Lactones
70 ( 41 188 ( 96 221 ( 138 90 ( 44 65 ( 28 74 ( 37 71 ( 24 117 ( 46 179 ( 74 1401 ( 466 58 ( 22 311 ( 112
Amides
3(8 249 ( 57 112 ( 23 48 ( 25
Dongbei style
Hunan style
notes
24 ( 6 93 ( 13 44 ( 25 45 ( 13 57 ( 33 69 ( 17 36 ( 30 91 ( 35 34 ( 41 80 ( 13 54 ( 20 102 ( 89 12 ( 20 59 ( 19 nd 63 ( 43 nd 96 ( 51 20 ( 32
12 ( 3 46 ( 9 21 ( 15 29 ( 6 23 ( 6 39 ( 8 39 ( 17 75 ( 22 65 ( 59 91 ( 26 66 ( 18 205 ( 46 25 ( 21 74 ( 13 nd 207 ( 68 nd 346 ( 45 nd
a a b a a a a a a a a a a a a a a a a
5(1 56 ( 10 76 ( 19 nd 20 ( 4 12 ( 3 9(4 6(4 1(2 5(6 3(4 nd 6(6 18 ( 9
4(1 28 ( 16 49 ( 12 nd 13 ( 3 10 ( 2 7(2 5(3 nd 6(2 4(2 nd 3(6 19 ( 8
b a a a a a a a a a a a a a
4501 ( 2735 405 ( 86 255 ( 46 287 ( 83 193 ( 54 359 ( 50 436 ( 104
1373 ( 519 174 ( 34 109 ( 25 120 ( 46 99 ( 37 141 ( 39 232 ( 75
a b b b b b b
504 ( 44 589 ( 45 386 ( 39 151 ( 19 64 ( 100 103 ( 54 844 ( 181 39 ( 51 1383 ( 330
213 ( 43 235 ( 46 197 ( 44 69 ( 12 20 ( 49 65 ( 95 401 ( 138 8 ( 19 594 ( 134
b b b b b b b b b
86 ( 35 172 ( 47 288 ( 135 119 ( 39 102 ( 34 120 ( 35 91 ( 11 157 ( 35 147 ( 42 1260 ( 182 7 ( 11 178 ( 35
53 ( 17 177 ( 67 134 ( 44 56 ( 15 39 ( 9 94 ( 33 37 ( 9 58 ( 30 84 ( 14 739 ( 135 15 ( 13 184 ( 37
b b b b b b b b b b b b
3(6 494 ( 137 165 ( 54 53 ( 28
nd 172 ( 47 92 ( 23 43 ( 13
b b b b
VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
101
TABLE 1. (Continued) Cantonese style
Sichuan style
Dongbei style
Hunan style
notes
hexanoic acid heptanoic acid octanoic acid nonanoic acid decanoic acid undecanoic acid dodecanoic acid tridecanoic acid tetradecanoic acid (myristic acid) pentadecanoic acid hexadecanoic acid (palmitic acid) heptadecanoic acid octadecanoic acid (stearic acid) nonadecanoic acid eicosanoic acid docosanoic acid tetracosanoic acid
Saturated Fatty Acids 85 ( 74 159 ( 45 13 ( 20 59 ( 11 32 ( 32 130 ( 32 60 ( 46 208 ( 35 14 ( 16 60 ( 18 3(7 37 ( 11 116 ( 60 247 ( 86 25 ( 28 108 ( 47 341 ( 114 1635 ( 809 102 ( 27 437 ( 200 10608 ( 2238 30121 ( 7982 118 ( 41 285 ( 121 4209 ( 996 6084 ( 1608 5(8 43 ( 10 257 ( 74 400 ( 119 355 ( 94 490 ( 189 22 ( 35 86 ( 80
144 ( 112 67 ( 29 137 ( 87 417 ( 243 81 ( 27 73 ( 26 208 ( 79 113 ( 34 801 ( 113 283 ( 43 23344 ( 3547 151 ( 28 2876 ( 533 29 ( 16 191 ( 33 52 ( 52 10 ( 24
103 ( 39 50 ( 18 96 ( 27 154 ( 38 44 ( 14 30 ( 7 305 ( 113 61 ( 15 730 ( 188 224 ( 80 14757 ( 2947 162 ( 37 3975 ( 799 23 ( 7 235 ( 54 268 ( 55 71 ( 29
a a a a a a a a a a a a a a a a a
9-hexadecenoic acid (palmitoleic acid) 9,12-octadecadienoic acid (Linoleic acid) 9-octadecenoic acid (oleic acid)
Unsaturated Fatty Acids 101 ( 39 255 ( 96 8677 ( 2233 13547 ( 5345 13775 ( 2663 29375 ( 8307
108 ( 37 3077 ( 904 18828 ( 3256
196 ( 60 10132 ( 2498 18044 ( 2723
b a a
Dicarboxylic Acids 49 ( 47 225 ( 64 4(7 80 ( 28 15 ( 17 129 ( 38 84 ( 42 246 ( 65 131 ( 75 413 ( 177 675 ( 284 1890 ( 828 97 ( 21 179 ( 62 14 ( 18 82 ( 43
262 ( 159 91 ( 42 109 ( 31 197 ( 73 189 ( 72 1043 ( 671 156 ( 43 62 ( 24
166 ( 43 56 ( 16 70 ( 13 154 ( 57 201 ( 61 975 ( 325 107 ( 29 50 ( 21
b a a a a a a b
30 ( 11 14 ( 9 554 ( 296
4(4 3(2 124 ( 33
a a a
293 ( 31 114 ( 19 84 ( 17
1080 ( 286 315 ( 84 516 ( 133
a a a
20 ( 6 170 ( 228
17 ( 4 82 ( 40
a b
butanedioic acid pentanedioic acid hexanedioic acid heptanedioic acid octanedioic acid nonanedioic acid decanedioic acid undecanedioic acid galactosan mannosan levoglucosan
Monosaccharide Anhydrides nd 20 ( 7 nd 7(3 282 ( 147 218 ( 56
β-sitosterol cholesterol stigmasterol
1168 ( 181 261 ( 61 619 ( 104
benzoic acid funan, 2-pentyl-
Sterols
1313 ( 527 353 ( 125 621 ( 245
Other Compounds 39 ( 11 30 ( 6 237 ( 127 237 ( 119
a Notes: (1) a, authentic quantitative standard; b, authentic quantitative standard for similar compounds; nd, not detected; (2) Standard deviation is calculated based on six samples which were collected from the same restaurant.
FIGURE 3. Distributions of fatty acids emitted from tested restaurants.
FIGURE 2. Distributions of n-alkanes emitted from the tested restaurants. saturated and unsaturated fatty acids, are major constituents of triglycerides in seed oils and fats of higher land animals (29, 30). Saturated fatty acids from C6 to C24 are quantified. The distribution of saturated fatty acids shows a distinct evento-odd preference and palmitic acid displays the highest emission (Figure 3). This distribution as the function of carbon 102
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 1, 2007
number is similar to those in meat cooking (15), seed oil cooking (12), and Chinese cooking in Shenzhen (17). The two most common unsaturated fatty acids are oleic acid and linoleic acid. Of Guangzhou Chinese restaurants, Cantonese style produces the lowest amount of saturated and unsaturated fatty acids, with only 39 µg of fatty acids/mg of POM. Hunan style and Dongbei style produced 50 and 51 µg of fatty acids/ mg of POM. Sichuan style releases about 2-fold more fatty acids compared with Cantonese style. Cantonese style and Hunan style in Shenzhen release more fatty acids of per mg of POM than in Guangzhou.
Regarding fatty acids, the compound of the highest concentration is different among Guangzhou restaurants, oleic acid in Cantonese style and Hunan style, palmitic acid in Dongbei style; the concentration of palmitic acid is approximate to that of oleic acid in Sichuan style. The compound with the highest concentration in Shenzhen Chinese cooking is linoleic acid (17). Dicarboxylic Acids. Dicarboxylic acids are measured in the range from C4 to C11 with nonanedioic acid being the predominant one. Their distributions in four tested restaurants in Guangzhou are similar to each other and to those in Chinese restaurants in Shenzhen (17). Meat cooking produced dicarboxylic acids ranging from C4 to C8 and hexanedioic acid had the highest concentration (15); and only hexanedioic acid and octanedioic acid were reported in study of cooking with seed oils and the concentration of C8 is higher (12). Dicarboxylic acids are the oxidation products of dialdehydes formed during the auto-oxidation process of unsaturated lipids (15). The total concentrations of dicarboxylic acids show different patterns from that of unsaturated fatty acids in the emissions from four tested restaurants, but the patterns of the concentrations of nonanedioic acid in four restaurants are in agreement with those of oleic acid and linoleic acid. Polycyclic Aromatic Hydrocarbons. Organic substances containing carbon and hydrogen produce PAHs during incomplete combustion or pyrolysis. The percentages of PAHs in total quantified compounds are small in the emissions of Chinese cooking, like PAHs emitted from Chinese restaurants in Shenzhen constituting no more than 0.15% of quantified compounds (17). In this study, Cantonese style releases the highest concentration of PAHs with 540 ng/mg of POM and accounts for 2% of total quantified POM. The second highest concentration of PAHs occurs in Sichuan style (430 ng/mg of POM), then Dongbei style (100 ng/mg of POM). The concentration decreases by 8-fold in Hunan style compared with Cantonese style. The compound of the highest concentration is Pyrene in tested restaurants. He et al. (17) reported that Pyrene was also the highest one in Chinese restaurants in Shenzhen. However, Chrysene is the compound with the highest concentration, released from meat cooking and seed oils cooking (12, 15, 16). Benzo[a]pyrene, a known carcinogen, is also detected in the emissions from Chinese restaurants in Guangzhou. Molecular Biomarkers. Molecular biomarkers are organic compounds of biological origin that have the same chemical structure as the original natural products found in living organisms. Such molecules are characterized by their restricted occurrence, source specificity, molecular stability, and suitable concentration for analytical detection (31). The major biomarkers identified in the Chinese restaurants of Guangzhou are sterols and monosaccharide anhydrides. Sterols. Sterols are widely present in animal and vegetable body tissues. Plant lipid membranes and waxes are generally comprised of the C28 and C29 phytosterol compounds, like β-sitosterol and stigmasterol; cholesterol is biosynthesized by higher animals and found in all body tissues, especially in animal fats and oils (15, 30, 32, 33). β-sitosterol is the most common sterol in PM2.5 emitted from Chinese cooking of Guangzhou, followed by cholesterol and stigmasterol. The pattern is consistent with the results of the study of Chinese cooking in Shenzhen (17). However, Cholesterol is the only sterol detected in meat cooking (1416) and has served as tracers to estimate the contribution of meat cooking to atmospheric particles (9). In POM emitted from Guangzhou tested restaurants, the concentrations of the sterols have minor differences between restaurants except for Dongbei style.
Monosaccharide Anhydrides. The major product of the breakdown of cellulose is levoglucosan, accompanied by galactosan and mannosan. Of monosaccharide anhydrides, levoglucosan has been regarded as the tracer of the products from wood burning (34-37). No monosaccharide anhydrides are reported in the emissions of meat cooking (15, 16). Levoglucosan is the most common monosaccharide anhydride detected in POM emitted from Chinese cooking, followed by galactosan and mannosan. The concentration of levoglucosan emitted from Chinese cooking is much less than those released by field burning of straws and wood (32, 33, 38), but the average ratio ()12) of levoglucosan/ (mannosan + galactosan) for Chinese cooking operations are greater than the average ratio ()3) for field burning of straws and wood. The ratio for Chinese cooking is close to that ()18) for family fireplaces (39-41), but these cases are few in the cities of China.
Other Compounds Lactones. Lactones are detected in POM emitted from the tested restaurants in Guangzhou, ranging from C7 to C18. They can serve as tracers for the emissions from food cooking (12, 15). N-alkanals and N-alkanones. Seven n-alkanals ranging from C9 to C15 are detected in the POM emitted from the Guangzhou restaurants. The distributions of n-alkanals in emissions from different tested restaurants are similar. Nonanal is the compound of the highest concentration among n-alkanals series, and the concentrations of other n-alkanals are close to each other. Of all identified n-alkan2-ones, the concentration of 2-pentadecanone is much greater than other homologues, which is also the prominent one of the n-alkan-2-ones produced by seed oil cooking (12). The cause may be that cooking oils contain a large amount of triglycerides (15).
Comparison of Different Cooking Styles Though the concentrations of compounds vary significantly between the tested restaurants in Guangzhou, the total percentages of quantified compounds and types of homologues detected are approximately consistent and fatty acids contribute the same level to the concentrations of POM in PM2.5. However, the predominant compound varies among different tested restaurants; the reason can be that the ratio of meat (with a higher content of saturated fats than monounsaturated fats) to cooking oil varies since cooking oils used by each restaurant contain more monosaturated fats than saturated fats or the same amount; and the boiling point of oleic acid is lower than those of palmitic acid. Fatty acids emitted from Chinese cooking in Shenzhen (17) account for a larger fraction of quantified compounds than the same cooking style in Guangzhou and linoleic acid is the compound of the highest concentration. The reason may be that different cooking oils are used in the tested restaurants since the cooking styles are the same. For example, the different types of cooking oils have different fractions of saturated, unsaturated, and polyunsaturated fats, like peanut oil (18:49:33) and sunflower oil (11:20:69) (42). For meat fats, polyunsaturated fats constitute a smaller fraction than monounsaturated fats (42). Dongbei style in Guangzhou releases the highest concentration of monosaccharide anhydrides, but the lowest concentration of β-sitosterol, which can result from different types of vegetables used in cooking operations. Candidates of Tracers. The compounds detected in POM emitted from Chinese cooking in Guangzhou are also released by other primary sources. β-sitosterol, stigmasterol and levoglucosan detected in the emissions from burning of wood and straw (32, 33, 38, 41); Cholesterol is produced by burning of cigarette (43, 44), debris of plant and road dust (43). A VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
103
single compound is usually emitted by more than one type of source, but a group of compounds can reveal a unique pattern. In addition, these tracers are expected to reflect the emissions and transformation of POM during Chinese cooking operations. The reported possible tracers of meat cooking comprise of cholesterol, myristic acid, palmitic acid, stearic acid, oleic acid, nonanal, 2-decanone, and lactones (15). Some of them listed above have been used to calculate the contribution of meat cooking to atmospheric particles (9, 11, 45). In the case of palmitic acid, stearic acid, and oleic acid, they show unique ratios of palmitic acid/stearic acid and oleic acid/stearic acid, compared with other reported primary sources. The ratios of levoglucosan/(mannosan + galactosan) also display special ratios for Chinese cooking in Guangzhou(see the Monosaccharide Anhydrides section). Nonanal and lactones may be produced during cooking or emissions; therefore, they can reflect reactions which change the compositions of emissions of Chinese cooking (12, 15). Compared with the composition of atmospheric POM, n-alkanoic acids, monosaccharide anhydrides, and lactones with carbon number greater than 14 occur in high concentrations in atmospheric samples; however, the amounts of cholesterol and 2-decanone in most atmospheric samples are lower than the detection limit. Therefore, the candidates for organic tracers for Chinese cooking comprise tetradecanoic acid, hexadecanoic acid, octadecanoic acid, oleic acid, levoglucosan, mannosan, galactosan, nonanal, and lactones(5-decyldihydro-2(3H)furanone, 5-dodecyldihydro-2(3H)-furanone and 5-tetradecyldihydro-2(3H)-furanone). In addition, sterols can be the good tracers for emissions of Chinese cooking if they can be detected in atmospheric particles. Some of recommended tracers can be used to estimate the contribution of Chinese cooking to special species in atmosphere. For example, the average ratio of levoglucosan/ (mannosan + galactosan) and the average ratio of hexadecanoic acid/levoglucosan can be used to estimate the contribution of Chinese cooking to the concentration of hexadecanoic acid in atmosphere. These average ratios of levoglucosan/(mannosan + galactosan) are 12 for Chinese cooking and three for field burning of wood and straw, 9.9 for atmospheric PM2.5. The average ratios of hexadecanoic acid to levoglucosan are 84 for Chinese cooking, wood burning: 0.6∼3.7, average ) 1.7 for temperate climate conifers, 3.9∼7.4, average ) 5.8 for deciduous trees (32, 33), and 1.4 for atmospheric PM2.5. The contribution is calculated by two methods in this study through building equations: in the first method, the emissions from family fireplaces and fatty acids from fossil fuel are not introduced. However, the equation cannot be solved. In fact, the wood and straw are still used as fuel to cook food in some areas. Therefore, in the second method, the source of the wood or straw as fuel to cook food is incorporated and assumes that the pollutants from it are similar to those from family fireplaces, for which the average ratio of hexadecanoic acid to levoglucosan is about 0.02 (39, 40). By calculation, the highest contribution of Chinese cooking to the concentration of hexadecanoic acid in atmospheric PM2.5 is 1.3%.
Acknowledgments This work was supported the “863” project (2005AA649030), the National Basic Research Program (2002CB211605) from Ministry of Science & Technology, China and the Research Fund for the Doctoral Program of Higher Education (20040001055). We thank Dr. Lingyan He, Mr. Yan Zhang, and Miss Nan Sheng for their help with collecting samples, and Mr. Yuanxun Zhang for his advice during measurement of the samples. 104
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 1, 2007
Literature Cited (1) Durant, J. L.; Lafleur, A. L.; Plummer, E. F.; Taghizadeh, K.; Busby, W. F.; Thilly, W. G. Human lymphoblast mutagens in urban airborne particles. Environ. Sci. Technol. 1998, 32 (13), 18941906. (2) Bi, X. H.; Sheng, G. Y.; Peng, P. A.; Zhang, Z. Q.; Fu, J. M. Extractable organic matter in PM from LiWan district of 10 Guangzhou City, PR China. Sci. Total Environ. 2002, 300, 213228. (3) Bi, X. H.; Sheng, G. Y.; Peng P. A.; Chen, Y. J.; Fu, J. M. Size distribution of n-alkanes and polycyclic aromatic hydrocarbons (PAHs) in urban and rural atmospheres of Guangzhou, China. Atmos. Environ. 2005, 39, 477-487. (4) Cruz, C. N.; Pandis, S. N. A study of the ability of pure secondary organic aerosol to act as cloud condensation nuclei. Atmos. Environ. 1997, 31, 2205-2214. (5) Novakov, T.; Penner, J. E. Large contribution of organic aerosols to cloud -condensation nuclei concentrations. Nature 1993, 365, 323-365. (6) Facchici, M. C.; Mircea, M.; Fuzzi, S.; Charison, R. J. Cloud albedo enhancement by suface-active organic solutes in growing droplets. Nature 1999, 401 (16), 257-259. (7) Wexler, A. S., Ge, Z. Hydrophobic particles can activate at lower relative humidity than slightly hygroscopic ones: a Kohler theory incorporating surface fixed charge. J. Geophys. Res. 1998, 103, 6083-6088. (8) Hildemann, L. M.; Mazurek, M. A.; Cass, G. R. Chemical composition of emissions from urban sources of fine organic aerosol. Environ. Sci. Technol. 1991, 25, 744-759. (9) Schauer, J. J.; Rogge, W. F.; Hildemann, M. A; Cass, G. R. Source apportionment of airborne particulate matter using organic compounds as tracers. Atmos. Environ. 1996, 30 (22), 38373855. (10) Schauer, J. J.; Cass, G. R. Source apportionment of wintertime gas-phase and particle-phase air pollutants using organic compounds as tracers. Environ. Sci. Technol. 2000, 34, 18211832. (11) Zheng, M.; Cass, G. R.; Schauer, J. J.; Edgerton, E. Source apportionment of PM2.5 in the southeastern united states using solvent-extractable organic compounds as tracers. Environ. Sci. Technol. 2002, 36, 2361-2371. (12) Schauer, J. J.; Kleeman, M. J.; Cass, G. R.; Simoneit, B. R. T. Measurement of emissions from air pollution sources. 4. C1C27. organic compounds from cooking with seed oils. Environ. Sci. Technol. 2002, 36, 567-575. (13) Venkataraman, C.; Friedlander, S. K. Source resolution of fine particulate polycyclic aromatic hydrocarbons using a receptor model modified for reactivity. J. Air Waste Manage. Assoc. 1994, 44, 1103-1108. (14) McDonald, J. D.; Zielinska, B.; Fujita, E. M.; Sagebiel, J. C.; Chow, J. C.; Watson, J. G. Emissions from charbroiling and grilling of chicken and beef. J. Air Waste Manage. Assoc. 2003, 53, 185194. (15) Rogge, W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. Sources of fine organic aerosol. 1. charbroilers and meat cooking operations. Environ. Sci. Technol. 1991, 25, 1112-1125. (16) Schauer, J. J.; Kleeman, M. J.; Cass, G. R.; Simoneit, B. R. T. Measurement of emissions from air pollution sources. 1. C1 through C29 organic compounds from meat charbroiling. Environ. Sci. Technol. 1999, 33 (10), 1566-1577. (17) He, L. Y.; Hu, M.; Huang, X. F.; Yu, B. D.; Zhang, Y. H.; Liu, D. Q. Measurement of emissions of fine particulate organic matter from Chinese cooking. Atmos. Environ. 2004, 38, 6557-6564. (18) Labensky, S. R. International cuisine: Turkey travels the global. http://www.eatturkey.com/foodsrv/manual. (19) Chen, B. H.; Chen, Y. C. Formation of PAH in the smoke from heated model lipids and food lipids. J. Agric. Food Chem. 2001, 49, 5238-5243. (20) Chiang, T. A.; Wu, P. F.; Ko, Y. C. Identification of carcinogens in cooking oil fumes. Environ. Res. 1999, 81A, 18-22. (21) Chiang, T. A.; Wu, P. F.; Wang, L. F.; Lee, H.; Lee, CH. H.; Ying, C. K. Mutagenicity and polycyclic aromatic hydrocarbon content of fumes from heated cooking oils produced in Taiwan. Mutat. Res. 1997, 381, 157-161. (22) Li, S. G.; Pan, D. H.; Wang, G. X. Analysis of PAH in cooking oil fumes. Arch. Environ. Health 1994, 49 (2), 119-122. (23) Qu, Y. H.; Xu, G. X.; Chen, T. D. Genotoxicity of heated cooking oil vapors. Mutat. Res. 1992, 298, 105-111. (24) Wu, P. F.; Chiang, T. A.; Ko, Y. C.; Lee, H. Genotoxicity of fumes from heated cooking oils produced in Taiwan. Environ. Res. 1999, 80A, 122-126.
(25) Zhu, L. Z.; Wang, J. Sources and patterns of polycyclic aromatic hydrocarbons pollution in kitchen air, China. Chemosphere 2003, 50, 611-618. (26) Lee, S. C.; Li, W. M.; Chan, L. Y. Indoor air quality at restaurants with different styles of cooking in metropolitan Hong Kong. Sci. Total Environ. 2001, 279, 181-193. (27) Simoneit, B. R. T. Characterization of organic constituents in aerosols in relation to their origin and transport: a review. Int. J. Environ. Chem. 1986, 23, 207-237. (28) Rogge, W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. Sources of fine organic aerosol. 4. particulate abrasion products from leaf surfaces of urban plants. Environ. Sci. Technol. 1993, 27, 2700-2711. (29) Plant Chemistry; An, Y. L., Eds.; Northeast Forestry University Press: Shanghai, 1996; pp 311-312. (30) Chemical Basis of Cooking; Ji, H. K., Cun, G.; Y., Zhao, G. D., Eds.; Shanghai Science and Technology Press: Shanghai; 1993; pp 417-422, 441-445. (31) Mazurek, M. A.; Simoneit, B. R. T.; Cass, G. R.; Gray, H. A. Quantitative high-resolution gas chromatography and chromatograph/mass spectrometry analyses of carbonaceous fine aerosol particles. Intern. J. Anal. Chem. 1987, 29, 119-139. (32) Oros, D. R.; Simoneit, B. R. T. Identification and emission factors of molecular tracers in organic aerosols from biomass burning part 1. Temperate climate conifers. Appl. Geochem. 2001a, 16, 1513-1544. (33) Oros, D. R.; Simoneit, B. R. T. Identification and emission factors of molecular tracers in organic aerosols from biomass burning part 2. Deciduous trees. Appl. Geochem. 2001b, 16, 1545-1565. (34) Simoneit, B. R. T.; Schauer, J. J.; Nolte, C. G.; Oros, D. R.; Elias, V. O.; Fraser, M. P.; Rogge, W. F.; Cass, G. R. Levoglucosan: a tracer for cellulose in biomass burning and atmospheric particles. Atmos. Environ. 1999, 33, 173-182. (35) Simoneit, B. R. T, Elias, V. O. Organic tracers from biomass burning in atmospheric particulate matter over the ocean. Mar. Chem. 2000, 69, 301-312. (36) Oros, D. R.; Simoneit, B. R. T. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel 2000, 79, 515-536.
(37) Abas, M. R. B.; Oros, D. R.; Simoneit, B. R. T. Biomass burning as the main source of organic aerosol particulate matter in Malaysia during haze episodes. Chemosphere 2004, 55, 10891095. (38) Zhu, X. L. The study of chemical compositions and organic tracers of fine aerosol emissions from the burning of agriculture straws in typical areas. Post-doctoral report, Peking University, 2004. (39) Nolte, C. G.; Schauer, J. J.; Cass, G. R.; Simoneit, B. R. T. Highly polar organic compounds present in wood smoke and in the ambient atmosphere. Environ. Sci. Technol. 2001, 35, 19121919. (40) Fine, M. P.; Cass, G. R.; Simoneit, B. R. T. Chemical characterization of fine particle emissions from fireplace combustion of woods grown in the northeastern United States. Environ. Sci. Technol. 2001, 35, 2665-2675. (41) Rogge, W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. Sources of fine organic aerosol. 9. pine, oak, and synthetic log combustion in residential fireplaces. Environ. Sci. Technol. 1998, 32, 13-22. (42) Types of oils and their characteristics. http://en.wikipedia.org/ wiki/Cooking_oils. (43) Nolte, C. G.; Schauer, J. J.; Cass, G. R.; Simoneit, B. R. T. Trimethylsilyl derivatives of organic compounds in source samples and in atmospheric fine particulate matter. Environ. Sci. Technol. 2002, 36, 4273-4281. (44) Rogge, W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass, G. R.; Simoneit, B. R. T. Sources of fine organic aerosol: 6. cigarette smoke in the urban atmosphere. Environ. Sci. Technol. 1994, 28, 1375-1388. (45) Fraser, M. P.; Yue, Z. W.; Buzcu, B. Source apportionment of fine particulate matter in Houston, TX, using organic molecular markers. Atmos. Environ. 2003, 37, 2117-2123.
Received for review June 19, 2006. Revised manuscript received October 16, 2006. Accepted October 17, 2006. ES0614518
VOL. 41, NO. 1, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
105
