Polycyclic aromatic hydrocarbons (PAHs) in total suspended particulates (TSP) collected at six rural and urban stations in Hong Kong from 1993–1995 using high volume air samplers were identified using GC-MS (gas chromatography-mass spectrometry). The results showed that the PAHs exhibited distinct spatial and seasonal variability. The total PAH content (PAH) in the samples ranged from 0.41 to 48 ng m-3. The unsubstituted analogs are the characteristic products of high temperature combustion processes. The highest average PAH was measured at the street-level station in Mong Kok indicating that vehicles were high PAHs contributors. The rural station at Hok Tsui exhibited the lowest PAH level, however; influences of city plumes could be seen when northerly or northeasterly winds prevailed in the winter. All stations experienced the highest loading of PAHs in autumn and the lowest in summer; the former was 2.8 times greater than the latter. This seasonal variability is due to the Asian monsoon system, precipitation, micrometeorology, and the rate of photodegradation. In summer, Hong Kong experiences relatively clean oceanic air and high rates of precipitation and photodegradation, while upon the onset of the winter monsoon, local air mass is replaced by polluted air streams from the Asian continent. Benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(e)pyrene, indeno(1,2,3-cd)pyrene and benzo(ghi)perylene were the dominant species in the samples. The PAH distribution patterns at different stations were similar within each season. However, seasonal variations existed. For example, phenanthrene contributed up to 14% of the total PAH in summer, while the dominance of benzo(b)fluoranthene, benzo(k)fluoranthene was more significant in autumn and winter.
Correlations between 35 different organic and inorganic species or tracers in the aerosols of Hong Kong were derived based on the GC−MS organic results from our 1993−1994 study and Hong Kong Environmental Protection Department inorganic data. Source apportionment is essential in fingerprinting of air pollutants for cross-boundary studies. It is also important in the development of emissions inventory data, control strategies, and legislature. Traditionally, emission sources are identified from either the inorganic or organic tracers in the aerosols, but not both. Each of these techniques yields much source information; however, each by itself provides only a partial picture due to the complicated nature of ambient aerosols. Since many of the organic and inorganic species in aerosols are from the same source, correlations between the two are to be expected. In this study, such correlations were indeed found. For example, benzo[b,k]fluoranthene, the most abundant PAH (polycyclic aromatic hydrocarbon) in all the PAHs detected in Hong Kong, exhibited good correlation with Pb, Zn, As, nss SO42- (nonseasalt sulfate), total PAHs, and benzo[a]pyrene. This paper further demonstrates that a combination of the two techniques could be an improved method for source apportionment than using either method alone.
The results from a year-long monthly study of the solvent extractable organic compounds (SEOC) in PM2.5 of the ambient aerosols in Hong Kong are reported. A total of 18 samples were analyzed. The extracted organic compounds were separated into four major fractions (n-alkanes, fatty acids, alkanols and PAHs (polycyclic aromatic hydrocarbons)) and identified with GC-MS (gas chromatography–mass spectrometry). The percentage of each fraction in total yield is as follows: fatty acids at 46–80%, alkanes at 10–34%, alkanols at 4–21%, and PAHs at 1–6%. Compared to the TSP (total suspended particulates) samples from our previous studies, the PAH fraction was higher in PM2.5 than TSP (0.5–2.5%). The bias of PAHs in PM2.5 suggests potential implications in health impact because PM2.5 is respirable. The total yield (defined as the sum of the four fractions) in PM2.5 was in the range of 56.4–233.6ngm−3. The sources of SEOC in PM2.5 could be attributed to vehicular, biogenic and microbial origins. On the average, 79% of the alkanes in PM2.5 came from vehicular emissions. U:R (ratio of unresolved to resolved components), an index to assess the magnitude of petroleum residues in aerosols, exhibited higher values in PM2.5 (average 3.2, range 1.0–5.9) than TSP (average 2.4, range 1.1–3.3). CPI (carbon preference index) of alkanes showed an inverse relationship with U:R, and a positive correlation with the percentage of alkanes from higher plant wax. Although U:R and CPI have been widely used as indices in source apportionment, their effectiveness was demonstrated for the first time from the statistical point of view in this paper. It was discovered that the CPI=1 (a characteristic of petroleum hydrocarbons) envelope and Cmax shifted towards lower carbon numbers in winter, suggesting there was more contribution from vehicular emission. PAHs had the following range 0.7–12.2ngm−3. The positive correlation with benzo(ghi)perylene suggested that they were of vehicular origin. Distinct seasonal variation in PAH concentration was found with higher concentrations in the winter samples. The concentration range of fatty acids in PM2.5 (31.1–168.8ngm−3) indicated that the microbial contribution in the PM2.5 samples was systematically lower in late fall and winter (average 52%) while larger in the hotter weather (over 80%), suggesting higher microbial activities in the summer. A ratio as high as 1.93 in summer for C18:1/C18:0 and only 0.69 in winter suggested that there were more aged aerosols in winter while more fresh emissions were found in summer. The results of SEOC in the PM2.5 aerosols indicated that vehicular emission was the characteristic of Hong Kong's aerosols.