以雲端同步雙層析儀進行揮發性有機氣體污染之連續分析田野調查與數據統計方法研究

本研究以雲端同步應用兩台微型氣相層析儀（µGC）即時連續分析 16 天觀音樹林國小的揮發性有機化合物濃度，結合氣象資訊探討樹林國小受觀音工業區污染的狀況，並採用不同統計方法及污染源計算方法找出可能污染來源方向。所定量的化合物包含丙酮、乙酸甲酯、丁酮、正己烷、苯、甲苯和二甲苯，濃度都在個位數到百位數 ppb 範圍。丙酮、乙酸甲酯和丁酮在 µGC-1 的濃度比 µGC-2 高，表示 µGC-1 比較靠近丙酮、乙酸甲酯和丁酮的排放來源；正己烷、苯、甲苯、二甲苯的濃度是 µGC-2高於 µGC-1，表示 µGC-2 比較靠近正己烷、苯、甲苯、二甲苯的排放來源。眾多氣象資訊裡以風向為主要影響化合物濃度的氣象因素，兩台 µGC 的丁酮、正己烷、苯和甲苯主要源自於西南方，二甲苯主要源自於西北方，兩台 µGC 的丙酮和乙酸甲酯的主要來源都不同。本研究嘗試從不同的統計方法探討各化合物和各風向的污染狀況，方法包含相關性分析、集群分析以及主成份分析。在相關性分析裡，兩台 µGC 的丙酮和乙酸甲酯都呈低度相關；兩台 µGC 的丁酮、正己烷、苯和二甲苯呈中度相關；兩台 µGC 的二甲苯呈高度相關。至於集群分析，兩台 µGC 將化合物分為三群，分別為乙酸甲酯一群、二甲苯一群以及丙酮、丁酮、正己烷、苯和甲苯一群。兩台 µGC 的化合物在主成份分析的結果稍微不同，從化合物的成分矩陣得知，兩台 µGC 的乙酸甲酯有來自西南風和西北風的貢獻；丁酮、正己烷、苯和甲苯主要由西南風貢獻；二甲苯主要來自西北風的貢獻；µGC-1 的丙酮來自西南風的貢獻較高，µGC-2 的丙酮貢獻則跟西南風的貢獻較低。除了統計，本研究亦採用污染源機率分佈方法計算各化合物的高污染源機率方向，丙酮、乙酸甲酯、丁酮、正己烷、苯和甲苯在西南和東南方都有污染源機率，只是西南方的機率較高，而且每個化合物高機率的角度有所不同；二甲苯的污染最高機率來源是西北方。從以上結果知道，本研究方法得以在短時間內得到大量數據作統計分析以及計算污染源機率分佈得到不同化合物和風向的關係，這些結果可作為後續空污管理的參考。

關鍵字

微型氣相層析儀；即時連續分析；揮發性有機化合物；統計方法；定位污染源

並列摘要

In this study, two µGCs are used to analyze VOCs concentrations real-time and continuously at a 15 minutes interval in a school adjacent to an industrial area located at northern Taiwan for 16 days. Meteorological data are also collected to know its impact on VOCs concentration. The key VOCs were acetone, methyl acetate, butanone, hexane, benzene, toluene and m/p-xylene by canister/GC-MS confirmation, their concentrations measured by μGC ranged from units digit to hundreds ppb. The concentration of acetone, methyl acetate and butanone were higher in µGC-1 than in µGC-2, which means µGC-1 was near to the pollution source of these VOCs. On the other hand, the concentration of hexane, benzene, toluene and m/p-xylene were higher in µGC-2 than in µGC-1, which means µGC-2 was near to the pollution source of these VOCs. Among the various meteorological information that we had, we observed the wind direction is the main determinants of VOC concentrations which pointed to the possible source of pollution. The direction of high concentration sources of each VOC can be determined from the pollution rose plots. Butanone, hexane, benzene and toluene were mainly came from southwestern direction in both µGCs, m/p-xylene were mainly came from northwestern direction in both µGCs. The main pollution source of acetone and methyl acetate were different in both µGCs, which indicates the sources of these VOCs may be varied. These observations can be presented and analyzed by statistical method in detail: correlation analysis, cluster analysis and principle component analysis. In correlation anaylsis, the correlation of acetone and methyl acetate in both µGCs were low, the other VOCs were high correlated in both µGCs. In cluster anaylsis, three cluster of VOCs are grouped for both µGCs with distance of eight: methyl acetate in one cluster, m/p-xylene in one cluster, the rest of the compounds in one cluster. However, the result of component plot of VOCs for both µGCs are different in principle component analysis. By refering the component score of 16 main wind directions, we can observe that the methyl acetate was contributed by both southwestern and northwestern wind in both µGCs; m/p-xylene was contributed by northwestern wind in both µGCs; butanone, hexane, benzene, toluene were contributed by southwestern wind in both µGCs; acetone was more contributed by southwestern wind in µGC-1 than µGC-2. Besides, probability distribution maps of emission source were drawn for each VOC to know the highest probability of emission source of different VOCs. For acetone, methyl acetate, butanone, hexane, benzene and toluene, the probability were found in both southwest and southeast, but the probability of southwest is higher than southeast. Although they were all having high pobability in southwest, but they have different specific degree in southwest. For m/p-xylene, the highest probability were found in northwest. In summary, this study reveals plenty of data can be collected in a short period of time which can apply them with statistical method and calculate the probability distribution in order to study the relationship between VOC and wind direction, these results can be used as a reference for air pollution management.

並列關鍵字

micro gas chromatograph ； real-time analysis ； volatile organic compound ； statistical methods ； locating emission source

參考文獻

1. 歷年各季國內生產毛額依行業分. 2017: 行政院主計總處

Google Scholar

2. 105年工廠校正及營運調查最終統計概況分析. 2017: 經濟部統計處

Google Scholar

3. Volatile Organic Compounds in Consumer and Commercial Products. Available from: http://www.ec.gc.ca/cov-voc/.

Google Scholar

4. Jian, R.-S., Huang, Y.-S., Lai, S.-L., Sung, L.-Y., and Lu, C.-J., Compact instrumentation of a μ-GC for real time analysis of sub-ppb VOC mixtures. Microchemical Journal, 2013. 108: p. 161-167.

Google Scholar

5. Jian, R.-S., Sung, L.-Y., and Lu, C.-J., Measuring real-time concentration trends of individual VOC in an elementary school using a sub-ppb detection μGC and a single GC–MS analysis. Chemosphere, 2014. 99: p. 261-266.

Google Scholar

主題瀏覽