科學工業園區戶外與室內空氣污染物質監測模式探討：(1)某半導體工廠之空氣污染物質排放與蓄積特性之監測策略探討(2)某半導體工廠附屬幼稚園之室內水損環境與學童缺勤率之關係探討

Exploration of strategy for the outdoor and indoor air pollutants monitoring in a science industrial park: (1) Characteristics and impact of air pollutants emission and accumulation in a semiconductor plant (2) Relationship of the indoor water-damaged environment with student absences of an affiliated kindergarten of a semiconductor plant

曹永杰(Yung-Chieh Tsao)

指導教授 ： 黃耀輝

國立臺灣大學/公共衛生學院/職業醫學與工業衛生研究所/博士(2013年)

https://doi.org/10.6342/NTU.2013.02702

摘要

科學工業園區為半導體及光電產業工廠群聚的高科技產業園區，製造生產所排放的空氣污染物質導致園區形成特殊而複雜的空氣品質環境。本論文以科學工業園區之廠區為研究對象，藉由量測大氣環境與室內環境之空氣污染物質組成、發生與汙染特性，探討廠區空氣污染物監測模式之規劃與成效評估，研究分成二個主題，第一個主題利用傅立葉紅外光譜儀(Fourier transform infrared, FTIR)調查戶外與室內環境空氣污染物發生及蓄積現象，探討連續性監測設備在異味、發生源及污染物質流布調查策略的應用規劃、設計與影響。第二個主題以科學園區常見的密閉式中央空調大樓環境為研究對象，評估廠區幼稚園水損因子對於學童出缺勤率的影響，探討一般室內環境危害物質監測策略。
第一主題包含兩部分研究內容，第一部份利用三部傅立葉紅外光譜儀，同步調查某一半導體廠區空氣污染物質及其排放來源，連續三天共蒐集736筆資料，發現環境異味率為43.9%，包括氨氣、乙酸丁酯、臭氧與丙二醇單甲基醚酯等化學物質濃度超過異味閥值。以邏輯式回歸模式探討傅立葉紅外光譜儀量測效能，氣象條件在時序上持續變化的狀況下，南風時的量測效能較其他風向時高(勝算比=3.8, 95%信賴區間 2.9~5.0)；此外，測線配置與污染物質排放擴散區域之距離也會影響量測效能，最靠近發生源下風處約30公尺的測線二之量測效能即高於其他測線(勝算比=5.1, 95%信賴區間 3.6~7.2)，說明氣象條件與測線配置對量測效能有相當大幅的影響。另外一個干擾效能因子為發生源特性，高科技產業廠區污染物發生源為多重管道排放形式，不同類型污染物質在不同排放煙囪上升氣流速度差異，導致單一測線配置無法同步量測不同擴散沉降距離的污染物質，讓異味調查更複雜困難。簡而言之，傳統上單一測線傅立葉紅外光譜儀佈線策略已無法提供完整的量測數據，多重測線的傅立葉紅外光譜儀規劃應用於高科技工業園區或煙囪密集聚集園區的異味調查確實有其必要性，平面二維空間甚至三維立體空間的採樣規劃都應被考慮。
第一主題的第二項研究為同步利用三部傅立葉紅外光譜儀，探討戶外與室內空氣污染物質逸散與濃度變化，在連續四天的1,032筆資料分析中，發現戶外與室內濃度具有相關性的氣體包含四氯化碳、一氧化二氮、一氧化碳與甲烷等四種。約有62.2%~73.1%的量測值，其室內與戶外濃度比值大於1.0，室內與戶外濃度分布在時序上呈現動態變化關係。分析戶外與室內濃度之迴歸模式，一氧化二氮在低濃度時，室內sub Fab濃度最高、外氣入口之MAU次之，以及室內Fab環境最低，由三者濃度間關係可以說明其主要發生源為室內環境sub Fab，並且蓄積於戶外大氣環境的現象；高濃度時，室內Fab或sub Fab濃度均高於戶外MAU，濃度關係改變起因於化學污染物係由室內產生，且80%無塵室循環供氣來自於室內環境再循環，無法有效排出室內空氣污染物，導致室內環境濃度升高。在固定污染源密集的科學園區，利用同步傅立葉紅外光譜儀連續量測戶外與室內濃度關係，可有效監測污染物質排放與擴散行為，提供對於空氣品質污染物高度敏感的高科技產業一項可靠的室內與戶外污染物質量測與監控的方法。
第二主題的研究為追蹤某半導體廠區內附設幼稚園受水損教室26位學童與27位來自非水損教室學童，觀察連續42週期間學童之出缺席狀況與原因(1,953人週)。水損教室學童的週缺勤率為2.99%，高於非水損教室的1.28% (p<0.001)。利用GEE(generalized estimating equation model)模式分析發現，年齡及性別在學童週缺勤率統計上並無差異，但是六歲學童週缺勤率略高於五歲學童(勝算比=2.02; 95%信賴區間0.99-4.11, P=0.054)；但水損教室學童的週缺勤率確實高於非水損教室學童，勝算比為2.45 (95%信賴區間1.15-5.24, P<0.05)。比較室內空氣指標結果顯示，由於研究教室有良好的空調系統運作，室內受水損與非水損教室濕度未出現明顯差異，但是水損教室發現黴菌濃度(993 CFU/m3)則明顯高於非水損環境(404 CFU/m3) 但全空調大樓的室內水損情形不易以濕度量測來辨認。本研究結果說明定期黴菌量測可以作為評估室內環境水損狀況的先期指標。特別當溼度量測無法在中央空調的辦公大樓提供有效指引時，黴菌量測數值則更顯重要性及有效性。

關鍵字

科學園區 ； 半導體 ； 傅立葉紅外光譜儀 ； 空氣品質 ； 異味 ； 濃度比率 ； 水損 ； 缺勤率

並列摘要

Science industrial park in Taiwan accommodates high-tech industries, including semiconductor and opticelectronic manufacturing industries. Emitted air pollutants from these industries resulted in special and complicated scenario of ambient air quality. This study was conducted in a semiconductor manufacturing plant to explore the monitoring strategy and to evaluate its efficacy by examining chemical composition, emissions, and charactistics of the emitted air pollutants around indoor and ourdoor environments. There were two parts in this study: (1) to explore the strategy of continuous monitoring and its impact in the investigation of odor episode, emission sources and distribution of air pollutants by examining the occurrence and accumulation of air pollutants in a semiconductor plant using Fourier transform infrared (FTIR) spectrometers, (2) to explore a monitoring strategy for affecting factors of a general indoor environment by examining the relationship of indoor water-damaged environment with student absences rate of an affiliated kindergarten of a seminconductor plant, located in an envelop building equipped with mechanical ventilation.
The first part included two studies. Study I employed three Fourier transform infrared (FTIR) spectrometers to synchronously characterize air pollutants and emission sources in a semiconductor plant. Based on the 736 consecutive five-minute monitoring records in three consecutive days, some detectable chemicals, inclduding ammonia, ozone, butyl acetate, and propylene glycol monomethyl ether acetate (PGMEA), presented with concentrations exceeding their respective odor thresholds and resulted in an combined total odor rate of 43.9%. Results of logistic regression analysis indicated that the FTIR’s monitoring showed higher efficacy as south wind prevailing with odds ratio of 3.8 (95% confidence interval (CI): 2.9-5.0) as compared to other wind directions. Additionally, distance between the monitoring path and the emission source is another factor influencing the efficacy of FTIR performance. The closest monitoring path, 30 m from the emission source, in the down-wind direction of emission source showed the highest efficacy in detecting odorous samples with OR of 5.1 (95% CI: 3.6-7.2). These findings illustrated the importance of meteorological factors and location of monitoring path on the FTIR monitoring performance. Such monitoring performance might also be counfounded by the properties of the emission sources. For instance, among the high-tech industries, pollutants were usually ejected from multi-stacks in a plant along with different ejection flow velocities, producing different landing distance for the plume of emitted air pollutants and therefore biasing the performance of FTIR monitoring with single path. In conclusion, traditional single FTIR monitoring path is not capable in monitoring common multiple emission sources. Synchronous FTIR monitorings with multiple paths, dispatched either in two dimensions or even three dimensions, are recommended for investigation on relatively complicated episodes in high-tech industry parks or any other industrial zone with densely erected stacks to enhance the efficacy of FTIR monitoring.
Study II in the first part of the present study was conducted to explore the emission of air pollutants and their concentration fluctuations around the indoor and outdoor environmens of a semiconductor manaufactrure plant using four-consective-day monitoring data synchronously collected by three FTIR spectrometers. Based on a total of 1,032 continuous five-minute interval records, the percentages for indoor/outdoor concentration ratio higher than 1.0 ranged from 62.2% to 73.1% for carbon tetrafluoride, nitrous oxide, carbon monoxide, and methane, indicating indoor concentrations usually hihger than outdoor ones for these chemicsls. And, it was observed that the relationship between indoor and outdoor chemical concentrations dynamically changed from the temporary point of view. Regression analysis showed that, at relatively low level, the highest nitrous oxide concentration was present in the sub Fab, followed by the outdoor MAU, and then by the Fab, implying the original emission sources was in the sub Fab of the indoor environment and these chemicals might be able to accumulate in the outdoor environment. In contrast, results of regression analysis indicated that, at relatively high level, the indoor chemical concentrations measured in Fab or sub Fab were generally greater than the outdoor chemical concentration measured at the MAU. This situation could be attributed to the generation of chemical pollutants in indoor environment and 80% returned air from the sub Fab was used in the ventilation system, resulting in low ventilation efficacy for the indoor air pollutants and increased airborne chemical concentrations accordingly. It was demonstrated that, in the science industrial park with densely distributed fixed pollutant sources, synchronous multiple FTIR monitoring paths for continuous monitoring on the relationship between indoor and outdoor chemical cocentrations were effective measures in the monitoring of pollutant emission and dispersion. This method provided the high-tech manufacturing facilities, which are extremely sensitive to airborne molecular contaminants, a reliable measurement and monitoring method for indoor and outdoor pollutants.
The second part of this study, i.e., Study III, was conducted to explore the impact of a water-damaged indoor environment on children’s weekly absences resulting from upper respiratory infection in a kindergarten. Twenty-six and 27 children were recruited from water-damaged and non-water-damaged classrooms, respectively, in the same building, and were followed up during the study period of 42 weeks. Weekly absence rates were computed from daily absence records. The weekly absence rate was significantly higher for children in the water-damaged classroom, 2.99%, than that for those in the non-water-damaged classroom, 1.28%. Results of the generalized estimating equation model (GEE) indicated that the weekly absence rate of K-6 was slightly higher than K-5 (OR=2.02 / 95% CI: 0.99-4.11, P=0.054). After adjusting for gender and age, the odds ratio for absence was 2.45 (95% CI: 1.15－5.24, p<0.05) for the children in the water-damaged classroom. Additionally, fungal concentration was significantly higher in the water-damaged classroom (993 CFU/m3) than in the non-water damaged classroom (404 CFU/m3). The water-damaged issue in the air-conditioned building is hard to be identified by measuring humidity. It is tentatively concluded that the fungal concentration is a better indicator than humidity for early-stage water damage in the indoor environment. Fungal concentration is of greater importantance and effectiveness as humidity is not an appropriate indicator for water-damage in a well air-conditioned indoor envirionment.

並列關鍵字

science industrial park ； semiconductor ； Fourier transform infrared (FTIR) ； air quality ； odor ； indoor/outdoor concentration ratio ； water-damaged ； absence rate

參考文獻

Barnett HL, and Hunter BB. 1987. Illustrated genera of imperfect fungi: New York, Macmillan.

Blondeau P, Iordache V, Poupard O, Genin D, and Allard F. 2005. Realtionship between outdoor and indoor air quality in eight French schools. Indoor Air. 15(1):2-12.

Bornehag C, Blomquist G, Gyntelberg F, Jarvholm B, Malmberg P, Nordvall L, Nielsen A, Pershagen G, and Sundell S. 2001. Dampness in buildings and health nordic interdisciplinary review of the scientific evidence on associations between exposure to "dampness" in buildings and health effects (NORDDAMP). Indoor Air 11(2):72-86.

Bornehag CG, Sundell J, Bonini S, Custovic A, Malmberg P, Skerfving S, Sigsgaard T, and Verhoeff A. 2004. Dampness in buildings as a risk factor for health effects, EUROEXPO: a multidisciplinary review of the literature (1998–2000) on dampness and mite exposure in buildings and health effects. Indoor Air 14(4):243-257.

Bornehag CG, Sundell J, Hagerhed-Engman L, Sigsggard T, Janson S, and Aberg N. 2005. ‘Dampness’ at home and its association with airway, nose, and skin symptoms among 10,851 preschool children in Sweden: a cross-sectional study. Indoor Air 15:48-55.