風險相似暴露族群鑑定及定量推估模型應用-以半導體廠為例

Identification of Risk Similar Exposure Groups and Application of Quantitative Modeling - A Semiconductor manufacturing industry

吳嘉駿(Ka-Chon Ng)

指導教授 ： 吳章甫

共同指導教授 ： 曹永杰

國立臺灣大學/公共衛生學院/公共衛生碩士學位學程/碩士(2020年)

https://doi.org/10.6342/NTU202000456

摘要

半導體在製程及研發上使用多種類之新興化學品，特別是混合物，引起對作業暴露危害的關注及增加管理的複雜性。在實務進行大規模作業環境監測有其困難處，且具標準採樣分析及容許濃度之化學品有限，人員潛在危害可能被忽略。本實習研究基於AIHA暴露評估架構，研究目的是1. 整理事業單位之數據資料庫，以描述性統計分析2015-2018/9化學品使用量及含IARC致癌物質化學品數量，鑑定使用含IARC致癌物質化學品之暴露族群（定義為風險相似暴露族群）；2. 應用定量暴露模型—二暴露區模型，推估設備維修保養作業人員丙酮之暴露濃度，與實測值及法規容許濃度比較，評估其實務普及應用之可能性。事業單位共使用21797件化學品並篩選出1135件含IARC物質之化學品，並鑑定出120個有使用之相似暴露族群；使用軟體AIHA IH MOD 2.0進行二暴露區模型分析，人員呼吸帶之暴露濃度peak值為31.7 ppm，與現場實際測量濃度9.75 ppm高3.25倍，落在合理的預測範圍；另以風險管理角度調整參數，呼吸帶之暴露濃度peak值可能在15.83 – 55.99 ppm之間，而時量平均值在12.71-41.17 ppm之間。本研究以定性方式鑑定風險相似暴露族群，能深入了解化學品之致癌物質、種類及所使用的族群，實務管理上提供優先關注及健康追蹤，而二暴露區模型提供合理而保守的推估值，並在多重條件估算下peak值及時量平均都遠低於我國法規值200 ppm，加上其操作簡單便利、具科學性及低成本，因此本研究之結果支持兩者在化學品實務管理的應用。

關鍵字

相似暴露族群 ； 暴露評估 ； 定量推估模型 ； 二暴露區模型半導體製程

並列摘要

Using a variety of chemicals, especially emerging agents and mixtures, in semiconductor industries has raised concerns for potential exposure hazards and management complexity. There are difficulties in carrying out large-scale exposure monitoring of chemical agents. Meanwhile, chemicals with standard sampling analysis and permissible exposure limit are limited, as such the potential hazards of workers may be underestimated. This internship study is based on the AIHA exposure assessment framework. The purpose of the study is to: 1. Compile the databases and descriptively analyze the use of chemicals between 2015-2018/9, focusing on those containing IARC carcinogens and identifying the number of risk SEGs (defined as similar risk exposure groups using chemicals containing IARC carcinogens); 2. Apply the two-zone model to estimate the exposure concentration of acetone for preventive and maintenance workers and compare with the measured values ​​and permissible exposure limit. The company used a total of 21,797 chemicals, included 1,135 chemicals containing IARC chemicals, and 120 risk SEGs; AIHA IH MOD 2.0 software was used to perform quantitative modeling analysis. The estimated peak value is 31.7 ppm, which is 3.25 times higher than the measured concentration of 9.75 ppm, and falls within a reasonable prediction. In addition, several parameters are adjusted in the perspective of risk management. The peak value of the revised estimates ranges between 15.83-55.99 ppm, and time-weighted average (TWA) ranged between 12.71-41.17 ppm. Qualitative identification of risk SEGs can provide insights into the carcinogens, types, and exposure groups, providing prioritization for exposure and health managements. The two-zone model provides reasonable and conservative estimates. Under multiple scenarios, the concentration peak and TWA of acetone are below the permissible exposure limit of 200 ppm. This model is user-friendly, scientifically proven and low-cost. The results of this study support the application of these two approaches for chemical management in practice.

並列關鍵字

similar exposure groups ； exposure assessment ； quantitative modeling ； two- zone exposure model semiconductor industry

參考文獻

1 國家發展委員會: 臺灣經濟發展歷程與策略(2018).

2 行政院主計總處: 行業標準分類（第10次修訂）(2016).

3 Bender, T.J., C. Beall, H. Cheng, R.F. Herrick, A.R. Kahn, R. Matthews et al.: Cancer incidence among semiconductor and electronic storage device workers. Occup Environ Med 64(1): 30-36 (2007).

4 Kim, M.H., H. Kim, and D. Paek: The health impacts of semiconductor production: an epidemiologic review. Int J Occup Environ Health 20(2): 95-114 (2014).

5 謝功毅: 半導體員工生殖健康及癌症風險之探討(2005).