石化廠作業勞工苯暴露測定與風險推估

Measurements and Risk Estimation of Benzene Exposure for Workers in Petrochemical Factories

賴貫華(Guan-Hua Lai)

指導教授 ： 吳俊德

長榮大學/健康科學學院/職業安全與衛生學系(所)/碩士(2017年)

摘要

為了瞭解石化產業作業勞工苯暴露實態及暴露風險，本研究針對石化廠區工廠內作業勞工進行苯暴露的空氣濃度及尿液生物偵測指標物測定。研究以美國工業衛生協會(American Industrial Hygiene Association, AIHA)建議的全面性暴露評估策略對每一個廠區作業勞工劃分相似暴露群組(similar exposure groups, SEGs)，在每一個SEG以隨機抽樣方式選定勞工進行暴露採樣測定。勞工個人空氣中苯暴露濃度測定，以填充Carbopack X吸附劑的熱脫附管作為採樣介質，搭配個人空氣採樣幫浦進行採集。採集樣本以熱脫附儀搭配氣相層析儀(gas chromatography, GC)及火焰離子偵測器(flame ionization detector, FID)進行樣本分析；生物偵測指標物部分，採取勞工上工前、中午休息、下工前的尿液樣本，然後以頂空(headspace, HS)萃取進樣器及氣相層析質譜儀(gas chromatography-mass spectrometry, GC/MS)，量測尿液中苯本體、苯的代謝物質t,t-muconic acid(t,t-MA)及S-phenylmercapturic acid(SPMA)作為生物偵測指標物測定。於樣本收集同時，給予受測勞工一份工作紀錄表，紀錄個人基本資料及一天工作時間內每十五分鐘所執行的工作項目，了解可能造成勞工苯暴露的主要工作項目及影響因子。本研究完成對21個石化廠區(以代號A01~A21表示)，1485個勞工個人空氣苯暴露測定，464個尿液樣本生物偵測指標物測定。測定結果顯示：所有廠區的作業勞工空氣中苯暴露測定濃度平均值(±標準差)為4.59 (±35.56) ppb，各廠區勞工的空氣苯暴露實態濃度算術平均值範圍介於0.23~23.78 ppb，算術標準偏差範圍介於0.28~118.53 ppb，顯示大部分廠區勞工空氣苯暴露存在相當顯著的變異；少數廠區有高於100 ppb的最大測定濃度。整體而言，所有廠區勞工苯暴露測定濃度皆遠低於我國法定容許暴露濃度1 ppm。有關尿液生物偵測指標物t,t-MA和SPMA測定結果，顯示在勞工低濃度空氣苯暴露下，苯暴露濃度僅與下午尿液SPMA的測值存在低關聯性(r=0.27)，針對t,t-MA/SPMA的比值與空氣苯暴露測定濃度及抽菸習慣的迴歸分析，發現抽菸習慣較低濃度空氣苯暴露對此比值R的大小有更大影響，也確認抽菸習慣確實會造成勞工尿液t,t-MA代謝量的增加，但較不影響SPMA的代謝量。本研究使用貝氏統計分析針對21個廠區106個SEGs勞工苯暴露的風險等級機率進行苯暴露風險推估，發現有23個SEGs勞工苯暴露在等級4有大於5%的機率，這其中又有9個SEGs勞工苯暴露在等級4有大於20%的機率。雖然廠區勞工苯暴露仍符合法令規定，為了保護苯暴露易感受性勞工及降低勞工苯暴露風險，應將高風險等級的SEGs列為重點監測群組，需要對這些群組勞工的苯暴露進行更多的評估。

關鍵字

生物偵測 ； 個人採樣 ； 苯暴露 ； 空氣採樣 ； 石化廠

並列摘要

In order to understand the exposure profile and risk of benzene exposure for the workers in petrochemical manufacturing industries, this study aimed at the measurements of air exposure concentrations and biological markers of benzene for the workers in the petrochemical factories in a petrochemical complex area. The exposure sampling in this study adopted the comprehensive exposure assessment strategy suggested by American Industrial Hygiene Association (AIHA). Similar exposure groups (SEGs) were established for the workers in each factory. The workers in each SEG were randomly selected for benzene exposure measurements. Thermal desorption tubes packed with the adsorbent of Carbopack X were used as the sampling medium for benzene erxposure sampling by connncecting to a personal sampling pump. The measured workers were asked to provide three urine samples at the pre-shift, lunch break and post-shift times on the exposure sampling day. The collected air samples were analyzed by an automatic thermal desorption system equipped with gas chromatography (GC) system and a flame ionization detector (FID). The collected urine samples were quantified by the parent compound of benzene and its metabolites of t,t-muconic acid (t,t-MA) and S-phenylmercapturic acid (SPMA) by a headspace gas chromatography-mass spectrometry (GC/MS) system. Each worker received a data sheet for recording the work tasks he participated in each 15 minutes interval of a regular 8-hour workday during the sample collection day. The information recorded by the data sheet was coded into a data file for the analysis of exposure determinants. Twenty-one petrochemical factories (designated by A01~A21) were recruited into the study. A total of 1485 measurements of benzene exposure air samples and 464 urine samples were collected. The results of the analyses of the exposure measurement data indicated the arithmetic mean (±standard deviation) (AM±SD) of the benzene exposure concentrations of these all air samples was 4.59 (±35.56) ppb. The AMs of the exposure profiles of benzene for the SEGs in all the factories were in the range of 0.23~23.78 ppb, and the SDs were in the range of 0.28~118.53 ppb. This showed that significant variability of benzene exposure among the workers in the factories and seveal factories had the maximum measured concentrations greater than 100 ppb. In general, the benzene exposure of all the workers in the factories was less than the occupational permissible exposure limit of 1 ppm. The result of the measured concntrations of t,t-MA and SPMA in the urine samples showed there was only a low correlation (r=0.27) between the air exposure concentrations and the SPMA concentrations of the post-shift urine samples under the situation of low air benzene exposure levels. A linear regression model was built to evaluate the influence of air benzene exposure concentrations and cigarette smoking on the ratio of t,t-MA to SPMA. The result indicated the cigarette smoking had a greater influence than the air benzene exposure concentrations on the magnitude of the ratio of t,t-MA to SPMA. The cigarette smoking did result in the increase of the qunatity of t,t-MA but did not make any significant effect on the quantity of SPMA in the urine samples of the workers. Bayesian statistical analysis was used to estimate the benzene exposure risk rating for the 106 SEGs in these 21 factories. The results found 23 out of 106 SEGs had probabilities of the benzene exposure risk rating category 4 greater than 5%, and 9 out of these 23 SEGs had probabilities of the benzene exposure risk rating category 4 greater than 20%. Although the exposure measurements of the workers for the factories were in compliance with the permissible exposure limit of benzene, the focus of future exposure monitoring programs should put on the SEGs with high benzene exposure risk rating categories to protect the susceptible workers from benzene exposure and reduce the risk of benzene exposure for the workers. More exposure assessments for the workers in the high risk categories are warranted.

並列關鍵字

biological detection ； personal sampling ； benzene exposure ； air sampling ； petrochemical plant

參考文獻

楊秀宜和李聯雄 (2011) 生物偵測暴露指標建議值訂定機制研究(Strategies for Setting Biological Exposure Indices), 勞工安全衛生研究報告, 行政院勞動部勞動及職業安全衛生研究所, http://www.ilosh.gov.tw/wSite/ct?xItem=3131&ctNode=278&mp=11, Last accessed, June 25, 2016.

American Conference of Governmental Industrial Hygienists (ACGIH) (2016) Threshold Limit Values for Chemical Substances and Physical Agents and Biological Exposure Indices. Cincinnati (OH), ACGIH, ISBN: 978-1-607260-84-4.

Boogaard PJ, van Sittert NJ. (1995) Biological monitoring of exposure to benzene: a comparison between S-phenylmercapturic acid, trans,trans-muconic acid, and phenol, Occup Environ Med. 52(9):611-620.

Carrieri M, Tranfo G, Pigini D, Paci E, Salamon F, Scapellato ML, Fracasso ME, Manno M, Bartolucci GB. (2010) Correlation between environmental and biological monitoring of exposure to benzene in petrochemical industry operators, Toxicol Lett. 15;192(1):17-21.

Carrieri M, Bartolucci GB, Scapellato ML, Spatari G, Sapienza D, Soleo L, Lovreglio P, Tranfo G, Manno M, Trevisan A. (2012) Influence of glutathione S-transferases polymorphisms on biological monitoring of exposure to low doses of benzene, Toxicol Lett. 13;213(1):63-8.