This project is to conduct PTR-MS measurements of volatile organic compounds (VOCs) within the Formosa Naphtha No.6 industrial park, which is different from the previous relevant NIEA projects that were conducted around and outside the park. Three sites (P1, P2 and P3) were chosen including: 1. south of the employee dormitory (6/26 – 7/13, P1); 2. Formosa Sumco Technology (7/20 – 8/3, P2); 3. Oil refinery-1 (8/10 – 8/31, P3), where real-time measurements were lasted for at least two weeks at each site.
PTR-TOF has a major advantage of greater mass resolution than does QMS. It can perform rapid scans of air matrix and detect air-borne compounds released from potential sources. During the three monitoring periods, of all the compounds detected by PTR-TOF, their mixing ratios were in the range of several ppb (v/v) to several hundred ppb. Based on the observed concentration ranges, no major leaks existed considering the scale and complexity of the park.
The measurement results suggest that 11 VOCs were detected at P1, 15 compounds at P2 and 17 compounds at P3. These found VOCs mainly comprised of oxygenated VOCs (alcohol, aldehydes, ketones, organic acids, esters), non-methane hydrocarbons (alkene, aromatics), nitrogen-containing VOCs and sulfur-containing VOCs.
When summarizing the semi-quantitative results obtained by PTR-TOF at the three sites, major constituting VOCs in ambient air are categorized as follows: Alkenes including ethene (460.1 ppb), butenes (64.7 ppb), pentenes (11.5 ppb), butadienes (59.5 ppb). Aromatics included benzene (92.2 ppb), C2-benzene (28.7 ppb), styrene (26.7 ppb), and C3-benzene (64.5 ppb). The N-containing VOCs included N,N-dimethylformamide (154.6 ppb), morpholine (38.7 ppb). O-containing VOCs included acetaldehyde (44.6 ppb) and acetone/propanal (54.5 ppb). Furthermore, there were two compounds of trace levels, namely formic acid (7.2 ppb) and methyl methacrylate (0.9 ppb), which were only observed at P3. These aforementioned compounds will form a useful list of reference to be adopted in future forensic investigation in case of industrial hazards.
We found that most observed VOCs were sensitive to the change of wind direction, which is reasonable as the sources of high variety were in proximity to the monitoring site. A common feature would be pronounced concentration spikes (such as for propylene) during the course of real-time monitoring.
The measurement data of PTR-TOF were able to be compared with those of QMS at P1 to assure the data quality of TOFMS. Consistent results were noted between the two instruments, which suggests the reliability of TOFMS measurements for most VOCs. An automated gas chromatograph (GC) was employed to assist the validation of measurements of TOFMS at P2 and P3 when QMS malfunctioned and was not available for mutual validation. As a result, a rugged and simple GC can be useful to compliment the more fragile and delicate instruments such as PTR-MS in field measurements.
