PM2.5 is a complex mixture from various sources, undergoing numerous atmospheric processes and posing significant health risks. To reduce PM2.5 concentrations, understanding its sources is crucial. This study uses the Positive Matrix Factorization (PM...
PM2.5 is a complex mixture from various sources, undergoing numerous atmospheric processes and posing significant health risks. To reduce PM2.5 concentrations, understanding its sources is crucial. This study uses the Positive Matrix Factorization (PMF) model to identify PM2.5 sources and the Conditional Bivariate Probability Function (CBPF) to estimate local sources. The oxidative potential of PM2.5 was measured using the dithiothreitol (DTT) assay. Trace elements, key markers for anthropogenic sources, were analyzed using X-ray fluorescence (XRF) and inductively coupled plasma mass spectrometry (ICP-MS) with same samples. Differences in these methods can affect not only with concentration but also with source identification and the correlation between oxidative potential and PMF results. This study aimed to compare the results from XRF and ICP-MS in terms of concentration levels, PMF results, and their correlation with oxidative potential. From November 2021 to December 2022, 107 PM2.5 samples were collected in Seoul, and analyzed for carbonaceous species, ionic species, and trace element species. The average PM2.5 concentration during this period was 20.5 μg/m³, exceeding the annual ambient air quality standard of 15 μg/m³. Trace element concentrations averaged 0.756 μg/m³ with XRF and 0.825 μg/m³ with ICP-MS, showing a slightly higher concentration with ICP-MS. PMF analysis using both XRF and ICP-MS identified ten major sources of PM2.5: secondary nitrate, secondary sulfate, mobile sources, biomass burning, incinerator, oil combustion, coal combustion, soil, industry, and aged sea salt. Although the identified sources were consistent, some marker species and their contributions varied between XRF-PMF and ICP-MS-PMF. Significant differences were observed in the contributions of mobile sources, biomass burning, oil combustion. CBPF results indicated slight variations due to differing trace element portions. The DTT assay results showed an oxidative potential with DTTv of 0.444 nmol/min/m³ and DTTm of 0.251 nmol/min/μg. OC had the highest correlation with DTTv (r=0.864). Secondary nitrate and coal combustion correlated with DTTm in XRF-PMF results, whereas industry and oil combustion correlated in ICP-MS-PMF results. This study shows that different analytical methods can yield varying results, affecting chemical species concentrations, PMF results, and oxidative stress estimates. The correlation with oxidative stress was influenced by concentration data. However, PMF's consideration of both concentration and uncertainty data ensures reliable results. Therefore, selecting the appropriate instrument is crucial depending on the specific PM2.5 study being conducted.