Under a very tough situation that there has been increasing concern to the air quality in underground subway spaces, this study set sights on the thorough estimation of the chemical properties and source apportionment of particulate matter (PM) collected on an underground subway platform by a cooperative approach of semi-bulk and single particle analyses. The size-resolved PMs were intensively collected on the platform of Miasageori station on the Seoul Subway Line-4, and then, they were semibulkily analyzed by a PIXE and the TOR<TEX>$^{(R)}$</TEX> method, and individually analyzed by a SEM-EDX. Overwhelmingly enriched iron was a notable feature of elemental concentration of <TEX>$PM_{2.5}$</TEX>. Source classification of iron in <TEX>$PM_{10-2.5}$</TEX> and <TEX>$PM_{2.5}$</TEX> performed along with their elemental concentrations, indicates that the railway originated iron accounts for 95.71% and 66.39% of total iron in <TEX>$PM_{10-2.5}$</TEX> and <TEX>$PM_{2.5}$</TEX>, respectively. Via a stoichiometric categorization, <TEX>$Fe_2O_3$</TEX>, <TEX>$CaAl_2Si_2O_8$</TEX>, <TEX>$Al_2O_3$</TEX>, and <TEX>$CaCO_3$</TEX> show more than 85% abundance ratio in individual coarse particles. The result of theoretical estimation of the subway derived organic carbon (<TEX>$OC_{Subway}$</TEX>) suggests that <TEX>$OC_{Subway}$</TEX> in <TEX>$PM_1$</TEX> and <TEX>$PM_{2.5-1}$</TEX> account for 75.86% and 51.88% of total organic carbon, respectively.