Chemical compositions and source identification of particulate matter (PM2.5 and PM2.5–10) from a scrap iron and steel smelting industry along the Ife–Ibadan highway, Nigeria

Abstract

To determine the chemical compositions and source identification of PM2.5 and PM2.5–10 fractions, airborne particulate matter (PM) samples were collected from May, 2011 through April, 2012 at three sites: up and downwind and within a scrap iron and steel smelting industry, Ife–Ibadan highway, south western Nigeria. Samples of PM2.5 (fine) and PM2.5–10 (coarse) were collected on Nuclepore polycarbonate filters using a low volume GENT sampler equipped with a stacked filter unit (SFU). A total of 200 samples were collected (100 of each fraction). The mass concentration of the sampled fine and coarse PM fraction ranged between 14.4–986.5μg/m3 and 11.2–3 250μg/m3, respectively. These values exceed the permissible daily limit (NAAQS) of 35μg/m3 for PM2.5 and 150μg/m3 for PM10. The samples were analyzed for black carbon (BC) using an optical transmissometer and for elemental concentrations using X–Ray Fluorescence (XRF). The size–resolved data sets were analyzed using Positive Matrix Factorization (PMF) to identify possible sources and estimate the contribution of these sources to the fine and coarse PM mass concentrations. Four source categories, providing stable profiles, were identified for both fine and coarse fractions. The identified sources and their contributions for the fine fraction are coking coal (83%), soil (10%), metallurgical industry (6%), and electronic waste processing (1%). For the coarse fraction, the identified sources are metallurgical production plus electronic waste (53%), suspended input materials (28%), soil (18%), and galvanized steel scrap with cadmium (1%). Conditional probability function (CPF) identified the local sources for both the fine and coarse PM samples. This work presents the first known major use of PMF in Nigeria for source identification in particulate matter (PM) studies.