干旱区生态环境知识资源中心

A single particle aerosol mass spectrometer (SPAMS) was employed to characterize the particles in urban Beijing in the spring of 2011. The whole sampling time was divided into clear, hazy, and dusty day periods. Based on SPAMS results, the chemical composition of size-resolved particles in five major classes, including K-rich, carbonaceous, industry metal, dust, and Na-rich particles, and their corresponding sources were compared during clear, hazy, and dusty days. Air mass back trajectories are used to identify likely source regions for each period. Under stagnant meteorological conditions, local and regional emissions are a key factor in haze formation. K-rich and carbonaceous particles dominated under all conditions (25-54%), with higher contributions (23% and 29%) from industrial metal particles during hazy days, which could be attributed to the emission of biomass burning, coal combustion, and vehicle exhaust during all periods, whereas industrial metal processes are also a likely source for K-rich particles during hazy days. The aging of particles was observed over the whole sampling period, as evidenced by the presence of secondary species in all particle classes. Industrial metal particles possibly originate from industrial metal processes, vehicle exhaust, and coal combustion etc, while road and soil dust is also a possible source during dusty days. Dust particles increased greatly (17%) in a dust storm event with a coarse mode distribution, which are mainly associated with the mineral dust from deserted regions, whereas they could originate from local dust and coal combustion during non-dusty days. Last, Na-rich particles are mainly associated with industrial metal particles, coal, and fuel combustion during hazy days, and mineral dust during dusty days, respectively. Sea salt is a possible source of Na-rich particles under all conditions. Our analysis could shed a light on the understanding of the different source apportionment of aerosol particles under different meteorology/pollution conditions. (C) 2015 Elsevier Ltd. All rights reserved.