The water-soluble ionic composition of PM2.5 in Shanghai and Beijing, China

Abstract

A year-long field study to characterize the ionic species in PM2.5 was carried out in Shanghai and Beijing, China, in 1999–2000. Weekly samples of PM2.5 were collected using a special low flow rate (0.4 l min⁻¹) sampler. In Shanghai, SO₄²⁻ NO₃⁻ and NH₄⁺ were the dominant ionic species, which accounted for 46%, 18% and 17% of the total mass of ions, respectively. Local SO₂ emissions were an important source of SO₄²⁻ in PM2.5 because the SO₄²⁻ concentration was correlated with the SO₂ concentration (r=0.66). The relatively stable SO₄²⁻/SO₂ mass ratio over a large range of temperatures suggests that gas-phase oxidation of SO₂ played a minor role in the formation of SO₄²⁻. The sum of SO₄²⁻ and NO₃⁻ was highly correlated with NH₄⁺ (r=0.96), but insufficient ammonium was present to totally neutralize the aerosol. In Beijing, SO₄²⁻, NO₃⁻ and NH₄⁺ were also the dominant ionic species, constituting 44%, 25% and 16% of the total mass of water-soluble ions, respectively. Local SO₂ emissions were an important source of SO₄²⁻ in the winter since SO₄²⁻ was correlated with SO₂ (r=0.83). The low-mass SO₄²⁻/SO₂ ratio (0.27) during winter, which had low humidity, suggests that gas-phase oxidation of SO₂ was a major route of sulfate formation. In the summer, however, much higher mass ratios of SO₄²⁻/SO₂ (5.6) were observed and were ascribed to in-cloud sulfate formation. The annual average ratio of NO₃⁻/SO₄²⁻ was 0.4 and 0.6 in Shanghai and in Beijing, respectively, suggesting that stationary emissions were still a dominant source in these two cities.

Introduction

Shanghai and Beijing are two of the largest cities in China. Both cities have populations of over 10 million. These two urban areas have experienced a rapid increase in the use of vehicles, concurrent with large increases in energy consumption. Particulate pollution has become a major problem. Many short-term studies on TSP and PM10 (particle size smaller than 10 μm) have been published since the 1980s (Winchester et al., 1981; Huebert et al., 1988). However, data on PM2.5, especially regarding the semi-volatile species in PM2.5, e.g., NO₃⁻ and Cl⁻ and dicarboxylic acids, is very limited. A year-long study characterizing PM2.5 in Shanghai and Beijing, sponsored by the General Motors Co., was initiated in March 1999. The composition and seasonal trends of PM2.5 in Beijing and Shanghai have been described by He et al. (2001) and Ye et al. (2002), respectively. SO₄²⁻, NO₃⁻ and NH₄⁺ are the dominant ionic species in PM2.5, which account for about one-third of the total PM2.5 mass in Shanghai and Beijing. The sum of SO₄²⁻, NO₃⁻ and NH₄⁺, the crustal species and the carbonaceous matter (organic and element carbon) were within 5% of the PM2.5 mass in Shanghai and 20% of the PM2.5 mass in Beijing (He et al., 2001; Ye et al., 2002). In their studies, crustal species were believed to originate from local soil and/or dust storms and carbonaceous species principally originated from primary sources. The formation of SO₄²⁻ and NO₃⁻ in PM2.5 was not investigated, but it should be done since an understanding of the origin of these species impacts control decisions. In Beijing, the vehicular population has grown at a rate of 15% per year since 1990. The resulting increase in NO_x emissions is expected to have an impact on the PM2.5 nitrate concentration. On the other hand, clean fuels such as natural gas have replaced coal used for small domestic heating boilers in the central urban area in Beijing. This has reduced the SO₂ and HCl emissions. In this paper, we focus on the composition, source, and the formation of the secondary ionic species, with an emphasis on sulfate, nitrate, ammonium, chloride and dicarboxylic acids.

Measurements of semi-volatile species are complicated by sampling artifacts, resulting from gas–particle and particle–particle interactions (Sickles et al., 1999). Denuders for absorbing gases prior to particle collection and back filters for absorbing HNO₃, HCl and NH₃ evaporated from collected particles have been widely accepted as effective tools to minimize sampling artifacts (Chow and Watson, 1998). In this study, a special low flow rate (0.4 l min⁻¹) sampler equipped with three parallel cassettes (one cassette equipped with an Na₂CO₃-coated denuder and an nylon back filter) was used to collect PM2.5. Correlation analysis along with seasonal variation in ratios will be examined to provide insight into the formation of secondary species and the relative importance of stationary and mobile sources of nitrogen and sulfur in PM2.5.

Dicarboxylic acids and/or their salts are important constituents in the water-soluble organic fraction of PM (Kawamura and Kaplan 1987; Kerminen et al., 2000) and are expected to be useful in source apportionment. Kawamura and Kaplan (1987) proposed that vehicle exhaust is an important source of dicarboxylic acids in PM. Given that organic carbon was the most abundant species in PM2.5 in Shanghai and Beijing (He et al., 2001; Ye et al., 2002), obtaining more information on its source is very important for control purposes. In this study, the composition of dicarboxylic acids in PM2.5 was determined only for samples collected after December 1999. Correlation analyses between dicarboxylic acids and other species, such as organic carbon, sulfate and nitrate in PM2.5 and O₃, was performed to understand the source of dicarboxylic acids and the formation of secondary species.