Two-year measurements of particle number size distribution (3 nm-10 mu m) were conducted in Beijing, China since March 2004. Their seasonal, weekly and diurnal variations and dependencies on meteorological parameters were investigated. The annual average particle number concentrations of the nucleation mode (3-20 nm), Aitken mode (20100 nm), and accumulation mode (0.1-1 mu m) are 9000 cm(-3), 15,900 cm(-3), and 7800 cm(-3), respectively. Particle number concentrations in Beijing are generally higher than that in cities of developed countries, especially for the accumulation mode particles. Both the highest total particle number concentration and the lowest volume concentration occurred in spring due to the frequent nucleation events. However, the minimum particle number concentration was observed in summer, and the maximum volume concentration in fall. The diurnal variation of the nucleation mode particles was mainly influenced by nucleation events, primarily in spring and winter. The diurnal variation of number concentration of the Aitken mode particles closely correlates with the traffic densities in all the four seasons. At the same time, obvious contribution of the growth of the nucleation mode to the number concentration of the Aitken mode particle has been also found in spring, Summer, and fall. Significant differences in diurnal patterns of particle number concentrations between workdays and weekends are not observed in Beijing. Local wind speed plays an important role in shaping the particle number size distributions in the urban area of Beijing. With increasing wind speed, the nucleation and coarse mode particle number concentrations increase, while the number concentrations of the Aitken mode and accumulation mode particles decrease. A "U-shape relationship" between the total particle volume concentration and wind speed is observed. Frequently low wind speed (lower than 3 m s(-1)) in Beijing is one of key factors leading to the poor air quality and low visibility. (C) 2008 Elsevier Ltd. All rights reserved.
Concentration and source rate of precursor vapors participating in particle formation and subsequent growth were investigated during the Pearl River Delta intensive campaign (PRD2004, October 2004) in southeastern China. Four new particle formation event days and a typical non-event day were selected for our analysis. Atmospheric sulphuric acid, the important precursor vapor in nucleation and growth, were simulated with a pseudo steady-state model based on the measurements of SO2, NOx, O-3; CO, non-methane hydrocarbon (NMHC) and ambient particle number concentrations as well as modeled photolysis frequencies obtained from measurements. The maximum midday sulphuric acid concentrations vary from 4.53 x 10(7) to 2.17 x 10(8) molecules cm(-3), the corresponding source rate via reaction of OH and SO2 range between 2.37 x 10(6) and 1.16 x 10(7) molecules cm(-3) s(-1). Nucleation mode growth rate was derived from size spectral evolution during the events to be 6:8-13.8 nm h(-1). Based on the growth rate, concentration of the vapors participating in subsequent growth were estimated to vary from 1.32 x 10(8) to 2.80 x 10(8) molecules cm(-3) with corresponding source rate between 7.26 x 10(6) and 1.64 x 10(7) molecules cm(-3) s(-1). Our results show the degree of pollution is larger in PRD. Sulphuric acid concentrations are fairly high and have a close correlation with new particle formation events. Budget analysis shows that sulphuric acid alone is not enough for required growth; other nonvolatile vapors are needed. However, sulphuric acid plays an important role in growth; the contribution of sulphuric acid to growth in PRD is 12.4%-65.2%.
Real-time measurements of acidic trace gases (HCl, HNO(3), HONO, and SO(2)), ammonia, and water-soluble ions in PM(2.5) were conducted at Xinken, a coastal site downwind of Guangzhou, from 4 October to 4 November 2004, as part of the Pearl River Delta (PRD) intensive field campaign. The average concentrations of HCl, HONO, HNO(3), SO(2), and NH(3) are 2.8, 2.9, 6.3, 55.4, and 7.3 mu g m(-3) respectively, and 2.4, 7.2, 24.1, and 9.2 mu g m(-3) for Cl(-), NO(3)(-), SO(4)(2-), and NH(4)(+) in PM(2.5). The diurnal variations of both HCl and HNO(3) showed higher concentrations during daytime and lower concentrations at night, and aerosol Cl(-) and NO(3)(-) showed an opposite diurnal patterns to HCl and HNO(3). The diurnal variation of NH(3) showed the similar pattern to that of aerosol NH(4)(+) with lower concentration during daytime and higher concentration at night. The average concentration of SO(2) during daytime was higher than that at night. The transportation of urban plumes to the sampling site could explain the higher concentration of SO(2) during daytime. HONO showed a clear diurnal variation with lower concentration during daytime and higher concentration at night. The HONO concentrations were positively correlated with the particle surface area concentrations, suggesting the formation of HONO through the heterogeneous conversion on particle surfaces could be significant. The ionic charge balance analysis included the cations derived from filter measurements indicates that the contribution of the cations in fine particle (PM(1.8)) to the charge balance is not pronounced. The theoretical equilibrium constant (K(e)) of NH(4)NO(3) is higher than the observed concentration product (K(m) = [NH(3)] x [HNO(3)]) during daytime, and lower than K(m) at night. This indicates that the atmospheric conditions during the sampling period did not favor the formation of NH(4)NO(3) during daytime. (c) 2008 Elsevier Ltd. All rights reserved.
Continuous measurements of aerosol number size distribution in the range of 3 nm-10 mu m were performed in Pearl River Delta (PRD), China. These measurements were made during the period of 3 October to 5 November in 2004 at rural/coastal site, Xinken (22 degrees 37'N, 113 degrees 35'E, 6m above sea level), in the south suburb of Guangzhou City (22 degrees 37'N, 113 degrees 35'E, 6m abovesea level), using a Twin Differential Mobility Particle Sizer (TDMPS) combined with an Aerodynamic Particle Sizer (APS). The aerosol particles at Xinken were divided into four groups according to the observation results: nucleation mode particles (3-30 nm), Aitken mode particles (30-130 nm), accumulation mode particles (130-1000 nm) and coarse mode particles (1-10 mu m). Concentrations of nucleation mode, Aitken mode and accumulation mode particles were observed in the same order of magnitude (about 10,000 cm(-3)), among which the concentration of Aitken mode particle was the highest. The Aitken mode particles usually had two peaks: the morning peak may be caused by the land-sea circulation, which is proven to be important for transporting aged aerosols back to the sampling site, while the noon peak was ascribed to the condensational growth of new particles. New particle formation events were found on 7 days of 27 days, the new particle growth rates ranged from 2.2 to 19.8 nm h(-1) and the formation rates ranged from 0.5 to 5.2 cm(-3) s(-1), both of them were in the range of typical observed formation rates (0.01-10 cm(-3) s(-1)) and typical particle growth rates (1-20 nm h(-1)). The sustained growth of the new particles for several hours under steady northeast wind indicated that the new particle formation events may occur in a large homogeneous air mass. (c) 2008 Elsevier Ltd. All rights reserved.