The particle emission characteristics and chemical components of six kinds of textiles containing silver nanoparticles and three kinds of silver nano-sprays were tested in the present study. For the textiles containing silver nanoparticles, the sock A with nanoMark contains low silver nanoparticle concentration of 12.63 ± 0.57 μg/g, which were distributed uniformly within the whole sock. The sock B and C without nanoMark contain higher silver nanoparticle concentration of 2499.10 ± 122.59 and 34.42 ± 4.74 μg/g, respectively, as comparted with the sock A, in which the silver nanoparticles were concentrated on toes’ spot. The silver nanoparticle concentration in the sock D was only observed on the toes’ spot of 2.89 μg/g. The towels and the handkerchief with nanoMark contain 8.75 and 4.73 μg /g silver nanoparticle concentration, respectively. After the first washing, textiles containing silver nanoparticles were found to release more metals related to the materials added in the textiles, as compared to the tested socks, towels and handkerchiefs. The released metal concentrations decreased with increasing times of washing. After 20 times of washing, the sock A emitted the lowest particle concentration and released the lowest metal concentration among the all textiles. Besides, according to the results analyzed by TEM and EDS, the silver signal is very low, suggesting that the silver nanoparticles added in the sock A is not easy to be emitted by rubbing. For the nano-sprays containing silver, the experimental results show that risk of nanoparticle exposure for human body is very high within 20 to 60 seconds emission test. The mass concentration distributions of three nano-sprays containing silver measured by MOUDI were slightly higher than those measured by ELPI of 8, 6 and 8 %. The mass concentration distributions of nano-sprays A and B showed bimodal distributions, in which the first peak was located between 1-3 μm and the second peak was located between 0.18-0.48 μm. Nano-sprays C shows single modal mass concentration distribution with peak in the size range between 0.29-0.44 μm. Another work item in this project is to measure the number concentration and mass concentration distribution of atmospheric nanoparticles. The average number concentration measured in the Zhongshan site was 1.8 ± 1.2×104 #/cm3, which highly correlated with NOx and CO influenced by traffic emissions. The average number concentration was 8.0 ± 5.4×103 #/cm3 in Shalu site, which slightly correlated with O3 concentration influenced by photochemical reaction and industries emissions. The average number concentration was measured to be 4.7 ± 1.8×103 #/cm3 in Douliou site and the correlation between SO2 concentration and the average number concentration is high due to traffic emissions and photochemical reaction. Based on the number and mass concentration measured in the sampling as the aforementioned, the effective density was calculated to convert the mass concentration measured by MOUDI into number concentration and surface concentration. The results showed that although the mass ratio of PM0.1/PM2.5 of 7.9 ± 4.4 % was low, the number concentration ratio of PM0.1/PM2.5 was 89.0 ± 5.5 %, and the surface concentration ratio was 42.1 ± 12.8 %. Therefore, we should pay attention to the adverse effect of PM0.1 on human health. The experimental results showed that the submicron particle number concentration measured by APS (Aerodynamic particle sizer, Model 3321, TSI Incorporated, USA) was lower than those measured by OPC (Optical particle counter, Model 1.108, GRIMM technology inc., USA) because the APS detection efficiency for particles with size smaller than 0.7 μm is low. Besides, the mass concentration measured by PM2.5 manual sampler PQ-200 (BGI Inc., USA) was highly correlated with those measured by DustTrak (DustTrak DRX 8533, TSI Incorporated, USA) (R2= 0.70). The correlation of real-time mass concentration measured by DustTrak and by OPC is high with R2 of 0.82。 In this study, the applicability of the real-time measurement of atmospheric PM2.5 water soluble ions and trace acidic and basic gases concentration was investigated. The results showed that the atmospheric PM2.5 water soluble ions and trace acidic and basic gases concentrations measured by Parallel plate wet denuder particle-into-liquid sampler (PPWD-PILS) were in good agreement with those measured by Porous metal denuder sampler (PDS) with R2 > 0.75. In NCTU and Zhudong, NH4NO3 and (NH4)2SO4 were measured to be the main components of the PM2.5 inorganic salts. Inorganic salts in NCTU accounted for 54.0 ± 9.5% of PM2.5, which were highly aged aerosols. OM and EC measured to be higher than those in NCTU were mainly came from traffic emissions. PM2.5 concentration in Zhudong varied significantly, which mainly came from north-eastern side transported by north-eastern monsoon. The OC/EC ratio suggested the OM and EC in Zhudong were mainly attributed to combustion of fussil fuel.

nanoparticles, effective density, particle-into-liquid sampler