Volatile organic compounds emissions from traditional and clean domestic heating appliances in Guanzhong Plain, China: Emission factors, source profiles, and effects on regional air quality

作者：Jian Sun, Zhenxing Shen, Leiming Zhang, Yue Zhang, Tian Zhang, Yali Lei, Xinyi Niu, Qian Zhang, Wei Dang, Wenping Han, Junji Cao, Hongmei Xu, Pingping Liu, Xuxiang Li · 发表于：Environment International · 年份：2019 · DOI：10.1016/j.envint.2019.105252 · 被引用次数：78 · 研究领域：Atmospheric chemistry and aerosols、Air Quality and Health Impacts、Vehicle emissions and performance

Solid (biomass and coal) fuels burned for residential heating are major sources of atmospheric volatile organic compounds (VOCs). In this study, VOC samples were collected in-situ from chimneys in 10 typical heating scenarios in rural areas of the Guanzhong Plain. A modified SUMA canister approach was then employed, followed by gas chromatography/mass spectrometry analysis. Emission factors (EFs) (as received basis) of targeted non-methane VOCs (NMVOCs) varied from 90.3 ± 29.3 to 12300 ± 1510 mg kg−1 in descending order of fuel wood > maize straw > bitumite ≫ anthracite (p < 0.05). Both clean stove and coal briquetting technologies effectively reduced VOC EFs compared with traditional heating methods. The EFs of methane (CH4) had similar characteristics to those of NMVOCs. However, they yielded different correlations with CO because of their differing mechanisms of formation. Coefficient of divergence (CD) values showed that a semi-gasifier has a limited effect on changing VOC profiles compared with a traditional stove using the same fuels. However, different types of fuel produce CD values over 0.50, which should therefore be classified as different sub-categories in source apportionment models. Correlation analysis showed that volatile matter content (V%) and modified combustion efficiency (MCE) were the two primary factors influencing NMVOC and CH4 emissions. A stepwise linear regression analysis showed that V%, MCE, and element nitrogen content (N%) can be used to predict...