Keywords: 
• organic chemicals /
• (subcooled) liquid vapor pressure(P_L) /
• temperature-dependence /
• partial least square(PLS) regression /
• support vector machine(SVM)

Abstract: The (subcooled) liquid vapor pressure (P_L) is an important parameter to assess the distribution, transport and fate of organic chemicals in the environment. P_L possesses strong temperature-dependence. In order to fitl the gap of the data for the ecological risk assessment, it is of significance to develop an effective tool for predicting P_L at different temperatures. In this study, overall 10 478 logP_L values of 661 compounds at different temperatures are collected and used to develop two models, a partial least square (PLS) regression linear model and a support vector machine (SVM) nonlinear model. Results reveal that both models have satis-factory goodness-of-fit, robustness and external predictive performance. The predictive performance of SVM 2 2 model is slightly better than that of PLS model (PLS:Radj.tra=0.912, RMSE, r.=0.477, Q_ext²=0.910; SVM:R2adjtra=0.997, RMSE_tra=0.092, Q_ext²=0.967). Mechanistic explanation shows that the P_L of organic chemicals is mainly influenced by the temperature and Xlsol (intermolecular dispersive forces). Increasing temperature and decreasing intermolecular dispersive forces lead to the increase oflogP_L values. Moreover, the number of hydrogen bond, dipole and the molecular structure can also exert influence on P_L values. The applicability domain of the PLS model was characterized by the Williams plot. The developed models can be used to predict logP_L values of alkanes, alkenes, ketones, carboxylie acids, benzene, phenol, biphenyl, halogenatearomatic hydrocarbonsas well as N-and S-containing chemicals.