Tong Chinghang, Blanco Mario, Goddard William A, Seinfeld John H
Department of Environmental Science and Engineering, Material and Process Simulation Center, California Institute of Technology, Pasadena, California 91125, USA.
Environ Sci Technol. 2004 Jul 15;38(14):3941-9. doi: 10.1021/es0354216.
Ambient particulate matter contains polar multifunctional oxygenates that partition between the vapor and aerosol phases. Vapor pressure predictions are required to determine the gas-particle partitioning of such organic compounds. We present here a method based on atomistic simulations combined with the Clausius-Clapeyron equation to predict the liquid vapor pressure, enthalpies of vaporization, and heats of sublimation of atmospheric organic compounds. The resulting temperature-dependent vapor pressure equation is a function of the heat of vaporization at the normal boiling point [deltaHvap(Tb)], normal boiling point (Tb), and the change in heat capacity (liquid to gas) of the compound upon phase change [deltaCp(Tb)]. We show that heats of vaporization can be estimated from calculated cohesive energy densities (CED) of the pure compound obtained from multiple sampling molecular dynamics. The simulation method (CED) uses a generic force field (Dreiding) and molecular models with atomic charges determined from quantum mechanics. The heats of vaporization of five dicarboxylic acids [malonic (C3), succinic (C4), glutaric (C5), adipic (C6), and pimelic (C7)] are calculated at 500 K. Results are in agreement with experimental values with an averaged error of about 4%. The corresponding heats of sublimation at 298 K are also predicted using molecular simulations. Vapor pressures of the five dicarboxylic acids are also predicted using the derived Clausius-Clapeyron equation. Predicted liquid vapor pressures agree well with available literature data with an averaged error of 29%, while the predicted solid vapor pressures at ambient temperature differ considerably from a recent study by Bilde et al. (Environ. Sci. Technol. 2003, 37, 1371-1378) (an average of 70%). The difference is attributed to the linear dependence assumption thatwe used in the derived Clausius-Clapeyron equation.
环境颗粒物包含在气相和气溶胶相之间分配的极性多功能含氧化合物。需要蒸气压预测来确定此类有机化合物的气-粒分配。我们在此提出一种基于原子模拟并结合克劳修斯-克拉佩龙方程的方法,以预测大气有机化合物的液体蒸气压、汽化焓和升华热。所得的温度依赖性蒸气压方程是正常沸点下的汽化热[ΔHvap(Tb)]、正常沸点(Tb)以及化合物在相变时(液体到气体)的热容变化[ΔCp(Tb)]的函数。我们表明,汽化热可以从通过多次采样分子动力学获得的纯化合物的计算内聚能密度(CED)中估算出来。模拟方法(CED)使用通用力场(Dreiding)和由量子力学确定原子电荷的分子模型。在500 K下计算了五种二元羧酸[丙二酸(C3)、琥珀酸(C4)、戊二酸(C5)、己二酸(C6)和庚二酸(C7)]的汽化热。结果与实验值一致,平均误差约为4%。还使用分子模拟预测了298 K时相应的升华热。使用推导的克劳修斯-克拉佩龙方程也预测了五种二元羧酸的蒸气压。预测的液体蒸气压与现有文献数据吻合良好,平均误差为29%,而预测的环境温度下的固体蒸气压与比尔德等人最近的一项研究(《环境科学与技术》,2003年,37卷,1371 - 1378页)有很大差异(平均为70%)。这种差异归因于我们在推导克劳修斯-克拉佩龙方程时使用的线性依赖假设。