Vapor–liquid equilibria of mixtures containing nitrogen, oxygen, carbon dioxide, and ethane
Vapor–liquid equilibria of mixtures containing nitrogen, oxygen, carbon dioxide, and ethane
复制标题
含有氮气、氧气、二氧化碳和乙烷的混合物的汽液平衡
DOI：
--
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
H. Hasse
中科院分区：
文献类型：
--
作者：
J. Stoll;J. Vrabec;H. Hasse
采用双中心Lennard-Jones Plus点四极模型，用分子模拟方法研究了含氮(N_2)、氧(O_2)、二氧化碳(CO_2)和乙烷(C_2H_6)的二元和三元混合物的汽液平衡。纯组件模型取自最近的工作。混合物用Lorentz-Berthelot组合规则描述。仅从纯组分数据对汽液平衡的预测与实验数据符合得很好，例如，正确地预测了二氧化碳+乙烷体系的共沸行为。通过将合并规则能量项中的一个参数调整为二进制数据，实现了进一步的改进。为此，提出了一种简单有效的方法。本文所研究的五个二元体系(N_2+O_2、CO_2+C_2H_6、O_2+CO_2、N_2+CO_2、N_2+C_2H_6)的汽液平衡、饱和密度和汽化热的分子模型与实验数据符合得很好。在不进一步调整模型参数的情况下，对两种三元混合物(N_2+O_2+CO_2，N_2+CO_2+C_2H_6)的汽液平衡进行了较好的预测。使用相同的数据将分子模拟的结果与Peng-Robinson状态方程、PC-SAFT状态方程和BACKONE状态方程的结果进行比较，以确定模型参数。分子模拟和状态方程与系统特有的二元相互作用参数的关联质量相似，分子模拟的预测能力明显优于分子模拟。
Vapor–liquid equilibria of binary and ternary mixtures containing nitrogen (N2), oxygen (O2), carbon dioxide (CO2) and ethane (C2H6) are studied by molecular simulation using two-center Lennard-Jones plus point quadrupole models. Pure-component models are taken from recent work. Mixtures are described using the Lorentz-Berthelot combining rules. Predictions of vapor–liquid equilibria from pure-component data alone agree well with experimental data, for example, the azeotropic behavior of the carbon dioxide + ethane system is predicted correctly. Further improvements are achieved by adjusting one parameter in the energetic term of the combining rule to binary data. For this purpose, a simple and efficient procedure is proposed. Excellent agreement between the molecular models and experimental data for vapor-liquid equilibria, saturated densities, and enthalpies of vaporization is observed for the five binary systems studied in the present work (N2+O2, CO2+C2H6, O2+CO2, N2+CO2, N2+C2H6). Vapor–liquid equilibria of two ternary mixtures (N2+O2+CO2, N2+CO2+C2H6) are predicted well without any further adjustment of model parameters. Results from molecular simulation are compared to those from the Peng-Robinson equation of state, the PC-SAFT equation of state, and the BACKONE equation of state using the same data to determine model parameters. The quality of correlations with system-specific binary interaction parameters from molecular simulation and equations of state is similar, and the predictive power of molecular simulation is clearly superior.