关键词: 烷基苯, Mie势力场, 气液相平衡, Monte Carlo模拟, 优化

Abstract: The realization of molecular management of refinery processes relies on physical property data. However, experimental data are scarce for many molecules present in petroleum fractions. To bridge this data gap, molecular simulation has emerged as an effective approach for generating high-quality property data. To address the insufficient accuracy of existing general force fields (e.g., TraPPE) in predicting key properties like saturated vapor pressure, a high-precision and transferable united-atom Mie potential force field applicable for vapor-liquid equilibrium simulations of alkylbenzenes was developed. This development was based on the high-precision n-6 Mie potential and Gibbs ensemble Monte Carlo simulation methods, accelerated via an analytical equation-of-state surrogate optimization approach. Results demonstrate that the developed Mie potential force field achieves average relative deviations of 1%–2% in predicting saturated vapor pressure, vaporization enthalpy, and liquid density for benzene and ethylbenzene across a wide temperature range, significantly outperforming TraPPE series force fields. Simultaneously demonstrating excellent prediction accuracy for vapor-liquid equilibrium compositions in binary systems such as benzene-cyclopentane, ethylbenzene-cyclohexane, and m-xylene-cyclohexane; Notably, the developed Mie potentialforce field parameters exhibit excellent transferability, with an average absolute deviation of only 2.95 K in predicting the boiling points of various alkylbenzene compounds, offering a highly promising approach for expanding physical property data in refining and chemical processes.

Key words: alkylbenzenes, Mie potential force field, vapor-liquid equilibrium, Monte Carlo simulation, optimization