Abstract

In this work, we determine the vapor-liquid (VL) coexistence and interfacial properties of the hydroquinone (HQ) pure system from NVT molecular dynamics simulations. We employ the direct coexistence technique to put in contact both phases in the same simulation box and generate the VL interface. Five different models have been tested to describe the HQ molecule in order to assess the performance of different approaches. The first two models are based on the Transferable Parameters Potentials for Phase Equilibria (TraPPE) force field. The first TraPPE model is the original one (Rai and Siepmann, 2007 [21] and Rai and Siepmann, 2013 [22]) based on an all-atoms approach (TraPPE-AA). The second TraPPE model is proposed for the first time in this work and is based on an united-atoms approach where the –CH groups from the aromatic ring are modeled as a single interaction site (TraPPE-Hybrid-UA). We also use two HQ models based on the Optimized Potentials for Liquid Simulations (OPLS) force fields. Both OPLS models have already been reported in the literature (Jorgensen et al., 1996 [25] and Comesaña et al., 2018 [14]), but this is the first time that are used to describe VL equilibria and interfacial behavior. In addition, we propose a new Coarse Grain (CG) HQ model based on the Statistical Associating Fluid Theory (SAFT) framework. We determine density profiles, coexistence densities, vapor pressure, interfacial thicknesses, and interfacial tensions as obtained from the NVT simulations with the five different models. We explore the VL behavior of pure HQ system from 500 to 750K. Remarkably good agreement has been found between the simulation results obtained by the TraPPE-AA, CG, and both OPLS models. Unfortunately, the results obtained by the TraPPE-Hybrid-UA model proposed in this work show discrepancies with the rest of the HQ models. Finally, we also determine the critical temperature, density, and pressure from the analysis of the coexistence densities, vapor pressure, and interfacial tension. The critical temperature predicted by the TraPPE-AA is in excellent agreement with the experimental data taken from the literature. The CG and both OPLS HQ models slightly underestimate experimental data, and the TraPPE-Hybrid-UA model clearly overestimates it.