QI Chang, LU Diannan, LIU Yongmin. Prediction of thermodynamic properties of n-alkanes based on temperature-corrected force field[J]. CIESC Journal, 2018, (8): 3338-3347.
DOI:
QI Chang, LU Diannan, LIU Yongmin. Prediction of thermodynamic properties of n-alkanes based on temperature-corrected force field[J]. CIESC Journal, 2018, (8): 3338-3347.DOI: 10.11949/j.issn.0438-1157.20180294.
Prediction of thermodynamic properties of n-alkanes based on temperature-corrected force field
Aim to accurately predict the thermodynamic properties of n-alkanes and their mixtures at different temperature
in this paper
the non-bond Lennard-Jones (L-J) interaction parameter
ε
was optimized by using n-C4-C10 as the training set and all-atom molecular dynamics simulations. Here an empirical relationship between the L-J parameter (ε) and the reduced temperature (Tr) and carbon atomic number of n-alkanes (NC) was obtained. According to above-mentioned relationship
the values of ε at different temperatures were successfully calculated. Further
the properties of viscosity
density and diffusion coefficient of different n-alkanes and their mixtures were predicted using traditional and non-equilibrium molecular dynamics simulations with all-atom OPLS-AA force field corrected by temperature-dependent ε. Compared with previous theoretical prediction values and experimental data
the prediction with temperature-dependent ε showed that the physical properties of n-alkanes and their mixtures predicted by this method were in the best agreement with the experimental values. The relative deviations between the predicted values of density
viscosity and diffusion coefficient and their corresponding experimental counterparts were less than 2%
5% and 10%
respectively. The new predicted values are significantly superior to those predicted by both other theoretical methods and molecular dynamics simulations with the original OPLS-AA force field. Therefore
the established temperature-dependent force field has important practical application value for accurately predicting the thermodynamic properties of n-alkanes and their mixtures by molecular dynamics simulations.