Transport property prediction and inhomogeneity analysis of supercritical n-Dodecane by molecular dynamics simulation

Predicting the transport properties of fuels in high-pressure conditions such as those for supercritical combustion has been a major challenge. In this study, molecular dynamics method of optimised potentials for liquid simulations force field is used to model the diesel fuel surrogate, i.e. n-Dodecane molecule. The Green-Kubo and Einstein methods are used to predict basic transport properties with conditions ranging from the high-pressure nozzle condition (150 MPa, 363 K) to the supercritical chamber environment (6 MPa, 900 K). The computed values agree well with experimental results in conditions with temperature and pressure far away from the critical point. But large discrepancies are observed in the neighborhood of the critical region due to density fluctuation and the critical divergence. The bulk viscosity of n-Dodecane is calculated and found to be hundreds of times greater than shear viscosity in supercritical conditions which cannot be ignored in computational fluid dynamics simulations. The inhomogeneity of supercritical fluids associated with the Widom and Frenkel lines is studied by mapping the loci of the extrema and ridges for compressibility coefficient, heat capacity, density fluctuation and velocity autocorrelation function on the P-T phase diagram. In addition, a new criterion is developed to distinguish inhomogeneity and structural transition of n-Dodecane into the rigid-like, liquid-like and gas-like fluids, by evaluating the decay of radial distribution function peak heights quantitatively.