General method for prediction of thermal conductivity for well-characterized hydrocarbon mixtures and fuels up to extreme conditions using entropy scaling

Abstract A general and efficient technique is developed to predict the thermal conductivity of well-characterized hydrocarbon mixtures, rocket propellant (RP) fuels, and jet fuels up to high temperatures and high pressures (HTHP). The technique is based upon entropy scaling using the group contribution method coupled with the Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT) equation of state. The mixture number averaged molecular weight and hydrogen to carbon ratio are used to define a single pseudo-component to represent the compounds in a well-characterized hydrocarbon mixture or fuel. With these two input parameters, thermal conductivity predictions are less accurate when the mixture contains significant amounts of iso-alkanes, but the predictions improve when a single thermal conductivity data point at a reference condition is used to fit one model parameter. For eleven binary mixtures and three ternary mixtures at conditions from 288 to 360 K and up to 4,500 bar, thermal conductivities are predicted with mean absolute percent deviations (MAPDs) of 16.0 and 3.0% using the two-parameter and three-parameter models, respectively. Thermal conductivities are predicted for three RP fuels and three jet fuels at conditions from 293 to 598 K and up to 700 bar with MAPDs of 14.3 and 2.0% using the two-parameter and three-parameter models, respectively.