Prediction of Properties of Fluids by Molecular Simulation Techniques

Abstract

Chemical Engineering Thermodynamics and Transport Phenomena are driven by the need to have physical description of phase properties for pure components and fluid mixtures. Literature is rich with the experimental data about the various transport properties of alkanes, alcohols and water. A few research works are available where molecular simulation techniques were used to model such fluids and predict the properties. However, the literature data is scanty when it comes to lower alkanes and also mixture of alkanes. In the present study, Green-Kubo formulae have been used to predict the viscosity for such lower alkanes at a set of temperatures and at different box sizes. In the present study, both MC and MD simulation technique have been applied to study the properties of a basic fluid system of Argon in the temperature range of 150 to 700 K and reduced density range of 0.05 = ? = 1.0. The average of the physical properties like potential energy and pressure was calculated in MATLAB environment. The standard deviation for MC and MD simulation was found out to be 0.502 and 0.227 respectively. Then we gradually moved to complex fluids such as alkanes, water and alcohol. So, MC simulations was carried out to predict the average energy and density of alcohols, alkane and three different water models and the results are successfully validated with the literature data with a standard deviation of 0.054. MD simulation was used to predict the viscosity and average density of ethane, propane, butane and isobutene at different temperatures and different box sizes. The viscosity of a mixture of ethane and propane in the ratio 1:1 was also studied at different parameters. The effect of variation of temperature and number of molecules in the box on the accuracy of the prediction of viscosity was also studied.