Simulation of Liquid-Liquid Equilibria with Molecular Models Optimized to Vapor-Liquid Equilibria and Model Development for Hydrazine and Two of Its Derivatives

In the chemical industry, knowledge on fluid phase equilibria is crucial for design and optimization of many technical processes. In a chemical plant, the costs for separation facilities constitute one of the highest investment outlays, typically in the order of 40–80% [15]. Not only vapor-liquid equilibrium(VLE) data are of interest, e.g. for distillation columns, but also other types of phase equilibria. For example, liquid-liquid equilibrium (LLE) data provide the basis for extraction processes. Classically, thermodynamic data for the design of such processes have to be measured experimentally and have to be aggregated by empirical correlations. For practical applications this leads to problems. For example, it is not possible to describe the entire fluid phase behavior consistently with a single model and set of parameters. Thus LLE data cannot be predicted reliably from VLE data (or vice versa) based on such correlations. Furthermore, the effort for measurements in the laboratory is very high, because every single fluid system of interest has to be measured individually. This approach particularly reaches its limits when multicomponent fluids or systems with multiple phases are of interest due to the sheer amount of independent variables. In a recent study by Hendriks et al. [10] about the demand of thermodynamic and transport properties in the chemical industry, the urgent need for a reliable and predictive approach to describe VLE as well as LLE with a single model and parameter set is pointed out.