Precursor solubility is a crucial factor in industrial applications, dominating the outcome of reactions and purification steps. The outcome and success of thermodynamic modelling of this industrially important property with equations of states, such as Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT), vastly depends on the quality of the pure-component parameters. The pure-component parameters for low-volatile compounds such as ionic liquids (ILs) have been commonly estimated using mixture properties, e. g. the osmotic pressure of aqueous solutions. This leads to parameters that depend on the solvent, and transferability to other mixtures often causes poor modeling results. Mixture-independent experimental properties would be a more suitable basis for the parameter estimation offering a way to universal parameter sets. Model parameters for ILs are available in the literature [10.1016/j.fluid.2012.05.029], but they were estimated using pure-IL density data. The present work focuses on a step towards a more universal estimation strategy that includes new experimental vapor-pressure data of the pure IL. ILs exhibit an almost negligible vapor pressure in magnitude of usually 10^-5 Pa even at elevated temperatures. In this work, such vapor-pressure data of a series of 1-ethyl-3-methyl-imidazolium-based [C₂ mim]-ILs with various IL-anions (e. g. tetrafluoroborate [BF₄ ]^- , hexafluorophosphate [PF₆ ]^- , bis(trifluoromethylsulfonyl)imide [NTf₂ ]^- ) were experimentally determined and subsequently used for PC-SAFT parameter estimation. The so-determined parameters were used to predict experimental molecular precursor solubility in ILs and infinitely diluted activity coefficients of various solvents in ILs. The parameters were further compared to modeling results using classical parametrization methods (use of liquid-density data only for the molecular PC-SAFT and the ion-based electrolyte PC-SAFT). As a result, the modeled precursor solubilities using the new approach are much more precise than using the classical parametrization methods, and required binary parameters were found to be much smaller (if needed). In sum, including the pure-component vapor-pressure data of ILs opens the door towards parameter estimation that is not biased by mixture data. This procedure might be suitable also for polymers and for all kind of ionic species but needs extension to ion-specific parametrization in the long term.

	Year	Citations

About