(293f) Thermodynamic Modeling for Surfactant Based Separation Processes

The micellar extraction and the micellar chromatography are well known methods for the separation and analysis of products from aqueous systems. Both methods are based on the formation of aggregates of different size and shape (micelles), which are formed in aqueous surfactant solutions due to self assembly above the critical micellar concentration (cmc). The partition of target molecules is mainly caused by hydrophobic and electrostatic interactions between the micelles and the solute. The most relevant factor is the partition coefficient of all related species, which defines the selectivity of the process. Efficient separation processes can be designed if an appropriate surfactant or surfactant combination is selected for a multi component system. The dominating factors influencing a micellar separation process are the surfactant itself, its concentration, the characteristics of the solute, the pH-value of the solution and the ionic strength. Effective processes need fully optimized systems, however the vast amount of parameters need a theoretical description and modeling of the present interactions.

In this work, the potential of surfactant based separation processes for isolation of biorelated products is discussed. Partition coefficients of various biorelated solutes in micellar solutions were measured using the Micellar Liquid Chromatography (MLC).This fast and efficient screening technique for partition coefficients in micellar solutions provides the experimental basis for the design of micellar extraction processes. Further, the partition coefficients were predicted using the thermodynamic model COSMO-RS. It is demonstrated that the partition coefficients can be predicted quantitatively based solely on the chemical structure of the substances. Furthermore, the model COSMO-RS is extended in order to account for ionic strength, and dissociation of the compounds at corresponding pH. Regarding the vast amount of different parameters like the influence of e. g. buffers, salts and alcohols in situ predictions using COSMO-RS is an effective tool to optimize surfactant based separation processes. The results are validated in a countercurrent column for micellar extraction.

The described modeling is extended for other types of liquid-liquid equilibria, especially in the field of drug delivery and biotechnology. Modeling of the product isolation from the biotechnological production based on salt/sugar induced liquid-liquid phase split is presented.