Model Development

Model Development

Model Development                                                                                   

For the synthesis and design of thermal separation processes and for solving special problems (e.g. in environmental protection, biotechnology, ...) thermodynamic models for the correlation and estimation of pure component as well as mixture properties are required. The research is focused on different types of models:

Activity Coefficient Models (gE -Models)

Based on the group contribution approach, reliable models with a wide range of applicability are being developed (UNIFAC, mod. UNIFAC , ASOG).

Equations of State

Work is focused on the further development of EOS of practical importance (Soave - Redlich - Kwong, Peng - Robinson) as well as reliable gE - mixing rules. On the basis of this work, group contribution equations of state were developed (PSRK / VTPR).

Octanol-Water Partition Coefficients

Based on the UNIFAC model, a new group contribution method for the estimation of octanol-water partition coefficients was developed, which shows several advantages compared to the method of Leo and Hansch. Besides KOW, also Henry coefficients and water solubilities can calculated.

Electrolyte Mixtures

After extensive tests of various existing models for the description of phase equilibria of mixtures containing electrolytes, based on a large data base new models were developed, which features an improved description of this type of mixtures with strong electrolytes (LIQUAC, LIFAC ).

Adsorption Equilibria

Using a large database and own measurements, different models for the correlation and prediction of adsorption equilibria were tested and a new predictive model was developed.

Group Contribution Methods for the Estimation of Pure Component Properties

With the help of an automatic procedure for the fragmentation of molecular structures and a huge database, a large number of group contribution methods for the prediction of different pure component properties were implemented. Based on the results of extensive tests, the development of new reliable methods with a large range of applicability is planned.

(Stand: 21.08.2020)