Five nonelectrolyte solution models are used to predict infinite dilution activity coefficients (γ∞) of five linear, four branched, and two cyclic alkanes in 67 solvents at 25 °C, and the results are compared with experimental data. The models use two distinct approaches to the prediction of γ∞. The solution of groups concept provides the basis for three versions of the UNIFAC model: original UNIFAC, γ∞-based UNIFAC, and modified UNIFAC (Dortmund). The MOSCED and the SPACE models avoid the group concept and use only pure component parameters. For a database of 737 limiting activity coefficients, the SPACE model gave an average absolute error of 8.1%, and in only 13.3% of the cases were the errors worse than 15%. The modified UNIFAC model gave an absolute average error of 9.8%, and 32% of the predicted γ∞ had errors larger than 15%. The SPACE approach also produced the most reliable estimations over a wide range of activity coefficient values.
No takes yet. Share an insight, caveat, or question.
Castells et al. (1999) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: