Heterosegmental Modeling of Long-Chain Molecules and Related Mixtures Using PC-SAFT: 2. Associating Compounds

Abstract

Renewable resources such as natural fats and oils are progressively used as feedstock for the chemical industry due to the depletion of fossil resources. Natural fats and oils primarily comprise triglycerides that are the basis for long-chain molecules such as fatty alcohols, fatty amines, and fatty acids. The saturated representatives of these individual families are composed of an identical functional head moiety and an n-alkylic residue that only varies in chain length within a homologous series. In the first part of our work (Haarmann et al., Ind. Eng. Chem. Res. 2019, 58 (7), 2551−2574) a heterosegmental approach of the Perturbed Chain Statistical Associating Fluid Theory was proposed with the focus on polar long-chain molecules. Within the heterosegmental approach, a (saturated and linear) long-chain molecule is modeled as comprising a tail domain described as an n-alkane and a universal head domain. In this work, this heterosegmental approach is extended to associating long-chain molecules with the focus on the homologous series of n-alcohols, n-amines, and n-alkyl carboxylic acids. Applying the heterosegmental approach, vapor pressures and saturated-liquid densities of the pure associating long-chain compounds were predicted in very good agreement with the experimental data. The same holds for the molar excess enthalpies of the binary mixtures long-chain compound + n-alkane, which were modeled for a broad range of chain lengths without using any additional model parameter. Moreover, the mutual solubilities in the binary mixtures long-chain compound + water could be predicted in remarkable agreement with the available experimental data for a broad range of chain lengths of the long-chain compounds.