Optimisation of the dynamical behaviour of the anisotropic united atom model of branched alkanes: application to the molecular simulation of fuel gasoline

• Published in: Molecular simulation Vol. 34; no. 2; pp. 211 - 230
• Main Authors: Nieto-Draghi, Carlos, Bocahut, Anthony, Creton, Benoît, Have, Pascal, Ghoufi, Aziz, Wender, Aurélie, Boutin, Anne, Rousseau, Bernard, Normand, Laurent
• Format: Journal Article
• Language: English
• Published: Taylor & Francis Group 01.02.2008; Taylor & Francis
• Subjects: anisotropic united atom; branched alkanes; Chemical Sciences; fuel gasoline; lumping methodology; olefins; shear viscosity; transport properties; Physical Sciences
• ISSN: 0892-7022; 1029-0435
• DOI: 10.1080/08927020801993370

In the present work, we have optimised the dynamical behaviour of the anisotropic united atom (AUA) intermolecular potential for branched alkanes developed by Bourasseau et al. [E. Bourasseau, P. Ungerer, A. Boutin, and A.H. Fuchs, Monte Carlo simulation of branched alkanes and long chain n-alkanes with anisotropic united atoms intermolecular potential, Mol. Sim. 28 (2002), pp. 317-336], by a modification of the energetic barrier of the torsion potential. The new potential (AUA(4m)) preserves all the intermolecular parameters and only explores an increment in the trans-gauche and gauche+-gauche − transition barrier of the torsion potential. This modification better reproduces transport properties like the shear viscosity, keeping the accuracy achieved in the original work for equilibrium properties. An extensive investigation of the shear viscosity of 12 different types of branched alkanes in a wide range of pressures and temperatures, shows that the AUA(4m) improves the accuracy of the original AUA4, reducing the absolute average deviation from 24 to 15%. In addition, molecular simulation results of the shear viscosity of olefins reveal that the original AUA potential is accurate enough to reproduce the experimental data with less than 12% of deviation. Finally, we present a consistent lumping methodology to perform molecular simulations on complex multi-component systems such as fuel gasoline by representing the real system by a simplified mixture with only tenths of species.