470185 Integrating Molecular Modelling in Systems Engineering

Monday, November 14, 2016
Grand Ballroom B (Hilton San Francisco Union Square)
Erich A. Muller1, Claire S. Adjiman2, George Jackson3 and Amparo Galindo3, (1)Department of Chemical Engineering, Imperial College London, London, United Kingdom, (2)Department of Chemical Engineering, Centre for Process Systems Engineering, Imperial College London, London, United Kingdom, (3)Centre for Process Systems Engineering, Department of Chemical Engineering, Imperial College London, London, United Kingdom

The Molecular Systems Engineering team at Imperial College has championed the close integration of molecular modelling towards the intelligent molecular design of materials for engineering applications. A workhorse of the group is the development of the newest version of the Statistical Associating Fluid Theory equation of state (SAFT-γ) [1-3], that is effectively used to bridge the scales from molecular modelling to process and product design [4,5]. The SAFT model employs intermolecular potential parameters that are estimated directly from macroscopic thermophysical properties (e.g., fluid-phase equilibria) using the equation of state. This procedure provides a robust route to obtain coarse grained force fields which can then be used to obtain transport and interfacial properties [6-10].

Examples are provided in the areas of modelling of non-ionic surfactant behaviour, (direct calculation of critical micelle concentrations of CiEj alkyl-ethoxylates, superspreading of siloxane-based surfactants, interfacial tensions of light switching surfactants, liquid-liquid separations of polystyrene-solvent mixtures) where coarse grained models are used in a group-contribution fashion to obtain phase behaviour and molecular structure and self-assembly properties.

The presentation furthers discusses the “bottled SAFT” approach [11] where as the parameters for coarse grained moieties can be approximated through a corresponding states formulation of SAFT-γ. The resources in Bottled SAFT allow not only the fetch of the parameters but also the immediate production of scripts to be used in common MD suites. It currently hosts parameters for over 5000 molecules.

The SAFT model exemplifies a systematic and universal way to obtain transferable and representative potentials which can be employed to make molecular modelling directly accessible to non-experts by removing the barriers to developing intermolecular potentials.


[1] V Papaioannou, T Lafitte, C Aveñdano, CS Adjiman, G Jackson, EA Müller, and A Galindo, Journal of Chemical Physics, 140, 054107 (2014)

[2] T Lafitte, A Apostolakou, C Avendaño, A Galindo, CS Adjiman, EA Müller, and G Jackson, Journal of Chemical Physics, 139, 154504 (2013)

[3] S Dufal, T Lafitte, AJ Haslam, A Galindo, GNI Clark, C Vega, G Jackson,
Molecular Physics 113, 948-984 (2015).

[4] J Burger, V Papaioannou, S Gopinath, G Jackson, A Galindo, CS Adjiman,
AIChE Journal 61, 3249-3269 (2015)

[5] CS Adjiman, A Galindo, G Jackson, Computer Aided Chemical Engineering 34, 55 (2014)

[6] C Avendano, T Lafitte, A Galindo, CS Adjiman, G Jackson, EA Müller, Journal of Physical Chemistry B 115, 11154-11169 (2011)

[7] C Avendaño, T Lafitte, CS Adjiman, A Galindo, EA Müller, G Jackson, Journal of Physical Chemistry B 117, 2717-2733 (2013)

[8] T Lafitte, C Avendaño, V Papaioannou, A Galindo, CS Adjiman, G Jackson, EA Muller, Molecular Physics 110, 1189-1203 (2012)

[9] O Lobanova, C Avendaño, T Lafitte, EA Müller, G Jackson, Molecular Physics 113, 1228-1249 (2015)

[10] EA Müller, G Jackson, Annual Review of Chemical and Biomolecular Engineering 5, 405-427 (2014)

[11] www.bottledsaft.org

Extended Abstract: File Not Uploaded