Silicate and organosilicate molecules are extremely important in a wide variety of natural and industrial processes, including geochemistry, biosilicification, bone tissue growth, glass and ceramics production, polymer manufacture (e.g., PDMS), and synthesis of nanoporous materials. Although silica is normally encountered in the form of crystalline solids (e.g, quartz, zeolites), the processes mentioned above almost always involve (organo)silicates in the liquid state or in aqueous solution. Moreover, enhancing our understanding of those processes necessarily involves delving into the molecular-level structures and interactions involving silicates and other components of the system. With the exception of a few studies devoted to specific molecules (e.g., silicic acid) [1,2], there is no general molecular model that can be applied to organosilicates – in fact, the silicon atom is conspicuously absent from most general liquid-state force fields, such as OPLS [3].
In this work, we focus on parameterising a new transferable atomistic force field for organosilicate molecules suitable for the liquid state and aqueous solutions of these molecules. The basis for our new force field is the United Atom (UA) form of the Transferable Potentials for Phase Equilibria (TraPPE) force field [4], and we extend it to include parameters for silicon and adjacent oxygen, hydrogen and carbon atoms. We have used quantum chemistry (QM) calculations to calibrate parameters for the torsional potentials. The results of QM were also fitted to obtain point charges on the molecules of interest. With these parameters in place, we ran molecular dynamics (MD) simulations and fitted the Lennard-Jones parameters to match experimental data for density and heat of vaporisation of selected organosilicate molecules. The parameters that we have obtained from this first approach are reasonable but show some systematic discrepancies for the heats of vaporisation, which we hypothesise is due to neglecting polarisation effects. As such, special attention has been paid to producing electrostatic charges which correspond accurately to the liquid- or solution-like environment of the molecules, thus hoping to capture polarisation effects in an implicit way within a non-polarisable model. The parameters of the new model are validated against a range of thermodynamic properties, including free energies of solvation. We expect this new model to find applicability in processes such as those mentioned above, which rely on interactions involving silicate molecules in solution.
[1] J.C.G. Pereira, C. R. A. Catlow, G. D. Price; J. Phys. Chem. A., 2002, 106, 130-148
[2] Miguel Jorge, José R. B. Gomes, M. Natália D. S. Cordeiro, Nigel A. Seaton; J. Phys. Chem. B.; 2009, 113, 708-718
[3] W. L. Jorgensen, D.S. Maxwell, J. Tirado-Rives; 1996, J. Am. Chem. Soc. 118, 11225-11236
[4] Marcus G. Martin, and J. Ilja Siepmann; J. Phys. Chem. B, 1998, 102, 2569-2577
