(615a) Prediction of the Interfacial Tension of Sugar-Based Surfactants through Molecular Modelling

Sugar-based surfactants (SBS) are amphiphiles with potential applications in various industrial settings, including enhanced oil recovery, pharmaceutical products, and personal and household care. The large industrial-scale production of these “green” surfactants has become possible due to the availability of feedstocks synthesised from biomass through biorefineries ^[1]. As opposed to the more common petrochemical surfactants (e.g., linear alkylbenzene sulfonates, fatty alcohol ethoxylates), SBSs are carbon-neutral, non-toxic, potentially cost-effective, and can be suitably tailored by modifying their morphology ^[2].

A typical SBS is composed of a polar head derived from glucose or maltose, a linker and a hydrocarbon (alkane) tail (see Figure 1). There is, however, a large and diverse chemical ‘space’ to be explored if one is to design surfactants for specific applications, as small differences in the morphology of these components (number and length of tails, morphology of head groups, substituents, linkers etc.) will affect the surfactant properties ^[3]. Experimental screening of candidate surfactants is costly and slow due to the combinatorial nature of the problem and the prior need to synthesise and purify prototypes chemically. In silico calculations employing molecular dynamics simulations can rapidly assess interfacial properties over a wide range of operating conditions. This work presents a novel and computationally efficient methodology that allows the quantitative prediction of the interfacial tension/concentration isotherm. The procedure is based on combining small scale simulations of non-saturated water/air interface with free energy calculations that relate the surfactant surface density with the bulk concentration.

Materials and Methods

A molecular modelling approach is presented exemplified by predicting the surface tension of non-ionic alkylpolyglucoside (APG₁₂) on the water/air interface. Canonical all-atom Molecular Dynamics (MD) simulations are performed, including both phases and the interfacial region with different levels of surface coverage up to the critical micelle concentration. The simulations consist of roughly 12000 molecules of water and a variable amount of surfactant molecules ranging from 25 to 170 APG₁₂ molecules. The air is represented by a vacuum in which specific water molecules can escape from the liquid phase at random. Each state point is run for a minimum of 12 ns at 298.15 K with a time-step of 1 fs using a modified version of the pcff+ force field.

Adsorption data obtained from the MD simulations, relating the interfacial tension, γ, (or surface pressure, Π) with surfactant surface coverage, Γ, are used to build a two-dimensional (2D) simulation-based equation of state. The equation of state is used to derive a model for the surfactant excess chemical potentials at the interface. The adsorption isotherm of APG₁₂ (the γ – bulk concentration, c, curve) is determined from the derivation of a thermodynamic relationship that incorporates a free energy transfer. An enhanced sampling method is used to compute the free energy change associated with transferring a single surfactant molecule from the bulk solution phase to the interface (the Henry’s law constant), which is one of the inputs to the model.

Results and Discussion

Figure 1 shows the predicted and the experimental interfacial tensions as a function of the bulk solution surfactant concentration, c. While experimental data (including ours) show some appreciable degree of uncertainty, it is seen that the simulation predictions are quantitatively correct. Both the shape of the adsorption isotherm and the critical micelle concentration are well described. This is remarkable if one notes that the pcff+ force fields is parameterised to match bulk properties of organic moieties and, as such, was not specifically designed to reproduce interfacial properties accurately.

Significance

A framework to predict adsorption isotherms of non-ionic surfactant solutions is presented. This framework is based solely on the chemical structures of the surfactant molecules, and devoid of any empirical fitted parameters. In the future, the role of MD simulations in facilitating the surfactant development process is likely to grow substantially with the increasing computer power and advancements in the development of force fields and enhanced MD methodologies.

References

[1] A. Demirbas Biorefineries, for biomass upgrading facilities. London: Springer; 2010.

[2] N. Negm, and S. Tawfik, S. Environmentally Friendly Surface-Active Agents and Their Applications. Surfactants in Tribology, Volume 3, 147–194 (2013).

[3] G. Théophile G. et al. Adv. Colloid Interf. Sci., 270, 87 – 100 (2019).

[4] S. Kozo, Y. Tokio, H. Ryohei. Bull. Chem. Soc. Jpn 34: 237-241 (1961).

[5] S. Matsumura, K. Imai, S. Yoshikawa, K. Kawada, T. Uchibori. J Am Oil Chem Soc, 67: 996-1001 (1996).

[6] J. Krawczyk. Colloids Surf., A, 533, 61-67 (2017).

[7] P. Wydro. J. Colloid Interface Sci., 316, 107-113 (2007).