Final Report Summary - SIDIS (Simulation of Dielectric Spectra)
Dielectric spectroscopy is a powerful experimental technique that allows for investigation into the dynamics of molecules in liquids by probe collective molecular relaxations times in a wide spectral range, from the picosecond scale up to several hours. The interpretation of dielectric spectra and in general dielectric permittivity measurement, however, is not always straightforward, due to the complex interaction between neighbouring molecules, and simple analytical models can provide often only a qualitative picture of the real molecular dynamics. Computer simulations, especially at atomistic detail, can on the other hand offer a very detailed insight into the microscopic dynamics of liquid.
With the SIDIS project we introduced a new approach based on the analysis of the contribution of collective electric currents, rather than dipole moments, to investigate the origin of a number of features of the dielectric spectra of system of relevance (water, simple salt aqueous solutions, ionic liquid/water mixtures, and glycerol).
The analysis of the spectra of nine different, popular models for water was performed with the aim of providing future reference and guidance in choosing the most appropriate one, from the point of view of their dielectric spectrum. While qualitatively all models were able to reproduce the main features of the spectrum, on the quantitative side the performance of the models were mixed, with some better suited to approximate the static permittivity and other the Debye relaxation frequency. Surprisingly, the more refined and computationally demanding polarisable models did not outperform some of the venerable water models. In the terahertz region we performed a detailed analysis of the spectrum to understand its complex feature, and discovered that the librational absorption peak's broadness is partially due to the contributions at different frequencies coming from groups of water molecules having different hydrogen bond coordination numbers. The results of this investigation have been published in J. Chem. Phys. A, 119 (2015), 1539.
Regarding ionic systems, we obtained important results about simple salt aqueous solutions and ionic liquid/water mixtures. In simple salt solutions, we showed that a subtle kinetic effect in the static permittivity predicted decades ago by Hubbard and Onsager can be indeed be seen simulations in the limit of very diluted solutions, and we developed a semi-phenomenological theory for finite concentrations (J. Chem. Phys., 140, (2014) 211101 ; Phys. Chem. Chem. Phys., 17, (2015) 130). In ionic liquids / water mixtures we have calculated the static permittivity and dielectric spectra, decomposed into translational and rotational contributions, at different volume fraction of ionic liquid (1-butyl-3-methylimidazolium tetrafluoroborate). We formulated a geometrical model to estimate the fraction of hydration water, and with this information we modeled the dependence of the static permittivity as a function of the ionic liquid content. The large volume of the ionic liquids prevents any pairing, the rotational depolarization with strong dipolar interactions between the ions and water, points towards a general picture which is closer to that of dipolar mixture rather than electrolyte solutions (J. Chem. Phys., 140, (2014) 204505).
The results obtained for glycerol are still preliminary due to a reschedule required after the project reviews and have not been yet published, therefore they are not discussed in this public summary.
With the SIDIS project we introduced a new approach based on the analysis of the contribution of collective electric currents, rather than dipole moments, to investigate the origin of a number of features of the dielectric spectra of system of relevance (water, simple salt aqueous solutions, ionic liquid/water mixtures, and glycerol).
The analysis of the spectra of nine different, popular models for water was performed with the aim of providing future reference and guidance in choosing the most appropriate one, from the point of view of their dielectric spectrum. While qualitatively all models were able to reproduce the main features of the spectrum, on the quantitative side the performance of the models were mixed, with some better suited to approximate the static permittivity and other the Debye relaxation frequency. Surprisingly, the more refined and computationally demanding polarisable models did not outperform some of the venerable water models. In the terahertz region we performed a detailed analysis of the spectrum to understand its complex feature, and discovered that the librational absorption peak's broadness is partially due to the contributions at different frequencies coming from groups of water molecules having different hydrogen bond coordination numbers. The results of this investigation have been published in J. Chem. Phys. A, 119 (2015), 1539.
Regarding ionic systems, we obtained important results about simple salt aqueous solutions and ionic liquid/water mixtures. In simple salt solutions, we showed that a subtle kinetic effect in the static permittivity predicted decades ago by Hubbard and Onsager can be indeed be seen simulations in the limit of very diluted solutions, and we developed a semi-phenomenological theory for finite concentrations (J. Chem. Phys., 140, (2014) 211101 ; Phys. Chem. Chem. Phys., 17, (2015) 130). In ionic liquids / water mixtures we have calculated the static permittivity and dielectric spectra, decomposed into translational and rotational contributions, at different volume fraction of ionic liquid (1-butyl-3-methylimidazolium tetrafluoroborate). We formulated a geometrical model to estimate the fraction of hydration water, and with this information we modeled the dependence of the static permittivity as a function of the ionic liquid content. The large volume of the ionic liquids prevents any pairing, the rotational depolarization with strong dipolar interactions between the ions and water, points towards a general picture which is closer to that of dipolar mixture rather than electrolyte solutions (J. Chem. Phys., 140, (2014) 204505).
The results obtained for glycerol are still preliminary due to a reschedule required after the project reviews and have not been yet published, therefore they are not discussed in this public summary.