Polarisation forces in molecular ionic liquids

Molecular ionic liquids are molten salts with a melting point below 373 K. Most of them consist of a imidazolium based cation and a weakly basic anion like triflate, dicyanoamide or bis(trifluoromethane)sulfonimide. Changing the cation or anion respectively, one can generate a large diversity of physical properties. Although remnant features of a charge ordered salt are visible, molecular ionic liquids show a translational and rotational dynamics which is slower but comparable to that of neutral molecular liquids. So far, the structure and single-particle dynamics of these systems was studied by simulations using pairwise additive forces. They do not consider the response of the molecular charge distribution to the local environment. This project tries to model the reorganisation of the charge distribution by atomic polarisation forces in three different ways: First, the fluctuating charge model changes the atomic partial charges as a function of the molecular neighbourhood. Second, the induced-point-dipole method emulates the same effect by an auxiliary set of induced mathematical dipoles. Third, Drude oscillators work with physical dipoles which are represented by a pair of opposite auxiliary charges. In a first project phase, we will evaluate the three models with respect to their efficiency, stability and reality in case of molecular ionic liquids. With the most appropriate model at hand we will interpret, analyse and decompose the experimental dielectric spectra for different cation-anion combinations. A special emphasis lies on the influence of the induced dipoles and their coupling with the translational and rovibrational components. The last project phase is dedicated to the interpretation of solvation dynamics of the model solute Coumarin as measured by the dynamics Stokes shift. Thereby, the mutual relation between dielectric properties and solvation phenomena will be investigated in detail. The parameter-free Voronoi tessellation will be used for the decomposition of various solvation properties into shell specific contributions. In this way, the range of the solute`s influence on the solvent ionic liquids can be rationalised.

Pairwise additive forces have a long-tradition in computational studies of soft matter and the scientific community has gained a lot of experience in handling and calibrating them. However, there is a drawback of this feasibility: The charge distribution of a molecularspecies is not specific to the respective local environment, but uniform. The easiest and computationally feasible to mimic this local adjustment of molecular charge distributions is the concept of molecular or atomic polarisability. Thereby the permanent charge distribution is augmented by additional charges or dipoles whose ''strength'' is a - usually linear - function of the local electric field exerted by all peers on a molecule or atom. Two variants have been developed using physical or mathematical dipoles, respectively. The physical dipoles are realized by a pair of opposite charges. Uniting one of them with the atomic permanent charge its partner charge is tethered by a spring. The elongation of the spring is a - usually linear function of the local electric field and linear coefficient is called ''polarisability''.While this ''Drude oscillator model'' stays within the point charge concept, hydrogen atoms cannot be treated because of technical problems. This problem can be circumvented by ''mathematical dipole'' vectors which can change length and orientation, but not translational position. This achievement comes at the cost of additional programming code for charge-dipole and dipole-dipole interactions, particularly for their long-range contributions.In molecular ionic liquids polarization forces are necessary to mitigate the strong and highly directional electrostatic forces stemming from permanent charges which may retard molecular motion by an order of magnitude. In fact, we have found that ''polarisable dipoles'' act as an ''inner solvent'' screening electrostatic interactions between charged molecular cations and anions an ionic liquid is composed of. In addition to the anisotropic shape forces they enable a more realistic representation of interactions in ionic liquid.While the initial phase of the project had a focus on methodic development and evaluation of polarisation models, these methods were subsequently applied to interpret dielectric spectra as well as to study the solvation of a chromophore. Altogether the promised working program was fulfilled completely. The pool of knowledge created in this way was further exploited in dual sense: On the one hand, the magnetic reaction field governing the nuclear Overhauser effect (NOE) in NMR was analysed. On the other hand, the hybrid nature of molecular ionic liquids being composed of a charged, polar head and a hydrophobic tail provided knowledge to start research with surfactants forming reverse micelles.