Trees, Queues and Alcohols

Doktorsavhandling, 2013

Hydrogen bonded (H-bonded) materials, such as water, alcohols, sugars, and even DNA, are extremely important for biology, as well as chemical industry. Alcohols are used as solvents in paints, in perfumes, as cleaners, anti-freezers, or as an alternative to petrol in combustion engines. Crucial in most of the applications are the effects the hydrogen bonds have on the physical properties of the liquid and its functionality. This thesis is concerned with the H-bonding structure and dynamics in some of the simplest H-bonding material: small molecule alcohols. To investigate the structure and dynamics of the H-bonded clusters we use a combination of exper- imental, computational, and theoretical methods. More specifically, a statistical model of the hydrogen bonded clusters is developed that describes the distri- bution of cluster sizes and their properties. The clusters that we find, have a tree-like topology, and a broad distribution of cluster sizes. The model properties are in good agreement with results from Monte Carlo simulations as well as EPSR simulations based on neutron diffraction data. The model is also shown to be compatible with spectroscopic IR- and Raman data. The dynamics of the clusters are captured in a model inspired by queuing theory, with monomers leaving and joining the clusters. The dipole correlation spectrum of the dynamic model explains the Debye peak seen in dielectric spec- tra, and also the different time scales measured by NMR and neutron scattering techniques.