Study of Slow Dynamics in Supercooled Water by Molecular Dynamics and Quasi-Elastic Neutron Scattering

The slow dynamics of supercooled water is studied by modelling the spectrum of test particle fluctuations: intermediate scattering function (ISF). The theoretical models are compared with experimental measurements by quasi-elastic neutron scattering (QENS) and molecular dynamics (MD) simulation results. The dynamics of supercooled water can be decoupled into a product of translational and rotational dynamics. While the translational dynamics is modelled well by a product of a ISF representing short time in-cage vibrations and long time cage relaxation processes, the rotational dynamics is an aspect we shall study in this thesis. We introduce a model for the first, second and third order rotational correlation functions, which are required for the computation of the rotational intermediate scattering function corresponding to the incoherent QENS spectra from supercooled water. The model is tested against MD data generated from an extended-simple-point-charge (SPC/E) model of water and is found to be satisfactory. The analysis can be used as a practical method for extracting rotational relaxation parameters from QENS spectra measured at large Q from supercooled bulk water or interfacial water in porous materials. By confining water in nano-pores of silica glass, we can bypass the crystallization and study the pressure effect on the dynamical behavior in deeply supercooled state using incoherent QENS technique. we investigated the dynamics of water confined in nanoporous alumino-silicate matrices, from ambient temperature to the deeply supercooled state. We collected data on three instruments, with widely different resolutions, the Fermi chopper (FCS), the disk chopper (DCS) and the backscattering (HFBS) spectrometers at the NIST Center for Neutron Research (NIST NCNR). The confining systems were lab synthesized nanoporous glasses MCM-41-S and MCM-48-S. Inside the pores of these matrices the freezing process of water is retarded so that freezing occurs at a temperature about 50 K less than in bulk water. Thus, with the combined use of different instruments and nano-pore confinement, we were able to study the system over a wide temperature range, 160-325 K, with different energy resolutions. The data from all the three 3 spectrometers were analyzed using a single consistent expression based on the relaxing cage model (RCM) for the translational and rotational dynamics. A remarkable slowing down of the translational and rotational relaxation times has been observed. The behavior of shear viscosity η or equivalently the structural relaxation time τ of a supercooled liquid as it approaches the glass transition temperature is called ‘fragile’ if it exhibits non-Arrhenius character, such as that described by the VogelFulcher-Tammann (VFT) law; otherwise, with η and τ obeying Arrhenius law, it is called ‘strong’. The fragile behavior is typical to ionic and van der Waals systems. In contrast, a liquid being strong reflects that its structural makeup, to a large extent by strong (commonly covalent) bonds forming a network structure. Bulk water is considered as a fragile liquid at room temperature but for supercooled water, a ‘fragileto-strong’ (F-S) transition at around 228 K has been proposed to occur at around 228 K, based on a thermodynamic argument. The F-S transition in a molecular liquid like water may be interpreted as a variant of kinetic glass transition predicted by the ideal Mode-Coupling Theory (MCT), where the real structural arrest transition is avoided by an activated hoping mechanism below the transition. Experiments at DCS and HFBS for water in MCM-41-S with 14 A pore diameter were also done under selective pressures, from ambient to 2400 bar. We observe a clear evidence of a cusp-like F-S dynamic transition at pressures lower than 1600 bar. Here we show that the transition temperature decreases steadily with an increasing pressure, until it intersects the homogenous nucleation temperature line of bulk water at a pressure of 1600 bar. Above this pressure, it is no longer possible to discern the characteristic feature of the F-S transition. The discussion part of this thesis concludes that the high-temperature liquid corresponds to the high-density liquid (HDL) of which the hydrogen bond network is not fully developed to conform a locally tetrahedral coordination, while the lowtemperature liquid corresponds to the low-density liquid (LDL) of which the more open, locally 4-coordinated, ice-like hydrogen bond network is fully developed. Identification of the end point of F-S transition with a possible second critical point is also discussed. Thesis Supervisor: Sow-Hsin Chen Title: Professor of Nuclear Science and Engineering