Physics of supercooled water and amorphous ices

In this thesis, I perform molecular dynamics simulations to investigate water in two metastable states: supercooled liquid and amorphous ice. Supercooled water is liquid water at temperatures below the freezing point. Amorphous ice is glassy water, where molecules are arranged in a disordered structure similar to the one found in liquids.; In Part I, I give a brief introduction to the concepts needed to understand the main results of this thesis.; In Part II, I investigate the microscopic dynamics in supercooled water. I describe the dynamics using the potential energy landscape (PEL) approach, and using an alternative approach proposed by Adam and Gibbs (AG). In agreement with the AG theory, I find that dynamics in water is heterogeneous. The AG theory is based on the vague concept of "cooperatively rearranging regions" (CRR). From my simulations I give a precise definition of the CRR and study the CRR properties with temperature. I relate the presence of CRR to the dynamics of the system as described in the PEL approach.; Cooling a liquid fast enough below the glass transition temperature Tg produces a glass. In Part III, I study the glass transition in water. The effect of cooling/heating rates and aging are shown to be crucial in the determination of Tg. Our simulations support the view that the glass transition in water is not experimentally accessible. The PEL properties sampled during the glass transition indicate that the common view of the glass as a "frozen" liquid is not always correct.; Amorphous ice can exist in at least two different forms, low-density amorphous (LDA) and high-density amorphous (HDA). In Part IV, I show that the LDA-HDA transition can be found in computer simulations and describe it using the PEL approach. I also explore the entire phase diagram of amorphous ice not accessible to experiments. I find that the recently suggested third glass found experimentally in water, very-high density amorphous (VHDA) ice, is not a new glass but a result of aging effects. Furthermore, I find that a continuum of amorphous ices can be identified in glassy water.