Nicolas Giovambattista

Glass-transition temperature of water: A simulation study

We report a computer simulation study of the glass transition for water using the extended simple point charge potential. To mimic the difference between standard and hyperquenched glass, we generate glassy configurations with different cooling rates, and we calculate the temperature dependence of the specific heat on heating. The absence of crystallization phenomena allows us, for properly annealed samples, to detect in the specific heat the simultaneous presence of a weak prepeak (shadow transition) and an intense glass transition peak at higher temperature. Our results support the view-point that the glass transition temperature is higher than the conventionally accepted value 136 K. We also compare our simulation results with the Tool-Narayanaswamy-Moynihan phenomenological model.

Models for a liquid–liquid phase transition

We use molecular dynamics simulations to study two- and three-dimensional models with the isotropic double-step potential which in addition to the hard core has a repulsive soft core of larger radius. Our results indicate that the presence of two characteristic repulsive distances (hard core and soft core) is sufficient to explain liquid anomalies and a liquid–liquid phase transition, but these two phenomena may occur independently. Thus liquid–liquid transitions may exist in systems like liquid metals, regardless of the presence of the density anomaly. For 2D, we propose a model with a specific set of hard core and soft core parameters, that qualitatively reproduces the phase diagram and anomalies of liquid water. We identify two solid phases: a square crystal (high density phase), and a triangular crystal (low density phase) and discuss the relation between the anomalies of liquid and the polymorphism of the solid. Similarly to real water, our 2D system may have the second critical point in the metastable liquid phase beyond the freezing line. In 3D, we find several sets of parameters for which two fluid–fluid phase transition lines exist: the first line between gas and liquid and the second line between high-density liquid (HDL) and low-density liquid (LDL). In all cases, the LDL phase shows no density anomaly in 3D. We relate the absence of the density anomaly with the positive slope of the LDL–HDL phase transition line.

Connection between Adam-Gibbs theory and spatially heterogeneous dynamics.

We investigate the spatially heterogeneous dynamics in the extended simple point charge model of water using molecular dynamics simulations. We relate the average mass n* of mobile particle clusters to the diffusion constant and the configurational entropy. Hence, n* can be interpreted as the mass of the cooperatively rearranging regions that form the basis of the Adam-Gibbs theory of the dynamics of supercooled liquids. We also examine the time and temperature dependence of these transient clusters.

Waterlike anomalies for core-softened models of fluids: Two-dimensional systems

We use molecular dynamics simulations in two dimensions to investigate the possibility that a core-softened potential can reproduce static and dynamic anomalies found experimentally in liquid water: (i) the increase in specific volume upon cooling, (ii) the increase in isothermal compressibility upon cooling, and (iii) the increase in the diffusion coefficient with pressure. We relate these anomalies to the shape of the potential. We obtain the phase diagram of the system and identify two solid phases: a square crystal (high density phase), and a triangular crystal (low density phase). We also discuss the relation between the anomalies observed and the polymorphism of the solid. Finally, we compare the phase diagram of our model system with experimental data, noting especially the line of temperatures of maximum density, line of pressures of maximum diffusion constant, and line of temperatures of minimum isothermal compressibility.

Relation between the high density phase and the very-high density phase of amorphous solid water.

It has been suggested that high-density amorphous (HDA) ice is a structurally arrested form of high-density liquid (HDL) water, while low-density amorphous ice is a structurally arrested form of low-density liquid (LDL) water. Recent experiments and simulations have been interpreted to support the possibility of a second distinct high-density structural state, named very high-density amorphous (VHDA) ice, questioning the LDL-HDL hypothesis. We test this interpretation using extensive computer simulations and find that VHDA is a more stable form of HDA and that, in fact, VHDA should be considered as the amorphous ice of the quenched HDL.

Applications of the Stell–Hemmer Potential to Understanding Second Critical Points in Real Systems

We consider the novel properties of the Stell–Hemmer core-softened potentials. First we explore how the theoretically predicted second critical point for these potentials is related to the occurrence of the experimentally observed solid–solid isostructural critical point. We then discuss how this class of potentials can generate anomalies analogous to those found experimentally in liquid water.

Phase diagram of amorphous solid water: low-density, high-density, and very-high-density amorphous ices.

We calculate the phase diagram of amorphous solid water by performing molecular dynamics simulations using the extended simple point charge (SPC/E) model. Our simulations follow different paths in the phase diagram: isothermal compression/decompression, isochoric cooling/heating, and isobaric cooling/heating. We are able to identify low-density amorphous (LDA), high-density amorphous (HDA), and very-high density amorphous (VHDA) ices. The density rho of these glasses at different pressure P and temperature T agree well with experimental values. We also study the radial distribution functions of glassy water. In agreement with experiments, we find that LDA, HDA, and VHDA are characterized by a tetrahedral hydrogen-bonded network and that, as compared to LDA, HDA has an extra interstitial molecule between the first and second shell. VHDA appears to have two such extra molecules. We obtain VHDA, as in experiment, by isobaric heating of HDA. We also find that other forms of glassy water can be obtained upon isobaric heating of LDA, as well as amorphous ices formed during the transformation of LDA to HDA. We argue that these other forms of amorphous ices, as well as VHDA, are not altogether new glasses but rather are the result of aging induced by heating. Samples of HDA and VHDA with different densities are recovered at normal P, showing that there is a continuum of glasses. Furthermore, the two ranges of densities of recovered HDA and recovered VHDA overlap at ambient P. Our simulations reproduce the experimental findings of HDA --> LDA and VHDA --> LDA transformations. We do not observe a VHDA --> HDA transformation, and our final phase diagram of glassy water together with equilibrium liquid data suggests that for the SPC/E model the VHDA --> HDA transformation cannot be observed with the present heating rates accessible in simulations. Finally, we discuss the consequences of our findings for the understanding of the transformation between the different amorphous ices and the two hypothesized phases of liquid water.

Transitions between Inherent Structures in Water

The energy landscape approach has been useful to help understand the dynamic properties of supercooled liquids and the connection between these properties and thermodynamics. The analysis in numerical models of the inherent structure (IS) trajectories-the set of local minima visited by the liquid-offers the possibility of filtering out the vibrational component of the motion of the system on the potential energy surface and thereby resolving the slow structural component more efficiently. Here we report an analysis of an IS trajectory for a widely studied water model, focusing on the changes in hydrogen bond connectivity that give rise to many ISs separated by relatively small energy barriers. We find that while the system travels through these ISs, the structure of the bond network is continuously modified, exchanging linear bonds for bifurcated bonds and usually reversing the exchange to return to nearly the same initial configuration. For the 216-molecule system we investigate, the time scale of these transitions is as small as the simulation time scale ( approximately 1 fs). Hence, for water, the transition between each of these ISs is relatively small and eventual relaxation of the system occurs only by many of these transitions. We find that during IS changes the molecules with the greatest displacements move in small clusters of 1-10 molecules with displacements of approximately 0.02-0.2 nm, not unlike simpler liquids. However, for water these clusters appear to be somewhat more branched than the linear stringlike clusters formed in a supercooled Lennard-Jones system found by Glotzer and her collaborators.

Potential-Energy Landscape Study of the Amorphous-Amorphous Transformation in H2O

We study the potential energy landscape explored during a compression-decompression cycle for the simple point charge extended model of water. During the cycle, the system changes from low density amorphous (LDA) ice to high density amorphous ice. After the cycle, the system does not return to the same region of the landscape, supporting the interesting possibility that more than one significantly different configuration corresponds to LDA. We find that the regions of the landscape explored during this transition have properties remarkably different from those explored in thermal equilibrium in the liquid phase.

Application of statistical physics to understand static and dynamic anomalies in liquid water

We present an overview of recent research applying ideas of statistical mechanics to try to better understand the statics and especially the dynamic puzzles regarding liquid water. We discuss recent molecular dynamics simulations using the Mahoney–Jorgensen transferable intermolecular potential with five points (TIP5P), which is closer to real water than previously-proposed classical pairwise additive potentials. Simulations of the TIP5P model for a wide range of deeply supercooled states, including both positive and negative pressures, reveal (i) the existence of a non-monotonic temperature of maximum density line and a non-reentrant spinodal, (ii) the presence of a low-temperature phase transition. The take-home message for the static aspects is that what seems to “matter” more than previously appreciated is local tetrahedral order, so that liquid water has features in common with SiO2 and P, as well as perhaps Si and C. To better understand dynamic aspects of water, we focus on the role of the number of diffusive directions in the potential energy landscape. What seems to “matter” most is not values of thermodynamic parameters such as temperature T and pressure P, but only the value of a parameter characterizing the potential energy landscape—just as near a critical point what matters is not the values of T and P but rather the values of the correlation length.