The anomalous properties of cold and supercooled water, such as the fact that at sufficiently low temperatures it becomes more compressible and less dense when cooled, and more fluid when compressed, have attracted the attention of physical scientists for a long time. The discovery in the 1970s that several thermodynamic and transport properties of supercooled water exhibit a pronounced temperature dependence and appear to diverge slightly below the homogeneous nucleation temperature inspired a large number of experimental and theoretical studies. Likewise, an important body of work on glassy water has been stimulated by experiments, starting in the mid-1980s and continuing to this date, which suggest that vitreous water can exist in at least two apparently distinct forms. A coherent theory of the thermodynamic and transport properties of supercooled and glassy water does not yet exist. Nevertheless, significant progress towards this goal has been made during the past 20 years. This article summarizes the known experimental facts and reviews critically theoretical and computational work aimed at interpreting the observations and providing a unified viewpoint on cold, non-crystalline, metastable states of water.

https://iopscience.iop.org/article/10.1088/0953-8984/15/45/R01/pdf

Supercooled and glassy water

In a review of the present title, the first requirement is to ensure that the title words are understood. While the term “glass” is broadly familiar, and the origin of the “glass transition” in terms of the crossing of system and experimental time scales is generally agreed upon, there are at least three different definitions of the material property “glass transition temperature” (Tg) in current use.1,2 Furthermore, these may differ from each other by as much as 50 K in certain higher Tg cases. The difference is a consequence of the magnitude of the “glass transformation range” within which the Tg is variously defined. This magnitude, the “width of the glass transition”, can vary greatly from system to system, for reasons discussed below. Since the quantity Tg will recur frequently in this review, it is necessary to deal adequately with the definition problem, and this will be done by reference to Figures 1 and 2 below. Depending on the liquid in question, glass transitions may be observed occurring over an enormous range in temperature, from below 50 K to above 1500 K. The reason for this range is clearly to do with the strength of the interparticle interactions, i.e., the “bonds” that are being broken as the particles rearrange. However, the reason that some glass transitions are “sharp” (meaning narrow glass transformation range, or “transition width”) and others very spread out in temperature is not so clear. It is largely to do with the “fragility” of the glassformer, but may C. A. Angell was born in Canberra, Australia, and studied chemistry and metallurgy at the University of Melbourne. After working on molten salts with J. O’M Bockris at the University of Pennsylvania for two years, he became the Stanley Elmore Fellow at Imperial College of Science, London, where he completed his Ph.D. under the direction of John W. Tomlinson. There he was awarded the Armstrong medal for graduate research 1959− 1961. He returned to Australia as lecturer in chemical metallurgy but after two years came back to the U.S. as a post-doc with Dieter Gruen at Argonne National Laboratory. In 1966, he joined Purdue University as Assistant Professor, becoming full Professor in 1971. In 1989, he moved to Arizona State University where he is now Regents’ Professor of Chemistry and Biochemistry. He has enjoyed and profited from sabbatical leaves at the University of Amsterdam, the Australian National University, Institute Laue-Langevin, Grenoble, the Ecole de Physique et Chemie Industrielle, Paris, University of Rennes-Beaulieu, Sydney University, and the University of Rome. His research interests range from rechargeable lithium batteries and fuel cells, through the phenomenology of the glass transition and the origin of fragility in liquids, to the anomalous properties of water and geochemical fluids and their relation to polyamorphism. Currently, he is focusing on the annealing behavior of hyperquenched liquids and solutions, particularly denatured protein solutions, with a sideline on nanoporous network glasses as gas storage media. 2627 Chem. Rev. 2002, 102, 2627−2650

Liquid fragility and the glass transition in water and aqueous solutions.

The vitrification of pure water is compared with that of molecular solutions rich in water, and gross differences are noted. Thermodynamic reasoning and direct observations on noncrystallizing nanoconfined water indicate that the glass transition in ambient-pressure water is qualitatively distinct from that found in the usual molecular liquids. It belongs instead to the order-disorder class of transition seen in molecular and ionic crystalline materials. The distinctive "folding funnel" energy landscape for this type of system explains the extreme weakness of the glass transition of water as well as the consequent confusion that has characterized its scientific history; it also explains the very small excess entropy at the glass transition temperature. The relation of confined water behavior to that of bulk is discussed, and the "fragile-to-strong" transition for supercooled water is interpreted by adding a "critical point-free" scenario to the two competing scenarios for understanding supercooled bulk water.

https://science.sciencemag.org/content/sci/319/5863/582.full.pdf

Insights into phases of liquid water from study of its unusual glass-forming properties.

Transition between the low‐ and high‐density amorphous H2O was observed with a piston‐cylinder apparatus from 77 K to 140 K, almost the glass‐transition temperature of the amorphs. The transition was found to occur reversibly and abruptly at ∼135 K and ∼0.2 GPa with a volume change of 0.20±∼0.01 cm3/g and some hysteresis. This confirmed the transition to be first order. Absence of intermediate amorphous states between the low‐density and high‐density amorphs suggested existence of two substantially different liquid phases near the glass‐transition temperature.

Reversible first‐order transition between two H2O amorphs at ∼0.2 GPa and ∼135 K

The glass transition temperature (Tg) in water is still uncertain, with conflicting values reported in the literature. As with other hyperquenched glasses, water exhibits a large relaxation exotherm on reheating at the normal rate of 10 kelvin (K) per minute. This release of heat indicates the transformation of a high enthalpy state to a lower one found in slow-cooled glasses. When the exotherm temperature is scaled by Tg, the good glass-formers show a common pattern. However, for hyperquenched water, when this analysis is performed using the commonly accepted Tg = 136 K, its behavior appears completely different, but this should not be the case because enthalpy relaxation is fundamental to the calorimetric glass transition. With Tg = 165 +/- 5 K, normal behavior is restored in comparison with other hyperquenched glasses and with the binary solution behavior of network-former systems (H2O, ZnCl2, or BeF2 plus a second component). This revised value has relevance to the understanding of water- biomolecule interactions.

https://science.sciencemag.org/content/sci/294/5550/2335.full.pdf

The glass transition of water, based on hyperquenching experiments.

The heat capacity of low-density amorphous ice, made by warming and annealing high-density amorphous ice, has been measured in the range 90-136 K. A glass transition was observed at 124 K for samples that were annealed in the range 124-130 K. The glass transition observed is presumably the same as that reported in the literature for amorphous ice prepared by vapor deposition or rapid quenching of water.

D. D. Klug

Papers

Heat capacity and glass transition behavior of amorphous ice