At what temperature between 136 and 165 K the glass transition of water occurs is still controversial, while the crystallization of water prevents the determination. To confine water in nanopores stabilizes its liquid state down to low temperatures. Heat capacities and enthalpy relaxation effects of the water confined within MCM-41 nanopores with diameters in the range 1.5-5.0 nm were measured in this work by using adiabatic calorimetry. No fusion of the confined water was detected up to 2.0 nm, part of the water exhibited fusion in 2.1 nm pores, and the whole internal water which excludes the molecules interacting with the pore-wall atoms crystallized within pores with diameter of 2.3 nm and above. A glass transition of the internal water occurred at a temperature T(g) = 160-165 K for pore diameters in the range 1.5-2.0 nm and at 205-210 K for diameters of 2.0 and 2.1 nm; thus, the T(g) jumped from 165 to 205 K at 2.0 nm. The jump is connected to the development of hydrogen-bond network to a more complete one as the diameter is increased, and is conjectured as caused by the increase in the number, from three to four, of hydrogen bonds formed by each molecule. These imply that the glass transition of bulk water occurs at 210 K, which is much higher than 136 or 165 K debated so far.

How low-temperature water develops the formation of strong hydrogen bonds with some network structure is still open to a question. Heat capacities of the water confined within silica MCM-41 nanopores with different diameters in the range 1.7-4.2 nm were measured by adiabatic calorimetry. They revealed a hump with its maximum at 233 and 240 K for ordinary and heavy water, respectively. The maximum temperatures were essentially independent of the pore diameter, whereas the maximum values increased only in proportion to the fraction of the internal water molecules within the pores. It was concluded that the manner in which the hydrogen-bond formation progresses in bulk water is essentially the same as that in nanopore water and that strong hydrogen bonds are formed on cooling by arranging the neighboring water molecules at tetrahedral positions but keeping their network structure irregular to make striking contrast with ice structure.

No abstract

Glass transition behaviors of dilute aqueous solutions are currently unclear because the water crystallizes immediately below the fusion temperatures to prevent the determination. The behaviors of methanol aqueous solutions [(CH(3)OH)(x)(H(2)O)(1 - x)] were studied here by confining the solutions within silica-gel pores and following the enthalpy relaxation associated with the glass transitions by adiabatic calorimetry. The dilution of the solutions in the composition range x < 0.3 brought both abrupt increase in the glass transition temperature T(g) as referred to the composition dependence expected from the behavior in x > 0.3 and appearance of a new glass transition at around 115 K. It was conjectured from the results that a hydrogen-bond network inherent to water starts to develop at around x = 0.3, and that molecules on the pore wall cannot join the network by forming tetrahedrally extended hydrogen-bonds so that they should constitute a mobile layer as an interfacial one. Such a special layer is understood as absent above x > 0.3, indicating that no network structure inherent to water is developed in the solutions.