Disruption of Hydrogen-Bonding Network Eliminates Water Anomalies

Apr 11, 2017 - We show that lithium chloride prevents hydrogen-bonding network completion in water on cooling, as manifested, in particular, in changi...
2 downloads 0 Views 2MB Size
Article pubs.acs.org/JPCB

Disruption of Hydrogen-Bonding Network Eliminates Water Anomalies Normally Observed on Cooling to Its Calorimetric Glass Transition Jose M. Borreguero† and Eugene Mamontov*,‡ †

Neutron Data Analysis and Visualization Division, Neutron Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States ‡ Chemical and Engineering Materials Division, Neutron Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831, United States ABSTRACT: The calorimetric glass-transition temperature of water is 136 K, but extrapolation of thermodynamic and relaxation properties of water from ambient temperature to below its homogeneous nucleation temperature TH = 235 K predicts divergence at TS = 228 K. The “noman’s land” between the TH and glassy water crystallization temperature of 150 K, which is encountered on warming up from the vitrified state, precludes a straightforward reconciliation of the two incompatible temperature dependences of water properties, above 235 K and below 150 K. The addition of lithium chloride to water allows bypassing both TH and TS on cooling, resulting in the dynamics with no features except the calorimetric glass transition, still at 136 K. We show that lithium chloride prevents hydrogen-bonding network completion in water on cooling, as manifested, in particular, in changing microscopic diffusion mechanism of the water molecules. Thus thermodynamic and relaxation peculiarities exhibited by pure water on cooling to its glass transition, such as the existence of the TH and TS, must be associated specifically with the hydrogen-bonding network. counterparts.16 To avoid these complications, we have advocated the use of bulk water−lithium chloride mixtures to study low-temperature relaxations in aqueous systems.17−21 Pioneering studies carried out almost five decades ago showed that, uniquely among salts, lithium chloride did not change the glass-transition temperature of water even at very high concentrations.22 These studies were performed on warming up of the vitrified aqueous solutions of lithium chloride, with a maximum reported temperature (for very high concentrations of LiCl) below 163 K. Nevertheless, similarities between lithium chloride aqueous solutions and pure water have been also reported at higher temperatures, similar to those probed in the present study; for example, the temperature of the apparent dynamic crossover in confined water of ca. 225 K remains the same in a bulk solution of lithium chloride.23 For the present study, we choose a slightly off-eutectic composition of (H2O)6(LiCl) for its superior glass-forming properties and nanoscopic homogeneity,24 which may not be the case for other compositions.24−26 The temperature dependence of the water molecules relaxations in aqueous solution of lithium chloride has been previously addressed in some detail.17−26 Here we explore a different question: Does the addition of lithium chloride induce a major change in the microscopic structural relaxation mechanism in water that could be associated with bypassing both TH and TS on cooling? That is, if some

1. INTRODUCTION On cooling, supercooled water either crystallizes at its homogeneous nucleation temperature of TH = 235 K or becomes glassy when extremely fast cooling rates, in excess of 105 K/s, are applied. By the standard definition of glass transition as the temperature when the system structural relaxation time is ∼100 s, the glass-transition temperature of water is ∼136 K, as suggested by calorimetric,1 diffusion,2 and relaxation3 experiments. A problem, however, arises if one extrapolates viscosity,4 diffusivity,5 compressibility,6 or relaxation time7 from the data measured above 235 K; such extrapolations suggest divergence at a temperature of TS = 228 K, much above 136 K. It is not possible to study glassy water on warming up above ca. 150 K because of crystallization, which precludes a straightforward experimental reconciliation of the two seemingly incompatible temperature dependences of water properties, one above 235 K and another below 150 K. A transition from the high-temperature super-Arrhenius to the low-temperature Arrhenius behavior, known as “fragile”-to“strong” transition, which would link the high- and lowtemperature regimes in water, has been suggested.8 However, its experimental verification would need to rely, with few notable exceptions,9 on measurements of strongly confined water (typically in pores of