Background

 

Premelting of ice is traditionally defined as a reversible formation of a thin, liquid-like layer at the ice/vapor interface or at an interface with another solid at temperatures significantly below the bulk melting point¹. Such liquid-like or “quasi-liquid” layers (QLL) are the precursors to the complete melting of ice.  A QLL may be only a few molecules thick at the onset of interfacial melting; however, its thickness rapidly increases as the temperature approaches the bulk melting point. Over the past decades, QLL’s at ice interfaces were observed using a number of experimental techniques. Some of the data imply the onset of premelting at temperatures as low as –100 °C.  The onset and extent of interfacial premelting depends on the nature of the interface and is likely to demonstrate a complex dependence on concentration of ionic impurities¹.

 

Crystallization of an aqueous solution often produces polycrystalline ice with a significant fraction of the impurities situated at the grain boundaries. The impurity-rich boundaries between relatively pure ice crystallites, tri-grain junctions (veins), and vein intersections (nodes) represent another type of locations for premelting. Although there is little doubt that that, even at low concentrations, ionic impurities are capable of inducing premelting at grain boundaries and other localities in polycrystalline ice, the effect of other (molecular) chemical species on this type of phase transition remains mainly unexplored.

 

At moderate concentrations of molecular impurities, the segregation of the solute in polycrystalline ice is likely to result in dispersed monomers or small clusters of organic molecules and, thus, may not result in interfacial premelting in a traditional sense.  Nevertheless, taking into account the sensitivity of the hydrogen bond network to perturbations, the ice may still undergo premelting in the immediate vicinity of a guest molecule. Taking into account the significance of structure of aqueous phase for a variety of applied fields of science, we plan to explore molecular-scale premelting in the vicinity of organic guest species using a combination of FTDS and TOF mass spectrometry at temperatures near 0 °C. The immediate goal of these studies will be quantitative characterization of molecular scale surface tension forces in aqueous matrix surrounding particular organic species in polycrystalline ice.

 

 TOF-FTDS Experiments

 

We believe that detailed information on molecular structure and dynamics of the aqueous phase localized in the vicinity of impurities can be obtained by monitoring changes in the TOF distribution of non-polar organic species vaporizing from polycrystalline ice films during FTDS experiments at various temperatures. The central idea of these experiments is described below. If the guest molecules trapped in the condensed aqueous phase are non-polar, their average velocity after effusion from the film may be significantly higher than that of the surrounding H₂O molecules. For example, the average energies of carboxylic acid dimers evaporating from water at -20 °C exceed the temperature of the surface by 100-200 °C ².

 

The excess energy of non-polar desorption products correlates with the size of the molecules and the local surface tension of the surrounding aqueous phase² and can be explained in the following fashion: a large volatile hydrophobic molecule trapped in ice or water is essentially a single molecule gas bubble that bursts, when it is sufficiently close to the surface. The consequent relaxation of the aqueous film surface propels the contents of the bubble (a single non-polar molecule) into the gas phase with a velocity higher than thermal. The surface tension of water must differ from that of ice². Thus, by monitoring changes in the TOF distribution of the non-polar organic species escaping into the gas phase along the grain boundaries we should be able to gain insights into molecular-scale phase transitions localized in the vicinity of the impurity. 

 

In the past, we attempted measurements of TOF distributions of organic species evolving from polycrystalline ice near 0 °C. Although these earlier TOF-FTDS experiments seemed to produce results supporting the hyperthermal vaporization mechanism (see Fig. 1), we could not rule out possible perturbations to TOF distributions of relatively large organic species by collisions with co-desorbing water molecules. In other words, the filaments employed in our early TOF experiments were too large to ensure collision-free desorption of organic impurities.

 

Recently, we were able to use submicron Pt wire (approximately 700 nm in diameter, 4 mm long) as a substrate for aqueous film growth and vaporization in FTDS experiments. The 700 nm diameter of an ultrathin Pt filament (see Figure 2) is almost fifteen times smaller than the mean free path of H₂O molecules in saturated water vapor at 0 °C (vapor pressure is 4.7 Torr), which nearly eliminate the probability of collisions in our experimental arrangement³. In summary, by employing the submicrometer Pt filament as a substrate for growth of 100-300 nm thick polycrystalline doped ice films, we will be able to measure collision-free TOF distributions of volatile organic species. Note that a surface curvature of the filament (1/350 nm) is still sufficiently small to significantly affect the vaporization of ice. Also note that a few hundred nanometers (1000 monolayers) thick ice films are likely to retain the properties of bulk samples essential for the phenomena under investigation.   Combined with molecular dynamics simulation and other theoretical results the FTDS-TOF studies will make it possible to gain molecular-level view of structure, properties and interactions of ice with various gas species.

 

 

References:

 

2.  J. G. Dash, A. W. Rempel and J. S. Wettlaufer, “The physics of premelted ice and its geophysical consequences”, Rev. Mod. Phys. 78, 695, (2006) and references therein.

 

2. M. Faubel, T. Kisters, “Non-equilibrium evaporation of carboxylic acid dimmers”, Nature, 339, 527, (1989).

 

3. V. Sadtchenko, M. Brindza, M. Chonde, B. Palmore, R. Eom., “The Vaporization Rate of Ice at Temperatures near Its Melting Point”, J. Chem. Phys. 121, 11980, (2004).

Time-of-Flight and Fast Thermal Desorption Spectroscopy (TOF-FTDS) studies of organic species vaporization from polycrystalline ice

Figure 1. TOF distribution of benzene evaporating from a polycrystalline ice film on a 10 micrometers filament at -5 ^oC. The equilibrium TOF distribution is shown for comparison. The mole fraction of benzene in ice was approximately 0.05.

Vlad Sadtchenko, GW Chemistry