Cryosphere Nonequilibrium Science

We are aiming to reveal dynamics of various nonequilibrium phenomena found in a cryosphere. In particular, we are focusing on "phase transition of water", which holds the key to wide spectrum of phenomena, such as weather, environment-related issues, living animal and plant life, chemical evolution of the cosmos, etc.

After Prof. Ukichiro Nakaya succeeded in growing snow crystals artificially (1936), many excellent studies, such as studies on a quasi-liquid layer present at a water-ice interface, relations between morphology and growth conditions, growth suppression by antifreeze proteins, etc., have been carried out so far.

However surface morphology of ice/snow crystals has not yet been visualized at a molecular scale, although such micro-morphology (e.g. elementary steps) includes great insight into elementary processes of growth and melting/sublimation of ice and snow crystals. Hence we are developing various advanced optical microscopy and trying to study growth/melting/sublimation mechanisms of ice/snow crystals. We have obtained the following achievements.

Direct visualization of "elementary steps" at air-ice interfaces

Growing ends of molecular layers, so called "elementary steps", were so far observed on surfaces of various crystals other than ice (see Notice and Fig. 3). In the case of ice crystals, no one yet succeeded in visualizing elementary steps during growth. Hence mechanisms of phase transition and other many physical/chemical processes on ice crystal surfaces have been a matter of speculation.

To achieve breakthrough in such situation, we and Olympus Co. have further improved laser confocal microscopy combined with differential interference microscopy (LCM-DIM), which was originally developed by us and Olympus on 2004. Then we tried to observe surfaces of ice crystals grown from vapor (i.e. snow crystal surfaces) by improved LCM-DIM.

Figure 1 shows a typical example of the observation on a basal face of a Ih ice crystal. We could observe the appearance of many two-dimensional (2D) islands and the subsequent lateral growth of the 2D islands. When adjacent 2D islands coalesced, the contrast of steps of the islands always disappeared completely, as indicated by cross marks in Fig. 1. This disappearance of the step contrast demonstrates that the height of the adjacent molecular layers was the same. Since such disappearances of the step contrast could be always observed all over the crystal surfaces, we could conclude that the molecular layers visualized in this study were "elementary steps" (0.37 nm thickness), which has the minimum height determined from a crystal structure. We could also visualize elementary steps on a prism face (Fig. 2). Furthermore we also observed spiral growth of ice crystals. LCM-DIM, which was further improved in this study, enabled us to elucidate growth processes of ice crystals in the molecular level, for the first time.

We expect that such molecular-level in-situ visualization can be a promising means in the future to explore many long-standing questions of ice crystals, such as role of ice crystal surfaces in ozone degradation, the formation mechanisms of quasi-liquid layers, roughening transitions of steps and crystal surfaces, evaluation of the step ledge free energy, etc., in addition to the growth kinetics of elementary steps.

(In details: G. Sazaki, et al., Proc. Nat. Acad. Sci. USA., 107, 19702-19707 (2010): doi:10.1073/pnas.1008866107)

noneq01 noneq02
Figure 1. On a basal face of an ice crystal, many two-dimensional (2D) islands appeared, and grew in the lateral direction. Cross marks show the regions at which contrast between the coalesced steps disappeared. Images were taken at 0.57 s time intervals. (Movies [avi:4.8M, avi:86M]) Figure 2. The appearances and subsequent lateral growth of two-dimensional islands on a prism face of an ice crystal. Cross marks show the regions at which contrast between the coalesced steps disappeared. Images were taken at 0.99 s time intervals. (Movie [avi:66M])
(Notice) A crystal grows by one molecular layer by one molecular layer, like in the case of assembling bricks one by one. A molecular layer is grown in the lateral direction by the incorporation of a molecule (shown in red in Fig. 3) into a edge of a molecular layer (elementary step). Elementary steps, which become the bases of crystal growth, have the thickness determined from the size of a molecule and a crystal structure. noneq03

Quasi-liquid layers on ice crystal surfaces are made up of two different phases

Ice crystal surfaces melt even below the melting point (0°C), and then quasi-liquid layers (QLLs) is formed. It is generally acknowledged that QLLs formed by the surface melting play crucially important roles in the slipperiness of a skating rink, regelation, frost heave, recrystallization and coarsening of ice grains, morphological change of snow crystals, cryopreservation and electrification of thunderclouds. Hence the molecular-level understanding of QLLs holds the key to unlocking the secrets of those phenomena. Although surface melting was first proposed in the 1850s by Faraday, who pioneered the studies of electromagnetics, revealing of the dynamic behavior of QLLs has remained an experimental challenge.

Utilizing laser confocal microscopy combined with differential interference contrast microscopy (LCM-DIM), by which 0.37-nm-thick elementary steps on ice crystal surfaces can be visualized directly, we observed surfaces of ice crystals grown from water vapor. Then we found that QLLs on basal faces of ice crystals are made up of two different liquid phases (Fig. 4). Below -1.5°C, ice crystals just grew without any QLL. However, at temperatures higher than -1.5~-0.4°C, round liquid-like droplets (α-QLL) appeared (white arrowhead in Fig. 4). In addition, at temperatures higher than -1.0~-0.1°C, thin liquid-like layers (β-QLLs) appeared (red arrowhead in Fig. 4). The difference in the morphologies of α- and β-QLLs indicates that physical and chemical properties of these two QLLs are significantly different. The two QLLs moved around and coalesced each other on the ice crystal surfaces. So far, it was thought that only one QLL phase appears homogeneously on ice crystal surfaces. The novel picture of QLLs found in our study is quite different from the previous one. The insights into the nature of QLLs obtained from direct visualization may play a crucially important roles in understanding a variety of phenomena in which QLLs play a vital role, from the slipperiness of a skating rink to the electrification of thunderclouds.

Water and oil are not immiscible since they are composed of quite different molecules. However, immiscible two QLLs composed of same water molecules are quite interesting from fundamental scientific viewpoint.

(In details: G. Sazaki, et al., Proc. Nat. Acad. Sci. USA, 109, 1052-1055 (2012).)

Fig. 4. Two types of QLL phases formed on a basal face of an ice crystal. Photomicrographs (left) and schematic drawings (right). Although the appearing temperature of QLL phases exhibited slight variations, β-QLLs always appeared at a higher temperature than α-QLLs in the same run. (Movies [avi:1.3M])

Advanced optical microscopy

noneq05 noneq06
Figure 5. Elementary steps (2.8 nm in height) on a {101} face of a tetragonal lysozyme crystal. Arrows show the regions at which contrast between the coalesced steps disappeared. Many disappearances of the contrast of the coalesced steps indicate that the steps observed were elementary ones. (Movies [avi:36MB, mov:3MB, gif:16MB]) Figure 6. Diffusion of fluorescent-labeled lysozyme molecules at an interface between a solution and a lysozyme crystal. Individual diffusing molecules were observed by total internal reflection fluorescent microscopy of a thin layer type. One fluorescent spot corresponds to one fluorescent-labeled lysozyme molecule. (Movies [avi:16MB, mov:6MB, gif:7MB])

Themes under study

Contact address

Gen Sazaki, Professor, Concurrently studying with the Research Group for Phase Transition Dynamics of Ice
Research building 304, Phone & Fax +81-11-706-6880 (Office), 6889 (Exp. Room)
E-mail: (please replace "_a_" with @)

Page Top