No Arabic abstract
We present a series of experiments investigating the growth of ice crystals from water vapor in the presence of a background gas. We measured growth dynamics at temperatures ranging from -2 C to -25 C, at supersaturations between 0.5 and 30 percent, and with background gases of nitrogen, argon, and air at a pressure of one bar. We compared our data with numerical models of diffusion-limited growth based on cellular automata to extract surface growth parameters at different temperatures and supersaturations. These data represent a first step toward obtaining precision ice growth measurements as a function of temperature, supersaturation, background gas pressure and gas constituents. From these investigations we hope to better understand the surface molecular dynamics that determine crystal growth rates and growth morphologies.
We present the design of a general-purpose convection chamber that produces a stable environment for studying the growth of ice crystals from water vapor in the presence of a background gas. Crystals grow in free fall inside the chamber, where the temperature and supersaturation are well characterized and surprisingly uniform. As crystals fall and land on a substrate, their dimensions are measured using direct imaging and broad-band interferometry. We also present a parameterized model of the supersaturation inside the chamber that is based on differential hygrometer measurements. Using this chamber, we are able to observe the growth and morphology of ice crystals over a broad range of conditions, as a function of temperature, supersaturation, gas constituents, gas pressure, growth time, and other parameters.
I examine the molecular dynamics of ice growth from water vapor, focusing on how the attachment kinetics can be augmented by edge-dependent surface diffusion. Although there are significant uncertainties in developing an accurate physical model of this process, it is possible to make some reasonable estimates of surface diffusion rates and admolecule density enhancements, derived from our basic understanding of ice-crystal growth processes. A quantitative model suggests that edge-dependent surface diffusion could substantially enhance terrace nucleation on narrow faceted surfaces, especially at the onset of surface premelting. This result supports our hypothesized mechanism for structure-dependent attachment kinetics, which readily explains the changes in snow crystal growth morphology with temperature depicted in the well-known Nakaya diagram. Many of the model features described here may be amenable to further quantitative investigation using existing computational models of the molecular structure and dynamics of the ice surface.
I examine a variety of snow crystal growth measurements taken at a temperature of -5 C, as a function of supersaturation, background gas pressure, and crystal morphology. Both plate-like and columnar prismatic forms are observed under different conditions at this temperature, along with a diverse collection of complex dendritic structures. The observations can all be reasonably understood using a single comprehensive physical model for the basal and prism attachment kinetics, together with particle diffusion of water vapor through the surrounding medium and other well-understood physical processes. A critical model feature is structure-dependent attachment kinetics (SDAK), for which the molecular attachment kinetics on a faceted surface depend strongly on the nearby mesoscopic structure of the crystal.
I examine a variety snow crystal growth experiments performed at temperatures near -2 C, as a function of supersaturation, background gas pressure, and crystal morphology. Although the different experimental data were obtained using quite diverse experimental techniques, the resulting measurements can all be reasonably understood using a single comprehensive physical model for the basal and prism attachment kinetics, together with particle diffusion of water vapor through the surrounding medium and other well-understood physical processes. As with the previous paper in this series, comparing and reconciling different data sets at a single temperature yields significant insights into the underlying physical processes that govern snow crystal growth dynamics.
I describe a new approach to the classification of snow crystal morphologies that focuses on the most common growth behaviors that appear in normal air under conditions of constant applied temperature and water-vapor supersaturation. The resulting morphological structures are generally robust with respect to small environmental changes and thus should be especially amenable to computational modeling. Because spontaneous structure formation depends on initial conditions, the choice of seed crystal can be an important consideration, and I have found that slender c-axis ice needles provide an exceptionally good starting point for this series of investigations. A sharp needle tip exposes a single basal surface that often simplifies subsequent morphological development, and the absence of a nearby substrate allows for the exploration of a broad range of supersaturations with well-controlled boundary conditions. The overarching goal of this endeavor is to facilitate detailed quantitative comparisons between laboratory ice-growth experiments and corresponding computational models, which will should greatly improve our understanding of the ice/vapor molecular attachment kinetics as well as our ability to model diffusion-limited growth dynamics in the ice/vapor system. This specific case-study of water ice connects broadly to many areas in aqueous chemistry, cryobiology, and environmental science, while the physical principles of molecular attachment kinetics and diffusion-limited growth apply more generally to other systems in crystal growth and materials science.