An Improved Apparatus For Measuring the Growth of Ice Crystals from Water Vapor

We describe an apparatus designed for obtaining precise measurements of the growth rates of ice crystals from water vapor over a range of experimental conditions. Our aim is to produce clean, high-quality test crystals in a well controlled environment for investigating the detailed molecular dynamics that controls the basic physics of ice crystal growth. In this paper we describe the nucleation and initial growth of test crystals, their transport and selection into a experimental chamber, the creation of a stable and controllable supersaturation, hardware and calibration issues, and the crystal measurement via direct imaging and broad-band interferometry. [Note: The figures in this paper are presented at reduced resolution to facilitate rapid downloading. The paper is available with higher quality figures at http://www.its.caltech.edu/atomic/publist /kglpub.htm, or by contacting the author.] 1 Investigating the Physics of Ice Crystal Growth The goal of our ice growth experiments is to observe the growth of individual ice crystals in a carefully controlled environment, and an idealized schematic diagram of our experimental set-up is shown in Figure 1. The top surface of the chamber is a thermal conductor with a uniform temperature TIR, and its inside surface is covered with a layer of ice crystals that make up the ice reservoir. At the beginning of each measurement a single test crystal is positioned near the center of the bottom substrate surface held at temperature Tsubst, separated from the ice reservoir by thermally insulating side walls with a vertical spacing of 1.0 mm. The temperature difference ΔT = TIR−Tsubst determines the effective supersaturation seen by the test crystal. After placing a test crystal we then increase ΔT and observe its size and thickness as a function of time, and from this extract growth velocities under various conditions. Our ultimate goal is to understand the crystal growth dynamics as a function of temperature, supersaturation, crystal morphology and history, chemical make-up of the substrate, and the pressure and chemical make-up of the background gas in the experimental chamber. The means by which we produce these experimental conditions are described in detail in this paper. As experience has shown repeatedly in investigations of ice crystal growth, variations with regard to initial crystal nucleation, growth history, and minute chemical impurities are inevitable in any experimental system. As a result each individual crystal grows somewhat differently, making it difficult to obtain experimental results that describe theoretically perfect ice crystals. A good fraction of this paper is therefore devoted to examining the absolute precision and reproducibility in our measurements under a variety of conditions, and we have also strived to produce quantitative measures of all processes that significantly affect ice growth rates.