Fundamental Measurements Using Microfluidics: The Rate of Ice Nucleation in Supercooled Water

We developed a microfluidic experiment that studies the freezing of supercooled water. Supercooled water, still liquid at temperatures below the freezing point of water, exists in large amounts on Earth - drops of supercooled water are one of the main components of clouds. Supercooling is possible because freezing must be initiated by a phenomenon that requires free energy: ice nucleation, which is the creation of a small amount of ice in the bulk of liquid water. A better understanding of ice nucleation is relevant to climate studies and to many other branches of science, including the physics and chemistry of water, and the biology of life in cold climates. Ice nucleation is a stochastic phenomenon and its experimental investigation is difficult because it requires repeated measurements on supercooled water, which is a phase out of thermodynamic equilibrium. We built an apparatus that supercools and freezes tens of thousands of small drops of water (~100-micron diameter), and measures with high accuracy the temperature of each drop down to their freezing that occurs at temperatures below -35°C. This instrument can record the largest sets of individual freezing temperatures, has the fastest data acquisition rate, and the best optical and temporal resolutions among instruments designed to study nucleation of ice. I will present the development of this instrument, our measurements of the rate of homogenous ice nucleation in drops of pure water, and our recent experiments on the effect of electric fields on ice nucleation.