The study of snow and ice goes back centuries, but mysteries still remain as to how it forms under certain conditions.
With a better understanding of ice nucleation in clouds – that is, the moment when water molecules first cluster up to form ice crystals – we could develop better climate prediction tools as well as materials that both promote and prevent freezing.
To that end, Amir Haji-Akbari, assistant professor of chemical & environmental engineering, has conducted a study on what effects the air-water interface has on initiating freezing, with the findings published this week in the Proceedings of the National Academy of Sciences.
Clouds are composed of, among other things, liquid water microdroplets. When one of these droplets freezes, it immediately becomes more stable than all liquid droplets nearby, and its vapor pressure drops significantly. As a result, water vapor from neighboring liquid droplets deposits onto this frozen droplet, and makes it larger. Also, some of those liquid droplets occasionally collide with the frozen droplet and immediately freeze on contact. These processes can eventually make the frozen droplet large enough for gravity to pull it from the cloud. Almost all freezing events in the atmosphere involve foreign particles but ice can form with pure water, especially in high-altitude clouds. In either case, the molecular mechanism cannot be readily observed or understood with conventional experimental techniques.
To observe ice nucleation close to vapor-liquid interfaces, Haji-Akbari used a computer simulations using the TIP4/Ice model – considered one of the most accurate models of water molecules – to calculate ice nucleation rates in a water nanofilm and see how they vary in comparison to the nucleation rate in the bulk, or the main body, at the same thermodynamic conditions.
Computer models are crucial to understanding the mechanisms of ice formation, said Haji-Akbari, who joined the Yale faculty this semester.
“In real experimental systems, there’s no way to check this now, because these nucleation events are very small – around a few nanometers – and you usually have a narrow window of observability for these transitions to occur,” said Haji-Akbari. The simulations, he said, allow researchers to “sample and explore a wide range of time scales and observe trends of how these time scales are affected by different changes.”
For the study, the researchers observed a freestanding film of water with a thickness of four nanometers. Haji-Akbari said they expected that freezing events would begin closer to the surface of the film, at the water-liquid interfaces.
“What we found was rather surprising – it was just the opposite,” said Haji-Akbari. “They actually emerge not on the surface, but in the central region that has certain bulk-like features.”
This happens because the region away from the surface is better for the formation of what’s known as double-diamond cages (pictured at left), a particular molecular structure that serves as the building blocks of certain crystallites found in high-altitude clouds.
These structures develop at a faster rate than crystallites built from hexagonal cages (pictured at right), which are more stable and make up the kind of ice we’re most likely to see in everyday life. In other words, these 4-nm films do not have a conventional bulk-like region, but one that has some bulk characteristics but lacks others. This allows their behavior to be strongly impacted by vapor-liquid interfaces.
This information could contribute to a number of practical benefits.
“The long-term goal of this work is basically to provide better predictive tools for the climate,” he said. Also, by observing through computer models the processes that lead to crystallites built from double diamond cages, “we could think of strategies to manipulate these freezing processes.” For instance, he said, they could lead to ways of developing materials that inhibit freezing – a potentially valuable innovation to prevent ice from forming on the wings of aircrafts.
Source: Yale University