Scientists have been perplexed by water for a long time. For the last 30 years or so, they've thought that when water is cooled down to a very low temperature, it could be able to separate into two different densities. These phases don't mix and may help explain how water becomes less dense as it cools.
It's almost impossible to study this phenomenon in a lab because it's so fast. Machine learning models are being used by the Georgia Institute of Technology to better understand the phases of water. Computational evidence in support of water's liquid-liquid transition can be applied to real-world systems that use water.
"We are doing this with very detailed quantum chemistry calculations that are trying to be as close as possible to the actual physics and physical chemistry of water," said ThomasGartner, an assistant professor in the School of Chemical and Biomolecular Engineering at Georgia Tech. This is the first time anyone has been able to study this transition in detail.
The paper, " Liquid-liquid transition in water from first principles", was presented in the journal Physical Review Letters.
There is a simulation of water.
In order to understand how water interacts, the researchers ran simulations on the world's most powerful computers.
He said that if you had a powerful microscope, you could zoom in all the way down to the level of the individual molecule. This is what we are doing by making a movie.
The researchers compared the liquid structure at different water temperatures and pressures to see how the molecule moves. They ran some simulations for up to a year, collecting extensive data and fine tuning their methods.
It wouldn't have been possible a decade ago to run such simulations. The researchers used a machine learning method to figure out how water interacts with each other The model was able to perform the calculation much faster than traditional methods.
The simulations improved the accuracy of the predictions. The researchers tested their predictions with different types of simulations. Multiple simulations can give similar results.
There's not a lot of data that we can compare to because it's difficult to study. It's important that we try to do this using multiple different techniques because we are pushing the boundaries here.
Beyond water is something else.
Extremes that probably don't exist on Earth directly, but could be found in various water environments of the solar system, from the oceans of Europa to water in the center of comets, are some of the conditions the researchers tested. Researchers could use the findings to better explain and predict water's strange and complex physical chemistry, inform water's use in industrial processes, and develop better climate models.
The work is a lot more general. Water is a well-studied research area, but it could be expanded to other difficult-to-simulate materials.
He said that the question of whether water can undergo this phase transition has been a longstanding problem. There's a lot of room to move the field forward even though we don't know what the boundaries are.
There is more information about Liquid-liquid transition in water from first principles. There is a book titled "PhysRevLett.129.255702."
Journal information: Physical Review Letters
All Rights Reserved © 2024
