Skip to main content

Towards a molecular-level understanding of heterogeneous ice nucleation

Final Report Summary - HETEROICE (Towards a molecular-level understanding of heterogeneous ice nucleation)

Ice formation is ubiquitous, affecting global issues like climate change and processes happening at the nanoscale, like intracellular freezing. Surprisingly, it is rather difficult to freeze pure water into ice. In fact, the formation of ice in nature happens almost exclusively heterogeneously, thanks to the presence of foreign substances such atmospheric mineral dust, which plays a role of great importance when it comes to ice formation in clouds.

One might think that, since the formation of ice is one of those most basic of everyday processes, surely it must have been studied to death and surely we must understand everything there is to know about it. Certainly it has been widely studied but to say that we fully understand the freezing of water could not be further from the truth. In fact, whilst a large body of excellent experimental work on ice formation exists, it has not yet been possible to simultaneously obtain the temporal and spatial resolution that would shed light on the molecular-level details of nucleation, that is the process responsible for the formation of ice at the molecular level.

The main objective of this project was simple and yet very ambitious: to use computer simulations to further our microscopic understanding of the heterogeneous formation of ice. Through the development and application of a range of state of the art computational techniques we achieved this aim. During the project we, for example:

i) Elucidated the role surface structure and symmetry plays in heterogeneous ice nucleation (J. Am. Chem. Soc. 137 , 13658 (2015))
ii) Identified failings in standard classical nucleation theory for heterogeneous ice nucleation (Nature Communications, 2017, 8 pp. 2257)
iii) Made a suggestion for the high ice nucleating ability of feldspar (Science, 2017, 355 (6323), pp. 367-371)
iv) Developed an improved set of 2-dimensional ice rules (J. Am. Chem. Soc. , 2017, 139 pp. 6403)
v) Understood the connection between structure and dynamics during ice nucleation (Proceedings of the National Academy of Sciences USA, 2019, 116 pp. 2009-2014)
vi) Developed improved computational techniques with high accuracy (Phys. Rev. B, 93, 241118(R) (2016); Proc. Natl. Acad. Sci. USA, 2018, 115 (8), pp. 1724-1729)

Overall our findings so far represent important steps within the community toward a reliable description of realistic heterogeneous ice nucleation scenarios, much closer to the experimental reality than has been achieved previously.