Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 70: Topical Session (Symposium MM): Big Data in Materials Science - Managing and exploiting the raw material of the 21st century
MM 70.5: Vortrag
Freitag, 16. März 2018, 10:45–11:00, H 0107
Identifying synthesisable ice structures from first principles — •Edgar Engel1,2, Andrea Anelli1, Michele Ceriotti1, Chris Pickard2, and Richard Needs2 — 1TCM Group, Cavendish Laboratory, UoCambridge, UK — 2Laboratory of Computational Science and Modeling, IMX, EPFL, Lausanne, Switzerland
We present a comprehensive density-functional-theory study of the crystalline phases of water ice. We construct candidate ice structures on the basis of more than five million tetrahedral networks listed in the Treacy, Deem, and IZA databases, collecting 15,882 locally-stable ice structures. The search for the few synthesisable structures among them is a needle-in-a-haystack kind of problem, which is conventionally tackled using a convex hull construction to identify structures which are stabilised by manipulation of a particular constraint (such as density) chosen on the basis of experimental evidence or intuition. This heavily constrains which stabilisable structures are identified and does not account for the uncertainties inherent to computed structure properties. Hence, we instead employ a recently developed probabilistic generalised convex hull construction to stochastically sample the likelihood of each structure to be stabilised by application of appropriate thermodynamic constraints. We thereby recover (entirely a priori) all known ice phases except the known-to-be metastable ice IV. We further identify several new promising candidates for experimental synthesis, providing a much needed starting point for the determination of accurate structural properties and possible synthetic pathways.