First principles methods for the prediction of the structures and chemistry of materials have delivered a powerful tool for computational exploration. While early studies focused on the exotic properties of relatively simple systems, typically the elements and binary compounds, much of the matter in the Universe is likely to be found in more complex mixtures.¹ At the same time, the promise of discovering materials with extreme properties relies on the ability of screen a wide variety of compounds.[2,3] I will reflect on why ab initio random structure searching (AIRSS) is particularly suited to these challenges, and the importance of visualising⁴ and exploiting⁵ the vast datasets we are now generating.

[1] Conway, Lewis J., Chris J. Pickard, and Andreas Hermann. “Rules of formation of H-C-N-O compounds at high pressure and the fates of planetary ices.” Proceedings of the National Academy of Sciences 118, no. 19 (2021).
[2] Lu, Ziheng, Bonan Zhu, Benjamin WB Shires, David O. Scanlon, and Chris J. Pickard. “Ab initio random structure searching for battery cathode materials.” The Journal of Chemical Physics 154, no. 17: 174111 (2021).
[3] Shipley, Alice M., Michael J. Hutcheon, Richard J. Needs, and Chris J. Pickard. “High- throughput discovery of high-temperature conventional superconductors.” Phys. Rev. B 104, 054501 (2021).
[4] Shires, Ben W.B., and Chris J. Pickard. “Visualising energy landscapes through manifold learning”, Phys. Rev. X 11, 041026 (2021)
[5] Chris J. Pickard, “Ephemeral data derived potentials for random structure search”, Phys. Rev. B 106, 014102 (2022)