​Discovering new materials is a challenging process, as there boundless potential combinations of atoms, a great many of those combinations will not result in materials that are stable, and which can be synthesised. The researchers' computer-guided strategy identifies regions of compositional space likely to yield new materials. Their algorithm uses knowledge of crystal chemistry to suggest probe structures which are used to approximate the composition energy landscape and identify promising regions of chemical composition. Their calculations led them to the experimental discovery of two new materials, one of the structures was determined using neutron powder diffraction on HRPD, together with X-ray powder and electron diffraction to confirm the crystallographic symmetry. The structure obtained is a triclinic, 64-fold super structure of perovskite containing 102 atomic sites, without the high resolution data the structure would have been insoluble due to the high degree of pseudo symmetry. This combined approach will accelerate the systematic discovery of new materials for many applications, including energy generation and storage.

Related publication: C. Collins et al. “Accelerated discovery of two crystal structure types in a complex inorganic phase field" Nature 546(2017), 280-284, DOI: 10.1038/nature22374​