Structure-mining: an automated tool to find candidate structures
What is it about?
Though powerful for understanding structure of complex materials, structural modeling of advanced x-ray or neutron scattering data such as pair distribution function (PDF) analysis is difficult and presents a steep learning curve for new users. There are two major challenges. The first is that PDF structure refinement requires a satisfactory plausible starting model to achieve a successful result. The second is that the refinement process is complex and requires significant user inputs to guide it to the best fit whilst avoiding overfitting. This study presents a new approach, called structure-mining, to address both issues. It pulls from structural databases all the known structures meeting the experimenter's search criteria and automatically performs structure refinements on them without human intervention. Tests on various material systems show the effectiveness and robustness of the algorithm in finding the correct atomic crystal structure. It works on crystalline and nanocrystalline materials including complex oxide nanoparticles and nanowires, low-symmetry and locally distorted structures, and complicated doped and magnetic materials.
Why is it important?
Next generation technologies for sustainable energy, environmental remediation and health will rely on the discovery of novel, high performing materials, and the determination of their structures is a critical but difficult step. This development will speed up discovery of the structure of novel samples and help in the development of these new technologies.