A New Data-Driven Interacting-Defect Model Describing Nanoscopic Grain Boundary Compositions in Ceramics
Preprints are manuscripts made publicly available before they have been submitted for formal peer review and publication. They might contain new research findings or data. Preprints can be a draft or final version of an author's research but must not have been accepted for publication at the time of submission.
A new data-driven interacting-defect model has quantitatively described the nanoscopic com- position of high solute concentrations at grain boundaries in ion-conducting ceramics. The successful model is a data-driven Cahn-Hilliard methodology for interfaces and surfaces, introduced and demonstrated in this report. The model is applied to high spatial resolu- tion composition data gathered at grain boundaries in calcium-doped ceria. The statistical methodology for the data-driven procedure shows definitively that gradient terms are re- quired to quantitatively describe the local grain boundary composition data. The model additionally shows co-accumulation of negatively-charged acceptor dopants and positively- charged oxygen vacancies at the interface, which is qualitatively in accordance with atom probe tomography evidence in acceptor-doped ceria.