English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Interface structures in complex oxides remain one of the active areas ofcondensed matter physics research, largely enabled by recent advances inscanning transmission electron microscopy (STEM). Yet the nature of the STEMcontrast in which the structure is projected along the given directionprecludes separation of possible structural models. Here, we utilize deepconvolutional neural networks (DCNN) trained on simulated 4D scanningtransmission electron microscopy (STEM) datasets to predict structuraldescriptors of interfaces. We focus on the widely studied interface betweenLaAlO3 and SrTiO3, using dynamical diffraction theory and leveraging highperformance computing to simulate thousands of possible 4D STEM datasets totrain the DCNN to learn properties of the underlying structures on which thesimulations are based. We validate the DCNN on simulated data and show that itis possible (with >95% accuracy) to identify a physically rough from achemically diffuse interface and achieve 85% accuracy in determination ofburied step positions within the interface. The method shown here is generaland can be applied for any inverse imaging problem where forward models arepresent.

Deep learning of interface structures from the 4D STEM data: cation intermixing vs. roughening