Single-Atom Level Determination of 3D Surface Atomic Structure via Neural Network-Assisted Atomic Electron Tomography

Functional properties of nanomaterials strongly depend on their surface atomic structures, but they often become largely different from their bulk structures, exhibiting surface reconstructions and relaxations However, most of the surface characterization methods are either limited to 2D measurements or not reaching to true 3D atomic-scale resolution, and single-atom level determination of the 3D surface atomic structure for general 3D nanomaterials still remains elusive. Atomic electron tomography (AET) has been developed as a powerful tool to determine the 3D atomic structure of nanomaterials at the single-atom level However, often due to geometrical limitations, only part of a full tomographic angular range is experimentally measurable (so-called “missing wedge” problem), which results in elongation and Fourier ringing artifacts along the direction of the missing information in the reconstructed tomogram. The missing wedge artifact negatively affects the accuracy of the 3D surface atomic structure determined from the tomogram. We network-assisted AET missing data missing wedge learning 3D-unet