No abstract

The excitation of surface plasmons in metallic nanostructures provides an opportunity to localize light at the nanoscale, well below the scale of the wavelength of the light. The high local electromagnetic field intensities generated in the vicinity of the nanostructures through this nanofocusing effect are exploited in surface enhanced Raman spectroscopy (SERS). At narrow interparticle gaps, so‐called hot‐spots, the nanofocusing effect is particularly pronounced. Hence, the engineering of substrates with a consistently high density of hot‐spots is a major challenge in the field of SERS. Here, a simple bottom‐up approach is described for the fabrication of highly SERS‐active gold core‐satellite nanostructures, using electrostatic and DNA‐directed self‐assembly. It is demonstrated that well‐defined core‐satellite gold nanostructures can be fabricated without the need for expensive direct‐write nanolithography tools such as electron‐beam lithography (EBL). Self‐assembly also provides excellent control over particle distances on the nanoscale. The as‐fabricated core‐satellite nanostructures exhibit SERS activities that are superior to commercial SERS substrates in signal intensity and reproducibility. This also highlights the potential of bottom‐up self‐assembly strategies for the fabrication of complex, well‐defined functional nanostructures with future applications well beyond the field of sensing.

Crystallographic defects can now be routinely imaged at atomic resolution with aberration-corrected scanning transmission electron microscopy (STEM) at high speed, with the potential for vast volumes of data to be acquired in relatively short times or through autonomous experiments that can continue over very long periods. Automatic detection and classification of defects in the STEM images are needed in order to handle the data in an efficient way. However, like many other tasks related to object detection and identification in artificial intelligence, it is challenging to detect and identify defects from STEM images. Furthermore, it is difficult to deal with crystal structures that have many atoms and low symmetries. Previous methods used for defect detection and classification were based on supervised learning, which requires human-labeled data. In this work, we develop an approach for defect detection with unsupervised machine learning based on a one-class support vector machine (OCSVM). We introduce two schemes of image segmentation and data preprocessing, both of which involve taking the Patterson function of each segment as inputs. We demonstrate that this method can be applied to various defects, such as point and line defects in 2D materials and twin boundaries in 3D nanocrystals.