Microstructural analysis of YBa2Cu3O7−x superconductor and low-dose scanning transmission electron microscopy

Sunghwan Hwang, Purdue University


High temperature superconductors (HTS), especially YBa2Cu3O7-x (YBCO), are among the most promising materials and have been widely studied for commercial applications. The main challenge in optimizing the critical current density of the HTS is the immobilization of the vortices by means of artificial flux pinning centers, which can be controlled by the microstructure of high-temperature superconductor. In this study, we focused on the Dy-doped YBCO samples that have different process conditions used for loading oxygen into the Y0.5Dy0.5Ba2Cu3O7-x post reaction. The microstructure of Y(Dy)BCO films were analyzed by high-resolution transmission electron microscopy. In the Y(Dy)BCO layer, individual nanoparticles, stacking faults and particles with facets were identified. The low temperature optimized YBCO sample has the average size of precipitates of 26.5 nm and densities of stacking faults of 0.0184 nm-1 while the sample for 77 K has the average size of precipitates of 23.3 nm, densities of stacking faults of 0.0256 nm-1, and more small nanoparticles. The lower density of small nanoparticles and stacking faults in low temperature optimized YBCO can explain the lower and nearly isotropic critical current density dependence on the orientation of an applied magnetic field. In another part of this thesis, low-dose imaging in scanning transmission electron microscopy (STEM) is discussed. High signal-to-noise ratio and sub-angstrom resolution imaging has been achieved with STEM by recent technological advances. However, the high current density of the electron probe can cause electron beam damage to electron-beam sensitive materials. Therefore, methods to reduce electron dose are required for STEM analysis of electron-beam sensitive materials. A new image acquisition method with random sparse-scan and model-based iterative reconstruction (MBIR) was applied to overcome the radiation damage problem in this study. This is the first experimental result using the random sparse-scan method on the FEI Titan 80-300 S/TEM without any modification of hardware or alignment. We selected only a small portion of the specimen to be scanned by the electron probe, whereas the conventional STEM scans all areas of the specimen. The sparsely acquired image was reconstructed by MBIR. In order to compare the performance of reconstruction with a different sampling ratio, profile plots of intensities along a line were presented. In addition, reconstruction errors of reconstructed images were calculated to quantitatively estimate the quality of the reconstructions. The reconstructed STEM image, even from the sampling ratio of 5%, had a comparable quality to the full STEM image acquired by the conventional method. These results will have far reaching consequences for improving the formulations of the preexisting protocol of image acquisition on STEM and the possibility to reduce electron dose by two orders of magnitude.




Ortalan, Purdue University.

Subject Area

Condensed matter physics|Materials science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server