Introduction

The discovery of cuprates exhibiting superconductivity at relatively high temperatures has opened up new prospects for the application of superconductivity in many areas, in particular in sensor systems and in electronics [13.1, 13.2]. In this respect the superconducting quantum interference device (SQUID) is one of the most attractive developments. Many different designs have been fabricated and studied, and modern SQUIDs on the basis of YBa2Cu3O7 have reached field sensitivity and performance levels not far different from those known for devices produced with classical low temperature superconductors [13.3, 13.4].

The physical properties of cuprate superconductors depend sensitively on the preparation conditions and the resulting microstructure. Owing to the essentially two-dimensional superconductivity and the very small coherence length in the cuprates, grain boundaries in general reduce the critical current density by orders of magnitude compared with the bulk value. Therefore much attention has been paid to the development of techniques for growing high quality epitaxial thin films of superconducting and appropriate non-superconducting materials required for active and passive electronic devices on suitable single-crystalline substrates. The remarkable progress achieved is closely related to both the development of proper materials preparation techniques and high quality materials characterization. In fact, device production requires atomic or close to atomic structural perfection. High-resolution transmission electron microscopy, which developed atomic resolution in many materials during the early 1980s, i.e. shortly before the new materials were discovered, has contributed substantially to the understanding of the structural properties of cuprate superconductors and to their use in electronic devices.

Besides instrumental resolution, two important technical factors are decisive for the enormous potential of high-resolution electron microscopy for superconductivity research.