Transmission electron microscopy is uniquely powerful when it comes to imaging point or line defects in a solid, or when one wants to image the structure of a single molecular layer in a film. Thus, TEM provides information about buried, localized features of thin films which is difficult to obtain in any other fashion.
The TEM images taken as part of this study have strikingly confirmed previous hypotheses about the structure of heterostructures and multilayers grown with the ALL-MBE technique. In particular, the TEM images demonstrate that single layers of 2212 and 2201 can be grown as part of a multilayer. Furthermore, single layers of metastable or unstable phases such as 2278 and 1278 can also be grown. These TEM micrographs provide the first direct evidence for the existence of these phases. While growth flaws like twins, stacking faults and antiphase boundaries do occur, especially in overgrowths on metastable barrier layers, the images are most notable for the long, unbroken continuity of most 2212 planes. Up to this point, ALL-MBE growth of BiSrCaCuO superconductors have not reached the perfection that is observed in MBE growth of GaAs, AlAs and related compounds. However, planned improvements in the ALL-MBE process such as rotation of the substrate during growth may make comparably low defect densities possible.
The TEM images presented here support previous ideas about the two-dimensional nature of superconductivity in the 2212 compound. In addition, they provide new information about the ways in which growth of inherently highly anisotropic phases may differ from growth of cubic materials. Studies of the growth of high-temperature superconductors may in the end have as much to teach us about anisotropic solids as about the fundamental nature of superconductivity.
Acknowledgements We thank Sandia National Laboratory and George Thomas for use of the TEM. We thank P.A. Sterne and M.J. Fluss for useful discussions, R.G. Musket for the RBS measurements, J.M. Yoshiyama for the EDAX measurements, and D.L. Phinney for the SIMS. Part of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract number W-7405- ENG-48. The work at Varian was supported in part by NRL and ONR via contracts N00014-93-C-2055 and N00014-94-C-2011.