INTERFACES IN GAAS/AIAS - PERFECTION AND APPLICATIONS

This article presents a short review of our present understanding of the structure of semiconductor interfaces and the way they may be used to investigate solid state processes at the atomic level. Until recently it was customary to claim the fabrication of atomically perfect interfaces, based on the results of photoluminescence (PL) experiments. This was based on the splitting of PL lines into closely spaced satellites, thought to stem from the recombination of excitons within large, atomically perfect terraces. However, quantitative microscopic data, obtained by chemical mapping [A. Ourmazd, F. Baumann, M. Bode, and Y. Kim, Ultramicroscopy 34, 237 (1990)], clearly indicate the presence of significant atomic scale roughness [A. Ourmazd, D. W. Taylor, J. Cunningham, and C. W. Tu, Phys. Rev. Lett. 62, 933 (1989)]. This has been confirmed more recently by high resolution PL and photoluminescence excitation experiments [C. A. Warwick, W. Y. Jan, A. Ourmazd, and T. D. Harris, Appl. Phys. Lett. 56, 2666 (1990)], by Raman scattering [D. Gammon, B. V. Shannabrook, and D. S. Katzer, Phys. Rev. Lett. 67, 1547 (1991)] and by high resolution x-ray measurements [M. Lagally, presented at the MRS Fall Meeting in Boston, 1991, Ca5.2]. All these experiments reveal a more complex interfacial configuration, produced by the interplay of roughness at different length scales. The quantitative microscopic techniques developed for the study of interfacial structure also makes possible a class of experiments, where the interface is used as a "photographic emulsion" layer, capable of recording the passage of defects. This allows one to investigate solid state processes at the atomic level. For example, the intermixing caused by the passage of an implanted ion through a GaAs/AlAs interface of a multilayer may be used to track the ion and investigate the microscopics of its interaction with the host material [M. Bode, A. Ourmazd, J. A. Rentschler, M. Hong, L. C. Feldman, and J. P. Mannaerts, Mater. Res. Soc. Symp. Proc. 1989 157, 197 (1990)]. From this type of experiments it is possible to deduce the microscopic damage signature produced by individual implanted Ga+ ions [M. Bode, A. Ourmazd, J. Cunningham, and M. Hong, Phys. Rev. Lett. 67, 843 (1991)]. Intriguingly, the damage is charged, and thus strongly influenced by internal electric fields in the solid. This may be exploited to microscopically steer ion implantation damage in solids.