A major advantage of the ALL-MBE method of film deposition is that its precise control of layering makes possible the growth of BiSrCaCuO phases which are not bulk equilibrium phases. For example, ALL-MBE has previously been used to grow films of the 2234 and 2245 compounds. The non-equilibrium BiSrCaCuO phases have the bounding double BiO and single SrO layers that 2212 has, but the number of CaCuO2 units internal to the unit cell may be varied in a large range as long as the unstable layer is epitaxially stabilized by growth on 2212. Figure 3 shows the proposed crystal structure of the 2278 phase and the well-known structure of 2212 for comparison. The c-axis lattice constant of 2278 is calculated to be 71.2 Å by assuming that all interplanar distances are the same as in 2212. This assumption is probably not exact due to the different charge-balance in 2278. In addition to growing non-equilibrium Bi2Sr2Can-1CunO 2n+4 phases, layers with only one bounding BiO layer have also been deposited which have the stoichiometry Bi1Sr2Can-1CunO 2n+3. Thus a Bi-1278 layer can be grown in a similar fashion to Bi-2278.
X-ray diffraction has been used to provide evidence for the successful growth of several of these higher-n phases. Figure 4 shows a layer of the 2278 (n=7) material sandwiched between two thick films of 2212. While examination of TEM images is an inexact method of lattice constant determination, the figure shows that the layer spacing of the 2212 layers (15.4 Å) is slightly less than half the thickness of the nominal 2278 layer, which is calculated to be 35.6 Å thick. The 2278 phase is of considerable interest because when doped with Dy it is useful as a tunnel barrier in Josephson junctions. 2278 and 1278 are also superconducting in their own right.[17,18]
TEM cross-sectional images (not shown) have also been taken of nominal 1278 tunnel barriers. The nominal layer thickness of 1278 (which should be a simple tetragonal rather than a body-centered tetragonal structure) is calculated to be 32.5 Å. It was not possible from TEM to distinguish them from the 2278 barriers and to determine whether a single BiO layer structure has actually been achieved. A microscopy technique which provides elemental contrast may be necessary to get at this information or to determine the rare-earth-atom spatial distribution in the Dy-doped barrier layers. At this stage it is possible to report only that the transport properties of tunnel junctions with 2278 and 1278 barriers are quite different.