Srilankite has an orthorhombic unit cell with sides a=4.70, b=5.55, and c=5.02 Angstroms. The question: Can compositional ordering on the 4 Zr/Ti sites in each cell, across multiple cells, explain the diffraction and HREM images below? Shutting off the O atoms and bonds, as well as perspective, in the model from the right-click menu might help sort this out. We can add javascript buttons to do this sort of thing more directly later.
For example, the bright lattice spacings in the diffraction pattern below may index as multiples of (002) in the direction of the streaks, and as multiples of (220) perpendicular to the streaks (thus parallel to the wire axis). The superlattice (weaker) spots in that case may require Zr substitution periodicities something like two unit cells in length (4 native periodicity units) in each of those directions. Propose a substitution rule that might give rise to this pattern, and we'll simulate some HREM images for more detailed comparison with intensities in experimental images like that below.
Above find a diffraction pattern, and below
a HREM image, of the same nanowire.
These independently
contain evidence of an interesting set of
projected periodicities.
To get a start on this answer, you might ask yourself if all Zr/Ti sites are equivalent. Also, what do we know about propensities that Zr atoms might have for either avoiding one another, or for getting together? You might also ask about the reason why the (220) spacing is found in the diffraction pattern above, but the superlattice row in from there (toward the DC peak) does not show a (110) spacing as well. Instead, one finds spots in that row canted by 10 to 11 degrees up and down from the (110) direction. In that context, would the model projection below, with extra strong (or weak) scattering columns circled, give us the diffraction pattern above?
Here's a first proposal, and a comparison to experiment...