It can happen that, by chance, exactly where a band could be, an intensity accumulation occurs that actually looks like a band profile.
To make matters worse, most band profiles do not look as ideal as is commonly thought, but have a rather atypical shape. With CALM, you can inspect any profile after defining the band positions.
Of course, it can also happen that single strong bands of another pattern become visible in one pattern because both superimpose in the image. These bands are usually easy to identify and are automatically ignored because they do not match the other bands, neither in direction nor in width.
The situation becomes more difficult when crystallographic intergrowths occur (e.g. single or multiple twinning) and such patterns overlap. There a separation can become difficult, especially if by chance both lattices generate a superlattice.
Wrong [hkl]* selected in a pattern from anatase (TiO2)
It is not very difficult to describe a band position [hkl]*j by another very close [hkl]*i, e.g. by [3hi 3ki (3li+1)]*. Depending on the (unknown) lattice parameters, the angular distance of [hkl]*i and [hkl]*j can be as small as a few tenths of a degree only.
Please consider: During definition of the trace positions the indexing is totally unknown. High and low indexing of [hkl]* is later decided by their bandwidth. However, by visual inspection of the circles around [hkl]* you can predict how likely the defined pole might be, or whether there is another connection of two narrow bands (small circles) which might match better.
Such an example is described in the following. In the left image below the uper right quarter of the Funk transformation of a Kukichi pattern from anatase is shown. Beside several ten poles one pole is displayed as red dot. It is defined by the intersection of two great circle. The one containing many [hkl]* is certainly reasonable. The other one, connecting one pole without and one with an intermediate band width (circle diameter) is only possibly reasonable
Selecting the pole (Ctrl+LMB) the band profile will be displayed in CALM. It is shown in the centered diagram by the white curve and does not really look like a reliable band profile. However, its 1st derivative looks promising since the asymmetry of the extrema positions (green vertical lines) is with θasym = ‒0.018° very low. However, the resulting indexing of the diffracting lattice plane is quite high: (0 -15 7), which indicates a big Bragg angle and a large distance of the reciprocal lattice point from the origin.
This is shown in the right image where this distance is so big that it is no more visible in this reciprocal latticeplane (uvw)*. There are already a few reciprocal lattice points, but this plane very likel does not match to this zone.
Correct description
We can try a slightly different direction [hkl]* using two alternative great circles (red colored in the left image below). Since the solution was already found before – see the green circles around each [hkl]* – the indexing of the red colored [hkl]* is given in the band profile diagram and delivers the diffracting lattice plane (0 -2 1).
Also for this band, the white colored band profile is not much different to the former one and looks again quite awful. Neither the bigger θasym = ‒0.047° and the slightly lower Iampl (derived maximum intensity) explain the preference of this solution compared to the upper one. However, the right image showing the zone in reciprocal space proves that both the direction and the distance between the two opposite reciprocal lattice points (bandwidth) perfectly match the translation lattice already found for this (uvw)*.
This again emphasizes that lower band edge asymmetry or apparently higher band intensity are not inevitable criteria for the detection of a band. The only criterion is that the derived reciprocal lattice point must match those already present, or it forms a matching high-order sublattice.
So be careful if only one lattice point occurs which forces a redefinition of all other lattice points.