The Search Begins

Course Material Index  Section Index  Previous Page  Next Page 

The Search Begins

Having collected a powder diffraction pattern for some unknown material that has been presented to us, and having a suitable database such as the ICDD-PDF, how do we proceed to search for suitable candidate materials? Plodding through the database entries, one at a time comparing each pattern to our unknown, is clearly not an intelligent strategy for a human or even computer-aided search. We need to have some essential distillation of the database that we can view more quickly than the complete data set. Such a distillation is the so called Hanawalt Index. It is just a list of the three strongest d spacings for each material contained in the database: they are the first three values in the top left hand side box in the card image shown on the previous page. The units are understood to be Angstroms (Å) and, since d spacings rather than 2θ values are used, the values do not depend on the X-ray wavelength.

These three numbers, the Hanawalt index, are usually sufficient in practice to narrow down the search to one or a few suspects which could then be examined in greater detail (this combined process is referred to as search/match). But to search using the Hanawalt index we obviously need to have a corresponding list compiled of all Hanawalt possibilities taken from the database and to have this list ordered in a fashion which is easy to interrogate. Such a system is termed the Hanawalt Search Manual, a page of which can be seen by clicking here (53 kBytes).

Each line contains first the three strongest d spacings in bold (i.e. the Hanawalt index) followed by the next 5 strongest d spacings, plus the formula/name and address (the whole 3+5 line entry is sometimes referred to as an extended Hanawalt index). The lines are grouped so that the strongest line is put in order first, then the second strongest followed by the third strongest line; one searches by trying to first locate the group with the same strongest line (d spacing), then moving onto the second, then to the third by which time the range should have narrowed considerably. However it is quite common for there to be multiple entries (i.e. other permutations) for each Hanawalt index. This is because, obviously with such a lot hanging on the choice of the three strongest lines, if that choice is in serious error (e.g. due to preferred orientation in the sample or a bad measurement) then the Hanawalt search is doomed to fail. To guard against this in marginal cases (e.g. when the second line is only slightly less intense than the first) the Hanawalt index will appear more than once but with the order changed. Another pitfall concerns inaccuracies in observed d spacings and for this some "error overlap" has been built into the Hanawalt groupings to guard against the complications from a d spacing being close to the border of two groupings, i.e. the searcher placing it in the wrong group. It is now time to try a simple example of the use of Hanawalt searching:

Course Material Index  Section Index  Previous Page  Next Page