The following options for search-match are available on the "Search-Match" page of the "Options" dialog:
Using this Search-Match option it is possible to restrict the 2θ range in which the figure-of-merit (FoM) is calculated to the range of the diffraction pattern of the unknown sample. This can be defined independently both for the low-angle as well as for the high-angle part of the diffraction pattern.
These options can be adjusted automatically (depending on the experimental diffraction pattern), if the Automatic option on the right-hand side is checked.
Please note:
Why is it useful to have this option? If the diffraction data have only been collected up to a certain more or less low 2θ angle or in a rather limited 2θ range, it is possible that there are more strong/important peaks with smaller or larger 2θ values that have not been recorded. In this case, peaks of reference database patterns with 2θ values smaller or larger than the currently outermost experimental peak should not be taken into account: If the data had been collected from smaller or up to larger angles, the peaks corresponding to the reference entries most probably would have been detected as well. Hence, it is reasonable to limit the calculation of the FoM to the range of the experimental diffraction data, and to put no negative weight on the "missing" peaks.
However, in other cases (especially with samples revealing a low scattering power, e.g. organic samples) there are generally no more peaks detected to the right of the unknown diffraction pattern, even if this range had been scanned. In these cases, reference entries which have peaks to the right of the right-most peak in the unknown pattern are most probably not matching. In order to account for this in the FoM calculation, the (mis-)matching of these peaks has to be included.
Let's summarize: If you have a sample with low scattering power, you should disable the "Restrict to experimental range" option(s), in order to prevent entries with peaks at larger 2θ-values to be present at the top of the candidate list. If you have a sample with large scattering power, you should enable this option so that you do not miss corresponding reference entries which also have peaks at low or high 2θ-angles.
As the name suggests, a "search-match" typically consists of two phases: First, the presence of N strongest peaks (N=3 in case of Match!) is checked ("search"), whereupon
a full "match" is run only for those reference database entries whose N strongest peaks were present.
You can disable this behaviour using this option, so that a full match is run for all patterns of the selected reference database.
Note that activating this option might prevent minor phases from being detected, since
a peak can easily be missing for minor phases which might prevent the phase from being listed in the candidate list if this option is active! If you suspect that minor
phases are missing, you should deactivate this option and run a new search-match
calculation.
This option only works for reference databases created with Match! version 1.8 or later!
The 2theta window used for peak correlations is normally adjusted automatically (check box "Automatically...")
before the search-match calculation, based on the average FWHM
(full width at half maximum) of the experimental peaks. As a consequence, broad peaks (large
average FWHM) result in large peak correlation windows, so a large number of peaks (entries)
may be considered as "matching".
You can disable the automatic determination of the 2theta peak correlation window size by deselecting the "Automatic..." check box below "Check 3 strongest peak...". Try to use/adjust a smaller value for
"Δ2θ" if you obtain a large number of entries from your search-match.
If this Search-Match option is active, the zero point shift will be optimized individually for each comparison "reference entry" vs. "unknown pattern" during the search-match calculation, in order to reach the best possible agreement.
If this option is active, Match! will run another search-match calculation after the selection of a matching phase. The purpose is to check all entries which have not yet been selected as matching against the "residual pattern", taking into account the peaks (or profile areas) "occupied" by already matched entries.
There are important issues about automatic residual searching you should always keep in mind:
For profile-fitting search-match (PFSM), the follwing settings can be adjusted:
Sometimes it can be useful to define a minimum number of reference pattern peaks that can be correlated to experimental peaks, e.g. if the 2theta-range of your diffraction pattern is rather small. In this
case, there may be a large number of reference patterns/phases where only few peaks are located within the experimental 2theta-window, so that the probability that one or two of them can be correlated is rather large.
If you increase the minimum number of correlated peaks, many of these "wrong" entries/phases are no longer
considered due to their reduced figure-of-merit (FoM) values, however, you should carefully check for
matching high-symmetry phases that may also be excluded by doing so!
The default value for this parameter is 2.
You can define a minimum intensity value (given as relative intensity between 0 and 999)
for peaks to be correlated. Only peaks whose relative intensity is larger than this value will be
correlated and hence be taken into account in the FoM calculation (i.e. qualitative phase analysis).
By using this parameter you can mask small peaks that might just be noise, e.g. determined by
accident by the automatic peak searching function.
You can also modify this parameter by moving the small red bar on the intensity axis
of the pattern graphics up or down.
The default value for this parameter is 1.
Note: This parameter is also used for determining which peaks can be automatically selected for indexing
if no peaks have been marked by the user; see the indexing options.
The values of these two parameters are identical.
The following numerical parameters affecting the FoM calculation during the search-match process can be adjusted using the sliders:
The weight of the 2θ-differences of correlated peaks on the overall FoM-value. The range of the parameter value is 0.0 to 1.0. A value of 0.0 means that the differences of the 2θ-values of correlated peaks does not have any influence on the figure-of-merit and hence on the result of the search-match (of course, this is not a recommended value !). The default value is 0.5.
The weight of the intensity-differences of correlated peaks on the overall FoM-value. By reducing this parameter you can (at least to a certain degree) account for possible "bad" intensities due to preferred orientation. The range of the parameter value is 0.0 to 1.0. A value of 0.0 means that the differences of the intensity-values of correlated peaks does not have any influence on the figure-of-merit and hence on the result of the search-match. The default value is 0.5.
This parameter can be used in order to tell Match! if single phases shall be preferred. The range of the parameter value is 0.0 to 1.0. A value of 0.0 means that single phases covering as much of the unknown sample's diffraction pattern as possible are preferred. Low values of this parameter should be used only if you are sure that you are looking for a single phase! The default value is 0.5.
Using this parameter you can put more (or less) weight on the numerical impact of a reference pattern's number of peaks that cannot be correlated to experimental peaks. The default value is 0.25.
The minimum figure-of-merit for an entry in order to be accepted in the candidate list (answer set) during the search-match process. The smaller this value, the more entries will result from the search-match and vice versa. The range of the parameter value is 0.0 to 1.0. A value of 0.0 would result in ALL reference database entries being loaded into the candidate list (don't try this !). The default value is 0.6.
Only reference entries with a FoM larger than this value are taken into account during the automatic selection of matching phases. The default value is 0.75.