Chemistry Reference
In-Depth Information
(1) The peak positions (the variables that determine the peak positions include the
unit cell parameters and the zero-point shift parameter)
(2) The background intensity distribution
(3) The peak widths
(4) The peak shapes
(5) The peak intensities
With regard to (1), the input values of the unit cell parameters used in the profile
fitting procedure are those obtained in the indexing stage, and the refined values that
result from the profile fitting procedure represent a more accurate set of unit cell
parameters. It is important to emphasize that no structural model is used in the
profile fitting procedure (except in so far as the unit cell parameters are used to
determine the peak positions), and the intensities in (5) represent a set of intensity
variables that are refined to give an optimal fit to the experimental powder XRD
pattern without reference to any structural model. Thus, the aim of the profile fitting
procedure is not to determine the crystal structure, but rather to obtain reliable
values of the variables that describe different features of the powder XRD profile
[i.e. (1)-(5) above] in preparation for the subsequent stages of the structure deter-
mination process. As now discussed, it is also important to note that different
strategies for structure solution make use of different combinations of the variables
in (1)-(5) as input information.
The traditional approach for structure solution requires, as input data, the
integrated peak intensities extracted from the experimental powder XRD pattern -
i.e. the fitted intensity values (5) obtained from the profile fitting procedure.
In addition, some implementations of the direct-space strategy for structure solution
are based on comparison of integrated peak intensities, and also make use of the
intensity values (5) obtained at the profile fitting stage. After extracting the integrated
peak intensities (5), these approaches do not make any further use of the experimental
powder XRD profile during the structure solution stage (although the experimental
powder XRD pattern is again used in the Rietveld refinement stage).
Alternatively, several other implementations of the direct-space structure solu-
tion strategy involve comparison between experimental and calculated data using a
whole-profile figure-of-merit such as
R
wp
. In this case, the intensity data (5) extracted
from the experimental powder XRDpattern in the profile fitting procedure are not used
in the structure solution stage. Instead, the variables (1)-(4) determined in the profile
fitting procedure are used (together with intensities calculated for trial structural
models) to construct the calculated powder XRD pattern for each trial structure
generated in the direct-space structure solution calculation.
Following the profile fitting procedure, the space group can be assigned by
identifying the conditions for systematic absences in the intensity data (5). If the
space group cannot be assigned uniquely, structure solution calculations should be
carried out separately for each of the plausible space groups. Knowledge of unit cell
volume and space group, together with density considerations, allows the contents
of the asymmetric unit to be established. Information from other experimental
techniques (particularly high-resolution solid-state NMR) may be particularly
