Chemistry Reference
In-Depth Information
3.3.3 Neutron Powder Diffraction
As with X-ray diffraction studies, the simplicity of powder methods means that
such studies have proved more popular than single-crystal neutron studies. Diffrac-
tion data have been collected using both angle- and energy-dispersive methods,
typically at reactor and pulsed (spallation) neutron sources, respectively. However,
the early high-pressure study of Bi by Brugger et al. [
172
] used a reactor source but
with a mechanical chopper in order to utilise time-of-flight diffraction methods.
While detailed crystallographic studies were conducted using neutron powder-
diffraction methods prior to 1990, using both high-pressure gas [
173
,
174
] and
clamped piston-cylinder [
175
,
176
] cells, the upper pressures were limited to some
1 GPa for gas cells, and 2-3 GPa for clamped cells.
The scattering of the neutrons from the pressure cell body and/or anvils can be
greatly reduced, or removed completely, in energy-dispersive experiments per-
formed on spallation neutron sources by collecting the diffraction data at 90
to
the incident beam. With careful collimation of both incident and diffracted beams,
only the sample is both illuminated by the incident beam and viewed by the
detector. The P-E cell, when used on such a source, is mounted so that the incident
beam is incident through one anvil, while the diffracted beam exits at 90
through
the gasket (see Fig.
2
). This enables very high quality diffraction data to be
collected, free from contamination from the pressure cell components. The data
can also be corrected accurately for pressure cell absorption and other systematic
errors [
177
], with the result that accurate relative peak intensities can be collected
to very high scattering vectors. At the ISIS facility, this ability has been used to
make detailed studies of crystalline samples, for example ammonia and gas hydrates
[
102
,
178
-
180
], as well as both liquid water [
181
], and the various forms of
amorphous ice that exist at high pressures and low temperatures [
182
,
183
].
In angle-dispersive studies, using the high-pressure gas or piston-cylinder
clamped cells, an intense background can arise from the cell body. However, the
contaminant diffraction peaks are typically few in number, arising fromAl or Al
2
O
3
,
and those regions of the diffraction pattern are typically omitted from a Rietveld fit.
The intensity of these peaks can be greatly reduced by increased collimation of the
detectors. When P-E cells are employed on reactor sources, the use of highly-
absorbing boron nitride anvils results in almost all of the scattering from the anvils
being absorbed, giving high quality diffraction profiles, ideally suited to Rietveld
refinement [
184
].
3.3.4 Neutron Single-Crystal Diffraction
As with neutron powder methods, both monochromatic and time-of-flight (Laue)
techniques have been used for single-crystal neutron studies. Such studies are much
less numerous, however, as a result of the difficulty of compressing suitably sized
(
1mm
3
) single-crystal sample to the required pressure while maintaining the
crystal quality. The piston cylinder cell design of McWhan and others [
128
] has
been used at the ILL reactor source for a number of neutron structural studies to
>
