Ph.D. thesis (pdf) - dirac

122 Mean squared displacement
collected at 7 detectors, which for the experiments reported here were positioned
such that they spanned a range in angle (2θ) from 14 ◦ to 156 ◦ corresponding to a
Q range of 0.2 Å −1 to 2 Å −1 .
The monochromation of the incoming beam as well as the scattered beam is done by
use of crystals in backscattering geometry. The use of backscattering geometry optimizes
the energy resolution. The (1 1 1) reflection of Si was used in our experiment
yielding an energy resolution of FWHM=1 µeV. This energy resolution corresponds
to a timescale in the order of a few nanoseconds. Hence, intensity corresponding to
processes that are slower than this, will contribute to the measured elastic intensity
(see section 4.3.10).
Cumene, m-toluidine, and DBP were studied at atmospheric as well as at elevated
pressure, while DHIQ was studied at atmospheric pressure only. See appendix A for
details on the samples.
The elastic scattering was measured as a function of temperature in a temperature
range of of 2 K to ∼ 1.5T g ≈ 300 K. This was done in cooling in order to avoid
crystallization of the sample. The cooling rate was approximately 0.5 K/min.
The experiments at atmospheric pressure were performed using a standard cylindrical
aluminum cell for liquid samples. This cell has about the size of the beam,
namely about 4 cm 2 . The sample thickness was 0.05 mm. This yields a transmission
of roughly 95% for organic liquids, depending on density and the exact composition
of the sample.
The experiments at elevated pressure were performed using clamp cells (of slightly
different dimensions depending on pressure range). The cells had outer diameters
of 1 cm and a wall thickness of about 2 mm. The compression was in the case of
cumene performed in the pressure lab and the cell was subsequently sealed. However,
we found that this method yields a large uncertainty on the pressure 1 . The
measurements on DBP were performed while applying in situ compression and readings
of the pressure. This allowed us to adjust the pressure during the cooling and
to have a more precise reading of the pressure.
The cell (and the sample) only cover a small part of the beam. We therefore placed
a cadmium mask in front of the sample in order to absorb the excess part of the
beam and thereby reduce the background signal.
We place a small cylindrical aluminum inset in the center of the cell in order to avoid
too thick a sample, since this leads to multiple scattering. The sample thickness was
1 The actual pressure in the cell can be estimated from the point of fusion which is easily seen
in the raw data when heating.