198 Topics in Current Chemistry Editorial Board: A. de Meijere KN ...

184 M.R. Caira
theory, the area under the curve in a DSC trace is directly proportional to the
enthalpy change for the thermal event, but special precautions are necessary to
obtain accurate values for this quantity [18]. An alternative method for determining
enthalpies of desolvation of pseudopolymorphs, based on direct vapour
pressure measurement of the liberated solvent as a function of temperature, has
been described [50]. The DSC traces in Fig. 7 display endotherms (peaks A,
above the baseline) whose temperature ranges correlate well with those of the
TGA traces, confirming that these peaks represent desolvation processes. The
bimodal nature of these endotherms is difficult to interpret but evidently reflects
complex desolvation mechanisms. The difference in the DSC temperature
ranges for loss of DMF (100–125 °C) and loss of DMSO (160–235 °C) is striking.
These pseudopolymorphs were subsequently shown by single crystal X-ray
analysis to contain their respective solvent guest molecules in constricted
channels formed by the nitrofurantoin host framework; each DMF molecule is
bound to one drug molecule by a single hydrogen bond, whereas each DMSO
molecule forms two hydrogen bonds with neighbouring drug molecules, thus
providing a possible explanation for the significantly higher onset temperature
for the loss of DMSO. In this connection, the use of the simple parameter T on–T b
(T on= extrapolated DSC desolvation onset temperature, T b=boiling point of the
guest solvent) has been proposed as a possible measure of the relative thermal
stabilities of inclusion compounds [50]. For both pseudopolymorphs referred to
in Fig. 7, desolvation produced the b-polymorph, as confirmed by the characteristic
sharp fusion endotherm at 273 °C, followed by a decomposition exotherm
peaking at 279 °C.
Recent reviews of thermoanalytical methods and their application to polymorphic
systems [18, 23] include descriptions of modern instrumentation,
discussions of the influence of instrumental and sample parameters on measured
data, as well as guides to the interpretation of DSC traces. One review [23]
includes a summary of the thermodynamic rules established by Burger [107] for
distinguishing monotropic and enantiotropic transitions. It is worth emphasising
that in the absence of supporting techniques such as hot stage microscopy
(HSM), X-ray diffraction (XRD) and Fourier transform infrared spectroscopy
(FT-IR), errors in the interpretation of DSC data may result. Reference to HSM
was made earlier in connection with its use in screening recrystallized batches
of a compound to identify polymorphs and pseudopolymorphs. Apart from its
ability to yield melting points and transition temperatures for polymorphic
changes or glass transitions, HSM is an invaluable adjunct to the DSC technique
in that visual recording of thermal events (e.g. desolvation, recrystallization,
sublimation) may corroborate assignments of endothermic or exothermic
peaks appearing in DSC traces.As pointed out recently [108], visual observation
can sometimes be more sensitive indicators of phase changes than DSC measurements.
The report describes the construction and use of the device shown
in Fig. 8, which allows simultaneous DSC measurement and microscopic observations
with a video-camera. A second camera is used to give a direct readout
of the varying sample temperature.
Solid-state infrared (IR) spectroscopy in the spectral range 400–4000 cm –1
has been used extensively to distinguish different polymorphic forms as well as