Chemistry Reference
In-Depth Information
heated to the temperature of interest. By holding the samples isothermally until
crystallization could be detected from the appearance of an exothermic peak, the
induction times for the crystallization process could be evaluated. For
flurbiprofen and
tolbutamide dispersions with PVP, it was observed that the induction time for
crystallization increased with an increasing amount of PVP; in contrast, increasing
the temperature decreased the induction time. By assuming that the induction time is
inversely proportional to the nucleation rate, it was concluded that PVP decreased the
nucleation rate. PVP was more effective at
increasing the induction times for
tolbutamide compared with
flurbiprofen, and this is thought to be due to better
speci
flurbi-
profen and PVP. Using the same method, induction times for solid dispersions of
naproxen and HPMCAS were also measured [116]. In this instance, it was observed
that the induction time for naproxen crystallization decreased with temperature until a
minimum value was obtained at around 335K and then started to increase with
increasing temperature. This result highlights the complex temperature dependence of
nucleation whereby it seems that the nucleation rate reached a maximum at 335K. For
these examples, induction times were relatively short, making it possible to study them
with DSC. Unfortunately, for many systems of interest, the long induction times
preclude direct measurement of the crystallization event using this approach.
An alternative approach, which does not require long periods of equilibration in
the instrument, is to store samples under certain conditions, and then analyze them by
DSC using a more typical scanning mode. The crystallization of solid dispersions of
PVP with either nifedipine or phenobarbital was studied using this approach [30].
Solid dispersions were stored between 25 and 80
c interactions between tolbutamide and PVP relative to those between
C in a dry environment, and at
certain time intervals samples were analyzed using DSC. Nifedipine dispersions
crystallized when heated above
T
g
and the magnitude of this exotherm was used
to estimate the amount of amorphous material remaining. With increasing storage
times, more material crystallized and thus the size of the recrystallization exotherm
decreased. The fraction of amorphous nifedipine remaining was calculated from the
ratio of the exotherm obtained following storage at a particular time to that of a freshly
prepared amorphous sample. For this experimental approach, the inherent assumption
is that all of the remaining amorphous nifedipine in the solid dispersion will be able to
crystallize during heating. For the phenobarbital
°
PVP dispersions, an alternative
approach was used. For these systems, the change in the heat capacity as the system
passed through
T
g
was evaluated using modulated DSC. For a pure compound, it has
been demonstrated that the change in heat capacity at
T
g
correlates well with the extent
of crystallinity [117]. This is because, as the drug crystallizes, the
-
C
p
value for the
T
g
event of the remaining amorphous fraction decreases in a manner proportional to the
loss of the amorphous material due to crystallization. By assuming that a similar
relationship holds for a solid dispersion, the fraction of amorphous drug remaining at
time
R
(
t
) can be calculated [30]:
Δ
R
t
Δ
C
p
T
g
t
C
p
T
g
0
;
(5.12)
Δ
