Chemistry Reference
In-Depth Information
Figure 3.4.
Example of miscibility changing with concentration. Values of
T
g
for binary
mixtures of citric acid
-
indomethacin as a function of weight fraction of citric acid. The
square symbols correspond to the values of
T
g2
, while the circles show the values of
T
g1
.
(Reproduced with permission from Ref. 53. Copyright 1998, Springer.)
miscibility can change with concentration [53] as shown with indomethacin and citric
acid in Figure 3.4; below a citric acid weight fraction of 0.3, only one
T
g
is observed
suggesting a miscible system, but above this weight fraction two
T
g
values are observed
indicating a nonmiscible system. This observation should be applicable to API:polymer
systems as well.
Based on miscibility studies, it is important to look at a range of concentrations of
API and polymer. A common perception is that more polymer will provide more
stability, but it has been shown that even small amounts of polymer (such as 5%) can
provide stabilization for years [54]. Many pharma applications target polymer levels
above 50% polymer, but the
final dose of the API also has to be considered, since a large
amount of polymer can signi
cantly increase the size of the capsule or tablet to be
ingested. A suggested initial range is 20
50% API, depending on the formulation plans
of the compound. Guidelines shown in Figure 3.2 can also be used if applicable.
The minimum amount of polymer needed to maintain physical stability is another
aspect to consider. One approach that has been reported is to make dispersions with low
polymer concentrations, equilibrate under extreme stress conditions, and quantify the
amount of crystalline API that is produced [55]. An improved kinetics approach
describes the physical stability of amorphous solid dispersions. A low level of polymer
is used to encourage crystallization of the API for quantitation, and a plot of the weight
fraction crystallized versus the amount of polymer results in a minimum value of polymer
needed to maintain an amorphous material, as shown in Figure 3.5 for an efavirenz:PVP
dispersion. In this example, a minimum polymer concentration of 24% PVP was
determined, and this value was con
-
rmed using a 75:25 API:PVP sample stored at
72
C/53% RH that showed no crystallization for 24 months. For early development
projects, collecting physical stability data on three to
°
five dispersions at extreme
