Real time PCR - European Pharmaceutical Review

THERMAL ANALYSIS ISSUE 2009
system, TA Instruments LLC) was
employed to quantify the extent of
any disorder.
The use of IGPC for amorphous
content has become popular and the
technique offers many advantages 6 .
Briefly, in a gas-perfusion experiment
the sample is held in an ampoule
while a carrier gas flows over the
sample. The relative vapour pressure
(RVP) of the gas is controlled by
proportional mixing of dry (0% RVP)
and wet (100% RVP) gas streams.
The system can thus be programmed
to follow a specific RVP program.
IGPC is convenient for quantifying
amorphous material; the sample is
held dry and allowed to reach thermal
equilibrium. The RVP is then
increased (usually to ca. 90%) which
results in solvent uptake by the
sample. This plasticises the sample,
lowering the glass transition
temperature and causing
recrystallisation. The RVP is then
returned to zero, which dries the
sample. Subtraction of the areas
under the wetting and drying peaks
thus indicates the presence of any
irreversible events.
A calibration curve of amorphous
content versus heat of crystallisation
was prepared by mixing proportional
mass ratios of the crystalline and
amorphous SS samples. Full
experimental details are available
upon request.
The calibration curve obtained of
SS amorphous content versus
measured heat was linear (see
Figure 2 on page 29). It was thus
possible to use the data to quantify
extents of disorder in milled SS
samples, although it should be noted
that the use of this type of calibration
curve has some limitations 6,7 .
Principally, the material used to
prepare the calibration curve (a
mixture of wholly amorphous and
wholly crystalline particles) is
physically distinct from the study
material, which in this case will have
disordered regions (that is, a
combination of crystal dislocations
and amorphous areas) forming a
Figure 3: Particle size and amorphous content data as a function of milling time for ball
milled salbutamol sulphate
corona around an otherwise
crystalline substrate. The magnitude
of this effect will vary according to the
sample studied. In the case of
lactose6 significant issues arise
because of the existence of 2 anomers
and 3 crystal forms and the effect of
muta-rotation must be considered. For
SS the case is much simpler and
hence confidence in the correlation
between the data sets is higher, since
there are no isomeric forms. Since the
calibration curve was prepared with
totally amorphous material (because
it was spray-dried), the values
reported below for milled SS samples
are expressed in amorphous content
(% w/w) but it is noted that this term
really means extent of disorder and
accounts for crystal dislocations as
well as true amorphous regions.
SS samples were milled for various
time periods up to one hour. Samples
were subsequently removed from the
mill and analysed for particle size and
amorphous content immediately (to
reduce the risk of relaxation or
recrystallisation increasing the
particle size or reducing amorphous
content value). The average particle
size and extent of disorder data are
plotted in Figure 3. Over short milling
times there is a sharp reduction in
particle size, with no increase in
extent of disorder detected with the
calorimeter. After ca. five minute
milling time, the particle size
distribution remains constant and
there is a concomitant increase in
extent of disorder.
The data clearly shows that the
hypothesis that particle size reduction
occurs before generation of
amorphous content is correct. The
immediate benefit of this is that it
appears feasible to control the
micronising process to produce a
product with a defined particle size
distribution and a crystalline surface.
Although we did not attempt to make
similar measurements on a larger
scale, there would seem to be no
reason why this outcome would not
apply to larger ball milling apparatus;
the break-point time at which
maximum particle size reduction is
achieved would simply increase.
Although not explicitly explored in
this study, IGPC analysis may have
application to air-jet milling also. In
this case, samples are ejected from
the mill under centrifugal force, but
changing the pressure of the input
gas would modulate ejection time and
attrition rate.
One further conclusion can be
drawn from the data in Figure 3.
At longer milling times, the extent of
disorder starts to reach a plateau
while the particle size starts to
increase. It is likely that at these
longer milling times the conditions in
the mill are such that the ambient RH
or, more likely, temperature become
30
www.europeanpharmaceuticalreview.com