Practical Applications of HyperDSC in a Pharmaceutical Laboratory

A P P L I C AT I O N N O T E
Differential Scanning Calorimetry
Authors
Svenja Goth
PerkinElmer, Inc.
Rodgau, Germany
Didier Clénet
Sanofi-Aventis
Analytical Sciences Department
Physical Characterization Laboratory
Vitry-sur-Seine, France
Practical Applications
of HyperDSC in a
Pharmaceutical
Laboratory
HyperDSC™ (or High Speed DSC), a
thermal technique that complements
conventional calorimetry, has
found applications in the fields
of pharmaceutical, polymers and
compounds.
This technique features measurements taken at high heating (or cooling)
rates, from 100 to 500 °C/min. It is an approach that makes it easier to detect
such events as glass transition or the melting of a compound when they
are concealed by kinetic phenomena like, for example, water vaporization,
crystallization or chemical degradation.
Furthermore, HyperDSC offers greatly enhanced analysis
sensitivity due to the concentration of the energy phenomena
measured into a very brief space of time. Calorimetric
analyses on samples of very low mass (below 10 µg) are
thus made possible. The detection limits permitted by this
technique can be lowered considerably as compared with
conventional DSC, down to values < 1% when quantifying
physical forms (amorphous or crystalline). HyperDSC analyses
can be carried out on the power compensation DSC 8500
from PerkinElmer shown in Figure 1.
Figure 1. DSC 8500.
Presentation and relevance of HyperDSC
A direct, reliable measurement of heat flux up to
750 °C/min.
First, a quick review of the differences between power
compensation DSC and heat-flux DSC is presented. Its
design and working principle makes power compensation
DSC the best technique for analysis at high heating and
cooling rates.
The melting peak of an Indium standard has reliable onset
and enthalpy values at any chosen heating speed, even at
500 °C/min as shown in Figure 4 and the inset Table.
The small low-mass furnace
(Figure 2) of the power
compensation DSC 8500 is
highly responsive with respect
to the chosen temperature
programs. Power compensation
provides a direct measurement
of the heat flow given off or
absorbed by the sample, as
well as constant temperature
Figure 2. Sample furnace of a
readings up to rates of
power-compensation DSC.
500 °C/min.
After calibration – Indium at increasing scan rate
120
110
HeatingOnsetHeat
Rate
Temp.Flow
(˚C/min)(˚C) (J/g)
100
Heat Flow Endo Up (mW)
90
20
156.6028.81
80
100
156.7428.35
70
200
156.6728.43
60
300
156.7428.58
50
400
156.6928.32
500
156.7528.57
40
30
20
10
0
130
135 140
145 150
155
160
165 170
175
180
185 190
195
Temperature (˚C)
A very short startup transient
Figure 4. Melting peaks of an Indium standard at different heating rates.
As is shown in Figure 3, it takes the DSC 8500 less than
10 seconds to reach a selected temperature setting, even
at a rate as high as 750 °C/min. Thus, for an experiment
beginning at -65 °C and carried out at 200 °C/min,
the apparatus reaches equilibrium at about –30 °C, the
temperature from which any transition can be measured.
By comparison, a classical heat-flux DSC can take 30 to
60 seconds and is not ready for the detection of a transition
before +35 °C or more likely +135 °C, which therefore
makes it necessary to start the analysis at a much lower
temperature.
HyperDSC’s enhanced sensitivity
Fast Scan DSC provides increased sensitivity with decreasing
experiment time because the output of a DSC is mW or J/sec.
Figure 5 depicts the increase in heat flow signal at higher
heating rates.
35
HyperDSC – higher sensitivity at greater speeds
30
Thermoplastic Polymer
300 ˚C/min
25
121.1
20
Heat Flow Endo Up (mW)
Transient duration less than 8 seconds
100
Heat Flow Endo Up (mW)
80
60
40
20
5
10 ˚C/min
0
-5
2
4
6
8
10
12
14
16
18
20
22
24
Time (min)
-20
Figure 5. Increased sensitivity at higher scanning rates.
0.02
0.04
0.06
0.08
0.10
0.12
0.14
0.16
0.18
0.20
Time (min)
Figure 3. Illustration of the equilibration time for the DSC 8500 furnaces at
different heating rates.
2
100 ˚C/min
10
0.5
0
-40
-44.81
0.00
15
The enthalpy measured remains the same independent
on the scanning rate; only the intensity of the signal is
reinforced because of the concentration of energy absorbed
or released in a very short time frame.
Analysis of a polymorph mixture
The first sample shown is a mixture of Carbamezipine
stable Form I and metastable Form III. Figure 6 shows an
experiment carried out at 500 °C/min in which a very weak
endotherm was detected for the melting of Carbamazepine
Form III, a polymorph making up only 1% of the sample
content.
Peak = 179.203 ˚C
Peak Height = 0.0599 mW
-256
-280
Onset =
167.557 ˚C
Area = 0.053 mJ
Delta H = 0.024 J/g
-300
Peak = 200.642 ˚C
Peak Height =
172.1672 mW
Heat Flow Endo Up (mW)
-320
Area = 577.496 mJ
Delta H = 166.425 J/g
-340
-360
-380
Onset =
191.183 ˚C
-400
-420
Application of HyperDSC to several Sanofi-Aventis
samples
Polymorphism
This active ingredient comprises a mixture of physical forms
(phases α + b).
A DSC measurement under standard conditions at 10 °C/min
causes form α to transform into form b, which prevents the
two crystalline forms from being quantified by comparing
melting enthalpies.
In this case, HyperDSC inhibits phase transformation, and
the melting endotherms of the two forms can thus be used
to quantify the mixture. Figure 8 shows an experiment at
100 °C/min, carried out on a reconstituted mixture of form
α / form b at 95 / 5%.
-440
3.5
-460
Peak = 149.94 ˚C
3.0
-480
-145.2 150
160
170
180
190
200
210
220
228
Temperature (˚C)
Figure 6. Detection of 1% Carbamezipine Form III using HyperDSC.
Analysis of a weak glass transition (Tg)
The detection of a glass transition in an amorphous material
is also improved by HyperDSC (Figure 7). With HyperDSC
detection limits < 1% (w/w) can be reached in order
to quantify the amorphous material in pharmaceutical
compounds.1
At such heating rates, thermal gradients within the sample
can have an impact, widening the signals detected and
superimposing thermal phenomena inside the material. One
way to limit this thermal gradient is to perform analyses on
samples of very small mass, less than a milligram.2
1.478
Glass Transition of Polypropylene
1.4
1.6 mg sample
150 ˚C/min
1.0
Form a/form b at 95/5%
Area = 18.671 mJ
Delta H = 124.4722 J/g
2.0
1.5
Limit (˚C) Area (mJ) Percent (%)
160.5317.610 94.32
176.2218.671 100.00
1.0
0.5
0.0
X1 = 130.00 ˚C
Y1 = 0.0839 mW
-0.5
110
120
150
160
170
180
190
200
Transition separation by HyperDSC
Glass transition and evaporation
This amorphous active ingredient has a high water content
(about 10%) whose evaporation conceals glass transition during
a DSC analysis under conventional conditions (5 °C/min). An
experiment at 300 °C/min pushes water evaporation up to high
temperatures, making it possible to detect glass transition, as
is shown in Figure 9. The decomposition of the compound is
clearly identified after evaporation.
Vaporization H2O
50
50 ˚C/min
0.6
40
10 ˚C/min
0.4
-10
-5
0
5
10
Temperature (˚C)
Figure 7. Detection of a Tg at different heating rates.
15
20
25
Tg
Decomposition
30
0.2
-15.64
140
Figure 8. Detection of 5% active ingredient from b by HyperDSC.
0.8
0.09019
130
X2 = 176.22 ˚C
Y2 = 0.1647 mW
Temperature (˚C)
29.42
Heat Flow Endo Up (mW)
Heat Flow Endo Up (mW)
1.2
2.5
Heat Flow Endo Up (mW)
-494.7
20
Tg: Half Cp
Extrapolated = 68.92 ˚C
10
0
-10
-20
-30
25
50
100
150
200
250
300
350 370
Temperature (˚C)
Figure 9. Separation of events by HyperDSC.
3
Melting and decomposition
Melting
50.59
Figure 10 shows the capability of HyperDSC to distinguish
melting from the chemical decomposition of the molecule.
Decomposition
Peak = 232.04 ˚C
40
Heat Flow Endo Up (mW)
30
20
Similar for forms B and D, two polymorphs of another
component, HyperDSC clearly separates the melting
endotherms of each form from the decomposition exotherm
of the active ingredient as shown in Figure 11. It should be
noted that the melting temperatures and enthalpies for the
two polymorphs are different while the decomposition profile
is the same at higher temperature.
5 ˚C/min
10
300 ˚C/min
Onset = 223.42 ˚C
0
-10
-20
-30
-32.61 160
180
200
220
240
260
280
300
320
340
360
380 394.2
Temperature (˚C)
Figure 10. HyperDSC detection of the melting endotherm before decomposition
of the active ingredient.
40
Peak = 208.70 ˚C
35
Area = 44.575 mJ
Delta H = 55.1668 J/g
Heat Flow Endo Up (mW)
Onset = 196.84 ˚C
Form D
Peak = 186.61 ˚C
20
15
Onset = 171.69 ˚C
Form B
10
Area = 30.501 mJ
Delta H = 32.9388 J/g
5
0
-5
-10
20
50
100
150
The analyzed lyophilisate is made up half of glycine and half of
an active ingredient in amorphous form. Figure 12 shows how
HyperDSC contributes to the detection of the glass transition,
in comparison with conventional DSC.
Summary
30
25
Enhanced sensitivity for a Lyophilisate
200
250
300
350 370
Temperature (˚C)
Figure 11. HyperDSC detection of the melting endotherm before decomposition
of the active ingredient.
HyperDSC is an excellent tool to enhance material
characterization. It provides multiple benefits in the
pharmaceutical analysis. As demonstrated – HyperDSC
increases the sensitivity of DSC analysis, separates transitions
and allows better evaluation of polymorphic materials. The
analysis time is very short and permits you to increase your
sample throughput or to use this technique as a screening tool
for new materials.
References
1.M. Hurtta; I. Pitkänen; Thermochimica Acta 419 (2004)
19-29.
45
Increased Sensitivity
2.Pijpers, Thijs F.J.; Mathot, Vincent B.F.; Goderis, Bart;
Scherrenberg, Rolf L.; van der Vegte, Eric W.; Macromolecules
2002, 35, 3601-3613.
40
Heat Flow Endo Up (mW)
35
300 ˚C/min
30
25
Tg: Half Cp Extrapolated = 75.63 ˚C
20
Delta Cp =
0.933 J/g* ˚C
5 ˚C/min
15
10
0
20
40
60
80
100
120
140
160
180
200
220 225
Temperature (˚C)
Figure 12. Comparison of different scannig rates on a lyophylisate.
PerkinElmer, Inc.
940 Winter Street
Waltham, MA 02451 USA
P: (800) 762-4000 or
(+1) 203-925-4602
www.perkinelmer.com
For a complete listing of our global offices, visit www.perkinelmer.com/ContactUs
Copyright ©2006-2012, PerkinElmer, Inc. All rights reserved. PerkinElmer® is a registered trademark of PerkinElmer, Inc. All other trademarks are the property of their respective owners.
007522A_01