Chemistry Reference
In-Depth Information
Figure 4.12.
1
H
T
1ρ
values measured by SSNMR as a function of temperature for a 1:1 co-melt of
nifedipine and PVP (open squares) and a 1:1 co-melt of indomethacin and PVP (closed
diamonds). (Reproduced with permission from Ref. 140.)
cant change in
1
H
T
1
ρ
times is observed near to the observed
T
g
for each system. The nifedipine
DSC [139]. The results are shown in Figure 4.12, where a signi
-
PVP dispersion
shows signi
cant increases in mobility starting at 50
C, well below the measured
T
g
. Such
°
a
nding is consistent with themobility expected above theKauzmann temperature
T
K
. The
ability to detect changes in mobility below the measured
T
g
(e.g., to assess relaxation
behavior) may also be useful for studies of amorphous solid dispersions prepared by
different processes. The pulse sequence used tomeasure
1
H
T
1
ρ
times involves a simple 90
°
RF pulse on the
1
H channel followed by a variable-duration spin-lock pulse and direct
1
H
signal acquisition under moderate spinning rate MAS conditions. As with heteronuclear-
detected experiments, a pseudo-2D data set is obtained, usually involving 8
-
16 spin-lock
fitted to obtain
1
H
T
1
ρ
times, and
the experiment is repeated at different temperatures. Because of the sensitivity of
1
H
detection, each
1
H
T
1
ρ
experiment can be completed in a matter of minutes, allowing a full
temperature pro
pulse time increments. The resulting exponential decay is
le such as that shown in Figure 4.12 to be built in a reasonable time.
Finally, the use of
13
C as a detection nucleus for measurement of
1
H
T
1
,
1
H
T
1
ρ
,
13
C
T
1
,or
13
C
T
1
ρ
relaxation times, which is usually seen in the literature to date, often results
in long experimental durations because of the need to collect several hundred or thousand
transients for each of 8
16 time points. Sensitivity is limited because the drug
concentration in many dispersions of interest ranges from 10 to 50% (w/w). In some
cases, more sensitive nuclei can be employed. For example,
-
19
F-detected
1
H
T
1
measurements were employed for the ef
cient study of a dispersion containing 15%
(w/w) voriconazole, while
13
C-detected
1
H
T
1
measurements were used to separately
obtain values for the high-concentration polymer [119]. Several different relaxation
times of a solid dispersion of ezetimibe in mesoporous silica were observed via the
19
F
nucleus, allowing for more sensitive experiments [122].
Dynamic nuclear polarization (DNP) using microwave irradiation of biradicals
inserted into a sample is an emerging technique that can greatly enhance the sensitivity of
SSNMR experiments for less sensitive nuclei [140]. In the past, DNP was applied to
