dehydration. Therefore, the cryoprotectant solution is suboptimal
for slow freezing and storage at elevated subzero temperatures. The
frozen sample was warmed from À130
C at a rate of 3
C per min.
DSC analysis has revealed that rapidly frozen sample is unstable at
elevated subzero temperatures. First, after glass transition at
À121
C upon warming, devitrification is observed at approximately À110
C (T d ). The devitrification peak is very small as
compared to the ice-melting peak that occurs above À10
C, indicating that bulk ice formation happened during rapid cooling when
the sample is plunged into liquid nitrogen. Second, there is a broad
second-order transition between À90 and À50
C. This broad
transition shows the presence of nonhomogeneous amorphous
domains. The blood cell suspension has been frozen rapidly, resulting in the formation of multiple amorphous domains with varying
degrees of freeze concentration. During warming, glass transitions
of these domains overlapped one another, giving rise a continuous
baseline shift over a broad temperature range. Third, between À45
and À10
C just before ice melting, there are glass transition of a
less concentrated amorphous domain starting at À38
C, recrystallization in freeze-concentrated domains around À20 to À25
C,
and glass transition of the maximally freeze-concentrated domain at
À17
C. To study the stability of frozen cells at elevated subzero
temperatures, the authors stored the frozen sample at À40
C for
4 days before it was cooled again for DSC analysis at a warming rate
of 2
C/min (see Fig. 10 inset). Storage at À40
C eliminated all
thermal transitions below À40
C and also avoids recrystallization
between À25
C and À20
C. These changes (thermal instability)
have important implications to the survival of frozen cells. Devitrification (crystallization) is a major event damaging rapidly frozen
cells during slow warming and storage at elevated subzero
temperature.
In the system studied, devitrification around À110
C does not
cause significant damage to frozen blood cells. When frozen samples are transferred from À196 to À78
C for isothermal storage,
time-dependent hemolysis remains less than 4% within the first 6 h,
and maximum hemolysis has been reported to be 10–12% after
extended storage. Following the broad transition above À80
C,
the stability of the frozen cells declined significantly. Within this
temperature range, no particular DSC-detectable thermal event has
been linked to the instability of frozen erythrocytes. Slow recrystallization (undetectable by DSC) would occur at this temperature
range. The abrupt increase in hemolysis occurs at temperatures
above À20
C, which is associated with the glass transition of
freeze-concentrated amorphous domains and probably also to the
melting of ice.
DSC Analysis of Thermophysical Properties
299
for slow freezing and storage at elevated subzero temperatures. The
frozen sample was warmed from À130
C at a rate of 3
C per min.
DSC analysis has revealed that rapidly frozen sample is unstable at
elevated subzero temperatures. First, after glass transition at
À121
C upon warming, devitrification is observed at approximately À110
C (T d ). The devitrification peak is very small as
compared to the ice-melting peak that occurs above À10
C, indicating that bulk ice formation happened during rapid cooling when
the sample is plunged into liquid nitrogen. Second, there is a broad
second-order transition between À90 and À50
C. This broad
transition shows the presence of nonhomogeneous amorphous
domains. The blood cell suspension has been frozen rapidly, resulting in the formation of multiple amorphous domains with varying
degrees of freeze concentration. During warming, glass transitions
of these domains overlapped one another, giving rise a continuous
baseline shift over a broad temperature range. Third, between À45
and À10
C just before ice melting, there are glass transition of a
less concentrated amorphous domain starting at À38
C, recrystallization in freeze-concentrated domains around À20 to À25
C,
and glass transition of the maximally freeze-concentrated domain at
À17
C. To study the stability of frozen cells at elevated subzero
temperatures, the authors stored the frozen sample at À40
C for
4 days before it was cooled again for DSC analysis at a warming rate
of 2
C/min (see Fig. 10 inset). Storage at À40
C eliminated all
thermal transitions below À40
C and also avoids recrystallization
between À25
C and À20
C. These changes (thermal instability)
have important implications to the survival of frozen cells. Devitrification (crystallization) is a major event damaging rapidly frozen
cells during slow warming and storage at elevated subzero
temperature.
In the system studied, devitrification around À110
C does not
cause significant damage to frozen blood cells. When frozen samples are transferred from À196 to À78
C for isothermal storage,
time-dependent hemolysis remains less than 4% within the first 6 h,
and maximum hemolysis has been reported to be 10–12% after
extended storage. Following the broad transition above À80
C,
the stability of the frozen cells declined significantly. Within this
temperature range, no particular DSC-detectable thermal event has
been linked to the instability of frozen erythrocytes. Slow recrystallization (undetectable by DSC) would occur at this temperature
range. The abrupt increase in hemolysis occurs at temperatures
above À20
C, which is associated with the glass transition of
freeze-concentrated amorphous domains and probably also to the
melting of ice.
DSC Analysis of Thermophysical Properties
299