phase as follows: the burst phase was estimated
by calculating AbsCyAbsA .rwAbsByAbsC.q
AbsCyAbsA .x where Abs denotes the absor-
bance spectrum.. We first took the spectrum be-
fore the conformation transition AbsA ., then injected
ethanol]D2O mixture to allow the conformation
transition to begin. From the series of
dynamic spectra, we chose the last spectrum ;60
min. and the first spectrum ;0.1 min. to get
AbsB and AbsC. Using these spectra as a basis
for calculation, we can ensure that the sample
position had minimal changes during the measurement
and also ensure the membrane thickness
changes were minimal. Afterwards, using
baseline corrections to correct background dif-
ferences., we subtracted AbsByAbsC same
background, small error. and AbsCyAbsA dif-
ferent background, larger relative error. to calculate
the burst phase. This analysis indicates a
burst phase of 50"10%.
3.2. Time-resol¨ed FTIR spectra collected during the
conformation transition process of regenerated B.
mori silk fibroin membrane induced by ethanol
Fig. 2 shows the infrared absorbance spectra of
the regenerated B. mori silk fibroin membrane
during the conformation transition after the addition
of ethanol, from 0.1 to 60 min. With increasing
time, characteristic changes in the absorbance
spectra are seen, i.e. the amide I band gradually
changes from ;1660 cmy1 to ;1620 cmy1.
These time-resolved changes are emphasized in
the difference spectra Fig. 3. which have been
calculated according to At]A0 , where At are the
absorbance spectra at a running time and A0 is
the first absorbance spectrum, i.e. the one
recorded at 0.0975 min Fig. 1c.. Note that in the
difference spectra, the main difference features
are identical with those identified by the second
derivative analysis in Fig. 1b,c. The increasing
positive band at 1618 cmy1 reflects the increasing
b-sheet structure that develops with time after
addition of ethanol. Correspondingly, the one