Summary: | 博士 === 國立中央大學 === 化學工程學系 === 85 === The morphological effect on the soft segment thermal
transition of a series of 4, 4''-diphenylmethane diisocyanate
(MDI) based cis-polybutadiene (cis-PB) polyurethanes was studied
by differential scanning calorimetry and small-angle X-ray
scattering (SAXS). The cis-PB polyurethanes were synthesized by
two-step solution polymerization. Number average molecular
weight of the soft segments ranged from 1000 to 4200. The hard
segment content ranged from 20 to 69%. For the cis-polybutadiene
soft segment, thermal transition behaviors of the glass
transition temperature (Tg), the breadth of the glass transition
temperature (△B), and the specific heat capacity (△Cp) were
investigated. Experimental results indicated that increases in
Tg and △B that corresponded to a decrease in the soft segment
molecular weight are attributed to decrease in free volume and
mobility due to bonding of soft segments to the rigid urethane
blocks. In each series of polyurethanes with a constant soft
segment molecular weight, both Tg and △B remain constant as
hard segment content increased from 20 to 69%, indicating the
phase separation between the soft and the hard segments is
completed even down to the interfacial region. This well-defined
two phase system simplifies the study on the soft segment △Cp.
By definition, △Cp is the heat capacity change per unit mass of
soft segment at the glass transition temperature. Thus, △Cp
relates to the entropy increase, △Q/T, for specific amount of
soft segment at glass transition temperature. Since a soft
segment chain''s conformational freedom can be restricted by the
nearby hard segment phase, an entropy restriction model was
proposed to account for the effect of morphological changes on
△Cp. This effect can easily cause a 50% decrease in △Cp for
conventional polyurethanes with a soft segment molecular weight
and hard segment content ranging from 1500-2000 and 30 to 55%,
respectively. Since △Cp was the most important property in
quantitatively accessing the degree of phase separation in
segment polyurethane systems, neglecting to consider entropy
restriction effect would indicate a serious inadequacy to
accurately characterize a phase separated polyurethane system.
The lack of complete glass transition thermograms for segment
polyurethanes with wide range of composition has given rise to
controversies on the thermal property and phase composition for
half a century. By using a new series of complete phase
separated polyurethanes and under melt-quenched condition, this
study indicates that the glass transition temperature (Tg) of
the amorphous hard phase is highly dependent on both the soft
segment and hard segment molecular weights, whereas the
amorphous soft phase Tg is only slightly dependent on its own
molecular weight. On the other hand, the specific heat capacity
changes (△Cp) in either phase are dependent both on segmental
length and morphology. These results not only throw fresh light
on the behavior of polyurethane''s glass transition, but also add
to fundamental understanding of the thermal properties of
amorphous polymeric nanomaterials that have a domain size much
smaller than the critical molecular weight between entanglement.
The long-term confusing glass transition temperature, Tg, and
specific heat capacity, △Cp, of the 4,4''-diphenylmethane
diisocyanate (MDI) and 1,4-butanediol (BD) hard segment in
segment polyurethanes have been clearly evidenced by using a new
series of completely phase mixed polyurethanes while
extrapolating to 100% hard segment content. The hard segment Tg
and △Cp thus obtained are 108℃ and 0.38 J/g℃, respectively,
which are equivalent to the reported Tg and △Cp of a high
molecular weight MDI-BD homopolymer. In addition, if a single
homogeneous phase is present, the observed △Cp and Tg are found
to be given by the linear weighted combination of the pure
constituent values. This provide a simple relationship to access
the composition of the individual microdomain in segment
polyurethanes, providing that the phase separate morphology does
not affect the microdomain glass transition behavior.
|