136
L. V. Karabanova et al.
a consequence of interaction of the PU chains with the surface of nanotubes and
restriction of glass transition dynamics, i.e., prevailing role of interfacial dynamics
(“constrained dynamics” effect, see for example reviews [29, 41]). However, this
peak may also retain the initial temperature position and even slightly shift to
lower temperatures indicating acceleration of glass transition dynamics. As for the
shoulder of tan δ peak located in the region of −40 to −50 ◦ C, its contribution to the
intensity decreases in the composites’ spectra, but its temperature position remains
almost unchangeable. It means that the considerable part of the PU nanoregions is
not disturbed by the addition of MWCNTs. As the segments contributing to lowtemperature relaxations are definitely soft poly(propylene glycol) ones, we could
expect that nanotubes are concentrated close to the hard nanodomains of PU matrix
since chemical reaction of functionalised MWCNTs with isocyanate groups occurs.
The impact of nanotubes on the PU dynamics turned out to be minimal for the
nanocomposite with MWCNT-hemin (Fig. 8.9a). After filling with 0.1% nanotubes,
the main tan δ maximum displaces from +10 to −10 ◦ C, and the temperature
position of the other constituent relaxations remains almost unchangeable; this is
due apparently to some loosening of the molecular packing in PU. Introducing of
0.25% MWCNT-hemin results in the restoration of the temperature position of the
main tan δ maximum, a small decreasing mechanical losses over −40 to +20 ◦ C
range, but a slight increasing of mechanical losses in the temperature region of the
higher-temperature shoulder. The hemin-modified nanotube surface is characterized
with the highest content of functional COOH groups, which act as points for
covalent PU attaching (Table 8.1); however, the hemin coating is immobilized on
the nanotube surface only through non-covalent, van der Waals interactions with the
π electron network of the MWCNTs; this led to decreased interfacial interactions.
After incorporating MWCNT-hemin-red into PU, similar effects of some relaxation peak displacement to lower temperatures (loosening molecular packing and
acceleration of dynamics) are observed with nanofiller contents of 0.01 or 0.1%
(Fig. 8.9b). At the same time, using 0.25% of nanotubes resulted in substantial
constraining dynamics: a plateau at 20–60 ◦ C arises. The latter effect is registered
despite the total disappearance of carboxyl and phenol functional groups as
“covalent bonding points” at MWCNT-hemin-red surface (Table 8.1). This may be
explained presumably by the above-mentioned fact that aliphatic alcohol groups,
undetectable via the titrations used, are likely to be present in this case on the
MWCNT surface serving as such anchoring sites.
The most dramatic changes are registered, however, in the mechanical loss
spectrum after filling PU with MWCNT-ox (Fig. 8.9c) which possesses a high total
number of covalently bound surface carboxylic and phenol groups (Table 8.1) and
perhaps also undetectable at titration alcohol groups generated by acid oxidation.
The tan δ (T) contour slightly changes at 0.01% nanotubes, but it is strongly
transformed at 0.1 and especially 0.25% nanotubes: the low-temperature shoulder
and main peak II decrease, whereas the high-temperature shoulder increases by far
with the displacement to higher temperatures. The plateau covering the temperature
range from – 60 to +80 ◦ C rises at 0.1% nanofiller, which indicates that a good
dispersion of MWCNT-ox in PU matrix [26] led to the formation of a large
L. V. Karabanova et al.
a consequence of interaction of the PU chains with the surface of nanotubes and
restriction of glass transition dynamics, i.e., prevailing role of interfacial dynamics
(“constrained dynamics” effect, see for example reviews [29, 41]). However, this
peak may also retain the initial temperature position and even slightly shift to
lower temperatures indicating acceleration of glass transition dynamics. As for the
shoulder of tan δ peak located in the region of −40 to −50 ◦ C, its contribution to the
intensity decreases in the composites’ spectra, but its temperature position remains
almost unchangeable. It means that the considerable part of the PU nanoregions is
not disturbed by the addition of MWCNTs. As the segments contributing to lowtemperature relaxations are definitely soft poly(propylene glycol) ones, we could
expect that nanotubes are concentrated close to the hard nanodomains of PU matrix
since chemical reaction of functionalised MWCNTs with isocyanate groups occurs.
The impact of nanotubes on the PU dynamics turned out to be minimal for the
nanocomposite with MWCNT-hemin (Fig. 8.9a). After filling with 0.1% nanotubes,
the main tan δ maximum displaces from +10 to −10 ◦ C, and the temperature
position of the other constituent relaxations remains almost unchangeable; this is
due apparently to some loosening of the molecular packing in PU. Introducing of
0.25% MWCNT-hemin results in the restoration of the temperature position of the
main tan δ maximum, a small decreasing mechanical losses over −40 to +20 ◦ C
range, but a slight increasing of mechanical losses in the temperature region of the
higher-temperature shoulder. The hemin-modified nanotube surface is characterized
with the highest content of functional COOH groups, which act as points for
covalent PU attaching (Table 8.1); however, the hemin coating is immobilized on
the nanotube surface only through non-covalent, van der Waals interactions with the
π electron network of the MWCNTs; this led to decreased interfacial interactions.
After incorporating MWCNT-hemin-red into PU, similar effects of some relaxation peak displacement to lower temperatures (loosening molecular packing and
acceleration of dynamics) are observed with nanofiller contents of 0.01 or 0.1%
(Fig. 8.9b). At the same time, using 0.25% of nanotubes resulted in substantial
constraining dynamics: a plateau at 20–60 ◦ C arises. The latter effect is registered
despite the total disappearance of carboxyl and phenol functional groups as
“covalent bonding points” at MWCNT-hemin-red surface (Table 8.1). This may be
explained presumably by the above-mentioned fact that aliphatic alcohol groups,
undetectable via the titrations used, are likely to be present in this case on the
MWCNT surface serving as such anchoring sites.
The most dramatic changes are registered, however, in the mechanical loss
spectrum after filling PU with MWCNT-ox (Fig. 8.9c) which possesses a high total
number of covalently bound surface carboxylic and phenol groups (Table 8.1) and
perhaps also undetectable at titration alcohol groups generated by acid oxidation.
The tan δ (T) contour slightly changes at 0.01% nanotubes, but it is strongly
transformed at 0.1 and especially 0.25% nanotubes: the low-temperature shoulder
and main peak II decrease, whereas the high-temperature shoulder increases by far
with the displacement to higher temperatures. The plateau covering the temperature
range from – 60 to +80 ◦ C rises at 0.1% nanofiller, which indicates that a good
dispersion of MWCNT-ox in PU matrix [26] led to the formation of a large