5.5 Nanofiller Clustering as Revealed by Synchrotron Radiation
97
subsection, further deductive reasoning on our semiflexible nanofiller network structure, which has mainly been estimated by an inductive reasoning from the 3D-TEM
observations, is to be presented as the concluding remarks of this chapter.
5.6 Semiflexible Nanofiller Networks with Bound Rubber
and Proposal of a Tentative Rheological Model
As shown in Fig. 4.16a, the nearest distance, d p , converges to a value of ca. 3 nm at the
larger amount of CB than 40 phr. It means that the increase of CB amount more than
40 phr does not influence on the distance, which is actually a bound rubber thickness
at relatively high loadings of CB. Together with d p value of ca. 1.3 nm for particulate
silica, it is notable that 3D-TEM has uniquely afforded bound rubber thickness in
spite of the fact that electron beam does not recognize the presence of the specific
rubber layer around the particle surface at all. In this connection, X-ray scattering
techniques do not manage to do it, too, and NMR and AFM techniques have been
mainly employed for the recognition. This uniquely observed thickness of rubber is
named as CB/NR interaction layer (CNIL) [49, 57]. CNIL is reasonably assumed to
be the distance between the nearest CB particles at the high dense packing of CB,
while the bound rubber on CB surface in general is thicker than 3 nm.
Taking the results so far obtained and the idea of CNIL into account, the following
way of expression of the scenario (see the Sect. 5.4) is proposed on the clustering or
structuring of primary CB aggregates in the rubber matrix:
The CB aggregate initially grows in size via CNIL formation, and around the CB content of
20 phr the growth becomes accompanied by the cross-linking between the higher aggregates
again via CNIL. Between 20 phr and 40 phr CB contents, CB network formation process
become more and more dominant due to the dense packing of CB up to gelation or percolation,
and finally a semi-flexible nanofiller network is formed covering the whole system as the
agglomerate.
In this expression, CNIL is given the role of mediating the aggregation and
cross-linking, and it is responsible for the semiflexibility of the resultant nanofiller
networks.
Figure 5.11 shows 3D-TEM images of CNIL in the CB-loaded NR vulcanizates
whose bound rubber thicknesses are set constant at 3 nm. Note that Fig. 5.11 is
derived from Fig. 4.14 as well as Fig. 4.18. The images show the CB aggregates
consisting of CB and CNIL. The gelation (CB network formation) scheme stated
just above is not irrational when comparison of the images with those in Fig. 4.18 is
made and taking the changes of CNILs from CB10 to CB80 in Fig. 5.11 into account.
The size of CNIL of CB20 is found to be almost equal to the phase-delay region in
the AFM data [49].
Figure 5.12 shows CB amount dependence of dynamic storage modulus (G’) of
the CB-loaded NR vulcanizates measured at 293 K [49]. W CB is the loaded amount
of CB in phr. G’ increases nonlinearly with the increase of W CB . In accordance with
97
subsection, further deductive reasoning on our semiflexible nanofiller network structure, which has mainly been estimated by an inductive reasoning from the 3D-TEM
observations, is to be presented as the concluding remarks of this chapter.
5.6 Semiflexible Nanofiller Networks with Bound Rubber
and Proposal of a Tentative Rheological Model
As shown in Fig. 4.16a, the nearest distance, d p , converges to a value of ca. 3 nm at the
larger amount of CB than 40 phr. It means that the increase of CB amount more than
40 phr does not influence on the distance, which is actually a bound rubber thickness
at relatively high loadings of CB. Together with d p value of ca. 1.3 nm for particulate
silica, it is notable that 3D-TEM has uniquely afforded bound rubber thickness in
spite of the fact that electron beam does not recognize the presence of the specific
rubber layer around the particle surface at all. In this connection, X-ray scattering
techniques do not manage to do it, too, and NMR and AFM techniques have been
mainly employed for the recognition. This uniquely observed thickness of rubber is
named as CB/NR interaction layer (CNIL) [49, 57]. CNIL is reasonably assumed to
be the distance between the nearest CB particles at the high dense packing of CB,
while the bound rubber on CB surface in general is thicker than 3 nm.
Taking the results so far obtained and the idea of CNIL into account, the following
way of expression of the scenario (see the Sect. 5.4) is proposed on the clustering or
structuring of primary CB aggregates in the rubber matrix:
The CB aggregate initially grows in size via CNIL formation, and around the CB content of
20 phr the growth becomes accompanied by the cross-linking between the higher aggregates
again via CNIL. Between 20 phr and 40 phr CB contents, CB network formation process
become more and more dominant due to the dense packing of CB up to gelation or percolation,
and finally a semi-flexible nanofiller network is formed covering the whole system as the
agglomerate.
In this expression, CNIL is given the role of mediating the aggregation and
cross-linking, and it is responsible for the semiflexibility of the resultant nanofiller
networks.
Figure 5.11 shows 3D-TEM images of CNIL in the CB-loaded NR vulcanizates
whose bound rubber thicknesses are set constant at 3 nm. Note that Fig. 5.11 is
derived from Fig. 4.14 as well as Fig. 4.18. The images show the CB aggregates
consisting of CB and CNIL. The gelation (CB network formation) scheme stated
just above is not irrational when comparison of the images with those in Fig. 4.18 is
made and taking the changes of CNILs from CB10 to CB80 in Fig. 5.11 into account.
The size of CNIL of CB20 is found to be almost equal to the phase-delay region in
the AFM data [49].
Figure 5.12 shows CB amount dependence of dynamic storage modulus (G’) of
the CB-loaded NR vulcanizates measured at 293 K [49]. W CB is the loaded amount
of CB in phr. G’ increases nonlinearly with the increase of W CB . In accordance with