5.5 Nanofiller Clustering as Revealed by Synchrotron Radiation
95
In this figure, structures from an atomic level at a to bulk in rubber matrix at g are
displayed. The a and b show molecular structure of rubber. The c in the figure shows
primary CB particle of r TEM (radius determined by TEM) = 13 nm and of D s (surface
fractal) = 2.6, which suggests that practically the surface cannot be a perfect sphere
as shown in Fig. 5.3 at the left-hand side. The d is the primary aggregate consisting
of nine (maximum) CB particles whose radius, R a , found maximally 27 nm, too. The
e is named agglomerate at level 1 and is assumed to be a unit of CB dispersion in
rubber, which is consisting of a few primary aggregates (W is the number): W = 2
for SBR, and W = 4 for IR. The R a was 27 nm, and hence, radius of gyration of
the level 1 agglomerate R d is W × 27 nm. The f shows an agglomerate at level 2
consisting of several dispersion units, i.e., level 1 agglomerates. Its mass fractal was
determined as D m = 2.3, and hence, they also named as ‘mass fractal’ structure. The
size of mass fractal, l u , was found 10 µm for SBR and 20 µm was a minimum value
for IR. The g shows the dispersion of the mass fractal in rubber matrix.
These results are at least qualitatively, or even semiquantitatively, in compatible
with our models shown in Figs. 5.3 and 5.4. Their aggregate is equivalent to our
primary aggregate, level 1 agglomerate is equivalent to our higher aggregates, and
their mass fractal may be equivalent to our agglomerate since both are assumed to be
the ultimate of structuring CB in rubber matrix. However, one of the most important
failures of them is the neglect of the filler-to-rubber interaction, which in this case is
bound rubber. This failure is due to using the scattering technique, not due to their
mistake. In the e, rubber molecules were in the picture, and the black points on the
CB surface seem to suggest their claim of chemical bonding of rubber onto the CB
surface. However, the effect of rubber-to-filler interaction was not discussed at all,
even though they had recognized a few differences between SBR and IR. This may
be one of the reasons for their missing of the nanofiller network structure in their
view on the filler clustering, contrary to our proposition by Figs. 5.5 and 5.6, which
are based on 3D-TEM. Their loading amount of 20 vol% may be another reason for
their failure of observing the CB network structure in SBR and IR. At the higher
loadings, they might have observed the gelation as a whole. Or, they did not think
of the nanofiller network at all. Instead, they have assumed the mass fractal being
equivalent to the pregel state like as CB agglomerate, which is their final morphology
of CB and is not equivalent to the CB network. If so, this is another difference from
our result showing the gelation, i.e., a whole CB network formation in NR matrix at
higher CB loading.
Hagita et al. reported the results of 2D-USAXS (ultrasmall-angle X-ray scattering) and 2D-SAXS measurements on monodispersed particulate silicas (radii, 100
and 300 nm) in SBR [70, 71]. The measurements were dynamic ones, i.e., the samples were under uniaxial deformation and used 2D RMC (reverse Monte Carlo)
method to visualize the position of silica particles in rubber matrix upon elongation.
From the results, it is suggested that the particles which have been connected before
the deformation stay unchanged even under the higher elongation. The majority of
deformation seems to be due to the deformation of the rubber phase not that of the
silica aggregates. Their results on silica/SBR are apparently against our results on
95
In this figure, structures from an atomic level at a to bulk in rubber matrix at g are
displayed. The a and b show molecular structure of rubber. The c in the figure shows
primary CB particle of r TEM (radius determined by TEM) = 13 nm and of D s (surface
fractal) = 2.6, which suggests that practically the surface cannot be a perfect sphere
as shown in Fig. 5.3 at the left-hand side. The d is the primary aggregate consisting
of nine (maximum) CB particles whose radius, R a , found maximally 27 nm, too. The
e is named agglomerate at level 1 and is assumed to be a unit of CB dispersion in
rubber, which is consisting of a few primary aggregates (W is the number): W = 2
for SBR, and W = 4 for IR. The R a was 27 nm, and hence, radius of gyration of
the level 1 agglomerate R d is W × 27 nm. The f shows an agglomerate at level 2
consisting of several dispersion units, i.e., level 1 agglomerates. Its mass fractal was
determined as D m = 2.3, and hence, they also named as ‘mass fractal’ structure. The
size of mass fractal, l u , was found 10 µm for SBR and 20 µm was a minimum value
for IR. The g shows the dispersion of the mass fractal in rubber matrix.
These results are at least qualitatively, or even semiquantitatively, in compatible
with our models shown in Figs. 5.3 and 5.4. Their aggregate is equivalent to our
primary aggregate, level 1 agglomerate is equivalent to our higher aggregates, and
their mass fractal may be equivalent to our agglomerate since both are assumed to be
the ultimate of structuring CB in rubber matrix. However, one of the most important
failures of them is the neglect of the filler-to-rubber interaction, which in this case is
bound rubber. This failure is due to using the scattering technique, not due to their
mistake. In the e, rubber molecules were in the picture, and the black points on the
CB surface seem to suggest their claim of chemical bonding of rubber onto the CB
surface. However, the effect of rubber-to-filler interaction was not discussed at all,
even though they had recognized a few differences between SBR and IR. This may
be one of the reasons for their missing of the nanofiller network structure in their
view on the filler clustering, contrary to our proposition by Figs. 5.5 and 5.6, which
are based on 3D-TEM. Their loading amount of 20 vol% may be another reason for
their failure of observing the CB network structure in SBR and IR. At the higher
loadings, they might have observed the gelation as a whole. Or, they did not think
of the nanofiller network at all. Instead, they have assumed the mass fractal being
equivalent to the pregel state like as CB agglomerate, which is their final morphology
of CB and is not equivalent to the CB network. If so, this is another difference from
our result showing the gelation, i.e., a whole CB network formation in NR matrix at
higher CB loading.
Hagita et al. reported the results of 2D-USAXS (ultrasmall-angle X-ray scattering) and 2D-SAXS measurements on monodispersed particulate silicas (radii, 100
and 300 nm) in SBR [70, 71]. The measurements were dynamic ones, i.e., the samples were under uniaxial deformation and used 2D RMC (reverse Monte Carlo)
method to visualize the position of silica particles in rubber matrix upon elongation.
From the results, it is suggested that the particles which have been connected before
the deformation stay unchanged even under the higher elongation. The majority of
deformation seems to be due to the deformation of the rubber phase not that of the
silica aggregates. Their results on silica/SBR are apparently against our results on