by [25] on silica-filled SBR rubbers, by modifying the silica surface using different
silanes. Their results showed that the presence of chemical links at the silica surface
reduces the amplitude of the effect. Based on that result, the authors attempted to
provide a theoretical explanation of the Payne effect: the polymer chains are initially
bound to the filler surface, but under the applied stress, they de-bond, thus causing
the nonlinear mechanical behaviour. The de-bonding effect has also been used in a
recently proposed model for the explanation of the nonlinear, large-strain mechanical behaviour of nanofilled elastomers [41].
The majority of the above-mentioned studies have attempted to explain the Payne
effect by focusing on the importance of the polymer-filler interactions and the
development of glassy bridges between the nanoparticles. Indeed, the behaviour of
a polymer under conditions of strong confinement is different than that in the bulk of
the material. However, the assumed increase of rigidity due to overlapping glassy
layers at high filler volume fractions, and its consequent drop under deformation, has
not been verified experimentally. The difficulty arises due to the minuscule length
scales that are involved: the approaching surfaces of neighbouring filler particles
may result in a geometric confinement with a size on the scale of only a few
nanometres, which is difficult to probe experimentally. Therefore, in order to provide
insight on the nanoscopic mechanisms which might affect the mechanical properties
of polymer nanocomposites, it is important to understand the dynamical and
mechanical behaviour of polymer films – strongly confined systems [36].
Various attempts to study the influence of the degree of confinement on the
mechanical properties of polymer nanocomposites have deployed a model polymer
film, aiming to establish a quantitative equivalence between the thermomechanical
properties of the two systems [99]. Indeed, the properties of both systems are
strongly influenced by adhesive interactions and by confinement effects, and,
depending on the volume fraction of fillers in the nanocomposite, similarities have
been identified. In highly filled polymer nanocomposites (40 wt%), it has been
shown [36] that the changes in the glass transition temperature with decreasing
interparticle spacing are quantitatively equivalent to the corresponding thin-film
data. On the other hand, in materials with low filler concentration (less than 1.0 wt
%), only a qualitative equivalence has been established [36, 100]. The glass transition of thin polymer films has been also widely examined experimentally [101–
114]. The general conclusion thus far is that the glass transition of a polymer film
(and by extension its dynamic response to external perturbations) depends mainly on
the degree of confinement, the presence of free interfaces and the polymer-wall
interactions. However, due to the fact that the measured properties seem to be
influenced by the employed experimental technique as well as by the preparation
procedure of the samples, the reported results have shown disagreement among
different laboratories and among different experimental methods [115–
122]. Dynamic fragility (a measure of the glass transition abruptness of glassforming materials) has also been employed in order to explain why different
118
C. Raffaelli et al.
silanes. Their results showed that the presence of chemical links at the silica surface
reduces the amplitude of the effect. Based on that result, the authors attempted to
provide a theoretical explanation of the Payne effect: the polymer chains are initially
bound to the filler surface, but under the applied stress, they de-bond, thus causing
the nonlinear mechanical behaviour. The de-bonding effect has also been used in a
recently proposed model for the explanation of the nonlinear, large-strain mechanical behaviour of nanofilled elastomers [41].
The majority of the above-mentioned studies have attempted to explain the Payne
effect by focusing on the importance of the polymer-filler interactions and the
development of glassy bridges between the nanoparticles. Indeed, the behaviour of
a polymer under conditions of strong confinement is different than that in the bulk of
the material. However, the assumed increase of rigidity due to overlapping glassy
layers at high filler volume fractions, and its consequent drop under deformation, has
not been verified experimentally. The difficulty arises due to the minuscule length
scales that are involved: the approaching surfaces of neighbouring filler particles
may result in a geometric confinement with a size on the scale of only a few
nanometres, which is difficult to probe experimentally. Therefore, in order to provide
insight on the nanoscopic mechanisms which might affect the mechanical properties
of polymer nanocomposites, it is important to understand the dynamical and
mechanical behaviour of polymer films – strongly confined systems [36].
Various attempts to study the influence of the degree of confinement on the
mechanical properties of polymer nanocomposites have deployed a model polymer
film, aiming to establish a quantitative equivalence between the thermomechanical
properties of the two systems [99]. Indeed, the properties of both systems are
strongly influenced by adhesive interactions and by confinement effects, and,
depending on the volume fraction of fillers in the nanocomposite, similarities have
been identified. In highly filled polymer nanocomposites (40 wt%), it has been
shown [36] that the changes in the glass transition temperature with decreasing
interparticle spacing are quantitatively equivalent to the corresponding thin-film
data. On the other hand, in materials with low filler concentration (less than 1.0 wt
%), only a qualitative equivalence has been established [36, 100]. The glass transition of thin polymer films has been also widely examined experimentally [101–
114]. The general conclusion thus far is that the glass transition of a polymer film
(and by extension its dynamic response to external perturbations) depends mainly on
the degree of confinement, the presence of free interfaces and the polymer-wall
interactions. However, due to the fact that the measured properties seem to be
influenced by the employed experimental technique as well as by the preparation
procedure of the samples, the reported results have shown disagreement among
different laboratories and among different experimental methods [115–
122]. Dynamic fragility (a measure of the glass transition abruptness of glassforming materials) has also been employed in order to explain why different
118
C. Raffaelli et al.