The present study investigates the effect of the active
chain density
and cross-link length on the Mullins softening in natural rubber by
using a wide range of vulcanization systems (conventional, semiefficient,
and efficient). For that purpose, sulfur [S] and accelerator [A] contents,
as well as their ratio [A]/[S], have been varied. Different vulcanization
conditions, corresponding to various times and temperatures, have
also been tested. In order to investigate how the fillers affect the
relationship between the rubber network and the softening, the tests
were conducted with the same natural rubber, unfilled and filled with
carbon black. The active chain density was determined from the equilibrium
swelling test by applying the Flory–Rehner relationship and
by removing the effect of the insoluble components. The Mullins softening
was evaluated from sets of uniaxial tension cycles by increasing maximum
stretches. The mechanical response showed that Mullins softening in
unfilled NR is significantly impacted by the active chain density.
The higher the active chain density, the higher the softening. Furthermore,
our results confirm that strain-induced crystallization is necessary
to induce Mullins softening. The effect of the cross-link length was
found to be negligible. In the filled NR, the addition of fillers
drastically increases the Mullins softening and is the first effect
compared to that of the active chain density. Moreover, the vulcanization
conditions tested were found to have no significant effect on the
softening. This study gathers a large amount of data on the effects
of the rubber network and fillers on the Mullins softening and therefore
provides information on the importance of better modeling the softening
with physically motivated approaches, especially those inspired from
the molecular network description.