unloaded (red cycle) and (2) where the hydrogel was stretched to 200% strain and
unloaded (green cycle) and then immediately re-stretched to 400% strain
and unloaded (blue cycle). The dashed black cycle in Fig. 12 shows the loading
and unloading data for a fresh NF8 hydrogel sample stretched to 400% strain.
For each two-cycle experiment, the stress required to re-stretch the specimen to
either 300 or 400% strain after a prior loading and unloading cycle to 100 or 200%
strain, respectively, was significantly lower than for the initial loading and unloading
cycle. In both experiments, where the strain for the second cycle was greater than the
maximum strain used in the first cycle, the loading curve for the second cycle followed
the same path as the single 400% loading curve, i.e., the black dashed curve. However,
in the part of the second cycle when the strains were below the maximum strain used
in the first cycle, the stresses were considerably lower than the dashed loading curve.
That is, for subsequent loading and unloading cycles, the hydrogels were softer over
the strain range corresponding to the previous stress–strain history, but for strains
greater than the maximum strain previously achieved, the sample deforms exactly like
a fresh sample and is unaffected by the past mechanical history.
The Mullins effect is most commonly observed for filled elastomers and neat
elastomers that can strain-crystallize [67], but it has also been reported for unfilled
polyampholyte hydrogels [59]; microphase-separated thermoplastic elastomers [68];
DN hydrogels [64]; and clay [69], graphene [70], and silica [71] nanocomposite
hydrogels. Various explanations have been proposed for the origin of the Mullins
softening effect, including the detachment of chains from filler particles, the slippage
of chains bound to the reinforcing filler, the rupture of filler clusters, chain disentanglement (which requires either large-scale movement of the particles to which the
chains are bound or breakage and reformation of covalent or physical bonds), and
changes of a “double-layer” microstructure composed of carbon black aggregates
surrounded by polymer layers [67]. With the notable exceptions of the supramolecular NFx and polyampholyte hydrogels [59], one common observation for all the
other materials that exhibit a Mullins effect is that the phenomenon coincides with a
permanent deformation of the material. All of the mechanisms of stress softening
delineated above have in common either the breaking or jumping (i.e., breaking and
reforming) of a reversible bond or a microstructure change. For the supramolecular
hydrogels that mechanism may be the reversible breaking of the supramolecular
bonds or orientation of nanodomains. The ability of the supramolecular hydrogels to
fully or near-fully recover their dimensions with no or insignificant permanent
deformation suggests that reformation of the broken supramolecular bonds occurs
spatially at essentially the same positions in the gel where they were originally,
which may be a consequence of highly non-affine deformation – i.e., that the
microscopic deformation of the nanodomain microstructure is very small compared
with the macroscopic deformation. In that case, most of the deformation is provided
by changing conformations of the network chains, such that the deformation appears
to be nearly elastic. That reversibility is a notable advantage of these materials
compared with tough hydrogels that exhibit plastic deformation, such as DN
hydrogels. The small microscopic and large macroscopic deformation could be
achieved by intradomain jumping of the hydrophobic groups. The core of the
Hydrophobically Associating Hydrogels with Microphase-Separated Morphologies
187
unloaded (green cycle) and then immediately re-stretched to 400% strain
and unloaded (blue cycle). The dashed black cycle in Fig. 12 shows the loading
and unloading data for a fresh NF8 hydrogel sample stretched to 400% strain.
For each two-cycle experiment, the stress required to re-stretch the specimen to
either 300 or 400% strain after a prior loading and unloading cycle to 100 or 200%
strain, respectively, was significantly lower than for the initial loading and unloading
cycle. In both experiments, where the strain for the second cycle was greater than the
maximum strain used in the first cycle, the loading curve for the second cycle followed
the same path as the single 400% loading curve, i.e., the black dashed curve. However,
in the part of the second cycle when the strains were below the maximum strain used
in the first cycle, the stresses were considerably lower than the dashed loading curve.
That is, for subsequent loading and unloading cycles, the hydrogels were softer over
the strain range corresponding to the previous stress–strain history, but for strains
greater than the maximum strain previously achieved, the sample deforms exactly like
a fresh sample and is unaffected by the past mechanical history.
The Mullins effect is most commonly observed for filled elastomers and neat
elastomers that can strain-crystallize [67], but it has also been reported for unfilled
polyampholyte hydrogels [59]; microphase-separated thermoplastic elastomers [68];
DN hydrogels [64]; and clay [69], graphene [70], and silica [71] nanocomposite
hydrogels. Various explanations have been proposed for the origin of the Mullins
softening effect, including the detachment of chains from filler particles, the slippage
of chains bound to the reinforcing filler, the rupture of filler clusters, chain disentanglement (which requires either large-scale movement of the particles to which the
chains are bound or breakage and reformation of covalent or physical bonds), and
changes of a “double-layer” microstructure composed of carbon black aggregates
surrounded by polymer layers [67]. With the notable exceptions of the supramolecular NFx and polyampholyte hydrogels [59], one common observation for all the
other materials that exhibit a Mullins effect is that the phenomenon coincides with a
permanent deformation of the material. All of the mechanisms of stress softening
delineated above have in common either the breaking or jumping (i.e., breaking and
reforming) of a reversible bond or a microstructure change. For the supramolecular
hydrogels that mechanism may be the reversible breaking of the supramolecular
bonds or orientation of nanodomains. The ability of the supramolecular hydrogels to
fully or near-fully recover their dimensions with no or insignificant permanent
deformation suggests that reformation of the broken supramolecular bonds occurs
spatially at essentially the same positions in the gel where they were originally,
which may be a consequence of highly non-affine deformation – i.e., that the
microscopic deformation of the nanodomain microstructure is very small compared
with the macroscopic deformation. In that case, most of the deformation is provided
by changing conformations of the network chains, such that the deformation appears
to be nearly elastic. That reversibility is a notable advantage of these materials
compared with tough hydrogels that exhibit plastic deformation, such as DN
hydrogels. The small microscopic and large macroscopic deformation could be
achieved by intradomain jumping of the hydrophobic groups. The core of the
Hydrophobically Associating Hydrogels with Microphase-Separated Morphologies
187