We report the first multi‐system study of a layered‐silicate dispersion in polysiloxane/layered‐silicate nanocomposites. A variety of layered silicates (montmorillonite, synthetic fluoromica, laponite, and fluorohectorite) and cationic modifiers (single‐, twin‐, and triple‐tailed surfactants with tails of varying lengths and both primary and quaternary head‐groups) are combined to form organically modified layered silicates, which are then screened for compatibility with low‐molecular‐weight silanol‐terminated poly(dimethylsiloxane) (PDMS). Promising combinations are then selected and studied in greater depth with respect to both molecular weight and polysiloxane end‐group and substituent chemistry. We find that the PDMS backbone is generally incompatible with the layered silicates, regardless of modification type, and that dispersion in PDMS systems results from the presence of polar end‐groups, a result unprecedented in the field of polymer nanocomposites. We go on to quantify the substituent effect, not only with respect to end‐group chemistry, but taking into account changes in the polysiloxane backbone itself. For instance, in the absence of polar end‐groups we observe dispersion in the case of poly(methylphenylsiloxane) but not poly(3,3,3‐trifluoropropylmethylsiloxane). Finally, we apply a new epoxy/amine PDMS curing chemistry to PDMS‐nanocomposite production and show higher levels of layered‐silicate dispersion than observed in comparable silanol‐terminated PDMS‐based systems. Our findings serve as an indication of what is necessary to achieve a layered‐silicate dispersion in polysiloxane/layered‐silicate nanocomposites, and may indicate a more general approach for improving dispersion in systems where the polymer backbone is otherwise incompatible with the layered silicate.

Silica-filled polydimethylsiloxane networks are submitted to successive stretching cycles, in order to get the stabilized stretching curve, at variable temperature. This study explains the peculiar temperature dependence of the first stretching curve of filled rubbers, and highlights the molecular origin of the stress-softening phenomenon, known as Mullins effect. Thanks to the comparison between the strain dependence of stress and the molecular orientation, this effect is attributed to the detachment from the filler surface or slippage on the filler surface, of chains having reached their limit of extensibility. Moreover, by taking advantage of Atomic Force Microscopy observations on stretched samples, the Mullins effect is shown to take place mainly in regions of high local concentration of silica. The experimental results are also compared to Bueche's model for the Mullins effect.

The polymer relaxation dynamic of a sample, stretched up to the stress hardening regime, is measured, at room temperature, as a function of the strain λ for a wide range of the strain ratė γ, by an original dielectric spectroscopy set up. The mechanical stress modifies the shape of the dielectric spectra mainly because it affects the dominant polymer relaxation time τ , which depends on λ and is a decreasing function ofγ. The fastest dynamics is not reached at yield but in the softening regime. The dynamics slows down during the hardening, with a progressive increase of τ . A small influence ofγ and λ on the relative dielectric strength cannot be excluded.

No abstract