By adding nanoparticles to polymer melts, nanocomposites with improved mechanical and thermal properties are obtained. For a given polymer-filler couple, the improvement of properties appears related to the filler size, volume fraction and dispersion. Nevertheless no universal model is established at the present time. Whereas the filler size and volume fraction are determined by the formulation, the dispersion is more difficult to control and investigate. In the first part of the talk, we will show how we can tune the particles dispersion in model nanolatex systems (polymethacrylate-silica) using different strategies: the control of the filler charge, the modulation of the latex/filler size ratio (R) and the use of different polymer chain length. Using small angle scattering and TEM analysis we will demonstrate that repulsive aggregates of varying size can be obtained playing with the pH of the casting solution1. A good filler dispersion is obtained for R=1, whereas a network of well-organized silica particles around the latex beads is highlighted for high latex/filler size ratio. Finally, decreasing the polymer molecular weight, fractals aggregates of increasing size are generated. The effect of these various structures on mechanical properties will be discussed. The mechanical properties of nanocomposite materials are controlled to a large extent by the filler-filler interactions, nevertheless, another important contribution - less well understood - is due to the polymer chain-filler interactions. In a second part, we will display results of an investigation of the polymer chains in such nanocomposites2. Small angle neutron scattering experiments performed in zero average contrast conditions (mixing hydrogenated and deuterated latex beads) enabled to follow the latex beads dissolution using an original model. Experimental results demonstrate that the dissolution dynamic of polymer chains is significantly slowed down by the presence of silica nanoparticles.

Mixing inorganic nanoparticles (ca. 10-20 nm) with polymer leads to the formation of nanocomposites. Their main reason of existence is that they combine the best of both worlds, i.e. nanoparticle strength and polymer flexibility, which has important applications, e.g. for the car tire industry. In this talk, a simplified industrial model system based on filler particles of industrial origin but with a highly reduced chemistry will be presented. It is generally recognized that the filler structure has a strong impact on the rheological properties, and it is thus important to be able to characterize it in detail. Here, microscopic observations of the dispersion of the nanoparticles in the polymer matrix by electron microscopy and scattering will be discussed and related to the microscopic structure to their macroscopic rheological properties. The structure of SBR-silica nanocomposites designed to reproduce key features of industrial samples, but of simplified composition, has been studied on length scales extending from the nanometric primary particles to microns. We propose an original method for scattering data analysis of such multi-scale systems by including self-consistent polydisperse form and structure factors of aggregates [1]. The complex structure of the silica within the nanocomposites is quantitatively modeled in a step-by-step manner, starting with the primary silica beads as basic building units (10 - 20 nm range). These beads are found to be aggregated in small clusters, the typical radius of which is in the 40 nm range. These aggregates are themselves concentrated in large-scale fractal branches (thickness ca. 150 nm, extending over microns), cf. the TEM cliché below. A quantitative analysis of the TEM data is used to estimate the volume fraction of fractal branches, which is an input for structure factor calculations of aggregates inside branches using numerical simulations as indicated in the Figure on the right. Then, the influence of polymer grafting is explored, yielding a strong effect on the aggregate mass [2]. In parallel, the rheological properties of these silica structures in the SBR-matrix are characterized with oscillatory shear. The resulting reinforcement curve of the high-frequency storage modulus can be described using a combination of hydrodynamic reinforcement for silica fraction below a critical percolation fraction, and a percolation law above. It is interesting to note that the aggregate compacity obtained from the structural analysis (SAXS and TEM) is fully compatible with the reinforcement data.

Nous proposons l'étude de matériaux composites industriels simplifiés constitués de caoutchouc non réticulé (copolymère styrène-butadiène " SBR ") renforcés par des charges nanométriques de silice hautement dispersible. Afin d'identifier les mécanismes physico-chimiques responsables de ce renforcement et être capable de l'optimiser, nous devons comprendre les corrélations existantes entre les propriétés macroscopiques du matériau et la structure des charges à différentes échelles. Pour cela, une large campagne d'expériences de diffusion des rayons-X aux petits angles (DXPA) ainsi que de nombreux clichés de microscopie électronique ont été réalisés. En couplant ces données avec des simulations Monte-Carlo, il a été notamment possible de mettre en avant la présence d'une organisation à trois niveaux en partant de billes élémentaires d'une dizaine de nanomètres formant des agrégats eux-mêmes arrangés selon un réseau tridimensionnel branché existant à travers tout l'échantillon. Afin de déceler les corrélations existantes entre structure et propriétés, nous nous sommes attachés à décrire systématiquement l'influence de paramètres-clés tels que la fraction volumique en silice, le type de polymère employé (greffable sur la silice ou pas) ou leur masse molaire sur la morphologie des charges (par exemple la taille des agrégats) ainsi que sur le comportement mécanique (par exemple leur module d'élasticité) des composites. Ce travail a permis d'identifier la densité de greffage des chaines comme le paramètre définissant la structure des composites et impactant significativement le renforcement. Ces travaux résolument tournés vers la compréhension fondamentale, sont aussi une contribution à la recherche d'une loi de comportement décrivant l'effet de la structure des charges sur les performances des pneumatiques. Cette dernière doit permettre de répondre à des problématiques rencontrées en ingénierie telles que la résistance à l'usure, l'adhérence, ou la résistance au roulement.

An un-cross-linked SBR-system filled with precipitated silica nanoparticles of radius ≈10 nm by mixing is studied as a function of the fraction of graftable matrix chains (140 kg mol-1) varying from 0% to 100%, for a low (ΦSi = 8.5 vol %) and high (16.7 vol %) silica volume fraction. The linear rheology in shear shows a strong impact of the grafting on the terminal flow regime, and a shift to longer relaxation times with increasing grafting. Simultaneously, the plateau modulus stays approximately constant for the low ΦSi, suggesting a link to the silica content. The microstructure of the silica is characterized by using a combination of transmission electron microscopy and small-angle X-ray scattering data. We apply a quantitative model of interacting aggregates, and determine the average aggregation number (decreasing from 160 to 30 with grafting), aggregate size (50 to 30 nm), and compacity (55% to 35%). While the linear rheology seems to be dominated by the matrix composition, both the mixing rheology and the structure display a saturation with increasing grafting fraction. A closer analysis of this effect indicates that a critical amount of grafting is needed to trigger structural evolution. To summarize, a quantitative study of complex nanocomposites with several features of industrial systems demonstrates that the grafting density can be used as a fine-tuning parameter of rheology and microstructure.

Model systems of nanocomposites are important for our understanding of the structural and dynamical contributions of the constituents to macroscopic properties, and in particular mechanical reinforcement. We study silica-latex nanocomposites obtained from drying an aqueous colloidal mixture of polyacrylate nanolatex and silica nanoparticles. In these systems, the filler microstructure is controlled by the precursor solution pH and it is possible to work at constant aggregation number for different filler fractions [1]. We have followed the evolution of the polymer structure during nanocomposite formation and annealing using small angle neutron scattering under zero-average contrast conditions for the silica nanoparticles. The progressive disappearance of the latex beads by chain interdiffusion and release in the nanocomposites is analyzed with a model for the scattered intensity of hairy latex beads and an RPA description of the free chains (see Figure 1). In silica-free matrices and nanocomposites of low (7%v) and high (20%v) silica content, the annealing procedure results in a molecular dispersion of chains. However, the interdiffusion of individual chain is delayed in presence of 20% of silica. The radius of gyration was not found to be affected by the silica loading in these samples with similar aggregation state (typically ten primary silica particles per aggregate, percolating at high concentration). [2] This important result opens the way for a systematic study of the chain structure in complex environments. Fig.1: Modelling of the structural evolution of the polymer chains in nanocomposites during annealing. References [1] M. Tatou et al., Macromolecules 44, 9029 (2011) [2] A.C. Genix et al., Macromolecules 45, 1663 (2012)