Understanding microscopic parameters that control steepness of the temperature variations of segmental relaxation (fragility) and the glass transition phenomenon remains a challenge. We present dielectric and mechanical relaxation studies of segmental dynamics in various polymers with different side groups and backbone structures. The results have been analyzed in terms of flexibility of backbone and side groups of polymeric molecules, as suggested by the recent theoretical works by Dudowicz et al. A comparison of structures with identical backbones and varying side groups and identical side groups but different backbones reveals that the flexibility of side groups relative to the flexibility of the backbone is the most important factor controlling fragility in polymers, while the glass transition temperature T g depends primarily on the backbone flexibility and the side group bulkiness (occupied volume). Based on these results and analysis of literature data we formulated a modified approach to understand the role of chemical structure in segmental dynamics: (i) Polymers with stiff backbones always have high T g and fragility, while (ii) polymers with flexible backbones and no side groups are the strongest; (iii) however, for the most common type of polymeric structure, C-C or Si-O backbone with side groups, fragility increases with increasing "relatiVe" stiffness of side groups versus the backbone. In this class of polymers, lowest fragility is expected when the side groups are of similar chemical structure (or flexibility) as the backbone, as in the case of polyisobutylene, one of the strongest polymers known.

The viscoelastic glass-to-rubber softening transition is analyzed for various cross-linked polymers reinforced with filler particles. We find that the loss modulus peak corresponding to the segmental relaxation process (glass transition) is not significantly affected by the particle surface area in carbon black-filled polybutadiene or by silane chemical coupling of poly(styrene-co-butadiene) to silica. Large differences in shape and magnitude of the peak in the loss tangent (tan δ) vs temperature are noted for these materials; however, this is due to variations in the storage modulus at small strains in the rubbery state, which is influenced by the nature of the jammed filler network. The use of a simple relaxation model demonstrates this feature of the viscoelastic glass transition in filled rubber. It is not necessary to invoke concepts involving a mobility-restricted polymer layer near the filler surfaces to explain the viscoelastic results. Atomic force microscopy conducted with an ultrasharp tungsten tip indicates that there may be some stiffening of the elastomer in the proximity of filler particles, but this does not translate into an appreciable effect on the segmental dynamics in these materials. IntroductionDespite significant research activity on the effect of nanoscale confinement on the glass transition temperature (T g ) of polymers, many controversial issues remain unresolved, as recently reviewed by Alcoutlabi and McKenna.1 Of particular relevance to the field of elastomers is the influence of reinforcing particles on the polymer T g . It is reasonable to expect that physical adsorption or chemical attachment of polymer chains to rigid particles can slow down the polymer dynamics, which might increase the glass transition of the polymer chains near particle surfaces. However, while some published studies show increases in T g upon the addition of carbon black, silica, or other fillers, others report no change in T g or even T g decreases.2-26 The nature of the interfacial interactions between the polymer and particles may account for some of the disparate results concerning the effect of fillers on T g .

The effect of nanosized silica particles on the properties of poly(vinyl acetate) (PVAc) was investigated for a range of silica concentrations encompassing the percolation threshold. The quantity of polymer adsorbed to the particles ("bound rubber") increased systematically with silica content and was roughly equal to the quantity shielded from shear stresses ("occluded rubber"). This bound and occluded polymer attained a level of ∼12% at a silica volume content of 28%; nevertheless, the glass transition properties of the PVAc, including the glass transition temperature, local segmental relaxation function and relaxation times, and the changes in thermal expansion coefficient and heat capacity at T g , were unaffected by the interfacial material. That is, there is no indication that the local segmental dynamics of the chains adjacent to silica particles differ from the motions of the bulk chains. Interestingly, the volume sensitivity of the segmental dynamics, as determined from the scaling exponent γ in the relation T g ∼ V g -γ in which V g is the specific volume at the glass transition, becomes stronger with increasing silica concentration. Moreover, this dependence of γ increases abruptly at the filler percolation threshold. The implication of this result and possible directions for new research are considered.

We review the literature concerned with the effect of proximity to a filler surface on the local segmental mobility of polymer chains. This mobility is commonly assessed from either the glass transition temperature, Tg, or the segmental relaxation times measured by mechanical, dielectric, or NMR spectroscopy. Published studies report increases, decreases, or no change in Tg upon the addition of carbon black, silica, and other reinforcing fillers. Similarly, the segmental relaxation times have been found to increase or be invariant to the presence of nanometer-sized particles. Some of these discrepancies can be ascribed to ambiguous methods of data analysis; others likely reflect the variation in filler-polymer interaction among different systems. There are unequivocal examples of polymers that have segmental dynamics and glass transitions unaffected by nano-particle reinforcement. However, the general principles governing the behavior remain to be clarified, with further work, focusing on the micromechanics at the particle interface, required for resolution of this important aspect of rubber science and technology.

No abstract

Despite significant experimental and theoretical efforts, a fundamental understanding of how the chemical structure influences various dynamic processes in glass-forming materials and polymers remains a topic of active discussion. The present study analyzes the influence of polar interactions on the temperature dependences of segmental and chain dynamics in polymers. We found that segmental dynamics slow down (the glass transition temperature T g increases) and have steeper temperature dependence (higher fragility index m) when a polar group is attached directly to the polymer backbone. However, when a polar group is separated from the backbone by a side group, both T g and m become complex functions of the monomer's polarity and the relative position of the polar group. Our analysis revealed unexpected effect of polar interactions on chain dynamics: chain modes in polar polymers are coupled to the segmental dynamics stronger than in nonpolar polymers with similar fragilities. This results in a steeper temperature dependence of chain dynamics in polar polymers. How the polar interactions affect the coupling of chain and segmental modes remains unclear. ■ INTRODUCTIONFor the past several decades, a significant amount of research was aimed at understanding the effect that chemical structure has on various dynamic processes in polymeric systems. 1−7 It is well-known that parameters like the glass transition temperature, T g , and the steepness of the temperature dependence of segmental relaxation (fragility index, m) in polymers strongly depend on backbone/side group chemical structure. 1,2,4,6,8 However, most of the research in the field of segmental dynamics was focused either on materials with weak van der Waals interactions or relatively polar polymers were grouped together with nonpolar polymers in the analysis. 1,2,4,6 Limited attention was paid to the influence that polar interactions themselves may have on properties such as T g and m. The presence of additional intermolecular interactions does, without a doubt, shift the delicate balance between the entropic and enthalpic variables that control the segmental relaxation process. A naive picture suggests that the presence of polar interactions would lead to slower segmental dynamics due to higher friction between relaxing units; i.e., the glass transition temperature would rise. An increase in the strength of the intermolecular interactions could also result in a higher degree of cooperativity/heterogeneity of the segmental dynamics, potentially leading to a higher fragility index value for a polymer. However, the presence of a polar group may have a different effect on the volumetric and energetic activation barriers for segmental motion, making it rather difficult to predict the exact behavior of T g and m.The effect that strong intermolecular interactions impose on chain dynamics is even less clear. Several recent studies revealed an intriguing aspect of polymer dynamics: chemical structure seems to have very little impact on the steepness of temperature d...

Tsagaropoulos and Eisenberg [Macromolecules 1995, 28, 396; Macromolecules 1995, 28, 6067] reported a second loss tangent (tan δ) peak in temperature-dependent viscoelastic data for various un-cross-linked polymers filled with nanometer-sized silica particles. This peak, occurring at temperatures as much as 100 °C above the primary tan δ peak (glass-to-rubber softening transition), was ascribed to the glass transition of immobilized chains near the particles. This research is often cited as support for the existence of severely retarded segmental motion of polymer near the surfaces of small particles (glassy polymer shell). We offer a different interpretation of the results from Tsagaropoulos and Eisenberg, reinforced by our recent measurements on particle-filled polybutadiene and consideration of other literature data. Particles restrict the flow of some polymer chains, thus resulting in incomplete terminal relaxation, and partial cross-linking of unfilled polymers produces the same higher temperature tan δ peak. Polymer chains which are adsorbed onto filler surfaces are immobilized from a flow relaxation (chain diffusion/reptation) standpoint, but segmental relaxation (glass transition) is not substantially altered by small particles as a general rule.

The understanding of size-dependent properties is key to the implementation of nanotechnology. One controversial and unresolved topic is the influence of characteristic size on the glass transition temperature (T(g)) for ultrathin films and other nanoscale geometries. We show that T(g) does depend on size for polystyrene spherical domains with diameters from 20 to 70 nm which are formed from phase separation of diblock copolymers containing a poly(styrene-co-butadiene) soft block and a polystyrene hard block. A comparison of our data with published results on other block copolymer systems indicates that the size dependence of T(g) is a consequence of diffuse interfaces and does not reflect an intrinsic size effect. This is supported by our measurements on 27 nm polystyrene domains in a styrene-isobutylene-styrene triblock copolymer which indicate only a small T(g) depression (3 K) compared to bulk behavior. We expect no effect of size on T(g) in the limit as the solubility parameters of the hard and soft blocks diverge from each other. This strongly segregated limiting behavior agrees with published data for dry and aqueous suspensions of small polystyrene spheres but is in sharp contrast to the strong influence of film thickness on T(g) noted in the literature for free standing ultrathin polystyrene films.