We report the first measurements of the glass transition temperature T g for thin freely standing polystyrene (PS) films. We have used Brillouin light scattering to measure T g for freely standing films of different thicknesses. We find that T g decreases linearly with film thickness h for h # 700 Å, with a reduction of 70 K for a film with h 290 Å. These measurements characterize unambiguously the effects of the free surface on T g of thin polymer films. Results are compared to similar results for supported PS films [Keddie et al., Europhys. Lett. 27, 59 (1994)], and we find that their measured values are influenced strongly by the substrate. [S0031-9007(96)01093-9] PACS numbers: 64.70. Pf, 68.60.Bs, 78.35.+c From organic liquids to metals to polymers, almost any substance can be transformed into a glassy state. One of the fundamental parameters describing a glass is the glass transition temperature T g . For temperatures greater than T g , the material is a viscous liquid. As the liquid is cooled below this temperature, the material forms an amorphous solid. Despite the technological significance of glass forming materials, the glass transition itself is poorly understood. In particular, it is not clear whether the glass transition is a thermodynamic transition [1] or a purely kinetic phenomenon [2]. Adam and Gibbs [3] introduced the concept of cooperative rearrangement in an attempt to unify these two views of the glass transition by demonstrating that such cooperativity, coupled with a thermodynamic glass transition, resulted naturally in system dynamics such as those described by the WLF equation [4] for temperatures near T g . Donth [5] has estimated the size of such cooperatively rearranging regions (CRR) to be of the order of 10 Å for a number of glass forming materials. The introduction of such a length scale suggests that studying samples with dimensions comparable to the CRR may lead to the observation of finite size effects.An attractive choice for studies of finite size effects on the glass transition is the use of polymer molecules. For polymers the molecule can be characterized by the endto-end distance R EE ϳ 2R g , where R g is the radius of gyration of the molecule. R EE is typically much larger than the size of the CRR, and R EE can be adjusted by changing the molecular weight M w of the molecules. Polymer molecules can be confined by preparing samples in a thin film geometry. The relevant length scale is the film thickness h which can be adjusted to be comparable to, much larger than or smaller than R EE so that the effects of chain confinement can be investigated. Glassy polymer films can be cast onto any substrate, even those that the polymer melt itself does not wet.The first direct measurements of the glass transition temperature in thin polymer films were performed recently by Keddie, Jones, and Cory [6]. For polystyrene (PS) films on hydrogen-passivated Si(111), the measured T g values were lower than the bulk value T g ͑bulk͒ for films with thicknesses h # 400 Å. Because the changes ...

No abstract

We have used Brillouin light scattering and ellipsometry to measure the glass transition temperature T g of thin polystyrene ͑PS͒ films as a function of the film thickness h for two different molecular weights M w. Three different film geometries were studied: freely standing films, films supported on a SiO x surface with the other film surface free ͑uncapped supported͒, and films supported on a SiO x surface and covered with a SiO x layer ͑capped supported͒. For freely standing films T g is reduced dramatically from the bulk value by an amount that depends on both h and M w. For hՇR EE ͑the average end-to-end distance of the unperturbed polymer mol-ecules͒, T g decreases linearly with decreasing h with reductions as large as 60 K for both M w values. We observe a large M w dependence of the T g reductions for freely standing films which provides the first strong evidence of the importance of chain confinement effects on the glass transition temperature of thin polymer films. For both the uncapped and capped supported films, T g is reduced only slightly (Ͻ10 K) from the bulk value, with only small differences in T g (Ͻ4 K) observed between uncapped and capped supported films of the same thickness. The results of our experiments demonstrate that the polymer-substrate interaction is the dominant effect in determining the glass transition temperature of PS films supported on SiO x .

The motion of polymer chain segments cooled below the glass transition temperature slows markedly; with sufficient cooling, segmental motion becomes completely arrested. There is debate as to whether the chain segments near the free surface, or in thin films, are affected in the same way as the bulk material. By partially embedding and then removing gold nanospheres, we produced a high surface coverage of well-defined nanodeformations on a polystyrene surface; to probe the surface dynamics, we measured the time-dependent relaxation of these surface deformations as a function of temperature from 277 to 369 kelvin. Surface relaxation was observed at all temperatures, providing strong direct evidence for enhanced surface mobility relative to the bulk. The deviation from bulk alpha relaxation became more pronounced as the temperature was decreased below the bulk glass transition temperature. The temperature dependence of the relaxation time was much weaker than that of the bulk alpha relaxation of polystyrene, and the process exhibited no discernible temperature dependence between 277 and 307 kelvin.

The past 20 years have seen a substantial effort to understand dynamics and the glass transition in thin polymer films. In this Perspective, we consider developments in this field and offer a consistent interpretation of some major findings. We discuss recent experiments that directly measure mobility at or near the surface of glassy polymers. These experiments indicate that enhanced mobility near the free surface can exceed bulk mobility by several orders of magnitude and extend for several nanometers into the bulk polymer. Enhanced mobility near the free surface allows a qualitative understanding of many of the observations of a reduced glass transition temperature T g in thin films. For thin films, knowledge of T g by itself is less useful than for bulk materials. Because of this, new experimental methods that directly measure important material properties are being developed.

We have used transmission ellipsometry to perform a comprehensive study of the glass transition temperature T(g) of freely standing polystyrene films. Six molecular weights M(w), ranging from 575 x 10(3) to 9100 x 10(3), were used in the study. For each M(w) value, large reductions in T(g) (as much as 80 degrees C below the bulk value) were observed as the film thickness h was decreased. We have studied in detail the dependence of the T(g) reductions on M(w) in a regime dominated by chain confinement effects. The empirical analysis presented is highly suggestive of the existence of a mechanism of mobility in thin freely standing films that is inhibited in the bulk and distinct from the usual cooperative motion associated with the glass transition.

We have used Brillouin light scattering to make a detailed study of the behavior of the glass transition temperature T(g) in ultrathin, free-standing polystyrene films. The glass transitions were experimentally identified as near discontinuities in the thermal expansion. The effects of film thickness, molecular weight, and thermal history on the measured T(g) values have been investigated. While the size of the glass transition effects was comparable for all molecular weights, a complicated M(n) dependence suggested a separation of the results into two regimes, each dominated by a different length scale: a low M(n) regime controlled by a length scale intrinsic to the glass transition and a high M(n) region, where polymer chain confinement induced effects take over.

We report measurements of the glass transition temperature, T(g), in free standing polymer films in a low M(n) limit where chain confinement effects are not observed. The measured T(g) values are accurately described by a layer model incorporating a mobile surface layer with a size determined by the length scale of cooperative dynamics. The analysis leads to a surface T(g) value and length scale of cooperative motion near bulk T(g) which quantitatively agree with independently determined values. The model and parameters provide a framework within which all previous measurements of T(g) values in thin supported films may be understood and provides values for the length scale of cooperative motion over an extended range of temperatures below the bulk T(g) value.