Eur. Phys. J. Special Topics 141, 143–151 (2007) c  EDP Sciences, Springer-Verlag 2007 DOI: 10.1140/epjst/e2007-00032-0 T HE EUROPEAN P HYSICAL JOURNAL SPECIAL TOPICS Fluorescence studies of confinement in polymer films and nanocomposites: Glass transition temperature, plasticizer effects, and sensitivity to stress relaxation and local polarity M.K. Mundra 1 , C.J. Ellison 2 , P. Rittigstein 2 , and J.M. Torkelson 1,2 1 Department of Materials Science and Engineering, Northwestern University, Evanston, IL 60208, USA 2 Department of Chemical and Biological Engineering, Northwestern University, Evanston, IL 60208, USA Abstract. Confinement effects in polystyrene and poly(methyl methacrylate) films and nanocomposites are studied by fluorescence. The ability to employ an inten- sive measurable, the excited-state fluorescence lifetime, in defining the glass tran- sition temperature, T g , of polymers is demonstrated and compared to the use of an extensive measurable, fluorescence intensity. In addition, intrinsic fluorescence from the phenyl groups in polystyrene is used to determine the T g -confinement effect in films as thin as ∼15 nm. The decrease in T g with decreasing film thickness (below ∼60 nm) agrees well with results obtained by extrinsic pyrene fluorescence. Dye label fluorescence is used to quantify the enhancement in T g observed with decreasing thickness (below ∼90 nm) in poly(methyl methacrylate) films; addition of 2–4 wt% dioctyl phthalate plasticizer reduces or eliminates the T g -confinement effect in films down to 20nm thickness. Intrinsic polystyrene fluorescence, which is sensitive to local conformation, is used to quantify the time scales (some tens of minutes) associated with stress relaxation in thin and ultrathin spin-coated films at T g + 10 K. Finally, the shape of the fluorescence spectrum of pyrene doped at trace levels in polystyrene films and polystyrene-silica nanocomposites is used to determine effects of confinement on microenvironment polarity. 1 Introduction As demonstrated more than a decade ago in seminal studies by Jackson and McKenna [1], Reiter [2], and Keddie et al. [3], confinement of glass-forming materials at the nanoscale can lead to significant deviations in glass transition temperature from the bulk value. This effect is present in low molecular weight glass formers confined to nanoporous systems [1] as well as polymeric glass formers confined in ultrathin films [2,3]. In particular, large deviations from bulk T g values were reported in the 1990’s by a number of research groups. Keddie et al. [3] reported a T g depression relative to bulk Tg of 23 K in a 15-nm-thick polystyrene (PS) film supported on a silica substrate while van Zanten et al. [4] reported a T g enhancement relative to bulk T g of ∼50 K in an 8-nm-thick poly(2-vinyl pyridine) (P2VP) film supported on a silica substrate. These different responses can be rationalized by the role of hydrogen-bonding interactions at the polymer-substrate interfaces; such interactions are present in P2VP films supported on silica (with hydroxyl groups present on the substrate surface) but absent in PS films. Greater deviations from bulk T g values were reported by Forrest et al. [5] who characterized T g values in freely standing PS films and found T g depressions relative to bulk T g of 70 K in 20-nm-thick