Interfacial Response of a Fluorescent Dye Grafted to Glass Joseph L. Lenhart, † John H. van Zanten, ‡ Joy P. Dunkers, § and Richard S. Parnas* ,§ Chemical Engineering Department, Johns Hopkins University, Baltimore, Maryland 21218, Chemical Engineering Department, North Carolina State University, Raleigh, North Carolina 27695, and Polymers Division, National Institute of Standards and Technology, Gaithersburg, Maryland 20899 Received April 12, 2000 The properties of an epoxy/glass interfacial region are studied by covalently grafting a fluorescent probe to the glass surface. A dimethylaminonitrostilbene fluorophore is tethered to a triethoxysilane-coupling agent, generating a fluorescently labeled silane coupling agent (FLSCA). The glass surface is coated with a silane layer that was doped with small amounts of FLSCA. When the FLSCA-doped, silane-coated glass is immersed in epoxy resin, a 42-nm blue shift in fluorescence occurs during resin cure over the grafted FLSCA layer. When the dye is dissolved in bulk epoxy a 64-nm blue shift occurs during resin cure. The difference in blue shift is attributed to higher polarity and enhanced mobility in the buried interface. Introduction The surface properties of polymers are known to be different from those of the bulk polymers and many techniques have been used to probe these surface proper- ties. The mobility and glass transition behavior of a polymer at surfaces and interfaces may be different from the bulk polymer behavior. Preferential segregation of low-molecular-weight polymer fractions, or chain ends to the interface, 1 and lower entanglement density near the interface 2 could reduce the glass transition in this region. For linear polymers, the perturbed polymer structure will typically extend on the order of a radius of gyration away from the surface. Mobility in the interfacial region is dominated by the interaction between the substrate and the polymer. Keddie et al. used ellipsometry to observe that the glass transition of thin polystyrene films on silicon oxide surfaces was lower than the bulk glass-transition temperature (T g ). 3 Reiter observed, using X-ray reflecto- metry, that thin polystyrene films could dewet a silicon wafer even below the glass transition of the bulk polymer, suggesting a lower effective T g in the thinner films. 4 The interaction between the silicon oxide surface and poly- styrene is weak. When poly(methyl methacrylate) (PMMA) films are cast on a silicon wafer, the interfacial mobility can decrease because of the strong interaction between the oxide surface and PMMA. For example, the glass transition of PMMA films on a silicon wafer can increase by ∼30 °C. 5 Also, the thermal expansivity of thin PMMA films on a silicon wafer were lower than the bulk PMMA, both above and below the T g . 6 For thermosetting resins near a solid surface, the problem is even more complicated than with linear polymers. In addition to the effects of the solid surface, a thermoset is reacting in the presence of the substrate. Thermosetting resins can contain impurities, and often a mixture of many monomers, which can vary both in composition and molecular weight. 7,8 Preferential diffusion of different monomers or low-molecular-weight plastizers to the surface can change the interfacial cure behavior. In addition, the substrate surface is often coated with a sizing layer (often a silane coupling agent) to promote adhesion and durability at the interface. 9-11 The sizing layer can vary in thickness from tens of angstroms to hundreds of nanometers depending on the application. In fiber-reinforced composites, the sizing layer may also include processing aids such as a film former, anti-static agent, and lubricant, in addition to the coupling agent. The presence of the sizing layer, interpenetration between the sizing layer and the curing thermoset, and potential reactivity between the sizing and resin can further alter the interfacial structure. Because of these factors, the interfacial region for a thermosetting resin can extend 100-500 nm away from the substrate and may extend several microns in certain cases (much farther than a radius of gyration of a typical linear polymer). The cure difference near the interface causes a composition gradient that extends from the substrate into the resin. This compositional and resultant morphological difference may cause the interfacial region to have different properties from the bulk resin. These differences are manifested by changes in the T g , coefficient of thermal expansion, and the viscoelastic properties of the interfacial region when compared with the bulk resin. 12 These differences will * To whom correspondence should be sent. † Johns Hopkins University. ‡ North Carolina State University. § National Institute of Standards and Technology. Identification of any commercial products is made only to facilitate experimental reproducibility and describe experimental procedure. It does not imply endorsement by NIST or imply that the particular product is necessarily the best for the experiment. (1) Mayes, A. M. Macromolecules 1994, 27, 3114. (2) Brown, H. R.; Russell, T. P. Macromolecules 1996, 29, 798. (3) Keddie, J. L.; Jones, R. A. L.; Cory, R. A. Europhys. Lett. 1994, 27, 59. (4) Reiter, G. Macromolecules 1994, 27, 3046. (5) Wallace, W. E.; van Zanten, J. H.; Wu, W. L. Phys. Rev. E. 1995, 52, R3329. (6) Wu, W. L.; van Zanten, J. H.; Orts, W. J. Macromolecules 1995, 28, 771. (7) Drzal, L. T. The Interphase in Epoxy Composites. In Advances in Polymer Science; Dusek, K., Ed.; Springer-Verlag: Berlin, 1986; p 1. (8) Morris, C. E. M.; Pearce, P. J.; Davidson, R. G. J. Adhes. 1982, 15, 1. (9) Plueddemann, E. P. Composite Materials: Interfaces in Polymer Matrix Composites; Academic Press: New York, 1974; Vol. 6. (10) Larson, B. K.; Drzal, L. T. Composites 1994, 25, 711. (11) Thomason, J. L. Composites 1995, 26, 487. (12) Palmese, G. R.; McCullough, R. L. J. Appl. Polym. Sci. 1992, 46, 1863. 8145 Langmuir 2000, 16, 8145-8152 10.1021/la000553n CCC: $19.00 © 2000 American Chemical Society Published on Web 09/15/2000