Substrate Effects on Glass Transition and Free Surface Viscoelasticity of Ultrathin Polystyrene Films Heedong Yoon and Gregory B. McKenna* Department of Chemical Engineering, Whitacre College of Engineering, Texas Tech University, Lubbock, Texas 79409-4121, United States ABSTRACT: We describe results from experiments in which spontaneous embedment of 20 nm silica particles was used to probe under-layer effects on the free surface dynamics and the glass transition temperature (T g ) of polystyrene. Both 13 and 20 nm thick polystyrene (PS) films were prepared and placed on different under-layer substrates, which in turn were supported on a silicon wafer (Si) substrate. The under-layer substrates used were PS, poly(2-vinylpyridine) (P2VP), and poly(methyl methacrylate) (PMMA) with thicknesses ranging from 13 to 350 nm. The particle height change during the embedment was monitored using an atomic force microscope. For the PS films supported on the PS and P2VP under-layers, experimental temperatures varied from T g - 10 K to T g + 5 K. In the case of the PMMA under-layer, experimental temperatures varied from T g - 10 K to T g + 10 K. The Hutcheson and McKenna model [Phys. Rev. Lett. 2005, 94 (7)] was applied to the particle embedment depth to obtain the surface rheological temperatures [Eur. Phys. J. E 2007, 22 (4), 281-286] and the T g . It is found that the dynamics of the top-layer PS films were faster than the bulk material below the macroscopic T g and slower above it for all under-layers considered. The T g of both the 20 and 13 nm top-layer PS films were found to be essentially independent of under-layer thickness and reduced by less than 7 °C. Upon replacing PS under-layers with the same thickness of P2VP and PMMA as the under-layers, the T g of the 20 nm PS top-layer films changed by less than 5 K. 1. INTRODUCTION The glass transition is the phenomenon in which a liquid in the equilibrium state is rapidly cooled and falls into a non- equilibrium state for kinetic reasons. 3-6 There is a great deal of ongoing research in this area, as a substance’s viscoelastic and mechanical properties are closely related to its glass transition temperature (T g ). 3,4,6-18 Although the glass transition behavior of a bulk material is still not fully understood, there has developed much interest in nanoconfinement effects on the glass transition temperature and associated behaviors. 1-3,7-66 Consider, for example, how the T g of an ultrathin film of polystyrene compares with the T g of the bulk material. The T g of supported ultrathin films of polystyrene has been found to either remain unchanged or reflect a reduced value when compared to the bulk T g . 2,3,8-10,19,21,22,28-30,33,34,43,45,47 How- ever, in the case of free-standing ultrathin films of polystyrene (PS), a reduction in T g has been commonly ob- served. 3,7,14,15,39,41,44,48 Similarly, different materials and differ- ent methods give different T g changes at the nanometer scale. 3,7,8,19,20,27,29,37,42,43,46,49 The reasons for the changes in T g for ultrathin films are still unclear, but there are some commonly proposed ideas for them: the role of free surface and substrate, size effects, and molecular weight effects. 3,7,20,35,36,39,43,48-66 Among these ideas, free surface effects 1,3,7,28,29,33,35,38,39,41-43,50,66 are, perhaps, the most prevalent. The free surface view considers that the segmental mobility at the free surface is what causes a change in T g in the remainder of the film. 3,7,28-30,33,43,66 One consequence of this is a gradient of mobility 64 in the films. For example, the work from Torkelson’ s group using fluorescence intensity measurements on multilayer films suggests that free surfaces and substrate or under-layers can affect the T g to distances of several tens of nanome- ters. 9-11,28-30,32-34 In that work, the T g was determined by measuring a change in the slope of fluorescence intensity as a function of temperature, 9-11,28-30 i.e., in a pseudo-thermody- namic mode. 3 They found, for instance, that the T g of a 15 nm top layer of a polystyrene (PS) film deposited on PS under- layers changed with the PS under-layer thickness. 11,34 They also observed no T g reduction in the PS top layer with a poly(2- vinylpyridine) (P2VP) under-layer and interpreted this to imply that strong interactions could affect the segmental mobility of a film surface to fairly large length scales and, consequently, prevent the T g reduction 11,34 seen with, e.g., PS on a PS under-layer. One of the direct measurements of free surface dynamics and their interpretation with respect to effects on T g was in the work by Teichroeb and Forrest. 42,53 They used a spontaneous Received: August 7, 2014 Revised: November 25, 2014 Article pubs.acs.org/Macromolecules © XXXX American Chemical Society A dx.doi.org/10.1021/ma501630g | Macromolecules XXXX, XXX, XXX-XXX