Nano-scale friction of polystyrene in air and in vacuum Sophie Bistac a, * , Marjorie Schmitt a , Achraf Ghorbal a , Enrico Gnecco b , Ernst Meyer b a Universite´ de Haute Alsace, CNRS,15 rue Jean Starcky, 68057 Mulhouse, France b NCCR Nanoscale Science and Institute of Physics, University of Basel, Klingelbergstrasse 82, CH-4056 Basel, Switzerland article info Article history: Received 20 May 2008 Accepted 17 June 2008 Available online 25 June 2008 Keywords: Friction AFM Polymer abstract Using atomic force microscopy (AFM), we measured friction between an AFM tip and a polystyrene surface at 25  C, as a function of the sliding velocity and the applied normal load, both in air and under vacuum conditions. The objective was to analyze the influence of humidity on the frictional behavior of polystyrene. Our experimental results as a function of sliding velocity revealed a logarithmic increase of the friction force in air whereas a logarithmic decrease of this force is found in vacuum. Our comparative results unveil that two different dissipation mechanisms are dominating the frictional behavior of polystyrene in air and in vacuum. We propose a tentative explanation. Ó 2008 Elsevier Ltd. All rights reserved. 1. Introduction Atomic force microscopy (AFM) is a versatile tool helping to de- velop fundamental understanding of interfacial molecular phe- nomena [1]. AFM can provide information on surface topography, structure, or organization [2–8], and is also able to probe (by in- dentation experiments) the thermo-mechanical properties of polymer surfaces, strongly linked to the mobility of polymeric chains at these surfaces. Moreover, the phase contrast mode of AFM allows to distinguish differences in viscoelastic response [9]. Local surface properties such as surface modulus or glass transition tem- perature (T g ) can indeed differ from the corresponding properties in the bulk. Lateral force microscopy (LFM), also called frictional force mi- croscopy, measures the lateral or friction force between a surface and a sliding AFM tip, on the nanometer scale [10]. LFM can be used in order to get information about molecular mobility at polymer surfaces [11–13], and is often used to quantify nano-friction prop- erties. Scanning probe methods have been indeed applied to the investigation of nano-scale tribology [14–21]. Performing parallel investigations of identical systems on both macro and nano-scales, nano-friction measurements allow a better understanding of macrotribology properties. However, the correlation between macro and nano-scale results may be delicate in some cases, given that contact areas and velocity ranges being different. Nano-friction is usually strongly dependant on adhesion [22], molecular confor- mation [17] and surface energy [23]. In addition, the advantage of LFM is also to provide a single asperity (tip) in contact with the polymer surface [24], able to simulate what is happening during friction in nanostructures such as microelectromechanical systems (MEMS) [25,26]. At this contact scale, noting that the surface to volume ratio of the tip is quite large, humidity and presence of adsorbed water (specially on the hydrophilic surface of the tip) can have a major influence on nano-friction due to capillary effects [27,28]. Grigg et al. [29] demonstrated that such capillary forces can be several times larger than chemical interactions between a tip and a sample. Performing AFM experiments in vacuum represents a way to exclude the effect of adsorbed water, i.e. the contribution of capil- lary forces is eliminated [29–31]. However, it cannot be completely ruled out that polymer surface properties such as chain confor- mation and mobility are modified in vacuum. The goal of this work was to compare nano-scale friction of polystyrene (an amorphous glassy polymer, frequently used as model polymer) in air and in vacuum conditions. The first objective was to identify the effect of humidity on nano-friction. Thus, the evolution of the friction force between an AFM tip and the poly- styrene surface was analysed as a function of applied normal force and friction speed. Finally, we developed some hypotheses able to explain the experimental results. 2. Materials and methods Amorphous atactic polystyrene (PS) was purchased from Sigma–Aldrich and films were prepared by spin coating (1000 rpm) 10 wt% polymer solution in toluene onto silicon (100) wafers, which have been washed and sonicated in acetone, rinsed * Corresponding author. E-mail address: sophie.bistac-brogly@uha.fr (S. Bistac). Contents lists available at ScienceDirect Polymer journal homepage: www.elsevier.com/locate/polymer 0032-3861/$ – see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymer.2008.06.032 Polymer 49 (2008) 3780–3784