Surface roughness criteria for cement paste nanoindentation Mahalia Miller, Christopher Bobko, Matthieu Vandamme, Franz-Josef Ulm ⁎ Massachusetts Institute of Technology, Cambridge MA, United States Received 5 September 2007; accepted 29 November 2007 Abstract Analysis of nanoindentation experiments assumes that the indentation occurs on a flat surface. As a result, the accuracy of nanoindentation depends on reducing the surface roughness to a tolerable level. Within the context of statistical nanoindentation techniques suitable for heterogeneous materials, this study presents a criterion for roughness of cement paste surfaces for nanoindentation, and describes a method for obtaining the desired roughness. Through a systematic experimental study, we show the evolution of roughness and nanomechanical properties from indentation as a function of increased polishing. We conclude that the root-mean-squared (RMS) roughness of the sample, taken over a square area with edge dimensions of 200 times the average indentation depth of the dominating phase of the material, should be less than five times the average indentation depth of the dominating phase of the material. © 2007 Elsevier Ltd. All rights reserved. Keywords: Nanoindentation; Atomic Force Microscopy; Roughness; Cement paste; Sample preparation 1. Introduction The grid nanoindentation technique has been proven to provide useful, quantitative information about the mechanical behavior of cement pastes at the nanoscale [1–5]. The technique extends nanoindentation tools that had previously been limited to homogeneous materials and thin films to complex heterogeneous composites. Its attractiveness stems largely from the fact that properties of mechanically meaningful material phases can be identified in situ by performing large grids of indentations on highly heterogeneous samples, with a proper choice of the indentation depth to ensure the self-similar properties of classical continuum indentation analysis [6]. One challenge in the appli- cation of nanoindentation to cement pastes, however, is the development of a surface preparation technique that minimizes both sample disturbance and surface roughness. A further challenge is understanding how rough a surface can be without affecting the results of nanoindentation. Analysis of individual indentation tests using the conven- tionally applied Oliver and Pharr method [7] assumes that the initial surface is perfectly flat. Classical tools of indentation analysis, based on the infinite half-space model, can then be applied. Given this assumption, indentation with a Berkovich indenter can be considered to be self-similar. Since the infinite half-space model has by definition no length scale, dimensional analysis reveals that, for a homogeneous material, results of an indentation test do not depend on any other length scale than the indentation depth [8,9]. It is then readily understood that the presence of significant surface roughness introduces a new length scale into the dimensional analysis, which breaks the self-similarity of the indentation test and introduces a link between measured properties and indentation depth. Indeed, experimental evidence from prior research shows that the presence of significant surface roughness tends to increase the scatter in measured indentation modulus and indentation hardness, along with an overall reduction in these properties [10–12]. An ISO Standard dealing with nanoindentation warns that “surface finish has a significant influence on the test results” [15]. The question to be solved is, how small must the surface roughness be, in comparison with the indentation depth, to not have an effect on the measured mechanical properties? Some researchers have approached a slightly different question and introduced corrections to the contact depth based on measurements of the roughness of the sample to be indented [10,11]. These procedures have two important limitations. First, Available online at www.sciencedirect.com Cement and Concrete Research 38 (2008) 467 – 476 ⁎ Corresponding author. Tel.: +1 617 253 3544; fax: +1 617 253 6044. E-mail address: ulm@mit.edu (F.-J. Ulm). 0008-8846/$ - see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.cemconres.2007.11.014