Comparative study of direct and indirect image-based profilometry in characterization of surface roughness Ž. Pavlović, a D. Risović b * and D. Novaković a In this study, we have investigated and compared two different approaches to the characterization of surface structure and roughness. The results of the stylus profilometric method are compared with the results obtained using a relatively novel, indirect, image-based profilometry. The aim was to evaluate the performance and practical usefulness of the indirect method in the characterization of a surface topography. The indirect approach involved the use of Gwyddion software for analysis of scanning electron microscope (SEM) images and calculation of standard profilometric parameters. It is well known that SEM micrographs provide an excellent tool for visualization and qualitative description of surface topography, including the estimation of corresponding fractal dimensions. The results of this study demonstrate that it is also possible to obtain profilometric parameters from analysis of SEM micrographs with appropriately calibrated gray scale intensity distributions, and that the values of the parameters are comparable to those obtained by classical (stylus) profilometry. Better agreement with results of the direct profilometric method was achieved for parameters related to the distribution of heights than those related to the distribution of depths. Regarding these differences, we have provided arguments indicating that the values of these depths – related parameters obtained by indirect profilometry – are closer to the “true” values than those inferred from the direct profilometry. Generally, the results of this comparative study indicate that indirect image-based profilometry is a valuable and efficient tool in the characterization of various surface’s topographies. Copyright © 2012 John Wiley & Sons, Ltd. Keywords: profilometry; surface roughness; surface topography; SEM micrograph; printing plate; aluminium oxide Introduction Functional properties of materials used in many engineering industries are often defined by the surface structure and its char- acteristics. This is the reason why the precise determination and characterization of roughness and surface topography is of utmost importance. There are many methods for the analysis and description of surface topographies. Among these, imaging methods such as scanning electron microscopy (SEM) and atomic force microscopy (AFM) are widely used for surface visualization and characterization. In addition to these methods for visualiza- tion of surface topography, various profilometric methods are used to provide quantitative topographical information. These can be further divided into contact and non-contact profilometric methods. The former group includes mechanical stylus profilo- metry and the latter group includes optical/interferometer and laser-based profilometry. These methods are used to quantify the results of profilometric measurements in terms of different surface roughness parameters. [1–4] However, these parameters cannot successfully describe surface irregularity or complexity. Especially, precise topography characterization of surfaces with asymmetric roughness cannot be achieved through conventional roughness parameters because there is a shift between the mean surface level and the true surface. [5] To describe complex surface topographies, the concept of fractals has been introduced. It is based on self-similarity of sur- faces at different scales. The advantage of this concept lies in its insensitivity to the structural details as well as in the fact that the structure is characterized by a single descriptor only, the fractal dimension. [6] The fractal dimension can be conveniently estimated from SEM or AFM images/micrographs [7–11] and seems to be well correlated to the profilometric parameters. [9,12] More- over, it is related to basic material properties [13] and different mechanisms that influence the surface topography. [14,15] A further advantage of this imaging method is the possibility of inferring information on three-dimensional (3D) topographical surface data from a stereoscopic pair of SEM images. [16–19] Another recent development, software for the extraction of profilometric para- meters from SEM/AFM image analysis, [20] seems to be a promising tool in the characterization and quantification of surface topogra- phies. By analyzing spatial gray scale intensity distributions in a single SEM image, or several images obtained at different tilt angles and merged together producing a virtual 3D image, the software calculates the number of standard profilometric para- meters. Recently, a similar approach was used to determine the micron scale roughness of volcanic surfaces. [21] In this context, the aim of this study was to compare the results of a direct stylus profilometric method with the results obtained by application of the novel indirect image-based profilometry to evaluate the performance and practical usefulness of the indi- rect method in the characterization of a surface topography. * Correspondence to: D. Risović, Molecular Physics Laboratory, Rudjer Bosković Institute, PO Box 180, HR-10002, Zagreb, Croatia. E-mail: drisovic@irb.hr a Faculty of Technical Sciences, Graphic Engineering and Design, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia b Molecular Physics Laboratory, Rudjer Bosković Institute, PO Box 180, HR-10002, Zagreb, Croatia Surf. Interface Anal. 2012, 44, 825–830 Copyright © 2012 John Wiley & Sons, Ltd. Research article Received: 26 August 2011 Revised: 22 December 2011 Accepted: 16 January 2012 Published online in Wiley Online Library: 7 February 2012 (wileyonlinelibrary.com) DOI 10.1002/sia.4889 825