Wavelength-dependent measurement and evaluation of surface topographies:: application of a new concept of window roughness and surface transfer function

The technological performance of surfaces in fields such as tribology, biocompatibility and optics is often highly dependent on surface topography and roughness. The common practice of applying 'integral' roughness parameters, however, is often an incomplete and unsatisfactory way to describing surface topographies. Wavelength-dependent roughness evaluation is shown to be a successful method for the description of surface topographies in various characteristic roughness ranges, as well as being a useful indicator of the effect of surface treatment processes. This study examined the effects of common cut-off (COF) (high- and low-frequency) and average filtering (AFT) as well as fast Fourier transformation (FFT) techniques, as applied to synthetic and experimental profiles. Furthermore, the relationship between the roughness value R-q and the amplitudes C-n of the FFT power spectrum is demonstrated. To characterize a particular surface process consisting of several consecutive processes, a surface treatment transfer function is defined using individual FFT coefficients C-n(x) for each surface treatment step. To illustrate the application for industrial surfaces, two-dimensional (2-D) profiles on a micromachined steel surface (calibration sample) as well as on lacquered car body sheet and titanium implant surfaces were measured with a non-contact laser profilometer (LPM) and evaluated. FFT is shown to be a powerful method for the calculation of the wavelength-dependent roughness, as well as for the back-transformation of partial data sets into profiles in predefined wavelength ranges ("window roughness"). The more conventional filter methods give similar results, but some information are lost due to the fact that they do not correspond to a set of orthonormal functions. The concept of wavelength-dependent ("window") roughness parameters is demonstrated to be a concept that allows for a much more detailed description of surface topographical properties with two main merits: (a) it allows macroscopic physico-technological properties to be correlated with roughness contributions in selected wavelength ranges; (b) the overall effect of consecutive surface treatment processes can be separated into wavelength-dependent contributions from each treatment step. The concept is demonstrated for consecutive surface treatment processes in the pretreatment and lacquering of aluminium car body sheet and for blasting and etching processes in the case of titanium medical implants. (C) 2000 Elsevier Science S.A. All rights reserved.