Measurement of the practical adhesion of paint coatings to metallic sheets by the pull-off and three-point flexure tests

In this paper, we compare the practical adhesion measurements of alkyd, vinyl, epoxy, and polyurethane paints on cold-rolled steel using the pull-off and three-point flexure tests. It has been shown that (a) in the pull-off test, the ultimate load and ultimate strain values depend on the stud area. During trimming, cracks are created within the system which are responsible for a large decrease in the parameters measured. On the other hand, it is impossible to differentiate between the failure initiation and the failure propagation zones; (b) in the three-point flexure test, the area (W) subtended by the load/displacement curves corresponding to samples with and without stiffener does not depend on the bonded width or on the substrate compliance. After carrying out such a mechanical test, it is also possible to discriminate between failure initiation and failure propagation using appropriate tools for observation, such as scanning electron microscopy, and for analysis, such as electron microprobe analysis and Fourier transform infrared spectroscopy. W is shown to be representative of the failure initiation energy. To give a clear indication of the formation of the paint/substrate interphasial zone, we have studied the practical adhesion of different types of acrylic paint (thermoplastic acrylic binder filled with TiO₂, BaSO₄, or CaCO₃, with or without crosslinked polyurethane) applied onto aluminium substrates (treated in a sulphochromic acid bath). Five different types of failure initiation were observed, each type corresponding to quite different energies (W). Failure initiation takes place within the interphasial layer for thin coatings (thickness 70 μm). For thicker paint layers, cohesive failures were observed. A model showing the composition of the paint/metal interphase is proposed.