An analysis of adhesive properties of brittle coatings

Abstract

In a previous study [1] a pendular scratching test was used to perform single-tip abrasion under controlled conditions on massive samples. During operation a pendulum swung down and an abrasive tip, which was fixed radially to this rigid pendulum, scratched a horizontal specimen at the bottom of the trajectory. The length 2L and the maximum depth D of the arc-shaped groove produced could be varied by adjusting the radial position of the tip. A transition scratching depth D* from ductile to brittle abrasion was then defined: this transition depth was found to be correlated to the ratio Kc/H, where K c is the toughness and H the hardness. The relationship was similar to that obtained in static indentation [2, 3] and the transition depth D* was found to fit the relationship D* = ( Kc/H) 15 The incident velocity of the abrasive tip was 2 m s -1 . These scratching tests were run with half-conical indentors (apex angle 90°). The flat surface was in front of the indentor, at a right angle to the scratching direction during experiments. Scratching tests are also commonly used for testing the adhesive properties of brittle coatings [4, 5]. The applied load is increased until brittle chipping occurs, the phenomenon appears similar to that obtained on massive samples and described above. This critical load of brittle chipping is sometimes used to characterize the adhesive proper- ties of coatings. Pendular experiments were conducted in order to analyse the process of scratching of semibrittle coatings and to discern the effects of the brittleness (ratio H/Kc) from effects of the adhesive proper- ties. Tests were run with the same half-conical indentors described above and,with conical inden- tors, apex angle 90 °. The incident velocity of the abrasive tip was 2ms -I. The energy consumed during abrasion was given by the loss of potential energy of the pendulum. A particular material was used in order to observe easily the different effects: aluminium (99.9% pure) coated with commercial unsaturated polyester (UP) resin. Samples were prepared in order to obtain good or poor adhesion of the coating. The thickness of the coating was e = 0.05 ram. Tests were also run on massive samples of aluminium and polyester. A typical morphology of the surface damage is shown in Fig. 1. When the indentor scratches only the coating, a first critical depth D1 is observed. This depth D1 characterizes the onset of lateral cracking of the coating and corresponds to the transition