Rubber Adhesion of Brass-Plated Steel Tire Cords: Fundamental Study of the Effects of Compound Formulation Variations on Adhesive Properties

Abstract

Abstract The rubber adhesion of brass-plated steel tire cord has been studied with emphasis on the effects on unaged and aged adhesion of compound formulation variations. A series of 16 test compounds was formulated in which the levels of sulfur, accelerator, stearic acid, zinc oxide and cobalt salt were varied. Adhesion data for two steel cord constructions—4 × 0.25 and 7 × 4 × 0.175 + 0.15—obtained under various unaged and aged conditions are presented. Despite the relatively minor compound differences, a wide variation of adhesion characteristics was observed. Good adhesion following overcure, humidity aging and steam aging was obtained with compounds containing high sulfur levels and high sulfur to accelerator ratios. The adhesion data do not correlate with compound crosslink densities or other physical properties. Rates of brass sulfidization do not correlate with adhesion data either. This was studied in model systems with squalene replacing the natural rubber. Analysis by high-resolution Auger electron spectroscopy revealed differences between the compositions of the sulfide films formed by the good and the poor mixtures of the series. The locus of the adhesion failure of the poor compounds was found to be the interface between copper sulfide and zinc oxide, the two reaction products formed when brass is bonded to rubber. A model was presented which accounted for the effects of compound formulation variations. It is postulated that brass is not sulfidized by the free sulfur in the mixture but by an intermediate containing reactive sulfur. The composition of this intermediate is dependent on the compound formulation. This leads to different types of sulfide films formed by good or poor compounds. This model is based on the chemistry of squalene model system mixtures, which was studied by chromatography. The ranking of crosslink densities observed for the various squalene mixtures was very similar to that determined for the corresponding NR mixtures. Some implications of the findings of this study for tire cord and compounding practice were discussed.