Failures in Rubber Bonding to Substrates

Bonds between rubber and substrates can fail for a number of reasons. Section 12.1 deals with some of the causes of rubber to metal bond failures. Section 12.2 examines the type of failures which are adhesion related, in fabric or cord reinforced power transmission belts. Section 12.3 discusses a phenomenon which causes service failures of rubber components, mainly in sealing applications. This phenomenon arises through a bond which is formed between the rubber (nitrile) and the metal mating surface of a valve or similar, which is of sufficient strength to rupture the rubber surface when the valve is opened. 

Bond failures in rubber to metal products are fortunately of relatively rare occurrence. When failures do occur they can stem from a number of fundamental areas and the faults are generally very characteristic of those problem areas. The main areas of bond 

Rubber to metal bonded components have been designed and manufactured since the early days of the rubber industry and their technology and manufacture has been discussed elsewhere [1]. However, the design of the rubber to metal component in the modern car engine mount has provided other problems for the rubber manufacturer not related to 

All rubber to metal product designers must consider the effects of stresses exerted between the metal and rubber at the interface. This is particularly important at the edges of a component. Correct shaping of this part of the moulding will remove the stresses away from vulnerable conjunction points of metal and rubber. 

Many bonding failures can be traced back to faulty metal preparation, poor application of bonding agents or careless handling techniques on the factory floor. 

---

These adhesives have been found to adhere strongly to metals, glass, wood, ceramics, masonry, asphalt, leather, and plastics like polystyrene, phenolics, polycarbonates, ABS, cellulose acetate, polyesters, rubbers, and some polyolefins. In general, the most favorable results are noted in the bonding of steel and aluminum, perhaps because the bond strengths are more easily observed before substrate failure. 

Some tests specify the use of two circular plates that may attach directly to the straining frame. Rubber is bonded between the plates, which are then separated in a mode perpendicular to the plane of the plates (see Tensile tests). Very high loads should be expected in this type of test. On failure of a satisfactory bond, there should be no sign of bonding agent - rubber or bonding agent - substrate separation. Details are in ASTM D 429, Method A. 

Cohesive failure was found to be the predominant mode of failure for each rubber compound containing Saret 633 (Figure 8.7). Therefore, it would be expected that as the Saret 633 concentration is increased, the rubber compound would become stronger due to additional crosslinking, which would result in an increase in adhesive strength at the interface between rubber and substrate. This proved to be the case and is shown in Figure 8.8 for EPDM bonded to untreated steel. As the Saret 633 concentration was increased from 0 to 20 phr, the shear adhesion increased from approximately 0.55 MPa for the control to over 11.0 MPa. Cohesive failure was the predominant mode of failure at each concentration. Similar performance was observed for other rubbers, such as nitrile, natural, polybutadiene, silicone and hydrogenated nitrile. 

Consideration is given to the different steps involved in rubber-to-metal bonding, including surface preparation of metal substrates, the application of primers and adhesives, and moulding, vulcanisation, curing and posttreatment processes. Factors which can lead to weak adhesion and bond failure are discussed, and approaches to the identification and correction of such problems are outlined. 

Figure 2 shows fracture surfaces for bonds between zinc and rubber-toughened epoxy resin (see Toughened adhesives). In Fig. 2(a), there is a region of cohesive failure within the resin the sites of bubbles, which may have initiated the fracture, can be seen. In Fig. 2(b), a piece of resin is seen adhering to what appears to be the bare zinc substrate. 

---