Silica reinforced tyre rubbers: mechanistic aspects of the role of coupling agents

Since the introduction of the "green" tyre, silica is more and more used as a reinforcing filler in tyre applications. Silica when compared to carbon black, strongly reduces the rolling resistance of a tyre, which in turn leads to a lower fuel consumption of the car. The disadvantage of the use of silica is the mixing problem in introduces. Due to its polar surface, silica is difficult to mix with a hydrocarbon rubber. The solution for this problem lies in the use of a coupling agent, which shields the polar silica surface on one hand and reacts with the rubber matrix on the other hand.The coupling agent most commonly used ofr these applications in industry is bis(triethoxy-silylpropyl) tetrasuphide (TESPT). The reaction of this coupling agent with the silica surface has been extensively studied by different authors. It is now widely accepted, that one silanol group of the coupling agent after hydrolysis reacts with a silanol group on the silica surface in a primary reaction. This primary reaction is followed by a secondary reaction in which the remaining silanol groups of the coupling agent react with silanol groups of neighbouring coupling agents. The reaction of the coupling agent with the rubber matrix however, has not been studied to such an extent as the reaction with the silica surface and is still not completely understood.The objective of the research was to investigate the reaction of the coupling agent with the rubber matrix, in an attempt to better understand the reinforcement mechanism for silica-filled rubbers. The interaction between silica and rubber via the coupling agent was studied using three different coupling agent systems: variations on TESPT, silane-accelerator combinations and thiophosphoryl compounds. Their effects on processing as well as on dynamic mechanical properties were investigated. 1H NMR relaxation measurements were used to obtain additional information concering the interaction between rubber and silica filler in presence and absence of a coupling agent.The coupling agent systems selected for investigation were synthesised and characterised. Variations of TESPT were easy to synthesise, while the sysntehsis of silane-accelarator combinations and of thiphosphoryl compounds was more complicated. Fortunately, at least one coupling agent for each selected system could be syntehs8ised. The influence of these different coupling agent systems on properties of a silica reinforced tyre tread compound was studied.The effect of the sulphur- and carbon chain length of a silane coupling agent showed the importance of the existence of a rubber-reactive functional group in a coupling agent. The length of the sulphur chain of a silane coupling agent like TESPT has a strong influence on the processing behaviour of a silica reinforced tyre tread compound, especially on scorch safety. Decreasing the sulphur rank of TESPT from four to two, results in a better scorch safety and therefore a broader processing window, while final properties are comparable to a TESPT reference compound when correction for the reduced amount of sulphur is applied in the curing package. For a sulphur free coupling agent an increase in the length of the carbon chain improves the processing behaviour. Less energy is consumed during mixing when the length of the carbon chain is increased. The Payne effect, which is comonly interpreted as a measure for filler dispersion, is lower for a sulphur free silane, indicating a better filler dispersion. Mechanical properties however, fail when compared to a TESPT reference compound. When sulphur correction is applied in the curing package the tensile properties improve due to an increase in crosslink density, but remain inferior to TESPT. The fact that a sulphur free silane with an equimolar amount of sulphur added to the curing package, still shows inferior tensile properties, compared to TESPT, can be interpreted as the effect of the absence of a chemical link between silica filler and rubber via the coupling agent. In teh presence of a rubber-reactive group in the form of sulphur built-in into the coupling agent, there is a strong indication of the formation of a rubber-to-filler link via the coupling agent.Triethoxysilylpropylbenzothiazole disulphide (TESBD) was investigated as a compound containing a silane and an accelarator functionality. A silica reinforced tyre tread compound containing TESBD shows a lower Payne effect, thus indicating a better filler dispersion. The accelerator built into TESBD turns out to be too slow to act as an accelerator for the curing reaction, which is also reflected in inferior mechanical properties of the compound containing TESBD when compared to TESPT. A small amount of amine in the form of DPG turns out to be sufficient to release the accelerator part of TESBD, resulting in properties grossly comparable to TESPT.Diisopropylthiophosphoryltriethoxysilylpropyl monosulphide (DIPTESM) was studied as a coupling agent containing a thiophosphoryl group. The Payne-effect of a DIPTESM-containing silica-reinforced tyre tread compound is far inferior to the TESPT reference compound. This high Payne-effect indicates a less pronounced shielding of the silica surface for DIPTESM, resulting in an inferior silica dispersion compared to the TESPT reference, which is also reflected in the mechanical properties. The release of teh thiophosphoryl accelerator part turns out to be not effective enough to result in an appreciable corsslinking density. Addition of an extra amount of accelerator is still necessary in order to obtain properties comparable to the TESPT reference.The indication above that a rubber-to-filler link is formed via the coupling agent is based on differences in mechanical properties, but could still not be convincingly proven with the techniques applied there. 1H_NMR T2 relaxation, which has proved to be a successful technique for the study of rubber-filler interactions, was therefore used to investigate interactions between Natural Rubber (NR) and pure as well as grafted silica. Different amounts of immobilised NR were found, depending on grafting material on the silica and grafting density. Based on the results described above a lower amount of immoblised NR was expected for a grafting material without a rubber-reactive group, propyltriethoxy silane (PTES). Indeed, using 1H NMR relaxation measurements, a lower proportion of immobilised NR is found for the PTES-grafted silica. A highter proportion of immobilised NR is measured for the silane-accelerator combination TESBD when compared to TESPT, by 1H NMR relaxation. The better overall properties of the silane-accelerator combination TESBD compared to TESPT can be interpreted as the result of a higher affinity of TESBD towards the rubber matrix when compared to TESPT. This then explains the higher amount of immobilised NR found for TESBD-grafted silica, when compared to the TESPT-grafted silica.The results of 1H NMR relaxation point towards the formation of a strong physical or chemical network around the silica surface. Based on the 1H NMR relaxation measurements no conclusion can be drawn either whether the network is physical or chemical in nature. The effects of the different silane coupling agents investigated, indicate the formation of a chemical network. Admittedly, all techniques add more or less evidence and none of them is fully conclusive. Combining all results gathered it may be stated that they all point to a chemical rather than a physical bond of the coupling agent to the rubber molecule, whereby it forms a chemical bridge between the polar silica and apolar rubber.