Copper is one of the most important components in brake pads and its amount can reach up to 14%. In spite of a number of positive features copper usage in brake pad formulations has recently become the subject of considerable discussions, primarily due to concerns about potential risks related to environmental impacts of copper particles. So, for developing new pad formulations with possible replacements of copper content, it is very important to understand the functionality of copper additions to brake friction materials. In the paper theoretical investigation of the role of copper as a pad ingredient was carried out on the basis of modelling by the method of movable cellular automata (MCA). Our previous studies show that copper as a constituent of the tribofilm formed during braking provides smooth sliding by forming a granular layer of mechanically mixed materials from the friction layers. In the present study the concentration of copper particles in a Fe3O4-matrix was varied systematically in the range 5.5-28 vol. % and compared to mixtures with the same amount of graphite nanoparticles. The sliding simulations were performed while assuming material properties at 500°C in order to assess the beneficial role of copper during severe braking conditions corresponding to fading cycles during dynamometer testing.
The method of movable cellular automata (MCA) and method of molecular dynamics (MD) were applied to simulate the friction and sliding behavior of model-tribofilms formed from a nanocomposite consisting of an epoxy matrix, 10 vol % micron-sized carbon fibers and 5 vol. % silica nanoparticles. Whereas MCA considered the tribofilm as an agglomerate of silica nanoparticles released from the composite and mixed with graphite particles, MD simulated the sliding behavior of an amorphous silica layer supported by stiff crystalline substrates on both sides. The MCA model provided reasonable quantitative results which corroborate experimental findings at moderate stressing conditions. The very low coefficient of friction observed experimentally under severe stressing conditions was not explained by this model. This could be attributed to the lack of mechanical data at the high temperature expected under these conditions. Although based on a simpler assumption of the tribofilm composition, MD-modelling could be easily applied to the expected high flash temperature and was able to predict friction reduction and smooth sliding under these conditions.
The composition and nanostructure of a beneficial tribofilm formed during sliding of a hybrid nanocomposite against steel were characterized comprehensively. A similar nanostructure was produced by high energy ball milling of the three identified tribofilm constituents: silica, hematite and graphite. By supplying powders to a pin-on-disc test it has been shown that neither silica, nor hematite, nor a mixture of both provide the low coefficient of friction (COF) observed for the hybrid composite. Only if graphite was blended with the oxides, the low COF was obtained. Thus, a film of finely dispersed stable inorganic wear products containing 15 vol% graphite provides low friction and wear in the considered case.
We applied a combination of Raman spectroscopy (RS) and cross-sectional transmission electron microscopy (X-TEM) to identify silicone oil residues and tribofilms at steel disc surfaces after tribological testing. Neither chemical cleaning nor mechanical removal of a 50 µm thick surface layer produced a surface without any silicone residue. Nevertheless, long-term tribological properties are not affected due to silicone degradation which has been proved by Raman spectroscopy. Excellent anti-wear and anti-friction properties of a nanocomposite at severe stressing conditions correlated with the formation of a silica-based tribofilm containing amorphous and graphite-like carbon nanoparticles. Since reliable carbon quantification by analytical TEM is difficult, RS is a useful complementary method for carbon identification at wear scars.
In part 1 it was shown that tribofilms usually are 100 nm thick and exhibit a multiphase nanocrystalline structure. The objective of our modelling efforts was to obtain a better understanding of the sliding behaviour and associated friction properties and to study the impact of internal and external parameters on these properties. The method of movable cellular automata (MCA) was used. The third bodies were considered as aggregates of linked nanoparticles which may decompose and form a layer of granulär material, the so-called mechanically mixed layer (MML), if certain fracture criteria are fulfilled. The basic model structure which consists of Fe3Ü4 nanoparticles with 13 % graphite inclusions was used. In order to assess the robustness of the model the following parameter studies were performed. The pressure ränge at an asperity contact was varied between 15 and 50 MPa. The mechanical properties of the oxide were varied between brittle and ductile behaviour corresponding to room temperature and high temperature behaviour. The mechanical properties of the soft ingredient were varied + 50 % of the properties of graphite. The influence