Research On Temperature Field Of Layered Composites Of Steel-Copper-Plastics Under End-face Frictional Conditions

The steel-copper-plastic material with layered structure, composed of low carbon steel, copper powder and polyoxymethylene (POM) or modified POM, is one kind of oil-less self-lubrication composites. It has high loading capacity besides its self-lubrication. The steel provides the composite high strength and loading capacity, the middle-layer of copper powder supplies the effects of rapid heat transfer ability and holding the polymeric surface layer which acts as solid lubrication. This kind of composites conquers the low strength of pure plastics, and has some typical performances, for example, excellent heat exchange, small linear expansion coefficient and high anti-abrasion. It still retains the self-lubrication of plastics. Therefore, it has been used widely to produce self-lubrication bearings.In this thesis, the friction and wear behaviors of the composite with the surface layer of pure POM or POM filled by nano-MoS₂ particles were firstly investigated via the end-face tribometer. It was found that the wear resistance and friction reduction of POM filled with nano-MoS₂ particles were higher those of pure POM. Secondly, a three-dimensional transient-state mathematical model described the heat transfer behavior which was resulted from the friction on the rubbing surface of the composites was established based on the heat transfer theory, tribological principle and finite element method. At last, it was studied that the finite element model of the composites by means of ANSYS software and the temperature fields on the rubbing surface under end-face conditions. The temperature distributions on the surface layer were studied in case of different polymer composites, and the simulation results were compared with the experimental results as well. The results showed the heat conductivity of POM filled by nano-MoS₂ particles was higher that that of POM. In the end, it was obtained that the temperature distribution curves based on different heat fluxes calculated by different methods, and found that the simulation results via the reverse-reduce method was the most close to experimental values. On the other hand, the results gained by the classical contacting heat flux distribution coefficient methods were far away the experiment data.