Monodispersed nano-copper particles were prepared, and the optimized method of preparation procedure was established through orthogonal experiments under four different factors. The experimental results show that the particle size of freeze-dried nano-copper particles can be adjusted by rotational speed, freezing rate, types of active agents and freezing mode, and freezing mode is the most important among all of the four factors. Tribological results of the nano-copper particles from the best optimized procedure can be improved 5%-10% than that of pure paroline oil and the lowest coefficient of friction 0.12 can be obtained for the sample of 0.05 wt%. The lubrication mechanism mostly attributed to the appropriate concentration of copper particles filling the surface pits, which is beneficial for the improvement of the tribological performance.

Ultra-smooth and low-damage processing of single-crystalline 4H-SiC has become a research focus as a substrate for third-generation semiconductor wafers. However, the high hardness and strong chemical inertia significantly affect 4H-SiC chemical-mechanical polishing (CMP) efficiency and accuracy. In this study, polishing process optimization experiments of 4H-SiC are conducted based on the grey relational analysis method to achieve low surface roughness (Ra) and high material removal rate (MRR). First, MRR and Ra of Si surface (0001) are obtained by orthogonal experiments considering down force, rotation speed, slurry flow rate and abrasive particle size as four key factors. Then the grey relational coefficient and grey relational grade of MRR and Ra are calculated by data processing. The results show that significant factors of the single-objective process are rotation speed, down force, particle size, and flow rate, while the factors of the multi-objective process are down force, flow rate, rotation speed, and particle size in turn. Finally, the MRR of 208.12 nm/h and Ra of 0.391 nm are polished using multi-objective optimization process parameters. The polishing efficiency and accuracy were improved, confirming the applicability of grey relational analysis in CMP.

No abstract

No abstract