The microstructure and properties of Copper-Graphite Composites (CGC) prepared by spark plasma sintering (SPS) based on two-step mixing and wet milling were investigated. The results showed that Cu powders were rolled into Cu flakes during milling, and their size significantly decreased from 23.2 to 10.9 μm when the graphite content increased from 1.0 wt.% to 2.5 wt.%. The oxidation of Cu powder was avoided during two-step mixing and wet milling. After spark plasma sintering, the graphite powders of the composites were mainly distributed at Cu grain boundaries in granular and flake shapes. The mean size of Cu grains was 9.4 um for 1.0 wt.% graphite content and reduced slightly with the increasing of graphite content. Compared with other conventional methods, the composite prepared by two-step mixing and SPS achieved higher relative density, electrical conductivity, and micro-hardness, which, respectively, reduced from 98.78%, 89.7% IACS (International annealed copper standard), and 64 HV (Vickers-hardness) to 96.56%, 81.3% IACS, and 55 HV when the graphite content increased from 1.0 wt.% to 2.5 wt.%. As the graphite content increases, the friction coefficient and wear rate of the composite decreases. When the graphite content of CGC is 1.0 wt.%, the main wear mechanism was plastic deformation, delamination, adhesive, and fatigue wear. The adhesive and fatigue wear disappeared gradually with the increasing of graphite content.

Copper–graphite sandwich composites are new functional composites with excellent tribological and mechanical properties. More and more researchers prepare the sandwich composites by accumulative roll-bonding processes due to severe plasticity deformation. The microstructure and properties of copper–graphite sandwich composites prepared by continual annealing and accumulative roll-bonding processes for eight rolling cycles were investigated. The results showed that with the increase of rolling cycles, the dispersion of graphite particles between two copper layers along a rolling direction became more uniform, the pores at the interface between graphite particles and the Cu matrix gradually disappeared, and the hardness of copper–graphite composites gradually increased. After eight rolling cycles, the Vickers hardness of the composites increased by 140.8% as compared with that of annealed pure copper. During friction, the steel balls can meet some graphite particles in the composites rolled by four cycles. The friction coefficient and wear loss of copper–graphite composites gradually decreased with the increase of rolling cycles. Compared with the composites rolled by one cycle, the friction coefficient and wear loss of the composites rolled by eight cycles were reduced by 32.5 and 49.0%, respectively. The main wear mechanism gradually evolved from fatigue wear and abrasive wear to adhesive wear, and then to fatigue wear.

Copper electrolysis is a crucial process in copper smelting. The surface of cathodic copper plates is often affected by various electrolytic process factors, resulting in the formation of nodule defects that significantly impact surface quality and disrupt the downstream production process, making the prompt detection of these defects essential. At present, the detection of cathode copper plate nodules is performed by manual identification. In order to address the issues with manual convex nodule identification on the surface of industrial cathode copper plates in terms of low accuracy, high effort, and low efficiency in the manufacturing process, a lightweight YOLOv5 model combined with the BiFormer attention mechanism is proposed in this paper. The model employs MobileNetV3, a lightweight feature extraction network, as its backbone, reducing the parameter count and computational complexity. Additionally, an attention mechanism is introduced to capture multi-scale information, thereby enhancing the accuracy of nodule recognition. Meanwhile, the F-EIOU loss function is employed to strengthen the model’s robustness and generalization ability, effectively addressing noise and imbalance issues in the data. Experimental results demonstrate that the improved YOLOv5 model achieves a precision of 92.71%, a recall of 91.24%, and a mean average precision (mAP) of 92.69%. Moreover, a single-frame detection time of 4.61 ms is achieved by the model, which has a size of 2.91 MB. These metrics meet the requirements of practical production and provide valuable insights for the detection of cathodic copper plate surface quality issues in the copper electrolysis production process.