In the semiconductor industry, especially in the manufacturing of Through-Glass Vias (TGVs), there is an increasing need to improve the quality and efficiency of manufacturing processes. To address the challenges such as lack of efficiency, requiring substantial manual labor, and falling short in precision of traditional methods in meeting high standards for TGVs manufacturing, the approach that combines deep learning and optimization techniques was introduced to achieve automatic quality assessment and refine laser drilling parameters for TGVs manufacturing. We have developed a residual U-Net model with an accuracy of up to 87.9% by training high-resolution scanning electron microscope (SEM) images of TGVs for automatic assessment of TGVs quality, closely matching the assessments made by human experts. We used Bayesian optimization to iteratively adjust the laser drilling parameters that are crucial for TGVs manufacturing, and the quality scores obtained by residual U-Net model enhanced 13.2% after 50 iterations, which confirms the effectiveness of the integration of U-Net architecture with Bayesian optimization in achieving optimal manufacturing results.

The micromechanical properties (i.e., hardness, elastic modulus, and stress–strain curve) of AlCu films were determined by an instrumented indentation test in this work. For three AlCu films with different thicknesses (i.e., 1 µm, 1.5 µm, and 2 µm), the same critical ratio (h_max/t) of 0.15 and relative indentation depth range of 0.15–0.5 existed, within which the elastic modulus (i.e., 59 GPa) and nanoindentation hardness (i.e., 0.75 GPa, 0.64 GPa and 0.63 GPa for 1 µm, 1.5 µm and 2 µm films) without pile-up and substrate influence can be determined. The yield strength (i.e., 0.754 GPa, 0.549 GPa and 0.471 GPa for 1 µm, 1.5 µm and 2 µm films) and hardening exponent (i.e., 0.073, 0.131 and 0.150 for 1 µm, 1.5 µm and 2 µm films) of Al-(4 wt.%)Cu films for MEMS were successfully reported for the first time using a nanoindentation reverse method. In dimensional analysis, the ideal representative strain ε_r was determined to be 0.038. The errors of residual depth h_r between the simulations and the nanoindentation experiments was less than 5% when the stress–strain curve obtained by the nanoindentation reverse method was used for simulation.

Bioresorbable magnesium-metal vascular stents are gaining popularity due to their biodegradable nature and good biological and mechanical properties. They are also suitable candidate materials for biodegradable stents. Due to the rapid degradation rate of Mg metal vascular scaffolds, a Mg/Zn bilayer composite was formed by a number of means, such as magnetron sputtering and physical vapor deposition, thus delaying the degradation time of the Mg metal vascular scaffolds while providing good radial support for the stenotic vessels. However, the interlaminar compounds at the metal interface have an essential impact on the mechanical properties of the bi-material interface, especially the cracking and delamination of the Mg matrix Zn coating vascular stent in the radially expanded process layer. Intermetallic compounds (IMCs) are commonly found in dual-layer composites, such as Mg/Zn composites and multi-layer structures. They are frequently overlooked in simulations aiming to predict mechanical properties. This paper analyses the interfacial failure processes and evolutionary mechanisms of interfacial fracture mechanics of a Mg/Zn interface with an intermetallic compound layer between coated Zn and Mg matrix metallic vascular stents. The simulation results show that the fracture mode in the Mg/Zn interface with an intermetallic compound involves typical ductile fracture under static tensile conditions. The dislocation line defects mainly occur on the side of the Mg, which induces the Mg/Zn interfacial crack to expand along the interface into the pure Mg. The stress intensity factor and the critical strain energy release rate decrease as the intermetallic compound layer’s thickness gradually increases, indicating that the intensity of stress and the force of the crack extending and expanding along the crack tip are weakened. The presence of intermetallic compounds at the interface can significantly strengthen the mechanical properties of the material interface and alleviate the crack propagation between the interfaces.

Mg metal vascular stents not only have good mechanical support properties but also can be entirely absorbed by the human body as a trace element beneficial to the human body, but because Mg metal is quickly dissolved and absorbed in the human body, magnesium metal alone cannot be ideally used as a vascular stent. Since the dense oxide Zn film formed by Zn contact with oxygen in the air has good anti-corrosion performance, the Zn nanolayer film deposited on the Mg matrix vascular scaffold by magnetron sputtering can effectively inhibit the rapid dissolution of Mg metal. However, there are few studies on the molecular dynamic structural defects of Mg-matrix Zn atomic magnetron sputtering, and the atomic level simulation of Mg-matrix Zn thin-film depositions can comprehensively understand the atomic level structural defects in the deposition process of Zn thin films from a theoretical perspective to further guide experimental research. Based on this, this research first studied and analyzed the atomic layer structure defects, surface morphology, surface roughness, atomic density of different deposited layers, radial distribution function, and residual stress of the thin-film deposition layer of 1500 deposited Zn atoms at the initial deposition stage, during and after deposition under Mg-matrix thermal layer 500K and a deposited velocity 5 Å/ps by molecular dynamics. At the same time, the effects of temperature and deposited velocity of the Mg-matrix thermal layer on the surface morphology, roughness, and biaxial stress of the deposited films were studied.