The directed energy deposition (DED) process of metal 3D printing technologies has been treated as an effective method for welding, repairing, and even 3-dimensional building of machinery parts. In this study, stainless steel 316L (STS316L) and Inconel 625 (IN625) alloy powders are additively manufactured using the DED process, and the microstructure of the fabricated STS316L/IN625 sample is investigated. In particular, there are no secondary phases in the interface between STS316L and the IN625 alloy. The EDS and Vickers hardness results clearly show compositionally and mechanically transient layers a few tens of micrometers in thickness. Interestingly, several cracks are only observed in the STS 316L rather than in the IN625 alloy near the interface. In addition, small-sized voids 200– 400 nm in diameter that look like trapped pores are present in both materials. The cracks present near the interface are formed by tensile stress in STS316L caused by the difference in the CTE (coefficient of thermal expansion) between the two materials during the DED process. These results can provide fundamental information for the fabrication of machinery parts that require joining of two materials, such as valves.

3-D IC integration enables the smallest form factor and highest performance due to the shortest and most plentiful interconnects between chips. Direct metal bonding has several advantages over the solder-based bonding, including lower electrical resistivity, better electromigration resistance and more reduced interconnect RC delay, while high process temperature is one of the major bottlenecks of metal direct bonding because it can negatively influence device reliability and manufacturing yield. We performed quantitative analyses of the interfacial properties of Al-Al bonds with varying process parameters, bonding temperature, bonding time, and bonding environment. A 4-point bending method was used to measure the interfacial adhesion energy. The quantitative interfacial adhesion energy measured by a 4-point bending test shows 1.33, 2.25, and 6.44 J/m2 for 400, 450, and 500˚C, respectively, in a N2 atmosphere. Increasing the bonding time from 1 to 4 hrs enhanced the interfacial fracture toughness while the effects of forming gas were negligible, which were correlated to the bonding interface analysis results. XPS depth analysis results on the delaminated interfaces showed that the relative area fraction of aluminum oxide to the pure aluminum phase near the bonding surfaces match well the variations of interfacial adhesion energies with bonding process conditions.

The effect of annealing treatment conditions on the interfacial adhesion energy between electrolessplated Ni film and polyimide substrate was evaluated using a 180˚ peel test. Measured peel strength values are 26.9±0.8, 22.4±0.8, 21.9±1.5, 23.1±1.3, 16.1±2.0 and 14.3±1.3g/mm for annealing treatment times during 0, 1, 3, 5, 10, and 20 hours, respectively, at 200˚C in ambient environment. XPS and AES analysis results on peeled surfaces clearly reveal that the peeling occurs cohesively inside polyimide. This implies a degradation of polyimide structure due to oxygen diffusion through interface between Ni and polyimide, which is also closely related to the decrease in the interfacial adhesion energy due to thermal treatment in ambient conditions.

1.5 μm-thick copper films deposited on silicon wafers were successfully bonded at 415˚C/25 kN for 40 minutes in a thermo-compression bonding method that did not involve a pre-cleaning or pre-annealing process. The original copper bonding interface disappeared and showed a homogeneous microstructure with few voids at the original bonding interface. Quantitative interfacial adhesion energies were greater than 10.4 J/m2 as measured via a four-point bending test. Post-bonding annealing at a temperature that was less than 300˚C had only a slight effect on the bonding energy, whereas an oxygen environment significantly deteriorated the bonding energy over 400˚C. This was most likely due to the fast growth of brittle interfacial oxides. Therefore, the annealing environment and temperature conditions greatly affect the interfacial bonding energy and reliability in Cu-Cu bonded wafer stacks.