Heavy wire bonding represents a pivotal technology in the field of power electronics, serving as the primary method for interconnecting semiconductors with their substrates. It is crucial for ensuring the longevity and reliability of power modules, particularly in applications that are subjected to thermal and mechanical stresses, such as those found in automotive and renewable energy systems. Prior research has predominantly concentrated on pure aluminum wires and copper bonding, with comparatively limited investigation of alloyed aluminum wires. 
The principal failure mechanism in heavy wire bonds is thermal fatigue, which is caused by a discrepancy in the coefficients of thermal expansion (CTE) between the wire material and the silicon (Si) die. This research addresses the existing gap in knowledge by testing a range of aluminum alloys under a variety of power cycling conditions and temperatures. The objective of this study is to gain insight into the impact of wire material and diameter on the longevity of wire bonds, as well as to ascertain whether alloying or wire size modifications can enhance reliability.

The study employed MOSFET power modules comprising Al₂O₃ ceramic-based direct copper bonded (DCB) substrates and a 5 μm AlSiCu0.5 topside MOSFET metallization. Wires derived from four distinct aluminum materials (Al-H11, Al-R, Al-CR, and AlX) and three wire diameters (125 μm, 300 μm, and 400 μm) were bonded onto the aforementioned MOSFETs. Power cycling tests were conducted with temperature swings (ΔT) of 75 K, 85 K, 95 K, and 105 K, in order to simulate operational conditions. A load voltage of 30 V, a heating time of 1.8 seconds, and maximum and minimum junction temperatures of 140°C and 30°C, respectively, were maintained throughout the experiment. The bonds were subjected to destructive shear testing at designated intervals to assess the extent of degradation. A microstructural analysis, including electron backscatter diffraction (EBSD), was conducted to evaluate changes in grain structure and orientation across different stages of power cycling. Furthermore, infrared thermography was employed to monitor the temperature distribution and ensure the maintenance of controlled heating across the samples.

• Wire Material Performance: The AlX wire demonstrated markedly enhanced reliability in comparison to Al-H11, Al-R, and Al-CR. The degradation patterns exhibited by Al-H11, Al-R, and Al-CR were found to be similar due to their comparable rotated cube (RC) texture formation during ultrasonic bonding. The superior performance of AlX can be attributed to its scandium (Sc) content, which provides precipitation hardening at temperatures between 200°C and 300°C, preventing the grain coarsening and softening that are typical of other aluminum alloys.
• Wire Diameter Influence: The results demonstrated that wire diameters of 125 μm exhibited significantly enhanced reliability during power cycling in comparison to larger diameters of 300 μm and 400 μm. Wires with a diameter of 125 μm exhibited a cycle life that was 10 to 15 times greater than that of wires with a diameter of 300 μm under comparable conditions. A lifetime model, utilising an exponential decay and linear regression fit function, corroborated the hypothesis that smaller diameters exhibited superior resistance to fatigue-induced shear force degradation, particularly in the context of higher temperature swings.
• Microstructural Insights: The results of the EBSD analysis indicated that for the Al-H11, Al-R, and Al-CR samples, a district (101)-A1 texture was observed to have formed above the chip metallization post-bonding. This texture, which exhibits a rotated cube (RC) orientation, was identified as the underlying cause of the comparable lifetimes observed in these wires. In contrast, the AlX wire bonds exhibited a more refined grain structure, accompanied by the presence of Sc precipitates at the grain boundaries. These precipitates were observed to play a stabilizing role under elevated temperatures. Crack propagation analysis demonstrated that in Al-H11 bonds, cracks developed at a depth of 10-20 μm above the chip metallization. In contrast, in AlX bonds, cracks propagated in a different manner, forming within the chip metallization itself, which contributed to its longer lifespan.
• Temperature Dependence: It was observed that elevated temperature fluctuations resulted in accelerated deterioration across all wire materials. At a temperature differential of 95 K, Al-H11, Al-R, and Al-CR exhibited a 50% reduction in shear force after 50,000 cycles, whereas AlX demonstrated durability beyond 500,000 cycles at comparable temperature conditions.

The findings indicate that AlX wire material, with its exceptional resistance to thermal fatigue, presents a substantial opportunity to extend the operational lifetime of power modules, particularly in high-temperature environments such as those encountered in automotive and renewable energy applications. The potential for industry adoption of AlX is that it could reduce the frequency of module replacements and improve system reliability. 
Furthermore, the research highlights the significance of wire diameter optimization for specific applications, with smaller diameters being more suitable for environments subjected to extensive thermal cycling. Future research could investigate additional alloy modifications, such as combining elements like Sc with other materials to enhance wire performance. Additionally, an investigation into alternative bonding techniques, such as ultrasonic or thermocompression bonding with these new materials, could offer further reliability gains.

A significant limitation of this study is its focus on aluminum-based alloys, which excludes other potential materials, such as copper alloys, that are known for their high electrical and thermal conductivity. Furthermore, the testing was constrained to a specific set of power cycling conditions with a defined range of temperature fluctuations, which may not fully encompass the spectrum of operational environments. 
Further research into other alloy compositions, alternative bonding surfaces, and the impact of different types of metallization (such as Ni or Ag) on wire bond longevity could enhance the current understanding. Furthermore, the study did not investigate long-term aging effects beyond the power cycling experiments. For instance, the impact of continuous thermal stress without cycling could reveal additional failure mechanisms.