Soldered joints can be made with a wide range of base materials and filler metals that allow the assembly to meet its performance and reliability requirements. Structural solder joints have, as their foremost requirement, to provide mechanical attachment between base material structures. The joint is typically subjected to one, or a combination of, three loading configurations: (a) tensile or compressive force, (b) shear force, or (c) peel force. Solder filler metals and in particular, the so-called “soft solders” based on tin (Sn), lead (Pb), and indium (In), generally have a bulk strength that is less than that of the base materials. Finally, deformation occurs largely in the solder when the joint is subjected to an applied force.
The assembly of ceramic components often uses soldering technologies to attach metal structures to the ceramic base material. Because many suitable solder alloys do not readily wet and spread on ceramics, a metallization layer is deposited on the latter to support wetting and spreading by the molten solder for completion of the joint The metallization layer must be sufficiently robust to retain its integrity through the soldering process as well as not negatively impact the long-term reliability of the joint A study was performed to evaluate the mechanical properties of solder joints made to a 0.200Ti/W-40Cu-2.0Pt-0.375Au (pm) thin-film metallization deposited on low-temperature co-fired ceramic (LTCC) base materials. The solder joints were made with the 63Sn-37Pb solder (wt-%, abbreviated Sn-Pb). A pin pull test was developed to measure the tensile strength of the solder joint as a function of soldering parameters. Failure mode analysis was a critical metric for assessing the roles of interfaces, bulk solder, and the ceramic on mechanical performance. The Sn-Pb solder joints experienced a nominal strength loss with increased severity of the soldering process parameters. The strength decline was attributed to changes in the solder joint microstructure, and not degradation to the thin film structures.
This report examines the role of interfaces in electronic packaging applications with the focus placed on soldering technology. Materials and processes are described with respect to their roles on the performance and reliability of associated interfaces. The discussion will also include interface microstructures created by coatings and finishes that are frequently used in packaging applications. Numerous examples are cited to illustrate the importance of interfaces in physical and mechanical metallurgy as well as the engineering function of interconnections. Regardless of the specific application, interfaces are non-equilibrium structures, which has important ramifications for the long-term reliability of electronic packaging.
This study examined the cause of nonwetted regions of the gold (Au) finish on iron-nickel (Fe-Ni) alloy lids that seal ceramic packages using the 80Au-20Sn solder (wt %, abbreviated Au-Sn) and their impact on the final lid-to-ceramic frame solder joint. The Auger electron spectroscopy (AES) surface and depth profile techniques identified surface and through-thickness contaminants in the Au metallization layer. In one case, the AES analysis identified background levels of carbon (C) contamination on the surface; however, the depth profile detected Fe and Ni contaminants that originated from the plating process. The Fe and Ni could impede the completion of wetting and spreading to the edge of the Au metallization. The Au layer of lids not exposed to a Au-Sn solder reflow step had significant surface and through-thickness C contamination. Inorganic contaminants were absent. Subsequent simulated reflow processes removed the C contamination from the Au layer without driving Ni diffusion from the underlying solderable layer. An Au metallization having negligible C contamination developed elevated C levels after exposure to a simulated reflow process due to C contamination diffusing into it from the underlying Ni layer. However, the second reflow step removed that contamination from the Au layer, thereby allowing the metallization to support the formation of lid-to-ceramic frame Au-Sn joints without risk to their mechanical strength or hermeticity.
Materials aging is a high-consequence failure mode in electronic systems. Such mechanisms can degrade the electrical properties of connectors, relays, wire bonds, and other interconnections. Lost performance will impact, not only that of the device, but also the function and reliability of next-level assemblies and the weapons system as a whole. The detections of changes to materials surfaces at the nanometer-scale resolution, provides a means to identify aging processes at their early stages before they manifest into latent failures that affect system-level performance and reliability. Diffusion will be studied on thin films that undergo accelerated aging using the nanometer scale characterization technique of Frequency Modulated Kelvin Probe Force Microscopy (FM-KPFM). The KPFM provides a relatively easy, non-destructive methodology that does not require high-vacuum facilities to obtain nanometer spatial resolution of surface chemistry changes. The KPFM method can provide the means to measure surface, and near-surface, elemental concentrations that allow the determination of diffusion rate kinetics. These attributes will be illustrated by assessing diffusion in a thin film couple. Validation data will obtained from traditional techniques: (a) Auger electron spectroscopy (AES), x-ray fluorescence (XRF), and xray photoelectron spectroscopy (XPS).
The run-out phenomenon was observed in Ag-Cu-Zr active braze joints made between the alumina ceramic and Kovar™ base material. Run-out introduces a significant yield loss by generating functional and/or cosmetic defects in brazements. A prior study identified a correlation between run-out and the aluminum (Al) released by the reduction/oxidation reaction with alumina and aluminum's reaction with the Kovar™ base material. A study was undertaken to understand the fundamental principles of run-out by examining the interface reaction between Ag-xAl filler metals (x = 2,5, and 10 wt-%) and Kovar™ base material. Sessile drop samples were fabricated using brazing temperatures of 965° (T769°F) or 995°C 0823°F) and times of 5 or 20 min. The correlation was made between the degree of wetting and spreading by the sessile drops and the run-out phenomenon. Wetting and spreading increased with Al content (x) of the. Ag-xAl filler metal, but was largely insensitive to the brazing process parameters. The increased Al concentration resulted in higher Al contents of the (Fe, Ni, Co)xAly reaction layer. Run-out was predicted when the filler metal has a locally elevated Al content exceeding 2-5 wt-%. Several mitigation strategies were proposed, based upon these findings.
Whether structural or electronic, all solder joints must provide the necessary level of reliability for the application. The Part 1 report examined the effects of filler metal properties and the soldering process on joint reliability. Filler metal solderability and mechanical properties, as well as the extents of base material dissolution and interface reaction that occur during the soldering process, were shown to affect reliability performance. The continuation of this discussion is presented in this Part 2 report, which highlights those factors that directly affect solder joint reliability. There is the growth of an intermetallic compound (IMC) reaction layer at the solder/base material interface by means of solid-state diffusion processes. In terms of mechanical response by the solder joint, fatigue remains as the foremost concern for long-term performance. Thermal mechanical fatigue (TMF), a form of low-cycle fatigue (LCF), occurs when temperature cycling is combined with mismatched values of the coefficient of thermal expansion (CTE) between materials comprising the solder joint “system.” Vibration environments give rise to high-cycle fatigue (HCF) degradation. Although accelerated aging studies provide valuable empirical data, too many variants of filler metals, base materials, joint geometries, and service environments are forcing design engineers to embrace computational modeling to predict the long-term reliability of solder joints.