Mechanical Property Evolution in Lead Free Solders Subjected to Various Thermal Exposure Profiles

Abstract

With the growth of electronic packaging used by various industries, ranging from automobiles to hand-held products, reliability is a major concern. Also, to keep pace with the increasing demand of end users for integration, miniaturization, light weight, high speed, and multi-functionality of portable electronic devices such as mobile phones, digital cameras, as well as personal digital assistant (PDA), electronic packaging industries are attempting to manufacture packages of high density and smaller dimensions. These packages consist of smaller solder joint interconnects with reduced spacing between the solder joints. The reliability of smaller electronic packages is greatly affected by the environment and application where it is being used. Solder joints provide mechanical support, and electrical and thermal interconnection between packaging levels in microelectronics assembly systems. Proper functioning of these interconnections and the reliability of the electronic packages depend largely on the mechanical properties of the solder joints. Lead free solders are common as interconnects in electronic packaging due to their relatively high melting point, attractive mechanical properties, thermal cycling reliability, and environment friendly chemical properties. However, Lead free electronic assemblies are often subjected to thermal cycling during qualification testing or during actual use. During the dwells at the constant high temperature extreme, the lead free solders joints will experience thermal aging phenomena, resulting in microstructural evolution and material property degradation. Additional aging effects can also occur in the ramp periods from low to high temperature. Also, CTE mismatches between the silicon die and PCB causes shear fatigue of the solder joints and affects the reliability of the entire package. Many studies have been conducted to investigate the effect of isothermal aging on lead free solder alloy properties and the reliability of assemblies. However, no study has focused on how thermal cycling with different profiles (ramp and dwell times) affects the mechanical property evolution in lead free solders. This research involves several projects to create a database of mechanical properties of bulk and real solder joints, and the associated mechanical and microstructural evolution under different thermal exposures. In the first project, several different SAC+Bi solder alloys with various bismuth contents were investigated. In particular, a family of SAC+Bi alloys with 1%, 2%, and 3% Bi were studied with four different thermal exposure profiles (isothermal aging, slow thermal cycling, thermal shock, and thermal ramping). The primary objective of this study was to determine how much bismuth is needed in the lead-free alloy to mitigate microstructure and material property evolutions during thermal exposures. Use of lower Bi content can lower solder cost and also increase reliability in high strain rate loadings such as shock/drop/vibration. Uniaxial test specimens were prepared by reflowing solder in rectangular cross-section glass tubes with a controlled temperature profile. After reflow solidification, the samples were placed into the environmental chamber and thermally cycled from -40 C to +125 C under a stress-free condition (no load). Several thermal cycling profiles were examined including: (1) 150 minute thermal cycles with 45 minutes ramps and 30 minutes dwells, (2) air-to-air thermal shock exposures with 30 minutes dwells and near instantaneous ramps, and (3) 90 minute cycles with 45 minutes ramps and 0 minutes dwells (thermal ramping only), (4) no cycling (simple aging at high temperature extreme). After the preconditioning, mechanical properties including stress-strain were explored at room temperature. Stress-strain behavior under different thermal exposures has been compared in terms of exposure time. Mechanical behavior under different thermal exposures has been compared to explore the most detrimental thermal exposures. In the second project, the effects of thermal exposures on the creep behaviors of SAC+Bi solders were studied. A controlled temperature profile was used during the preparation of rectangular cross-section uniaxial test specimens by reflow solidification. After fabrication, the samples were exposed to various types of thermal pre-conditioning (aging, cycling, or ramping) for varying amounts of time (0-20 days) in a stress-free environment (no load). Finally, creep testing at room temperature was performed on the pre-conditioned specimens. Creep tests were performed at three different stress levels (σ = 10, 12, and 15 MPa), and the secondary creep strain rates were measured. The changes in the creep strain rate for the SAC+Bi solder alloys were measured for each type of thermal exposure and for different exposure times. Results for the SAC+Bi alloys were then compared to prior results for SAC305, and it was observed that adding bismuth had the beneficial result of significantly reducing the creep rate degradations. Higher levels of bismuth led to increased mitigations of the thermal degradation effects. In the third project, the evolution of mechanical properties of both SAC305 and SAC+Bi solder joints under different thermal exposures have been explored. Mechanical properties were recorded as modulus, hardness, and yield strength. Mechanical properties were extracted using nanoindentation technique. For nanoindentation, samples were prepared by attaching the package on epoxy mold using glue followed by grinding, polishing, and finally optical microscopy (OM) to find out single grain joint for avoiding grain orientation effect on mechanical properties. After the sample preparation, all samples were preconditioned and then tested at room temperature to measure the mechanical properties. For each exposure time, 10 indents were made in a row and average was taken to extract the mechanical properties. Mechanical properties were compared under different thermal exposures with exposure time for each solder alloy. Also, mechanical properties evolution under different thermal exposures for both bulk solder and solder joint of SAC+Bi alloys were compared to observe the effect of thermal exposures on small scale and large scale specimen. In the final project, the microstructure evolutions occurring in lead free solders subjected to several different thermal exposure profiles (isothermal aging, slow thermal ramping, and slow thermal cycling) were investigated and then the observed changes in microstructure were correlated with the previously measured mechanical property and creep behavior evolutions. For the microstructural study, a slot was made on an epoxy mold and the sample was inserted into the slot followed by grinding, polishing, and exposed in a thermal chamber for 0, 1, 5, and 20 days. After exposing the samples to different thermal profiles, the samples were viewed in an SEM to depict their microstructures, and various attributes of the microstructures (e.g. IMC particle size, particle spacing, dendrite size, etc.) were measured as a function of the thermal profile and exposure time. Correlations of the observed microstructure evolutions with the corresponding changes in mechanical properties and creep behavior were performed for each alloy, and it was demonstrated that the Ag3Sn IMC particle size (diameter) was the most significant characteristic of the microstructure that controls the SAC lead free solder mechanical behavior. More importantly, it was observed that for given alloy, the mechanical property degradation depended on the IMC particle diameter in same manner irrespective of thermal profile that caused the mechanical and microstructural evolutions. This suggests that a single curve/relation can be used for each alloy to correlate mechanical behavior and microstructure, independent of the thermal history seen by the solder. This means that, one need merely look at the microstructure of the solder to know its mechanical behavior.