Arrhenius Fatigue Life Modeling for Lead-Free Solder Joints in Accelerated Combined Fatigue and Creep Tests at Different Operating Temperatures

Abstract

Frequent failures in interconnected materials in microelectronic packages play a vital role in determining the reliability of electronic devices. Differences in the thermal expansion coefficients of the solder joint, printed circuit board (PCB), and electronic package are a major source of applied stress on solder joints. Due to health concerns and governmental regulations associated with using leaded solder alloys, SAC solder alloys are commonly used as an alternative to Pb-alloys because of their outstanding solderability and reliability. The shear fatigue creep accelerated test at different load levels (16 MPa, 20 MPa, and 24 MPa) was implemented in this study by using a universal micro-testing machine to investigate solder joint reliability. The creep effects were demonstrated by employing 60 seconds of dwell time for each alternating stress value and comparing its impact on solder joint reliability with the fatigue test results. The accelerated test was applied to SAC305 (96.5% Sn, 3% Ag, 0.5% Cu) solder joints with OSP surface finish at different testing temperatures (25°C, 60°C, and 100°C). Seven replicates were considered as a sample size for the reliability analysis. A two-parameter Weibull distribution was employed as the dominant distribution to describe the fatigue behavior at each condition. The characteristic life and shape parameters at each condition were extracted from the obtained Weibull models. The stress–life equation was used to estimate the reliability performance of the solder joints against fluctuations in stress amplitude at differing testing temperatures. The inelastic work per cycle and plastic strain were calculated from the acquired hysteresis loop under each condition. Coffin–Manson and Morrow energy models illustrate the relationships between fatigue life versus plastic strain and inelastic work, respectively. The effects of the oscillated testing temperature on the stress–life model, Morrow energy model, and Coffin–Manson model were elaborated using the Arrhenius equation. The results indicated a severe degradation of fatigue life with increased testing temperature. Raising the stress amplitude had a less significant impact on fatigue life reduction than increasing the testing temperature. The reliability prediction models, as a function of the fatigue properties and testing temperature, were derived from the Arrhenius, Coffin–Manson, and Morrow energy models. The acquired prediction equations produced acceptable goodness-of-fit values. Finally, a general reliability model with a 99% adequacy value (R <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> ) was constructed as a function of testing temperature and stress amplitude.