|The eutectic Sn-Pb solder alloys are widely used in the electronic packaging industry. However, they are being replaced by new near-eutectic solder alloys because of environment and health concerns associated with lead. The performance of solder alloys can be improved by adding extra elements, such as nickel (Ni), bismuth (Bi), antimony (Sb), iron (Fe), indium (In), and aluminum (Al). Although much research has been done about mechanical, thermal, and fatigue properties of these solder alloys using the accelerated tests, the performance of individual solder joints under realistic service condition needs better understanding. Because the stress amplitude of solder joint will be frequently changed during its real service life, it is necessary to test the solder alloys under the varying amplitude condition in the accelerated test. Our study is focusing on the fatigue performance of individual SAC-Bi solder joints with isothermal aging under varying stress amplitude condition. Background introduction and literature review will be introduced in Chapter 1, Chapter 2, and Chapter 3. Chapter 4 will focus on explaining the experiment design and analysis method of our study.
In Chapter 5, three kinds of individual SAC-Bi solder joints along with SAC305 (Sn-3.0%Ag-0.5%Cu) were tested with Organic Solderability Preservative (OSP) surface finish under four single stress amplitudes condition. In this research, their fatigue properties were compared in aspect of inelastic work, plastic strain, and microstructure. Their characteristic fatigue life was investigated using Morrow Energy model and Coffin-Manson model. Typically, SAC-Q and SAC-I showed more fatigue resistance than SAC305 and SAC-R. SAC-Q and SAC-I showed similar fatigue performance, so we believe that the addition of Ni and Sb does not influence fatigue performance of SAC-I individual solder joints and Bi is the main reason. Also, a power equation of characteristic life versus stress amplitudes was fitted and we were able to find any characteristic life associated with a specific stress amplitude, which was a good reference for the following study in varying stress amplitude test.
In Chapter 6, we focused on the fatigue performance of SAC-Bi solder alloys (SAC305, SAC-Q, and SAC-R) under varying stress amplitude conditions. Typically, all the individual SAC-Bi solder joints were failed earlier than expected according to common damage accumulation model. SAC-Q showed better fatigue resistance than SAC305 and SAC-R under varying stress amplitude conditions. Hysteresis loops (Inelastic Work & Loading Slope) under varying stress amplitude conditions were analyzed. And they were compared with the case of single stress amplitude conditions. Basically, inelastic work was observed to step up after every switch between mild stress cycles and harsh stress cycles. Loading slope was observed to step down after every switch between mild stress cycles and harsh stress cycles.
In Chapter 7, we focused on the aging effect on the fatigue of SAC-Bi solder alloys (SAC305, SAC-Q, and SAC-R). After aging for 10 hours and 1000 hours aging under 125℃, all the solder alloys’ fatigue life were reduced under single stress amplitude test, especially SAC305. SAC305’s characteristic life under each single stress amplitude test would drop up to 60%. However, SAC-Q and SAC-R were seldomly influenced by aging in single stress amplitude test. Bi was believed as the main reason considering their compositions. Similar phenomena were observed under varying stress amplitude test. SAC305’s characteristic life (Switches) was significantly reduced after 10 hours aging and 1000 hours aging under 125℃. We could see a decreasing trend of characteristic life (Intervals), though that decrease was not significant. The more important is, all the solder alloys after aging were failed earlier than expected using common damage accumulation model. It indicates that a more accurate fatigue model is essential for individual solder joints under varying stress amplitude conditions.
In Chapter 8, we focused on the fatigue modeling of aged individual SAC-Bi solder joint under varying stress amplitude conditions. Because the “Step-up” phenomena of inelastic work were observed, we got a modified fatigue model based on the common damage accumulation model. In the proposed model, amplification factor was considered to quantify the amplification of inelastic work after every switching between mild and harsh stress cycles under varying stress amplitude conditions. For SAC-Bi solder alloys (SAC305, SAB-Q, and SAC-R), the amplification factor would increase linearly with the crack initiation and dramatically rise after crack propagation until total failure. The cut-off points were analyzed for all the cases and the results indicated that SAC-Q’s amplification factor would increase linearly in 83% of its total life, significantly higher than the cases of SAC305 and SAC-R. Additionally, we were surprised that the aging time would not significantly influence the cut-off point for different SAC-Bi solder alloys. Finally, the proposed fatigue model was showing better estimation of damage accumulation, when comparing with the common damage accumulation models.