Publication: Discrete phase model simulation and optimization of different types nano-reinforced solder fillet using taguchi analysis and neural network method
Loading...
Date
2021-12-01
Authors
Muhamed Mukhtar, Muhamed Abdul Fatah
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Recently, the electronic industry required electronic components to become reliable, lightweight, and miniature. In order to ensure the functionality of electronic devices, advanced
joining through the nano-reinforced solder material was required. In previous studies, the addition of different types of nanoparticles to lead-free solders has been studied. One of the strengthened elements added, such as NiO and TiO2, helps the solder fillet to have better mechanical properties. Nevertheless, there is still a large research gap in the miniaturized part assembly processes through nano-reinforced solder paste on the actual surface mount devices. Hence, this research aims to investigate the advanced joining of ultra-fine packages using the nano-reinforced solder paste. Nanoparticles material name, TiO2, Fe2O3, and NiO with (0.01wt. %, 0.05 wt. %, and 0.15 wt. %) were selected to be reinforced with lead-free solder paste (SAC 305 type 5) to form three different types of nanoparticles samples. The nano-reinforced solder paste was contrasted with the pure SAC305 solder paste in terms of material and mechanical properties. In the current analysis, a two-way interaction is implemented using both the fluid volume method (VOF) and the disperse phase method (DPM) to account for the interaction between both the nanoparticles and the molten solder. DPM simulation is capable of viewing the comprehensive trajectory of nanoparticles as it undergoes thermal reflow from SAC305 based on the comparison of the simulation the experimental result. Additionally, for all cases of nanoparticles being used, strong agreement can be seen between both experimental and simulation data collected. By using different experimental techniques, the microstructure,
fillet height, hardness and Modulus Young were investigated. The experimental results showed that the presence of nanoparticles generally strengthened the ultra-fine solder joint's
material and mechanical properties. The reflow soldering process parameters were optimized by Taguchi technique based on Taguchi’s L16 orthogonal array. The optimum weighted
percentage and nanoparticle material types were determined, and their percentage of contribution was estimated by applying the signal-to-noise ratio and analysis of variance. By
adding 0.15 wt. % of TiO2, Fe2O3, and NiO, respectively, it increased the hardness and shear strength of the solder joint. The optimal configuration and the highest mean value of hardness and the modulus Young were obtained by 0.15 wt. % of SAC305+TiO2 nanoparticles were0.2875GPa and 89.65GPa. Meanwhile, optimal configuration and the highest mean for fillet height were obtained by 0.05 wt. % of SAC305+NiO nanoparticles was 0.1719mm. Quasi Newton method of Neural Network (NN) was used to train the experimental values. In comparison between experimental and neural network model results, the percentage error of predictive models for fillet height, hardness, and modulus Young were -0.046% , 1.077% ,and 20.420 %. This research provides engineers with a deep understanding of the ultra-fine package features of the nano-reinforced solder joint in the microelectronics industry. The results are expected to provide an appropriate guide and reference for the electronics industries in order to develop nano-reinforced solder joints of miniaturized electronic packages in the future.