A data-driven framework to predict the morphology of interfacial Cu6Sn5 IMC in SAC/Cu system during laser soldering

A data-driven approach combining together the experimental laser soldering,finite element analysis and machine learning,has been utilized to predict the morphology of interracial intermetallic compound(IMC) in Sn-xAg-yCu/Cu(SAC/Cu) system.Six types of SAC solders with varying weight proportion of Ag and Cu,have been processed with fiber laser at different magnitudes of power(30-50 W) and scan speed(10-240 mm/min),and the resultant IMC morphologies characterized through scanning electron microscope are categorized as prismatic and scalloped ones.For the different alloy composition and laser parameters,finite element method(FEM) is employed to compute the transient distribution of temperature at the interface of solder and substrates.The FEM-generated datasets are supplied to a neural network that predicts the IMC morphology through the quantified values of temperature dependent Jackson parameter(αJ).The numerical value of αJ predicted from neural network is validated with experimental IMC morphologies.The critical scan speed for the morphology transition between prismatic and scalloped IMC is estimated for each solder composition at a given power.Sn-0.7 Cu having the largest critical scan speed at 30 W and Sn-3.5 Ag alloy having the largest critical scan speed at input power values of 40 W and 50 W,thus possessing the greatest likelihood of forming prismatic interfacial IMC during laser soldering,can be inferred as most suitable SAC solders in applications exposed to shear loads.

Integration of machine learning with phase field method to model the electromigration induced Cu_(6)Sn_(5) IMC growth at anode side Cu/Sn interface

Currently,in the era of big data and 5G communication technology,electromigration has become a serious reliability issue for the miniaturized solder joints used in microelectronic devices.Since the effective charge number(Z*)is considered as the driving force for electromigration,the lack of accurate experimental values for Z* poses severe challenges for the simulation-aided design of electronic materials.In this work,a data-driven framework is developed to predict the Z* values of Cu and Sn species at the anode based LIQUID,Cu_(6)Sn_(5) intermetallic compound(IMC)and FCC phases for the binary Cu-Sn system undergoing electromigration at 523.15 K.The growth rate constants(kem)of the anode IMC at several magnitudes of applied low current density(j=1×10^6 to 10×10^6A/m^2)are extracted from simulations based on a 1D multi-phase field model.A neural network employing Z* and j as input features,whereas utilizing these computed kemdata as the expected output is trained.The results of the neural network analysis are optimized with experimental growth rate constants to estimate the effective charge numbers.For a negligible increase in temperature at low j values,effective charge numbers of all phases are found to increase with current density and the increase is much more pronounced for the IMC phase.The predicted values of effective charge numbers Z* are then utilized in a 2D simulation to observe the anode IMC grain growth and electrical resistance changes in the multi-phase system.As the work consists of the aspects of experiments,theory,computation,and machine learning,it can be called the four paradigms approach for the study of electromigration in Pb-free solder.Such a combination of multiple paradigms of materials design can be problem-solving for any future research scenario that is marked by uncertainties regarding the determination of material properties.