Design of experiment assisted optimization of the bonding conditions of S235/TiAl6V4 composite

• S235 and TiAl6V4 identified as a promising material pair for laminated rotors • Regression model developed to predict joint shear strength vs bonding conditions • Thermodynamics prevailed over kinetics in forming intermetallic phases during bonding • Strength was independent of intermetallic layer thickness, while TiC increased it • Peak shear strength ∼249 MPa, likely the upper limit for the S235/TiAl6V4 pair In response to prevailing market demands with a view to reducing the carbon emissions, there is an increasing need for high-speed/high-power electrical machines. The synchronous reluctance machine (SRM) with an axially laminated rotor (ALA) is a relatively old concept, which has recently attracted renewed interest due to the new technological possibilities. It is considered a promising candidate for high-power applications. However, the optimal combination of materials for an ALA composite remains to be identified. The present paper focuses on a novel combination of materials for this application: titanium grade 5 (TiAl6V4) with mild steel (S235). The application of diffusion bonding can help with reliable joining of dissimilar materials; nevertheless, the optimisation of the bonding parameters remains a challenging aspect. Several studies have addressed the diffusion bonding of titanium with various alloys; however, it remains unclear, which combination of parameters (time, temperature, and pressure) yields the highest mechanical strength. In this work, experimental design (DoE) method was employed to evaluate these parameters, with the shear strength serving as a primary response. For understanding, how the bonding parameters affect the microstructure, scanning electron microscopy (SEM) was used. It was found that at the 95% confidence level, temperature was the only variable that maintained its statistical significance. The maximum shear strength was observed in joints, welded at 900°C, in agreement with the predictions of the DoE model.