PSI - Issue 79

Ibrahim T. Teke et al. / Procedia Structural Integrity 79 (2026) 17–25

18

1. Introduction Resistance spot-welding is one of the most widely used joining techniques in automotive, aerospace, and manufacturing industries due to its high speed, cost-effectiveness, and suitability for mass production. Its importance lies in providing strong, reliable joints for sheet metals while minimizing additional material use and preserving structural integrity (Akbulut, 2021; Akbulut & Ertas, 2021; Ertas, 2004, 2015; Ertas & Sonmez, 2008, 2009, 2011; Ertas, Yilmaz, & Baykara, 2008; Ertas, Vardar, Sonmez, & Solim, 2009; Ertas & Akbulut, 2021a, 2021b, 2021c).The integration of dissimilar materials such as aluminum and high-strength steels in automotive, aerospace, and structural systems is a cornerstone of modern lightweighting strategies. Aluminum alloys offer excellent strength-to-weight ratio and corrosion resistance, while steels contribute superior fatigue and impact resistance. The combination, however, introduces complexities in joining processes — especially under cyclic loads — due to thermomechanical incompatibilities, intermetallic formation, and stress localization. Resistance spot welding (RSW) remains the most prevalent joining method for thin-gauge dissimilar metals due to its low cost, process speed, and compatibility with automated production. However, welding aluminum to steel brings a series of challenges. Most critical is the formation of Fe- Al intermetallic compounds (IMCs) such as Fe₂Al₅ and FeAl₃ at the weld interface. These brittle phases significantly reduce joint ductility and fatigue resistance if their thickness exceeds 3 – 5 µm. Several studies (Shi et a l., 2019; Walker et al., 2024; Rao et al., 2018) have explored how process optimization, electrode geometry, and surface conditioning can suppress excessive IMC growth and improve weld integrity. Recent innovations in electrode design, such as the multi-ring domed (MRD) configuration, and tailored welding schedules like CSS (Conditioning – Shaping – Sizing), have been successful in refining nugget geometry and reducing interfacial reaction layers (Shi et al., 2022; Rao et al., 2018). However, coating type on the steel sheet (e.g., galvanized or Zn-coated) and electrode polarity can also influence the local heat generation and IMC formation, altering joint strength and consistency. Beyond metallurgical aspects, geometrical parameters such as notch root angle, weld nugget diameter, and sheet thickness ratio critically impact fatigue performance. As shown by Shi et al. (2020) and Ibrahim et al. (2016), increasing the notch root angle decreases local stress concentrations, thereby delaying crack initiation and transitioning the failure mode from interfacial separation to button pull-out or through-thickness fracture. Similarly, experimental studies have demonstrated that larger nugget diameters reduce principal strain amplitudes under cyclic shear loads, thus improving fatigue life. Despite advances in experimental characterization and empirical modeling, a major gap in current FE-based approaches is the insufficient representation of thermal strain and residual stress fields generated during the welding process. The mismatch in thermal expansion coefficients between aluminum and steel results in in-plane contraction and out-of-plane distortion during post-weld cooling — phenomena that significantly affect stress redistribution under service loads. While structural stress and principal strain approaches have been applied for fatigue life prediction (Kang et al., 2020; Haselhuhn et al., 2017), they often treat thermal effects implicitly. This simplification can lead to underestimation of crack-driving forces and inaccurate failure mode prediction. Moreover, dissimilar RSW joints show large scatter in fatigue life due to un-modeled variations in thermal fields and microstructural heterogeneity, particularly near the transition zones between nugget, heat-affected zone (HAZ), and base metal (Liu et al., 2013; Janardhan et al., 2021). To address these limitations, the current study proposes a novel simulation framework that integrates a thermal strain generation algorithm into a nonlinear finite element model. By mapping temperature gradients obtained from welding simulations or empirical data onto nodal displacements, the algorithm explicitly incorporates residual thermal stress fields into mechanical performance predictions. This enables more accurate assessment of fatigue-critical stress states and crack initiation zones. Validation is conducted using tensile-shear and fatigue tests across various aluminum – steel combinations, weld parameters (current, electrode force, hold time), and specimen configurations (tensile-shear, MTS, single-lap joint). The results not only affirm the impact of thermally induced strains on fatigue crack initiation, but also explain previously unexplained transitions between fracture modes under otherwise similar loading conditions.