Al–Si alloys are widely used for lightweight components in the automotive and aerospace industries because they can simultaneously provide excellent castability, high thermal conductivity, good corrosion resistance, and high specific strength. Owing...
Al–Si alloys are widely used for lightweight components in the automotive and aerospace industries because they can simultaneously provide excellent castability, high thermal conductivity, good corrosion resistance, and high specific strength. Owing to industrial requirements for complex-shaped parts, these components are primarily manufactured via casting. However, residual stresses form during processing due to non-uniform cooling rates between the surface and interior of the casting, which can cause warpage and distortion and thereby degrade fatigue life and structural integrity. Residual stress is the stress remaining in a material in the absence of external loads or temperature changes; therefore, accurate prediction and analysis are essential to ensure high reliability and quality. Recent prediction simulations based on solidification and heat-treatment behavior have been developed; however, their predictive accuracy remains limited because they do not sufficiently incorporate the effects of microscale residual stresses induced by the microstructure. Although prediction simulations based on solidification and heat-treatment behavior have recently been developed, their accuracy remains limited because they do not sufficiently account for microscale residual stresses induced by microstructure. Accordingly, this study sought to elucidate the principal factors influencing residual stress based on cast alloys with varying Si contents and actual cast products.
Based on the maximum solubility limit of Si, cast alloys of Al–1.2Si, Al–1.6Si, Al–2.0Si, and Al–10Si were fabricated with hypo- and hyper-additions of Si. In the low-Si alloys (Al–1.2Si, Al–1.6Si, and Al–2.0Si), a network-like eutectic Si phase was observed, whereas in Al–10Si alloy, the eutectic Si phase appeared predominantly in acicular and short-rod morphologies. The solid solubility of Si in each alloy was estimated using Vegard’s law. Owing to the negligible solid solubility in the low-Si alloys (Al–1.2Si, Al–1.6Si, and Al–2.0Si), a low level of compressive residual stress was observed. In contrast, in the Al–10Si alloy, stress concentration sites were formed due to the acicular and short-rod eutectic Si morphology, and it was confirmed that the combined effects of stress concentration and lattice contraction associated with Si dissolution (up to 1.22 wt.%) contributed to the residual stress behavior.
To examine the practical applicability of the trends observed in the binary alloys, a total of ten measurement locations were selected based on simulation results for an actual cast product, comprising five sites in the high-stress region and five sites in the low-stress region. At each location, the Si area fraction, Intermetallic compound (IMC) area fraction, and Si sphericity were selected as the primary microstructural factors, and the residual stress after polishing was used as the response variable in a design of experiments (DOE). As a result, Si sphericity was identified as a statistically significant factor at the 0.05 significance level (p ≤ 0.05).
As a result of comprehensively analyzing microstructural factors that affect residual-stress behavior in alloys with under- and over-additions based on the maximum solid solubility of Si and in actual cast products, it was confirmed that, in Al–Si alloys, the Si phase is a dominant variable in the formation and distribution behavior of residual stress. In addition, by applying the DOE, it was demonstrated that it is effective in clearly distinguishing the significance of interdependent variables. Therefore, the results of this study could be used as basic data for establishing a residual-stress prediction simulation model that reflects microstructural factors at the microscopic scale.