Laser-induced breakdown spectroscopy based on corrosion transfer methods: a reliable strategy for quantitative analysis of intergranular corrosion susceptibility in aluminum alloys

Abstract

The susceptibility of Al-Mg alloys to sensitization during thermal exposure, leading to intergranular corrosion (IGC) via grain boundary β-phase (Al₃Mg₂) precipitation, poses a significant threat to their structural integrity in marine applications. Quantitative assessment of the degree of sensitization (DoS) remains a challenge, as standard methods like the nitric acid mass loss test (NAMLT) are destructive and slow, while direct laser-induced breakdown spectroscopy (LIBS) suffers from insufficient spatial resolution and strong matrix effects. This work presents a novel corrosion transfer approach coupled with LIBS to overcome these limitations. The method selectively extracts grain boundary precipitates onto a filter paper substrate via an optimized ammonium persulfate etching process, effectively isolating the analyte from the aluminum matrix. This transfer eliminates matrix effects and bypasses spatial resolution constraints by concentrating the target phases. The strategy was evaluated on 5083 and 5456 aluminum alloys with varying DoS. The transferred samples were analyzed by LIBS, and the spectra were processed using Partial Least Squares Regression (PLSR) to establish a quantitative model against NAMLT reference values. The models for both alloy series achieved high predictive accuracy, with coefficients of determination (R²) exceeding 0.9, root mean square errors (RMSE) of ≤ 5 mg/cm², and average relative errors (ARE) reduced to approximately 20%. The results demonstrate that the corrosion transfer-LIBS technique provides a rapid, micro-destructive, and reliable strategy for quantifying IGC susceptibility. Its accuracy and efficiency meet the demands for industrial field inspection, showing significant potential for high-throughput quality control and lifetime assessment of critical aluminum components.