May 2026

Journal

Optimizing aluminum oxide passivation layers—Laser-induced plasmas for layer formation, real-time diagnostics, and layer evaluation

By:
Milota, Peggy T; Browning, Katie L; Sacci, Robert L; Manard, Benjamin T; Phongikaroon, Supathorn; Andrews, Hunter B
Journal Name:
Spectrochimica Acta Part B: Atomic Spectroscopy
Page Number:
107536
Volume:
241
Publication Date:
May 18, 2026
View DOI Listing:
https://doi.org/10.1016/j.sab.2026.107536

Abstract

Molten salts have beneficial thermophysical and electrolytic properties, but the aggressive nature of molten salts requires corrosion mitigation strategies, such as structural material surface treatment. In this study, a laser-induced breakdown spectroscopy (LIBS) system was used to simultaneously ablate and monitor de-excitation spectra to form and identify dense alumina passivation layers. Additionally, high-frequency (kHz) LIBS imaging allowed rapid elemental mapping and depth profiling for surface O/Al ratios via minimally destructive analysis. The paired t-test rejected the null hypothesis that air and Ar cover gases were equivalent (p = 0.00012), and the analysis of variance (ANOVA) showed that overlap between shots (p = 0.03112) and cover gas flow rate (p = 0.02729) were statistically significant. At 75% overlap, the O/Al ratio increased to 10% ± 2%, and 90% overlap rose further to 31% ± 2%. Regardless of overlap, tuning the gas flow from 0.5 to 2.5 L min−1 enhanced the O/Al ratio by 24% ± 4%. LIBS depth profiling showed a 3× increase in layer thickness from 75% to 90% overlap. Molecular band peaks of AlO, a precursor to Al2O3 found during treatment, suggest that band intensity could be used for real-time Al2O3 optimization. Ultimately, by adjusting laser spot overlap and cover gas flow rate, a 266 nm laser was shown to form Al2O3 layers on an Al6061 sample, with cover gas flow rate primarily affecting the O/Al ratio and layer thickness correlating with laser spot overlap. This combination of molecular spectra collection, imaging, and depth profiling demonstrated the robust layer analysis capabilities of LIBS.