Fast Quantification of Lithium Concentration in Non-Compliant Materials Using Laser-Induced Breakdown Spectroscopy

Featured Application: This work proposes combining the efficient and adaptable analytical technique of LIBS with a robust, accurate multivariate non-linear model. This approach offers a valid method for rapidly analyzing lithium in situ in various lithium-rich waste materials, such as ceramics, lubricants, and pharmaceuticals. Although approximately half of global lithium consumption is used in the rechargeable battery industry, lithium is also in demand for other specialized applications, such as high-temperature lubricants, ceramics, glass, and pharmaceuticals. The growing need for efficient lithium recovery and recycling underscores the importance of fast and accurate analytical tools for determining lithium concentrations in non-compliant and waste materials generated by industrial processes. In this paper, we present a machine learning-based procedure utilizing Laser-Induced Breakdown Spectroscopy (LIBS) to accurately quantify lithium concentrations in lithium-rich non-compliant materials derived from the industrial production of enamels used for coating metallic surfaces. This procedure addresses challenges such as strong self-absorption and matrix effects, which limit the effectiveness of conventional univariate calibration methods. By employing a multivariate approach, we developed a single model capable of quantifying lithium content across a wide concentration range. A comparison of the LIBS results with those obtained using conventional laboratory analysis (Inductively Coupled Plasma–Optical Emission Spectrometry, ICP-OES) confirms that LIBS can deliver the speed, precision, and reliability required for potential routine applications in the lithium recovery and recycling industry.