Photons produced through laser-induced excitation of a sample typically encounter absorption by atoms or ions of the same class along the radiation path as they propagate outward. This self-absorption phenomenon not only induces a dip at the top of the spectral line intensity but also results in an expansion of the spectral line width. Consequently, the accurate representation of the elemental composition of a sample is compromised, leading to a potential deviation in the precision of quantitative analysis. Spectra significantly affected by self-absorption can have detrimental effects on laser-induced breakdown spectroscopy (LIBS) applications. The objective of this study is to obtain laser-induced breakdown spectra free from self-absorption. We employ a novel method to achieve a precise quantitative analysis of elements within a sample with the aim of mitigating the effects of self-absorption on the accuracy of the obtained results.

If the upper level transition states corresponding to different transition line wavelengths of a selected component element are the same (or approximately the same) in the ionized state Z, the intensity ratio of the doublet lines is linked solely to the physical parameters of the transition associated with the two spectral lines. This ratio remains independent of the experimental device, conditions, and time evolution. Assuming similar energy-level structures and wavelengths for the two selected lines, their changes in intensity are anticipated to be similar. Experimental measurements involve determining the actual intensity ratio at different delay time. When the experimental ratio at a specific moment equals or approximates the theoretical value, that moment is considered the point at which optical thinning of the plasma occurs. Theoretically, this represents the moment in the radiation spectrum of the plasma without self-absorption, enabling the acquisition of the intensity of the plasma radiation line in the self-absorption-free state. Utilizing spectral values in this state can significantly enhance the accuracy of elemental concentration inversion.

We utilize aluminum as a representative element to validate the proposed method, specifically focusing on the spectral lines of Al 396.15 nm and Al 394.40 nm, where the upper energy levels of both lines are 3.143 eV. Utilizing the data provided in Table 1, the intensity ratio of these doublet lines is computed as 1.983. Precise control of the delay time following the emission of the laser excitation pulse is crucial for achieving accurate delay sampling, as shown in Fig. 1. The following experimental results are obtained: 1) The intensity ratio of Al 396.15 nm and Al 394.40 nm spectral lines decreases with time, indicating the occurrence of optical thinness (minimum self-absorption) between 200 ns and 400 ns of the acquisition time (Fig. 2). 2) Analyzing the relationship between the spectral signal-to-noise ratio (SNR) and the optimal time of the corresponding optically thin plasma under different integration time enables us to ensure the SNR before selecting the appropriate delay time (Fig. 3). 3) Samples with varying aluminum contents exhibit different time of optical thinness (minimum self-absorption), suggesting that samples with different aluminum contents require different delay time to minimize the self-absorption effects. This observation may explain the measurement errors associated with the fixed delay time method used in traditional LIBS (Fig. 4). 4) As the aluminum mass fraction increases from 0 to 19.5%, the occurrence time of optical thinness gradually decreases, approaching 0. This indicates that LIBS technology when is used to analyze samples with an aluminum mass fraction exceeding 19.5% will result in distorted spectra due to a significant self-absorption influence, thus rendering accurate aluminum content inversion unattainable (Fig. 5).

This study validates the technology for the precise quantitative analysis of aluminum using self-absorption-free laser-induced breakdown spectroscopy (SAF-LIBS). The theoretical intensity ratio I_{Al 396.15nm}/I_{Al 394.40nm}=1.983 is calculated using the Al spectroscopic parameters at 396.15 nm and 394.40 nm. This ratio serves as the optically thin criterion, and the temporal evolution of the experimental intensity ratio of Al 396.15 nm and Al 394.40 nm spectral lines is examined. Experimental results reveal the following: 1) The optimal spectral acquisition time depends on the elemental content of the sample. 2) The maximum element content that can be accurately measured by LIBS is constrained by the minimum optically thin time. Specifically, LIBS can precisely measure the aluminum mass fraction of samples ranging from 0 to 15.9%. However, for aluminum samples with mass fractions exceeding 19.5%, LIBS is unable to obtain spectra with minimal self-absorption effects, resulting in an inability to achieve precise measurements.