TY - JOUR
T1 - High-precision quantification and low detection limits of chlorine and fluorine in coal via laser-induced breakdown spectroscopy
AU - Wang, Yazi
AU - Yang, Xiaoning
AU - Yi, Wen
AU - Liu, Xiaodong
AU - Sun, Haohan
AU - Zhang, Xinyu
AU - Yang, Shuoyu
AU - Du, Yongyi
AU - Zhao, Jiaqi
AU - Lv, Yue
AU - Ding, Caihao
AU - Zhong, Lixiang
AU - Liu, Ruibin
N1 - Publisher Copyright:
© 2025 The Royal Society of Chemistry.
PY - 2025/8/27
Y1 - 2025/8/27
N2 - Chlorine (Cl) and fluorine (F), as halogen elements, hold significant detection value across numerous fields (e.g., Mars exploration, industrial production, and environmental protection). Conventional detection methods are not only intricate and time-consuming but also fail to facilitate real-time analysis. In contrast, laser-induced breakdown spectroscopy (LIBS) offers simplicity, rapidity, and the capability for simultaneous multi-element detection. However, detecting Cl and F in coal using LIBS presents substantial challenges, primarily due to coal's complex matrix effects and the difficulty of exciting halogens to generate high-intensity atomic peaks. To overcome these obstacles, this study investigated the correlation between F and Cl content and characteristic peaks using statistical methods. It was found that Ca emission lines exhibit a high correlation (>0.9) with Cl and F. The unification of high correlation and low Gibbs free energy laws elucidates the influence of underlying chemical reaction pathways from thermodynamic and statistical perspectives, offering a promising solution for detecting low - content elements in complex matrices. Ultimately, principal component analysis combined with partial least squares (PCA-PLS) modeling was applied to analyze preprocessed spectra. This approach effectively mitigated the matrix effects in coal and enhanced the model's robustness. The root mean square error (RMSE) for F and Cl prediction results in the out-of-model set were 0.0447 wt% and 0.0809 wt%, respectively. This study achieved highly accurate simultaneous quantitative analysis of F and Cl at low detection limits. Specifically, the limits of detection (LOD) are 0.04 wt% for F and 0.06 wt% for Cl, respectively. Compared to the traditional LIBS atomic line method, these LODs are lower, and the detection accuracies are higher. In this study, for the first time, LIBS technology has achieved synchronous low LOD detection of F and Cl in coal, and it is capable of meeting the detection requirements for practical applications. This provides a novel approach for the rapid, efficient, and simultaneous detection of hazardous elements in coal, contributing to reduced environmental pollution.
AB - Chlorine (Cl) and fluorine (F), as halogen elements, hold significant detection value across numerous fields (e.g., Mars exploration, industrial production, and environmental protection). Conventional detection methods are not only intricate and time-consuming but also fail to facilitate real-time analysis. In contrast, laser-induced breakdown spectroscopy (LIBS) offers simplicity, rapidity, and the capability for simultaneous multi-element detection. However, detecting Cl and F in coal using LIBS presents substantial challenges, primarily due to coal's complex matrix effects and the difficulty of exciting halogens to generate high-intensity atomic peaks. To overcome these obstacles, this study investigated the correlation between F and Cl content and characteristic peaks using statistical methods. It was found that Ca emission lines exhibit a high correlation (>0.9) with Cl and F. The unification of high correlation and low Gibbs free energy laws elucidates the influence of underlying chemical reaction pathways from thermodynamic and statistical perspectives, offering a promising solution for detecting low - content elements in complex matrices. Ultimately, principal component analysis combined with partial least squares (PCA-PLS) modeling was applied to analyze preprocessed spectra. This approach effectively mitigated the matrix effects in coal and enhanced the model's robustness. The root mean square error (RMSE) for F and Cl prediction results in the out-of-model set were 0.0447 wt% and 0.0809 wt%, respectively. This study achieved highly accurate simultaneous quantitative analysis of F and Cl at low detection limits. Specifically, the limits of detection (LOD) are 0.04 wt% for F and 0.06 wt% for Cl, respectively. Compared to the traditional LIBS atomic line method, these LODs are lower, and the detection accuracies are higher. In this study, for the first time, LIBS technology has achieved synchronous low LOD detection of F and Cl in coal, and it is capable of meeting the detection requirements for practical applications. This provides a novel approach for the rapid, efficient, and simultaneous detection of hazardous elements in coal, contributing to reduced environmental pollution.
UR - https://www.scopus.com/pages/publications/105014291532
U2 - 10.1039/d5ja00199d
DO - 10.1039/d5ja00199d
M3 - Article
AN - SCOPUS:105014291532
SN - 0267-9477
VL - 40
SP - 2462
EP - 2470
JO - Journal of Analytical Atomic Spectrometry
JF - Journal of Analytical Atomic Spectrometry
IS - 9
ER -